Hi,
This patch series (based on next-20210816) implements stricter (no struct member overflows) bounds checking for memcpy(), memmove(), and memset() under CONFIG_FORTIFY_SOURCE. To quote a later patch in the series:
tl;dr: In order to eliminate a large class of common buffer overflow flaws that continue to persist in the kernel, have memcpy() (under CONFIG_FORTIFY_SOURCE) perform bounds checking of the destination struct member when they have a known size. This would have caught all of the memcpy()-related buffer write overflow flaws identified in at least the last three years.
As this series introduces various helpers and performs several phases of treewide cleanups, I'm expecting to carry this series in my tree, so I'd love to get some Reviews and Acks. Given the size, I've mostly aimed this series at various mailing lists, otherwise the CC size got really big. :)
I have separated a few things off of this submission, as things continue to grow: - run-time bounds checking (which now has a better solution for unknown bounds besides "oh well") - __alloc_size enablement (sent separately[1]) - -Warray-bounds enablement (several cleanups have been sent)
The remaining portions (here) haven't seen as much change, so I think it's better to get these compile-time portions landed first -- they're big enough. My intention is to put this in -next soon via my "overflow" tree, assuming people are happy with it. :)
Another random thing of note, I'm using the "Enhanced-by:" tag to mean "this was improved by...", inspired by the existing "Fixed-by:" tag (for "folded fixes"). "Fixed-by:" seemed inaccurate (it wasn't broken; it is just better now), and "Suggested-by:" seemed too nebulous (what was suggested?).
Thanks!
-Kees
[1] https://lore.kernel.org/lkml/20210818050841.2226600-1-keescook@chromium.org/
v2: - fix CC_IS_CLANG typo (nathan) - bug.h hunk moved later (ndesaulnier) - various commit log typos (ndesaulnier) - omap3isp: fix read source (gustavo) - struct_group can be variadic (rasmus) - add struct_group_tagged (dan williams) - move raw __struct_group() to UAPI (gregkh, dan vetter) - ibmvscsi: add BUILD_BUG_ON() size checks - net: use memset_after() instead of struct_group() (jakob) - swap some memset_after() to memset_startat() for readability - add missing __NO_FORTIFY to arm boot stub (lkp) - powerpc signal32 struct_group() added (lkp) - powerpc sata_fsl struct_group() added - add fortify failure condition arguments to warning funcs to assist debugging - add ray_cs compile-time fix - add flexcan compile-time fix - add af_iucv compile-time fix - fix dropped strcpy() compile-time write overflow check - add missing strscpy() compile-time read overflow check - fix more boot stub build instances where fortification must be disabled - add reviewed-bys/acks v1: https://lore.kernel.org/lkml/20210727205855.411487-1-keescook@chromium.org/
Specifically, this series is logically split into several steps:
Clean up remaining simple compile-time memcpy() warnings: ipw2x00: Avoid field-overflowing memcpy() net/mlx5e: Avoid field-overflowing memcpy() rpmsg: glink: Replace strncpy() with strscpy_pad() pcmcia: ray_cs: Split memcpy() to avoid bounds check warning
Introduce struct_group() and apply it treewide to avoid compile-time memcpy() warnings: stddef: Introduce struct_group() helper macro cxl/core: Replace unions with struct_group() skbuff: Switch structure bounds to struct_group() bnxt_en: Use struct_group_attr() for memcpy() region mwl8k: Use struct_group() for memcpy() region libertas: Use struct_group() for memcpy() region libertas_tf: Use struct_group() for memcpy() region thermal: intel: int340x_thermal: Use struct_group() for memcpy() region iommu/amd: Use struct_group() for memcpy() region cxgb3: Use struct_group() for memcpy() region intersil: Use struct_group() for memcpy() region cxgb4: Use struct_group() for memcpy() region bnx2x: Use struct_group() for memcpy() region drm/amd/pm: Use struct_group() for memcpy() region staging: wlan-ng: Use struct_group() for memcpy() region drm/mga/mga_ioc32: Use struct_group() for memcpy() region net/mlx5e: Use struct_group() for memcpy() region HID: cp2112: Use struct_group() for memcpy() region media: omap3isp: Use struct_group() for memcpy() region sata_fsl: Use struct_group() for memcpy() region
Prepare fortify for additional hardening: compiler_types.h: Remove __compiletime_object_size() lib/string: Move helper functions out of string.c fortify: Move remaining fortify helpers into fortify-string.h fortify: Explicitly disable Clang support fortify: Fix dropped strcpy() compile-time write overflow check fortify: Prepare to improve strnlen() and strlen() warnings fortify: Allow strlen() and strnlen() to pass compile-time known lengths
Add compile-time and run-time tests: fortify: Add compile-time FORTIFY_SOURCE tests lib: Introduce CONFIG_TEST_MEMCPY
Enable new compile-time memcpy() and memmove() bounds checking: fortify: Detect struct member overflows in memcpy() at compile-time fortify: Detect struct member overflows in memmove() at compile-time
Clean up remaining simple compile-time memset() warnings: scsi: ibmvscsi: Avoid multi-field memset() overflow by aiming at srp
Introduce memset_after() and memset_startat() helpers and apply them (and struct_group()): string.h: Introduce memset_after() for wiping trailing members/padding xfrm: Use memset_after() to clear padding ipv6: Use memset_after() to zero rt6_info netfilter: conntrack: Use memset_startat() to zero struct nf_conn net: 802: Use memset_startat() to clear struct fields net: dccp: Use memset_startat() for TP zeroing net: qede: Use memset_startat() for counters mac80211: Use memset_after() to clear tx status ath11k: Use memset_startat() for clearing queue descriptors iw_cxgb4: Use memset_startat() for cpl_t5_pass_accept_rpl intel_th: msu: Use memset_startat() for clearing hw header IB/mthca: Use memset_startat() for clearing mpt_entry btrfs: Use memset_startat() to clear end of struct tracing: Use memset_startat() to zero struct trace_iterator drbd: Use struct_group() to zero algs cm4000_cs: Use struct_group() to zero struct cm4000_dev region KVM: x86: Use struct_group() to zero decode cache dm integrity: Use struct_group() to zero struct journal_sector HID: roccat: Use struct_group() to zero kone_mouse_event RDMA/mlx5: Use struct_group() to zero struct mlx5_ib_mr powerpc/signal32: Use struct_group() to zero spe regs ethtool: stats: Use struct_group() to clear all stats at once can: flexcan: Use struct_group() to zero struct flexcan_regs regions net/af_iucv: Use struct_group() to zero struct iucv_sock region powerpc: Split memset() to avoid multi-field overflow
Enable new compile-time memset() bounds checking: fortify: Detect struct member overflows in memset() at compile-time
Enable Clang support: fortify: Work around Clang inlining bugs
arch/arm/boot/compressed/string.c | 1 + arch/powerpc/include/asm/processor.h | 6 +- arch/powerpc/kernel/signal_32.c | 6 +- arch/s390/lib/string.c | 3 + arch/x86/boot/compressed/misc.c | 3 +- arch/x86/boot/compressed/misc.h | 2 + arch/x86/boot/compressed/pgtable_64.c | 2 + arch/x86/kvm/emulate.c | 3 +- arch/x86/kvm/kvm_emulate.h | 19 +- arch/x86/lib/memcpy_32.c | 1 + arch/x86/lib/string_32.c | 1 + drivers/ata/sata_fsl.c | 10 +- drivers/block/drbd/drbd_main.c | 3 +- drivers/block/drbd/drbd_protocol.h | 6 +- drivers/block/drbd/drbd_receiver.c | 3 +- drivers/char/pcmcia/cm4000_cs.c | 9 +- drivers/cxl/cxl.h | 61 +--- drivers/gpu/drm/amd/include/atomfirmware.h | 9 +- .../drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 +- .../drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 +- .../amd/pm/inc/smu13_driver_if_aldebaran.h | 3 +- .../gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 6 +- .../gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 12 +- .../drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 6 +- drivers/gpu/drm/mga/mga_ioc32.c | 27 +- drivers/hid/hid-cp2112.c | 14 +- drivers/hid/hid-roccat-kone.c | 2 +- drivers/hid/hid-roccat-kone.h | 12 +- drivers/hwtracing/intel_th/msu.c | 4 +- drivers/infiniband/hw/cxgb4/cm.c | 5 +- drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +- drivers/infiniband/hw/mthca/mthca_mr.c | 3 +- drivers/iommu/amd/init.c | 9 +- drivers/macintosh/smu.c | 3 +- drivers/md/dm-integrity.c | 9 +- drivers/media/platform/omap3isp/ispstat.c | 5 +- drivers/net/can/flexcan.c | 68 ++-- .../net/ethernet/broadcom/bnx2x/bnx2x_stats.c | 7 +- .../net/ethernet/broadcom/bnx2x/bnx2x_stats.h | 14 +- drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c | 4 +- drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h | 14 +- drivers/net/ethernet/chelsio/cxgb3/sge.c | 9 +- drivers/net/ethernet/chelsio/cxgb4/sge.c | 8 +- drivers/net/ethernet/chelsio/cxgb4/t4_msg.h | 2 +- drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h | 10 +- drivers/net/ethernet/chelsio/cxgb4vf/sge.c | 7 +- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 +- .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 4 +- .../net/ethernet/mellanox/mlx5/core/en_tx.c | 2 +- drivers/net/ethernet/qlogic/qede/qede_main.c | 2 +- drivers/net/wireguard/queueing.h | 4 +- drivers/net/wireless/ath/ath11k/hal_rx.c | 13 +- drivers/net/wireless/ath/carl9170/tx.c | 11 +- .../net/wireless/intel/ipw2x00/libipw_rx.c | 56 +--- .../net/wireless/intersil/hostap/hostap_hw.c | 5 +- .../wireless/intersil/hostap/hostap_wlan.h | 14 +- drivers/net/wireless/intersil/p54/txrx.c | 6 +- drivers/net/wireless/marvell/libertas/host.h | 10 +- drivers/net/wireless/marvell/libertas/tx.c | 5 +- .../marvell/libertas_tf/libertas_tf.h | 10 +- .../net/wireless/marvell/libertas_tf/main.c | 3 +- drivers/net/wireless/marvell/mwl8k.c | 10 +- drivers/net/wireless/ray_cs.c | 4 +- drivers/rpmsg/qcom_glink_native.c | 2 +- drivers/scsi/ibmvscsi/ibmvscsi.c | 3 +- drivers/staging/wlan-ng/hfa384x.h | 16 +- drivers/staging/wlan-ng/hfa384x_usb.c | 4 +- .../intel/int340x_thermal/acpi_thermal_rel.c | 5 +- .../intel/int340x_thermal/acpi_thermal_rel.h | 48 +-- fs/btrfs/root-tree.c | 6 +- include/linux/compiler-gcc.h | 2 - include/linux/compiler_types.h | 4 - include/linux/fortify-string.h | 308 +++++++++++++----- include/linux/if_vlan.h | 6 +- include/linux/skbuff.h | 9 +- include/linux/stddef.h | 47 +++ include/linux/string.h | 43 ++- include/linux/thread_info.h | 2 +- include/net/iucv/af_iucv.h | 10 +- include/net/mac80211.h | 7 +- include/uapi/drm/mga_drm.h | 22 +- include/uapi/linux/omap3isp.h | 21 +- include/uapi/linux/stddef.h | 21 ++ kernel/trace/trace.c | 4 +- lib/.gitignore | 2 + lib/Kconfig.debug | 7 + lib/Makefile | 35 ++ lib/string.c | 210 +----------- lib/string_helpers.c | 201 ++++++++++++ lib/test_fortify/read_overflow-memchr.c | 5 + lib/test_fortify/read_overflow-memchr_inv.c | 5 + lib/test_fortify/read_overflow-memcmp.c | 5 + lib/test_fortify/read_overflow-memscan.c | 5 + lib/test_fortify/read_overflow2-memcmp.c | 5 + lib/test_fortify/read_overflow2-memcpy.c | 5 + lib/test_fortify/read_overflow2-memmove.c | 5 + .../read_overflow2_field-memcpy.c | 5 + .../read_overflow2_field-memmove.c | 5 + lib/test_fortify/test_fortify.h | 35 ++ lib/test_fortify/write_overflow-memcpy.c | 5 + lib/test_fortify/write_overflow-memmove.c | 5 + lib/test_fortify/write_overflow-memset.c | 5 + lib/test_fortify/write_overflow-strcpy-lit.c | 5 + lib/test_fortify/write_overflow-strcpy.c | 5 + lib/test_fortify/write_overflow-strlcpy-src.c | 5 + lib/test_fortify/write_overflow-strlcpy.c | 5 + lib/test_fortify/write_overflow-strncpy-src.c | 5 + lib/test_fortify/write_overflow-strncpy.c | 5 + lib/test_fortify/write_overflow-strscpy.c | 5 + .../write_overflow_field-memcpy.c | 5 + .../write_overflow_field-memmove.c | 5 + .../write_overflow_field-memset.c | 5 + lib/test_memcpy.c | 288 ++++++++++++++++ net/802/hippi.c | 2 +- net/core/skbuff.c | 14 +- net/dccp/trace.h | 4 +- net/ethtool/stats.c | 15 +- net/ipv6/route.c | 4 +- net/iucv/af_iucv.c | 2 +- net/netfilter/nf_conntrack_core.c | 4 +- net/xfrm/xfrm_policy.c | 4 +- net/xfrm/xfrm_user.c | 2 +- scripts/test_fortify.sh | 59 ++++ security/Kconfig | 3 + 124 files changed, 1489 insertions(+), 686 deletions(-) create mode 100644 lib/test_fortify/read_overflow-memchr.c create mode 100644 lib/test_fortify/read_overflow-memchr_inv.c create mode 100644 lib/test_fortify/read_overflow-memcmp.c create mode 100644 lib/test_fortify/read_overflow-memscan.c create mode 100644 lib/test_fortify/read_overflow2-memcmp.c create mode 100644 lib/test_fortify/read_overflow2-memcpy.c create mode 100644 lib/test_fortify/read_overflow2-memmove.c create mode 100644 lib/test_fortify/read_overflow2_field-memcpy.c create mode 100644 lib/test_fortify/read_overflow2_field-memmove.c create mode 100644 lib/test_fortify/test_fortify.h create mode 100644 lib/test_fortify/write_overflow-memcpy.c create mode 100644 lib/test_fortify/write_overflow-memmove.c create mode 100644 lib/test_fortify/write_overflow-memset.c create mode 100644 lib/test_fortify/write_overflow-strcpy-lit.c create mode 100644 lib/test_fortify/write_overflow-strcpy.c create mode 100644 lib/test_fortify/write_overflow-strlcpy-src.c create mode 100644 lib/test_fortify/write_overflow-strlcpy.c create mode 100644 lib/test_fortify/write_overflow-strncpy-src.c create mode 100644 lib/test_fortify/write_overflow-strncpy.c create mode 100644 lib/test_fortify/write_overflow-strscpy.c create mode 100644 lib/test_fortify/write_overflow_field-memcpy.c create mode 100644 lib/test_fortify/write_overflow_field-memmove.c create mode 100644 lib/test_fortify/write_overflow_field-memset.c create mode 100644 lib/test_memcpy.c create mode 100644 scripts/test_fortify.sh
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
libipw_read_qos_param_element() copies a struct libipw_info_element into a struct libipw_qos_information_element, but is actually wanting to copy into the larger struct libipw_qos_parameter_info (the contents of ac_params_record[] is later examined). Refactor the routine to perform centralized checks, and copy the entire contents directly (since the id and len members match the elementID and length members):
struct libipw_info_element { u8 id; u8 len; u8 data[]; } __packed;
struct libipw_qos_information_element { u8 elementID; u8 length; u8 qui[QOS_OUI_LEN]; u8 qui_type; u8 qui_subtype; u8 version; u8 ac_info; } __packed;
struct libipw_qos_parameter_info { struct libipw_qos_information_element info_element; u8 reserved; struct libipw_qos_ac_parameter ac_params_record[QOS_QUEUE_NUM]; } __packed;
Cc: Stanislav Yakovlev stas.yakovlev@gmail.com Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- .../net/wireless/intel/ipw2x00/libipw_rx.c | 56 ++++++------------- 1 file changed, 17 insertions(+), 39 deletions(-)
diff --git a/drivers/net/wireless/intel/ipw2x00/libipw_rx.c b/drivers/net/wireless/intel/ipw2x00/libipw_rx.c index 5a2a723e480b..7a684b76f39b 100644 --- a/drivers/net/wireless/intel/ipw2x00/libipw_rx.c +++ b/drivers/net/wireless/intel/ipw2x00/libipw_rx.c @@ -927,7 +927,8 @@ static u8 qos_oui[QOS_OUI_LEN] = { 0x00, 0x50, 0xF2 }; static int libipw_verify_qos_info(struct libipw_qos_information_element *info_element, int sub_type) { - + if (info_element->elementID != QOS_ELEMENT_ID) + return -1; if (info_element->qui_subtype != sub_type) return -1; if (memcmp(info_element->qui, qos_oui, QOS_OUI_LEN)) @@ -943,57 +944,34 @@ static int libipw_verify_qos_info(struct libipw_qos_information_element /* * Parse a QoS parameter element */ -static int libipw_read_qos_param_element(struct libipw_qos_parameter_info - *element_param, struct libipw_info_element - *info_element) +static int libipw_read_qos_param_element( + struct libipw_qos_parameter_info *element_param, + struct libipw_info_element *info_element) { - int ret = 0; - u16 size = sizeof(struct libipw_qos_parameter_info) - 2; + size_t size = sizeof(*element_param);
- if ((info_element == NULL) || (element_param == NULL)) + if (!element_param || !info_element || info_element->len != size - 2) return -1;
- if (info_element->id == QOS_ELEMENT_ID && info_element->len == size) { - memcpy(element_param->info_element.qui, info_element->data, - info_element->len); - element_param->info_element.elementID = info_element->id; - element_param->info_element.length = info_element->len; - } else - ret = -1; - if (ret == 0) - ret = libipw_verify_qos_info(&element_param->info_element, - QOS_OUI_PARAM_SUB_TYPE); - return ret; + memcpy(element_param, info_element, size); + return libipw_verify_qos_info(&element_param->info_element, + QOS_OUI_PARAM_SUB_TYPE); }
/* * Parse a QoS information element */ -static int libipw_read_qos_info_element(struct - libipw_qos_information_element - *element_info, struct libipw_info_element - *info_element) +static int libipw_read_qos_info_element( + struct libipw_qos_information_element *element_info, + struct libipw_info_element *info_element) { - int ret = 0; - u16 size = sizeof(struct libipw_qos_information_element) - 2; + size_t size = sizeof(struct libipw_qos_information_element) - 2;
- if (element_info == NULL) + if (!element_info || !info_element || info_element->len != size - 2) return -1; - if (info_element == NULL) - return -1; - - if ((info_element->id == QOS_ELEMENT_ID) && (info_element->len == size)) { - memcpy(element_info->qui, info_element->data, - info_element->len); - element_info->elementID = info_element->id; - element_info->length = info_element->len; - } else - ret = -1;
- if (ret == 0) - ret = libipw_verify_qos_info(element_info, - QOS_OUI_INFO_SUB_TYPE); - return ret; + memcpy(element_info, info_element, size); + return libipw_verify_qos_info(element_info, QOS_OUI_INFO_SUB_TYPE); }
/*
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use flexible arrays instead of zero-element arrays (which look like they are always overflowing) and split the cross-field memcpy() into two halves that can be appropriately bounds-checked by the compiler.
We were doing:
#define ETH_HLEN 14 #define VLAN_HLEN 4 ... #define MLX5E_XDP_MIN_INLINE (ETH_HLEN + VLAN_HLEN) ... struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(wq, pi); ... struct mlx5_wqe_eth_seg *eseg = &wqe->eth; struct mlx5_wqe_data_seg *dseg = wqe->data; ... memcpy(eseg->inline_hdr.start, xdptxd->data, MLX5E_XDP_MIN_INLINE);
target is wqe->eth.inline_hdr.start (which the compiler sees as being 2 bytes in size), but copying 18, intending to write across start (really vlan_tci, 2 bytes). The remaining 16 bytes get written into wqe->data[0], covering byte_count (4 bytes), lkey (4 bytes), and addr (8 bytes).
struct mlx5e_tx_wqe { struct mlx5_wqe_ctrl_seg ctrl; /* 0 16 */ struct mlx5_wqe_eth_seg eth; /* 16 16 */ struct mlx5_wqe_data_seg data[]; /* 32 0 */
/* size: 32, cachelines: 1, members: 3 */ /* last cacheline: 32 bytes */ };
struct mlx5_wqe_eth_seg { u8 swp_outer_l4_offset; /* 0 1 */ u8 swp_outer_l3_offset; /* 1 1 */ u8 swp_inner_l4_offset; /* 2 1 */ u8 swp_inner_l3_offset; /* 3 1 */ u8 cs_flags; /* 4 1 */ u8 swp_flags; /* 5 1 */ __be16 mss; /* 6 2 */ __be32 flow_table_metadata; /* 8 4 */ union { struct { __be16 sz; /* 12 2 */ u8 start[2]; /* 14 2 */ } inline_hdr; /* 12 4 */ struct { __be16 type; /* 12 2 */ __be16 vlan_tci; /* 14 2 */ } insert; /* 12 4 */ __be32 trailer; /* 12 4 */ }; /* 12 4 */
/* size: 16, cachelines: 1, members: 9 */ /* last cacheline: 16 bytes */ };
struct mlx5_wqe_data_seg { __be32 byte_count; /* 0 4 */ __be32 lkey; /* 4 4 */ __be64 addr; /* 8 8 */
/* size: 16, cachelines: 1, members: 3 */ /* last cacheline: 16 bytes */ };
So, split the memcpy() so the compiler can reason about the buffer sizes.
"pahole" shows no size nor member offset changes to struct mlx5e_tx_wqe nor struct mlx5e_umr_wqe. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences and optimizations).
Cc: Saeed Mahameed saeedm@nvidia.com Cc: Leon Romanovsky leon@kernel.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Alexei Starovoitov ast@kernel.org Cc: Daniel Borkmann daniel@iogearbox.net Cc: Jesper Dangaard Brouer hawk@kernel.org Cc: John Fastabend john.fastabend@gmail.com Cc: netdev@vger.kernel.org Cc: linux-rdma@vger.kernel.org Cc: bpf@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 4 ++-- drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c | 4 +++- 2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 4f6897c1ea8d..8997476c20cc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -200,7 +200,7 @@ static inline int mlx5e_get_max_num_channels(struct mlx5_core_dev *mdev) struct mlx5e_tx_wqe { struct mlx5_wqe_ctrl_seg ctrl; struct mlx5_wqe_eth_seg eth; - struct mlx5_wqe_data_seg data[0]; + struct mlx5_wqe_data_seg data[]; };
struct mlx5e_rx_wqe_ll { @@ -216,7 +216,7 @@ struct mlx5e_umr_wqe { struct mlx5_wqe_ctrl_seg ctrl; struct mlx5_wqe_umr_ctrl_seg uctrl; struct mlx5_mkey_seg mkc; - struct mlx5_mtt inline_mtts[0]; + struct mlx5_mtt inline_mtts[]; };
extern const char mlx5e_self_tests[][ETH_GSTRING_LEN]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 2f0df5cc1a2d..efae2444c26f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -341,8 +341,10 @@ mlx5e_xmit_xdp_frame(struct mlx5e_xdpsq *sq, struct mlx5e_xmit_data *xdptxd,
/* copy the inline part if required */ if (sq->min_inline_mode != MLX5_INLINE_MODE_NONE) { - memcpy(eseg->inline_hdr.start, xdptxd->data, MLX5E_XDP_MIN_INLINE); + memcpy(eseg->inline_hdr.start, xdptxd->data, sizeof(eseg->inline_hdr.start)); eseg->inline_hdr.sz = cpu_to_be16(MLX5E_XDP_MIN_INLINE); + memcpy(dseg, xdptxd->data + sizeof(eseg->inline_hdr.start), + MLX5E_XDP_MIN_INLINE - sizeof(eseg->inline_hdr.start)); dma_len -= MLX5E_XDP_MIN_INLINE; dma_addr += MLX5E_XDP_MIN_INLINE; dseg++;
The use of strncpy() is considered deprecated for NUL-terminated strings[1]. Replace strncpy() with strscpy_pad() (as it seems this case expects the NUL padding to fill the allocation following the flexible array). This additionally silences a warning seen when building under -Warray-bounds:
./include/linux/fortify-string.h:38:30: warning: '__builtin_strncpy' offset 24 from the object at '__mptr' is out of the bounds of referenced subobject 'data' with type 'u8[]' {aka 'unsigned char[]'} at offset 24 [-Warray-bounds] 38 | #define __underlying_strncpy __builtin_strncpy | ^ ./include/linux/fortify-string.h:50:9: note: in expansion of macro '__underlying_strncpy' 50 | return __underlying_strncpy(p, q, size); | ^~~~~~~~~~~~~~~~~~~~ drivers/rpmsg/qcom_glink_native.c: In function 'qcom_glink_work': drivers/rpmsg/qcom_glink_native.c:36:5: note: subobject 'data' declared here 36 | u8 data[]; | ^~~~
[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nu...
Cc: Andy Gross agross@kernel.org Cc: Bjorn Andersson bjorn.andersson@linaro.org Cc: Ohad Ben-Cohen ohad@wizery.com Cc: Mathieu Poirier mathieu.poirier@linaro.org Cc: linux-arm-msm@vger.kernel.org Cc: linux-remoteproc@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Gustavo A. R. Silva gustavoars@kernel.org Link: https://lore.kernel.org/lkml/20210728020745.GB35706@embeddedor --- drivers/rpmsg/qcom_glink_native.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/rpmsg/qcom_glink_native.c b/drivers/rpmsg/qcom_glink_native.c index 05533c71b10e..c7b9de655080 100644 --- a/drivers/rpmsg/qcom_glink_native.c +++ b/drivers/rpmsg/qcom_glink_native.c @@ -1440,7 +1440,7 @@ static int qcom_glink_rx_open(struct qcom_glink *glink, unsigned int rcid, }
rpdev->ept = &channel->ept; - strncpy(rpdev->id.name, name, RPMSG_NAME_SIZE); + strscpy_pad(rpdev->id.name, name, RPMSG_NAME_SIZE); rpdev->src = RPMSG_ADDR_ANY; rpdev->dst = RPMSG_ADDR_ANY; rpdev->ops = &glink_device_ops;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Split memcpy() for each address range to help memcpy() correctly reason about the bounds checking. Avoids the future warning:
In function 'fortify_memcpy_chk', inlined from 'memcpy_toio' at ./include/asm-generic/io.h:1204:2, inlined from 'ray_build_header.constprop' at drivers/net/wireless/ray_cs.c:984:3: ./include/linux/fortify-string.h:285:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 285 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/ray_cs.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c index 590bd974d94f..d57bbe551630 100644 --- a/drivers/net/wireless/ray_cs.c +++ b/drivers/net/wireless/ray_cs.c @@ -982,7 +982,9 @@ AP to AP 1 1 dest AP src AP dest source if (local->net_type == ADHOC) { writeb(0, &ptx->mac.frame_ctl_2); memcpy_toio(ptx->mac.addr_1, ((struct ethhdr *)data)->h_dest, - 2 * ADDRLEN); + ADDRLEN); + memcpy_toio(ptx->mac.addr_2, ((struct ethhdr *)data)->h_source, + ADDRLEN); memcpy_toio(ptx->mac.addr_3, local->bss_id, ADDRLEN); } else { /* infrastructure */
Kernel code has a regular need to describe groups of members within a structure usually when they need to be copied or initialized separately from the rest of the surrounding structure. The generally accepted design pattern in C is to use a named sub-struct:
struct foo { int one; struct { int two; int three, four; } thing; int five; };
This would allow for traditional references and sizing:
memcpy(&dst.thing, &src.thing, sizeof(dst.thing));
However, doing this would mean that referencing struct members enclosed by such named structs would always require including the sub-struct name in identifiers:
do_something(dst.thing.three);
This has tended to be quite inflexible, especially when such groupings need to be added to established code which causes huge naming churn. Three workarounds exist in the kernel for this problem, and each have other negative properties.
To avoid the naming churn, there is a design pattern of adding macro aliases for the named struct:
#define f_three thing.three
This ends up polluting the global namespace, and makes it difficult to search for identifiers.
Another common work-around in kernel code avoids the pollution by avoiding the named struct entirely, instead identifying the group's boundaries using either a pair of empty anonymous structs of a pair of zero-element arrays:
struct foo { int one; struct { } start; int two; int three, four; struct { } finish; int five; };
struct foo { int one; int start[0]; int two; int three, four; int finish[0]; int five; };
This allows code to avoid needing to use a sub-struct named for member references within the surrounding structure, but loses the benefits of being able to actually use such a struct, making it rather fragile. Using these requires open-coded calculation of sizes and offsets. The efforts made to avoid common mistakes include lots of comments, or adding various BUILD_BUG_ON()s. Such code is left with no way for the compiler to reason about the boundaries (e.g. the "start" object looks like it's 0 bytes in length), making bounds checking depend on open-coded calculations:
if (length > offsetof(struct foo, finish) - offsetof(struct foo, start)) return -EINVAL; memcpy(&dst.start, &src.start, offsetof(struct foo, finish) - offsetof(struct foo, start));
However, the vast majority of places in the kernel that operate on groups of members do so without any identification of the grouping, relying either on comments or implicit knowledge of the struct contents, which is even harder for the compiler to reason about, and results in even more fragile manual sizing, usually depending on member locations outside of the region (e.g. to copy "two" and "three", use the start of "four" to find the size):
BUILD_BUG_ON((offsetof(struct foo, four) < offsetof(struct foo, two)) || (offsetof(struct foo, four) < offsetof(struct foo, three)); if (length > offsetof(struct foo, four) - offsetof(struct foo, two)) return -EINVAL; memcpy(&dst.two, &src.two, length);
In order to have a regular programmatic way to describe a struct region that can be used for references and sizing, can be examined for bounds checking, avoids forcing the use of intermediate identifiers, and avoids polluting the global namespace, introduce the struct_group() macro. This macro wraps the member declarations to create an anonymous union of an anonymous struct (no intermediate name) and a named struct (for references and sizing):
struct foo { int one; struct_group(thing, int two; int three, four; ); int five; };
if (length > sizeof(src.thing)) return -EINVAL; memcpy(&dst.thing, &src.thing, length); do_something(dst.three);
There are some rare cases where the resulting struct_group() needs attributes added, so struct_group_attr() is also introduced to allow for specifying struct attributes (e.g. __align(x) or __packed). Additionally, there are places where such declarations would like to have the struct be typed, so struct_group_typed() is added.
Given there is a need for a handful of UAPI uses too, the underlying __struct_group() macro has been defined in UAPI so it can be used there too.
Co-developed-by: Keith Packard keithp@keithp.com Signed-off-by: Keith Packard keithp@keithp.com Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Gustavo A. R. Silva gustavoars@kernel.org Link: https://lore.kernel.org/lkml/20210728023217.GC35706@embeddedor Enhanced-by: Rasmus Villemoes linux@rasmusvillemoes.dk Link: https://lore.kernel.org/lkml/41183a98-bdb9-4ad6-7eab-5a7292a6df84@rasmusvill... Enhanced-by: Dan Williams dan.j.williams@intel.com Link: https://lore.kernel.org/lkml/1d9a2e6df2a9a35b2cdd50a9a68cac5991e7e5f0.camel@... Enhanced-by: Daniel Vetter daniel.vetter@ffwll.ch Link: https://lore.kernel.org/lkml/YQKa76A6XuFqgM03@phenom.ffwll.local --- include/linux/stddef.h | 47 +++++++++++++++++++++++++++++++++++++ include/uapi/linux/stddef.h | 21 +++++++++++++++++ 2 files changed, 68 insertions(+)
diff --git a/include/linux/stddef.h b/include/linux/stddef.h index 998a4ba28eba..f2aefdb22d1d 100644 --- a/include/linux/stddef.h +++ b/include/linux/stddef.h @@ -36,4 +36,51 @@ enum { #define offsetofend(TYPE, MEMBER) \ (offsetof(TYPE, MEMBER) + sizeof_field(TYPE, MEMBER))
+/** + * struct_group(NAME, MEMBERS) + * + * Used to create an anonymous union of two structs with identical + * layout and size: one anonymous and one named. The former can be + * used normally without sub-struct naming, and the latter can be + * used to reason about the start, end, and size of the group of + * struct members. + * + * @NAME: The identifier name of the mirrored sub-struct + * @MEMBERS: The member declarations for the mirrored structs + */ +#define struct_group(NAME, MEMBERS...) \ + __struct_group(/* no tag */, NAME, /* no attrs */, MEMBERS) + +/** + * struct_group_attr(NAME, ATTRS, MEMBERS) + * + * Used to create an anonymous union of two structs with identical + * layout and size: one anonymous and one named. The former can be + * used normally without sub-struct naming, and the latter can be + * used to reason about the start, end, and size of the group of + * struct members. Includes structure attributes argument. + * + * @NAME: The identifier name of the mirrored sub-struct + * @ATTRS: Any struct attributes + * @MEMBERS: The member declarations for the mirrored structs + */ +#define struct_group_attr(NAME, ATTRS, MEMBERS...) \ + __struct_group(/* no tag */, NAME, ATTRS, MEMBERS) + +/** + * struct_group_tagged(TAG, NAME, MEMBERS) + * + * Used to create an anonymous union of two structs with identical + * layout and size: one anonymous and one named. The former can be + * used normally without sub-struct naming, and the latter can be + * used to reason about the start, end, and size of the group of + * struct members. Includes struct tag argument for the named copy. + * + * @TAG: The tag name for the named sub-struct + * @NAME: The identifier name of the mirrored sub-struct + * @MEMBERS: The member declarations for the mirrored structs + */ +#define struct_group_tagged(TAG, NAME, MEMBERS...) \ + __struct_group(TAG, NAME, /* no attrs */, MEMBERS) + #endif diff --git a/include/uapi/linux/stddef.h b/include/uapi/linux/stddef.h index ee8220f8dcf5..0fbdf2f711aa 100644 --- a/include/uapi/linux/stddef.h +++ b/include/uapi/linux/stddef.h @@ -4,3 +4,24 @@ #ifndef __always_inline #define __always_inline inline #endif + +/** + * __struct_group(TAG, NAME, ATTRS, MEMBERS) + * + * Used to create an anonymous union of two structs with identical layout + * and size: one anonymous and one named. The former's members can be used + * normally without sub-struct naming, and the latter can be used to + * reason about the start, end, and size of the group of struct members. + * The named struct can also be explicitly tagged, as well as both having + * struct attributes. + * + * @TAG: The tag name for the named sub-struct (usually empty) + * @NAME: The identifier name of the mirrored sub-struct + * @ATTRS: Any struct attributes (usually empty) + * @MEMBERS: The member declarations for the mirrored structs + */ +#define __struct_group(TAG, NAME, ATTRS, MEMBERS...) \ + union { \ + struct { MEMBERS } ATTRS; \ + struct TAG { MEMBERS } ATTRS NAME; \ + }
On Tue, Aug 17, 2021 at 11:06 PM Kees Cook keescook@chromium.org wrote:
Kernel code has a regular need to describe groups of members within a structure usually when they need to be copied or initialized separately from the rest of the surrounding structure. The generally accepted design pattern in C is to use a named sub-struct:
struct foo { int one; struct { int two; int three, four; } thing; int five; };
This would allow for traditional references and sizing:
memcpy(&dst.thing, &src.thing, sizeof(dst.thing));
However, doing this would mean that referencing struct members enclosed by such named structs would always require including the sub-struct name in identifiers:
do_something(dst.thing.three);
This has tended to be quite inflexible, especially when such groupings need to be added to established code which causes huge naming churn. Three workarounds exist in the kernel for this problem, and each have other negative properties.
To avoid the naming churn, there is a design pattern of adding macro aliases for the named struct:
#define f_three thing.three
This ends up polluting the global namespace, and makes it difficult to search for identifiers.
Another common work-around in kernel code avoids the pollution by avoiding the named struct entirely, instead identifying the group's boundaries using either a pair of empty anonymous structs of a pair of zero-element arrays:
struct foo { int one; struct { } start; int two; int three, four; struct { } finish; int five; }; struct foo { int one; int start[0]; int two; int three, four; int finish[0]; int five; };
This allows code to avoid needing to use a sub-struct named for member references within the surrounding structure, but loses the benefits of being able to actually use such a struct, making it rather fragile. Using these requires open-coded calculation of sizes and offsets. The efforts made to avoid common mistakes include lots of comments, or adding various BUILD_BUG_ON()s. Such code is left with no way for the compiler to reason about the boundaries (e.g. the "start" object looks like it's 0 bytes in length), making bounds checking depend on open-coded calculations:
if (length > offsetof(struct foo, finish) - offsetof(struct foo, start)) return -EINVAL; memcpy(&dst.start, &src.start, offsetof(struct foo, finish) - offsetof(struct foo, start));
However, the vast majority of places in the kernel that operate on groups of members do so without any identification of the grouping, relying either on comments or implicit knowledge of the struct contents, which is even harder for the compiler to reason about, and results in even more fragile manual sizing, usually depending on member locations outside of the region (e.g. to copy "two" and "three", use the start of "four" to find the size):
BUILD_BUG_ON((offsetof(struct foo, four) < offsetof(struct foo, two)) || (offsetof(struct foo, four) < offsetof(struct foo, three)); if (length > offsetof(struct foo, four) - offsetof(struct foo, two)) return -EINVAL; memcpy(&dst.two, &src.two, length);
In order to have a regular programmatic way to describe a struct region that can be used for references and sizing, can be examined for bounds checking, avoids forcing the use of intermediate identifiers, and avoids polluting the global namespace, introduce the struct_group() macro. This macro wraps the member declarations to create an anonymous union of an anonymous struct (no intermediate name) and a named struct (for references and sizing):
struct foo { int one; struct_group(thing, int two; int three, four; ); int five; }; if (length > sizeof(src.thing)) return -EINVAL; memcpy(&dst.thing, &src.thing, length); do_something(dst.three);
There are some rare cases where the resulting struct_group() needs attributes added, so struct_group_attr() is also introduced to allow for specifying struct attributes (e.g. __align(x) or __packed). Additionally, there are places where such declarations would like to have the struct be typed, so struct_group_typed() is added.
Given there is a need for a handful of UAPI uses too, the underlying __struct_group() macro has been defined in UAPI so it can be used there too.
Co-developed-by: Keith Packard keithp@keithp.com Signed-off-by: Keith Packard keithp@keithp.com Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Gustavo A. R. Silva gustavoars@kernel.org Link: https://lore.kernel.org/lkml/20210728023217.GC35706@embeddedor Enhanced-by: Rasmus Villemoes linux@rasmusvillemoes.dk Link: https://lore.kernel.org/lkml/41183a98-bdb9-4ad6-7eab-5a7292a6df84@rasmusvill... Enhanced-by: Dan Williams dan.j.williams@intel.com
Acked-by: Dan Williams dan.j.williams@intel.com
Use the newly introduced struct_group_typed() macro to clean up the declaration of struct cxl_regs.
