dri-devel January 2018

dri-devel@lists.freedesktop.org

226 participants
1677 discussions

[PATCH] drm/radeon: fill in rb backend map on evergreen/ni.
by Dave Airlie 11 Jan '18

11 Jan '18

From: Dave Airlie <airlied(a)redhat.com> This looks to have never gotten filled in, and it seems to trigger a bug in mesa. Reported-by: Roland Scheidegger <sroland(a)vmware.com> Signed-off-by: Dave Airlie <airlied(a)redhat.com> --- drivers/gpu/drm/radeon/evergreen.c | 1 + drivers/gpu/drm/radeon/ni.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 24fe66c..5712d63 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -3513,6 +3513,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev) tmp = r6xx_remap_render_backend(rdev, tmp, rdev->config.evergreen.max_backends, EVERGREEN_MAX_BACKENDS, disabled_rb_mask); } + rdev->config.evergreen.backend_map = tmp; WREG32(GB_BACKEND_MAP, tmp); WREG32(CGTS_SYS_TCC_DISABLE, 0); diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 9eccd0c..381b0255 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -1148,6 +1148,7 @@ static void cayman_gpu_init(struct radeon_device *rdev) rdev->config.cayman.max_shader_engines, CAYMAN_MAX_BACKENDS, disabled_rb_mask); } + rdev->config.cayman.backend_map = tmp; WREG32(GB_BACKEND_MAP, tmp); cgts_tcc_disable = 0xffff0000; -- 2.9.5

3 2

[PATCH] drm/amdkfd: Fix potential NULL pointer dereferences
by Gustavo A. R. Silva 10 Jan '18

10 Jan '18

In case kfd_get_process_device_data returns null, there are some null pointer dereferences in functions kfd_bind_processes_to_device and kfd_unbind_processes_from_device. Fix this by null checking pdd before dereferencing it. Addresses-Coverity-ID: 1463794 ("Dereference null return value") Addresses-Coverity-ID: 1463772 ("Dereference null return value") Signed-off-by: Gustavo A. R. Silva <garsilva(a)embeddedor.com> --- drivers/gpu/drm/amd/amdkfd/kfd_process.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index a22fb071..29d51d5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -461,6 +461,13 @@ int kfd_bind_processes_to_device(struct kfd_dev *dev) hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { mutex_lock(&p->mutex); pdd = kfd_get_process_device_data(dev, p); + + if (!pdd) { + pr_err("Process device data doesn't exist\n"); + mutex_unlock(&p->mutex); + return -EFAULT; + } + if (pdd->bound != PDD_BOUND_SUSPENDED) { mutex_unlock(&p->mutex); continue; @@ -501,6 +508,11 @@ void kfd_unbind_processes_from_device(struct kfd_dev *dev) mutex_lock(&p->mutex); pdd = kfd_get_process_device_data(dev, p); + if (!pdd) { + mutex_unlock(&p->mutex); + return; + } + if (pdd->bound == PDD_BOUND) pdd->bound = PDD_BOUND_SUSPENDED; mutex_unlock(&p->mutex); -- 2.7.4

