dri-devel March 2014

dri-devel@lists.freedesktop.org

144 participants
926 discussions

[PULL] drm-intel-next
by Daniel Vetter 17 Mar '14

17 Mar '14

Hi Dave, drm-intel-next-2014-03-07: - fine-grained display power domains for byt (Imre) - runtime pm prep patches for !hsw from Paulo - WiZ hashing flag updates from Ville - ppgtt setup cleanup and enabling of full 4G range on bdw (Ben) - fixes from Jesse for the inherited intial config code - gpu reset code improvements from Mika - per-pipe num_planes refactoring from Damien - stability fixes around bdw forcewake handling and other bdw w/a from Mika and Ken - and as usual a pile of smaller fixes all over The big thing here is the disabling of the full ppgtt code. I've figured I'll need to pull the plug a bit earlier to have enough time to test thing throughroughly before the 3.15 merge window opens. Cheers, Daniel The following changes since commit 4c0e552882114d1edb588242d45035246ab078a0: drm/i915: fix NULL deref in the load detect code (2014-02-14 17:23:12 +0100) are available in the git repository at: git://anongit.freedesktop.org/drm-intel tags/drm-intel-next-2014-03-07 for you to fetch changes up to 2fae6a860ca9adb0c881f6dcd633df775c2520e9: drm/i915: Go OCD on the Makefile (2014-03-07 22:37:00 +0100) ---------------------------------------------------------------- - fine-grained display power domains for byt (Imre) - runtime pm prep patches for !hsw from Paulo - WiZ hashing flag updates from Ville - ppgtt setup cleanup and enabling of full 4G range on bdw (Ben) - fixes from Jesse for the inherited intial config code - gpu reset code improvements from Mika - per-pipe num_planes refactoring from Damien - stability fixes around bdw forcewake handling and other bdw w/a from Mika and Ken - and as usual a pile of smaller fixes all over ---------------------------------------------------------------- Ben Widawsky (12): drm/i915: Move ppgtt_release out of the header drm/i915/bdw: Free PPGTT struct drm/i915/bdw: Reorganize PPGTT init drm/i915/bdw: Split ppgtt initialization up drm/i915: Make clear/insert vfuncs args absolute drm/i915/bdw: Reorganize PT allocations Revert "drm/i915/bdw: Limit GTT to 2GB" drm/i915: Update i915_gem_gtt.c copyright drm/i915: Split GEN6 PPGTT cleanup drm/i915: Split GEN6 PPGTT initialization up drm/i915/bdw: Kill ppgtt->num_pt_pages drm/i915/bdw: Add FBC support Brad Volkin (2): drm/i915: Refactor shmem pread setup drm/i915: Implement command buffer parsing logic Chris Wilson (9): drm/i915: Revert workaround for disabling L3 cache aging on IVB Revert "drm/i915: enable HiZ Raw Stall Optimization on IVB" drm/i915: Reject changes of fb base when we have a flip pending drm/i915: Accurately track when we mark the hardware as idle/busy drm/i915: Convert the forcewake worker into a timer func drm/i915: Perform pageflip using mmio if the GPU is terminally wedged drm/i915: Reset vma->mm_list after unbinding drm/i915: Rely on accurate request tracking for finding hung batches drm/i915: Record pid/comm of hanging task Damien Lespiau (8): drm/i915: Use a pipe variable to cycle through the pipes drm/i915: Don't declare unnecessary shadowing variable drm/i915: Replace a few for_each_pipe(i) by for_each_pipe(pipe) drm/i915: Add a for_each_sprite() macro drm/i915: Make num_sprites a per-pipe value drm/i915: Make i915_gem_retire_requests_ring() static drm/i915: Remove unused to_gem_object() macro drm/i915: Fix i915_switch_context() argument name in kerneldoc Daniel Vetter (7): drm/i915: tune down user-triggerable dmesg noise in the cursor/overlay code drm/i915: sprinkle static drm/i915: s/any_enabled/!fallback/ in fbdev_initial_config drm/i915: ignore bios output config if not all outputs are on drm/i915: reverse dp link param selection, prefer fast over wide again drm/i915: Disable full ppgtt by default drm/i915: Go OCD on the Makefile Imre Deak (21): drm/i915: use drm_i915_private everywhere in the power domain api drm/i915: switch order of power domain init wrt. irq install drm/i915: use power domain api to check vga power state drm/i915: move hsw power domain comment to its right place drm/i915: fold in __intel_power_well_get/put functions drm/i915: move modeset_update_power_wells earlier drm/i915: move power domain macros to intel_pm.c drm/i915: add init power domain to always-on power wells drm/i915: split power well 'set' handler to separate enable/disable/sync_hw drm/i915: add noop power well handlers instead of NULL checking them drm/i915: add port power domains drm/i915: get port power domain in connector detect handlers drm/i915: check port power domain when reading the encoder hw state drm/i915: check pipe power domain when reading its hw state