Cc: Alison Schofield alison.schofield@intel.com Cc: Vishal Verma vishal.l.verma@intel.com Cc: Ira Weiny ira.weiny@intel.com Cc: Ben Widawsky ben.widawsky@intel.com Cc: linux-cxl@vger.kernel.org Suggested-by: Dan Williams dan.j.williams@intel.com Link: https://lore.kernel.org/lkml/1d9a2e6df2a9a35b2cdd50a9a68cac5991e7e5f0.camel@... Signed-off-by: Kees Cook keescook@chromium.org --- drivers/cxl/cxl.h | 61 ++++++++++++++--------------------------------- 1 file changed, 18 insertions(+), 43 deletions(-)
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h index 53927f9fa77e..9db0c402c9ce 100644 --- a/drivers/cxl/cxl.h +++ b/drivers/cxl/cxl.h @@ -75,52 +75,27 @@ static inline int cxl_hdm_decoder_count(u32 cap_hdr) #define CXLDEV_MBOX_BG_CMD_STATUS_OFFSET 0x18 #define CXLDEV_MBOX_PAYLOAD_OFFSET 0x20
-#define CXL_COMPONENT_REGS() \ - void __iomem *hdm_decoder - -#define CXL_DEVICE_REGS() \ - void __iomem *status; \ - void __iomem *mbox; \ - void __iomem *memdev - -/* See note for 'struct cxl_regs' for the rationale of this organization */ -/* - * CXL_COMPONENT_REGS - Common set of CXL Component register block base pointers - * @hdm_decoder: CXL 2.0 8.2.5.12 CXL HDM Decoder Capability Structure - */ -struct cxl_component_regs { - CXL_COMPONENT_REGS(); -}; - -/* See note for 'struct cxl_regs' for the rationale of this organization */ -/* - * CXL_DEVICE_REGS - Common set of CXL Device register block base pointers - * @status: CXL 2.0 8.2.8.3 Device Status Registers - * @mbox: CXL 2.0 8.2.8.4 Mailbox Registers - * @memdev: CXL 2.0 8.2.8.5 Memory Device Registers - */ -struct cxl_device_regs { - CXL_DEVICE_REGS(); -}; - /* - * Note, the anonymous union organization allows for per - * register-block-type helper routines, without requiring block-type - * agnostic code to include the prefix. + * Using struct_group() allows for per register-block-type helper routines, + * without requiring block-type agnostic code to include the prefix. */ struct cxl_regs { - union { - struct { - CXL_COMPONENT_REGS(); - }; - struct cxl_component_regs component; - }; - union { - struct { - CXL_DEVICE_REGS(); - }; - struct cxl_device_regs device_regs; - }; + /* + * Common set of CXL Component register block base pointers + * @hdm_decoder: CXL 2.0 8.2.5.12 CXL HDM Decoder Capability Structure + */ + struct_group_tagged(cxl_component_regs, component, + void __iomem *hdm_decoder; + ); + /* + * Common set of CXL Device register block base pointers + * @status: CXL 2.0 8.2.8.3 Device Status Registers + * @mbox: CXL 2.0 8.2.8.4 Mailbox Registers + * @memdev: CXL 2.0 8.2.8.5 Memory Device Registers + */ + struct_group_tagged(cxl_device_regs, device_regs, + void __iomem *status, *mbox, *memdev; + ); };
struct cxl_reg_map {
On Tue, Aug 17, 2021 at 11:06 PM Kees Cook keescook@chromium.org wrote:
Use the newly introduced struct_group_typed() macro to clean up the declaration of struct cxl_regs.
Cc: Alison Schofield alison.schofield@intel.com Cc: Vishal Verma vishal.l.verma@intel.com Cc: Ira Weiny ira.weiny@intel.com Cc: Ben Widawsky ben.widawsky@intel.com Cc: linux-cxl@vger.kernel.org Suggested-by: Dan Williams dan.j.williams@intel.com
Looks good and tests ok here:
Reviewed-by: Dan Williams dan.j.williams@intel.com
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Replace the existing empty member position markers "headers_start" and "headers_end" with a struct_group(). This will allow memcpy() and sizeof() to more easily reason about sizes, and improve readability.
"pahole" shows no size nor member offset changes to struct sk_buff. "objdump -d" shows no object code changes (outside of WARNs affected by source line number changes).
Cc: "Jason A. Donenfeld" Jason@zx2c4.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Jonathan Lemon jonathan.lemon@gmail.com Cc: Alexander Lobakin alobakin@pm.me Cc: Jakub Sitnicki jakub@cloudflare.com Cc: Marco Elver elver@google.com Cc: Willem de Bruijn willemb@google.com Cc: wireguard@lists.zx2c4.com Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Gustavo A. R. Silva gustavoars@kernel.org Link: https://lore.kernel.org/lkml/20210728035006.GD35706@embeddedor --- drivers/net/wireguard/queueing.h | 4 +--- include/linux/skbuff.h | 9 ++++----- net/core/skbuff.c | 14 +++++--------- 3 files changed, 10 insertions(+), 17 deletions(-)
diff --git a/drivers/net/wireguard/queueing.h b/drivers/net/wireguard/queueing.h index 4ef2944a68bc..52da5e963003 100644 --- a/drivers/net/wireguard/queueing.h +++ b/drivers/net/wireguard/queueing.h @@ -79,9 +79,7 @@ static inline void wg_reset_packet(struct sk_buff *skb, bool encapsulating) u8 sw_hash = skb->sw_hash; u32 hash = skb->hash; skb_scrub_packet(skb, true); - memset(&skb->headers_start, 0, - offsetof(struct sk_buff, headers_end) - - offsetof(struct sk_buff, headers_start)); + memset(&skb->headers, 0, sizeof(skb->headers)); if (encapsulating) { skb->l4_hash = l4_hash; skb->sw_hash = sw_hash; diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 6bdb0db3e825..fee9041aa402 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -801,11 +801,10 @@ struct sk_buff { __u8 active_extensions; #endif
- /* fields enclosed in headers_start/headers_end are copied + /* Fields enclosed in headers group are copied * using a single memcpy() in __copy_skb_header() */ - /* private: */ - __u32 headers_start[0]; + struct_group(headers, /* public: */
/* if you move pkt_type around you also must adapt those constants */ @@ -922,8 +921,8 @@ struct sk_buff { u64 kcov_handle; #endif
- /* private: */ - __u32 headers_end[0]; + ); /* end headers group */ + /* public: */
/* These elements must be at the end, see alloc_skb() for details. */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index f9311762cc47..fd5ce57ccce6 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -991,12 +991,10 @@ void napi_consume_skb(struct sk_buff *skb, int budget) } EXPORT_SYMBOL(napi_consume_skb);
-/* Make sure a field is enclosed inside headers_start/headers_end section */ +/* Make sure a field is contained by headers group */ #define CHECK_SKB_FIELD(field) \ - BUILD_BUG_ON(offsetof(struct sk_buff, field) < \ - offsetof(struct sk_buff, headers_start)); \ - BUILD_BUG_ON(offsetof(struct sk_buff, field) > \ - offsetof(struct sk_buff, headers_end)); \ + BUILD_BUG_ON(offsetof(struct sk_buff, field) != \ + offsetof(struct sk_buff, headers.field)); \
static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old) { @@ -1008,14 +1006,12 @@ static void __copy_skb_header(struct sk_buff *new, const struct sk_buff *old) __skb_ext_copy(new, old); __nf_copy(new, old, false);
- /* Note : this field could be in headers_start/headers_end section + /* Note : this field could be in the headers group. * It is not yet because we do not want to have a 16 bit hole */ new->queue_mapping = old->queue_mapping;
- memcpy(&new->headers_start, &old->headers_start, - offsetof(struct sk_buff, headers_end) - - offsetof(struct sk_buff, headers_start)); + memcpy(&new->headers, &old->headers, sizeof(new->headers)); CHECK_SKB_FIELD(protocol); CHECK_SKB_FIELD(csum); CHECK_SKB_FIELD(hash);
For the drivers/net/wireguard part,
Reviewed-by: Jason A. Donenfeld Jason@zx2c4.com
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() around members queue_id, min_bw, max_bw, tsa, pri_lvl, and bw_weight so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of queue_id.
"pahole" shows no size nor member offset changes to struct bnxt_cos2bw_cfg. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences and optimizations).
Cc: Michael Chan michael.chan@broadcom.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Michael Chan michael.chan@broadcom.com Link: https://lore.kernel.org/lkml/CACKFLinDc6Y+P8eZ=450yA1nMC7swTURLtcdyiNR=9J6df... Reviewed-by: Gustavo A. R. Silva gustavoars@kernel.org Link: https://lore.kernel.org/lkml/20210728044517.GE35706@embeddedor --- drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c | 4 ++-- drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h | 14 ++++++++------ 2 files changed, 10 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c index 8a68df4d9e59..95c636f89329 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.c @@ -148,10 +148,10 @@ static int bnxt_hwrm_queue_cos2bw_qcfg(struct bnxt *bp, struct ieee_ets *ets) }
data = &resp->queue_id0 + offsetof(struct bnxt_cos2bw_cfg, queue_id); - for (i = 0; i < bp->max_tc; i++, data += sizeof(cos2bw) - 4) { + for (i = 0; i < bp->max_tc; i++, data += sizeof(cos2bw.cfg)) { int tc;
- memcpy(&cos2bw.queue_id, data, sizeof(cos2bw) - 4); + memcpy(&cos2bw.cfg, data, sizeof(cos2bw.cfg)); if (i == 0) cos2bw.queue_id = resp->queue_id0;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h index 6eed231de565..716742522161 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_dcb.h @@ -23,13 +23,15 @@ struct bnxt_dcb {
struct bnxt_cos2bw_cfg { u8 pad[3]; - u8 queue_id; - __le32 min_bw; - __le32 max_bw; + struct_group_attr(cfg, __packed, + u8 queue_id; + __le32 min_bw; + __le32 max_bw; #define BW_VALUE_UNIT_PERCENT1_100 (0x1UL << 29) - u8 tsa; - u8 pri_lvl; - u8 bw_weight; + u8 tsa; + u8 pri_lvl; + u8 bw_weight; + ); u8 unused; };
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct mwl8k_cmd_set_key around members key_material, tkip_tx_mic_key, and tkip_rx_mic_key so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of key_material.
"pahole" shows no size nor member offset changes to struct mwl8k_cmd_set_key. "objdump -d" shows no object code changes.
Cc: Lennert Buytenhek buytenh@wantstofly.org Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: wengjianfeng wengjianfeng@yulong.com Cc: Lv Yunlong lyl2019@mail.ustc.edu.cn Cc: Arnd Bergmann arnd@arndb.de Cc: Christophe JAILLET christophe.jaillet@wanadoo.fr Cc: Allen Pais allen.lkml@gmail.com Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/marvell/mwl8k.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/net/wireless/marvell/mwl8k.c b/drivers/net/wireless/marvell/mwl8k.c index 3bf6571f4149..a29277d5f9da 100644 --- a/drivers/net/wireless/marvell/mwl8k.c +++ b/drivers/net/wireless/marvell/mwl8k.c @@ -4225,9 +4225,11 @@ struct mwl8k_cmd_set_key { __le32 key_info; __le32 key_id; __le16 key_len; - __u8 key_material[MAX_ENCR_KEY_LENGTH]; - __u8 tkip_tx_mic_key[MIC_KEY_LENGTH]; - __u8 tkip_rx_mic_key[MIC_KEY_LENGTH]; + struct { + __u8 key_material[MAX_ENCR_KEY_LENGTH]; + __u8 tkip_tx_mic_key[MIC_KEY_LENGTH]; + __u8 tkip_rx_mic_key[MIC_KEY_LENGTH]; + } tkip; __le16 tkip_rsc_low; __le32 tkip_rsc_high; __le16 tkip_tsc_low; @@ -4375,7 +4377,7 @@ static int mwl8k_cmd_encryption_set_key(struct ieee80211_hw *hw, goto done; }
- memcpy(cmd->key_material, key->key, keymlen); + memcpy(&cmd->tkip, key->key, keymlen); cmd->action = cpu_to_le32(action);
rc = mwl8k_post_pervif_cmd(hw, vif, &cmd->header);
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct txpd around members tx_dest_addr_high and tx_dest_addr_low so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of queue_id.
"pahole" shows no size nor member offset changes to struct txpd. "objdump -d" shows no object code changes.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: libertas-dev@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/marvell/libertas/host.h | 10 ++++++---- drivers/net/wireless/marvell/libertas/tx.c | 5 +++-- 2 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/drivers/net/wireless/marvell/libertas/host.h b/drivers/net/wireless/marvell/libertas/host.h index dfa22468b14a..af96bdba3b2b 100644 --- a/drivers/net/wireless/marvell/libertas/host.h +++ b/drivers/net/wireless/marvell/libertas/host.h @@ -308,10 +308,12 @@ struct txpd { __le32 tx_packet_location; /* Tx packet length */ __le16 tx_packet_length; - /* First 2 byte of destination MAC address */ - u8 tx_dest_addr_high[2]; - /* Last 4 byte of destination MAC address */ - u8 tx_dest_addr_low[4]; + struct_group(tx_dest_addr, + /* First 2 byte of destination MAC address */ + u8 tx_dest_addr_high[2]; + /* Last 4 byte of destination MAC address */ + u8 tx_dest_addr_low[4]; + ); /* Pkt Priority */ u8 priority; /* Pkt Trasnit Power control */ diff --git a/drivers/net/wireless/marvell/libertas/tx.c b/drivers/net/wireless/marvell/libertas/tx.c index aeb481740df6..27304a98787d 100644 --- a/drivers/net/wireless/marvell/libertas/tx.c +++ b/drivers/net/wireless/marvell/libertas/tx.c @@ -113,6 +113,7 @@ netdev_tx_t lbs_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) p802x_hdr = skb->data; pkt_len = skb->len;
+ BUILD_BUG_ON(sizeof(txpd->tx_dest_addr) != ETH_ALEN); if (priv->wdev->iftype == NL80211_IFTYPE_MONITOR) { struct tx_radiotap_hdr *rtap_hdr = (void *)skb->data;
@@ -124,10 +125,10 @@ netdev_tx_t lbs_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) pkt_len -= sizeof(*rtap_hdr);
/* copy destination address from 802.11 header */ - memcpy(txpd->tx_dest_addr_high, p802x_hdr + 4, ETH_ALEN); + memcpy(&txpd->tx_dest_addr, p802x_hdr + 4, ETH_ALEN); } else { /* copy destination address from 802.3 header */ - memcpy(txpd->tx_dest_addr_high, p802x_hdr, ETH_ALEN); + memcpy(&txpd->tx_dest_addr, p802x_hdr, ETH_ALEN); }
txpd->tx_packet_length = cpu_to_le16(pkt_len);
In preparation for FORTIFY_SOURCE performing compile-time and run-time field array bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct txpd around members tx_dest_addr_high and tx_dest_addr_low so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of tx_dest_addr_high.
"pahole" shows no size nor member offset changes to struct txpd. "objdump -d" shows no object code changes.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Lee Jones lee.jones@linaro.org Cc: YueHaibing yuehaibing@huawei.com Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/marvell/libertas_tf/libertas_tf.h | 10 ++++++---- drivers/net/wireless/marvell/libertas_tf/main.c | 3 ++- 2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/drivers/net/wireless/marvell/libertas_tf/libertas_tf.h b/drivers/net/wireless/marvell/libertas_tf/libertas_tf.h index 5d726545d987..b2af2ddb6bc4 100644 --- a/drivers/net/wireless/marvell/libertas_tf/libertas_tf.h +++ b/drivers/net/wireless/marvell/libertas_tf/libertas_tf.h @@ -268,10 +268,12 @@ struct txpd { __le32 tx_packet_location; /* Tx packet length */ __le16 tx_packet_length; - /* First 2 byte of destination MAC address */ - u8 tx_dest_addr_high[2]; - /* Last 4 byte of destination MAC address */ - u8 tx_dest_addr_low[4]; + struct_group(tx_dest_addr, + /* First 2 byte of destination MAC address */ + u8 tx_dest_addr_high[2]; + /* Last 4 byte of destination MAC address */ + u8 tx_dest_addr_low[4]; + ); /* Pkt Priority */ u8 priority; /* Pkt Trasnit Power control */ diff --git a/drivers/net/wireless/marvell/libertas_tf/main.c b/drivers/net/wireless/marvell/libertas_tf/main.c index 71492211904b..02a1e1f547d8 100644 --- a/drivers/net/wireless/marvell/libertas_tf/main.c +++ b/drivers/net/wireless/marvell/libertas_tf/main.c @@ -232,7 +232,8 @@ static void lbtf_tx_work(struct work_struct *work) ieee80211_get_tx_rate(priv->hw, info)->hw_value);
/* copy destination address from 802.11 header */ - memcpy(txpd->tx_dest_addr_high, skb->data + sizeof(struct txpd) + 4, + BUILD_BUG_ON(sizeof(txpd->tx_dest_addr) != ETH_ALEN); + memcpy(&txpd->tx_dest_addr, skb->data + sizeof(struct txpd) + 4, ETH_ALEN); txpd->tx_packet_length = cpu_to_le16(len); txpd->tx_packet_location = cpu_to_le32(sizeof(struct txpd));
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct art around members weight, and ac[0-9]_max, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of weight.
"pahole" shows no size nor member offset changes to struct art. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences).
Cc: Zhang Rui rui.zhang@intel.com Cc: Daniel Lezcano daniel.lezcano@linaro.org Cc: Amit Kucheria amitk@kernel.org Cc: linux-pm@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- .../intel/int340x_thermal/acpi_thermal_rel.c | 5 +- .../intel/int340x_thermal/acpi_thermal_rel.h | 48 ++++++++++--------- 2 files changed, 29 insertions(+), 24 deletions(-)
diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c index a478cff8162a..e90690a234c4 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c @@ -250,8 +250,9 @@ static int fill_art(char __user *ubuf) get_single_name(arts[i].source, art_user[i].source_device); get_single_name(arts[i].target, art_user[i].target_device); /* copy the rest int data in addition to source and target */ - memcpy(&art_user[i].weight, &arts[i].weight, - sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2)); + BUILD_BUG_ON(sizeof(art_user[i].data) != + sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2)); + memcpy(&art_user[i].data, &arts[i].data, sizeof(art_user[i].data)); }
if (copy_to_user(ubuf, art_user, art_len)) diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h index 58822575fd54..78d942477035 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h @@ -17,17 +17,19 @@ struct art { acpi_handle source; acpi_handle target; - u64 weight; - u64 ac0_max; - u64 ac1_max; - u64 ac2_max; - u64 ac3_max; - u64 ac4_max; - u64 ac5_max; - u64 ac6_max; - u64 ac7_max; - u64 ac8_max; - u64 ac9_max; + struct_group(data, + u64 weight; + u64 ac0_max; + u64 ac1_max; + u64 ac2_max; + u64 ac3_max; + u64 ac4_max; + u64 ac5_max; + u64 ac6_max; + u64 ac7_max; + u64 ac8_max; + u64 ac9_max; + ); } __packed;
struct trt { @@ -47,17 +49,19 @@ union art_object { struct { char source_device[8]; /* ACPI single name */ char target_device[8]; /* ACPI single name */ - u64 weight; - u64 ac0_max_level; - u64 ac1_max_level; - u64 ac2_max_level; - u64 ac3_max_level; - u64 ac4_max_level; - u64 ac5_max_level; - u64 ac6_max_level; - u64 ac7_max_level; - u64 ac8_max_level; - u64 ac9_max_level; + struct_group(data, + u64 weight; + u64 ac0_max_level; + u64 ac1_max_level; + u64 ac2_max_level; + u64 ac3_max_level; + u64 ac4_max_level; + u64 ac5_max_level; + u64 ac6_max_level; + u64 ac7_max_level; + u64 ac8_max_level; + u64 ac9_max_level; + ); }; u64 __data[ACPI_NR_ART_ELEMENTS]; };
On Wed, Aug 18, 2021 at 8:08 AM Kees Cook keescook@chromium.org wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct art around members weight, and ac[0-9]_max, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of weight.
"pahole" shows no size nor member offset changes to struct art. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences).
Cc: Zhang Rui rui.zhang@intel.com Cc: Daniel Lezcano daniel.lezcano@linaro.org Cc: Amit Kucheria amitk@kernel.org Cc: linux-pm@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Rui, Srinivas, any comments here?
.../intel/int340x_thermal/acpi_thermal_rel.c | 5 +- .../intel/int340x_thermal/acpi_thermal_rel.h | 48 ++++++++++--------- 2 files changed, 29 insertions(+), 24 deletions(-)
diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c index a478cff8162a..e90690a234c4 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c @@ -250,8 +250,9 @@ static int fill_art(char __user *ubuf) get_single_name(arts[i].source, art_user[i].source_device); get_single_name(arts[i].target, art_user[i].target_device); /* copy the rest int data in addition to source and target */
memcpy(&art_user[i].weight, &arts[i].weight,
sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2));
BUILD_BUG_ON(sizeof(art_user[i].data) !=
sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2));
memcpy(&art_user[i].data, &arts[i].data, sizeof(art_user[i].data)); } if (copy_to_user(ubuf, art_user, art_len))
diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h index 58822575fd54..78d942477035 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h @@ -17,17 +17,19 @@ struct art { acpi_handle source; acpi_handle target;
u64 weight;
u64 ac0_max;
u64 ac1_max;
u64 ac2_max;
u64 ac3_max;
u64 ac4_max;
u64 ac5_max;
u64 ac6_max;
u64 ac7_max;
u64 ac8_max;
u64 ac9_max;
struct_group(data,
u64 weight;
u64 ac0_max;
u64 ac1_max;
u64 ac2_max;
u64 ac3_max;
u64 ac4_max;
u64 ac5_max;
u64 ac6_max;
u64 ac7_max;
u64 ac8_max;
u64 ac9_max;
);
} __packed;
struct trt { @@ -47,17 +49,19 @@ union art_object { struct { char source_device[8]; /* ACPI single name */ char target_device[8]; /* ACPI single name */
u64 weight;
u64 ac0_max_level;
u64 ac1_max_level;
u64 ac2_max_level;
u64 ac3_max_level;
u64 ac4_max_level;
u64 ac5_max_level;
u64 ac6_max_level;
u64 ac7_max_level;
u64 ac8_max_level;
u64 ac9_max_level;
struct_group(data,
u64 weight;
u64 ac0_max_level;
u64 ac1_max_level;
u64 ac2_max_level;
u64 ac3_max_level;
u64 ac4_max_level;
u64 ac5_max_level;
u64 ac6_max_level;
u64 ac7_max_level;
u64 ac8_max_level;
u64 ac9_max_level;
); }; u64 __data[ACPI_NR_ART_ELEMENTS];
};
2.30.2
On Tue, 2021-11-23 at 14:19 +0100, Rafael J. Wysocki wrote:
On Wed, Aug 18, 2021 at 8:08 AM Kees Cook keescook@chromium.org wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run- time field bounds checking for memcpy(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct art around members weight, and ac[0- 9]_max, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of weight.
"pahole" shows no size nor member offset changes to struct art. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences).
Cc: Zhang Rui rui.zhang@intel.com Cc: Daniel Lezcano daniel.lezcano@linaro.org Cc: Amit Kucheria amitk@kernel.org Cc: linux-pm@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Rui, Srinivas, any comments here?
Looks good. Reviewed-by: Srinivas Pandruvada srinivas.pandruvada@linux.intel.com
Thanks, Srinivas
.../intel/int340x_thermal/acpi_thermal_rel.c | 5 +- .../intel/int340x_thermal/acpi_thermal_rel.h | 48 ++++++++++------- -- 2 files changed, 29 insertions(+), 24 deletions(-)
diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c index a478cff8162a..e90690a234c4 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.c @@ -250,8 +250,9 @@ static int fill_art(char __user *ubuf) get_single_name(arts[i].source, art_user[i].source_device); get_single_name(arts[i].target, art_user[i].target_device); /* copy the rest int data in addition to source and target */ - memcpy(&art_user[i].weight, &arts[i].weight, - sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2)); + BUILD_BUG_ON(sizeof(art_user[i].data) != + sizeof(u64) * (ACPI_NR_ART_ELEMENTS - 2)); + memcpy(&art_user[i].data, &arts[i].data, sizeof(art_user[i].data)); }
if (copy_to_user(ubuf, art_user, art_len)) diff --git a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h index 58822575fd54..78d942477035 100644 --- a/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h +++ b/drivers/thermal/intel/int340x_thermal/acpi_thermal_rel.h @@ -17,17 +17,19 @@ struct art { acpi_handle source; acpi_handle target; - u64 weight; - u64 ac0_max; - u64 ac1_max; - u64 ac2_max; - u64 ac3_max; - u64 ac4_max; - u64 ac5_max; - u64 ac6_max; - u64 ac7_max; - u64 ac8_max; - u64 ac9_max; + struct_group(data, + u64 weight; + u64 ac0_max; + u64 ac1_max; + u64 ac2_max; + u64 ac3_max; + u64 ac4_max; + u64 ac5_max; + u64 ac6_max; + u64 ac7_max; + u64 ac8_max; + u64 ac9_max; + ); } __packed;
struct trt { @@ -47,17 +49,19 @@ union art_object { struct { char source_device[8]; /* ACPI single name */ char target_device[8]; /* ACPI single name */ - u64 weight; - u64 ac0_max_level; - u64 ac1_max_level; - u64 ac2_max_level; - u64 ac3_max_level; - u64 ac4_max_level; - u64 ac5_max_level; - u64 ac6_max_level; - u64 ac7_max_level; - u64 ac8_max_level; - u64 ac9_max_level; + struct_group(data, + u64 weight; + u64 ac0_max_level; + u64 ac1_max_level; + u64 ac2_max_level; + u64 ac3_max_level; + u64 ac4_max_level; + u64 ac5_max_level; + u64 ac6_max_level; + u64 ac7_max_level; + u64 ac8_max_level; + u64 ac9_max_level; + ); }; u64 __data[ACPI_NR_ART_ELEMENTS]; }; -- 2.30.2
On Wed, Nov 24, 2021 at 12:53 AM Srinivas Pandruvada srinivas.pandruvada@linux.intel.com wrote:
On Tue, 2021-11-23 at 14:19 +0100, Rafael J. Wysocki wrote:
On Wed, Aug 18, 2021 at 8:08 AM Kees Cook keescook@chromium.org wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run- time field bounds checking for memcpy(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct art around members weight, and ac[0- 9]_max, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of weight.
"pahole" shows no size nor member offset changes to struct art. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences).
Cc: Zhang Rui rui.zhang@intel.com Cc: Daniel Lezcano daniel.lezcano@linaro.org Cc: Amit Kucheria amitk@kernel.org Cc: linux-pm@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Rui, Srinivas, any comments here?
Looks good. Reviewed-by: Srinivas Pandruvada srinivas.pandruvada@linux.intel.com
Applied as 5.17 material, thank you!
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct ivhd_entry around members ext and hidh, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of ext.
"pahole" shows no size nor member offset changes to struct ivhd_entry. "objdump -d" shows no object code changes.
Cc: Joerg Roedel joro@8bytes.org Cc: Will Deacon will@kernel.org Cc: iommu@lists.linux-foundation.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/iommu/amd/init.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c index bdcf167b4afe..70506d6175e9 100644 --- a/drivers/iommu/amd/init.c +++ b/drivers/iommu/amd/init.c @@ -121,8 +121,10 @@ struct ivhd_entry { u8 type; u16 devid; u8 flags; - u32 ext; - u32 hidh; + struct_group(ext_hid, + u32 ext; + u32 hidh; + ); u64 cid; u8 uidf; u8 uidl; @@ -1377,7 +1379,8 @@ static int __init init_iommu_from_acpi(struct amd_iommu *iommu, break; }
- memcpy(hid, (u8 *)(&e->ext), ACPIHID_HID_LEN - 1); + BUILD_BUG_ON(sizeof(e->ext_hid) != ACPIHID_HID_LEN - 1); + memcpy(hid, &e->ext_hid, ACPIHID_HID_LEN - 1); hid[ACPIHID_HID_LEN - 1] = '\0';
if (!(*hid)) {
On Tue, Aug 17, 2021 at 11:04:43PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct ivhd_entry around members ext and hidh, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of ext.
"pahole" shows no size nor member offset changes to struct ivhd_entry. "objdump -d" shows no object code changes.
Cc: Joerg Roedel joro@8bytes.org Cc: Will Deacon will@kernel.org Cc: iommu@lists.linux-foundation.org Signed-off-by: Kees Cook keescook@chromium.org
Acked-by: Joerg Roedel jroedel@suse.de
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct rss_hdr around members imm_data and intr_gen, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of imm_data.
"pahole" shows no size nor member offset changes to struct rss_hdr. "objdump -d" shows no object code changes.
Cc: Raju Rangoju rajur@chelsio.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/chelsio/cxgb3/sge.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb3/sge.c b/drivers/net/ethernet/chelsio/cxgb3/sge.c index cb5c79c43bc9..1ab1bd86a3a6 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb3/sge.c @@ -126,8 +126,10 @@ struct rsp_desc { /* response queue descriptor */ struct rss_header rss_hdr; __be32 flags; __be32 len_cq; - u8 imm_data[47]; - u8 intr_gen; + struct_group(immediate, + u8 imm_data[47]; + u8 intr_gen; + ); };
/* @@ -929,7 +931,8 @@ static inline struct sk_buff *get_imm_packet(const struct rsp_desc *resp)
if (skb) { __skb_put(skb, IMMED_PKT_SIZE); - skb_copy_to_linear_data(skb, resp->imm_data, IMMED_PKT_SIZE); + BUILD_BUG_ON(IMMED_PKT_SIZE != sizeof(resp->immediate)); + skb_copy_to_linear_data(skb, &resp->immediate, IMMED_PKT_SIZE); } return skb; }
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct hfa384x_tx_frame around members frame_control, duration_id, addr1, addr2, addr3, and seq_ctrl, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of frame_control.
"pahole" shows no size nor member offset changes to struct hfa384x_tx_frame. "objdump -d" shows no object code changes.
Cc: Jouni Malinen j@w1.fi Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Lee Jones lee.jones@linaro.org Cc: Allen Pais allen.lkml@gmail.com Cc: Romain Perier romain.perier@gmail.com Cc: Arnd Bergmann arnd@arndb.de Cc: Vaibhav Gupta vaibhavgupta40@gmail.com Cc: Cong Wang xiyou.wangcong@gmail.com Cc: Sebastian Andrzej Siewior bigeasy@linutronix.de Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/intersil/hostap/hostap_hw.c | 5 +++-- drivers/net/wireless/intersil/hostap/hostap_wlan.h | 14 ++++++++------ 2 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/drivers/net/wireless/intersil/hostap/hostap_hw.c b/drivers/net/wireless/intersil/hostap/hostap_hw.c index 9a19046217df..cea8a9ddc4da 100644 --- a/drivers/net/wireless/intersil/hostap/hostap_hw.c +++ b/drivers/net/wireless/intersil/hostap/hostap_hw.c @@ -1812,8 +1812,9 @@ static int prism2_tx_80211(struct sk_buff *skb, struct net_device *dev) memset(&txdesc, 0, sizeof(txdesc));
/* skb->data starts with txdesc->frame_control */ - hdr_len = 24; - skb_copy_from_linear_data(skb, &txdesc.frame_control, hdr_len); + hdr_len = sizeof(txdesc.frame); + BUILD_BUG_ON(hdr_len != 24); + skb_copy_from_linear_data(skb, &txdesc.frame, hdr_len); if (ieee80211_is_data(txdesc.frame_control) && ieee80211_has_a4(txdesc.frame_control) && skb->len >= 30) { diff --git a/drivers/net/wireless/intersil/hostap/hostap_wlan.h b/drivers/net/wireless/intersil/hostap/hostap_wlan.h index dd2603d9b5d3..174735a137c5 100644 --- a/drivers/net/wireless/intersil/hostap/hostap_wlan.h +++ b/drivers/net/wireless/intersil/hostap/hostap_wlan.h @@ -115,12 +115,14 @@ struct hfa384x_tx_frame { __le16 tx_control; /* HFA384X_TX_CTRL_ flags */
/* 802.11 */ - __le16 frame_control; /* parts not used */ - __le16 duration_id; - u8 addr1[ETH_ALEN]; - u8 addr2[ETH_ALEN]; /* filled by firmware */ - u8 addr3[ETH_ALEN]; - __le16 seq_ctrl; /* filled by firmware */ + struct_group(frame, + __le16 frame_control; /* parts not used */ + __le16 duration_id; + u8 addr1[ETH_ALEN]; + u8 addr2[ETH_ALEN]; /* filled by firmware */ + u8 addr3[ETH_ALEN]; + __le16 seq_ctrl; /* filled by firmware */ + ); u8 addr4[ETH_ALEN]; __le16 data_len;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct fw_eth_tx_pkt_vm_wr around members ethmacdst, ethmacsrc, ethtype, and vlantci, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of ethmacdst.
"pahole" shows no size nor member offset changes to struct fw_eth_tx_pkt_vm_wr. "objdump -d" shows no object code changes.
Cc: Raju Rangoju rajur@chelsio.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/chelsio/cxgb4/sge.c | 8 +++++--- drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h | 10 ++++++---- drivers/net/ethernet/chelsio/cxgb4vf/sge.c | 7 ++----- 3 files changed, 13 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 6a099cb34b12..9080b2c5ffe8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -1842,8 +1842,10 @@ static netdev_tx_t cxgb4_vf_eth_xmit(struct sk_buff *skb, * (including the VLAN tag) into the header so we reject anything * smaller than that ... */ - fw_hdr_copy_len = sizeof(wr->ethmacdst) + sizeof(wr->ethmacsrc) + - sizeof(wr->ethtype) + sizeof(wr->vlantci); + BUILD_BUG_ON(sizeof(wr->firmware) != + (sizeof(wr->ethmacdst) + sizeof(wr->ethmacsrc) + + sizeof(wr->ethtype) + sizeof(wr->vlantci))); + fw_hdr_copy_len = sizeof(wr->firmware); ret = cxgb4_validate_skb(skb, dev, fw_hdr_copy_len); if (ret) goto out_free; @@ -1924,7 +1926,7 @@ static netdev_tx_t cxgb4_vf_eth_xmit(struct sk_buff *skb, wr->equiq_to_len16 = cpu_to_be32(wr_mid); wr->r3[0] = cpu_to_be32(0); wr->r3[1] = cpu_to_be32(0); - skb_copy_from_linear_data(skb, (void *)wr->ethmacdst, fw_hdr_copy_len); + skb_copy_from_linear_data(skb, &wr->firmware, fw_hdr_copy_len); end = (u64 *)wr + flits;
/* If this is a Large Send Offload packet we'll put in an LSO CPL diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index 0a326c054707..2419459a0b85 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -794,10 +794,12 @@ struct fw_eth_tx_pkt_vm_wr { __be32 op_immdlen; __be32 equiq_to_len16; __be32 r3[2]; - u8 ethmacdst[6]; - u8 ethmacsrc[6]; - __be16 ethtype; - __be16 vlantci; + struct_group(firmware, + u8 ethmacdst[ETH_ALEN]; + u8 ethmacsrc[ETH_ALEN]; + __be16 ethtype; + __be16 vlantci; + ); };
#define FW_CMD_MAX_TIMEOUT 10000 diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c index 7bc80eeb2c21..671ca93e64ab 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/sge.c @@ -1167,10 +1167,7 @@ netdev_tx_t t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) struct cpl_tx_pkt_core *cpl; const struct skb_shared_info *ssi; dma_addr_t addr[MAX_SKB_FRAGS + 1]; - const size_t fw_hdr_copy_len = (sizeof(wr->ethmacdst) + - sizeof(wr->ethmacsrc) + - sizeof(wr->ethtype) + - sizeof(wr->vlantci)); + const size_t fw_hdr_copy_len = sizeof(wr->firmware);
/* * The chip minimum packet length is 10 octets but the firmware @@ -1267,7 +1264,7 @@ netdev_tx_t t4vf_eth_xmit(struct sk_buff *skb, struct net_device *dev) wr->equiq_to_len16 = cpu_to_be32(wr_mid); wr->r3[0] = cpu_to_be32(0); wr->r3[1] = cpu_to_be32(0); - skb_copy_from_linear_data(skb, (void *)wr->ethmacdst, fw_hdr_copy_len); + skb_copy_from_linear_data(skb, &wr->firmware, fw_hdr_copy_len); end = (u64 *)wr + flits;
/*
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct nig_stats around members egress_mac_pkt0_lo, egress_mac_pkt0_hi, egress_mac_pkt1_lo, and egress_mac_pkt1_hi (and the respective members in struct bnx2x_eth_stats), so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of struct bnx2x_eth_stats's rx_stat_ifhcinbadoctets_hi.
"pahole" shows no size nor member offset changes to either struct. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences and optimizations).
Additionally adds BUILD_BUG_ON() to compare the separate struct group sizes.
Cc: Ariel Elior aelior@marvell.com Cc: Sudarsana Kalluru skalluru@marvell.com Cc: GR-everest-linux-l2@marvell.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c | 7 ++++--- drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h | 14 ++++++++++---- 2 files changed, 14 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c index 0b193edb73b8..2bb133ae61c3 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c @@ -849,7 +849,8 @@ static int bnx2x_hw_stats_update(struct bnx2x *bp)
memcpy(old, new, sizeof(struct nig_stats));
- memcpy(&(estats->rx_stat_ifhcinbadoctets_hi), &(pstats->mac_stx[1]), + BUILD_BUG_ON(sizeof(estats->shared) != sizeof(pstats->mac_stx[1])); + memcpy(&(estats->shared), &(pstats->mac_stx[1]), sizeof(struct mac_stx)); estats->brb_drop_hi = pstats->brb_drop_hi; estats->brb_drop_lo = pstats->brb_drop_lo; @@ -1634,9 +1635,9 @@ void bnx2x_stats_init(struct bnx2x *bp) REG_RD(bp, NIG_REG_STAT0_BRB_TRUNCATE + port*0x38); if (!CHIP_IS_E3(bp)) { REG_RD_DMAE(bp, NIG_REG_STAT0_EGRESS_MAC_PKT0 + port*0x50, - &(bp->port.old_nig_stats.egress_mac_pkt0_lo), 2); + &(bp->port.old_nig_stats.egress_mac_pkt0), 2); REG_RD_DMAE(bp, NIG_REG_STAT0_EGRESS_MAC_PKT1 + port*0x50, - &(bp->port.old_nig_stats.egress_mac_pkt1_lo), 2); + &(bp->port.old_nig_stats.egress_mac_pkt1), 2); }
/* Prepare statistics ramrod data */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h index d55e63692cf3..ae93c078707b 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.h @@ -36,10 +36,14 @@ struct nig_stats { u32 pbf_octets; u32 pbf_packet; u32 safc_inp; - u32 egress_mac_pkt0_lo; - u32 egress_mac_pkt0_hi; - u32 egress_mac_pkt1_lo; - u32 egress_mac_pkt1_hi; + struct_group(egress_mac_pkt0, + u32 egress_mac_pkt0_lo; + u32 egress_mac_pkt0_hi; + ); + struct_group(egress_mac_pkt1, + u32 egress_mac_pkt1_lo; + u32 egress_mac_pkt1_hi; + ); };
enum bnx2x_stats_event { @@ -83,6 +87,7 @@ struct bnx2x_eth_stats { u32 no_buff_discard_hi; u32 no_buff_discard_lo;
+ struct_group(shared, u32 rx_stat_ifhcinbadoctets_hi; u32 rx_stat_ifhcinbadoctets_lo; u32 tx_stat_ifhcoutbadoctets_hi; @@ -159,6 +164,7 @@ struct bnx2x_eth_stats { u32 tx_stat_dot3statsinternalmactransmiterrors_lo; u32 tx_stat_bmac_ufl_hi; u32 tx_stat_bmac_ufl_lo; + );
u32 pause_frames_received_hi; u32 pause_frames_received_lo;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in structs: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10 PPTable_t so the grouped members can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of the first member.
"pahole" shows no size nor member offset changes to any structs. "objdump -d" shows no object code changes.
Cc: "Christian König" christian.koenig@amd.com Cc: "Pan, Xinhui" Xinhui.Pan@amd.com Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch Cc: Hawking Zhang Hawking.Zhang@amd.com Cc: Feifei Xu Feifei.Xu@amd.com Cc: Lijo Lazar lijo.lazar@amd.com Cc: Likun Gao Likun.Gao@amd.com Cc: Jiawei Gu Jiawei.Gu@amd.com Cc: Evan Quan evan.quan@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Alex Deucher alexander.deucher@amd.com Link: https://lore.kernel.org/lkml/CADnq5_Npb8uYvd+R4UHgf-w8-cQj3JoODjviJR_Y9w9wqJ... --- drivers/gpu/drm/amd/include/atomfirmware.h | 9 ++++++++- .../gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 ++- drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 ++- .../gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 3 ++- drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 6 +++--- drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 12 ++++++++---- drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 6 +++--- 7 files changed, 28 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index 44955458fe38..7bf3edf15410 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -2081,6 +2081,7 @@ struct atom_smc_dpm_info_v4_5 { struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS + struct_group(dpm_info, // I2C Control struct smudpm_i2c_controller_config_v2 I2cControllers[8];
@@ -2159,7 +2160,7 @@ struct atom_smc_dpm_info_v4_5 uint32_t MvddRatio; // This is used for MVDD Vid workaround. It has 16 fractional bits (Q16.16)
uint32_t BoardReserved[9]; - + ); };
struct atom_smc_dpm_info_v4_6 @@ -2168,6 +2169,7 @@ struct atom_smc_dpm_info_v4_6 // section: board parameters uint32_t i2c_padding[3]; // old i2c control are moved to new area
+ struct_group(dpm_info, uint16_t maxvoltagestepgfx; // in mv(q2) max voltage step that smu will request. multiple steps are taken if voltage change exceeds this value. uint16_t maxvoltagestepsoc; // in mv(q2) max voltage step that smu will request. multiple steps are taken if voltage change exceeds this value.