2 3

[pull] amdgpu, radeon, ttm drm-next-4.16
by Alex Deucher 10 Jan '18

10 Jan '18

Hi Dave, A few fixes for 4.16: - Cleanup the the remains of ttm io_mem_pfn - A couple dpm quirks for SI - Add Chunming as another amdgpu maintainer - A few more huge page fixes - A few other misc fixes The following changes since commit fb8baefc7b2d7b93ad96abacbe63fa99e3d213d6: drm/amdgpu: use %pap format string for phys_addr_t (2018-01-09 10:59:28 +1000) are available in the git repository at: git://people.freedesktop.org/~agd5f/linux drm-next-4.16 for you to fetch changes up to ad8cec7df5d4bf3b1109fabbb1d61663857045ae: drm/amd/pp: Implement get_max_high_clocks for CI/VI (2018-01-10 15:44:55 -0500) ---------------------------------------------------------------- Alex Deucher (4): drm/radeon: Add dpm quirk for Jet PRO (v2) drm/amdgpu: Add dpm quirk for Jet PRO (v2) drm/amdgpu: add atpx quirk handling (v2) drm/amdgpu/gfx9: only init the apertures used by KGD (v2) Christian König (8): drm/ttm: specify DMA_ATTR_NO_WARN for huge page pools drm/amdgpu: loosen the criteria for huge pages a bit drm/amdgpu: minor optimize VM moved handling v2 drm/amdgpu: update VM PDs after the PTs drm/amdgpu: simplify huge page handling drm/amdgpu: optimize moved handling only when vm_debug is inactive drm/amdgpu: fix 64bit BAR detection MAINTAINERS: add David (Chunming) Zhou as additional amdgpu maintainer Evan Quan (1): drm/amd/powerplay: set pp_num_states as 0 on error situation Rex Zhu (1): drm/amd/pp: Implement get_max_high_clocks for CI/VI Tan Xiaojun (2): staging: remove the default io_mem_pfn set drm/ttm: remove ttm_bo_default_io_mem_pfn Yintian Tao (1): drm/amd/powerplay: fix memory leakage when reload (v2) MAINTAINERS | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c | 57 ++++++++++++++-- drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 8 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 8 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 78 +++++++++------------- drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 2 +- drivers/gpu/drm/amd/amdgpu/si_dpm.c | 5 ++ drivers/gpu/drm/amd/powerplay/amd_powerplay.c | 2 + drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c | 20 ++++++ drivers/gpu/drm/amd/powerplay/smumgr/smu7_smumgr.c | 6 ++ drivers/gpu/drm/radeon/si_dpm.c | 5 ++ drivers/gpu/drm/ttm/ttm_bo_vm.c | 11 +-- drivers/gpu/drm/ttm/ttm_page_alloc_dma.c | 8 ++- drivers/staging/vboxvideo/vbox_ttm.c | 1 - include/drm/ttm/ttm_bo_api.h | 11 --- 16 files changed, 139 insertions(+), 86 deletions(-)

1 0

[pull] drm/msm: msm-next for 4.16
by Rob Clark 10 Jan '18

10 Jan '18

Hi Dave, Updates for 4.16.. fairly small this time around, main thing is devfreq support for the gpu. The following changes since commit 9428088c90b6f7d5edd2a1b0d742c75339b36f6e: drm/qxl: reapply cursor after resetting primary (2017-12-08 13:37:02 +1000) are available in the Git repository at: git://people.freedesktop.org/~robclark/linux tags/drm-msm-next-2018-01-10 for you to fetch changes up to f91c14ab448af4d9d57350301dd9d6b6a7b6128a: drm/msm: Add devfreq support for the GPU (2018-01-10 14:30:03 -0500) ---------------------------------------------------------------- Archit Taneja (1): drm/msm: Fix NULL deref in adreno_load_gpu Arnd Bergmann (2): drm/msm: fix msm_rd_dump_submit prototype drm/msm: avoid false-positive -Wmaybe-uninitialized warning Colin Ian King (1): drm/msm: fix spelling mistake: "ringubffer" -> "ringbuffer" Jordan Crouse (9): drm/msm: gpu: Only sync fences on rings that exist drm/msm/adreno: Call dev_pm_opp_put() drm/msm/adreno: Remove a useless call to dev_pm_opp_get_freq() drm/msm/gpu: Remove unused bus scaling code drm/msm/adreno: Cleanup chipid parsing drm/msm/adreno: Move clock parsing to adreno_gpu_init() drm/msm/adreno: Read the speed bins for a5xx targets drm/msm/adreno: a5xx: Explicitly program the CP0 performance counter drm/msm: Add devfreq support for the GPU Lukas Wunner (1): drm/msm/mdp4: Deduplicate bus_find_device() by name matching Nicolas Dechesne (2): drm/msm: update adreno firmware path in MODULE_FIRMWARE drm/msm: add missing MODULE_FIRMWARE declarations Prakash Kamliya (1): drm/msm: fix leak in failed get_pages Rob Clark (1): drm/msm: free kstrdup'd cmdline drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 38 ++++++++ drivers/gpu/drm/msm/adreno/a5xx_power.c | 8 +- drivers/gpu/drm/msm/adreno/adreno_device.c | 140 ++++++++--------------------- drivers/gpu/drm/msm/adreno/adreno_gpu.c | 85 +++++++++++++++--- drivers/gpu/drm/msm/adreno/adreno_gpu.h | 4 - drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h | 7 +- drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c | 9 +- drivers/gpu/drm/msm/msm_drv.h | 3 +- drivers/gpu/drm/msm/msm_gem.c | 14 ++- drivers/gpu/drm/msm/msm_gpu.c | 112 +++++++++++++++++------ drivers/gpu/drm/msm/msm_gpu.h | 14 +-- 11 files changed, 264 insertions(+), 170 deletions(-)