drm/i915: vlv: keep first level vblank IRQs masked drm/i915: sanitize PUNIT register macro definitions drm/i915: factor out reset_vblank_counter drm/i915: sanity check power well sw state against hw state drm/i915: vlv: factor out valleyview_display_irq_install drm/i915: factor out intel_set_cpu_fifo_underrun_reporting_nolock drm/i915: power domains: add vlv power wells Jani Nikula (1): drm/i915: don't flood the logs about bdw semaphores Jesse Barnes (3): drm/i915: honor forced connector modes v2 drm/i915: re-add locking around hw state readout drm/i915: print connector mode list in display_info Kenneth Graunke (2): drm/i915: Add a partial instruction shootdown workaround on Broadwell. drm/i915: Add thread stall DOP clock gating workaround on Broadwell. Mika Kuoppala (8): drm/i915: Fix forcewake counts for gen8 drm/i915: Add error code into error state drm/i915: Add reason for capture in error state drm/i915: Add reset count to error state drm/i915: Add suspend count to error state drm/i915: Do forcewake reset on gen8 drm/i915: Don't access fifodbg registers on gen8 drm/i915: No need to put forcewake after a reset Patrik Jakobsson (1): drm/i915: Don't just say it, actually force edp vdd Paulo Zanoni (10): drm/i915: rename modeset_update_power_wells drm/i915: get/put runtime PM without holding rps.hw_lock drm/i915: put runtime PM only at the end of intel_mark_idle drm/i915: put runtime PM only when we actually release force_wake drm/i915: get runtime PM while trying to detect CRT drm/i915: get/put runtime PM in more places at i915_debugfs.c drm/i915: kill dev_priv->pc8.gpu_idle drm/i915: call assert_device_not_suspended at gen6_force_wake_work drm/i915: assert force wake is disabled when we runtime suspend drm/i915: assert we're not runtime suspended when accessing registers Shobhit Kumar (1): drm/i915: Update VBT data structures to have MIPI block enhancements Sinclair Yeh (1): drm/i915: Revert workaround for disabling L3 cache aging on BYT Thierry Reding (1): drm/i915: Remove dead code Ville Syrjälä (20): drm/i915: Fix SNB GT_MODE register setup drm/i915: Assume we implement WaStripsFansDisableFastClipPerformanceFix:snb drm/i915: There's no need to mask all 3D_CHICKEN bits on SNB drm/i915: Change IVB WIZ hashing mode to 16x4 drm/i915: Change HSW WIZ hashing mode to 16x4 drm/i915: Change BDW WIZ hashing mode to 16x4 drm/i915: Add a comment about WIZ hashing vs. thread counts drm/i915: Don't ban default context when stop_rings!=0 drm/i915: Fix VLV forcewake after reset drm/i915: Drop the forcewake count inc/dec around register read on VLV drm/i915: Streamline VLV forcewake handling drm/i915: Fix DDI port_clock for VGA output drm/i915: Use DIV_ROUND_UP() when calculating number of required FDI lanes drm/i915: Disable semaphore wait event idle message on BDW drm/i915: Implement WaDisableSDEUnitClockGating:bdw drm/i915: We implement WaDisableAsyncFlipPerfMode:bdw drm/i915: Don't clobber CHICKEN_PIPESL_1 on BDW drm/i915: Use RMW to update chicken bits in gen7_enable_fbc() drm/i915: Unify CHICKEN_PIPESL_1 register definitions drm/i915: Avoid div by zero when pixel clock is large drivers/gpu/drm/i915/Makefile | 83 ++-- drivers/gpu/drm/i915/i915_cmd_parser.c | 485 ++++++++++++++++++++++ drivers/gpu/drm/i915/i915_debugfs.c | 79 +++- drivers/gpu/drm/i915/i915_dma.c | 24 +- drivers/gpu/drm/i915/i915_drv.c | 19 +- drivers/gpu/drm/i915/i915_drv.h | 282 ++++++++----- drivers/gpu/drm/i915/i915_gem.c | 92 +++-- drivers/gpu/drm/i915/i915_gem_context.c | 52 ++- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 18 + drivers/gpu/drm/i915/i915_gem_gtt.c | 621 ++++++++++++++++++++--------- drivers/gpu/drm/i915/i915_gpu_error.c | 272 +++++++------ drivers/gpu/drm/i915/i915_irq.c | 208 +++++++--- drivers/gpu/drm/i915/i915_params.c | 5 + drivers/gpu/drm/i915/i915_reg.h | 69 +++- drivers/gpu/drm/i915/intel_bios.c | 4 +- drivers/gpu/drm/i915/intel_bios.h | 174 ++++++-- drivers/gpu/drm/i915/intel_crt.c | 57 ++- drivers/gpu/drm/i915/intel_ddi.c | 5 + drivers/gpu/drm/i915/intel_display.c | 363 ++++++++++------- drivers/gpu/drm/i915/intel_dp.c | 43 +- drivers/gpu/drm/i915/intel_drv.h | 21 +- drivers/gpu/drm/i915/intel_dsi.c | 18 +- drivers/gpu/drm/i915/intel_fbdev.c | 49 ++- drivers/gpu/drm/i915/intel_hdmi.c | 34 +- drivers/gpu/drm/i915/intel_overlay.c | 2 +- drivers/gpu/drm/i915/intel_pm.c | 575 +++++++++++++++++++++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +- drivers/gpu/drm/i915/intel_ringbuffer.h | 32 ++ drivers/gpu/drm/i915/intel_uncore.c | 227 ++++++----- 29 files changed, 2897 insertions(+), 1020 deletions(-) create mode 100644 drivers/gpu/drm/i915/i915_cmd_parser.c -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch

1 0

[PULL] drm-intel-fixes
by Jani Nikula 17 Mar '14

17 Mar '14

Hi Dave, Two 3.14 specific fixes, two cc: stable. I'm afraid I've accumulated a slight backlog of fixes. This pull is to get the tested stuff out of the way first. I'll be pushing more patches soon, and will send another pull once our QA has had some time to run them. BR, Jani. The following changes since commit 6375b768a9850b6154478993e5fb566fa4614a9c: drm/i915: Reject >165MHz modes w/ DVI monitors (2014-03-03 19:08:08 +0200) are available in the git repository at: git://anongit.freedesktop.org/drm-intel tags/drm-intel-fixes-2014-03-17 for you to fetch changes up to 5c673b60a9b3b23486f4eda75c72e91d31d26a2b: drm/i915: Don't enable display error interrupts from the start (2014-03-12 17:20:34 +0200) ---------------------------------------------------------------- Ben Widawsky (1): drm/i915: Fix PSR programming Daniel Vetter (1): drm/i915: Don't enable display error interrupts from the start Ville Syrjälä (2): drm/i915: Add a workaround for HSW scanline counter weirdness drm/i915: Fix scanline counter fixup on BDW drivers/gpu/drm/i915/i915_irq.c | 71 ++++++++++++++++++++++----------------- drivers/gpu/drm/i915/intel_dp.c | 2 +- 2 files changed, 42 insertions(+), 31 deletions(-) -- Jani Nikula, Intel Open Source Technology Center

1 0

[PATCH RFC 0/5] Move tda998x to a couple encoder/connector
by Jean-Francois Moine 17 Mar '14