@@ -2246,12 +2248,14 @@ struct atom_smc_dpm_info_v4_6
// reserved uint32_t boardreserved[10]; + ); };
struct atom_smc_dpm_info_v4_7 { struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS + struct_group(dpm_info, // I2C Control struct smudpm_i2c_controller_config_v2 I2cControllers[8];
@@ -2348,6 +2352,7 @@ struct atom_smc_dpm_info_v4_7 uint8_t Padding8_Psi2;
uint32_t BoardReserved[5]; + ); };
struct smudpm_i2c_controller_config_v3 @@ -2478,6 +2483,7 @@ struct atom_smc_dpm_info_v4_10 struct atom_common_table_header table_header;
// SECTION: BOARD PARAMETERS + struct_group(dpm_info, // Telemetry Settings uint16_t GfxMaxCurrent; // in Amps uint8_t GfxOffset; // in Amps @@ -2524,6 +2530,7 @@ struct atom_smc_dpm_info_v4_10 uint16_t spare5;
uint32_t reserved[16]; + ); };
/* diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h index 43d43d6addc0..8093a98800c3 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h @@ -643,6 +643,7 @@ typedef struct { // SECTION: BOARD PARAMETERS
// SVI2 Board Parameters + struct_group(v4_6, uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value.
@@ -728,10 +729,10 @@ typedef struct { uint32_t BoardVoltageCoeffB; // decode by /1000
uint32_t BoardReserved[7]; + );
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use - } PPTable_t;
typedef struct { diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h index 04752ade1016..0b4e6e907e95 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h @@ -725,6 +725,7 @@ typedef struct { uint32_t Reserved[8];
// SECTION: BOARD PARAMETERS + struct_group(v4, // I2C Control I2cControllerConfig_t I2cControllers[NUM_I2C_CONTROLLERS];
@@ -809,10 +810,10 @@ typedef struct { uint8_t Padding8_Loadline;
uint32_t BoardReserved[8]; + );
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use - } PPTable_t;
typedef struct { diff --git a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h b/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h index a017983ff1fa..5056d3728da8 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h +++ b/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h @@ -390,6 +390,7 @@ typedef struct { uint32_t spare3[14];
// SECTION: BOARD PARAMETERS + struct_group(v4_10, // Telemetry Settings uint16_t GfxMaxCurrent; // in Amps int8_t GfxOffset; // in Amps @@ -444,7 +445,7 @@ typedef struct {
//reserved uint32_t reserved[14]; - + ); } PPTable_t;
typedef struct { diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 8ab58781ae13..341adf209240 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -463,11 +463,11 @@ static int arcturus_append_powerplay_table(struct smu_context *smu) smc_dpm_table->table_header.format_revision, smc_dpm_table->table_header.content_revision);
+ BUILD_BUG_ON(sizeof(smc_pptable->v4_6) != sizeof(smc_dpm_table->dpm_info)); if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 6)) - memcpy(&smc_pptable->MaxVoltageStepGfx, - &smc_dpm_table->maxvoltagestepgfx, - sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_6, maxvoltagestepgfx)); + memcpy(&smc_pptable->v4_6, &smc_dpm_table->dpm_info, + sizeof(smc_dpm_table->dpm_info));
return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 2e5d3669652b..e8b6e25a7815 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -431,16 +431,20 @@ static int navi10_append_powerplay_table(struct smu_context *smu)
switch (smc_dpm_table->table_header.content_revision) { case 5: /* nv10 and nv14 */ - memcpy(smc_pptable->I2cControllers, smc_dpm_table->I2cControllers, - sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header)); + BUILD_BUG_ON(sizeof(smc_pptable->v4) != + sizeof(smc_dpm_table->dpm_info)); + memcpy(&smc_pptable->v4, &smc_dpm_table->dpm_info, + sizeof(smc_dpm_table->dpm_info)); break; case 7: /* nv12 */ ret = amdgpu_atombios_get_data_table(adev, index, NULL, NULL, NULL, (uint8_t **)&smc_dpm_table_v4_7); if (ret) return ret; - memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, - sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header)); + BUILD_BUG_ON(sizeof(smc_pptable->v4) != + sizeof(smc_dpm_table_v4_7->dpm_info)); + memcpy(&smc_pptable->v4, &smc_dpm_table_v4_7->dpm_info, + sizeof(smc_dpm_table_v4_7->dpm_info)); break; default: dev_err(smu->adev->dev, "smc_dpm_info with unsupported content revision %d!\n", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index c8eefacfdd37..492ba37bc514 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -407,11 +407,11 @@ static int aldebaran_append_powerplay_table(struct smu_context *smu) smc_dpm_table->table_header.format_revision, smc_dpm_table->table_header.content_revision);
+ BUILD_BUG_ON(sizeof(smc_pptable->v4_10) != sizeof(smc_dpm_table->dpm_info)); if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 10)) - memcpy(&smc_pptable->GfxMaxCurrent, - &smc_dpm_table->GfxMaxCurrent, - sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_10, GfxMaxCurrent)); + memcpy(&smc_pptable->v4_10, &smc_dpm_table->dpm_info, + sizeof(smc_dpm_table->dpm_info)); return 0; }
On 8/18/2021 11:34 AM, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in structs: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10 PPTable_t so the grouped members can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of the first member.
"pahole" shows no size nor member offset changes to any structs. "objdump -d" shows no object code changes.
Cc: "Christian König" christian.koenig@amd.com Cc: "Pan, Xinhui" Xinhui.Pan@amd.com Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch Cc: Hawking Zhang Hawking.Zhang@amd.com Cc: Feifei Xu Feifei.Xu@amd.com Cc: Lijo Lazar lijo.lazar@amd.com Cc: Likun Gao Likun.Gao@amd.com Cc: Jiawei Gu Jiawei.Gu@amd.com Cc: Evan Quan evan.quan@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Alex Deucher alexander.deucher@amd.com Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kerne...
drivers/gpu/drm/amd/include/atomfirmware.h | 9 ++++++++- .../gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 ++- drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 ++- .../gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 3 ++-
Hi Kees,
The headers which define these structs are firmware/VBIOS interfaces and are picked directly from those components. There are difficulties in grouping them to structs at the original source as that involves other component changes.
The driver_if_* files updates are frequent and it is error prone to manually group them each time we pick them for any update. Our usage of memcpy in this way is restricted only to a very few places.
As another option - is it possible to have a helper function/macro like memcpy_fortify() which takes the extra arguments and does the extra compile time checks? We will use the helper whenever we have such kind of usage.
Thanks, Lijo
drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 6 +++--- drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c | 12 ++++++++---- drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 6 +++--- 7 files changed, 28 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index 44955458fe38..7bf3edf15410 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -2081,6 +2081,7 @@ struct atom_smc_dpm_info_v4_5 { struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS
- struct_group(dpm_info, // I2C Control struct smudpm_i2c_controller_config_v2 I2cControllers[8];
@@ -2159,7 +2160,7 @@ struct atom_smc_dpm_info_v4_5 uint32_t MvddRatio; // This is used for MVDD Vid workaround. It has 16 fractional bits (Q16.16)
uint32_t BoardReserved[9];
); };
struct atom_smc_dpm_info_v4_6
@@ -2168,6 +2169,7 @@ struct atom_smc_dpm_info_v4_6 // section: board parameters uint32_t i2c_padding[3]; // old i2c control are moved to new area
- struct_group(dpm_info, uint16_t maxvoltagestepgfx; // in mv(q2) max voltage step that smu will request. multiple steps are taken if voltage change exceeds this value. uint16_t maxvoltagestepsoc; // in mv(q2) max voltage step that smu will request. multiple steps are taken if voltage change exceeds this value.
@@ -2246,12 +2248,14 @@ struct atom_smc_dpm_info_v4_6
// reserved uint32_t boardreserved[10];
); };
struct atom_smc_dpm_info_v4_7 { struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS
struct_group(dpm_info, // I2C Control struct smudpm_i2c_controller_config_v2 I2cControllers[8];
@@ -2348,6 +2352,7 @@ struct atom_smc_dpm_info_v4_7 uint8_t Padding8_Psi2;
uint32_t BoardReserved[5];
); };
struct smudpm_i2c_controller_config_v3
@@ -2478,6 +2483,7 @@ struct atom_smc_dpm_info_v4_10 struct atom_common_table_header table_header;
// SECTION: BOARD PARAMETERS
- struct_group(dpm_info, // Telemetry Settings uint16_t GfxMaxCurrent; // in Amps uint8_t GfxOffset; // in Amps
@@ -2524,6 +2530,7 @@ struct atom_smc_dpm_info_v4_10 uint16_t spare5;
uint32_t reserved[16];
); };
/*
diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h index 43d43d6addc0..8093a98800c3 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h @@ -643,6 +643,7 @@ typedef struct { // SECTION: BOARD PARAMETERS
// SVI2 Board Parameters
- struct_group(v4_6, uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value.
@@ -728,10 +729,10 @@ typedef struct { uint32_t BoardVoltageCoeffB; // decode by /1000
uint32_t BoardReserved[7];
);
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use
} PPTable_t;
typedef struct {
diff --git a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h index 04752ade1016..0b4e6e907e95 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h @@ -725,6 +725,7 @@ typedef struct { uint32_t Reserved[8];
// SECTION: BOARD PARAMETERS
- struct_group(v4, // I2C Control I2cControllerConfig_t I2cControllers[NUM_I2C_CONTROLLERS];
@@ -809,10 +810,10 @@ typedef struct { uint8_t Padding8_Loadline;
uint32_t BoardReserved[8];
);
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use
} PPTable_t;
typedef struct {
diff --git a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h b/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h index a017983ff1fa..5056d3728da8 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h +++ b/drivers/gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h @@ -390,6 +390,7 @@ typedef struct { uint32_t spare3[14];
// SECTION: BOARD PARAMETERS
- struct_group(v4_10, // Telemetry Settings uint16_t GfxMaxCurrent; // in Amps int8_t GfxOffset; // in Amps
@@ -444,7 +445,7 @@ typedef struct {
//reserved uint32_t reserved[14];
); } PPTable_t;
typedef struct {
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 8ab58781ae13..341adf209240 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -463,11 +463,11 @@ static int arcturus_append_powerplay_table(struct smu_context *smu) smc_dpm_table->table_header.format_revision, smc_dpm_table->table_header.content_revision);
- BUILD_BUG_ON(sizeof(smc_pptable->v4_6) != sizeof(smc_dpm_table->dpm_info)); if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 6))
memcpy(&smc_pptable->MaxVoltageStepGfx,
&smc_dpm_table->maxvoltagestepgfx,
sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_6, maxvoltagestepgfx));
memcpy(&smc_pptable->v4_6, &smc_dpm_table->dpm_info,
sizeof(smc_dpm_table->dpm_info));
return 0; }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 2e5d3669652b..e8b6e25a7815 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -431,16 +431,20 @@ static int navi10_append_powerplay_table(struct smu_context *smu)
switch (smc_dpm_table->table_header.content_revision) { case 5: /* nv10 and nv14 */
memcpy(smc_pptable->I2cControllers, smc_dpm_table->I2cControllers,
sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header));
BUILD_BUG_ON(sizeof(smc_pptable->v4) !=
sizeof(smc_dpm_table->dpm_info));
memcpy(&smc_pptable->v4, &smc_dpm_table->dpm_info,
break; case 7: /* nv12 */ ret = amdgpu_atombios_get_data_table(adev, index, NULL, NULL, NULL, (uint8_t **)&smc_dpm_table_v4_7); if (ret) return ret;sizeof(smc_dpm_table->dpm_info));
memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers,
sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header));
BUILD_BUG_ON(sizeof(smc_pptable->v4) !=
sizeof(smc_dpm_table_v4_7->dpm_info));
memcpy(&smc_pptable->v4, &smc_dpm_table_v4_7->dpm_info,
break; default: dev_err(smu->adev->dev, "smc_dpm_info with unsupported content revision %d!\n",sizeof(smc_dpm_table_v4_7->dpm_info));
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index c8eefacfdd37..492ba37bc514 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -407,11 +407,11 @@ static int aldebaran_append_powerplay_table(struct smu_context *smu) smc_dpm_table->table_header.format_revision, smc_dpm_table->table_header.content_revision);
- BUILD_BUG_ON(sizeof(smc_pptable->v4_10) != sizeof(smc_dpm_table->dpm_info)); if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 10))
memcpy(&smc_pptable->GfxMaxCurrent,
&smc_dpm_table->GfxMaxCurrent,
sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_10, GfxMaxCurrent));
memcpy(&smc_pptable->v4_10, &smc_dpm_table->dpm_info,
return 0; }sizeof(smc_dpm_table->dpm_info));
On Wed, Aug 18, 2021 at 05:12:28PM +0530, Lazar, Lijo wrote:
On 8/18/2021 11:34 AM, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in structs: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10 PPTable_t so the grouped members can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of the first member.
"pahole" shows no size nor member offset changes to any structs. "objdump -d" shows no object code changes.
Cc: "Christian König" christian.koenig@amd.com Cc: "Pan, Xinhui" Xinhui.Pan@amd.com Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch Cc: Hawking Zhang Hawking.Zhang@amd.com Cc: Feifei Xu Feifei.Xu@amd.com Cc: Lijo Lazar lijo.lazar@amd.com Cc: Likun Gao Likun.Gao@amd.com Cc: Jiawei Gu Jiawei.Gu@amd.com Cc: Evan Quan evan.quan@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Alex Deucher alexander.deucher@amd.com Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kerne...
drivers/gpu/drm/amd/include/atomfirmware.h | 9 ++++++++- .../gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 ++- drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 ++- .../gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 3 ++-
Hi Kees,
Hi! Thanks for looking into this.
The headers which define these structs are firmware/VBIOS interfaces and are picked directly from those components. There are difficulties in grouping them to structs at the original source as that involves other component changes.
So, can you help me understand this a bit more? It sounds like these are generated headers, yes? I'd like to understand your constraints and weight them against various benefits that could be achieved here.
The groupings I made do appear to be roughly documented already, for example:
struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS + struct_group(dpm_info,
Something emitted the "BOARD PARAMETERS" section heading as a comment, so it likely also would know where it ends, yes? The good news here is that for the dpm_info groups, they all end at the end of the existing structs, see: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10
The matching regions in the PPTable_t structs are similarly marked with a "BOARD PARAMETERS" section heading comment:
--- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h @@ -643,6 +643,7 @@ typedef struct { // SECTION: BOARD PARAMETERS
// SVI2 Board Parameters + struct_group(v4_6, uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value.
@@ -728,10 +729,10 @@ typedef struct { uint32_t BoardVoltageCoeffB; // decode by /1000
uint32_t BoardReserved[7]; + );
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use - } PPTable_t;
Where they end seems known as well (the padding switches from a "Board" to "MmHub" prefix at exactly the matching size).
So, given that these regions are already known by the export tool, how about just updating the export tool to emit a struct there? I imagine the problem with this would be the identifier churn needed, but that's entirely mechanical.
However, I'm curious about another aspect of these regions: they are, by definition, the same. Why isn't there a single struct describing them already, given the existing redundancy? For example, look at the member names: maxvoltagestepgfx vs MaxVoltageStepGfx. Why aren't these the same? And then why aren't they described separately?
Fixing that would cut down on the redundancy here, and in the renaming, you can fix the identifiers as well. It should be straight forward to write a Coccinelle script to do this renaming for you after extracting the structure.
The driver_if_* files updates are frequent and it is error prone to manually group them each time we pick them for any update.
Why are these structs updated? It looks like they're specifically versioned, and aren't expected to change (i.e. v4.5, v4.6, v4.10, etc).
Our usage of memcpy in this way is restricted only to a very few places.
True, there's 1 per PPTable_t duplication. With a proper struct, you wouldn't even need a memcpy().
Instead of the existing: memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header));
or my proposed: memcpy(&smc_pptable->v4, &smc_dpm_table_v4_7->dpm_info, sizeof(smc_dpm_table_v4_7->dpm_info));
you could just have: smc_pptable->v4 = smc_dpm_table_v4_7->dpm_info;
since they'd be explicitly the same type.
That looks like a much cleaner solution to this. It greatly improves readability, reduces the redundancy in the headers, and should be a simple mechanical refactoring.
Oh my, I just noticed append_vbios_pptable() in drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.c which does an open-coded assignment of the entire PPTable_t, including padding, and, apparently, the i2c address twice:
ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
As another option - is it possible to have a helper function/macro like memcpy_fortify() which takes the extra arguments and does the extra compile time checks? We will use the helper whenever we have such kind of usage.
I'd rather avoid special cases just for this, especially when the code here is already doing a couple things we try to avoid in the rest of the kernel (i.e. open coded redundant struct contents, etc).
If something mechanically produced append_vbios_pptable() above, I bet we can get rid of the memcpy()s entirely and save a lot of code doing a member-to-member assignment.
What do you think?
-Kees
On 8/19/2021 5:29 AM, Kees Cook wrote:
On Wed, Aug 18, 2021 at 05:12:28PM +0530, Lazar, Lijo wrote:
On 8/18/2021 11:34 AM, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in structs: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10 PPTable_t so the grouped members can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of the first member.
"pahole" shows no size nor member offset changes to any structs. "objdump -d" shows no object code changes.
Cc: "Christian König" christian.koenig@amd.com Cc: "Pan, Xinhui" Xinhui.Pan@amd.com Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch Cc: Hawking Zhang Hawking.Zhang@amd.com Cc: Feifei Xu Feifei.Xu@amd.com Cc: Lijo Lazar lijo.lazar@amd.com Cc: Likun Gao Likun.Gao@amd.com Cc: Jiawei Gu Jiawei.Gu@amd.com Cc: Evan Quan evan.quan@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Alex Deucher alexander.deucher@amd.com Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kerne...
drivers/gpu/drm/amd/include/atomfirmware.h | 9 ++++++++- .../gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 ++- drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 ++- .../gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 3 ++-
Hi Kees,
Hi! Thanks for looking into this.
The headers which define these structs are firmware/VBIOS interfaces and are picked directly from those components. There are difficulties in grouping them to structs at the original source as that involves other component changes.
So, can you help me understand this a bit more? It sounds like these are generated headers, yes? I'd like to understand your constraints and weight them against various benefits that could be achieved here.
The groupings I made do appear to be roughly documented already, for example:
struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS
- struct_group(dpm_info,
Something emitted the "BOARD PARAMETERS" section heading as a comment, so it likely also would know where it ends, yes? The good news here is that for the dpm_info groups, they all end at the end of the existing structs, see: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10
The matching regions in the PPTable_t structs are similarly marked with a "BOARD PARAMETERS" section heading comment:
--- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h @@ -643,6 +643,7 @@ typedef struct { // SECTION: BOARD PARAMETERS
// SVI2 Board Parameters
- struct_group(v4_6, uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value.
@@ -728,10 +729,10 @@ typedef struct { uint32_t BoardVoltageCoeffB; // decode by /1000
uint32_t BoardReserved[7];
);
// Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use
- } PPTable_t;
Where they end seems known as well (the padding switches from a "Board" to "MmHub" prefix at exactly the matching size).
So, given that these regions are already known by the export tool, how about just updating the export tool to emit a struct there? I imagine the problem with this would be the identifier churn needed, but that's entirely mechanical.
However, I'm curious about another aspect of these regions: they are, by definition, the same. Why isn't there a single struct describing them already, given the existing redundancy? For example, look at the member names: maxvoltagestepgfx vs MaxVoltageStepGfx. Why aren't these the same? And then why aren't they described separately?
Fixing that would cut down on the redundancy here, and in the renaming, you can fix the identifiers as well. It should be straight forward to write a Coccinelle script to do this renaming for you after extracting the structure.
The driver_if_* files updates are frequent and it is error prone to manually group them each time we pick them for any update.
Why are these structs updated? It looks like they're specifically versioned, and aren't expected to change (i.e. v4.5, v4.6, v4.10, etc).
Our usage of memcpy in this way is restricted only to a very few places.
True, there's 1 per PPTable_t duplication. With a proper struct, you wouldn't even need a memcpy().
Instead of the existing: memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header));
or my proposed: memcpy(&smc_pptable->v4, &smc_dpm_table_v4_7->dpm_info, sizeof(smc_dpm_table_v4_7->dpm_info));
you could just have: smc_pptable->v4 = smc_dpm_table_v4_7->dpm_info;
since they'd be explicitly the same type.
That looks like a much cleaner solution to this. It greatly improves readability, reduces the redundancy in the headers, and should be a simple mechanical refactoring.
Oh my, I just noticed append_vbios_pptable() in drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.c which does an open-coded assignment of the entire PPTable_t, including padding, and, apparently, the i2c address twice:
ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address; ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
As another option - is it possible to have a helper function/macro like memcpy_fortify() which takes the extra arguments and does the extra compile time checks? We will use the helper whenever we have such kind of usage.
I'd rather avoid special cases just for this, especially when the code here is already doing a couple things we try to avoid in the rest of the kernel (i.e. open coded redundant struct contents, etc).
If something mechanically produced append_vbios_pptable() above, I bet we can get rid of the memcpy()s entirely and save a lot of code doing a member-to-member assignment.
What do you think?
Hi Kees,
Will give a background on why there are multiple headers and why it's structured this way. That may help to better understand this arrangement.
This code is part of driver for AMD GPUs. These GPUs get to the consumers through multiple channels - AMD designs a few boards with those, there are add-in-board partners like ASRock, Sapphire etc. who take these ASICs and design their own boards, and others like OEM vendors who have their own design for boards in their laptops.
As you have noticed, this particular section in the structure carries information categorized as 'BOARD PARAMETERS'. Since there are multiple vendors designing their own boards, this gives the option to customize the parameters based on their board design.
There are a few components in AMD GPUs which are interested in these board parameters main ones being - Video BIOS (VBIOS) and power management firmware (PMFW). There needs to be a single source where a vendor can input the information and that is decided as VBIOS. VBIOS carries different data tables which carry other information also (some of which are used by driver), so this information is added as a separate data table in VBIOS. A board vendor can customize the VBIOS build with this information.
The data tables (and some other interfaces with driver) carried by VBIOS are published in this header - drivers/gpu/drm/amd/include/atomfirmware.h
There are multiple families of AMD GPUs like Navi10, Arcturus, Aldebaran etc. and the board specific details change with different families of GPUs. However, VBIOS team publishes a common header file for these GPUs and any difference in data tables (between GPU families) is maintained through a versioning scheme. Thus there are different tables like atom_smc_dpm_info_v4_5, atom_smc_dpm_info_v4_6 etc. which are relevant for a particular family of GPUs.
With newer VBIOS versions and new GPU families, there could be changes in the structs defined in atomfirmware.h and we pick the header accordingly. As mentioned earlier, one other user of the board specific information is power management firmware (PMFW). PMFW design is isolated from the actual source of board information. In addition to board specific information, PMFW needs some other info as well and driver is the one responsible for passing this info to the firmware. PMFW gives an interface header to driver providing the different struct formats which are used in driver<->PMFW interactions. Unlike VBIOS, these interface headers are defined per family of ASICs and those are smu11_driver_if_arcturus.h, smu11_driver_if_* etc. (in short driver_if_* files). Like VBIOS, with newer firmware versions, there could be changes in the different structs defined in these headers and we pick them accordingly.
Driver acts the intermediary between actual source of board information (VBIOS) and PMFW. So what is being done here is driver picks the board information from VBIOS table, strips the VBIOS table header and passes it as part of PPTable_t which defines all the information that is needed by PMFW from driver for enabling dynamic power management.
In summary, these headers are not generated and not owned by driver. They define the interfaces of two different components with driver, and are consumed by those components themselves. A simple change to group the information as a separate structure involves changes in multiple components like VBIOS, PMFW, software used to build VBIOS, Windows driver etc.
In all practical cases, this code is harmless as these structs (in both headers) are well defined for a specific family of GPUs. There is always a reserve field defined with some extra bytes so that the size is not affected if at all new fields need to be added.
The patch now makes us to modify the headers for Linux through script/manually whenever we pick them, and TBH that strips off the coherency with the original source. The other option is field by field copy. Now we use memcpy as a safe bet so that a new field added later taking some reserve space is not missed even if we miss a header update.
Thanks, Lijo
On Thu, Aug 19, 2021 at 10:33:43AM +0530, Lazar, Lijo wrote:
On 8/19/2021 5:29 AM, Kees Cook wrote:
On Wed, Aug 18, 2021 at 05:12:28PM +0530, Lazar, Lijo wrote:
On 8/18/2021 11:34 AM, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in structs: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10 PPTable_t so the grouped members can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of the first member.
"pahole" shows no size nor member offset changes to any structs. "objdump -d" shows no object code changes.
Cc: "Christian König" christian.koenig@amd.com Cc: "Pan, Xinhui" Xinhui.Pan@amd.com Cc: David Airlie airlied@linux.ie Cc: Daniel Vetter daniel@ffwll.ch Cc: Hawking Zhang Hawking.Zhang@amd.com Cc: Feifei Xu Feifei.Xu@amd.com Cc: Lijo Lazar lijo.lazar@amd.com Cc: Likun Gao Likun.Gao@amd.com Cc: Jiawei Gu Jiawei.Gu@amd.com Cc: Evan Quan evan.quan@amd.com Cc: amd-gfx@lists.freedesktop.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Alex Deucher alexander.deucher@amd.com Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kerne...
drivers/gpu/drm/amd/include/atomfirmware.h | 9 ++++++++- .../gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h | 3 ++- drivers/gpu/drm/amd/pm/inc/smu11_driver_if_navi10.h | 3 ++- .../gpu/drm/amd/pm/inc/smu13_driver_if_aldebaran.h | 3 ++-
Hi Kees,
Hi! Thanks for looking into this.
The headers which define these structs are firmware/VBIOS interfaces and are picked directly from those components. There are difficulties in grouping them to structs at the original source as that involves other component changes.
So, can you help me understand this a bit more? It sounds like these are generated headers, yes? I'd like to understand your constraints and weight them against various benefits that could be achieved here.
The groupings I made do appear to be roughly documented already, for example:
struct atom_common_table_header table_header; // SECTION: BOARD PARAMETERS
- struct_group(dpm_info,
Something emitted the "BOARD PARAMETERS" section heading as a comment, so it likely also would know where it ends, yes? The good news here is that for the dpm_info groups, they all end at the end of the existing structs, see: struct atom_smc_dpm_info_v4_5 struct atom_smc_dpm_info_v4_6 struct atom_smc_dpm_info_v4_7 struct atom_smc_dpm_info_v4_10
The matching regions in the PPTable_t structs are similarly marked with a "BOARD PARAMETERS" section heading comment:
--- a/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h +++ b/drivers/gpu/drm/amd/pm/inc/smu11_driver_if_arcturus.h @@ -643,6 +643,7 @@ typedef struct { // SECTION: BOARD PARAMETERS // SVI2 Board Parameters
- struct_group(v4_6, uint16_t MaxVoltageStepGfx; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value. uint16_t MaxVoltageStepSoc; // In mV(Q2) Max voltage step that SMU will request. Multiple steps are taken if voltage change exceeds this value.
@@ -728,10 +729,10 @@ typedef struct { uint32_t BoardVoltageCoeffB; // decode by /1000 uint32_t BoardReserved[7];
- ); // Padding for MMHUB - do not modify this uint32_t MmHubPadding[8]; // SMU internal use
- } PPTable_t;
Where they end seems known as well (the padding switches from a "Board" to "MmHub" prefix at exactly the matching size).
So, given that these regions are already known by the export tool, how about just updating the export tool to emit a struct there? I imagine the problem with this would be the identifier churn needed, but that's entirely mechanical.
However, I'm curious about another aspect of these regions: they are, by definition, the same. Why isn't there a single struct describing them already, given the existing redundancy? For example, look at the member names: maxvoltagestepgfx vs MaxVoltageStepGfx. Why aren't these the same? And then why aren't they described separately?
Fixing that would cut down on the redundancy here, and in the renaming, you can fix the identifiers as well. It should be straight forward to write a Coccinelle script to do this renaming for you after extracting the structure.
The driver_if_* files updates are frequent and it is error prone to manually group them each time we pick them for any update.
Why are these structs updated? It looks like they're specifically versioned, and aren't expected to change (i.e. v4.5, v4.6, v4.10, etc).
Our usage of memcpy in this way is restricted only to a very few places.
True, there's 1 per PPTable_t duplication. With a proper struct, you wouldn't even need a memcpy().
Instead of the existing: memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header));
or my proposed: memcpy(&smc_pptable->v4, &smc_dpm_table_v4_7->dpm_info, sizeof(smc_dpm_table_v4_7->dpm_info));
you could just have: smc_pptable->v4 = smc_dpm_table_v4_7->dpm_info;
since they'd be explicitly the same type.
That looks like a much cleaner solution to this. It greatly improves readability, reduces the redundancy in the headers, and should be a simple mechanical refactoring.
Oh my, I just noticed append_vbios_pptable() in drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_processpptables.c which does an open-coded assignment of the entire PPTable_t, including padding, and, apparently, the i2c address twice:
ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address; ppsmc_pptable->Vr2_I2C_address = smc_dpm_table.Vr2_I2C_address;
As another option - is it possible to have a helper function/macro like memcpy_fortify() which takes the extra arguments and does the extra compile time checks? We will use the helper whenever we have such kind of usage.
I'd rather avoid special cases just for this, especially when the code here is already doing a couple things we try to avoid in the rest of the kernel (i.e. open coded redundant struct contents, etc).
If something mechanically produced append_vbios_pptable() above, I bet we can get rid of the memcpy()s entirely and save a lot of code doing a member-to-member assignment.
What do you think?
Hi Kees,
Will give a background on why there are multiple headers and why it's structured this way. That may help to better understand this arrangement.
This code is part of driver for AMD GPUs. These GPUs get to the consumers through multiple channels - AMD designs a few boards with those, there are add-in-board partners like ASRock, Sapphire etc. who take these ASICs and design their own boards, and others like OEM vendors who have their own design for boards in their laptops.
As you have noticed, this particular section in the structure carries information categorized as 'BOARD PARAMETERS'. Since there are multiple vendors designing their own boards, this gives the option to customize the parameters based on their board design.
There are a few components in AMD GPUs which are interested in these board parameters main ones being - Video BIOS (VBIOS) and power management firmware (PMFW). There needs to be a single source where a vendor can input the information and that is decided as VBIOS. VBIOS carries different data tables which carry other information also (some of which are used by driver), so this information is added as a separate data table in VBIOS. A board vendor can customize the VBIOS build with this information.
The data tables (and some other interfaces with driver) carried by VBIOS are published in this header - drivers/gpu/drm/amd/include/atomfirmware.h
I understand this to mean that this header is shared by other projects?
If that's true, what compilers are processing this header? (i.e. so I can scope my suggestions to things that all the compilers will be able to deal with.)
How are edits of this file managed "upstream" from the Linux kernel?
Why does it have strange indentations like this:
uint8_t ledpin0; uint8_t ledpin1; uint8_t ledpin2; uint8_t padding8_4;
uint8_t pllgfxclkspreadenabled; uint8_t pllgfxclkspreadpercent; uint16_t pllgfxclkspreadfreq;
uint8_t uclkspreadenabled; uint8_t uclkspreadpercent; uint16_t uclkspreadfreq;
There are multiple families of AMD GPUs like Navi10, Arcturus, Aldebaran etc. and the board specific details change with different families of GPUs. However, VBIOS team publishes a common header file for these GPUs and any difference in data tables (between GPU families) is maintained through a versioning scheme. Thus there are different tables like atom_smc_dpm_info_v4_5, atom_smc_dpm_info_v4_6 etc. which are relevant for a particular family of GPUs.
With newer VBIOS versions and new GPU families, there could be changes in the structs defined in atomfirmware.h and we pick the header accordingly.
As mentioned earlier, one other user of the board specific information is power management firmware (PMFW). PMFW design is isolated from the actual source of board information. In addition to board specific information, PMFW needs some other info as well and driver is the one responsible for passing this info to the firmware. PMFW gives an interface header to driver providing the different struct formats which are used in driver<->PMFW interactions. Unlike VBIOS, these interface headers are defined per family of ASICs and those are smu11_driver_if_arcturus.h, smu11_driver_if_* etc. (in short driver_if_* files). Like VBIOS, with newer firmware versions, there could be changes in the different structs defined in these headers and we pick them accordingly.
Are these headers also shared between other projects?
What's needed to coordinate making these less redundant? (i.e. replacing the "BOARD PARAMETERS" portion of PPTable_t with the associated struct *_dpm_info_v* from atomfirmware.h?)
Driver acts the intermediary between actual source of board information (VBIOS) and PMFW. So what is being done here is driver picks the board information from VBIOS table, strips the VBIOS table header and passes it as part of PPTable_t which defines all the information that is needed by PMFW from driver for enabling dynamic power management.
In summary, these headers are not generated and not owned by driver. They define the interfaces of two different components with driver, and are consumed by those components themselves. A simple change to group the information as a separate structure involves changes in multiple components like VBIOS, PMFW, software used to build VBIOS, Windows driver etc.
In all practical cases, this code is harmless as these structs (in both headers) are well defined for a specific family of GPUs. There is always a reserve field defined with some extra bytes so that the size is not affected if at all new fields need to be added.
It sounds like it's unlikely that the headers will be able to change? If that's true, it seems like a good idea to mark those headers very clearly at the top with details like you describe here. Maybe something like:
/* * This header file is shared between VBIOS, Windows drivers, and Linux * drivers. Any changes need to be well justified and coordinated with * email@address... */
And in looking through these, I notice there's a typo in the Description:
header file of general definitions for OS nd pre-OS video drivers
nd -> and
The patch now makes us to modify the headers for Linux through script/manually whenever we pick them, and TBH that strips off the coherency with the original source. The other option is field by field copy. Now we use memcpy as a safe bet so that a new field added later taking some reserve space is not missed even if we miss a header update.
How does this look as a work-around for now:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 96e895d6be35..4605934a4fb7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1446,4 +1446,29 @@ static inline int amdgpu_in_reset(struct amdgpu_device *adev) { return atomic_read(&adev->in_gpu_reset); } + +/** + * memcpy_trailing - Copy the end of one structure into the middle of another + * + * @dst: Pointer to destination struct + * @first_dst_member: The member name in @dst where the overwrite begins + * @last_dst_member: The member name in @dst where the overwrite ends after + * @src: Pointer to the source struct + * @first_src_member: The member name in @src where the copy begins + * + */ +#define memcpy_trailing(dst, first_dst_member, last_dst_member, \ + src, first_src_member) \ +({ \ + size_t __src_offset = offsetof(typeof(*(src)), first_src_member); \ + size_t __src_size = sizeof(*(src)) - __src_offset; \ + size_t __dst_offset = offsetof(typeof(*(dst)), first_dst_member); \ + size_t __dst_size = offsetofend(typeof(*(dst)), last_dst_member) - \ + __dst_offset; \ + BUILD_BUG_ON(__src_size != __dst_size); \ + __builtin_memcpy((u8 *)(dst) + __dst_offset, \ + (u8 *)(src) + __src_offset, \ + __dst_size); \ +}) + #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 8ab58781ae13..1918e6232319 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -465,10 +465,8 @@ static int arcturus_append_powerplay_table(struct smu_context *smu)
if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 6)) - memcpy(&smc_pptable->MaxVoltageStepGfx, - &smc_dpm_table->maxvoltagestepgfx, - sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_6, maxvoltagestepgfx)); - + memcpy_trailing(smc_pptable, MaxVoltageStepGfx, BoardReserved, + smc_dpm_table, maxvoltagestepgfx); return 0; }
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 2e5d3669652b..b738042e064d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -431,16 +431,16 @@ static int navi10_append_powerplay_table(struct smu_context *smu)
switch (smc_dpm_table->table_header.content_revision) { case 5: /* nv10 and nv14 */ - memcpy(smc_pptable->I2cControllers, smc_dpm_table->I2cControllers, - sizeof(*smc_dpm_table) - sizeof(smc_dpm_table->table_header)); + memcpy_trailing(smc_pptable, I2cControllers, BoardReserved, + smc_dpm_table, I2cControllers); break; case 7: /* nv12 */ ret = amdgpu_atombios_get_data_table(adev, index, NULL, NULL, NULL, (uint8_t **)&smc_dpm_table_v4_7); if (ret) return ret; - memcpy(smc_pptable->I2cControllers, smc_dpm_table_v4_7->I2cControllers, - sizeof(*smc_dpm_table_v4_7) - sizeof(smc_dpm_table_v4_7->table_header)); + memcpy_trailing(smc_pptable, I2cControllers, BoardReserved, + smc_dpm_table_v4_7, I2cControllers); break; default: dev_err(smu->adev->dev, "smc_dpm_info with unsupported content revision %d!\n", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index c8eefacfdd37..a6fd7ee314a9 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -409,9 +409,8 @@ static int aldebaran_append_powerplay_table(struct smu_context *smu)
if ((smc_dpm_table->table_header.format_revision == 4) && (smc_dpm_table->table_header.content_revision == 10)) - memcpy(&smc_pptable->GfxMaxCurrent, - &smc_dpm_table->GfxMaxCurrent, - sizeof(*smc_dpm_table) - offsetof(struct atom_smc_dpm_info_v4_10, GfxMaxCurrent)); + memcpy_trailing(smc_pptable, GfxMaxCurrent, reserved, + smc_dpm_table, GfxMaxCurrent); return 0; }
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct hfa384x_tx_frame around members frame_control, duration_id, address[1-4], and sequence_control, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of frame_control.
"pahole" shows no size nor member offset changes to struct hfa384x_tx_frame. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences.)
Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org Cc: Rustam Kovhaev rkovhaev@gmail.com Cc: syzbot+22794221ab96b0bab53a@syzkaller.appspotmail.com Cc: Allen Pais apais@linux.microsoft.com Cc: Romain Perier romain.perier@gmail.com Cc: linux-staging@lists.linux.dev Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Link: https://lore.kernel.org/lkml/YQDvC4CghCazix4w@kroah.com --- drivers/staging/wlan-ng/hfa384x.h | 16 +++++++++------- drivers/staging/wlan-ng/hfa384x_usb.c | 4 +++- 2 files changed, 12 insertions(+), 8 deletions(-)
diff --git a/drivers/staging/wlan-ng/hfa384x.h b/drivers/staging/wlan-ng/hfa384x.h index 88e894dd3568..87eb87e3beab 100644 --- a/drivers/staging/wlan-ng/hfa384x.h +++ b/drivers/staging/wlan-ng/hfa384x.h @@ -476,13 +476,15 @@ struct hfa384x_tx_frame {
/*-- 802.11 Header Information --*/
- u16 frame_control; - u16 duration_id; - u8 address1[6]; - u8 address2[6]; - u8 address3[6]; - u16 sequence_control; - u8 address4[6]; + struct_group(p80211, + u16 frame_control; + u16 duration_id; + u8 address1[6]; + u8 address2[6]; + u8 address3[6]; + u16 sequence_control; + u8 address4[6]; + ); __le16 data_len; /* little endian format */
/*-- 802.3 Header Information --*/ diff --git a/drivers/staging/wlan-ng/hfa384x_usb.c b/drivers/staging/wlan-ng/hfa384x_usb.c index f2a0e16b0318..38aaae7a2d69 100644 --- a/drivers/staging/wlan-ng/hfa384x_usb.c +++ b/drivers/staging/wlan-ng/hfa384x_usb.c @@ -2516,7 +2516,9 @@ int hfa384x_drvr_txframe(struct hfa384x *hw, struct sk_buff *skb, cpu_to_le16s(&hw->txbuff.txfrm.desc.tx_control);
/* copy the header over to the txdesc */ - memcpy(&hw->txbuff.txfrm.desc.frame_control, p80211_hdr, + BUILD_BUG_ON(sizeof(hw->txbuff.txfrm.desc.p80211) != + sizeof(union p80211_hdr)); + memcpy(&hw->txbuff.txfrm.desc.p80211, p80211_hdr, sizeof(union p80211_hdr));
/* if we're using host WEP, increase size by IV+ICV */
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct drm32_mga_init around members chipset, sgram, maccess, fb_cpp, front_offset, front_pitch, back_offset, back_pitch, depth_cpp, depth_offset, depth_pitch, texture_offset, and texture_size, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of chipset.
"pahole" shows no size nor member offset changes to struct drm32_mga_init. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences and optimizations).
Note that since this is a UAPI header, __struct_group() is used directly.