1 0

[PATCH] drm/atomic: Fix memleak on ERESTARTSYS during non-blocking commits, v2.
by Maarten Lankhorst 10 Jan '18

10 Jan '18

From: "Leo (Sunpeng) Li" <sunpeng.li(a)amd.com> During a non-blocking commit, it is possible to return before the commit_tail work is queued (-ERESTARTSYS, for example). Since a reference on the crtc commit object is obtained for the pending vblank event when preparing the commit, the above situation will leave us with an extra reference. Therefore, if the commit_tail worker has not consumed the event at the end of a commit, release it's reference. Changes since v1: - Also check for state->event->base.completion being set, to handle the case where stall_checks() fails in setup_crtc_commit(). Fixes: 24835e442f28 ("drm: reference count event->completion") Cc: <stable(a)vger.kernel.org> # v4.11+ Signed-off-by: Leo (Sunpeng) Li <sunpeng.li(a)amd.com> Acked-by: Harry Wentland <harry.wentland(a)amd.com> #v1 Signed-off-by: Maarten Lankhorst <maarten.lankhorst(a)linux.intel.com> --- drivers/gpu/drm/drm_atomic_helper.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index b16f1d69a0bb..1d43f3e85a7d 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -3327,6 +3327,15 @@ EXPORT_SYMBOL(drm_atomic_helper_crtc_duplicate_state); void __drm_atomic_helper_crtc_destroy_state(struct drm_crtc_state *state) { if (state->commit) { + /* + * In the event that a non-blocking commit returns + * -ERESTARTSYS before the commit_tail work is queued, we will + * have an extra reference to the commit object. Release it, if + * the event has not been consumed by the worker. + */ + if (state->event && state->event->base.completion) + drm_crtc_commit_put(state->commit); + kfree(state->commit->event); state->commit->event = NULL; drm_crtc_commit_put(state->commit); -- 2.15.1

2 1

[radeon-alex:amd-staging-drm-next 374/952] sound/soc/amd/raven/acp3x.h:28:9: error: implicit declaration of function 'readl'; did you mean 'vread'?
by kbuild test robot 10 Jan '18