17 Mar '14

The 'slave encoder' structure of the tda998x driver asks for glue between the DRM driver and the encoder/connector structures. Changing the tda998x driver to a simple encoder/connector simplifies the code of the tilcdc driver. This change is permitted by Russell's infrastructure for componentised subsystems. The proposed patch set does not include changes to the Armada DRM driver which has no DT support yet and which will not work anymore if this patch set is applied. Jean-Francois Moine (5): drm/i2c: tda998x: Move tda998x to a couple encoder/connector drm/tilcdc: Change the interface with the tda998x driver drm/i2c: tda998x: Add required properties drm/tilcd: dts: Add the video sink ARM: AM33XX: dts: Change the tda998x description .../devicetree/bindings/drm/i2c/tda998x.txt | 12 + .../devicetree/bindings/drm/tilcdc/slave.txt | 18 - .../devicetree/bindings/drm/tilcdc/tilcdc.txt | 6 + arch/arm/boot/dts/am335x-base0033.dts | 28 +- arch/arm/boot/dts/am335x-boneblack.dts | 21 +- drivers/gpu/drm/i2c/tda998x_drv.c | 335 +++++++++-------- drivers/gpu/drm/tilcdc/Makefile | 1 - drivers/gpu/drm/tilcdc/tilcdc_drv.c | 81 +++- drivers/gpu/drm/tilcdc/tilcdc_slave.c | 406 --------------------- drivers/gpu/drm/tilcdc/tilcdc_slave.h | 26 -- 10 files changed, 297 insertions(+), 637 deletions(-) delete mode 100644 Documentation/devicetree/bindings/drm/tilcdc/slave.txt delete mode 100644 drivers/gpu/drm/tilcdc/tilcdc_slave.c delete mode 100644 drivers/gpu/drm/tilcdc/tilcdc_slave.h -- 1.9.0

1 5

[PATCH] drm/exynos: Fix (more) freeing issues in exynos_drm_drv.c
by Daniel Kurtz 17 Mar '14

17 Mar '14

The following commit [0] fixed a use-after-free, but left the subdrv open in the error path. [0] commit 6ca605f7c70895a35737435f17ae9cc5e36f1466 drm/exynos: Fix freeing issues in exynos_drm_drv.c Signed-off-by: Daniel Kurtz <djkurtz(a)chromium.org> --- Hi, I noticed this when reviewing some recent patches. I am only able to compile test this patch. drivers/gpu/drm/exynos/exynos_drm_drv.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 215131a..c204b4e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -172,20 +172,24 @@ static int exynos_drm_open(struct drm_device *dev, struct drm_file *file) ret = exynos_drm_subdrv_open(dev, file); if (ret) - goto out; + goto err_file_priv_free; anon_filp = anon_inode_getfile("exynos_gem", &exynos_drm_gem_fops, NULL, 0); if (IS_ERR(anon_filp)) { ret = PTR_ERR(anon_filp); - goto out; + goto err_subdrv_close; } anon_filp->f_mode = FMODE_READ | FMODE_WRITE; file_priv->anon_filp = anon_filp; return ret; -out: + +err_subdrv_close: + exynos_drm_subdrv_close(dev, file); + +err_file_priv_free: kfree(file_priv); file->driver_priv = NULL; return ret; -- 1.9.0.279.gdc9e3eb

2 1

[PULL 0/3] Pending patches for 3.15
by David Herrmann 16 Mar '14

16 Mar '14

Hi Dave As requested, these 3 pulls contain all my pending patches for 3.15. Branch #1 should be fairly trivial, all reviewed and minor fixes. Branch #2 contains the drm-minor rework, which have also been on the list and reviewed/tested by Daniel and Thierry. Branch #3 contains the drm-anon stuff again. Maybe you can try to get Al's final acked-by? I didn't include the simpledrm stuff for now. It depends on all three of these branches, so lets first deal with them and then have another look. Given that the simpledrm branch does not change any other DRM drivers, I don't think there's much gain in cooking it for 1 more release in -next. If nothing breaks this week, I will send the pull-request on Thursday/Friday, once I got the acked-by from Ingo. Thanks David