Cc: David Airlie airlied@linux.ie Cc: Lee Jones lee.jones@linaro.org Cc: dri-devel@lists.freedesktop.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Daniel Vetter daniel@ffwll.ch Link: https://lore.kernel.org/lkml/YQKa76A6XuFqgM03@phenom.ffwll.local --- drivers/gpu/drm/mga/mga_ioc32.c | 27 ++++++++++++++------------- include/uapi/drm/mga_drm.h | 22 ++++++++++++---------- 2 files changed, 26 insertions(+), 23 deletions(-)
diff --git a/drivers/gpu/drm/mga/mga_ioc32.c b/drivers/gpu/drm/mga/mga_ioc32.c index 4fd4de16cd32..894472921c30 100644 --- a/drivers/gpu/drm/mga/mga_ioc32.c +++ b/drivers/gpu/drm/mga/mga_ioc32.c @@ -38,16 +38,18 @@ typedef struct drm32_mga_init { int func; u32 sarea_priv_offset; - int chipset; - int sgram; - unsigned int maccess; - unsigned int fb_cpp; - unsigned int front_offset, front_pitch; - unsigned int back_offset, back_pitch; - unsigned int depth_cpp; - unsigned int depth_offset, depth_pitch; - unsigned int texture_offset[MGA_NR_TEX_HEAPS]; - unsigned int texture_size[MGA_NR_TEX_HEAPS]; + struct_group(always32bit, + int chipset; + int sgram; + unsigned int maccess; + unsigned int fb_cpp; + unsigned int front_offset, front_pitch; + unsigned int back_offset, back_pitch; + unsigned int depth_cpp; + unsigned int depth_offset, depth_pitch; + unsigned int texture_offset[MGA_NR_TEX_HEAPS]; + unsigned int texture_size[MGA_NR_TEX_HEAPS]; + ); u32 fb_offset; u32 mmio_offset; u32 status_offset; @@ -67,9 +69,8 @@ static int compat_mga_init(struct file *file, unsigned int cmd,
init.func = init32.func; init.sarea_priv_offset = init32.sarea_priv_offset; - memcpy(&init.chipset, &init32.chipset, - offsetof(drm_mga_init_t, fb_offset) - - offsetof(drm_mga_init_t, chipset)); + memcpy(&init.always32bit, &init32.always32bit, + sizeof(init32.always32bit)); init.fb_offset = init32.fb_offset; init.mmio_offset = init32.mmio_offset; init.status_offset = init32.status_offset; diff --git a/include/uapi/drm/mga_drm.h b/include/uapi/drm/mga_drm.h index 8c4337548ab5..2978a435dff9 100644 --- a/include/uapi/drm/mga_drm.h +++ b/include/uapi/drm/mga_drm.h @@ -279,20 +279,22 @@ typedef struct drm_mga_init {
unsigned long sarea_priv_offset;
- int chipset; - int sgram; + __struct_group(/* no tye */, always32bit, /* no attrs */, + int chipset; + int sgram;
- unsigned int maccess; + unsigned int maccess;
- unsigned int fb_cpp; - unsigned int front_offset, front_pitch; - unsigned int back_offset, back_pitch; + unsigned int fb_cpp; + unsigned int front_offset, front_pitch; + unsigned int back_offset, back_pitch;
- unsigned int depth_cpp; - unsigned int depth_offset, depth_pitch; + unsigned int depth_cpp; + unsigned int depth_offset, depth_pitch;
- unsigned int texture_offset[MGA_NR_TEX_HEAPS]; - unsigned int texture_size[MGA_NR_TEX_HEAPS]; + unsigned int texture_offset[MGA_NR_TEX_HEAPS]; + unsigned int texture_size[MGA_NR_TEX_HEAPS]; + );
unsigned long fb_offset; unsigned long mmio_offset;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct vlan_ethhdr around members h_dest and h_source, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of h_dest.
"pahole" shows no size nor member offset changes to struct vlan_ethhdr. "objdump -d" shows no object code changes.
Cc: Saeed Mahameed saeedm@nvidia.com Cc: Leon Romanovsky leon@kernel.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 2 +- include/linux/if_vlan.h | 6 ++++-- 2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index c63d78eda606..39942a952736 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -207,7 +207,7 @@ static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs) int cpy1_sz = 2 * ETH_ALEN; int cpy2_sz = ihs - cpy1_sz;
- memcpy(vhdr, skb->data, cpy1_sz); + memcpy(&vhdr->addrs, skb->data, cpy1_sz); vhdr->h_vlan_proto = skb->vlan_proto; vhdr->h_vlan_TCI = cpu_to_be16(skb_vlan_tag_get(skb)); memcpy(&vhdr->h_vlan_encapsulated_proto, skb->data + cpy1_sz, cpy2_sz); diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 41a518336673..45aad461aa34 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -46,8 +46,10 @@ struct vlan_hdr { * @h_vlan_encapsulated_proto: packet type ID or len */ struct vlan_ethhdr { - unsigned char h_dest[ETH_ALEN]; - unsigned char h_source[ETH_ALEN]; + struct_group(addrs, + unsigned char h_dest[ETH_ALEN]; + unsigned char h_source[ETH_ALEN]; + ); __be16 h_vlan_proto; __be16 h_vlan_TCI; __be16 h_vlan_encapsulated_proto;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct cp2112_string_report around members report, length, type, and string, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of report.
"pahole" shows no size nor member offset changes to struct cp2112_string_report. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences.)
Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/hid/hid-cp2112.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/hid/hid-cp2112.c b/drivers/hid/hid-cp2112.c index 477baa30889c..ece147d1a278 100644 --- a/drivers/hid/hid-cp2112.c +++ b/drivers/hid/hid-cp2112.c @@ -129,10 +129,12 @@ struct cp2112_xfer_status_report {
struct cp2112_string_report { u8 dummy; /* force .string to be aligned */ - u8 report; /* CP2112_*_STRING */ - u8 length; /* length in bytes of everyting after .report */ - u8 type; /* USB_DT_STRING */ - wchar_t string[30]; /* UTF16_LITTLE_ENDIAN string */ + struct_group_attr(contents, __packed, + u8 report; /* CP2112_*_STRING */ + u8 length; /* length in bytes of everything after .report */ + u8 type; /* USB_DT_STRING */ + wchar_t string[30]; /* UTF16_LITTLE_ENDIAN string */ + ); } __packed;
/* Number of times to request transfer status before giving up waiting for a @@ -986,8 +988,8 @@ static ssize_t pstr_show(struct device *kdev, u8 length; int ret;
- ret = cp2112_hid_get(hdev, attr->report, &report.report, - sizeof(report) - 1, HID_FEATURE_REPORT); + ret = cp2112_hid_get(hdev, attr->report, (u8 *)&report.contents, + sizeof(report.contents), HID_FEATURE_REPORT); if (ret < 3) { hid_err(hdev, "error reading %s string: %d\n", kattr->attr.name, ret);
On Tue, 17 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct cp2112_string_report around members report, length, type, and string, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of report.
"pahole" shows no size nor member offset changes to struct cp2112_string_report. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences.)
Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thanks.
On Fri, Aug 20, 2021, 6:01 AM Jiri Kosina jikos@kernel.org wrote:
On Tue, 17 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct cp2112_string_report around members report, length, type, and string, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of report.
"pahole" shows no size nor member offset changes to struct cp2112_string_report. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences.)
Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thanks.
Same for this one: it's part of the larger series.
-Kees
On Fri, Aug 20, 2021 at 03:01:43PM +0200, Jiri Kosina wrote:
On Tue, 17 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct cp2112_string_report around members report, length, type, and string, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of report.
"pahole" shows no size nor member offset changes to struct cp2112_string_report. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences.)
Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thanks.
I'm not sure if my other HTML email got through, but please don't apply these to separate trees -- struct_group() is introduced as part of this series.
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields. Wrap the target region in struct_group(). This additionally fixes a theoretical misalignment of the copy (since the size of "buf" changes between 64-bit and 32-bit, but this is likely never built for 64-bit).
FWIW, I think this code is totally broken on 64-bit (which appears to not be a "real" build configuration): it would either always fail (with an uninitialized data->buf_size) or would cause corruption in userspace due to the copy_to_user() in the call path against an uninitialized data->buf value:
omap3isp_stat_request_statistics_time32(...) struct omap3isp_stat_data data64; ... omap3isp_stat_request_statistics(stat, &data64);
int omap3isp_stat_request_statistics(struct ispstat *stat, struct omap3isp_stat_data *data) ... buf = isp_stat_buf_get(stat, data);
static struct ispstat_buffer *isp_stat_buf_get(struct ispstat *stat, struct omap3isp_stat_data *data) ... if (buf->buf_size > data->buf_size) { ... return ERR_PTR(-EINVAL); } ... rval = copy_to_user(data->buf, buf->virt_addr, buf->buf_size);
Regardless, additionally initialize data64 to be zero-filled to avoid undefined behavior.
Fixes: 378e3f81cb56 ("media: omap3isp: support 64-bit version of omap3isp_stat_data") Signed-off-by: Kees Cook keescook@chromium.org --- drivers/media/platform/omap3isp/ispstat.c | 5 +++-- include/uapi/linux/omap3isp.h | 21 +++++++++++++-------- 2 files changed, 16 insertions(+), 10 deletions(-)
diff --git a/drivers/media/platform/omap3isp/ispstat.c b/drivers/media/platform/omap3isp/ispstat.c index 5b9b57f4d9bf..68cf68dbcace 100644 --- a/drivers/media/platform/omap3isp/ispstat.c +++ b/drivers/media/platform/omap3isp/ispstat.c @@ -512,7 +512,7 @@ int omap3isp_stat_request_statistics(struct ispstat *stat, int omap3isp_stat_request_statistics_time32(struct ispstat *stat, struct omap3isp_stat_data_time32 *data) { - struct omap3isp_stat_data data64; + struct omap3isp_stat_data data64 = { }; int ret;
ret = omap3isp_stat_request_statistics(stat, &data64); @@ -521,7 +521,8 @@ int omap3isp_stat_request_statistics_time32(struct ispstat *stat,
data->ts.tv_sec = data64.ts.tv_sec; data->ts.tv_usec = data64.ts.tv_usec; - memcpy(&data->buf, &data64.buf, sizeof(*data) - sizeof(data->ts)); + data->buf = (uintptr_t)data64.buf; + memcpy(&data->frame, &data64.frame, sizeof(data->frame));
return 0; } diff --git a/include/uapi/linux/omap3isp.h b/include/uapi/linux/omap3isp.h index 87b55755f4ff..9a6b3ed11455 100644 --- a/include/uapi/linux/omap3isp.h +++ b/include/uapi/linux/omap3isp.h @@ -162,6 +162,7 @@ struct omap3isp_h3a_aewb_config { * struct omap3isp_stat_data - Statistic data sent to or received from user * @ts: Timestamp of returned framestats. * @buf: Pointer to pass to user. + * @buf_size: Size of buffer. * @frame_number: Frame number of requested stats. * @cur_frame: Current frame number being processed. * @config_counter: Number of the configuration associated with the data. @@ -176,10 +177,12 @@ struct omap3isp_stat_data { struct timeval ts; #endif void __user *buf; - __u32 buf_size; - __u16 frame_number; - __u16 cur_frame; - __u16 config_counter; + __struct_group(/* no type */, frame, /* no attrs */, + __u32 buf_size; + __u16 frame_number; + __u16 cur_frame; + __u16 config_counter; + ); };
#ifdef __KERNEL__ @@ -189,10 +192,12 @@ struct omap3isp_stat_data_time32 { __s32 tv_usec; } ts; __u32 buf; - __u32 buf_size; - __u16 frame_number; - __u16 cur_frame; - __u16 config_counter; + __struct_group(/* no type */, frame, /* no attrs */, + __u32 buf_size; + __u16 frame_number; + __u16 cur_frame; + __u16 config_counter; + ); }; #endif
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields.
Use struct_group() in struct command_desc around members acmd and fill, so they can be referenced together. This will allow memset(), memcpy(), and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of acmd:
In function 'fortify_memset_chk', inlined from 'sata_fsl_qc_prep' at drivers/ata/sata_fsl.c:534:3: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Jens Axboe axboe@kernel.dk Cc: linux-ide@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/ata/sata_fsl.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c index e5838b23c9e0..fec3c9032606 100644 --- a/drivers/ata/sata_fsl.c +++ b/drivers/ata/sata_fsl.c @@ -246,8 +246,10 @@ enum { struct command_desc { u8 cfis[8 * 4]; u8 sfis[8 * 4]; - u8 acmd[4 * 4]; - u8 fill[4 * 4]; + struct_group(cdb, + u8 acmd[4 * 4]; + u8 fill[4 * 4]; + ); u32 prdt[SATA_FSL_MAX_PRD_DIRECT * 4]; u32 prdt_indirect[(SATA_FSL_MAX_PRD - SATA_FSL_MAX_PRD_DIRECT) * 4]; }; @@ -531,8 +533,8 @@ static enum ata_completion_errors sata_fsl_qc_prep(struct ata_queued_cmd *qc) /* setup "ACMD - atapi command" in cmd. desc. if this is ATAPI cmd */ if (ata_is_atapi(qc->tf.protocol)) { desc_info |= ATAPI_CMD; - memset((void *)&cd->acmd, 0, 32); - memcpy((void *)&cd->acmd, qc->cdb, qc->dev->cdb_len); + memset(&cd->cdb, 0, sizeof(cd->cdb)); + memcpy(&cd->cdb, qc->cdb, qc->dev->cdb_len); }
if (qc->flags & ATA_QCFLAG_DMAMAP)
Since all compilers support __builtin_object_size(), and there is only one user of __compiletime_object_size, remove it to avoid the needless indirection. This lets Clang reason about check_copy_size() correctly.
Link: https://github.com/ClangBuiltLinux/linux/issues/1179 Suggested-by: Nick Desaulniers ndesaulniers@google.com Cc: Nathan Chancellor nathan@kernel.org Cc: Nick Desaulniers ndesaulniers@google.com Cc: Miguel Ojeda ojeda@kernel.org Cc: Sedat Dilek sedat.dilek@gmail.com Cc: Will Deacon will@kernel.org Cc: Marco Elver elver@google.com Cc: Arvind Sankar nivedita@alum.mit.edu Cc: Masahiro Yamada masahiroy@kernel.org Cc: Luc Van Oostenryck luc.vanoostenryck@gmail.com Cc: Andrew Morton akpm@linux-foundation.org Cc: Sami Tolvanen samitolvanen@google.com Cc: Thomas Gleixner tglx@linutronix.de Cc: Gabriel Krisman Bertazi krisman@collabora.com Cc: Andy Lutomirski luto@kernel.org Cc: Oleg Nesterov oleg@redhat.com Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/compiler-gcc.h | 2 -- include/linux/compiler_types.h | 4 ---- include/linux/thread_info.h | 2 +- 3 files changed, 1 insertion(+), 7 deletions(-)
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h index cb9217fc60af..01985821944b 100644 --- a/include/linux/compiler-gcc.h +++ b/include/linux/compiler-gcc.h @@ -41,8 +41,6 @@
#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__)
-#define __compiletime_object_size(obj) __builtin_object_size(obj, 0) - #define __compiletime_warning(message) __attribute__((__warning__(message))) #define __compiletime_error(message) __attribute__((__error__(message)))
diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h index e4ea86fc584d..c43308b0a9a9 100644 --- a/include/linux/compiler_types.h +++ b/include/linux/compiler_types.h @@ -290,10 +290,6 @@ struct ftrace_likely_data { (sizeof(t) == sizeof(char) || sizeof(t) == sizeof(short) || \ sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long))
-/* Compile time object size, -1 for unknown */ -#ifndef __compiletime_object_size -# define __compiletime_object_size(obj) -1 -#endif #ifndef __compiletime_warning # define __compiletime_warning(message) #endif diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index 0999f6317978..ad0c4e041030 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -203,7 +203,7 @@ static inline void copy_overflow(int size, unsigned long count) static __always_inline __must_check bool check_copy_size(const void *addr, size_t bytes, bool is_source) { - int sz = __compiletime_object_size(addr); + int sz = __builtin_object_size(addr, 0); if (unlikely(sz >= 0 && sz < bytes)) { if (!__builtin_constant_p(bytes)) copy_overflow(sz, bytes);
On Wed, Aug 18, 2021 at 8:14 AM Kees Cook keescook@chromium.org wrote:
Since all compilers support __builtin_object_size(), and there is only one user of __compiletime_object_size, remove it to avoid the needless indirection. This lets Clang reason about check_copy_size() correctly.
Sounds good to me, the indirection is making things more complex than they need to be.
Reviewed-by: Miguel Ojeda ojeda@kernel.org
Cheers, Miguel
The core functions of string.c are those that may be implemented by per-architecture functions, or overloaded by FORTIFY_SOURCE. As a result, it needs to be built with __NO_FORTIFY. Without this, macros will collide with function declarations. This was accidentally working due to -ffreestanding (on some architectures). Make this deterministic by explicitly setting __NO_FORTIFY and move all the helper functions into string_helpers.c so that they gain the fortification coverage they had been missing.
Cc: Andy Shevchenko andy@kernel.org Cc: Andrew Morton akpm@linux-foundation.org Cc: Nick Desaulniers ndesaulniers@google.com Cc: Andy Lavr andy.lavr@gmail.com Cc: Nathan Chancellor nathan@kernel.org Cc: Alexey Dobriyan adobriyan@gmail.com Cc: Stephen Rothwell sfr@canb.auug.org.au Cc: Bartosz Golaszewski bgolaszewski@baylibre.com Signed-off-by: Kees Cook keescook@chromium.org --- arch/arm/boot/compressed/string.c | 1 + arch/s390/lib/string.c | 3 + arch/x86/boot/compressed/misc.h | 2 + arch/x86/boot/compressed/pgtable_64.c | 2 + arch/x86/lib/string_32.c | 1 + lib/string.c | 210 +------------------------- lib/string_helpers.c | 193 +++++++++++++++++++++++ 7 files changed, 208 insertions(+), 204 deletions(-)
diff --git a/arch/arm/boot/compressed/string.c b/arch/arm/boot/compressed/string.c index 8c0fa276d994..fcc678fce045 100644 --- a/arch/arm/boot/compressed/string.c +++ b/arch/arm/boot/compressed/string.c @@ -5,6 +5,7 @@ * Small subset of simple string routines */
+#define __NO_FORTIFY #include <linux/string.h>
/* diff --git a/arch/s390/lib/string.c b/arch/s390/lib/string.c index cfcdf76d6a95..392fb9f4f4db 100644 --- a/arch/s390/lib/string.c +++ b/arch/s390/lib/string.c @@ -8,6 +8,9 @@ */
#define IN_ARCH_STRING_C 1 +#ifndef __NO_FORTIFY +# define __NO_FORTIFY +#endif
#include <linux/types.h> #include <linux/string.h> diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h index 31139256859f..49bde196da9b 100644 --- a/arch/x86/boot/compressed/misc.h +++ b/arch/x86/boot/compressed/misc.h @@ -14,6 +14,8 @@ #undef CONFIG_KASAN #undef CONFIG_KASAN_GENERIC
+#define __NO_FORTIFY + /* cpu_feature_enabled() cannot be used this early */ #define USE_EARLY_PGTABLE_L5
diff --git a/arch/x86/boot/compressed/pgtable_64.c b/arch/x86/boot/compressed/pgtable_64.c index 2a78746f5a4c..a1733319a22a 100644 --- a/arch/x86/boot/compressed/pgtable_64.c +++ b/arch/x86/boot/compressed/pgtable_64.c @@ -1,3 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "misc.h" #include <linux/efi.h> #include <asm/e820/types.h> #include <asm/processor.h> diff --git a/arch/x86/lib/string_32.c b/arch/x86/lib/string_32.c index d15fdae9656e..53b3f202267c 100644 --- a/arch/x86/lib/string_32.c +++ b/arch/x86/lib/string_32.c @@ -11,6 +11,7 @@ * strings. */
+#define __NO_FORTIFY #include <linux/string.h> #include <linux/export.h>
diff --git a/lib/string.c b/lib/string.c index 4fec38fc6e58..4e111d9dd6d5 100644 --- a/lib/string.c +++ b/lib/string.c @@ -6,20 +6,15 @@ */
/* - * stupid library routines.. The optimized versions should generally be found - * as inline code in <asm-xx/string.h> + * This file should be used only for "library" routines that may have + * alternative implementations on specific architectures (generally + * found in <asm-xx/string.h>), or get overloaded by FORTIFY_SOURCE. + * (Specifically, this file is built with __NO_FORTIFY.) * - * These are buggy as well.. - * - * * Fri Jun 25 1999, Ingo Oeser ioe@informatik.tu-chemnitz.de - * - Added strsep() which will replace strtok() soon (because strsep() is - * reentrant and should be faster). Use only strsep() in new code, please. - * - * * Sat Feb 09 2002, Jason Thomas jason@topic.com.au, - * Matthew Hawkins matt@mh.dropbear.id.au - * - Kissed strtok() goodbye + * Other helper functions should live in string_helpers.c. */
+#define __NO_FORTIFY #include <linux/types.h> #include <linux/string.h> #include <linux/ctype.h> @@ -254,40 +249,6 @@ ssize_t strscpy(char *dest, const char *src, size_t count) EXPORT_SYMBOL(strscpy); #endif
-/** - * strscpy_pad() - Copy a C-string into a sized buffer - * @dest: Where to copy the string to - * @src: Where to copy the string from - * @count: Size of destination buffer - * - * Copy the string, or as much of it as fits, into the dest buffer. The - * behavior is undefined if the string buffers overlap. The destination - * buffer is always %NUL terminated, unless it's zero-sized. - * - * If the source string is shorter than the destination buffer, zeros - * the tail of the destination buffer. - * - * For full explanation of why you may want to consider using the - * 'strscpy' functions please see the function docstring for strscpy(). - * - * Returns: - * * The number of characters copied (not including the trailing %NUL) - * * -E2BIG if count is 0 or @src was truncated. - */ -ssize_t strscpy_pad(char *dest, const char *src, size_t count) -{ - ssize_t written; - - written = strscpy(dest, src, count); - if (written < 0 || written == count - 1) - return written; - - memset(dest + written + 1, 0, count - written - 1); - - return written; -} -EXPORT_SYMBOL(strscpy_pad); - /** * stpcpy - copy a string from src to dest returning a pointer to the new end * of dest, including src's %NUL-terminator. May overrun dest. @@ -530,46 +491,6 @@ char *strnchr(const char *s, size_t count, int c) EXPORT_SYMBOL(strnchr); #endif
-/** - * skip_spaces - Removes leading whitespace from @str. - * @str: The string to be stripped. - * - * Returns a pointer to the first non-whitespace character in @str. - */ -char *skip_spaces(const char *str) -{ - while (isspace(*str)) - ++str; - return (char *)str; -} -EXPORT_SYMBOL(skip_spaces); - -/** - * strim - Removes leading and trailing whitespace from @s. - * @s: The string to be stripped. - * - * Note that the first trailing whitespace is replaced with a %NUL-terminator - * in the given string @s. Returns a pointer to the first non-whitespace - * character in @s. - */ -char *strim(char *s) -{ - size_t size; - char *end; - - size = strlen(s); - if (!size) - return s; - - end = s + size - 1; - while (end >= s && isspace(*end)) - end--; - *(end + 1) = '\0'; - - return skip_spaces(s); -} -EXPORT_SYMBOL(strim); - #ifndef __HAVE_ARCH_STRLEN /** * strlen - Find the length of a string @@ -704,101 +625,6 @@ char *strsep(char **s, const char *ct) EXPORT_SYMBOL(strsep); #endif
-/** - * sysfs_streq - return true if strings are equal, modulo trailing newline - * @s1: one string - * @s2: another string - * - * This routine returns true iff two strings are equal, treating both - * NUL and newline-then-NUL as equivalent string terminations. It's - * geared for use with sysfs input strings, which generally terminate - * with newlines but are compared against values without newlines. - */ -bool sysfs_streq(const char *s1, const char *s2) -{ - while (*s1 && *s1 == *s2) { - s1++; - s2++; - } - - if (*s1 == *s2) - return true; - if (!*s1 && *s2 == '\n' && !s2[1]) - return true; - if (*s1 == '\n' && !s1[1] && !*s2) - return true; - return false; -} -EXPORT_SYMBOL(sysfs_streq); - -/** - * match_string - matches given string in an array - * @array: array of strings - * @n: number of strings in the array or -1 for NULL terminated arrays - * @string: string to match with - * - * This routine will look for a string in an array of strings up to the - * n-th element in the array or until the first NULL element. - * - * Historically the value of -1 for @n, was used to search in arrays that - * are NULL terminated. However, the function does not make a distinction - * when finishing the search: either @n elements have been compared OR - * the first NULL element was found. - * - * Return: - * index of a @string in the @array if matches, or %-EINVAL otherwise. - */ -int match_string(const char * const *array, size_t n, const char *string) -{ - int index; - const char *item; - - for (index = 0; index < n; index++) { - item = array[index]; - if (!item) - break; - if (!strcmp(item, string)) - return index; - } - - return -EINVAL; -} -EXPORT_SYMBOL(match_string); - -/** - * __sysfs_match_string - matches given string in an array - * @array: array of strings - * @n: number of strings in the array or -1 for NULL terminated arrays - * @str: string to match with - * - * Returns index of @str in the @array or -EINVAL, just like match_string(). - * Uses sysfs_streq instead of strcmp for matching. - * - * This routine will look for a string in an array of strings up to the - * n-th element in the array or until the first NULL element. - * - * Historically the value of -1 for @n, was used to search in arrays that - * are NULL terminated. However, the function does not make a distinction - * when finishing the search: either @n elements have been compared OR - * the first NULL element was found. - */ -int __sysfs_match_string(const char * const *array, size_t n, const char *str) -{ - const char *item; - int index; - - for (index = 0; index < n; index++) { - item = array[index]; - if (!item) - break; - if (sysfs_streq(item, str)) - return index; - } - - return -EINVAL; -} -EXPORT_SYMBOL(__sysfs_match_string); - #ifndef __HAVE_ARCH_MEMSET /** * memset - Fill a region of memory with the given value @@ -1221,27 +1047,3 @@ void *memchr_inv(const void *start, int c, size_t bytes) return check_bytes8(start, value, bytes % 8); } EXPORT_SYMBOL(memchr_inv); - -/** - * strreplace - Replace all occurrences of character in string. - * @s: The string to operate on. - * @old: The character being replaced. - * @new: The character @old is replaced with. - * - * Returns pointer to the nul byte at the end of @s. - */ -char *strreplace(char *s, char old, char new) -{ - for (; *s; ++s) - if (*s == old) - *s = new; - return s; -} -EXPORT_SYMBOL(strreplace); - -void fortify_panic(const char *name) -{ - pr_emerg("detected buffer overflow in %s\n", name); - BUG(); -} -EXPORT_SYMBOL(fortify_panic); diff --git a/lib/string_helpers.c b/lib/string_helpers.c index 3806a52ce697..bde13612c25d 100644 --- a/lib/string_helpers.c +++ b/lib/string_helpers.c @@ -696,3 +696,196 @@ void kfree_strarray(char **array, size_t n) kfree(array); } EXPORT_SYMBOL_GPL(kfree_strarray); + +/** + * strscpy_pad() - Copy a C-string into a sized buffer + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @count: Size of destination buffer + * + * Copy the string, or as much of it as fits, into the dest buffer. The + * behavior is undefined if the string buffers overlap. The destination + * buffer is always %NUL terminated, unless it's zero-sized. + * + * If the source string is shorter than the destination buffer, zeros + * the tail of the destination buffer. + * + * For full explanation of why you may want to consider using the + * 'strscpy' functions please see the function docstring for strscpy(). + * + * Returns: + * * The number of characters copied (not including the trailing %NUL) + * * -E2BIG if count is 0 or @src was truncated. + */ +ssize_t strscpy_pad(char *dest, const char *src, size_t count) +{ + ssize_t written; + + written = strscpy(dest, src, count); + if (written < 0 || written == count - 1) + return written; + + memset(dest + written + 1, 0, count - written - 1); + + return written; +} +EXPORT_SYMBOL(strscpy_pad); + +/** + * skip_spaces - Removes leading whitespace from @str. + * @str: The string to be stripped. + * + * Returns a pointer to the first non-whitespace character in @str. + */ +char *skip_spaces(const char *str) +{ + while (isspace(*str)) + ++str; + return (char *)str; +} +EXPORT_SYMBOL(skip_spaces); + +/** + * strim - Removes leading and trailing whitespace from @s. + * @s: The string to be stripped. + * + * Note that the first trailing whitespace is replaced with a %NUL-terminator + * in the given string @s. Returns a pointer to the first non-whitespace + * character in @s. + */ +char *strim(char *s) +{ + size_t size; + char *end; + + size = strlen(s); + if (!size) + return s; + + end = s + size - 1; + while (end >= s && isspace(*end)) + end--; + *(end + 1) = '\0'; + + return skip_spaces(s); +} +EXPORT_SYMBOL(strim); + +/** + * sysfs_streq - return true if strings are equal, modulo trailing newline + * @s1: one string + * @s2: another string + * + * This routine returns true iff two strings are equal, treating both + * NUL and newline-then-NUL as equivalent string terminations. It's + * geared for use with sysfs input strings, which generally terminate + * with newlines but are compared against values without newlines. + */ +bool sysfs_streq(const char *s1, const char *s2) +{ + while (*s1 && *s1 == *s2) { + s1++; + s2++; + } + + if (*s1 == *s2) + return true; + if (!*s1 && *s2 == '\n' && !s2[1]) + return true; + if (*s1 == '\n' && !s1[1] && !*s2) + return true; + return false; +} +EXPORT_SYMBOL(sysfs_streq); + +/** + * match_string - matches given string in an array + * @array: array of strings + * @n: number of strings in the array or -1 for NULL terminated arrays + * @string: string to match with + * + * This routine will look for a string in an array of strings up to the + * n-th element in the array or until the first NULL element. + * + * Historically the value of -1 for @n, was used to search in arrays that + * are NULL terminated. However, the function does not make a distinction + * when finishing the search: either @n elements have been compared OR + * the first NULL element was found. + * + * Return: + * index of a @string in the @array if matches, or %-EINVAL otherwise. + */ +int match_string(const char * const *array, size_t n, const char *string) +{ + int index; + const char *item; + + for (index = 0; index < n; index++) { + item = array[index]; + if (!item) + break; + if (!strcmp(item, string)) + return index; + } + + return -EINVAL; +} +EXPORT_SYMBOL(match_string); + +/** + * __sysfs_match_string - matches given string in an array + * @array: array of strings + * @n: number of strings in the array or -1 for NULL terminated arrays + * @str: string to match with + * + * Returns index of @str in the @array or -EINVAL, just like match_string(). + * Uses sysfs_streq instead of strcmp for matching. + * + * This routine will look for a string in an array of strings up to the + * n-th element in the array or until the first NULL element. + * + * Historically the value of -1 for @n, was used to search in arrays that + * are NULL terminated. However, the function does not make a distinction + * when finishing the search: either @n elements have been compared OR + * the first NULL element was found. + */ +int __sysfs_match_string(const char * const *array, size_t n, const char *str) +{ + const char *item; + int index; + + for (index = 0; index < n; index++) { + item = array[index]; + if (!item) + break; + if (sysfs_streq(item, str)) + return index; + } + + return -EINVAL; +} +EXPORT_SYMBOL(__sysfs_match_string); + +/** + * strreplace - Replace all occurrences of character in string. + * @s: The string to operate on. + * @old: The character being replaced. + * @new: The character @old is replaced with. + * + * Returns pointer to the nul byte at the end of @s. + */ +char *strreplace(char *s, char old, char new) +{ + for (; *s; ++s) + if (*s == old) + *s = new; + return s; +} +EXPORT_SYMBOL(strreplace); + +void fortify_panic(const char *name) +{ + pr_emerg("detected buffer overflow in %s\n", name); + BUG(); +} +EXPORT_SYMBOL(fortify_panic);
On Tue, Aug 17, 2021 at 11:04:56PM -0700, Kees Cook wrote:
The core functions of string.c are those that may be implemented by per-architecture functions, or overloaded by FORTIFY_SOURCE. As a result, it needs to be built with __NO_FORTIFY. Without this, macros will collide with function declarations. This was accidentally working due to -ffreestanding (on some architectures). Make this deterministic by explicitly setting __NO_FORTIFY and move all the helper functions into string_helpers.c so that they gain the fortification coverage they had been missing.
No objections Acked-by: Andy Shevchenko andriy.shevchenko@linux.intel.com
Cc: Andy Shevchenko andy@kernel.org Cc: Andrew Morton akpm@linux-foundation.org Cc: Nick Desaulniers ndesaulniers@google.com Cc: Andy Lavr andy.lavr@gmail.com Cc: Nathan Chancellor nathan@kernel.org Cc: Alexey Dobriyan adobriyan@gmail.com Cc: Stephen Rothwell sfr@canb.auug.org.au Cc: Bartosz Golaszewski bgolaszewski@baylibre.com Signed-off-by: Kees Cook keescook@chromium.org
arch/arm/boot/compressed/string.c | 1 + arch/s390/lib/string.c | 3 + arch/x86/boot/compressed/misc.h | 2 + arch/x86/boot/compressed/pgtable_64.c | 2 + arch/x86/lib/string_32.c | 1 + lib/string.c | 210 +------------------------- lib/string_helpers.c | 193 +++++++++++++++++++++++ 7 files changed, 208 insertions(+), 204 deletions(-)
diff --git a/arch/arm/boot/compressed/string.c b/arch/arm/boot/compressed/string.c index 8c0fa276d994..fcc678fce045 100644 --- a/arch/arm/boot/compressed/string.c +++ b/arch/arm/boot/compressed/string.c @@ -5,6 +5,7 @@
- Small subset of simple string routines
*/
+#define __NO_FORTIFY #include <linux/string.h>
/* diff --git a/arch/s390/lib/string.c b/arch/s390/lib/string.c index cfcdf76d6a95..392fb9f4f4db 100644 --- a/arch/s390/lib/string.c +++ b/arch/s390/lib/string.c @@ -8,6 +8,9 @@ */
#define IN_ARCH_STRING_C 1 +#ifndef __NO_FORTIFY +# define __NO_FORTIFY +#endif
#include <linux/types.h> #include <linux/string.h> diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h index 31139256859f..49bde196da9b 100644 --- a/arch/x86/boot/compressed/misc.h +++ b/arch/x86/boot/compressed/misc.h @@ -14,6 +14,8 @@ #undef CONFIG_KASAN #undef CONFIG_KASAN_GENERIC
+#define __NO_FORTIFY
/* cpu_feature_enabled() cannot be used this early */ #define USE_EARLY_PGTABLE_L5
diff --git a/arch/x86/boot/compressed/pgtable_64.c b/arch/x86/boot/compressed/pgtable_64.c index 2a78746f5a4c..a1733319a22a 100644 --- a/arch/x86/boot/compressed/pgtable_64.c +++ b/arch/x86/boot/compressed/pgtable_64.c @@ -1,3 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "misc.h" #include <linux/efi.h> #include <asm/e820/types.h> #include <asm/processor.h> diff --git a/arch/x86/lib/string_32.c b/arch/x86/lib/string_32.c index d15fdae9656e..53b3f202267c 100644 --- a/arch/x86/lib/string_32.c +++ b/arch/x86/lib/string_32.c @@ -11,6 +11,7 @@
- strings.
*/
+#define __NO_FORTIFY #include <linux/string.h> #include <linux/export.h>
diff --git a/lib/string.c b/lib/string.c index 4fec38fc6e58..4e111d9dd6d5 100644 --- a/lib/string.c +++ b/lib/string.c @@ -6,20 +6,15 @@ */
/*
- stupid library routines.. The optimized versions should generally be found
- as inline code in <asm-xx/string.h>
- This file should be used only for "library" routines that may have
- alternative implementations on specific architectures (generally
- found in <asm-xx/string.h>), or get overloaded by FORTIFY_SOURCE.
- (Specifically, this file is built with __NO_FORTIFY.)
- These are buggy as well..
- Fri Jun 25 1999, Ingo Oeser ioe@informatik.tu-chemnitz.de
- Added strsep() which will replace strtok() soon (because strsep() is
- reentrant and should be faster). Use only strsep() in new code, please.
- Sat Feb 09 2002, Jason Thomas jason@topic.com.au,
Matthew Hawkins <matt@mh.dropbear.id.au>
- Kissed strtok() goodbye
*/
- Other helper functions should live in string_helpers.c.
+#define __NO_FORTIFY #include <linux/types.h> #include <linux/string.h> #include <linux/ctype.h> @@ -254,40 +249,6 @@ ssize_t strscpy(char *dest, const char *src, size_t count) EXPORT_SYMBOL(strscpy); #endif
-/**
- strscpy_pad() - Copy a C-string into a sized buffer
- @dest: Where to copy the string to
- @src: Where to copy the string from
- @count: Size of destination buffer
- Copy the string, or as much of it as fits, into the dest buffer. The
- behavior is undefined if the string buffers overlap. The destination
- buffer is always %NUL terminated, unless it's zero-sized.
- If the source string is shorter than the destination buffer, zeros
- the tail of the destination buffer.
- For full explanation of why you may want to consider using the
- 'strscpy' functions please see the function docstring for strscpy().
- Returns:
- The number of characters copied (not including the trailing %NUL)
- -E2BIG if count is 0 or @src was truncated.
- */
-ssize_t strscpy_pad(char *dest, const char *src, size_t count) -{
- ssize_t written;
- written = strscpy(dest, src, count);
- if (written < 0 || written == count - 1)
return written;
- memset(dest + written + 1, 0, count - written - 1);
- return written;
-} -EXPORT_SYMBOL(strscpy_pad);
/**
- stpcpy - copy a string from src to dest returning a pointer to the new end
of dest, including src's %NUL-terminator. May overrun dest.
@@ -530,46 +491,6 @@ char *strnchr(const char *s, size_t count, int c) EXPORT_SYMBOL(strnchr); #endif
-/**
- skip_spaces - Removes leading whitespace from @str.
- @str: The string to be stripped.
- Returns a pointer to the first non-whitespace character in @str.
- */
-char *skip_spaces(const char *str) -{
- while (isspace(*str))
++str;
- return (char *)str;
-} -EXPORT_SYMBOL(skip_spaces);
-/**
- strim - Removes leading and trailing whitespace from @s.
- @s: The string to be stripped.
- Note that the first trailing whitespace is replaced with a %NUL-terminator
- in the given string @s. Returns a pointer to the first non-whitespace
- character in @s.
- */
-char *strim(char *s) -{
- size_t size;
- char *end;
- size = strlen(s);
- if (!size)
return s;
- end = s + size - 1;
- while (end >= s && isspace(*end))
end--;
- *(end + 1) = '\0';
- return skip_spaces(s);
-} -EXPORT_SYMBOL(strim);
#ifndef __HAVE_ARCH_STRLEN /**
- strlen - Find the length of a string
@@ -704,101 +625,6 @@ char *strsep(char **s, const char *ct) EXPORT_SYMBOL(strsep); #endif
-/**
- sysfs_streq - return true if strings are equal, modulo trailing newline
- @s1: one string
- @s2: another string
- This routine returns true iff two strings are equal, treating both
- NUL and newline-then-NUL as equivalent string terminations. It's
- geared for use with sysfs input strings, which generally terminate
- with newlines but are compared against values without newlines.
- */
-bool sysfs_streq(const char *s1, const char *s2) -{
- while (*s1 && *s1 == *s2) {
s1++;
s2++;
- }
- if (*s1 == *s2)
return true;
- if (!*s1 && *s2 == '\n' && !s2[1])
return true;
- if (*s1 == '\n' && !s1[1] && !*s2)
return true;
- return false;
-} -EXPORT_SYMBOL(sysfs_streq);
-/**
- match_string - matches given string in an array
- @array: array of strings
- @n: number of strings in the array or -1 for NULL terminated arrays
- @string: string to match with
- This routine will look for a string in an array of strings up to the
- n-th element in the array or until the first NULL element.
- Historically the value of -1 for @n, was used to search in arrays that
- are NULL terminated. However, the function does not make a distinction
- when finishing the search: either @n elements have been compared OR
- the first NULL element was found.
- Return:
- index of a @string in the @array if matches, or %-EINVAL otherwise.
- */
-int match_string(const char * const *array, size_t n, const char *string) -{
- int index;
- const char *item;
- for (index = 0; index < n; index++) {
item = array[index];
if (!item)
break;
if (!strcmp(item, string))
return index;
- }
- return -EINVAL;
-} -EXPORT_SYMBOL(match_string);
-/**
- __sysfs_match_string - matches given string in an array
- @array: array of strings
- @n: number of strings in the array or -1 for NULL terminated arrays
- @str: string to match with
- Returns index of @str in the @array or -EINVAL, just like match_string().
- Uses sysfs_streq instead of strcmp for matching.
- This routine will look for a string in an array of strings up to the
- n-th element in the array or until the first NULL element.
- Historically the value of -1 for @n, was used to search in arrays that
- are NULL terminated. However, the function does not make a distinction
- when finishing the search: either @n elements have been compared OR
- the first NULL element was found.
- */
-int __sysfs_match_string(const char * const *array, size_t n, const char *str) -{
- const char *item;
- int index;
- for (index = 0; index < n; index++) {
item = array[index];
if (!item)
break;
if (sysfs_streq(item, str))
return index;
- }
- return -EINVAL;
-} -EXPORT_SYMBOL(__sysfs_match_string);
#ifndef __HAVE_ARCH_MEMSET /**
- memset - Fill a region of memory with the given value
@@ -1221,27 +1047,3 @@ void *memchr_inv(const void *start, int c, size_t bytes) return check_bytes8(start, value, bytes % 8); } EXPORT_SYMBOL(memchr_inv);
-/**
- strreplace - Replace all occurrences of character in string.