10 Jan '18

Hi Maruthi, FYI, the error/warning still remains. tree: git://people.freedesktop.org/~agd5f/linux.git amd-staging-drm-next head: 95b5f448bf04b220423808df1bb36ca1bd07b270 commit: c1888183e1764d55d51ae051bd8651e634febe4d [374/952] ASoC: AMD: enable ACP3x drivers build config: arm-allmodconfig (attached as .config) compiler: arm-linux-gnueabi-gcc (Debian 7.2.0-11) 7.2.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross git checkout c1888183e1764d55d51ae051bd8651e634febe4d # save the attached .config to linux build tree make.cross ARCH=arm All errors (new ones prefixed by >>): In file included from sound/soc/amd/raven/acp3x-pcm-dma.c:26:0: sound/soc/amd/raven/acp3x.h: In function 'rv_readl': >> sound/soc/amd/raven/acp3x.h:28:9: error: implicit declaration of function 'readl'; did you mean 'vread'? [-Werror=implicit-function-declaration] return readl(base_addr - ACP3x_PHY_BASE_ADDRESS); ^~~~~ vread sound/soc/amd/raven/acp3x.h: In function 'rv_writel': >> sound/soc/amd/raven/acp3x.h:33:2: error: implicit declaration of function 'writel'; did you mean 'vwrite'? [-Werror=implicit-function-declaration] writel(val, base_addr - ACP3x_PHY_BASE_ADDRESS); ^~~~~~ vwrite sound/soc/amd/raven/acp3x-pcm-dma.c: In function 'acp3x_audio_probe': >> sound/soc/amd/raven/acp3x-pcm-dma.c:638:22: error: implicit declaration of function 'devm_ioremap'; did you mean 'devm_kmemdup'? [-Werror=implicit-function-declaration] adata->acp3x_base = devm_ioremap(&pdev->dev, res->start, ^~~~~~~~~~~~ devm_kmemdup sound/soc/amd/raven/acp3x-pcm-dma.c:638:20: warning: assignment makes pointer from integer without a cast [-Wint-conversion] adata->acp3x_base = devm_ioremap(&pdev->dev, res->start, ^ cc1: some warnings being treated as errors vim +28 sound/soc/amd/raven/acp3x.h 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 25 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 26 static inline u32 rv_readl(void __iomem *base_addr) 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 27 { 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 @28 return readl(base_addr - ACP3x_PHY_BASE_ADDRESS); 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 29 } 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 30 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 31 static inline void rv_writel(u32 val, void __iomem *base_addr) 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 32 { 0360f007 Maruthi Srinivas Bayyavarapu 2017-03-27 @33 writel(val, base_addr - ACP3x_PHY_BASE_ADDRESS); :::::: The code at line 28 was first introduced by commit :::::: 0360f0076fa5ee321ca46b062374c17decdcfbe2 ASoC: AMD: add ACP3.0 PCI driver :::::: TO: Maruthi Srinivas Bayyavarapu <Maruthi.Bayyavarapu(a)amd.com> :::::: CC: Alex Deucher <alexander.deucher(a)amd.com> --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

1 0

[radeon-alex:amd-17.50 2016/2100] drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/ci_smc.c:579:62: warning: 'tmp' may be used uninitialized in this function
by kbuild test robot 10 Jan '18

10 Jan '18

Hi Rex, FYI, the error/warning still remains. tree: git://people.freedesktop.org/~agd5f/linux.git amd-17.50 head: 4b295aab3256d700354748dcd64017d76065d684 commit: 9c2014d08693830897cad6bbd359d905825d8a18 [2016/2100] drm/amd/powerplay: add CI asics support to smumgr config: i386-allyesconfig (attached as .config) compiler: gcc-7 (Debian 7.2.0-12) 7.2.1 20171025 reproduce: git checkout 9c2014d08693830897cad6bbd359d905825d8a18 # save the attached .config to linux build tree make ARCH=i386 Note: it may well be a FALSE warning. FWIW you are at least aware of it now. http://gcc.gnu.org/wiki/Better_Uninitialized_Warnings All warnings (new ones prefixed by >>): drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/ci_smc.c: In function 'ci_init_smc_table': >> drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/ci_smc.c:579:62: warning: 'tmp' may be used uninitialized in this function [-Wmaybe-uninitialized] smu_data->power_tune_table.FuzzyFan_PwmSetDelta = CONVERT_FROM_HOST_TO_SMC_US(tmp); ~^~~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/ci_smc.c:571:11: note: 'tmp' was declared here uint16_t tmp; ^~~ vim +/tmp +579 drivers/gpu/drm/amd/amdgpu/../powerplay/smumgr/ci_smc.c 568 569 static int ci_populate_fuzzy_fan(struct pp_hwmgr *hwmgr, uint32_t fuse_table_offset) 570 { 571 uint16_t tmp; 572 struct ci_smumgr *smu_data = (struct ci_smumgr *)(hwmgr->smumgr->backend); 573 574 if ((hwmgr->thermal_controller.advanceFanControlParameters.usFanOutputSensitivity & (1 << 15)) 575 || 0 == hwmgr->thermal_controller.advanceFanControlParameters.usFanOutputSensitivity) 576 tmp = hwmgr->thermal_controller.advanceFanControlParameters.usFanOutputSensitivity 577 = hwmgr->thermal_controller.advanceFanControlParameters.usDefaultFanOutputSensitivity; 578 > 579 smu_data->power_tune_table.FuzzyFan_PwmSetDelta = CONVERT_FROM_HOST_TO_SMC_US(tmp); 580 581 return 0; 582 } 583 --- 0-DAY kernel test infrastructure Open Source Technology Center https://lists.01.org/pipermail/kbuild-all Intel Corporation