1 3

[PATCH 1/4] drm/i915: preserve SSC if previously set v2
by Jesse Barnes 15 Mar '14

15 Mar '14

Some machines may have a broken VBT or no VBT at all, but we still want to use SSC there. So check for it and keep it enabled if we see it already on. Based on an earlier fix from Kristian. v2: honor modparam if set too (Daniel) read out at init time and store for panel_use_ssc() use (Jesse) Reported-by: Kristian Høgsberg <hoegsberg(a)gmail.com> Signed-off-by: Jesse Barnes <jbarnes(a)virtuousgeek.org> --- drivers/gpu/drm/i915/i915_drv.h | 2 ++ drivers/gpu/drm/i915/intel_display.c | 11 ++++++++++- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 70fbe90..c64f770 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1507,6 +1507,8 @@ typedef struct drm_i915_private { struct intel_opregion opregion; struct intel_vbt_data vbt; + bool bios_ssc; /* BIOS had SSC enabled at boot? */ + /* overlay */ struct intel_overlay *overlay; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 2bccc68..4b3e1c0 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4986,7 +4986,7 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv) { if (i915.panel_use_ssc >= 0) return i915.panel_use_ssc != 0; - return dev_priv->vbt.lvds_use_ssc + return (dev_priv->vbt.lvds_use_ssc || dev_priv->bios_ssc) && !(dev_priv->quirks & QUIRK_LVDS_SSC_DISABLE); } @@ -11732,9 +11732,18 @@ void intel_modeset_setup_hw_state(struct drm_device *dev, void intel_modeset_gem_init(struct drm_device *dev) { + struct drm_i915_private *dev_priv = dev->dev_private; struct drm_crtc *c; struct intel_framebuffer *fb; + /* + * There may be no VBT; and if the BIOS enabled SSC we can + * just keep using it to avoid unnecessary flicker. + */ + if ((HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) && + (I915_READ(PCH_DREF_CONTROL) & DREF_SSC1_ENABLE)) + dev_priv->bios_ssc = true; + intel_modeset_init_hw(dev); intel_setup_overlay(dev); -- 1.7.9.5

2 4

[Bug 25297] R600: glEdgeFlag doesn't work
by bugzilla-daemon＠freedesktop.org 15 Mar '14

15 Mar '14

https://bugs.freedesktop.org/show_bug.cgi?id=25297 Benjamin Bellec <b.bellec(a)gmail.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #7 from Benjamin Bellec <b.bellec(a)gmail.com> --- I didn't test with the softrender, but this bug is now fixed since: commit: 1337da51152db7fa1a71ac86b61b51a42d29d595 summary: r600g: implement edge flags author: Marek Olšák date: 2014-03-04 11:26:16 (GMT) r600g now render the square like your screenshot. Tested on AMD CYPRESS. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[PULL] drm documentation updates for 3.15
by Daniel Vetter 15 Mar '14