- @s: The string to operate on.
- @old: The character being replaced.
- @new: The character @old is replaced with.
- Returns pointer to the nul byte at the end of @s.
- */
-char *strreplace(char *s, char old, char new) -{
- for (; *s; ++s)
if (*s == old)
*s = new;
- return s;
-} -EXPORT_SYMBOL(strreplace);
-void fortify_panic(const char *name) -{
- pr_emerg("detected buffer overflow in %s\n", name);
- BUG();
-} -EXPORT_SYMBOL(fortify_panic); diff --git a/lib/string_helpers.c b/lib/string_helpers.c index 3806a52ce697..bde13612c25d 100644 --- a/lib/string_helpers.c +++ b/lib/string_helpers.c @@ -696,3 +696,196 @@ void kfree_strarray(char **array, size_t n) kfree(array); } EXPORT_SYMBOL_GPL(kfree_strarray);
+/**
- strscpy_pad() - Copy a C-string into a sized buffer
- @dest: Where to copy the string to
- @src: Where to copy the string from
- @count: Size of destination buffer
- Copy the string, or as much of it as fits, into the dest buffer. The
- behavior is undefined if the string buffers overlap. The destination
- buffer is always %NUL terminated, unless it's zero-sized.
- If the source string is shorter than the destination buffer, zeros
- the tail of the destination buffer.
- For full explanation of why you may want to consider using the
- 'strscpy' functions please see the function docstring for strscpy().
- Returns:
- The number of characters copied (not including the trailing %NUL)
- -E2BIG if count is 0 or @src was truncated.
- */
+ssize_t strscpy_pad(char *dest, const char *src, size_t count) +{
- ssize_t written;
- written = strscpy(dest, src, count);
- if (written < 0 || written == count - 1)
return written;
- memset(dest + written + 1, 0, count - written - 1);
- return written;
+} +EXPORT_SYMBOL(strscpy_pad);
+/**
- skip_spaces - Removes leading whitespace from @str.
- @str: The string to be stripped.
- Returns a pointer to the first non-whitespace character in @str.
- */
+char *skip_spaces(const char *str) +{
- while (isspace(*str))
++str;
- return (char *)str;
+} +EXPORT_SYMBOL(skip_spaces);
+/**
- strim - Removes leading and trailing whitespace from @s.
- @s: The string to be stripped.
- Note that the first trailing whitespace is replaced with a %NUL-terminator
- in the given string @s. Returns a pointer to the first non-whitespace
- character in @s.
- */
+char *strim(char *s) +{
- size_t size;
- char *end;
- size = strlen(s);
- if (!size)
return s;
- end = s + size - 1;
- while (end >= s && isspace(*end))
end--;
- *(end + 1) = '\0';
- return skip_spaces(s);
+} +EXPORT_SYMBOL(strim);
+/**
- sysfs_streq - return true if strings are equal, modulo trailing newline
- @s1: one string
- @s2: another string
- This routine returns true iff two strings are equal, treating both
- NUL and newline-then-NUL as equivalent string terminations. It's
- geared for use with sysfs input strings, which generally terminate
- with newlines but are compared against values without newlines.
- */
+bool sysfs_streq(const char *s1, const char *s2) +{
- while (*s1 && *s1 == *s2) {
s1++;
s2++;
- }
- if (*s1 == *s2)
return true;
- if (!*s1 && *s2 == '\n' && !s2[1])
return true;
- if (*s1 == '\n' && !s1[1] && !*s2)
return true;
- return false;
+} +EXPORT_SYMBOL(sysfs_streq);
+/**
- match_string - matches given string in an array
- @array: array of strings
- @n: number of strings in the array or -1 for NULL terminated arrays
- @string: string to match with
- This routine will look for a string in an array of strings up to the
- n-th element in the array or until the first NULL element.
- Historically the value of -1 for @n, was used to search in arrays that
- are NULL terminated. However, the function does not make a distinction
- when finishing the search: either @n elements have been compared OR
- the first NULL element was found.
- Return:
- index of a @string in the @array if matches, or %-EINVAL otherwise.
- */
+int match_string(const char * const *array, size_t n, const char *string) +{
- int index;
- const char *item;
- for (index = 0; index < n; index++) {
item = array[index];
if (!item)
break;
if (!strcmp(item, string))
return index;
- }
- return -EINVAL;
+} +EXPORT_SYMBOL(match_string);
+/**
- __sysfs_match_string - matches given string in an array
- @array: array of strings
- @n: number of strings in the array or -1 for NULL terminated arrays
- @str: string to match with
- Returns index of @str in the @array or -EINVAL, just like match_string().
- Uses sysfs_streq instead of strcmp for matching.
- This routine will look for a string in an array of strings up to the
- n-th element in the array or until the first NULL element.
- Historically the value of -1 for @n, was used to search in arrays that
- are NULL terminated. However, the function does not make a distinction
- when finishing the search: either @n elements have been compared OR
- the first NULL element was found.
- */
+int __sysfs_match_string(const char * const *array, size_t n, const char *str) +{
- const char *item;
- int index;
- for (index = 0; index < n; index++) {
item = array[index];
if (!item)
break;
if (sysfs_streq(item, str))
return index;
- }
- return -EINVAL;
+} +EXPORT_SYMBOL(__sysfs_match_string);
+/**
- strreplace - Replace all occurrences of character in string.
- @s: The string to operate on.
- @old: The character being replaced.
- @new: The character @old is replaced with.
- Returns pointer to the nul byte at the end of @s.
- */
+char *strreplace(char *s, char old, char new) +{
- for (; *s; ++s)
if (*s == old)
*s = new;
- return s;
+} +EXPORT_SYMBOL(strreplace);
+void fortify_panic(const char *name) +{
- pr_emerg("detected buffer overflow in %s\n", name);
- BUG();
+}
+EXPORT_SYMBOL(fortify_panic);
2.30.2
When commit a28a6e860c6c ("string.h: move fortified functions definitions in a dedicated header.") moved the fortify-specific code, some helpers were left behind. Moves the remaining fortify-specific helpers into fortify-string.h so they're together where they're used. This requires that any FORTIFY helper function prototypes be conditionally built to avoid "no prototype" warnings. Additionally removes unused helpers.
Cc: Andrew Morton akpm@linux-foundation.org Cc: Francis Laniel laniel_francis@privacyrequired.com Cc: Daniel Axtens dja@axtens.net Cc: Vincenzo Frascino vincenzo.frascino@arm.com Cc: Andrey Konovalov andreyknvl@google.com Cc: Dan Williams dan.j.williams@intel.com Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 7 +++++++ include/linux/string.h | 9 --------- lib/string_helpers.c | 2 ++ 3 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index c1be37437e77..7e67d02764db 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -2,6 +2,13 @@ #ifndef _LINUX_FORTIFY_STRING_H_ #define _LINUX_FORTIFY_STRING_H_
+#define __FORTIFY_INLINE extern __always_inline __attribute__((gnu_inline)) +#define __RENAME(x) __asm__(#x) + +void fortify_panic(const char *name) __noreturn __cold; +void __read_overflow(void) __compiletime_error("detected read beyond size of object (1st parameter)"); +void __read_overflow2(void) __compiletime_error("detected read beyond size of object (2nd parameter)"); +void __write_overflow(void) __compiletime_error("detected write beyond size of object (1st parameter)");
#if defined(CONFIG_KASAN_GENERIC) || defined(CONFIG_KASAN_SW_TAGS) extern void *__underlying_memchr(const void *p, int c, __kernel_size_t size) __RENAME(memchr); diff --git a/include/linux/string.h b/include/linux/string.h index b48d2d28e0b1..9473f81b9db2 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -249,15 +249,6 @@ static inline const char *kbasename(const char *path) return tail ? tail + 1 : path; }
-#define __FORTIFY_INLINE extern __always_inline __attribute__((gnu_inline)) -#define __RENAME(x) __asm__(#x) - -void fortify_panic(const char *name) __noreturn __cold; -void __read_overflow(void) __compiletime_error("detected read beyond size of object passed as 1st parameter"); -void __read_overflow2(void) __compiletime_error("detected read beyond size of object passed as 2nd parameter"); -void __read_overflow3(void) __compiletime_error("detected read beyond size of object passed as 3rd parameter"); -void __write_overflow(void) __compiletime_error("detected write beyond size of object passed as 1st parameter"); - #if !defined(__NO_FORTIFY) && defined(__OPTIMIZE__) && defined(CONFIG_FORTIFY_SOURCE) #include <linux/fortify-string.h> #endif diff --git a/lib/string_helpers.c b/lib/string_helpers.c index bde13612c25d..faa9d8e4e2c5 100644 --- a/lib/string_helpers.c +++ b/lib/string_helpers.c @@ -883,9 +883,11 @@ char *strreplace(char *s, char old, char new) } EXPORT_SYMBOL(strreplace);
+#ifdef CONFIG_FORTIFY_SOURCE void fortify_panic(const char *name) { pr_emerg("detected buffer overflow in %s\n", name); BUG(); } EXPORT_SYMBOL(fortify_panic); +#endif /* CONFIG_FORTIFY_SOURCE */
Hi.
Le mercredi 18 août 2021, 08:04:57 CEST Kees Cook a écrit :
When commit a28a6e860c6c ("string.h: move fortified functions definitions in a dedicated header.") moved the fortify-specific code, some helpers were left behind. Moves the remaining fortify-specific helpers into fortify-string.h so they're together where they're used. This requires that any FORTIFY helper function prototypes be conditionally built to avoid "no prototype" warnings. Additionally removes unused helpers.
Cc: Andrew Morton akpm@linux-foundation.org Cc: Francis Laniel laniel_francis@privacyrequired.com Cc: Daniel Axtens dja@axtens.net Cc: Vincenzo Frascino vincenzo.frascino@arm.com Cc: Andrey Konovalov andreyknvl@google.com Cc: Dan Williams dan.j.williams@intel.com Signed-off-by: Kees Cook keescook@chromium.org
include/linux/fortify-string.h | 7 +++++++ include/linux/string.h | 9 --------- lib/string_helpers.c | 2 ++ 3 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index c1be37437e77..7e67d02764db 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -2,6 +2,13 @@ #ifndef _LINUX_FORTIFY_STRING_H_ #define _LINUX_FORTIFY_STRING_H_
+#define __FORTIFY_INLINE extern __always_inline __attribute__((gnu_inline)) +#define __RENAME(x) __asm__(#x)
+void fortify_panic(const char *name) __noreturn __cold; +void __read_overflow(void) __compiletime_error("detected read beyond size of object (1st parameter)"); +void __read_overflow2(void) __compiletime_error("detected read beyond size of object (2nd parameter)"); +void __write_overflow(void) __compiletime_error("detected write beyond size of object (1st parameter)");
#if defined(CONFIG_KASAN_GENERIC) || defined(CONFIG_KASAN_SW_TAGS) extern void *__underlying_memchr(const void *p, int c, __kernel_size_t size) __RENAME(memchr); diff --git a/include/linux/string.h b/include/linux/string.h index b48d2d28e0b1..9473f81b9db2 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -249,15 +249,6 @@ static inline const char *kbasename(const char *path) return tail ? tail + 1 : path; }
-#define __FORTIFY_INLINE extern __always_inline __attribute__((gnu_inline)) -#define __RENAME(x) __asm__(#x)
-void fortify_panic(const char *name) __noreturn __cold; -void __read_overflow(void) __compiletime_error("detected read beyond size of object passed as 1st parameter"); -void __read_overflow2(void) __compiletime_error("detected read beyond size of object passed as 2nd parameter"); -void __read_overflow3(void) __compiletime_error("detected read beyond size of object passed as 3rd parameter"); -void __write_overflow(void) __compiletime_error("detected write beyond size of object passed as 1st parameter"); - #if !defined(__NO_FORTIFY) && defined(__OPTIMIZE__) && defined(CONFIG_FORTIFY_SOURCE) #include <linux/fortify-string.h> #endif diff --git a/lib/string_helpers.c b/lib/string_helpers.c index bde13612c25d..faa9d8e4e2c5 100644 --- a/lib/string_helpers.c +++ b/lib/string_helpers.c @@ -883,9 +883,11 @@ char *strreplace(char *s, char old, char new) } EXPORT_SYMBOL(strreplace);
+#ifdef CONFIG_FORTIFY_SOURCE void fortify_panic(const char *name) { pr_emerg("detected buffer overflow in %s\n", name); BUG(); } EXPORT_SYMBOL(fortify_panic); +#endif /* CONFIG_FORTIFY_SOURCE */
If I remember correctly, I let these helpers in string.h because I thought they could be used by code not related to fortify-string.h.
But you are right and I think it is better to have all the code related to one feature in the same place. I am happy to see the kernel is fortifying, and this contribution is good, so here is what I can give: Acked-by: Francis Laniel laniel_francis@privacyrequired.com
Best regards.
Clang has never correctly compiled the FORTIFY_SOURCE defenses due to a couple bugs:
Eliding inlines with matching __builtin_* names https://bugs.llvm.org/show_bug.cgi?id=50322
Incorrect __builtin_constant_p() of some globals https://bugs.llvm.org/show_bug.cgi?id=41459
In the process of making improvements to the FORTIFY_SOURCE defenses, the first (silent) bug (coincidentally) becomes worked around, but exposes the latter which breaks the build. As such, Clang must not be used with CONFIG_FORTIFY_SOURCE until at least latter bug is fixed (in Clang 13), and the fortify routines have been rearranged.
Update the Kconfig to reflect the reality of the current situation.
Signed-off-by: Kees Cook keescook@chromium.org --- security/Kconfig | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/security/Kconfig b/security/Kconfig index 0ced7fd33e4d..fe6c0395fa02 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -191,6 +191,9 @@ config HARDENED_USERCOPY_PAGESPAN config FORTIFY_SOURCE bool "Harden common str/mem functions against buffer overflows" depends on ARCH_HAS_FORTIFY_SOURCE + # https://bugs.llvm.org/show_bug.cgi?id=50322 + # https://bugs.llvm.org/show_bug.cgi?id=41459 + depends on !CC_IS_CLANG help Detect overflows of buffers in common string and memory functions where the compiler can determine and validate the buffer sizes.
The implementation for intra-object overflow in str*-family functions accidentally dropped compile-time write overflow checking in strcpy(), leaving it entirely to run-time. Add back the intended check.
Fixes: 6a39e62abbaf ("lib: string.h: detect intra-object overflow in fortified string functions") Cc: Daniel Axtens dja@axtens.net Cc: Francis Laniel laniel_francis@privacyrequired.com Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index 7e67d02764db..68bc5978d916 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -287,7 +287,10 @@ __FORTIFY_INLINE char *strcpy(char *p, const char *q) if (p_size == (size_t)-1 && q_size == (size_t)-1) return __underlying_strcpy(p, q); size = strlen(q) + 1; - /* test here to use the more stringent object size */ + /* Compile-time check for const size overflow. */ + if (__builtin_constant_p(size) && p_size < size) + __write_overflow(); + /* Run-time check for dynamic size overflow. */ if (p_size < size) fortify_panic(__func__); memcpy(p, q, size);
In order to have strlen() use fortified strnlen() internally, swap their positions in the source. Doing this as part of later changes makes review difficult, so reoroder it here; no code changes.
Cc: Francis Laniel laniel_francis@privacyrequired.com Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index 68bc5978d916..a3cb1d9aacce 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -56,6 +56,17 @@ __FORTIFY_INLINE char *strcat(char *p, const char *q) return p; }
+extern __kernel_size_t __real_strnlen(const char *, __kernel_size_t) __RENAME(strnlen); +__FORTIFY_INLINE __kernel_size_t strnlen(const char *p, __kernel_size_t maxlen) +{ + size_t p_size = __builtin_object_size(p, 1); + __kernel_size_t ret = __real_strnlen(p, maxlen < p_size ? maxlen : p_size); + + if (p_size <= ret && maxlen != ret) + fortify_panic(__func__); + return ret; +} + __FORTIFY_INLINE __kernel_size_t strlen(const char *p) { __kernel_size_t ret; @@ -71,17 +82,6 @@ __FORTIFY_INLINE __kernel_size_t strlen(const char *p) return ret; }
-extern __kernel_size_t __real_strnlen(const char *, __kernel_size_t) __RENAME(strnlen); -__FORTIFY_INLINE __kernel_size_t strnlen(const char *p, __kernel_size_t maxlen) -{ - size_t p_size = __builtin_object_size(p, 1); - __kernel_size_t ret = __real_strnlen(p, maxlen < p_size ? maxlen : p_size); - - if (p_size <= ret && maxlen != ret) - fortify_panic(__func__); - return ret; -} - /* defined after fortified strlen to reuse it */ extern size_t __real_strlcpy(char *, const char *, size_t) __RENAME(strlcpy); __FORTIFY_INLINE size_t strlcpy(char *p, const char *q, size_t size)
Under CONFIG_FORTIFY_SOURCE, it is possible for the compiler to perform strlen() and strnlen() at compile-time when the string size is known. This is required to support compile-time overflow checking in strlcpy().
Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 47 ++++++++++++++++++++++++++-------- 1 file changed, 36 insertions(+), 11 deletions(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index a3cb1d9aacce..e232a63fd826 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -10,6 +10,18 @@ void __read_overflow(void) __compiletime_error("detected read beyond size of obj void __read_overflow2(void) __compiletime_error("detected read beyond size of object (2nd parameter)"); void __write_overflow(void) __compiletime_error("detected write beyond size of object (1st parameter)");
+#define __compiletime_strlen(p) ({ \ + size_t ret = (size_t)-1; \ + size_t p_size = __builtin_object_size(p, 1); \ + if (p_size != (size_t)-1) { \ + size_t p_len = p_size - 1; \ + if (__builtin_constant_p(p[p_len]) && \ + p[p_len] == '\0') \ + ret = __builtin_strlen(p); \ + } \ + ret; \ +}) + #if defined(CONFIG_KASAN_GENERIC) || defined(CONFIG_KASAN_SW_TAGS) extern void *__underlying_memchr(const void *p, int c, __kernel_size_t size) __RENAME(memchr); extern int __underlying_memcmp(const void *p, const void *q, __kernel_size_t size) __RENAME(memcmp); @@ -60,21 +72,31 @@ extern __kernel_size_t __real_strnlen(const char *, __kernel_size_t) __RENAME(st __FORTIFY_INLINE __kernel_size_t strnlen(const char *p, __kernel_size_t maxlen) { size_t p_size = __builtin_object_size(p, 1); - __kernel_size_t ret = __real_strnlen(p, maxlen < p_size ? maxlen : p_size); + size_t p_len = __compiletime_strlen(p); + size_t ret; + + /* We can take compile-time actions when maxlen is const. */ + if (__builtin_constant_p(maxlen) && p_len != (size_t)-1) { + /* If p is const, we can use its compile-time-known len. */ + if (maxlen >= p_size) + return p_len; + }
+ /* Do no check characters beyond the end of p. */ + ret = __real_strnlen(p, maxlen < p_size ? maxlen : p_size); if (p_size <= ret && maxlen != ret) fortify_panic(__func__); return ret; }
+/* defined after fortified strnlen to reuse it. */ __FORTIFY_INLINE __kernel_size_t strlen(const char *p) { __kernel_size_t ret; size_t p_size = __builtin_object_size(p, 1);
- /* Work around gcc excess stack consumption issue */ - if (p_size == (size_t)-1 || - (__builtin_constant_p(p[p_size - 1]) && p[p_size - 1] == '\0')) + /* Give up if we don't know how large p is. */ + if (p_size == (size_t)-1) return __underlying_strlen(p); ret = strnlen(p, p_size); if (p_size <= ret) @@ -86,24 +108,27 @@ __FORTIFY_INLINE __kernel_size_t strlen(const char *p) extern size_t __real_strlcpy(char *, const char *, size_t) __RENAME(strlcpy); __FORTIFY_INLINE size_t strlcpy(char *p, const char *q, size_t size) { - size_t ret; size_t p_size = __builtin_object_size(p, 1); size_t q_size = __builtin_object_size(q, 1); + size_t q_len; /* Full count of source string length. */ + size_t len; /* Count of characters going into destination. */
if (p_size == (size_t)-1 && q_size == (size_t)-1) return __real_strlcpy(p, q, size); - ret = strlen(q); - if (size) { - size_t len = (ret >= size) ? size - 1 : ret; - - if (__builtin_constant_p(len) && len >= p_size) + q_len = strlen(q); + len = (q_len >= size) ? size - 1 : q_len; + if (__builtin_constant_p(size) && __builtin_constant_p(q_len) && size) { + /* Write size is always larger than destintation. */ + if (len >= p_size) __write_overflow(); + } + if (size) { if (len >= p_size) fortify_panic(__func__); __underlying_memcpy(p, q, len); p[len] = '\0'; } - return ret; + return q_len; }
/* defined after fortified strnlen to reuse it */
While the run-time testing of FORTIFY_SOURCE is already present in LKDTM, there is no testing of the expected compile-time detections. In preparation for correctly supporting FORTIFY_SOURCE under Clang, adding additional FORTIFY_SOURCE defenses, and making sure FORTIFY_SOURCE doesn't silently regress with GCC, introduce a build-time test suite that checks each expected compile-time failure condition.
As this is relatively backwards from standard build rules in the sense that a successful test is actually a compile _failure_, create a wrapper script to check for the correct errors, and wire it up as a dummy dependency to lib/string.o, collecting the results into a log file artifact.
Signed-off-by: Kees Cook keescook@chromium.org --- lib/.gitignore | 2 + lib/Makefile | 33 +++++++++++ lib/test_fortify/read_overflow-memchr.c | 5 ++ lib/test_fortify/read_overflow-memchr_inv.c | 5 ++ lib/test_fortify/read_overflow-memcmp.c | 5 ++ lib/test_fortify/read_overflow-memscan.c | 5 ++ lib/test_fortify/read_overflow2-memcmp.c | 5 ++ lib/test_fortify/read_overflow2-memcpy.c | 5 ++ lib/test_fortify/read_overflow2-memmove.c | 5 ++ lib/test_fortify/test_fortify.h | 35 +++++++++++ lib/test_fortify/write_overflow-memcpy.c | 5 ++ lib/test_fortify/write_overflow-memmove.c | 5 ++ lib/test_fortify/write_overflow-memset.c | 5 ++ lib/test_fortify/write_overflow-strcpy-lit.c | 5 ++ lib/test_fortify/write_overflow-strcpy.c | 5 ++ lib/test_fortify/write_overflow-strlcpy-src.c | 5 ++ lib/test_fortify/write_overflow-strlcpy.c | 5 ++ lib/test_fortify/write_overflow-strncpy-src.c | 5 ++ lib/test_fortify/write_overflow-strncpy.c | 5 ++ lib/test_fortify/write_overflow-strscpy.c | 5 ++ scripts/test_fortify.sh | 59 +++++++++++++++++++ 21 files changed, 214 insertions(+) create mode 100644 lib/test_fortify/read_overflow-memchr.c create mode 100644 lib/test_fortify/read_overflow-memchr_inv.c create mode 100644 lib/test_fortify/read_overflow-memcmp.c create mode 100644 lib/test_fortify/read_overflow-memscan.c create mode 100644 lib/test_fortify/read_overflow2-memcmp.c create mode 100644 lib/test_fortify/read_overflow2-memcpy.c create mode 100644 lib/test_fortify/read_overflow2-memmove.c create mode 100644 lib/test_fortify/test_fortify.h create mode 100644 lib/test_fortify/write_overflow-memcpy.c create mode 100644 lib/test_fortify/write_overflow-memmove.c create mode 100644 lib/test_fortify/write_overflow-memset.c create mode 100644 lib/test_fortify/write_overflow-strcpy-lit.c create mode 100644 lib/test_fortify/write_overflow-strcpy.c create mode 100644 lib/test_fortify/write_overflow-strlcpy-src.c create mode 100644 lib/test_fortify/write_overflow-strlcpy.c create mode 100644 lib/test_fortify/write_overflow-strncpy-src.c create mode 100644 lib/test_fortify/write_overflow-strncpy.c create mode 100644 lib/test_fortify/write_overflow-strscpy.c create mode 100644 scripts/test_fortify.sh
diff --git a/lib/.gitignore b/lib/.gitignore index 5e7fa54c4536..e5e217b8307b 100644 --- a/lib/.gitignore +++ b/lib/.gitignore @@ -4,3 +4,5 @@ /gen_crc32table /gen_crc64table /oid_registry_data.c +/test_fortify.log +/test_fortify/*.log diff --git a/lib/Makefile b/lib/Makefile index 5efd1b435a37..bd17c2bf43e1 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -360,3 +360,36 @@ obj-$(CONFIG_CMDLINE_KUNIT_TEST) += cmdline_kunit.o obj-$(CONFIG_SLUB_KUNIT_TEST) += slub_kunit.o
obj-$(CONFIG_GENERIC_LIB_DEVMEM_IS_ALLOWED) += devmem_is_allowed.o + +# FORTIFY_SOURCE compile-time behavior tests +TEST_FORTIFY_SRCS = $(wildcard $(srctree)/$(src)/test_fortify/*-*.c) +TEST_FORTIFY_LOGS = $(patsubst $(srctree)/$(src)/%.c, %.log, $(TEST_FORTIFY_SRCS)) +TEST_FORTIFY_LOG = test_fortify.log + +quiet_cmd_test_fortify = TEST $@ + cmd_test_fortify = $(CONFIG_SHELL) $(srctree)/scripts/test_fortify.sh \ + $< $@ "$(NM)" $(CC) $(c_flags) \ + $(call cc-disable-warning,fortify-source) + +targets += $(TEST_FORTIFY_LOGS) +clean-files += $(TEST_FORTIFY_LOGS) +clean-files += $(addsuffix .o, $(TEST_FORTIFY_LOGS)) +$(obj)/test_fortify/%.log: $(src)/test_fortify/%.c \ + $(src)/test_fortify/test_fortify.h \ + $(srctree)/include/linux/fortify-string.h \ + $(srctree)/scripts/test_fortify.sh \ + FORCE + $(call if_changed,test_fortify) + +quiet_cmd_gen_fortify_log = GEN $@ + cmd_gen_fortify_log = cat </dev/null $(filter-out FORCE,$^) 2>/dev/null > $@ || true + +targets += $(TEST_FORTIFY_LOG) +clean-files += $(TEST_FORTIFY_LOG) +$(obj)/$(TEST_FORTIFY_LOG): $(addprefix $(obj)/, $(TEST_FORTIFY_LOGS)) FORCE + $(call if_changed,gen_fortify_log) + +# Fake dependency to trigger the fortify tests. +ifeq ($(CONFIG_FORTIFY_SOURCE),y) +$(obj)/string.o: $(obj)/$(TEST_FORTIFY_LOG) +endif diff --git a/lib/test_fortify/read_overflow-memchr.c b/lib/test_fortify/read_overflow-memchr.c new file mode 100644 index 000000000000..2743084b32af --- /dev/null +++ b/lib/test_fortify/read_overflow-memchr.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memchr(small, 0x7A, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow-memchr_inv.c b/lib/test_fortify/read_overflow-memchr_inv.c new file mode 100644 index 000000000000..b26e1f1bc217 --- /dev/null +++ b/lib/test_fortify/read_overflow-memchr_inv.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memchr_inv(small, 0x7A, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow-memcmp.c b/lib/test_fortify/read_overflow-memcmp.c new file mode 100644 index 000000000000..d5d301ff64ef --- /dev/null +++ b/lib/test_fortify/read_overflow-memcmp.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcmp(small, large, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow-memscan.c b/lib/test_fortify/read_overflow-memscan.c new file mode 100644 index 000000000000..c1a97f2df0f0 --- /dev/null +++ b/lib/test_fortify/read_overflow-memscan.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memscan(small, 0x7A, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow2-memcmp.c b/lib/test_fortify/read_overflow2-memcmp.c new file mode 100644 index 000000000000..c6091e640f76 --- /dev/null +++ b/lib/test_fortify/read_overflow2-memcmp.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcmp(large, small, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow2-memcpy.c b/lib/test_fortify/read_overflow2-memcpy.c new file mode 100644 index 000000000000..07b62e56cf16 --- /dev/null +++ b/lib/test_fortify/read_overflow2-memcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcpy(large, instance.buf, sizeof(large)) + +#include "test_fortify.h" diff --git a/lib/test_fortify/read_overflow2-memmove.c b/lib/test_fortify/read_overflow2-memmove.c new file mode 100644 index 000000000000..34edfab040a3 --- /dev/null +++ b/lib/test_fortify/read_overflow2-memmove.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memmove(large, instance.buf, sizeof(large)) + +#include "test_fortify.h" diff --git a/lib/test_fortify/test_fortify.h b/lib/test_fortify/test_fortify.h new file mode 100644 index 000000000000..e1dfe64d79a3 --- /dev/null +++ b/lib/test_fortify/test_fortify.h @@ -0,0 +1,35 @@ +// SPDX-License-Identifier: GPL-2.0-only +#include <linux/kernel.h> +#include <linux/printk.h> +#include <linux/slab.h> +#include <linux/string.h> + +void do_fortify_tests(void); + +#define __BUF_SMALL 16 +#define __BUF_LARGE 32 +struct fortify_object { + int a; + char buf[__BUF_SMALL]; + int c; +}; + +#define LITERAL_SMALL "AAAAAAAAAAAAAAA" +#define LITERAL_LARGE "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA" +const char small_src[__BUF_SMALL] = LITERAL_SMALL; +const char large_src[__BUF_LARGE] = LITERAL_LARGE; + +char small[__BUF_SMALL]; +char large[__BUF_LARGE]; +struct fortify_object instance; +size_t size; + +void do_fortify_tests(void) +{ + /* Normal initializations. */ + memset(&instance, 0x32, sizeof(instance)); + memset(small, 0xA5, sizeof(small)); + memset(large, 0x5A, sizeof(large)); + + TEST; +} diff --git a/lib/test_fortify/write_overflow-memcpy.c b/lib/test_fortify/write_overflow-memcpy.c new file mode 100644 index 000000000000..3b3984e428fb --- /dev/null +++ b/lib/test_fortify/write_overflow-memcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcpy(instance.buf, large_src, sizeof(large_src)) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-memmove.c b/lib/test_fortify/write_overflow-memmove.c new file mode 100644 index 000000000000..640437c3b3e0 --- /dev/null +++ b/lib/test_fortify/write_overflow-memmove.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memmove(instance.buf, large_src, sizeof(large_src)) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-memset.c b/lib/test_fortify/write_overflow-memset.c new file mode 100644 index 000000000000..36e34908cfb3 --- /dev/null +++ b/lib/test_fortify/write_overflow-memset.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memset(instance.buf, 0x5A, sizeof(large_src)) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strcpy-lit.c b/lib/test_fortify/write_overflow-strcpy-lit.c new file mode 100644 index 000000000000..51effb3e50f9 --- /dev/null +++ b/lib/test_fortify/write_overflow-strcpy-lit.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strcpy(small, LITERAL_LARGE) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strcpy.c b/lib/test_fortify/write_overflow-strcpy.c new file mode 100644 index 000000000000..84f1c56a64c8 --- /dev/null +++ b/lib/test_fortify/write_overflow-strcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strcpy(small, large_src) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strlcpy-src.c b/lib/test_fortify/write_overflow-strlcpy-src.c new file mode 100644 index 000000000000..91bf83ebd34a --- /dev/null +++ b/lib/test_fortify/write_overflow-strlcpy-src.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strlcpy(small, large_src, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strlcpy.c b/lib/test_fortify/write_overflow-strlcpy.c new file mode 100644 index 000000000000..1883db7c0cd6 --- /dev/null +++ b/lib/test_fortify/write_overflow-strlcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strlcpy(instance.buf, large_src, sizeof(instance.buf) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strncpy-src.c b/lib/test_fortify/write_overflow-strncpy-src.c new file mode 100644 index 000000000000..8dcfb8c788dd --- /dev/null +++ b/lib/test_fortify/write_overflow-strncpy-src.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strncpy(small, large_src, sizeof(small) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strncpy.c b/lib/test_fortify/write_overflow-strncpy.c new file mode 100644 index 000000000000..b85f079c815d --- /dev/null +++ b/lib/test_fortify/write_overflow-strncpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strncpy(instance.buf, large_src, sizeof(instance.buf) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow-strscpy.c b/lib/test_fortify/write_overflow-strscpy.c new file mode 100644 index 000000000000..38feddf377dc --- /dev/null +++ b/lib/test_fortify/write_overflow-strscpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + strscpy(instance.buf, large_src, sizeof(instance.buf) + 1) + +#include "test_fortify.h" diff --git a/scripts/test_fortify.sh b/scripts/test_fortify.sh new file mode 100644 index 000000000000..a6d63871738b --- /dev/null +++ b/scripts/test_fortify.sh @@ -0,0 +1,59 @@ +#!/bin/sh +# SPDX-License-Identifier: GPL-2.0-only +set -e + +# Argument 1: Source file to build. +IN="$1" +shift +# Extract just the filename for error messages below. +FILE="${IN##*/}" +# Extract the function name for error messages below. +FUNC="${FILE#*-}" +FUNC="${FUNC%%-*}" +FUNC="${FUNC%%.*}" +# Extract the symbol to test for in build/symbol test below. +WANT="__${FILE%%-*}" + +# Argument 2: Where to write the build log. +OUT="$1" +shift +TMP="${OUT}.tmp" + +# Argument 3: Path to "nm" tool. +NM="$1" +shift + +# Remaining arguments are: $(CC) $(c_flags) + +# Clean up temporary file at exit. +__cleanup() { + rm -f "$TMP" +} +trap __cleanup EXIT + +status= +# Attempt to build a source that is expected to fail with a specific warning. +if "$@" -Werror -c "$IN" -o "$OUT".o 2> "$TMP" ; then + # If the build succeeds, either the test has failed or the the + # warning may only happen at link time (Clang). In that case, + # make sure the expected symbol is unresolved in the symbol list. + # If so, FORTIFY is working for this case. + if ! $NM -A "$OUT".o | grep -m1 "\bU ${WANT}$" >>"$TMP" ; then + status="warning: unsafe ${FUNC}() usage lacked '$WANT' symbol in $IN" + fi +else + # If the build failed, check for the warning in the stderr (gcc). + if ! grep -q -m1 "error:.*\b${WANT}'" "$TMP" ; then + status="warning: unsafe ${FUNC}() usage lacked '$WANT' warning in $IN" + fi +fi + +if [ -n "$status" ]; then + # Report on failure results, including compilation warnings. + echo "$status" | tee "$OUT" >&2 + cat "$TMP" | tee -a "$OUT" >&2 +else + # Report on good results, and save any compilation output to log. + echo "ok: unsafe ${FUNC}() usage correctly detected with '$WANT' in $IN" >"$OUT" + cat "$TMP" >>"$OUT" +fi
Before changing anything about memcpy(), memmove(), and memset(), add run-time tests to check basic behaviors for any regressions.
Signed-off-by: Kees Cook keescook@chromium.org --- lib/Kconfig.debug | 7 ++ lib/Makefile | 1 + lib/test_memcpy.c | 264 ++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 272 insertions(+) create mode 100644 lib/test_memcpy.c
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 139d362daa32..db2e8ffa6049 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2476,6 +2476,13 @@ config RATIONAL_KUNIT_TEST
If unsure, say N.
+config MEMCPY_KUNIT_TEST + tristate "Test memcpy(), memmove(), and memset() functions at runtime" if !KUNIT_ALL_TESTS + depends on KUNIT + default KUNIT_ALL_TESTS + help + Builds unit tests for memcpy(), memmove(), and memset() functions. + config TEST_UDELAY tristate "udelay test driver" help diff --git a/lib/Makefile b/lib/Makefile index bd17c2bf43e1..8a4c8bdb38a2 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -77,6 +77,7 @@ obj-$(CONFIG_TEST_MIN_HEAP) += test_min_heap.o obj-$(CONFIG_TEST_LKM) += test_module.o obj-$(CONFIG_TEST_VMALLOC) += test_vmalloc.o obj-$(CONFIG_TEST_OVERFLOW) += test_overflow.o +obj-$(CONFIG_TEST_MEMCPY) += test_memcpy.o obj-$(CONFIG_TEST_RHASHTABLE) += test_rhashtable.o obj-$(CONFIG_TEST_SORT) += test_sort.o obj-$(CONFIG_TEST_USER_COPY) += test_user_copy.o diff --git a/lib/test_memcpy.c b/lib/test_memcpy.c new file mode 100644 index 000000000000..be192b8e82b7 --- /dev/null +++ b/lib/test_memcpy.c @@ -0,0 +1,264 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Test cases for memcpy(), memmove(), and memset(). + */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <kunit/test.h> +#include <linux/device.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/overflow.h> +#include <linux/slab.h> +#include <linux/types.h> +#include <linux/vmalloc.h> + +struct some_bytes { + union { + u8 data[32]; + struct { + u32 one; + u16 two; + u8 three; + /* 1 byte hole */ + u32 four[4]; + }; + }; +}; + +#define check(instance, v) do { \ + int i; \ + BUILD_BUG_ON(sizeof(instance.data) != 32); \ + for (i = 0; i < sizeof(instance.data); i++) { \ + KUNIT_ASSERT_EQ_MSG(test, instance.data[i], v, \ + "line %d: '%s' not initialized to 0x%02x @ %d (saw 0x%02x)\n", \ + __LINE__, #instance, v, i, instance.data[i]); \ + } \ +} while (0) + +#define compare(name, one, two) do { \ + int i; \ + BUILD_BUG_ON(sizeof(one) != sizeof(two)); \ + for (i = 0; i < sizeof(one); i++) { \ + KUNIT_EXPECT_EQ_MSG(test, one.data[i], two.data[i], \ + "line %d: %s.data[%d] (0x%02x) != %s.data[%d] (0x%02x)\n", \ + __LINE__, #one, i, one.data[i], #two, i, two.data[i]); \ + } \ + kunit_info(test, "ok: " TEST_OP "() " name "\n"); \ +} while (0) + +static void memcpy_test(struct kunit *test) +{ +#define TEST_OP "memcpy" + struct some_bytes control = { + .data = { 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + }, + }; + struct some_bytes zero = { }; + struct some_bytes middle = { + .data = { 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + }, + }; + struct some_bytes three = { + .data = { 0x00, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x00, 0x00, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, 0x20, + }, + }; + struct some_bytes dest = { }; + int count; + u8 *ptr; + + /* Verify static initializers. */ + check(control, 0x20); + check(zero, 0); + compare("static initializers", dest, zero); + + /* Verify assignment. */ + dest = control; + compare("direct assignment", dest, control); + + /* Verify complete overwrite. */ + memcpy(dest.data, zero.data, sizeof(dest.data)); + compare("complete overwrite", dest, zero); + + /* Verify middle overwrite. */ + dest = control; + memcpy(dest.data + 12, zero.data, 7); + compare("middle overwrite", dest, middle); + + /* Verify argument side-effects aren't repeated. */ + dest = control; + ptr = dest.data; + count = 1; + memcpy(ptr++, zero.data, count++); + ptr += 8; + memcpy(ptr++, zero.data, count++); + compare("argument side-effects", dest, three); +#undef TEST_OP +} + +static void memmove_test(struct kunit *test) +{ +#define TEST_OP "memmove" + struct some_bytes control = { + .data = { 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + }, + }; + struct some_bytes zero = { }; + struct some_bytes middle = { + .data = { 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + }, + }; + struct some_bytes five = { + .data = { 0x00, 0x00, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x00, 0x00, 0x00, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + }, + }; + struct some_bytes overlap = { + .data = { 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + }, + }; + struct some_bytes overlap_expected = { + .data = { 0x00, 0x01, 0x00, 0x01, 0x02, 0x03, 0x04, 0x07, + 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, 0x99, + }, + }; + struct some_bytes dest = { }; + int count; + u8 *ptr; + + /* Verify static initializers. */ + check(control, 0x99); + check(zero, 0); + compare("static initializers", zero, dest); + + /* Verify assignment. */ + dest = control; + compare("direct assignment", dest, control); + + /* Verify complete overwrite. */ + memmove(dest.data, zero.data, sizeof(dest.data)); + compare("complete overwrite", dest, zero); + + /* Verify middle overwrite. */ + dest = control; + memmove(dest.data + 12, zero.data, 7); + compare("middle overwrite", dest, middle); + + /* Verify argument side-effects aren't repeated. */ + dest = control; + ptr = dest.data; + count = 2; + memmove(ptr++, zero.data, count++); + ptr += 9; + memmove(ptr++, zero.data, count++); + compare("argument side-effects", dest, five); + + /* Verify overlapping overwrite is correct. */ + ptr = &overlap.data[2]; + memmove(ptr, overlap.data, 5); + compare("overlapping write", overlap, overlap_expected); +#undef TEST_OP +} + +static void memset_test(struct kunit *test) +{ +#define TEST_OP "memset" + struct some_bytes control = { + .data = { 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + }, + }; + struct some_bytes complete = { + .data = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, + }, + }; + struct some_bytes middle = { + .data = { 0x30, 0x30, 0x30, 0x30, 0x31, 0x31, 0x31, 0x31, + 0x31, 0x31, 0x31, 0x31, 0x31, 0x31, 0x31, 0x31, + 0x31, 0x31, 0x31, 0x31, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + }, + }; + struct some_bytes three = { + .data = { 0x60, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x61, 0x61, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + }, + }; + struct some_bytes dest = { }; + int count, value; + u8 *ptr; + + /* Verify static initializers. */ + check(control, 0x30); + check(dest, 0); + + /* Verify assignment. */ + dest = control; + compare("direct assignment", dest, control); + + /* Verify complete overwrite. */ + memset(dest.data, 0xff, sizeof(dest.data)); + compare("complete overwrite", dest, complete); + + /* Verify middle overwrite. */ + dest = control; + memset(dest.data + 4, 0x31, 16); + compare("middle overwrite", dest, middle); + + /* Verify argument side-effects aren't repeated. */ + dest = control; + ptr = dest.data; + value = 0x60; + count = 1; + memset(ptr++, value++, count++); + ptr += 8; + memset(ptr++, value++, count++); + compare("argument side-effects", dest, three); +#undef TEST_OP +} + +static struct kunit_case memcpy_test_cases[] = { + KUNIT_CASE(memset_test), + KUNIT_CASE(memcpy_test), + KUNIT_CASE(memmove_test), + {} +}; + +static struct kunit_suite memcpy_test_suite = { + .name = "memcpy-test", + .test_cases = memcpy_test_cases, +}; +kunit_test_suite(memcpy_test_suite); + +MODULE_LICENSE("GPL");
memcpy() is dead; long live memcpy()
tl;dr: In order to eliminate a large class of common buffer overflow flaws that continue to persist in the kernel, have memcpy() (under CONFIG_FORTIFY_SOURCE) perform bounds checking of the destination struct member when they have a known size. This would have caught all of the memcpy()-related buffer write overflow flaws identified in at least the last three years.