1 0

[v2 PATCH] drm/msm: Add devfreq support for the GPU
by Jordan Crouse 10 Jan '18

10 Jan '18

Add support for devfreq to dynamically control the GPU frequency. By default try to use the 'simple_ondemand' governor which can adjust the frequency based on GPU load. v2: Fix __aeabi_uldivmod issue from the 0 day bot and use devfreq_recommended_opp() as suggested by Rob. Signed-off-by: Jordan Crouse <jcrouse(a)codeaurora.org> --- drivers/gpu/drm/msm/adreno/a5xx_gpu.c | 12 +++++ drivers/gpu/drm/msm/adreno/adreno_gpu.c | 1 - drivers/gpu/drm/msm/msm_gpu.c | 91 +++++++++++++++++++++++++++++++++ drivers/gpu/drm/msm/msm_gpu.h | 7 +++ 4 files changed, 110 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 56c2c44..7e09d44 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -600,6 +600,9 @@ static int a5xx_hw_init(struct msm_gpu *gpu) /* Select CP0 to always count cycles */ gpu_write(gpu, REG_A5XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT); + /* Select RBBM0 to countable 6 to get the busy status for devfreq */ + gpu_write(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_SEL_0, 6); + /* Increase VFD cache access so LRZ and other data gets evicted less */ gpu_write(gpu, REG_A5XX_UCHE_CACHE_WAYS, 0x02); @@ -1170,6 +1173,14 @@ static struct msm_ringbuffer *a5xx_active_ring(struct msm_gpu *gpu) return a5xx_gpu->cur_ring; } +static int a5xx_gpu_busy(struct msm_gpu *gpu, uint64_t *value) +{ + *value = gpu_read64(gpu, REG_A5XX_RBBM_PERFCTR_RBBM_0_LO, + REG_A5XX_RBBM_PERFCTR_RBBM_0_HI); + + return 0; +} + static const struct adreno_gpu_funcs funcs = { .base = { .get_param = adreno_get_param, @@ -1185,6 +1196,7 @@ static struct msm_ringbuffer *a5xx_active_ring(struct msm_gpu *gpu) #ifdef CONFIG_DEBUG_FS .show = a5xx_show, #endif + .gpu_busy = a5xx_gpu_busy, }, .get_timestamp = a5xx_get_timestamp, }; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index b4bac84..de63ff2 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -22,7 +22,6 @@ #include "msm_gem.h" #include "msm_mmu.h" - int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 5416fe8..bd376f9 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -21,12 +21,91 @@ #include "msm_fence.h" #include <linux/string_helpers.h> +#include <linux/pm_opp.h> +#include <linux/devfreq.h> /* * Power Management: */ +static int msm_devfreq_target(struct device *dev, unsigned long *freq, + u32 flags) +{ + struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev)); + struct dev_pm_opp *opp; + + opp = devfreq_recommended_opp(dev, freq, flags); + + if (IS_ERR(opp)) + return PTR_ERR(opp); + + clk_set_rate(gpu->core_clk, *freq); + dev_pm_opp_put(opp); + + return 0; +} + +static int msm_devfreq_get_dev_status(struct device *dev, + struct devfreq_dev_status *status) +{ + struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev)); + u64 cycles; + u32 freq = ((u32) status->current_frequency) / 1000000; + ktime_t time; + + status->current_frequency = (unsigned long) clk_get_rate(gpu->core_clk); + gpu->funcs->gpu_busy(gpu, &cycles); + + status->busy_time = ((u32) (cycles - gpu->devfreq.busy_cycles)) / freq; + + gpu->devfreq.busy_cycles = cycles; + + time = ktime_get(); + status->total_time = ktime_us_delta(time, gpu->devfreq.time); + gpu->devfreq.time = time; + + return 0; +} + +static int msm_devfreq_get_cur_freq(struct device *dev, unsigned long *freq) +{ + struct msm_gpu *gpu = platform_get_drvdata(to_platform_device(dev)); + + *freq = (unsigned long) clk_get_rate(gpu->core_clk); + + return 0; +} + +static struct devfreq_dev_profile msm_devfreq_profile = { + .polling_ms = 10, + .target = msm_devfreq_target, + .get_dev_status = msm_devfreq_get_dev_status, + .get_cur_freq = msm_devfreq_get_cur_freq, +}; + +static void msm_devfreq_init(struct msm_gpu *gpu) +{ + /* We need target support to do devfreq */ + if (!gpu->funcs->gpu_busy) + return; + + msm_devfreq_profile.initial_freq = gpu->fast_rate; + + /* + * Don't set the freq_table or max_state and let devfreq build the table + * from OPP + */ + + gpu->devfreq.devfreq = devm_devfreq_add_device(&gpu->pdev->dev, + &msm_devfreq_profile, "simple_ondemand", NULL); + + if (IS_ERR(gpu->devfreq.devfreq)) { + dev_err(&gpu->pdev->dev, "Couldn't initialize GPU devfreq\n"); + gpu->devfreq.devfreq = NULL; + } +} + static int enable_pwrrail(struct msm_gpu *gpu) { struct drm_device *dev = gpu->dev; @@ -140,6 +219,13 @@ int msm_gpu_pm_resume(struct msm_gpu *gpu) if (ret) return ret; + if (gpu->devfreq.devfreq) { + gpu->devfreq.busy_cycles = 0; + gpu->devfreq.time = ktime_get(); + + devfreq_resume_device(gpu->devfreq.devfreq); + } + gpu->needs_hw_init = true; return 0; @@ -151,6 +237,9 @@ int msm_gpu_pm_suspend(struct msm_gpu *gpu) DBG("%s", gpu->name); + if (gpu->devfreq.devfreq) + devfreq_suspend_device(gpu->devfreq.devfreq); + ret = disable_axi(gpu); if (ret) return ret; @@ -720,6 +809,8 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, gpu->pdev = pdev; platform_set_drvdata(pdev, gpu); + msm_devfreq_init(gpu); + gpu->aspace = msm_gpu_create_address_space(gpu, pdev, config->va_start, config->va_end); diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index 0de26b6..fccfccd 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -66,6 +66,7 @@ struct msm_gpu_funcs { /* show GPU status in debugfs: */ void (*show)(struct msm_gpu *gpu, struct seq_file *m); #endif + int (*gpu_busy)(struct msm_gpu *gpu, uint64_t *value); }; struct msm_gpu { @@ -120,6 +121,12 @@ struct msm_gpu { struct work_struct recover_work; struct drm_gem_object *memptrs_bo; + + struct { + struct devfreq *devfreq; + u64 busy_cycles; + ktime_t time; + } devfreq; }; /* It turns out that all targets use the same ringbuffer size */ -- 1.9.1