15 Mar '14

Hi Dave, Here's my drm documentation update and driver api polish pull request. Alex reviewed the entire pile, I've applied a little bit of spelling polish in a few places since then and otherwise the Usual Suspects (David, Rob, ...) don't seem up to have another look at it (I've poked them on irc). So I think it's as good as it gets ;-) Note that I've dropped the final imx breaker patch since that's blocked on imx getting sane. Once that's landed I'll ping you to pick up that straggler. Cheers, Daniel The following changes since commit 786a7828bc74b9b1466e83abb200b75f80f94121: Merge branch 'drm-next-3.15' of git://people.freedesktop.org/~deathsimple/linux into drm-next (2014-03-05 14:52:19 +1000) are available in the git repository at: ssh://people.freedesktop.org/~danvet/drm.git drm-docs for you to fetch changes up to c8e32cc1219fc15135b696b726421571f68bd97e: drm: kerneldoc polish for drm_crtc.c (2014-03-13 12:48:49 +0100) ---------------------------------------------------------------- Daniel Vetter (33): drm/doc: Clarify the dumb object interfaces drm/doc: Fix up kerneldoc in drm_edid.c drm/doc: Clean up and integrate kerneldoc for drm_gem.c drm/doc: Remove <term> from rendernode docs drm/doc: Reorganize driver documentation drm/doc: Move the vma offset manager to the right spot drm/doc: Remove the "command submissin and fencing" section drm/doc: No more drm perf counters drm/doc: Document drm_helper_resume_force_mode drm/doc: Hide legacy horrors better drm/docs: Include hdmi infoframe helper reference drm/doc: Clarify PRIME documentation drm/doc: Add PRIME function references drm/doc: Update copyright drm/mm: Remove MM_UNUSED_TARGET drm/doc: Overview documentation for drm_mm.c drm/doc: Add function reference documentation for drm_mm.c drm/kms: rip out drm_mode_connector_detach_encoder drm/doc: Integrate drm_modes.c kerneldoc drm/doc: Repleace LOCKING kerneldoc sections in drm_modes.c drm: move drm_mode related functions into drm_modes.c drm: extract drm_modes.h for drm_crtc.h functions drm/modes: remove drm_mode_height/width drm/modes: drop return value from drm_display_mode_from_videomode drm/modes: drop maxPitch from drm_mode_validate_size drm: polish function kerneldoc for drm_modes.[hc] drm: remove drm_display_mode->private_size drm/doc: Fix misplaced </para> drm: remove return value from drm_helper_mode_fill_fb_struct drm/crtc-helper: remove LOCKING from kerneldoc drm: drop error code for drm_helper_resume_force_mode drm: kerneldoc polish for drm_crtc_helper.c drm: kerneldoc polish for drm_crtc.c Documentation/DocBook/drm.tmpl | 500 ++++++++++++++------- drivers/gpu/drm/drm_crtc.c | 697 ++++++++++++++++++++++++----- drivers/gpu/drm/drm_crtc_helper.c | 193 ++++++-- drivers/gpu/drm/drm_crtc_internal.h | 38 ++ drivers/gpu/drm/drm_edid.c | 30 +- drivers/gpu/drm/drm_fb_helper.c | 4 +- drivers/gpu/drm/drm_gem.c | 63 ++- drivers/gpu/drm/drm_mm.c | 211 ++++++++- drivers/gpu/drm/drm_modes.c | 346 +++++++------- drivers/gpu/drm/drm_prime.c | 110 ++++- drivers/staging/imx-drm/imx-ldb.c | 2 - drivers/staging/imx-drm/parallel-display.c | 2 - include/drm/drmP.h | 29 -- include/drm/drm_crtc.h | 174 +------ include/drm/drm_crtc_helper.h | 6 +- include/drm/drm_mm.h | 154 +++++-- include/drm/drm_modes.h | 237 ++++++++++ include/linux/hdmi.h | 12 + 18 files changed, 2013 insertions(+), 795 deletions(-) create mode 100644 drivers/gpu/drm/drm_crtc_internal.h create mode 100644 include/drm/drm_modes.h -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch

1 1

[PATCH] drm/cma: remove to make sg_table when gem cma is created
by Joonyoung Shim 15 Mar '14

15 Mar '14

The sg_table made when gem cma is created isn't used anywhere. The sgt of struct drm_gem_cma_object will have only sg_tabel imported. Signed-off-by: Joonyoung Shim <jy0922.shim(a)samsung.com> --- drivers/gpu/drm/drm_gem_cma_helper.c | 19 ------------------- 1 file changed, 19 deletions(-) diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c index 6b51bf9..2c07cb9 100644 --- a/drivers/gpu/drm/drm_gem_cma_helper.c +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -79,7 +79,6 @@ struct drm_gem_cma_object *drm_gem_cma_create(struct drm_device *drm, unsigned int size) { struct drm_gem_cma_object *cma_obj; - struct sg_table *sgt = NULL; int ret; size = round_up(size, PAGE_SIZE); @@ -97,23 +96,9 @@ struct drm_gem_cma_object *drm_gem_cma_create(struct drm_device *drm, goto error; } - sgt = kzalloc(sizeof(*cma_obj->sgt), GFP_KERNEL); - if (sgt == NULL) { - ret = -ENOMEM; - goto error; - } - - ret = dma_get_sgtable(drm->dev, sgt, cma_obj->vaddr, - cma_obj->paddr, size); - if (ret < 0) - goto error; - - cma_obj->sgt = sgt; - return cma_obj; error: - kfree(sgt); drm_gem_cma_free_object(&cma_obj->base); return ERR_PTR(ret); } @@ -175,10 +160,6 @@ void drm_gem_cma_free_object(struct drm_gem_object *gem_obj) if (cma_obj->vaddr) { dma_free_writecombine(gem_obj->dev->dev, cma_obj->base.size, cma_obj->vaddr, cma_obj->paddr); - if (cma_obj->sgt) { - sg_free_table(cma_obj->sgt); - kfree(cma_obj->sgt); - } } else if (gem_obj->import_attach) { drm_prime_gem_destroy(gem_obj, cma_obj->sgt); } -- 1.8.1.2