Background and analysis:
While stack-based buffer overflow flaws are largely mitigated by stack canaries (and similar) features, heap-based buffer overflow flaws continue to regularly appear in the kernel. Many classes of heap buffer overflows are mitigated by FORTIFY_SOURCE when using the strcpy() family of functions, but a significant number remain exposed through the memcpy() family of functions.
At its core, FORTIFY_SOURCE uses the compiler's __builtin_object_size() internal[0] to determine the available size at a target address based on the compile-time known structure layout details. It operates in two modes: outer bounds (0) and inner bounds (1). In mode 0, the size of the enclosing structure is used. In mode 1, the size of the specific field is used. For example:
struct object { u16 scalar1; /* 2 bytes */ char array[6]; /* 6 bytes */ u64 scalar2; /* 8 bytes */ u32 scalar3; /* 4 bytes */ u32 scalar4; /* 4 bytes */ } instance;
__builtin_object_size(instance.array, 0) == 22, since the remaining size of the enclosing structure starting from "array" is 22 bytes (6 + 8 + 4 + 4).
__builtin_object_size(instance.array, 1) == 6, since the remaining size of the specific field "array" is 6 bytes.
The initial implementation of FORTIFY_SOURCE used mode 0 because there were many cases of both strcpy() and memcpy() functions being used to write (or read) across multiple fields in a structure. For example, it would catch this, which is writing 2 bytes beyond the end of "instance":
memcpy(&instance.array, data, 24);
While this didn't protect against overwriting adjacent fields in a given structure, it would at least stop overflows from reaching beyond the end of the structure into neighboring memory, and provided a meaningful mitigation of a subset of buffer overflow flaws. However, many desirable targets remain within the enclosing structure (for example function pointers).
As it happened, there were very few cases of strcpy() family functions intentionally writing beyond the end of a string buffer. Once all known cases were removed from the kernel, the strcpy() family was tightened[1] to use mode 1, providing greater mitigation coverage.
What remains is switching memcpy() to mode 1 as well, but making the switch is much more difficult because of how frustrating it can be to find existing "normal" uses of memcpy() that expect to write (or read) across multiple fields. The root cause of the problem is that the C language lacks a common pattern to indicate the intent of an author's use of memcpy(), and is further complicated by the available compile-time and run-time mitigation behaviors.
The FORTIFY_SOURCE mitigation comes in two halves: the compile-time half, when both the buffer size _and_ the length of the copy is known, and the run-time half, when only the buffer size is known. If neither size is known, there is no bounds checking possible. At compile-time when the compiler sees that a length will always exceed a known buffer size, a warning can be deterministically emitted. For the run-time half, the length is tested against the known size of the buffer, and the overflowing operation is detected. (The performance overhead for these tests is virtually zero.)
It is relatively easy to find compile-time false-positives since a warning is always generated. Fixing the false positives, however, can be very time-consuming as there are hundreds of instances. While it's possible some over-read conditions could lead to kernel memory exposures, the bulk of the risk comes from the run-time flaws where the length of a write may end up being attacker-controlled and lead to an overflow.
Many of the compile-time false-positives take a form similar to this:
memcpy(&instance.scalar2, data, sizeof(instance.scalar2) + sizeof(instance.scalar3));
and the run-time ones are similar, but lack a constant expression for the size of the copy:
memcpy(instance.array, data, length);
The former is meant to cover multiple fields (though its style has been frowned upon more recently), but has been technically legal. Both lack any expressivity in the C language about the author's _intent_ in a way that a compiler can check when the length isn't known at compile time. A comment doesn't work well because what's needed is something a compiler can directly reason about. Is a given memcpy() call expected to overflow into neighbors? Is it not? By using the new struct_group() macro, this intent can be much more easily encoded.
It is not as easy to find the run-time false-positives since the code path to exercise a seemingly out-of-bounds condition that is actually expected may not be trivially reachable. Tightening the restrictions to block an operation for a false positive will either potentially create a greater flaw (if a copy is truncated by the mitigation), or destabilize the kernel (e.g. with a BUG()), making things completely useless for the end user.
As a result, tightening the memcpy() restriction (when there is a reasonable level of uncertainty of the number of false positives), needs to first WARN() with no truncation. (Though any sufficiently paranoid end-user can always opt to set the panic_on_warn=1 sysctl.) Once enough development time has passed, the mitigation can be further intensified.
Given the potential frustrations of weeding out all the false positives when tightening the run-time checks, it is reasonable to wonder if these changes would actually add meaningful protection. Looking at just the last three years, there are 23 identified flaws with a CVE that mention "buffer overflow", and 11 are memcpy()-related buffer overflows.
(For the remaining 12: 7 are array index overflows that would be mitigated by systems built with CONFIG_UBSAN_BOUNDS=y: CVE-2019-0145, CVE-2019-14835, CVE-2019-14896, CVE-2019-14897, CVE-2019-14901, CVE-2019-17666, CVE-2021-28952. 2 are miscalculated allocation sizes which could be mitigated with memory tagging: CVE-2019-16746, CVE-2019-2181. 1 is an iovec buffer bug maybe mitigated by memory tagging: CVE-2020-10742. 1 is a type confusion bug mitigated by stack canaries: CVE-2020-10942. 1 is a string handling logic bug with no mitigation I'm aware of: CVE-2021-28972.)
At my last count on an x86_64 allmodconfig build, there are 25,018 calls to memcpy(). With callers instrumented to report all places where the buffer size is known but the length remains unknown (i.e. a run-time bounds check is added), we can count how many new run-time bounds checks are added when the destination and source arguments of memcpy() are changed to use "mode 1" bounds checking: 1540. In addition, there were 146 new compile-time warnings to evaluate and fix.
With this it's also possible to compare the places where the known 11 memcpy() flaw overflows happened against the resulting list of potential new bounds checks, as a measure of potential efficacy of the tightened mitigation. Much to my surprise, horror, and delight, all 11 flaws would have been detected by the newly added run-time bounds checks, making this a distinctly clear mitigation improvement: 100% coverage for memcpy() flaws, with a possible 2 orders of magnitude gain in coverage over existing but undiscovered run-time dynamic length flaws, against only 6% of all callers maybe gaining a false positive run-time check, with fewer than 150 new compile-time instances needing evaluation.
Specifically these would have been mitigated: CVE-2020-24490 https://git.kernel.org/linus/a2ec905d1e160a33b2e210e45ad30445ef26ce0e CVE-2020-12654 https://git.kernel.org/linus/3a9b153c5591548612c3955c9600a98150c81875 CVE-2020-12653 https://git.kernel.org/linus/b70261a288ea4d2f4ac7cd04be08a9f0f2de4f4d CVE-2019-14895 https://git.kernel.org/linus/3d94a4a8373bf5f45cf5f939e88b8354dbf2311b CVE-2019-14816 https://git.kernel.org/linus/7caac62ed598a196d6ddf8d9c121e12e082cac3a CVE-2019-14815 https://git.kernel.org/linus/7caac62ed598a196d6ddf8d9c121e12e082cac3a CVE-2019-14814 https://git.kernel.org/linus/7caac62ed598a196d6ddf8d9c121e12e082cac3a CVE-2019-10126 https://git.kernel.org/linus/69ae4f6aac1578575126319d3f55550e7e440449 CVE-2019-9500 https://git.kernel.org/linus/1b5e2423164b3670e8bc9174e4762d297990deff no-CVE-yet https://git.kernel.org/linus/130f634da1af649205f4a3dd86cbe5c126b57914 no-CVE-yet https://git.kernel.org/linus/d10a87a3535cce2b890897914f5d0d83df669c63
To accelerate the review of potential run-time false positives, it's also worth noting that it is possible to partially automate checking by examining the memcpy() buffer argument to check for the destination struct member having a neighboring array member. It is reasonable to expect that the vast majority of run-time false positives would look like the already evaluated and fixed compile-time false positives, where the most common pattern is neighboring arrays. (And, FWIW, several of the compile-time fixes were actual bugs.)
Implementation:
Tighten the memcpy() destination buffer size checking to use the actual ("mode 1") target buffer size as the bounds check instead of their enclosing structure's ("mode 0") size. Use a common inline for memcpy() (and memmove() in a following patch), since all the tests are the same. All new cross-field memcpy() uses must use the struct_group() macro or similar to target a specific range of fields, so that FORTIFY_SOURCE can reason about the size and safety of the copy.
For now, cross-member "mode 1" read detection at compile-time will be limited to W=1 builds, since it is, unfortunately, very common. As the priority is solving write overflows, read overflows can be part of the next phase.
For run-time, the "mode 0" size checking and mitigation is left unchanged, with "mode 1" to be added in stages. In this patch, no new run-time checks are added. Future patches will first bounds-check writes, and only perform a WARN() for now. This way any missed run-time false positives can be flushed out over the coming several development cycles, but system builders who have tested their workloads to be WARN()-free can enable the panic_on_warn=1 sysctl to immediately gain a mitigation against this class of buffer overflows. Once that is under way, run-time bounds-checking of reads can be similarly added.
Related classes of flaws that remain unmitigated:
- memcpy() with raw pointers (e.g. void *, char *, etc) have no good mitigation beyond memory tagging (and even that would only protect against inter-object overflow, not intra-object neighboring field overflows). Some kind of "fat pointer" solution is likely needed to gain proper size-of-buffer awareness.
- type confusion where a higher level type's allocation size does not match the resulting cast type eventually passed to a deeper memcpy() call where the compiler cannot see the true type. In theory, greater static analysis could catch these.
[0] https://gcc.gnu.org/onlinedocs/gcc/Object-Size-Checking.html [1] https://git.kernel.org/linus/6a39e62abbafd1d58d1722f40c7d26ef379c6a2f
Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 109 ++++++++++++++++-- include/linux/string.h | 5 +- lib/Makefile | 3 +- lib/string_helpers.c | 6 + .../read_overflow2_field-memcpy.c | 5 + .../write_overflow_field-memcpy.c | 5 + 6 files changed, 118 insertions(+), 15 deletions(-) create mode 100644 lib/test_fortify/read_overflow2_field-memcpy.c create mode 100644 lib/test_fortify/write_overflow_field-memcpy.c
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index e232a63fd826..25943442f532 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -8,7 +8,9 @@ void fortify_panic(const char *name) __noreturn __cold; void __read_overflow(void) __compiletime_error("detected read beyond size of object (1st parameter)"); void __read_overflow2(void) __compiletime_error("detected read beyond size of object (2nd parameter)"); +void __read_overflow2_field(size_t avail, size_t wanted) __compiletime_warning("detected read beyond size of field (2nd parameter); maybe use struct_group()?"); void __write_overflow(void) __compiletime_error("detected write beyond size of object (1st parameter)"); +void __write_overflow_field(size_t avail, size_t wanted) __compiletime_warning("detected write beyond size of field (1st parameter); maybe use struct_group()?");
#define __compiletime_strlen(p) ({ \ size_t ret = (size_t)-1; \ @@ -207,22 +209,105 @@ __FORTIFY_INLINE void *memset(void *p, int c, __kernel_size_t size) return __underlying_memset(p, c, size); }
-__FORTIFY_INLINE void *memcpy(void *p, const void *q, __kernel_size_t size) +/* + * To make sure the compiler can enforce protection against buffer overflows, + * memcpy(), memmove(), and memset() must not be used beyond individual + * struct members. If you need to copy across multiple members, please use + * struct_group() to create a named mirror of an anonymous struct union. + * (e.g. see struct sk_buff.) + * + * Mitigation coverage + * Bounds checking at: + * +-------+-------+-------+-------+ + * | Compile time | Run time | + * memcpy() argument sizes: | write | read | write | read | + * +-------+-------+-------+-------+ + * memcpy(known, known, constant) | y | y | n/a | n/a | + * memcpy(unknown, known, constant) | n | y | V | n/a | + * memcpy(known, unknown, constant) | y | n | n/a | V | + * memcpy(unknown, unknown, constant) | n | n | V | V | + * memcpy(known, known, dynamic) | n | n | b | B | + * memcpy(unknown, known, dynamic) | n | n | V | B | + * memcpy(known, unknown, dynamic) | n | n | b | V | + * memcpy(unknown, unknown, dynamic) | n | n | V | V | + * +-------+-------+-------+-------+ + * + * y = deterministic compile-time bounds checking + * n = cannot do deterministic compile-time bounds checking + * n/a = no run-time bounds checking needed since compile-time deterministic + * b = perform run-time bounds checking + * B = can perform run-time bounds checking, but current unenforced + * V = vulnerable to run-time overflow + * + */ +__FORTIFY_INLINE void fortify_memcpy_chk(__kernel_size_t size, + const size_t p_size, + const size_t q_size, + const size_t p_size_field, + const size_t q_size_field, + const char *func) { - size_t p_size = __builtin_object_size(p, 0); - size_t q_size = __builtin_object_size(q, 0); - if (__builtin_constant_p(size)) { - if (p_size < size) + /* + * Length argument is a constant expression, so we + * can perform compile-time bounds checking where + * buffer sizes are known. + */ + + /* Error when size is larger than enclosing struct. */ + if (p_size > p_size_field && p_size < size) __write_overflow(); - if (q_size < size) + if (q_size > q_size_field && q_size < size) __read_overflow2(); + + /* Warn when write size argument larger than dest field. */ + if (p_size_field < size) + __write_overflow_field(p_size_field, size); + /* + * Warn for source field over-read when building with W=1 + * or when an over-write happened, so both can be fixed at + * the same time. + */ + if ((IS_ENABLED(KBUILD_EXTRA_WARN1) || p_size_field < size) && + q_size_field < size) + __read_overflow2_field(q_size_field, size); } - if (p_size < size || q_size < size) - fortify_panic(__func__); - return __underlying_memcpy(p, q, size); + /* + * At this point, length argument may not be a constant expression, + * so run-time bounds checking can be done where buffer sizes are + * known. (This is not an "else" because the above checks may only + * be compile-time warnings, and we want to still warn for run-time + * overflows.) + */ + + /* + * Always stop accesses beyond the struct that contains the + * field, when the buffer's remaining size is known. + * (The -1 test is to optimize away checks where the buffer + * lengths are unknown.) + */ + if ((p_size != (size_t)(-1) && p_size < size) || + (q_size != (size_t)(-1) && q_size < size)) + fortify_panic(func); }
+#define __fortify_memcpy_chk(p, q, size, p_size, q_size, \ + p_size_field, q_size_field, op) ({ \ + size_t __fortify_size = (size_t)(size); \ + fortify_memcpy_chk(__fortify_size, p_size, q_size, \ + p_size_field, q_size_field, #op); \ + __underlying_##op(p, q, __fortify_size); \ +}) + +/* + * __builtin_object_size() must be captured here to avoid evaluating argument + * side-effects further into the macro layers. + */ +#define memcpy(p, q, s) __fortify_memcpy_chk(p, q, s, \ + __builtin_object_size(p, 0), __builtin_object_size(q, 0), \ + __builtin_object_size(p, 1), __builtin_object_size(q, 1), \ + memcpy) + __FORTIFY_INLINE void *memmove(void *p, const void *q, __kernel_size_t size) { size_t p_size = __builtin_object_size(p, 0); @@ -302,13 +387,14 @@ __FORTIFY_INLINE void *kmemdup(const void *p, size_t size, gfp_t gfp) return __real_kmemdup(p, size, gfp); }
-/* defined after fortified strlen and memcpy to reuse them */ +/* Defined after fortified strlen to reuse it. */ __FORTIFY_INLINE char *strcpy(char *p, const char *q) { size_t p_size = __builtin_object_size(p, 1); size_t q_size = __builtin_object_size(q, 1); size_t size;
+ /* If neither buffer size is known, immediately give up. */ if (p_size == (size_t)-1 && q_size == (size_t)-1) return __underlying_strcpy(p, q); size = strlen(q) + 1; @@ -318,14 +404,13 @@ __FORTIFY_INLINE char *strcpy(char *p, const char *q) /* Run-time check for dynamic size overflow. */ if (p_size < size) fortify_panic(__func__); - memcpy(p, q, size); + __underlying_memcpy(p, q, size); return p; }
/* Don't use these outside the FORITFY_SOURCE implementation */ #undef __underlying_memchr #undef __underlying_memcmp -#undef __underlying_memcpy #undef __underlying_memmove #undef __underlying_memset #undef __underlying_strcat diff --git a/include/linux/string.h b/include/linux/string.h index 9473f81b9db2..cbe889e404e2 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -261,8 +261,9 @@ static inline const char *kbasename(const char *path) * @count: The number of bytes to copy * @pad: Character to use for padding if space is left in destination. */ -static inline void memcpy_and_pad(void *dest, size_t dest_len, - const void *src, size_t count, int pad) +static __always_inline void memcpy_and_pad(void *dest, size_t dest_len, + const void *src, size_t count, + int pad) { if (dest_len > count) { memcpy(dest, src, count); diff --git a/lib/Makefile b/lib/Makefile index 8a4c8bdb38a2..ff80f09947c2 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -370,7 +370,8 @@ TEST_FORTIFY_LOG = test_fortify.log quiet_cmd_test_fortify = TEST $@ cmd_test_fortify = $(CONFIG_SHELL) $(srctree)/scripts/test_fortify.sh \ $< $@ "$(NM)" $(CC) $(c_flags) \ - $(call cc-disable-warning,fortify-source) + $(call cc-disable-warning,fortify-source) \ + -DKBUILD_EXTRA_WARN1
targets += $(TEST_FORTIFY_LOGS) clean-files += $(TEST_FORTIFY_LOGS) diff --git a/lib/string_helpers.c b/lib/string_helpers.c index faa9d8e4e2c5..961636c120b1 100644 --- a/lib/string_helpers.c +++ b/lib/string_helpers.c @@ -884,6 +884,12 @@ char *strreplace(char *s, char old, char new) EXPORT_SYMBOL(strreplace);
#ifdef CONFIG_FORTIFY_SOURCE +/* These are placeholders for fortify compile-time warnings. */ +void __read_overflow2_field(size_t avail, size_t wanted) { } +EXPORT_SYMBOL(__read_overflow2_field); +void __write_overflow_field(size_t avail, size_t wanted) { } +EXPORT_SYMBOL(__write_overflow_field); + void fortify_panic(const char *name) { pr_emerg("detected buffer overflow in %s\n", name); diff --git a/lib/test_fortify/read_overflow2_field-memcpy.c b/lib/test_fortify/read_overflow2_field-memcpy.c new file mode 100644 index 000000000000..de9569266223 --- /dev/null +++ b/lib/test_fortify/read_overflow2_field-memcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcpy(large, instance.buf, sizeof(instance.buf) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow_field-memcpy.c b/lib/test_fortify/write_overflow_field-memcpy.c new file mode 100644 index 000000000000..28cc81058dd3 --- /dev/null +++ b/lib/test_fortify/write_overflow_field-memcpy.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memcpy(instance.buf, large, sizeof(instance.buf) + 1) + +#include "test_fortify.h"
As done for memcpy(), also update memmove() to use the same tightened compile-time checks under CONFIG_FORTIFY_SOURCE.
Signed-off-by: Kees Cook keescook@chromium.org --- arch/x86/boot/compressed/misc.c | 3 ++- arch/x86/lib/memcpy_32.c | 1 + include/linux/fortify-string.h | 21 ++++--------------- .../read_overflow2_field-memmove.c | 5 +++++ .../write_overflow_field-memmove.c | 5 +++++ 5 files changed, 17 insertions(+), 18 deletions(-) create mode 100644 lib/test_fortify/read_overflow2_field-memmove.c create mode 100644 lib/test_fortify/write_overflow_field-memmove.c
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index 743f13ea25c1..83ff4354970e 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -34,10 +34,11 @@ * try to define their own functions if these are not defined as macros. */ #define memzero(s, n) memset((s), 0, (n)) +#ifndef memmove #define memmove memmove - /* Functions used by the included decompressor code below. */ void *memmove(void *dest, const void *src, size_t n); +#endif
/* * This is set up by the setup-routine at boot-time diff --git a/arch/x86/lib/memcpy_32.c b/arch/x86/lib/memcpy_32.c index e565d1c9019e..f19b7fd07f04 100644 --- a/arch/x86/lib/memcpy_32.c +++ b/arch/x86/lib/memcpy_32.c @@ -4,6 +4,7 @@
#undef memcpy #undef memset +#undef memmove
__visible void *memcpy(void *to, const void *from, size_t n) { diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index 25943442f532..0120d463ba33 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -307,22 +307,10 @@ __FORTIFY_INLINE void fortify_memcpy_chk(__kernel_size_t size, __builtin_object_size(p, 0), __builtin_object_size(q, 0), \ __builtin_object_size(p, 1), __builtin_object_size(q, 1), \ memcpy) - -__FORTIFY_INLINE void *memmove(void *p, const void *q, __kernel_size_t size) -{ - size_t p_size = __builtin_object_size(p, 0); - size_t q_size = __builtin_object_size(q, 0); - - if (__builtin_constant_p(size)) { - if (p_size < size) - __write_overflow(); - if (q_size < size) - __read_overflow2(); - } - if (p_size < size || q_size < size) - fortify_panic(__func__); - return __underlying_memmove(p, q, size); -} +#define memmove(p, q, s) __fortify_memcpy_chk(p, q, s, \ + __builtin_object_size(p, 0), __builtin_object_size(q, 0), \ + __builtin_object_size(p, 1), __builtin_object_size(q, 1), \ + memmove)
extern void *__real_memscan(void *, int, __kernel_size_t) __RENAME(memscan); __FORTIFY_INLINE void *memscan(void *p, int c, __kernel_size_t size) @@ -411,7 +399,6 @@ __FORTIFY_INLINE char *strcpy(char *p, const char *q) /* Don't use these outside the FORITFY_SOURCE implementation */ #undef __underlying_memchr #undef __underlying_memcmp -#undef __underlying_memmove #undef __underlying_memset #undef __underlying_strcat #undef __underlying_strcpy diff --git a/lib/test_fortify/read_overflow2_field-memmove.c b/lib/test_fortify/read_overflow2_field-memmove.c new file mode 100644 index 000000000000..6cc2724c8f62 --- /dev/null +++ b/lib/test_fortify/read_overflow2_field-memmove.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memmove(large, instance.buf, sizeof(instance.buf) + 1) + +#include "test_fortify.h" diff --git a/lib/test_fortify/write_overflow_field-memmove.c b/lib/test_fortify/write_overflow_field-memmove.c new file mode 100644 index 000000000000..377fcf9bb2fd --- /dev/null +++ b/lib/test_fortify/write_overflow_field-memmove.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memmove(instance.buf, large, sizeof(instance.buf) + 1) + +#include "test_fortify.h"
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Instead of writing beyond the end of evt_struct->iu.srp.cmd, target the upper union (evt_struct->iu.srp) instead, as that's what is being wiped.
Cc: Tyrel Datwyler tyreld@linux.ibm.com Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: "James E.J. Bottomley" jejb@linux.ibm.com Cc: "Martin K. Petersen" martin.petersen@oracle.com Cc: linux-scsi@vger.kernel.org Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Martin K. Petersen martin.petersen@oracle.com Link: https://lore.kernel.org/lkml/yq135rzp79c.fsf@ca-mkp.ca.oracle.com Acked-by: Tyrel Datwyler tyreld@linux.ibm.com Link: https://lore.kernel.org/lkml/6eae8434-e9a7-aa74-628b-b515b3695359@linux.ibm.... --- drivers/scsi/ibmvscsi/ibmvscsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c index 50df7dd9cb91..ea8e01f49cba 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.c +++ b/drivers/scsi/ibmvscsi/ibmvscsi.c @@ -1055,8 +1055,9 @@ static int ibmvscsi_queuecommand_lck(struct scsi_cmnd *cmnd, return SCSI_MLQUEUE_HOST_BUSY;
/* Set up the actual SRP IU */ + BUILD_BUG_ON(sizeof(evt_struct->iu.srp) != SRP_MAX_IU_LEN); + memset(&evt_struct->iu.srp, 0x00, sizeof(evt_struct->iu.srp)); srp_cmd = &evt_struct->iu.srp.cmd; - memset(srp_cmd, 0x00, SRP_MAX_IU_LEN); srp_cmd->opcode = SRP_CMD; memcpy(srp_cmd->cdb, cmnd->cmnd, sizeof(srp_cmd->cdb)); int_to_scsilun(lun, &srp_cmd->lun);
A common idiom in kernel code is to wipe the contents of a structure after a given member. This is especially useful in places where there is trailing padding. These open-coded cases are usually difficult to read and very sensitive to struct layout changes. Introduce a new helper, memset_after() that takes the target struct instance, the byte to write, and the member name after which the zeroing should start.
Additionally adds memset_startat() for wiping trailing members _starting_ at a specific member instead of after a member, which is more readable in certain circumstances, but doesn't include any preceding padding.
Cc: Steffen Klassert steffen.klassert@secunet.com Cc: Herbert Xu herbert@gondor.apana.org.au Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Andrew Morton akpm@linux-foundation.org Cc: Francis Laniel laniel_francis@privacyrequired.com Cc: Vincenzo Frascino vincenzo.frascino@arm.com Cc: Daniel Axtens dja@axtens.net Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/string.h | 29 +++++++++++++++++++++++++++++ lib/test_memcpy.c | 24 ++++++++++++++++++++++++ 2 files changed, 53 insertions(+)
diff --git a/include/linux/string.h b/include/linux/string.h index cbe889e404e2..fe56a1774207 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -272,6 +272,35 @@ static __always_inline void memcpy_and_pad(void *dest, size_t dest_len, memcpy(dest, src, dest_len); }
+/** + * memset_after - Set a value after a struct member to the end of a struct + * + * @obj: Address of target struct instance + * @v: Byte value to repeatedly write + * @member: after which struct member to start writing bytes + * + * This is good for clearing padding following the given member. + */ +#define memset_after(obj, v, member) do { \ + memset((u8 *)(obj) + offsetofend(typeof(*(obj)), member), v, \ + sizeof(*(obj)) - offsetofend(typeof(*(obj)), member)); \ +} while (0) + +/** + * memset_startat - Set a value starting at a member to the end of a struct + * + * @obj: Address of target struct instance + * @v: Byte value to repeatedly write + * @member: struct member to start writing at + * + * Note that if there is padding between the prior member and the target + * member, memset_after() should be used to clear the prior padding. + */ +#define memset_startat(obj, v, member) do { \ + memset((u8 *)(obj) + offsetof(typeof(*(obj)), member), v, \ + sizeof(*(obj)) - offsetof(typeof(*(obj)), member)); \ +} while (0) + /** * str_has_prefix - Test if a string has a given prefix * @str: The string to test diff --git a/lib/test_memcpy.c b/lib/test_memcpy.c index be192b8e82b7..50bc99552a17 100644 --- a/lib/test_memcpy.c +++ b/lib/test_memcpy.c @@ -215,6 +215,20 @@ static void memset_test(struct kunit *test) 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, }, }; + struct some_bytes after = { + .data = { 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x72, + 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, + 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, + 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, 0x72, + }, + }; + struct some_bytes startat = { + .data = { 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, 0x30, + 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, + 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, + 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, 0x79, + }, + }; struct some_bytes dest = { }; int count, value; u8 *ptr; @@ -245,6 +259,16 @@ static void memset_test(struct kunit *test) ptr += 8; memset(ptr++, value++, count++); compare("argument side-effects", dest, three); + + /* Verify memset_after() */ + dest = control; + memset_after(&dest, 0x72, three); + compare("memset_after()", dest, after); + + /* Verify memset_startat() */ + dest = control; + memset_startat(&dest, 0x79, four); + compare("memset_startat()", dest, startat); #undef TEST_OP }
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Clear trailing padding bytes using the new helper so that memset() doesn't get confused about writing "past the end" of the last struct member. There is no change to the resulting machine code.
Cc: Steffen Klassert steffen.klassert@secunet.com Cc: Herbert Xu herbert@gondor.apana.org.au Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/xfrm/xfrm_policy.c | 4 +--- net/xfrm/xfrm_user.c | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-)
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 37d17a79617c..1a06585022ab 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2486,9 +2486,7 @@ static inline struct xfrm_dst *xfrm_alloc_dst(struct net *net, int family) xdst = dst_alloc(dst_ops, NULL, 1, DST_OBSOLETE_NONE, 0);
if (likely(xdst)) { - struct dst_entry *dst = &xdst->u.dst; - - memset(dst + 1, 0, sizeof(*xdst) - sizeof(*dst)); + memset_after(xdst, 0, u.dst); } else xdst = ERR_PTR(-ENOBUFS);
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 03b66d154b2b..b7b986520dc7 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2912,7 +2912,7 @@ static int build_expire(struct sk_buff *skb, struct xfrm_state *x, const struct copy_to_user_state(x, &ue->state); ue->hard = (c->data.hard != 0) ? 1 : 0; /* clear the padding bytes */ - memset(&ue->hard + 1, 0, sizeof(*ue) - offsetofend(typeof(*ue), hard)); + memset_after(ue, 0, hard);
err = xfrm_mark_put(skb, &x->mark); if (err)
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_after() to clear everything after the dst_entry member of struct rt6_info.
Cc: "David S. Miller" davem@davemloft.net Cc: Hideaki YOSHIFUJI yoshfuji@linux-ipv6.org Cc: David Ahern dsahern@kernel.org Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/ipv6/route.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 6cf4bb89ca69..bd0ab3e436e7 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -327,9 +327,7 @@ static const struct rt6_info ip6_blk_hole_entry_template = {
static void rt6_info_init(struct rt6_info *rt) { - struct dst_entry *dst = &rt->dst; - - memset(dst + 1, 0, sizeof(*rt) - sizeof(*dst)); + memset_after(rt, 0, dst); INIT_LIST_HEAD(&rt->rt6i_uncached); }
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() to avoid confusing memset() about writing beyond the target struct member.
Cc: Pablo Neira Ayuso pablo@netfilter.org Cc: Jozsef Kadlecsik kadlec@netfilter.org Cc: Florian Westphal fw@strlen.de Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netfilter-devel@vger.kernel.org Cc: coreteam@netfilter.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/netfilter/nf_conntrack_core.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index d31dbccbe7bd..060503230519 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -1500,9 +1500,7 @@ __nf_conntrack_alloc(struct net *net, ct->status = 0; ct->timeout = 0; write_pnet(&ct->ct_net, net); - memset(&ct->__nfct_init_offset, 0, - offsetof(struct nf_conn, proto) - - offsetof(struct nf_conn, __nfct_init_offset)); + memset_after(ct, 0, __nfct_init_offset);
nf_ct_zone_add(ct, zone);
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Jes Sorensen jes@trained-monkey.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-hippi@sunsite.dk Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/802/hippi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/802/hippi.c b/net/802/hippi.c index f80b33a8f7e0..6f71ffc8bdd1 100644 --- a/net/802/hippi.c +++ b/net/802/hippi.c @@ -65,7 +65,7 @@ static int hippi_header(struct sk_buff *skb, struct net_device *dev, hip->le.src_addr_type = 2; /* 12 bit SC address */
memcpy(hip->le.src_switch_addr, dev->dev_addr + 3, 3); - memset(&hip->le.reserved, 0, 16); + memset_startat(&hip->le, 0, reserved);
hip->snap.dsap = HIPPI_EXTENDED_SAP; hip->snap.ssap = HIPPI_EXTENDED_SAP;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: dccp@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/dccp/trace.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/net/dccp/trace.h b/net/dccp/trace.h index 5062421beee9..5a43b3508c7f 100644 --- a/net/dccp/trace.h +++ b/net/dccp/trace.h @@ -60,9 +60,7 @@ TRACE_EVENT(dccp_probe, __entry->tx_t_ipi = hc->tx_t_ipi; } else { __entry->tx_s = 0; - memset(&__entry->tx_rtt, 0, (void *)&__entry->tx_t_ipi - - (void *)&__entry->tx_rtt + - sizeof(__entry->tx_t_ipi)); + memset_startat(__entry, 0, tx_rtt); } ),
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
The old code was doing the wrong thing: it starts from the second member and writes beyond int_info, clobbering qede_lock:
struct qede_dev { ... struct qed_int_info int_info;
/* Smaller private variant of the RTNL lock */ struct mutex qede_lock; ...
struct qed_int_info { struct msix_entry *msix; u8 msix_cnt;
/* This should be updated by the protocol driver */ u8 used_cnt; };
Cc: Ariel Elior aelior@marvell.com Cc: GR-everest-linux-l2@marvell.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/ethernet/qlogic/qede/qede_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index d400e9b235bf..0ed9a0c8452c 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -2419,7 +2419,7 @@ static int qede_load(struct qede_dev *edev, enum qede_load_mode mode, goto out; err4: qede_sync_free_irqs(edev); - memset(&edev->int_info.msix_cnt, 0, sizeof(struct qed_int_info)); + memset_startat(&edev->int_info, 0, msix_cnt); err3: qede_napi_disable_remove(edev); err2:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_after() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Additionally fix the common helper, ieee80211_tx_info_clear_status(), which was not clearing ack_signal, but the open-coded versions did. Johannes Berg points out this bug was introduced by commit e3e1a0bcb3f1 ("mac80211: reduce IEEE80211_TX_MAX_RATES") but was harmless.
Also drops the associated unneeded BUILD_BUG_ON()s, and adds a note to carl9170 about usage.
Cc: Johannes Berg johannes@sipsolutions.net Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/ath/carl9170/tx.c | 11 +++++------ drivers/net/wireless/intersil/p54/txrx.c | 6 +----- include/net/mac80211.h | 7 +------ 3 files changed, 7 insertions(+), 17 deletions(-)
diff --git a/drivers/net/wireless/ath/carl9170/tx.c b/drivers/net/wireless/ath/carl9170/tx.c index 88444fe6d1c6..aa95d1a65882 100644 --- a/drivers/net/wireless/ath/carl9170/tx.c +++ b/drivers/net/wireless/ath/carl9170/tx.c @@ -275,12 +275,11 @@ static void carl9170_tx_release(struct kref *ref) if (WARN_ON_ONCE(!ar)) return;
- BUILD_BUG_ON( - offsetof(struct ieee80211_tx_info, status.ack_signal) != 20); - - memset(&txinfo->status.ack_signal, 0, - sizeof(struct ieee80211_tx_info) - - offsetof(struct ieee80211_tx_info, status.ack_signal)); + /* + * Should this call ieee80211_tx_info_clear_status() instead of clearing + * manually? txinfo->status.rates do not seem to be used here. + */ + memset_after(&txinfo->status, 0, rates);
if (atomic_read(&ar->tx_total_queued)) ar->tx_schedule = true; diff --git a/drivers/net/wireless/intersil/p54/txrx.c b/drivers/net/wireless/intersil/p54/txrx.c index 873fea59894f..8414aa208655 100644 --- a/drivers/net/wireless/intersil/p54/txrx.c +++ b/drivers/net/wireless/intersil/p54/txrx.c @@ -431,11 +431,7 @@ static void p54_rx_frame_sent(struct p54_common *priv, struct sk_buff *skb) * Clear manually, ieee80211_tx_info_clear_status would * clear the counts too and we need them. */ - memset(&info->status.ack_signal, 0, - sizeof(struct ieee80211_tx_info) - - offsetof(struct ieee80211_tx_info, status.ack_signal)); - BUILD_BUG_ON(offsetof(struct ieee80211_tx_info, - status.ack_signal) != 20); + memset_after(&info->status, 0, rates);
if (entry_hdr->flags & cpu_to_le16(P54_HDR_FLAG_DATA_ALIGN)) pad = entry_data->align[0]; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index d8a1d09a2141..4c469b04de37 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1197,12 +1197,7 @@ ieee80211_tx_info_clear_status(struct ieee80211_tx_info *info) /* clear the rate counts */ for (i = 0; i < IEEE80211_TX_MAX_RATES; i++) info->status.rates[i].count = 0; - - BUILD_BUG_ON( - offsetof(struct ieee80211_tx_info, status.ack_signal) != 20); - memset(&info->status.ampdu_ack_len, 0, - sizeof(struct ieee80211_tx_info) - - offsetof(struct ieee80211_tx_info, status.ampdu_ack_len)); + memset_after(&info->status, 0, rates); }
On Tue, 2021-08-17 at 23:05 -0700, Kees Cook wrote:
@@ -275,12 +275,11 @@ static void carl9170_tx_release(struct kref *ref) if (WARN_ON_ONCE(!ar)) return;
- BUILD_BUG_ON(
offsetof(struct ieee80211_tx_info, status.ack_signal) != 20);
- memset(&txinfo->status.ack_signal, 0,
sizeof(struct ieee80211_tx_info) -
offsetof(struct ieee80211_tx_info, status.ack_signal));
- /*
* Should this call ieee80211_tx_info_clear_status() instead of clearing
* manually? txinfo->status.rates do not seem to be used here.
*/
Since you insist, I went digging :)
It should not, carl9170_tx_fill_rateinfo() has filled the rate information before we get to this point.
johannes
On Wed, 2021-08-18 at 09:08 +0200, Johannes Berg wrote:
On Tue, 2021-08-17 at 23:05 -0700, Kees Cook wrote:
@@ -275,12 +275,11 @@ static void carl9170_tx_release(struct kref *ref) if (WARN_ON_ONCE(!ar)) return;
- BUILD_BUG_ON(
offsetof(struct ieee80211_tx_info, status.ack_signal) != 20);
- memset(&txinfo->status.ack_signal, 0,
sizeof(struct ieee80211_tx_info) -
offsetof(struct ieee80211_tx_info, status.ack_signal));
- /*
* Should this call ieee80211_tx_info_clear_status() instead of clearing
* manually? txinfo->status.rates do not seem to be used here.
*/
Since you insist, I went digging :)
It should not, carl9170_tx_fill_rateinfo() has filled the rate information before we get to this point.
Otherwise, looks fine, FWIW.
Are you going to apply all of these together somewhere? I (we) can't, since memset_after() doesn't exist yet.
johannes
On Wed, Aug 18, 2021 at 10:06:51AM +0200, Johannes Berg wrote:
On Wed, 2021-08-18 at 09:08 +0200, Johannes Berg wrote:
On Tue, 2021-08-17 at 23:05 -0700, Kees Cook wrote:
@@ -275,12 +275,11 @@ static void carl9170_tx_release(struct kref *ref) if (WARN_ON_ONCE(!ar)) return;
- BUILD_BUG_ON(
offsetof(struct ieee80211_tx_info, status.ack_signal) != 20);
- memset(&txinfo->status.ack_signal, 0,
sizeof(struct ieee80211_tx_info) -
offsetof(struct ieee80211_tx_info, status.ack_signal));
- /*
* Should this call ieee80211_tx_info_clear_status() instead of clearing
* manually? txinfo->status.rates do not seem to be used here.