1 0

[git pull] vmwgfx-fixes-4.15
by Thomas Hellstrom 10 Jan '18

10 Jan '18

Dave, Two important fixes for vmwgfx. The off-by-one fix could cause a malicious user to potentially crash the kernel. The framebuffer map cache fix can under some circumstances enable a user to read from or write to freed pages. The following changes since commit b0bb222440a5c8273f67dd37946707e6ba6ad832: Merge branch 'linux-4.15' of git://github.com/skeggsb/linux into drm-fixes (2018-01-09 12:03:10 +1000) are available in the git repository at: git://people.freedesktop.org/~thomash/linux vmwgfx-fixes-4.15 for you to fetch changes up to 0d9cac0ca0429830c40fe1a4e50e60f6221fd7b6: drm/vmwgfx: Potential off by one in vmw_view_add() (2018-01-10 15:21:39 +0100) ---------------------------------------------------------------- Dan Carpenter (1): drm/vmwgfx: Potential off by one in vmw_view_add() Thomas Hellstrom (1): drm/vmwgfx: Don't cache framebuffer maps drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 2 ++ drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 6 ----- drivers/gpu/drm/vmwgfx/vmwgfx_kms.h | 2 +- drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c | 41 ++++++++++----------------------- 4 files changed, 15 insertions(+), 36 deletions(-)