2 1

[PATCH 1/3] drm/gma500: Code cleanup - inline documentation
by Arthur Borsboom 15 Mar '14

15 Mar '14

Improve readability by adding/changing inline documentation Signed-off-by: Arthur Borsboom <arthurborsboom(a)gmail.com> --- drivers/gpu/drm/gma500/psb_drv.c | 56 +++++++++++++++++++++++++++++++++------- drivers/gpu/drm/gma500/psb_drv.h | 24 +++++++++++------ 2 files changed, 63 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/gma500/psb_drv.c b/drivers/gpu/drm/gma500/psb_drv.c index 1199180..5c6cdd0 100644 --- a/drivers/gpu/drm/gma500/psb_drv.c +++ b/drivers/gpu/drm/gma500/psb_drv.c @@ -44,7 +44,20 @@ static int psb_probe(struct pci_dev *pdev, const struct pci_device_id *ent); MODULE_PARM_DESC(trap_pagefaults, "Error and reset on MMU pagefaults"); module_param_named(trap_pagefaults, drm_psb_trap_pagefaults, int, 0600); - +/* + * The table below contains a mapping of the PCI vendor ID and the PCI Device ID + * to the different groups of PowerVR 5-series chip designs + * + * 0x8086 = Intel Corporation + * + * PowerVR SGX535 - Poulsbo - Intel GMA 500, Intel Atom Z5xx + * PowerVR SGX535 - Moorestown - Intel GMA 600 + * PowerVR SGX535 - Oaktrail - Intel GMA 600, Intel Atom Z6xx, E6xx + * PowerVR SGX540 - Medfield - Intel Atom Z2460 + * PowerVR SGX544MP2 - Medfield - + * PowerVR SGX545 - Cedartrail - Intel GMA 3600, Intel Atom D2500, N2600 + * PowerVR SGX545 - Cedartrail - Intel GMA 3650, Intel Atom D2550, D2700, N2800 + */ static DEFINE_PCI_DEVICE_TABLE(pciidlist) = { { 0x8086, 0x8108, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops }, { 0x8086, 0x8109, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops }, @@ -207,8 +220,7 @@ static int psb_driver_unload(struct drm_device *dev) { struct drm_psb_private *dev_priv = dev->dev_private; - /* Kill vblank etc here */ - + /* TODO: Kill vblank etc here */ if (dev_priv) { if (dev_priv->backlight_device) @@ -268,7 +280,22 @@ static int psb_driver_unload(struct drm_device *dev) return 0; } - +/* + * psb_driver_load - setup chip and create an initial config + * @dev: DRM device + * @flags: startup flags, containing the driver_data field belonging to + * the PCI device ID + * + * The driver load routine has to do several things: + * - allocating and initializing driver private data + * - performing resource allocation and mapping + * - initialize the memory manager + * - setup the DRM framebuffer with the allocated memory + * - install the IRQ handler + * - setup vertical blanking handling + * - mode setting + * - set inital output configuration + */ static int psb_driver_load(struct drm_device *dev, unsigned long chipset) { struct drm_psb_private *dev_priv; @@ -278,6 +305,7 @@ static int psb_driver_load(struct drm_device *dev, unsigned long chipset) struct drm_connector *connector; struct gma_encoder *gma_encoder; + /* allocating and initializing driver private data */ dev_priv = kzalloc(sizeof(*dev_priv), GFP_KERNEL); if (dev_priv == NULL) return -ENOMEM; @@ -369,6 +397,9 @@ static int psb_driver_load(struct drm_device *dev, unsigned long chipset) acpi_video_register(); + /* + * Setup vertical blanking handling + */ ret = drm_vblank_init(dev, dev_priv->num_pipe); if (ret) goto