*/
Since you insist, I went digging :)
It should not, carl9170_tx_fill_rateinfo() has filled the rate information before we get to this point.
Ah-ha! Thanks for checking. I'll update the comment to explain the rationale here.
Otherwise, looks fine, FWIW.
Thanks!
Are you going to apply all of these together somewhere? I (we) can't, since memset_after() doesn't exist yet.
Right, given the dependencies, I am expecting to just carry the whole series, since the vast majority of it has no binary changes, etc. I'm hoping to get it into -next soon, but we're uncomfortably close to the merge window. :P
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally split up a later field-spanning memset() so that memset() can reason about the size.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: ath11k@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/wireless/ath/ath11k/hal_rx.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/net/wireless/ath/ath11k/hal_rx.c b/drivers/net/wireless/ath/ath11k/hal_rx.c index 325055ca41ab..0bab425f5dc9 100644 --- a/drivers/net/wireless/ath/ath11k/hal_rx.c +++ b/drivers/net/wireless/ath/ath11k/hal_rx.c @@ -29,8 +29,7 @@ static int ath11k_hal_reo_cmd_queue_stats(struct hal_tlv_hdr *tlv, FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
desc = (struct hal_reo_get_queue_stats *)tlv->value; - memset(&desc->queue_addr_lo, 0, - (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr))); + memset_startat(desc, 0, queue_addr_lo);
desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED; if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS) @@ -62,8 +61,7 @@ static int ath11k_hal_reo_cmd_flush_cache(struct ath11k_hal *hal, struct hal_tlv FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
desc = (struct hal_reo_flush_cache *)tlv->value; - memset(&desc->cache_addr_lo, 0, - (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr))); + memset_startat(desc, 0, cache_addr_lo);
desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED; if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS) @@ -101,8 +99,7 @@ static int ath11k_hal_reo_cmd_update_rx_queue(struct hal_tlv_hdr *tlv, FIELD_PREP(HAL_TLV_HDR_LEN, sizeof(*desc));
desc = (struct hal_reo_update_rx_queue *)tlv->value; - memset(&desc->queue_addr_lo, 0, - (sizeof(*desc) - sizeof(struct hal_reo_cmd_hdr))); + memset_startat(desc, 0, queue_addr_lo);
desc->cmd.info0 &= ~HAL_REO_CMD_HDR_INFO0_STATUS_REQUIRED; if (cmd->flag & HAL_REO_CMD_FLG_NEED_STATUS) @@ -762,15 +759,17 @@ void ath11k_hal_reo_qdesc_setup(void *vaddr, int tid, u32 ba_window_size, * size changes and also send WMI message to FW to change the REO * queue descriptor in Rx peer entry as part of dp_rx_tid_update. */ - memset(ext_desc, 0, 3 * sizeof(*ext_desc)); + memset(ext_desc, 0, sizeof(*ext_desc)); ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED, HAL_DESC_REO_QUEUE_EXT_DESC, REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_1); ext_desc++; + memset(ext_desc, 0, sizeof(*ext_desc)); ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED, HAL_DESC_REO_QUEUE_EXT_DESC, REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_2); ext_desc++; + memset(ext_desc, 0, sizeof(*ext_desc)); ath11k_hal_reo_set_desc_hdr(&ext_desc->desc_hdr, HAL_DESC_REO_OWNED, HAL_DESC_REO_QUEUE_EXT_DESC, REO_QUEUE_DESC_MAGIC_DEBUG_PATTERN_3);
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally split up a later field-spanning memset() so that memset() can reason about the size.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: ath11k@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
To avoid conflicts I prefer taking this via my ath tree.
On Thu, Aug 19, 2021 at 04:19:37PM +0300, Kalle Valo wrote:
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally split up a later field-spanning memset() so that memset() can reason about the size.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: ath11k@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
To avoid conflicts I prefer taking this via my ath tree.
The memset helpers are introduced as part of this series, so that makes things more difficult. Do you want me to create a branch with the helpers that you can merge?
-Kees
Kees Cook keescook@chromium.org writes:
On Thu, Aug 19, 2021 at 04:19:37PM +0300, Kalle Valo wrote:
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally split up a later field-spanning memset() so that memset() can reason about the size.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: ath11k@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
To avoid conflicts I prefer taking this via my ath tree.
The memset helpers are introduced as part of this series, so that makes things more difficult. Do you want me to create a branch with the helpers that you can merge?
Is this patch really worth the extra complexity? Why can't I apply this ath11k patch after the helpers have landed Linus' tree? That would be very simple.
On Sat, Aug 21, 2021 at 01:17:36PM +0300, Kalle Valo wrote:
Kees Cook keescook@chromium.org writes:
On Thu, Aug 19, 2021 at 04:19:37PM +0300, Kalle Valo wrote:
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally split up a later field-spanning memset() so that memset() can reason about the size.
Cc: Kalle Valo kvalo@codeaurora.org Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: ath11k@lists.infradead.org Cc: linux-wireless@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
To avoid conflicts I prefer taking this via my ath tree.
The memset helpers are introduced as part of this series, so that makes things more difficult. Do you want me to create a branch with the helpers that you can merge?
Is this patch really worth the extra complexity? Why can't I apply this ath11k patch after the helpers have landed Linus' tree? That would be very simple.
Not singularly, no. But I have a bit of a catch-22 in that I can't turn on greater FORTIFY strictness without first fixing the false positives, and I can't fix the false positives in "other" trees without those trees first having the helpers that get introduced by the FORTIFY series. :)
Anyway, since we're close to the merge window anyway, the FORTIFY series won't land in 1 release at this point regardless, so I'll just get the helpers landed and we can do the individual pieces once the merge window closes.
Wheee :)
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct. Additionally, since everything appears to perform a roundup (including allocation), just change the size of the struct itself and add a build-time check to validate the expected size.
Cc: Potnuri Bharat Teja bharat@chelsio.com Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: Raju Rangoju rajur@chelsio.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-rdma@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/infiniband/hw/cxgb4/cm.c | 5 +++-- drivers/net/ethernet/chelsio/cxgb4/t4_msg.h | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 291471d12197..6519ea8ebf23 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -2471,7 +2471,8 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, skb_get(skb); rpl = cplhdr(skb); if (!is_t4(adapter_type)) { - skb_trim(skb, roundup(sizeof(*rpl5), 16)); + BUILD_BUG_ON(sizeof(*rpl5) != roundup(sizeof(*rpl5), 16)); + skb_trim(skb, sizeof(*rpl5)); rpl5 = (void *)rpl; INIT_TP_WR(rpl5, ep->hwtid); } else { @@ -2487,7 +2488,7 @@ static int accept_cr(struct c4iw_ep *ep, struct sk_buff *skb, opt2 |= CONG_CNTRL_V(CONG_ALG_TAHOE); opt2 |= T5_ISS_F; rpl5 = (void *)rpl; - memset(&rpl5->iss, 0, roundup(sizeof(*rpl5)-sizeof(*rpl), 16)); + memset_after(rpl5, 0, iss); if (peer2peer) isn += 4; rpl5->iss = cpu_to_be32(isn); diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h index fed5f93bf620..26433a62d7f0 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_msg.h @@ -497,7 +497,7 @@ struct cpl_t5_pass_accept_rpl { __be32 opt2; __be64 opt0; __be32 iss; - __be32 rsvd; + __be32 rsvd[3]; };
struct cpl_act_open_req {
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Alexander Shishkin alexander.shishkin@linux.intel.com Signed-off-by: Kees Cook keescook@chromium.org --- drivers/hwtracing/intel_th/msu.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c index 432ade0842f6..70a07b4e9967 100644 --- a/drivers/hwtracing/intel_th/msu.c +++ b/drivers/hwtracing/intel_th/msu.c @@ -658,13 +658,11 @@ static void msc_buffer_clear_hw_header(struct msc *msc)
list_for_each_entry(win, &msc->win_list, entry) { unsigned int blk; - size_t hw_sz = sizeof(struct msc_block_desc) - - offsetof(struct msc_block_desc, hw_tag);
for_each_sg(win->sgt->sgl, sg, win->nr_segs, blk) { struct msc_block_desc *bdesc = sg_virt(sg);
- memset(&bdesc->hw_tag, 0, hw_sz); + memset_startat(bdesc, 0, hw_tag); } } }
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Alexander Shishkin alexander.shishkin@linux.intel.com Signed-off-by: Kees Cook keescook@chromium.org
Acked-by: Alexander Shishkin alexander.shishkin@linux.intel.com
drivers/hwtracing/intel_th/msu.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c index 432ade0842f6..70a07b4e9967 100644 --- a/drivers/hwtracing/intel_th/msu.c +++ b/drivers/hwtracing/intel_th/msu.c @@ -658,13 +658,11 @@ static void msc_buffer_clear_hw_header(struct msc *msc)
list_for_each_entry(win, &msc->win_list, entry) { unsigned int blk;
size_t hw_sz = sizeof(struct msc_block_desc) -
offsetof(struct msc_block_desc, hw_tag);
for_each_sg(win->sgt->sgl, sg, win->nr_segs, blk) { struct msc_block_desc *bdesc = sg_virt(sg);
memset(&bdesc->hw_tag, 0, hw_sz);
} }memset_startat(bdesc, 0, hw_tag);
}
2.30.2
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: Max Gurtovoy maxg@mellanox.com Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/infiniband/hw/mthca/mthca_mr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c index ce0e0867e488..1208e92ca3d3 100644 --- a/drivers/infiniband/hw/mthca/mthca_mr.c +++ b/drivers/infiniband/hw/mthca/mthca_mr.c @@ -469,8 +469,7 @@ int mthca_mr_alloc(struct mthca_dev *dev, u32 pd, int buffer_size_shift, mpt_entry->start = cpu_to_be64(iova); mpt_entry->length = cpu_to_be64(total_size);
- memset(&mpt_entry->lkey, 0, - sizeof *mpt_entry - offsetof(struct mthca_mpt_entry, lkey)); + memset_startat(mpt_entry, 0, lkey);
if (mr->mtt) mpt_entry->mtt_seg =
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Chris Mason clm@fb.com Cc: Josef Bacik josef@toxicpanda.com Cc: David Sterba dsterba@suse.com Cc: linux-btrfs@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- fs/btrfs/root-tree.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 702dc5441f03..12ceb14a1141 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -39,10 +39,8 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot, need_reset = 1; } if (need_reset) { - memset(&item->generation_v2, 0, - sizeof(*item) - offsetof(struct btrfs_root_item, - generation_v2)); - + /* Clear all members from generation_v2 onwards. */ + memset_startat(item, 0, generation_v2); generate_random_guid(item->uuid); } }
On 18.08.21 г. 9:05, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Chris Mason clm@fb.com Cc: Josef Bacik josef@toxicpanda.com Cc: David Sterba dsterba@suse.com Cc: linux-btrfs@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Reviewed-by: Nikolay Borisov nborisov@suse.com
On Tue, Aug 17, 2021 at 11:05:19PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() so memset() doesn't get confused about writing beyond the destination member that is intended to be the starting point of zeroing through the end of the struct.
Cc: Chris Mason clm@fb.com Cc: Josef Bacik josef@toxicpanda.com Cc: David Sterba dsterba@suse.com Cc: linux-btrfs@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Acked-by: David Sterba dsterba@suse.com
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() to avoid confusing memset() about writing beyond the target struct member.
Cc: Steven Rostedt rostedt@goodmis.org Cc: Ingo Molnar mingo@redhat.com Signed-off-by: Kees Cook keescook@chromium.org --- kernel/trace/trace.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 13587e771567..9ff8c31975cd 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6691,9 +6691,7 @@ tracing_read_pipe(struct file *filp, char __user *ubuf, cnt = PAGE_SIZE - 1;
/* reset all but tr, trace, and overruns */ - memset(&iter->seq, 0, - sizeof(struct trace_iterator) - - offsetof(struct trace_iterator, seq)); + memset_startat(iter, 0, seq); cpumask_clear(iter->started); trace_seq_init(&iter->seq); iter->pos = -1;
On Tue, 17 Aug 2021 23:05:20 -0700 Kees Cook keescook@chromium.org wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() to avoid confusing memset() about writing beyond the target struct member.
Cc: Steven Rostedt rostedt@goodmis.org Cc: Ingo Molnar mingo@redhat.com Signed-off-by: Kees Cook keescook@chromium.org
kernel/trace/trace.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 13587e771567..9ff8c31975cd 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6691,9 +6691,7 @@ tracing_read_pipe(struct file *filp, char __user *ubuf, cnt = PAGE_SIZE - 1;
/* reset all but tr, trace, and overruns */
- memset(&iter->seq, 0,
sizeof(struct trace_iterator) -
offsetof(struct trace_iterator, seq));
- memset_startat(iter, 0, seq);
I can't find memset_startat() in mainline nor linux-next. I don't see it in this thread either, but since this has 63 patches, I could have easily missed it.
This change really should belong to a patch set that just introduces memset_startat() (and perhaps memset_after()) and then updates all the places that should use it. That way I can give it a proper review. In other words, you should break this patch set up into smaller, more digestible portions for the reviewers.
Thanks,
-- Steve
cpumask_clear(iter->started); trace_seq_init(&iter->seq); iter->pos = -1;
On Wed, Aug 18, 2021 at 09:33:49AM -0400, Steven Rostedt wrote:
On Tue, 17 Aug 2021 23:05:20 -0700 Kees Cook keescook@chromium.org wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Use memset_startat() to avoid confusing memset() about writing beyond the target struct member.
Cc: Steven Rostedt rostedt@goodmis.org Cc: Ingo Molnar mingo@redhat.com Signed-off-by: Kees Cook keescook@chromium.org
kernel/trace/trace.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 13587e771567..9ff8c31975cd 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6691,9 +6691,7 @@ tracing_read_pipe(struct file *filp, char __user *ubuf, cnt = PAGE_SIZE - 1;
/* reset all but tr, trace, and overruns */
- memset(&iter->seq, 0,
sizeof(struct trace_iterator) -
offsetof(struct trace_iterator, seq));
- memset_startat(iter, 0, seq);
I can't find memset_startat() in mainline nor linux-next. I don't see it in this thread either, but since this has 63 patches, I could have easily missed it.
Sorry, it isn't called out in the Subject, but it's part of this patch: https://lore.kernel.org/lkml/20210818060533.3569517-38-keescook@chromium.org...
This change really should belong to a patch set that just introduces memset_startat() (and perhaps memset_after()) and then updates all the places that should use it. That way I can give it a proper review. In other words, you should break this patch set up into smaller, more digestible portions for the reviewers.
I will split memset_after() and memset_startat() introduction patches. I already split up each use into individual cases, so that those changes could be checked one step at a time for differences in pahole struct layout and object code.
Thanks for taking a look!
-Kees
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the algs so that memset() can correctly reason about the size.
Cc: Philipp Reisner philipp.reisner@linbit.com Cc: Lars Ellenberg lars.ellenberg@linbit.com Cc: Jens Axboe axboe@kernel.dk Cc: drbd-dev@lists.linbit.com Cc: linux-block@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/block/drbd/drbd_main.c | 3 ++- drivers/block/drbd/drbd_protocol.h | 6 ++++-- drivers/block/drbd/drbd_receiver.c | 3 ++- 3 files changed, 8 insertions(+), 4 deletions(-)
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 55234a558e98..b824679cfcb2 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -729,7 +729,8 @@ int drbd_send_sync_param(struct drbd_peer_device *peer_device) cmd = apv >= 89 ? P_SYNC_PARAM89 : P_SYNC_PARAM;
/* initialize verify_alg and csums_alg */ - memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); + BUILD_BUG_ON(sizeof(p->algs) != 2 * SHARED_SECRET_MAX); + memset(&p->algs, 0, sizeof(p->algs));
if (get_ldev(peer_device->device)) { dc = rcu_dereference(peer_device->device->ldev->disk_conf); diff --git a/drivers/block/drbd/drbd_protocol.h b/drivers/block/drbd/drbd_protocol.h index dea59c92ecc1..a882b65ab5d2 100644 --- a/drivers/block/drbd/drbd_protocol.h +++ b/drivers/block/drbd/drbd_protocol.h @@ -283,8 +283,10 @@ struct p_rs_param_89 {
struct p_rs_param_95 { u32 resync_rate; - char verify_alg[SHARED_SECRET_MAX]; - char csums_alg[SHARED_SECRET_MAX]; + struct_group(algs, + char verify_alg[SHARED_SECRET_MAX]; + char csums_alg[SHARED_SECRET_MAX]; + ); u32 c_plan_ahead; u32 c_delay_target; u32 c_fill_target; diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index 1f740e42e457..6df2539e215b 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -3921,7 +3921,8 @@ static int receive_SyncParam(struct drbd_connection *connection, struct packet_i
/* initialize verify_alg and csums_alg */ p = pi->data; - memset(p->verify_alg, 0, 2 * SHARED_SECRET_MAX); + BUILD_BUG_ON(sizeof(p->algs) != 2 * SHARED_SECRET_MAX); + memset(&p->algs, 0, sizeof(p->algs));
err = drbd_recv_all(peer_device->connection, p, header_size); if (err)
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct cm4000_dev that should be initialized to zero.
Cc: Harald Welte laforge@gnumonks.org Cc: Arnd Bergmann arnd@arndb.de Cc: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Kees Cook keescook@chromium.org Acked-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Link: https://lore.kernel.org/lkml/YQDvxAofJlI1JoGZ@kroah.com --- drivers/char/pcmcia/cm4000_cs.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/drivers/char/pcmcia/cm4000_cs.c b/drivers/char/pcmcia/cm4000_cs.c index 8f1bce0b4fe5..2f43e7088e16 100644 --- a/drivers/char/pcmcia/cm4000_cs.c +++ b/drivers/char/pcmcia/cm4000_cs.c @@ -116,8 +116,9 @@ struct cm4000_dev { wait_queue_head_t atrq; /* wait for ATR valid */ wait_queue_head_t readq; /* used by write to wake blk.read */
- /* warning: do not move this fields. + /* warning: do not move this struct group. * initialising to zero depends on it - see ZERO_DEV below. */ + struct_group(init, unsigned char atr_csum; unsigned char atr_len_retry; unsigned short atr_len; @@ -140,12 +141,10 @@ struct cm4000_dev {
struct timer_list timer; /* used to keep monitor running */ int monitor_running; + ); };
-#define ZERO_DEV(dev) \ - memset(&dev->atr_csum,0, \ - sizeof(struct cm4000_dev) - \ - offsetof(struct cm4000_dev, atr_csum)) +#define ZERO_DEV(dev) memset(&dev->init, 0, sizeof(dev->init))
static struct pcmcia_device *dev_table[CM4000_MAX_DEV]; static struct class *cmm_class;
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct x86_emulate_ctxt that should be initialized to zero.
Cc: Paolo Bonzini pbonzini@redhat.com Cc: Sean Christopherson seanjc@google.com Cc: Vitaly Kuznetsov vkuznets@redhat.com Cc: Wanpeng Li wanpengli@tencent.com Cc: Jim Mattson jmattson@google.com Cc: Joerg Roedel joro@8bytes.org Cc: Thomas Gleixner tglx@linutronix.de Cc: Ingo Molnar mingo@redhat.com Cc: Borislav Petkov bp@alien8.de Cc: x86@kernel.org Cc: "H. Peter Anvin" hpa@zytor.com Cc: kvm@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- arch/x86/kvm/emulate.c | 3 +-- arch/x86/kvm/kvm_emulate.h | 19 +++++++++++-------- 2 files changed, 12 insertions(+), 10 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 2837110e66ed..2608a047e769 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5377,8 +5377,7 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop)
void init_decode_cache(struct x86_emulate_ctxt *ctxt) { - memset(&ctxt->rip_relative, 0, - (void *)&ctxt->modrm - (void *)&ctxt->rip_relative); + memset(&ctxt->decode_cache, 0, sizeof(ctxt->decode_cache));
ctxt->io_read.pos = 0; ctxt->io_read.end = 0; diff --git a/arch/x86/kvm/kvm_emulate.h b/arch/x86/kvm/kvm_emulate.h index 68b420289d7e..9b8afcb8ad39 100644 --- a/arch/x86/kvm/kvm_emulate.h +++ b/arch/x86/kvm/kvm_emulate.h @@ -341,14 +341,17 @@ struct x86_emulate_ctxt { * the rest are initialized unconditionally in x86_decode_insn * or elsewhere */ - bool rip_relative; - u8 rex_prefix; - u8 lock_prefix; - u8 rep_prefix; - /* bitmaps of registers in _regs[] that can be read */ - u32 regs_valid; - /* bitmaps of registers in _regs[] that have been written */ - u32 regs_dirty; + struct_group(decode_cache, + bool rip_relative; + u8 rex_prefix; + u8 lock_prefix; + u8 rep_prefix; + /* bitmaps of registers in _regs[] that can be read */ + u32 regs_valid; + /* bitmaps of registers in _regs[] that have been written */ + u32 regs_dirty; + ); + /* modrm */ u8 modrm; u8 modrm_mod;
On Tue, Aug 17, 2021, Kees Cook wrote:
arch/x86/kvm/emulate.c | 3 +-- arch/x86/kvm/kvm_emulate.h | 19 +++++++++++-------- 2 files changed, 12 insertions(+), 10 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 2837110e66ed..2608a047e769 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5377,8 +5377,7 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop)
void init_decode_cache(struct x86_emulate_ctxt *ctxt) {
- memset(&ctxt->rip_relative, 0,
(void *)&ctxt->modrm - (void *)&ctxt->rip_relative);
memset(&ctxt->decode_cache, 0, sizeof(ctxt->decode_cache));
ctxt->io_read.pos = 0; ctxt->io_read.end = 0;
diff --git a/arch/x86/kvm/kvm_emulate.h b/arch/x86/kvm/kvm_emulate.h index 68b420289d7e..9b8afcb8ad39 100644 --- a/arch/x86/kvm/kvm_emulate.h +++ b/arch/x86/kvm/kvm_emulate.h @@ -341,14 +341,17 @@ struct x86_emulate_ctxt { * the rest are initialized unconditionally in x86_decode_insn * or elsewhere */
- bool rip_relative;
- u8 rex_prefix;
- u8 lock_prefix;
- u8 rep_prefix;
- /* bitmaps of registers in _regs[] that can be read */
- u32 regs_valid;
- /* bitmaps of registers in _regs[] that have been written */
- u32 regs_dirty;
- struct_group(decode_cache,
This is somewhat misleading because half of this struct is the so called "decode cache", not just these six fields.
KVM's "optimization" is quite ridiculous as this has never been such a hot path that saving a few mov instructions would be noticeable. And hilariously, the "optimization" is completely unnecessary because both gcc and clang are clever enough to batch the first five into a movq even when zeroing the fields individually.
So, I would much prefer to go with the following:
From dbdca1f4cd01fee418c252e54c360d518b2b1ad6 Mon Sep 17 00:00:00 2001
From: Sean Christopherson seanjc@google.com Date: Wed, 18 Aug 2021 08:03:08 -0700 Subject: [PATCH] KVM: x86: Replace memset() "optimization" with normal per-field writes
Explicitly zero select fields in the emulator's decode cache instead of zeroing the fields via a gross memset() that spans six fields. gcc and clang are both clever enough to batch the first five fields into a single quadword MOV, i.e. memset() and individually zeroing generate identical code.
Removing the wart also prepares KVM for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset().
No functional change intended.
Reported-by: Kees Cook keescook@chromium.org Signed-off-by: Sean Christopherson seanjc@google.com --- arch/x86/kvm/emulate.c | 9 +++++++-- arch/x86/kvm/kvm_emulate.h | 6 +----- 2 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 2837110e66ed..bf81fd017e7f 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5377,8 +5377,13 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop)
void init_decode_cache(struct x86_emulate_ctxt *ctxt) { - memset(&ctxt->rip_relative, 0, - (void *)&ctxt->modrm - (void *)&ctxt->rip_relative); + /* Clear fields that are set conditionally but read without a guard. */ + ctxt->rip_relative = false; + ctxt->rex_prefix = 0; + ctxt->lock_prefix = 0; + ctxt->rep_prefix = 0; + ctxt->regs_valid = 0; + ctxt->regs_dirty = 0;
ctxt->io_read.pos = 0; ctxt->io_read.end = 0; diff --git a/arch/x86/kvm/kvm_emulate.h b/arch/x86/kvm/kvm_emulate.h index 68b420289d7e..bc1fecacccd4 100644 --- a/arch/x86/kvm/kvm_emulate.h +++ b/arch/x86/kvm/kvm_emulate.h @@ -336,11 +336,7 @@ struct x86_emulate_ctxt { fastop_t fop; }; int (*check_perm)(struct x86_emulate_ctxt *ctxt); - /* - * The following six fields are cleared together, - * the rest are initialized unconditionally in x86_decode_insn - * or elsewhere - */ + bool rip_relative; u8 rex_prefix; u8 lock_prefix; -- 2.33.0.rc1.237.g0d66db33f3-goog
bool rip_relative;
u8 rex_prefix;
u8 lock_prefix;
u8 rep_prefix;
/* bitmaps of registers in _regs[] that can be read */
u32 regs_valid;
/* bitmaps of registers in _regs[] that have been written */
u32 regs_dirty;
- );
- /* modrm */ u8 modrm; u8 modrm_mod;
-- 2.30.2
On Wed, Aug 18, 2021 at 03:11:28PM +0000, Sean Christopherson wrote:
On Tue, Aug 17, 2021, Kees Cook wrote:
arch/x86/kvm/emulate.c | 3 +-- arch/x86/kvm/kvm_emulate.h | 19 +++++++++++-------- 2 files changed, 12 insertions(+), 10 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 2837110e66ed..2608a047e769 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5377,8 +5377,7 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop)
void init_decode_cache(struct x86_emulate_ctxt *ctxt) {
- memset(&ctxt->rip_relative, 0,
(void *)&ctxt->modrm - (void *)&ctxt->rip_relative);
memset(&ctxt->decode_cache, 0, sizeof(ctxt->decode_cache));
ctxt->io_read.pos = 0; ctxt->io_read.end = 0;
diff --git a/arch/x86/kvm/kvm_emulate.h b/arch/x86/kvm/kvm_emulate.h index 68b420289d7e..9b8afcb8ad39 100644 --- a/arch/x86/kvm/kvm_emulate.h +++ b/arch/x86/kvm/kvm_emulate.h @@ -341,14 +341,17 @@ struct x86_emulate_ctxt { * the rest are initialized unconditionally in x86_decode_insn * or elsewhere */
- bool rip_relative;
- u8 rex_prefix;
- u8 lock_prefix;
- u8 rep_prefix;
- /* bitmaps of registers in _regs[] that can be read */
- u32 regs_valid;
- /* bitmaps of registers in _regs[] that have been written */
- u32 regs_dirty;
- struct_group(decode_cache,
This is somewhat misleading because half of this struct is the so called "decode cache", not just these six fields.
KVM's "optimization" is quite ridiculous as this has never been such a hot path that saving a few mov instructions would be noticeable. And hilariously, the "optimization" is completely unnecessary because both gcc and clang are clever enough to batch the first five into a movq even when zeroing the fields individually.
So, I would much prefer to go with the following:
Sounds good to me!
From dbdca1f4cd01fee418c252e54c360d518b2b1ad6 Mon Sep 17 00:00:00 2001 From: Sean Christopherson seanjc@google.com Date: Wed, 18 Aug 2021 08:03:08 -0700 Subject: [PATCH] KVM: x86: Replace memset() "optimization" with normal per-field writes
Explicitly zero select fields in the emulator's decode cache instead of zeroing the fields via a gross memset() that spans six fields. gcc and clang are both clever enough to batch the first five fields into a single quadword MOV, i.e. memset() and individually zeroing generate identical code.
Removing the wart also prepares KVM for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset().
No functional change intended.
Reported-by: Kees Cook keescook@chromium.org Signed-off-by: Sean Christopherson seanjc@google.com
Reviewed-by: Kees Cook keescook@chromium.org
Do you want me to take this patch into my tree, or do you want to carry it for KVM directly?
Thanks!
-Kees
arch/x86/kvm/emulate.c | 9 +++++++-- arch/x86/kvm/kvm_emulate.h | 6 +----- 2 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 2837110e66ed..bf81fd017e7f 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5377,8 +5377,13 @@ static int fastop(struct x86_emulate_ctxt *ctxt, fastop_t fop)
void init_decode_cache(struct x86_emulate_ctxt *ctxt) {
- memset(&ctxt->rip_relative, 0,
(void *)&ctxt->modrm - (void *)&ctxt->rip_relative);
/* Clear fields that are set conditionally but read without a guard. */
ctxt->rip_relative = false;
ctxt->rex_prefix = 0;
ctxt->lock_prefix = 0;
ctxt->rep_prefix = 0;
ctxt->regs_valid = 0;
ctxt->regs_dirty = 0;
ctxt->io_read.pos = 0; ctxt->io_read.end = 0;
diff --git a/arch/x86/kvm/kvm_emulate.h b/arch/x86/kvm/kvm_emulate.h index 68b420289d7e..bc1fecacccd4 100644 --- a/arch/x86/kvm/kvm_emulate.h +++ b/arch/x86/kvm/kvm_emulate.h @@ -336,11 +336,7 @@ struct x86_emulate_ctxt { fastop_t fop; }; int (*check_perm)(struct x86_emulate_ctxt *ctxt);
- /*
* The following six fields are cleared together,
* the rest are initialized unconditionally in x86_decode_insn
* or elsewhere
*/
- bool rip_relative; u8 rex_prefix; u8 lock_prefix;
-- 2.33.0.rc1.237.g0d66db33f3-goog
bool rip_relative;
u8 rex_prefix;
u8 lock_prefix;
u8 rep_prefix;
/* bitmaps of registers in _regs[] that can be read */
u32 regs_valid;
/* bitmaps of registers in _regs[] that have been written */
u32 regs_dirty;
- );
- /* modrm */ u8 modrm; u8 modrm_mod;
-- 2.30.2
On Wed, Aug 18, 2021, Kees Cook wrote:
On Wed, Aug 18, 2021 at 03:11:28PM +0000, Sean Christopherson wrote:
From dbdca1f4cd01fee418c252e54c360d518b2b1ad6 Mon Sep 17 00:00:00 2001 From: Sean Christopherson seanjc@google.com Date: Wed, 18 Aug 2021 08:03:08 -0700 Subject: [PATCH] KVM: x86: Replace memset() "optimization" with normal per-field writes
Explicitly zero select fields in the emulator's decode cache instead of zeroing the fields via a gross memset() that spans six fields. gcc and clang are both clever enough to batch the first five fields into a single quadword MOV, i.e. memset() and individually zeroing generate identical code.
Removing the wart also prepares KVM for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset().
No functional change intended.
Reported-by: Kees Cook keescook@chromium.org Signed-off-by: Sean Christopherson seanjc@google.com
Reviewed-by: Kees Cook keescook@chromium.org
Do you want me to take this patch into my tree, or do you want to carry it for KVM directly?
That's a Paolo question :-)
What's the expected timeframe for landing stricter bounds checking? If it's 5.16 or later, the easiest thing would be to squeak this into 5.15.
On Wed, Aug 18, 2021 at 10:53:58PM +0000, Sean Christopherson wrote:
On Wed, Aug 18, 2021, Kees Cook wrote:
On Wed, Aug 18, 2021 at 03:11:28PM +0000, Sean Christopherson wrote:
From dbdca1f4cd01fee418c252e54c360d518b2b1ad6 Mon Sep 17 00:00:00 2001 From: Sean Christopherson seanjc@google.com Date: Wed, 18 Aug 2021 08:03:08 -0700 Subject: [PATCH] KVM: x86: Replace memset() "optimization" with normal per-field writes
Explicitly zero select fields in the emulator's decode cache instead of zeroing the fields via a gross memset() that spans six fields. gcc and clang are both clever enough to batch the first five fields into a single quadword MOV, i.e. memset() and individually zeroing generate identical code.
Removing the wart also prepares KVM for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset().
No functional change intended.
Reported-by: Kees Cook keescook@chromium.org Signed-off-by: Sean Christopherson seanjc@google.com
Reviewed-by: Kees Cook keescook@chromium.org
Do you want me to take this patch into my tree, or do you want to carry it for KVM directly?
That's a Paolo question :-)
What's the expected timeframe for landing stricter bounds checking? If it's 5.16 or later, the easiest thing would be to squeak this into 5.15.
I'm hoping to land all the "compile time" stuff for 5.15, but realistically, some portions may not get there. I'll just carry this patch for now and if we need to swap trees we can do that. :)
Thanks!
-Kees
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct journal_sector that should be initialized to zero.
Cc: Alasdair Kergon agk@redhat.com Cc: Mike Snitzer snitzer@redhat.com Cc: dm-devel@redhat.com Signed-off-by: Kees Cook keescook@chromium.org --- drivers/md/dm-integrity.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/md/dm-integrity.c b/drivers/md/dm-integrity.c index 40f8116c8e44..59deea0dd305 100644 --- a/drivers/md/dm-integrity.c +++ b/drivers/md/dm-integrity.c @@ -119,8 +119,10 @@ struct journal_entry { #define JOURNAL_MAC_SIZE (JOURNAL_MAC_PER_SECTOR * JOURNAL_BLOCK_SECTORS)
struct journal_sector { - __u8 entries[JOURNAL_SECTOR_DATA - JOURNAL_MAC_PER_SECTOR]; - __u8 mac[JOURNAL_MAC_PER_SECTOR]; + struct_group(sectors, + __u8 entries[JOURNAL_SECTOR_DATA - JOURNAL_MAC_PER_SECTOR]; + __u8 mac[JOURNAL_MAC_PER_SECTOR]; + ); commit_id_t commit_id; };
@@ -2856,7 +2858,8 @@ static void init_journal(struct dm_integrity_c *ic, unsigned start_section, wraparound_section(ic, &i); for (j = 0; j < ic->journal_section_sectors; j++) { struct journal_sector *js = access_journal(ic, i, j); - memset(&js->entries, 0, JOURNAL_SECTOR_DATA); + BUILD_BUG_ON(sizeof(js->sectors) != JOURNAL_SECTOR_DATA); + memset(&js->sectors, 0, sizeof(js->sectors)); js->commit_id = dm_integrity_commit_id(ic, i, j, commit_seq); } for (j = 0; j < ic->journal_section_entries; j++) {
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/hid/hid-roccat-kone.c | 2 +- drivers/hid/hid-roccat-kone.h | 12 +++++++----- 2 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/hid/hid-roccat-kone.c b/drivers/hid/hid-roccat-kone.c index 1ca64481145e..ea17abc7ad52 100644 --- a/drivers/hid/hid-roccat-kone.c +++ b/drivers/hid/hid-roccat-kone.c @@ -857,7 +857,7 @@ static int kone_raw_event(struct hid_device *hdev, struct hid_report *report, memcpy(&kone->last_mouse_event, event, sizeof(struct kone_mouse_event)); else - memset(&event->tilt, 0, 5); + memset(&event->wipe, 0, sizeof(event->wipe));
kone_keep_values_up_to_date(kone, event);
diff --git a/drivers/hid/hid-roccat-kone.h b/drivers/hid/hid-roccat-kone.h index 4a1a9cb76b08..65c800e3addc 100644 --- a/drivers/hid/hid-roccat-kone.h +++ b/drivers/hid/hid-roccat-kone.h @@ -152,11 +152,13 @@ struct kone_mouse_event { uint16_t x; uint16_t y; uint8_t wheel; /* up = 1, down = -1 */ - uint8_t tilt; /* right = 1, left = -1 */ - uint8_t unknown; - uint8_t event; - uint8_t value; /* press = 0, release = 1 */ - uint8_t macro_key; /* 0 to 8 */ + struct_group(wipe, + uint8_t tilt; /* right = 1, left = -1 */ + uint8_t unknown; + uint8_t event; + uint8_t value; /* press = 0, release = 1 */ + uint8_t macro_key; /* 0 to 8 */ + ); } __attribute__ ((__packed__));
enum kone_mouse_events {
On Tue, 17 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
On Fri, Aug 20, 2021, 6:02 AM Jiri Kosina jikos@kernel.org wrote:
On Tue, 17 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
Eek! No, this will break the build: struct_group() is not yet in the tree. I can carry this with an Ack, etc.
-Kees
On Fri, 20 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
Eek! No, this will break the build: struct_group() is not yet in the tree. I can carry this with an Ack, etc.
I was pretty sure I saw struct_group() already in linux-next, but that was apparently a vacation-induced brainfart, sorry. Dropping.
On Fri, Aug 20, 2021 at 05:27:35PM +0200, Jiri Kosina wrote:
On Fri, 20 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
Eek! No, this will break the build: struct_group() is not yet in the tree. I can carry this with an Ack, etc.
I was pretty sure I saw struct_group() already in linux-next, but that was apparently a vacation-induced brainfart, sorry. Dropping.
Cool, no worries. Sorry for the confusion!
On Fri, Aug 20, 2021 at 05:27:35PM +0200, Jiri Kosina wrote:
On Fri, 20 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
Eek! No, this will break the build: struct_group() is not yet in the tree. I can carry this with an Ack, etc.
I was pretty sure I saw struct_group() already in linux-next, but that was apparently a vacation-induced brainfart, sorry. Dropping.
Oh, for these two patches, can I add your Acked-by while I carry them?
Thanks!
On Fri, 20 Aug 2021, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct kone_mouse_event that should be initialized to zero.
Cc: Stefan Achatz erazor_de@users.sourceforge.net Cc: Jiri Kosina jikos@kernel.org Cc: Benjamin Tissoires benjamin.tissoires@redhat.com Cc: linux-input@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Applied, thank you Kees.
Eek! No, this will break the build: struct_group() is not yet in the tree. I can carry this with an Ack, etc.
I was pretty sure I saw struct_group() already in linux-next, but that was apparently a vacation-induced brainfart, sorry. Dropping.
Oh, for these two patches, can I add your Acked-by while I carry them?
Yes, thanks, and sorry for the noise.
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct mlx5_ib_mr that should be initialized to zero.
Cc: Leon Romanovsky leon@kernel.org Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bf20a388eabe..f63bf204a7a1 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -644,6 +644,7 @@ struct mlx5_ib_mr { struct ib_umem *umem;
/* This is zero'd when the MR is allocated */ + struct_group(cleared, union { /* Used only while the MR is in the cache */ struct { @@ -691,12 +692,13 @@ struct mlx5_ib_mr { bool is_odp_implicit; }; }; + ); };
/* Zero the fields in the mr that are variant depending on usage */ static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr) { - memset(mr->out, 0, sizeof(*mr) - offsetof(struct mlx5_ib_mr, out)); + memset(&mr->cleared, 0, sizeof(mr->cleared)); }
static inline bool is_odp_mr(struct mlx5_ib_mr *mr)
On Tue, Aug 17, 2021 at 11:05:26PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct mlx5_ib_mr that should be initialized to zero.
Cc: Leon Romanovsky leon@kernel.org Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bf20a388eabe..f63bf204a7a1 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -644,6 +644,7 @@ struct mlx5_ib_mr { struct ib_umem *umem;
/* This is zero'd when the MR is allocated */
- struct_group(cleared, union { /* Used only while the MR is in the cache */ struct {
@@ -691,12 +692,13 @@ struct mlx5_ib_mr { bool is_odp_implicit; }; };
- );
};
/* Zero the fields in the mr that are variant depending on usage */ static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr) {
- memset(mr->out, 0, sizeof(*mr) - offsetof(struct mlx5_ib_mr, out));
- memset(&mr->cleared, 0, sizeof(mr->cleared));
}
Why not use the memset_after(mr->umem) here?
Jason
On Thu, Aug 19, 2021 at 09:27:16AM -0300, Jason Gunthorpe wrote:
On Tue, Aug 17, 2021 at 11:05:26PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct mlx5_ib_mr that should be initialized to zero.
Cc: Leon Romanovsky leon@kernel.org Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bf20a388eabe..f63bf204a7a1 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -644,6 +644,7 @@ struct mlx5_ib_mr { struct ib_umem *umem;
/* This is zero'd when the MR is allocated */
- struct_group(cleared, union { /* Used only while the MR is in the cache */ struct {
@@ -691,12 +692,13 @@ struct mlx5_ib_mr { bool is_odp_implicit; }; };
- );
};
/* Zero the fields in the mr that are variant depending on usage */ static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr) {
- memset(mr->out, 0, sizeof(*mr) - offsetof(struct mlx5_ib_mr, out));
- memset(&mr->cleared, 0, sizeof(mr->cleared));
}
Why not use the memset_after(mr->umem) here?
I can certainly do that instead. In this series I've tended to opt for groupings so the position of future struct member additions are explicitly chosen. (i.e. reducing the chance that a zeroing of the new member be a surprise.)