1 0

[PATCH 2/3] drm: export gem dmabuf_ops for drivers to reuse
by Samuel Li 10 Jan '18

10 Jan '18

Change-Id: I03c22a890d2305f3243d88019d1a28bddd4ddda7 Signed-off-by: Samuel Li <Samuel.Li(a)amd.com> --- drivers/gpu/drm/drm_prime.c | 53 ++++++++++++++++++++++++++------------------- include/drm/drm_prime.h | 22 +++++++++++++++++++ 2 files changed, 53 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c index 8de93a2..68a69e9 100644 --- a/drivers/gpu/drm/drm_prime.c +++ b/drivers/gpu/drm/drm_prime.c @@ -180,9 +180,8 @@ static int drm_prime_lookup_buf_handle(struct drm_prime_file_private *prime_fpri return -ENOENT; } -static int drm_gem_map_attach(struct dma_buf *dma_buf, - struct device *target_dev, - struct dma_buf_attachment *attach) +int drm_gem_map_attach(struct dma_buf *dma_buf, struct device *target_dev, + struct dma_buf_attachment *attach) { struct drm_prime_attachment *prime_attach; struct drm_gem_object *obj = dma_buf->priv; @@ -200,9 +199,10 @@ static int drm_gem_map_attach(struct dma_buf *dma_buf, return dev->driver->gem_prime_pin(obj); } +EXPORT_SYMBOL(drm_gem_map_attach); -static void drm_gem_map_detach(struct dma_buf *dma_buf, - struct dma_buf_attachment *attach) +void drm_gem_map_detach(struct dma_buf *dma_buf, + struct dma_buf_attachment *attach) { struct drm_prime_attachment *prime_attach = attach->priv; struct drm_gem_object *obj = dma_buf->priv; @@ -227,6 +227,7 @@ static void drm_gem_map_detach(struct dma_buf *dma_buf, kfree(prime_attach); attach->priv = NULL; } +EXPORT_SYMBOL(drm_gem_map_detach); void drm_prime_remove_buf_handle_locked(struct drm_prime_file_private *prime_fpriv, struct dma_buf *dma_buf) @@ -253,8 +254,8 @@ void drm_prime_remove_buf_handle_locked(struct drm_prime_file_private *prime_fpr } } -static struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach, - enum dma_data_direction dir) +struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach, + enum dma_data_direction dir) { struct drm_prime_attachment *prime_attach = attach->priv; struct drm_gem_object *obj = attach->dmabuf->priv; @@ -289,13 +290,15 @@ static struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach, return sgt; } +EXPORT_SYMBOL(drm_gem_map_dma_buf); -static void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach, - struct sg_table *sgt, - enum dma_data_direction dir) +void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach, + struct sg_table *sgt, + enum dma_data_direction dir) { /* nothing to be done here */ } +EXPORT_SYMBOL(drm_gem_unmap_dma_buf); /** * drm_gem_dmabuf_export - dma_buf export implementation for GEM @@ -346,47 +349,52 @@ void drm_gem_dmabuf_release(struct dma_buf *dma_buf) } EXPORT_SYMBOL(drm_gem_dmabuf_release); -static void *drm_gem_dmabuf_vmap(struct dma_buf *dma_buf) +void *drm_gem_dmabuf_vmap(struct dma_buf *dma_buf) { struct drm_gem_object *obj = dma_buf->priv; struct drm_device *dev = obj->dev; return dev->driver->gem_prime_vmap(obj); } +EXPORT_SYMBOL(drm_gem_dmabuf_vmap); -static void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr) +void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr) { struct drm_gem_object *obj = dma_buf->priv; struct drm_device *dev = obj->dev; dev->driver->gem_prime_vunmap(obj, vaddr); } +EXPORT_SYMBOL(drm_gem_dmabuf_vunmap); -static void *drm_gem_dmabuf_kmap_atomic(struct dma_buf *dma_buf, - unsigned long page_num) +void *drm_gem_dmabuf_kmap_atomic(struct dma_buf *dma_buf, + unsigned long page_num) { return NULL; } +EXPORT_SYMBOL(drm_gem_dmabuf_kmap_atomic); -static void drm_gem_dmabuf_kunmap_atomic(struct dma_buf *dma_buf, - unsigned long page_num, void *addr) +void drm_gem_dmabuf_kunmap_atomic(struct dma_buf *dma_buf, + unsigned long page_num, void *addr) { } -static void *drm_gem_dmabuf_kmap(struct dma_buf *dma_buf, - unsigned long page_num) +EXPORT_SYMBOL(drm_gem_dmabuf_kunmap_atomic); + +void *drm_gem_dmabuf_kmap(struct dma_buf *dma_buf, unsigned long page_num) { return NULL; } +EXPORT_SYMBOL(drm_gem_dmabuf_kmap); -static void drm_gem_dmabuf_kunmap(struct dma_buf *dma_buf, - unsigned long page_num, void *addr) +void drm_gem_dmabuf_kunmap(struct dma_buf *dma_buf, unsigned long page_num, + void *addr) { } +EXPORT_SYMBOL(drm_gem_dmabuf_kunmap); -static int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, - struct vm_area_struct *vma) +int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma) { struct drm_gem_object *obj = dma_buf->priv; struct drm_device *dev = obj->dev; @@ -396,6 +404,7 @@ static int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, return dev->driver->gem_prime_mmap(obj, vma); } +EXPORT_SYMBOL(drm_gem_dmabuf_mmap); static const struct dma_buf_ops drm_gem_prime_dmabuf_ops = { .attach = drm_gem_map_attach, diff --git a/include/drm/drm_prime.h b/include/drm/drm_prime.h index 9cd9e36..4d5f5d6 100644 --- a/include/drm/drm_prime.h +++ b/include/drm/drm_prime.h @@ -54,6 +54,9 @@ struct device; struct dma_buf_export_info; struct dma_buf; +struct dma_buf_attachment; + +enum dma_data_direction; struct drm_device; struct drm_gem_object; @@ -79,6 +82,25 @@ int drm_gem_prime_fd_to_handle(struct drm_device *dev, struct dma_buf *drm_gem_dmabuf_export(struct drm_device *dev, struct dma_buf_export_info *exp_info); void drm_gem_dmabuf_release(struct dma_buf *dma_buf); +int drm_gem_map_attach(struct dma_buf *dma_buf, struct device *target_dev, + struct dma_buf_attachment *attach); +void drm_gem_map_detach(struct dma_buf *dma_buf, + struct dma_buf_attachment *attach); +struct sg_table *drm_gem_map_dma_buf(struct dma_buf_attachment *attach, + enum dma_data_direction dir); +void drm_gem_unmap_dma_buf(struct dma_buf_attachment *attach, + struct sg_table *sgt, + enum dma_data_direction dir); +void *drm_gem_dmabuf_vmap(struct dma_buf *dma_buf); +void drm_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr); +void *drm_gem_dmabuf_kmap_atomic(struct dma_buf *dma_buf, + unsigned long page_num); +void drm_gem_dmabuf_kunmap_atomic(struct dma_buf *dma_buf, + unsigned long page_num, void *addr); +void *drm_gem_dmabuf_kmap(struct dma_buf *dma_buf, unsigned long page_num); +void drm_gem_dmabuf_kunmap(struct dma_buf *dma_buf, unsigned long page_num, + void *addr); +int drm_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct *vma); int drm_prime_sg_to_page_addr_arrays(struct sg_table *sgt, struct page **pages, dma_addr_t *addrs, int max_pages); -- 2.7.4

6 7

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel January 2018