out_err; @@ -416,11 +447,11 @@ static int psb_driver_load(struct drm_device *dev, unsigned long chipset) return ret; psb_intel_opregion_enable_asle(dev); #if 0 - /*enable runtime pm at last*/ + /* Enable runtime pm at last */ pm_runtime_enable(&dev->pdev->dev); pm_runtime_set_active(&dev->pdev->dev); #endif - /*Intel drm driver load is done, continue doing pvr load*/ + /* Intel drm driver load is done, continue doing pvr load */ return 0; out_err: psb_driver_unload(dev); @@ -561,7 +592,7 @@ static int psb_mode_operation_ioctl(struct drm_device *dev, void *data, arg->data = resp; } - /*do some clean up work*/ + /* Do some clean up work */ if (mode) drm_mode_destroy(dev, mode); mode_op_out: @@ -614,8 +645,8 @@ static long psb_unlocked_ioctl(struct file *filp, unsigned int cmd, /* FIXME: do we need to wrap the other side of this */ } - -/* When a client dies: +/* + * When a client dies: * - Check for and clean up flipped page state */ static void psb_driver_preclose(struct drm_device *dev, struct drm_file *priv) @@ -655,6 +686,13 @@ static const struct file_operations psb_gem_fops = { .read = drm_read, }; +/* + * DRM driver structure initialization + * + * The drm_driver structure contains static information that describes + * the driver and features it supports, and pointers to methods that DRM + * core will call to implement DRM API. + */ static struct drm_driver driver = { .driver_features = DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | \ DRIVER_MODESET | DRIVER_GEM , diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h index 5ad6a03..85c560e 100644 --- a/drivers/gpu/drm/gma500/psb_drv.h +++ b/drivers/gpu/drm/gma500/psb_drv.h @@ -34,6 +34,19 @@ #include "opregion.h" #include "oaktrail.h" +/* + * Driver definitions + */ + +/* + * Driver history (deprecated) + * + * The driver version was intended for ABI changes and it's not used anymore. + * There are better ways to tell userspace what features we expose. + * + * TODO: describe better ways + */ + /* Append new drm mode definition here, align with libdrm definition */ #define DRM_MODE_SCALE_NO_SCALE 2 @@ -88,15 +101,11 @@ enum { #define _PSB_PGETBL_ENABLED 0x00000001 #define PSB_SGX_2D_SLAVE_PORT 0x4000 -/* To get rid of */ +/* TODO: To get rid of */ #define PSB_TT_PRIV0_LIMIT (256*1024*1024) #define PSB_TT_PRIV0_PLIMIT (PSB_TT_PRIV0_LIMIT >> PAGE_SHIFT) /* - * SGX side MMU definitions (these can probably go) - */ - -/* * Flags for external memory type field. */ #define PSB_MMU_CACHED_MEMORY 0x0001 /* Bind to MMU only */ @@ -519,7 +528,7 @@ struct drm_psb_private { uint32_t num_pipe; /* - * OSPM info (Power management base) (can go ?) + * OSPM info (Power management base) (TODO: can go ?) */ uint32_t ospm_base; @@ -766,9 +775,8 @@ extern void psb_mmu_remove_pages(struct psb_mmu_pd *pd, uint32_t desired_tile_stride, uint32_t hw_tile_stride); /* - *psb_irq.c + * psb_irq.c */ - extern irqreturn_t psb_irq_handler(int irq, void *arg); extern int psb_irq_enable_dpst(struct drm_device *dev); extern int psb_irq_disable_dpst(struct drm_device *dev); -- 1.9.0

2 6

← Newer
1
...
77
78
79
80
81
82
83
...
93
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel March 2014