-Kees
On Thu, Aug 19, 2021 at 09:19:08AM -0700, Kees Cook wrote:
On Thu, Aug 19, 2021 at 09:27:16AM -0300, Jason Gunthorpe wrote:
On Tue, Aug 17, 2021 at 11:05:26PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct mlx5_ib_mr that should be initialized to zero.
Cc: Leon Romanovsky leon@kernel.org Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bf20a388eabe..f63bf204a7a1 100644 +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -644,6 +644,7 @@ struct mlx5_ib_mr { struct ib_umem *umem;
/* This is zero'd when the MR is allocated */
- struct_group(cleared, union { /* Used only while the MR is in the cache */ struct {
@@ -691,12 +692,13 @@ struct mlx5_ib_mr { bool is_odp_implicit; }; };
- );
};
/* Zero the fields in the mr that are variant depending on usage */ static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr) {
- memset(mr->out, 0, sizeof(*mr) - offsetof(struct mlx5_ib_mr, out));
- memset(&mr->cleared, 0, sizeof(mr->cleared));
}
Why not use the memset_after(mr->umem) here?
I can certainly do that instead. In this series I've tended to opt for groupings so the position of future struct member additions are explicitly chosen. (i.e. reducing the chance that a zeroing of the new member be a surprise.)
I saw the earlier RDMA patches where using other memset techniques though? Were there flex arrays or something that made groups infeasible?
Jason
On Thu, Aug 19, 2021 at 01:47:57PM -0300, Jason Gunthorpe wrote:
On Thu, Aug 19, 2021 at 09:19:08AM -0700, Kees Cook wrote:
On Thu, Aug 19, 2021 at 09:27:16AM -0300, Jason Gunthorpe wrote:
On Tue, Aug 17, 2021 at 11:05:26PM -0700, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct mlx5_ib_mr that should be initialized to zero.
Cc: Leon Romanovsky leon@kernel.org Cc: Doug Ledford dledford@redhat.com Cc: Jason Gunthorpe jgg@ziepe.ca Cc: linux-rdma@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org drivers/infiniband/hw/mlx5/mlx5_ib.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bf20a388eabe..f63bf204a7a1 100644 +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -644,6 +644,7 @@ struct mlx5_ib_mr { struct ib_umem *umem;
/* This is zero'd when the MR is allocated */
- struct_group(cleared, union { /* Used only while the MR is in the cache */ struct {
@@ -691,12 +692,13 @@ struct mlx5_ib_mr { bool is_odp_implicit; }; };
- );
};
/* Zero the fields in the mr that are variant depending on usage */ static inline void mlx5_clear_mr(struct mlx5_ib_mr *mr) {
- memset(mr->out, 0, sizeof(*mr) - offsetof(struct mlx5_ib_mr, out));
- memset(&mr->cleared, 0, sizeof(mr->cleared));
}
Why not use the memset_after(mr->umem) here?
I can certainly do that instead. In this series I've tended to opt for groupings so the position of future struct member additions are explicitly chosen. (i.e. reducing the chance that a zeroing of the new member be a surprise.)
I saw the earlier RDMA patches where using other memset techniques though? Were there flex arrays or something that made groups infeasible?
Which do you mean? When doing the conversions I tended to opt for struct_group() since it provides more robust "intentionality". Strictly speaking, the new memset helpers are doing field-spanning writes, but the "clear to the end" pattern was so common it made sense to add the helpers, as they're a bit less disruptive. It's totally up to you! :)
On Thu, Aug 19, 2021 at 11:14:37AM -0700, Kees Cook wrote:
Which do you mean? When doing the conversions I tended to opt for struct_group() since it provides more robust "intentionality". Strictly speaking, the new memset helpers are doing field-spanning writes, but the "clear to the end" pattern was so common it made sense to add the helpers, as they're a bit less disruptive. It's totally up to you! :)
Well, of the patches you cc'd to me only this one used the struct group..
Jason
On Fri, Aug 20, 2021 at 09:34:00AM -0300, Jason Gunthorpe wrote:
On Thu, Aug 19, 2021 at 11:14:37AM -0700, Kees Cook wrote:
Which do you mean? When doing the conversions I tended to opt for struct_group() since it provides more robust "intentionality". Strictly speaking, the new memset helpers are doing field-spanning writes, but the "clear to the end" pattern was so common it made sense to add the helpers, as they're a bit less disruptive. It's totally up to you! :)
Well, of the patches you cc'd to me only this one used the struct group..
Understood. I've adjusted this for v3. Thanks!
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org --- arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE - unsigned long evr[32]; /* upper 32-bits of SPE regs */ - u64 acc; /* Accumulator */ + struct_group(spe, + unsigned long evr[32]; /* upper 32-bits of SPE regs */ + u64 acc; /* Accumulator */ + ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */ diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */ - unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs, - ELF_NEVRREG * sizeof(u32), failed); + unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs, + sizeof(current->thread.spe), failed); current->thread.used_spe = true; } else if (current->thread.used_spe) - memset(current->thread.evr, 0, ELF_NEVRREG * sizeof(u32)); + memset(¤t->thread.spe, 0, sizeof(current->thread.spe));
/* Always get SPEFSCR back */ unsafe_get_user(current->thread.spefscr, (u32 __user *)&sr->mc_vregs + ELF_NEVRREG, failed);
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org
arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE
- unsigned long evr[32]; /* upper 32-bits of SPE regs */
- u64 acc; /* Accumulator */
- struct_group(spe,
unsigned long evr[32]; /* upper 32-bits of SPE regs */
u64 acc; /* Accumulator */
- ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */
unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
ELF_NEVRREG * sizeof(u32), failed);
unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs,
sizeof(current->thread.spe), failed);
This makes me nervous, because the ABI is that we copy ELF_NEVRREG * sizeof(u32) bytes, not whatever sizeof(current->thread.spe) happens to be.
ie. if we use sizeof an inadvertent change to the fields in thread_struct could change how many bytes we copy out to userspace, which would be an ABI break.
And that's not that hard to do, because it's not at all obvious that the size and layout of fields in thread_struct affects the user ABI.
At the same time we don't want to copy the right number of bytes but the wrong content, so from that point of view using sizeof is good :)
The way we handle it in ptrace is to have BUILD_BUG_ON()s to verify that things match up, so maybe we should do that here too.
ie. add:
BUILD_BUG_ON(sizeof(current->thread.spe) == ELF_NEVRREG * sizeof(u32));
Not sure if you are happy doing that as part of this patch. I can always do it later if not.
cheers
Le 20/08/2021 à 09:49, Michael Ellerman a écrit :
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org
arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE
- unsigned long evr[32]; /* upper 32-bits of SPE regs */
- u64 acc; /* Accumulator */
- struct_group(spe,
unsigned long evr[32]; /* upper 32-bits of SPE regs */
u64 acc; /* Accumulator */
- ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */
unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
ELF_NEVRREG * sizeof(u32), failed);
unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs,
sizeof(current->thread.spe), failed);
This makes me nervous, because the ABI is that we copy ELF_NEVRREG * sizeof(u32) bytes, not whatever sizeof(current->thread.spe) happens to be.
ie. if we use sizeof an inadvertent change to the fields in thread_struct could change how many bytes we copy out to userspace, which would be an ABI break.
And that's not that hard to do, because it's not at all obvious that the size and layout of fields in thread_struct affects the user ABI.
At the same time we don't want to copy the right number of bytes but the wrong content, so from that point of view using sizeof is good :)
The way we handle it in ptrace is to have BUILD_BUG_ON()s to verify that things match up, so maybe we should do that here too.
ie. add:
BUILD_BUG_ON(sizeof(current->thread.spe) == ELF_NEVRREG * sizeof(u32));
You mean != I guess ?
Not sure if you are happy doing that as part of this patch. I can always do it later if not.
cheers
Christophe Leroy christophe.leroy@csgroup.eu writes:
Le 20/08/2021 à 09:49, Michael Ellerman a écrit :
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org
arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE
- unsigned long evr[32]; /* upper 32-bits of SPE regs */
- u64 acc; /* Accumulator */
- struct_group(spe,
unsigned long evr[32]; /* upper 32-bits of SPE regs */
u64 acc; /* Accumulator */
- ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */
unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
ELF_NEVRREG * sizeof(u32), failed);
unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs,
sizeof(current->thread.spe), failed);
This makes me nervous, because the ABI is that we copy ELF_NEVRREG * sizeof(u32) bytes, not whatever sizeof(current->thread.spe) happens to be.
ie. if we use sizeof an inadvertent change to the fields in thread_struct could change how many bytes we copy out to userspace, which would be an ABI break.
And that's not that hard to do, because it's not at all obvious that the size and layout of fields in thread_struct affects the user ABI.
At the same time we don't want to copy the right number of bytes but the wrong content, so from that point of view using sizeof is good :)
The way we handle it in ptrace is to have BUILD_BUG_ON()s to verify that things match up, so maybe we should do that here too.
ie. add:
BUILD_BUG_ON(sizeof(current->thread.spe) == ELF_NEVRREG * sizeof(u32));
You mean != I guess ?
Gah. Yes I do :)
cheers
On Fri, Aug 20, 2021 at 05:49:35PM +1000, Michael Ellerman wrote:
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org
arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE
- unsigned long evr[32]; /* upper 32-bits of SPE regs */
- u64 acc; /* Accumulator */
- struct_group(spe,
unsigned long evr[32]; /* upper 32-bits of SPE regs */
u64 acc; /* Accumulator */
- ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */
unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
ELF_NEVRREG * sizeof(u32), failed);
unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs,
sizeof(current->thread.spe), failed);
This makes me nervous, because the ABI is that we copy ELF_NEVRREG * sizeof(u32) bytes, not whatever sizeof(current->thread.spe) happens to be.
ie. if we use sizeof an inadvertent change to the fields in thread_struct could change how many bytes we copy out to userspace, which would be an ABI break.
And that's not that hard to do, because it's not at all obvious that the size and layout of fields in thread_struct affects the user ABI.
At the same time we don't want to copy the right number of bytes but the wrong content, so from that point of view using sizeof is good :)
The way we handle it in ptrace is to have BUILD_BUG_ON()s to verify that things match up, so maybe we should do that here too.
ie. add:
BUILD_BUG_ON(sizeof(current->thread.spe) == ELF_NEVRREG * sizeof(u32));
Not sure if you are happy doing that as part of this patch. I can always do it later if not.
Sounds good to me; I did that in a few other cases in the series where the relationships between things seemed tenuous. :) I'll add this (as !=) in v3.
Thanks!
Kees Cook keescook@chromium.org writes:
On Fri, Aug 20, 2021 at 05:49:35PM +1000, Michael Ellerman wrote:
Kees Cook keescook@chromium.org writes:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add a struct_group() for the spe registers so that memset() can correctly reason about the size:
In function 'fortify_memset_chk', inlined from 'restore_user_regs.part.0' at arch/powerpc/kernel/signal_32.c:539:3:
include/linux/fortify-string.h:195:4: error: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Werror=attribute-warning]
195 | __write_overflow_field(); | ^~~~~~~~~~~~~~~~~~~~~~~~
Cc: Michael Ellerman mpe@ellerman.id.au Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Paul Mackerras paulus@samba.org Cc: Christophe Leroy christophe.leroy@csgroup.eu Cc: Sudeep Holla sudeep.holla@arm.com Cc: linuxppc-dev@lists.ozlabs.org Reported-by: kernel test robot lkp@intel.com Signed-off-by: Kees Cook keescook@chromium.org
arch/powerpc/include/asm/processor.h | 6 ++++-- arch/powerpc/kernel/signal_32.c | 6 +++--- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h index f348e564f7dd..05dc567cb9a8 100644 --- a/arch/powerpc/include/asm/processor.h +++ b/arch/powerpc/include/asm/processor.h @@ -191,8 +191,10 @@ struct thread_struct { int used_vsr; /* set if process has used VSX */ #endif /* CONFIG_VSX */ #ifdef CONFIG_SPE
- unsigned long evr[32]; /* upper 32-bits of SPE regs */
- u64 acc; /* Accumulator */
- struct_group(spe,
unsigned long evr[32]; /* upper 32-bits of SPE regs */
u64 acc; /* Accumulator */
- ); unsigned long spefscr; /* SPE & eFP status */ unsigned long spefscr_last; /* SPEFSCR value on last prctl call or trap return */
diff --git a/arch/powerpc/kernel/signal_32.c b/arch/powerpc/kernel/signal_32.c index 0608581967f0..77b86caf5c51 100644 --- a/arch/powerpc/kernel/signal_32.c +++ b/arch/powerpc/kernel/signal_32.c @@ -532,11 +532,11 @@ static long restore_user_regs(struct pt_regs *regs, regs_set_return_msr(regs, regs->msr & ~MSR_SPE); if (msr & MSR_SPE) { /* restore spe registers from the stack */
unsafe_copy_from_user(current->thread.evr, &sr->mc_vregs,
ELF_NEVRREG * sizeof(u32), failed);
unsafe_copy_from_user(¤t->thread.spe, &sr->mc_vregs,
sizeof(current->thread.spe), failed);
This makes me nervous, because the ABI is that we copy ELF_NEVRREG * sizeof(u32) bytes, not whatever sizeof(current->thread.spe) happens to be.
ie. if we use sizeof an inadvertent change to the fields in thread_struct could change how many bytes we copy out to userspace, which would be an ABI break.
And that's not that hard to do, because it's not at all obvious that the size and layout of fields in thread_struct affects the user ABI.
At the same time we don't want to copy the right number of bytes but the wrong content, so from that point of view using sizeof is good :)
The way we handle it in ptrace is to have BUILD_BUG_ON()s to verify that things match up, so maybe we should do that here too.
ie. add:
BUILD_BUG_ON(sizeof(current->thread.spe) == ELF_NEVRREG * sizeof(u32));
Not sure if you are happy doing that as part of this patch. I can always do it later if not.
Sounds good to me; I did that in a few other cases in the series where the relationships between things seemed tenuous. :) I'll add this (as !=) in v3.
Thanks.
cheers
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark region of struct stats_reply_data that should be initialized, which can now be done in a single memset() call.
Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: Ido Schimmel idosch@nvidia.com Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- net/ethtool/stats.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/net/ethtool/stats.c b/net/ethtool/stats.c index ec07f5765e03..a20e0a24ff61 100644 --- a/net/ethtool/stats.c +++ b/net/ethtool/stats.c @@ -14,10 +14,12 @@ struct stats_req_info {
struct stats_reply_data { struct ethnl_reply_data base; - struct ethtool_eth_phy_stats phy_stats; - struct ethtool_eth_mac_stats mac_stats; - struct ethtool_eth_ctrl_stats ctrl_stats; - struct ethtool_rmon_stats rmon_stats; + struct_group(stats, + struct ethtool_eth_phy_stats phy_stats; + struct ethtool_eth_mac_stats mac_stats; + struct ethtool_eth_ctrl_stats ctrl_stats; + struct ethtool_rmon_stats rmon_stats; + ); const struct ethtool_rmon_hist_range *rmon_ranges; };
@@ -117,10 +119,7 @@ static int stats_prepare_data(const struct ethnl_req_info *req_base, /* Mark all stats as unset (see ETHTOOL_STAT_NOT_SET) to prevent them * from being reported to user space in case driver did not set them. */ - memset(&data->phy_stats, 0xff, sizeof(data->phy_stats)); - memset(&data->mac_stats, 0xff, sizeof(data->mac_stats)); - memset(&data->ctrl_stats, 0xff, sizeof(data->ctrl_stats)); - memset(&data->rmon_stats, 0xff, sizeof(data->rmon_stats)); + memset(&data->stats, 0xff, sizeof(data->stats));
if (test_bit(ETHTOOL_STATS_ETH_PHY, req_info->stat_mask) && dev->ethtool_ops->get_eth_phy_stats)
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark both regions of struct flexcan_regs that get initialized to zero. Avoid the future warnings:
In function 'fortify_memset_chk', inlined from 'memset_io' at ./include/asm-generic/io.h:1169:2, inlined from 'flexcan_ram_init' at drivers/net/can/flexcan.c:1403:2: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'fortify_memset_chk', inlined from 'memset_io' at ./include/asm-generic/io.h:1169:2, inlined from 'flexcan_ram_init' at drivers/net/can/flexcan.c:1408:3: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Cc: Wolfgang Grandegger wg@grandegger.com Cc: Marc Kleine-Budde mkl@pengutronix.de Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-can@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- drivers/net/can/flexcan.c | 68 +++++++++++++++++++-------------------- 1 file changed, 34 insertions(+), 34 deletions(-)
diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c index 7734229aa078..12b60ad95b02 100644 --- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -290,31 +290,33 @@ struct flexcan_regs { u32 dbg1; /* 0x58 */ u32 dbg2; /* 0x5c */ u32 _reserved3[8]; /* 0x60 */ - u8 mb[2][512]; /* 0x80 - Not affected by Soft Reset */ - /* FIFO-mode: - * MB - * 0x080...0x08f 0 RX message buffer - * 0x090...0x0df 1-5 reserved - * 0x0e0...0x0ff 6-7 8 entry ID table - * (mx25, mx28, mx35, mx53) - * 0x0e0...0x2df 6-7..37 8..128 entry ID table - * size conf'ed via ctrl2::RFFN - * (mx6, vf610) - */ - u32 _reserved4[256]; /* 0x480 */ - u32 rximr[64]; /* 0x880 - Not affected by Soft Reset */ - u32 _reserved5[24]; /* 0x980 */ - u32 gfwr_mx6; /* 0x9e0 - MX6 */ - u32 _reserved6[39]; /* 0x9e4 */ - u32 _rxfir[6]; /* 0xa80 */ - u32 _reserved8[2]; /* 0xa98 */ - u32 _rxmgmask; /* 0xaa0 */ - u32 _rxfgmask; /* 0xaa4 */ - u32 _rx14mask; /* 0xaa8 */ - u32 _rx15mask; /* 0xaac */ - u32 tx_smb[4]; /* 0xab0 */ - u32 rx_smb0[4]; /* 0xac0 */ - u32 rx_smb1[4]; /* 0xad0 */ + struct_group(init, + u8 mb[2][512]; /* 0x80 - Not affected by Soft Reset */ + /* FIFO-mode: + * MB + * 0x080...0x08f 0 RX message buffer + * 0x090...0x0df 1-5 reserved + * 0x0e0...0x0ff 6-7 8 entry ID table + * (mx25, mx28, mx35, mx53) + * 0x0e0...0x2df 6-7..37 8..128 entry ID table + * size conf'ed via ctrl2::RFFN + * (mx6, vf610) + */ + u32 _reserved4[256]; /* 0x480 */ + u32 rximr[64]; /* 0x880 - Not affected by Soft Reset */ + u32 _reserved5[24]; /* 0x980 */ + u32 gfwr_mx6; /* 0x9e0 - MX6 */ + u32 _reserved6[39]; /* 0x9e4 */ + u32 _rxfir[6]; /* 0xa80 */ + u32 _reserved8[2]; /* 0xa98 */ + u32 _rxmgmask; /* 0xaa0 */ + u32 _rxfgmask; /* 0xaa4 */ + u32 _rx14mask; /* 0xaa8 */ + u32 _rx15mask; /* 0xaac */ + u32 tx_smb[4]; /* 0xab0 */ + u32 rx_smb0[4]; /* 0xac0 */ + u32 rx_smb1[4]; /* 0xad0 */ + ); u32 mecr; /* 0xae0 */ u32 erriar; /* 0xae4 */ u32 erridpr; /* 0xae8 */ @@ -328,9 +330,11 @@ struct flexcan_regs { u32 fdcbt; /* 0xc04 - Not affected by Soft Reset */ u32 fdcrc; /* 0xc08 */ u32 _reserved9[199]; /* 0xc0c */ - u32 tx_smb_fd[18]; /* 0xf28 */ - u32 rx_smb0_fd[18]; /* 0xf70 */ - u32 rx_smb1_fd[18]; /* 0xfb8 */ + struct_group(init_fd, + u32 tx_smb_fd[18]; /* 0xf28 */ + u32 rx_smb0_fd[18]; /* 0xf70 */ + u32 rx_smb1_fd[18]; /* 0xfb8 */ + ); };
static_assert(sizeof(struct flexcan_regs) == 0x4 * 18 + 0xfb8); @@ -1400,14 +1404,10 @@ static void flexcan_ram_init(struct net_device *dev) reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ; priv->write(reg_ctrl2, ®s->ctrl2);
- memset_io(®s->mb[0][0], 0, - offsetof(struct flexcan_regs, rx_smb1[3]) - - offsetof(struct flexcan_regs, mb[0][0]) + 0x4); + memset_io(®s->init, 0, sizeof(regs->init));
if (priv->can.ctrlmode & CAN_CTRLMODE_FD) - memset_io(®s->tx_smb_fd[0], 0, - offsetof(struct flexcan_regs, rx_smb1_fd[17]) - - offsetof(struct flexcan_regs, tx_smb_fd[0]) + 0x4); + memset_io(®s->init_fd, 0, sizeof(regs->init_fd));
reg_ctrl2 &= ~FLEXCAN_CTRL2_WRMFRZ; priv->write(reg_ctrl2, ®s->ctrl2);
On 17.08.2021 23:05:29, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark both regions of struct flexcan_regs that get initialized to zero. Avoid the future warnings:
In function 'fortify_memset_chk', inlined from 'memset_io' at ./include/asm-generic/io.h:1169:2, inlined from 'flexcan_ram_init' at drivers/net/can/flexcan.c:1403:2: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In function 'fortify_memset_chk', inlined from 'memset_io' at ./include/asm-generic/io.h:1169:2, inlined from 'flexcan_ram_init' at drivers/net/can/flexcan.c:1408:3: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Cc: Wolfgang Grandegger wg@grandegger.com Cc: Marc Kleine-Budde mkl@pengutronix.de Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-can@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
Acked-by: Marc Kleine-Budde mkl@pengutronix.de
- memset_io(®s->mb[0][0], 0,
offsetof(struct flexcan_regs, rx_smb1[3]) -
offsetof(struct flexcan_regs, mb[0][0]) + 0x4);
- memset_io(®s->init, 0, sizeof(regs->init));
This even makes the code more readable! \o/
if (priv->can.ctrlmode & CAN_CTRLMODE_FD)
memset_io(®s->tx_smb_fd[0], 0,
offsetof(struct flexcan_regs, rx_smb1_fd[17]) -
offsetof(struct flexcan_regs, tx_smb_fd[0]) + 0x4);
memset_io(®s->init_fd, 0, sizeof(regs->init_fd));
reg_ctrl2 &= ~FLEXCAN_CTRL2_WRMFRZ; priv->write(reg_ctrl2, ®s->ctrl2);
regards, Marc
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark the region of struct iucv_sock that gets initialized to zero. Avoid the future warning:
In function 'fortify_memset_chk', inlined from 'iucv_sock_alloc' at net/iucv/af_iucv.c:476:2: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Cc: Julian Wiedmann jwi@linux.ibm.com Cc: Karsten Graul kgraul@linux.ibm.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-s390@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org --- include/net/iucv/af_iucv.h | 10 ++++++---- net/iucv/af_iucv.c | 2 +- 2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/include/net/iucv/af_iucv.h b/include/net/iucv/af_iucv.h index ff06246dbbb9..df85d19fbf84 100644 --- a/include/net/iucv/af_iucv.h +++ b/include/net/iucv/af_iucv.h @@ -112,10 +112,12 @@ enum iucv_tx_notify {
struct iucv_sock { struct sock sk; - char src_user_id[8]; - char src_name[8]; - char dst_user_id[8]; - char dst_name[8]; + struct_group(init, + char src_user_id[8]; + char src_name[8]; + char dst_user_id[8]; + char dst_name[8]; + ); struct list_head accept_q; spinlock_t accept_q_lock; struct sock *parent; diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 18316ee3c692..9446e2771d31 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -473,7 +473,7 @@ static struct sock *iucv_sock_alloc(struct socket *sock, int proto, gfp_t prio, atomic_set(&iucv->msg_recv, 0); iucv->path = NULL; iucv->sk_txnotify = afiucv_hs_callback_txnotify; - memset(&iucv->src_user_id , 0, 32); + memset(&iucv->init, 0, sizeof(iucv->init)); if (pr_iucv) iucv->transport = AF_IUCV_TRANS_IUCV; else
On 18/08/2021 08:05, Kees Cook wrote:
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Add struct_group() to mark the region of struct iucv_sock that gets initialized to zero. Avoid the future warning:
In function 'fortify_memset_chk', inlined from 'iucv_sock_alloc' at net/iucv/af_iucv.c:476:2: ./include/linux/fortify-string.h:199:4: warning: call to '__write_overflow_field' declared with attribute warning: detected write beyond size of field (1st parameter); maybe use struct_group()? [-Wattribute-warning] 199 | __write_overflow_field(p_size_field, size); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Cc: Julian Wiedmann jwi@linux.ibm.com Cc: Karsten Graul kgraul@linux.ibm.com Cc: "David S. Miller" davem@davemloft.net Cc: Jakub Kicinski kuba@kernel.org Cc: linux-s390@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Kees Cook keescook@chromium.org
include/net/iucv/af_iucv.h | 10 ++++++---- net/iucv/af_iucv.c | 2 +- 2 files changed, 7 insertions(+), 5 deletions(-)
No objections. Acked-by: Karsten Graul kgraul@linux.ibm.com
Thank you.
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Instead of writing across a field boundary with memset(), move the call to just the array, and an explicit zeroing of the prior field.
Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Qinglang Miao miaoqinglang@huawei.com Cc: "Gustavo A. R. Silva" gustavoars@kernel.org Cc: Hulk Robot hulkci@huawei.com Cc: Wang Wensheng wangwensheng4@huawei.com Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Michael Ellerman mpe@ellerman.id.au Link: https://lore.kernel.org/lkml/87czqsnmw9.fsf@mpe.ellerman.id.au --- drivers/macintosh/smu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index 94fb63a7b357..59ce431da7ef 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -848,7 +848,8 @@ int smu_queue_i2c(struct smu_i2c_cmd *cmd) cmd->read = cmd->info.devaddr & 0x01; switch(cmd->info.type) { case SMU_I2C_TRANSFER_SIMPLE: - memset(&cmd->info.sublen, 0, 4); + cmd->info.sublen = 0; + memset(&cmd->info.subaddr, 0, 3); break; case SMU_I2C_TRANSFER_COMBINED: cmd->info.devaddr &= 0xfe;
Le 18/08/2021 à 08:05, Kees Cook a écrit :
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Instead of writing across a field boundary with memset(), move the call to just the array, and an explicit zeroing of the prior field.
Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Qinglang Miao miaoqinglang@huawei.com Cc: "Gustavo A. R. Silva" gustavoars@kernel.org Cc: Hulk Robot hulkci@huawei.com Cc: Wang Wensheng wangwensheng4@huawei.com Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Michael Ellerman mpe@ellerman.id.au Link: https://lore.kernel.org/lkml/87czqsnmw9.fsf@mpe.ellerman.id.au
drivers/macintosh/smu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index 94fb63a7b357..59ce431da7ef 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -848,7 +848,8 @@ int smu_queue_i2c(struct smu_i2c_cmd *cmd) cmd->read = cmd->info.devaddr & 0x01; switch(cmd->info.type) { case SMU_I2C_TRANSFER_SIMPLE:
memset(&cmd->info.sublen, 0, 4);
cmd->info.sublen = 0;
memset(&cmd->info.subaddr, 0, 3);
subaddr[] is a table, should the & be avoided ? And while at it, why not use sizeof(subaddr) instead of 3 ?
break;
case SMU_I2C_TRANSFER_COMBINED: cmd->info.devaddr &= 0xfe;
On Wed, Aug 18, 2021 at 08:42:18AM +0200, Christophe Leroy wrote:
Le 18/08/2021 à 08:05, Kees Cook a écrit :
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memset(), avoid intentionally writing across neighboring fields.
Instead of writing across a field boundary with memset(), move the call to just the array, and an explicit zeroing of the prior field.
Cc: Benjamin Herrenschmidt benh@kernel.crashing.org Cc: Qinglang Miao miaoqinglang@huawei.com Cc: "Gustavo A. R. Silva" gustavoars@kernel.org Cc: Hulk Robot hulkci@huawei.com Cc: Wang Wensheng wangwensheng4@huawei.com Cc: linuxppc-dev@lists.ozlabs.org Signed-off-by: Kees Cook keescook@chromium.org Reviewed-by: Michael Ellerman mpe@ellerman.id.au Link: https://lore.kernel.org/lkml/87czqsnmw9.fsf@mpe.ellerman.id.au
drivers/macintosh/smu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/macintosh/smu.c b/drivers/macintosh/smu.c index 94fb63a7b357..59ce431da7ef 100644 --- a/drivers/macintosh/smu.c +++ b/drivers/macintosh/smu.c @@ -848,7 +848,8 @@ int smu_queue_i2c(struct smu_i2c_cmd *cmd) cmd->read = cmd->info.devaddr & 0x01; switch(cmd->info.type) { case SMU_I2C_TRANSFER_SIMPLE:
memset(&cmd->info.sublen, 0, 4);
cmd->info.sublen = 0;
memset(&cmd->info.subaddr, 0, 3);
subaddr[] is a table, should the & be avoided ?
It results in the same thing, but it's better form to not have the &; I will fix this.
And while at it, why not use sizeof(subaddr) instead of 3 ?
Agreed. :)
Thanks!
As done for memcpy(), also update memset() to use the same tightened compile-time bounds checking under CONFIG_FORTIFY_SOURCE.
Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 54 ++++++++++++++++--- .../write_overflow_field-memset.c | 5 ++ 2 files changed, 51 insertions(+), 8 deletions(-) create mode 100644 lib/test_fortify/write_overflow_field-memset.c
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index 0120d463ba33..7de4673dfe2c 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -198,17 +198,56 @@ __FORTIFY_INLINE char *strncat(char *p, const char *q, __kernel_size_t count) return p; }
-__FORTIFY_INLINE void *memset(void *p, int c, __kernel_size_t size) +__FORTIFY_INLINE void fortify_memset_chk(__kernel_size_t size, + const size_t p_size, + const size_t p_size_field) { - size_t p_size = __builtin_object_size(p, 0); + if (__builtin_constant_p(size)) { + /* + * Length argument is a constant expression, so we + * can perform compile-time bounds checking where + * buffer sizes are known. + */
- if (__builtin_constant_p(size) && p_size < size) - __write_overflow(); - if (p_size < size) - fortify_panic(__func__); - return __underlying_memset(p, c, size); + /* Error when size is larger than enclosing struct. */ + if (p_size > p_size_field && p_size < size) + __write_overflow(); + + /* Warn when write size is larger than dest field. */ + if (p_size_field < size) + __write_overflow_field(p_size_field, size); + } + /* + * At this point, length argument may not be a constant expression, + * so run-time bounds checking can be done where buffer sizes are + * known. (This is not an "else" because the above checks may only + * be compile-time warnings, and we want to still warn for run-time + * overflows.) + */ + + /* + * Always stop accesses beyond the struct that contains the + * field, when the buffer's remaining size is known. + * (The -1 test is to optimize away checks where the buffer + * lengths are unknown.) + */ + if (p_size != (size_t)(-1) && p_size < size) + fortify_panic("memset"); }
+#define __fortify_memset_chk(p, c, size, p_size, p_size_field) ({ \ + size_t __fortify_size = (size_t)(size); \ + fortify_memset_chk(__fortify_size, p_size, p_size_field), \ + __underlying_memset(p, c, __fortify_size); \ +}) + +/* + * __builtin_object_size() must be captured here to avoid evaluating argument + * side-effects further into the macro layers. + */ +#define memset(p, c, s) __fortify_memset_chk(p, c, s, \ + __builtin_object_size(p, 0), __builtin_object_size(p, 1)) + /* * To make sure the compiler can enforce protection against buffer overflows, * memcpy(), memmove(), and memset() must not be used beyond individual @@ -399,7 +438,6 @@ __FORTIFY_INLINE char *strcpy(char *p, const char *q) /* Don't use these outside the FORITFY_SOURCE implementation */ #undef __underlying_memchr #undef __underlying_memcmp -#undef __underlying_memset #undef __underlying_strcat #undef __underlying_strcpy #undef __underlying_strlen diff --git a/lib/test_fortify/write_overflow_field-memset.c b/lib/test_fortify/write_overflow_field-memset.c new file mode 100644 index 000000000000..2331da26909e --- /dev/null +++ b/lib/test_fortify/write_overflow_field-memset.c @@ -0,0 +1,5 @@ +// SPDX-License-Identifier: GPL-2.0-only +#define TEST \ + memset(instance.buf, 0x42, sizeof(instance.buf) + 1) + +#include "test_fortify.h"
To enable FORTIFY_SOURCE support for Clang, the kernel must work around a pair of bugs, related to Clang's inlining.
Change all the fortified APIs into macros with different inline names to bypass Clang's broken inline-of-a-builtin detection: https://bugs.llvm.org/show_bug.cgi?id=50322
Lift all misbehaving __builtin_object_size() calls into the macros to bypass Clang's broken __builtin_object_size() arguments-of-an-inline visibility: https://github.com/ClangBuiltLinux/linux/issues/1401
Thankfully, due to how the inlining already behaves in GCC, this change has no effect on GCC builds, but allows Clang to finally gain full FORTIFY coverage.
However, because of a third bug which had no work-arounds, FORTIFY_SOURCE will only work with Clang version 13 and later. Update the Kconfig to reflect the new requirements.
Signed-off-by: Kees Cook keescook@chromium.org --- include/linux/fortify-string.h | 55 +++++++++++++++++++++------------- security/Kconfig | 2 +- 2 files changed, 36 insertions(+), 21 deletions(-)
diff --git a/include/linux/fortify-string.h b/include/linux/fortify-string.h index 7de4673dfe2c..e62d3633a329 100644 --- a/include/linux/fortify-string.h +++ b/include/linux/fortify-string.h @@ -48,10 +48,10 @@ extern char *__underlying_strncpy(char *p, const char *q, __kernel_size_t size) #define __underlying_strncpy __builtin_strncpy #endif
-__FORTIFY_INLINE char *strncpy(char *p, const char *q, __kernel_size_t size) +#define strncpy(p, q, s) __fortify_strncpy(p, q, s, __builtin_object_size(p, 1)) +__FORTIFY_INLINE char *__fortify_strncpy(char *p, const char *q, + __kernel_size_t size, size_t p_size) { - size_t p_size = __builtin_object_size(p, 1); - if (__builtin_constant_p(size) && p_size < size) __write_overflow(); if (p_size < size) @@ -71,9 +71,10 @@ __FORTIFY_INLINE char *strcat(char *p, const char *q) }
extern __kernel_size_t __real_strnlen(const char *, __kernel_size_t) __RENAME(strnlen); -__FORTIFY_INLINE __kernel_size_t strnlen(const char *p, __kernel_size_t maxlen) +#define strnlen(p, s) __fortify_strnlen(p, s, __builtin_object_size(p, 1)) +__FORTIFY_INLINE __kernel_size_t __fortify_strnlen(const char *p, size_t maxlen, + size_t p_size) { - size_t p_size = __builtin_object_size(p, 1); size_t p_len = __compiletime_strlen(p); size_t ret;
@@ -108,10 +109,14 @@ __FORTIFY_INLINE __kernel_size_t strlen(const char *p)
/* defined after fortified strlen to reuse it */ extern size_t __real_strlcpy(char *, const char *, size_t) __RENAME(strlcpy); -__FORTIFY_INLINE size_t strlcpy(char *p, const char *q, size_t size) +#define strlcpy(p, q, s) __fortify_strlcpy(p, q, s, \ + __builtin_object_size(p, 1), \ + __builtin_object_size(q, 1)) +__FORTIFY_INLINE size_t __fortify_strlcpy(char *p, const char *q, + size_t size, + const size_t p_size, + const size_t q_size) { - size_t p_size = __builtin_object_size(p, 1); - size_t q_size = __builtin_object_size(q, 1); size_t q_len; /* Full count of source string length. */ size_t len; /* Count of characters going into destination. */
@@ -135,12 +140,15 @@ __FORTIFY_INLINE size_t strlcpy(char *p, const char *q, size_t size)
/* defined after fortified strnlen to reuse it */ extern ssize_t __real_strscpy(char *, const char *, size_t) __RENAME(strscpy); -__FORTIFY_INLINE ssize_t strscpy(char *p, const char *q, size_t size) +#define strscpy(p, q, s) __fortify_strscpy(p, q, s, \ + __builtin_object_size(p, 1), \ + __builtin_object_size(q, 1)) +__FORTIFY_INLINE ssize_t __fortify_strscpy(char *p, const char *q, + size_t size, + const size_t p_size, + const size_t q_size) { size_t len; - /* Use string size rather than possible enclosing struct size. */ - size_t p_size = __builtin_object_size(p, 1); - size_t q_size = __builtin_object_size(q, 1);
/* If we cannot get size of p and q default to call strscpy. */ if (p_size == (size_t) -1 && q_size == (size_t) -1) @@ -181,11 +189,13 @@ __FORTIFY_INLINE ssize_t strscpy(char *p, const char *q, size_t size) }
/* defined after fortified strlen and strnlen to reuse them */ -__FORTIFY_INLINE char *strncat(char *p, const char *q, __kernel_size_t count) +#define strncat(p, q, count) __fortify_strncat(p, q, count, \ + __builtin_object_size(p, 1), \ + __builtin_object_size(q, 1)) +__FORTIFY_INLINE char *__fortify_strncat(char *p, const char *q, size_t count, + size_t p_size, size_t q_size) { size_t p_len, copy_len; - size_t p_size = __builtin_object_size(p, 1); - size_t q_size = __builtin_object_size(q, 1);
if (p_size == (size_t)-1 && q_size == (size_t)-1) return __underlying_strncat(p, q, count); @@ -352,7 +362,8 @@ __FORTIFY_INLINE void fortify_memcpy_chk(__kernel_size_t size, memmove)
extern void *__real_memscan(void *, int, __kernel_size_t) __RENAME(memscan); -__FORTIFY_INLINE void *memscan(void *p, int c, __kernel_size_t size) +#define memscan(p, c, s) __fortify_memscan(p, c, s) +__FORTIFY_INLINE void *__fortify_memscan(void *p, int c, __kernel_size_t size) { size_t p_size = __builtin_object_size(p, 0);
@@ -363,7 +374,8 @@ __FORTIFY_INLINE void *memscan(void *p, int c, __kernel_size_t size) return __real_memscan(p, c, size); }
-__FORTIFY_INLINE int memcmp(const void *p, const void *q, __kernel_size_t size) +#define memcmp(p, q, s) __fortify_memcmp(p, q, s) +__FORTIFY_INLINE int __fortify_memcmp(const void *p, const void *q, __kernel_size_t size) { size_t p_size = __builtin_object_size(p, 0); size_t q_size = __builtin_object_size(q, 0); @@ -379,7 +391,8 @@ __FORTIFY_INLINE int memcmp(const void *p, const void *q, __kernel_size_t size) return __underlying_memcmp(p, q, size); }
-__FORTIFY_INLINE void *memchr(const void *p, int c, __kernel_size_t size) +#define memchr(p, c, s) __fortify_memchr(p, c, s) +__FORTIFY_INLINE void *__fortify_memchr(const void *p, int c, __kernel_size_t size) { size_t p_size = __builtin_object_size(p, 0);
@@ -391,7 +404,8 @@ __FORTIFY_INLINE void *memchr(const void *p, int c, __kernel_size_t size) }
void *__real_memchr_inv(const void *s, int c, size_t n) __RENAME(memchr_inv); -__FORTIFY_INLINE void *memchr_inv(const void *p, int c, size_t size) +#define memchr_inv(p, c, s) __fortify_memchr_inv(p, c, s) +__FORTIFY_INLINE void *__fortify_memchr_inv(const void *p, int c, size_t size) { size_t p_size = __builtin_object_size(p, 0);
@@ -415,7 +429,8 @@ __FORTIFY_INLINE void *kmemdup(const void *p, size_t size, gfp_t gfp) }
/* Defined after fortified strlen to reuse it. */ -__FORTIFY_INLINE char *strcpy(char *p, const char *q) +#define strcpy(p, q) __fortify_strcpy(p, q) +__FORTIFY_INLINE char *__fortify_strcpy(char *p, const char *q) { size_t p_size = __builtin_object_size(p, 1); size_t q_size = __builtin_object_size(q, 1); diff --git a/security/Kconfig b/security/Kconfig index fe6c0395fa02..530a15566b1d 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -193,7 +193,7 @@ config FORTIFY_SOURCE depends on ARCH_HAS_FORTIFY_SOURCE # https://bugs.llvm.org/show_bug.cgi?id=50322 # https://bugs.llvm.org/show_bug.cgi?id=41459 - depends on !CC_IS_CLANG + depends on !CC_IS_CLANG || CLANG_VERSION >= 130000 help Detect overflows of buffers in common string and memory functions where the compiler can determine and validate the buffer sizes.
dri-devel@lists.freedesktop.org