aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2017-08-11drm/mediatek: switch to drm_*_get(), drm_*_put() helpersCihangir Akturk2-3/+3
Use drm_*_get() and drm_*_put() helpers instead of drm_*_reference() and drm_*_unreference() helpers. drm_*_reference() and drm_*_unreference() functions are just compatibility alias for drm_*_get() and drm_*_put() and should not be used by new code. So convert all users of compatibility functions to use the new APIs. Generated by: scripts/coccinelle/api/drm-get-put.cocci Signed-off-by: Cihangir Akturk <[email protected]> Reviewed-by: Philipp Zabel <[email protected]> Signed-off-by: Sean Paul <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-11drm/i915/fbc: only update no_fbc_reason when activeDaniel Vetter1-1/+3
In our snb farm in CI we have plenty of underruns, but not enough stolen memory to enable fbc. Which means every time there's an underrun the no_fbc_reason swichtes to something that makes kms_frontbuffer_tracking fail instead of skip, adding massive amounts of additional noise to igt test runs. Make sure we don't try to disable fbc when it's off already. v2: Squash in additional WARN_ON suggestion from Chris. Cc: Paulo Zanoni <[email protected]> Reviewed-by: Chris Wilson <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-11Merge branch 'linus' into locking/core, to resolve conflictsIngo Molnar29-208/+239
Conflicts: include/linux/mm_types.h mm/huge_memory.c I removed the smp_mb__before_spinlock() like the following commit does: 8b1b436dd1cc ("mm, locking: Rework {set,clear,mm}_tlb_flush_pending()") and fixed up the affected commits. Signed-off-by: Ingo Molnar <[email protected]>
2017-08-11drm: Clean up drm_dev_unplugDaniel Vetter1-12/+3
Use drm_dev_unregister to unregister the interfaces, which also allows us to simplify the open_count == 0 case. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-11drm: Only lastclose on unload for legacy driversDaniel Vetter1-1/+2
The only thing modern drivers are supposed to do in lastclose is restore the fb emulation state. Which is entirely optional, and there's really no reason to do that. So restrict it to legacy drivers (where the driver cleanup essentially happens in lastclose). This will also allow us to share the unregister function with drm_dev_unplug(). Quoting my reply to Alex on dri-devel: On Thu, Aug 3, 2017 at 1:17 AM, Daniel Vetter <[email protected]> wrote: > On Wed, Aug 2, 2017 at 10:50 PM, Alex Deucher <[email protected]> wrote: >> On Wed, Aug 2, 2017 at 7:56 AM, Daniel Vetter <[email protected]> wrote: >>> The only thing modern drivers are supposed to do in lastclose is >>> restore the fb emulation state. Which is entirely optional, and >>> there's really no reason to do that. So restrict it to legacy drivers >>> (where the driver cleanup essentially happens in lastclose). >> >> vga_switcheroo_process_delayed_switch() gets called in lastclose. >> Won't that need to get moved elsewhere for this to work? > > Hm right, I forgot the lazy way to do runtime pm by keeping the device > alive as long as anyone has an open fd for it ... This shouldn't be a > problem, since you need to unregister from vgaswitcheroo anyway on > unload. Maybe that blows up, I'll check the code and augment the patch > as needed. So I think there's 3 cases: - Trying to unload the module. You can't do that while anyone has the fd still open, so lastclose is guaranteeed to run. - Forcefully unbinding the driver through sysfs. Not any better, since the can_switch stuff checks for the open count, and so will delay the delayed switch no matter what. - Actual hotremoval: a) not implemented since none of the drivers taking part in vgaswitcheroo correctly unplug the drm device and b) you can't hotremove a chip from a laptop. v2: Extend commit message with m-l discussion. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-11drm: Document device unplug infrastructureDaniel Vetter10-13/+31
While at it, also ocd and give them a consistent drm_dev_ prefix, like the other device instance functionality. Plus move the functions into the right places. Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-11drm/imx: ipuv3-plane: fix YUV framebuffer scanout on the base planePhilipp Zabel1-4/+2
Historically, only RGB framebuffers could be assigned to the primary plane. This changed with universal plane support. Since no colorspace conversion was set up for the IPUv3 full plane, assigning YUV frame buffers to the primary plane caused incorrect output. Fix this by enabling color space conversion also for the primary plane. Signed-off-by: Philipp Zabel <[email protected]>
2017-08-11gpu: ipu-v3: add DRM dependencyArnd Bergmann1-0/+1
The new PRE/PRG driver code causes a link failure when IPUv3 is built-in, but DRM is built as a module: drivers/gpu/ipu-v3/ipu-pre.o: In function `ipu_pre_configure': ipu-pre.c:(.text.ipu_pre_configure+0x18): undefined reference to `drm_format_info' drivers/gpu/ipu-v3/ipu-prg.o: In function `ipu_prg_format_supported': ipu-prg.c:(.text.ipu_prg_format_supported+0x8): undefined reference to `drm_format_info' Adding a Kconfig dependency on DRM means we don't run into this problem any more. If DRM is disabled altogether, the IPUv3 driver is built without PRE/PRG support. Fixes: ea9c260514c1 ("gpu: ipu-v3: add driver for Prefetch Resolve Gasket") Link: https://patchwork.kernel.org/patch/9636665/ Signed-off-by: Arnd Bergmann <[email protected]> [[email protected]: changed the dependency from DRM to DRM || !DRM, since the link failure only happens when DRM=m and IPUV3_CORE=y. Modified the commit message to reflect this.] Signed-off-by: Philipp Zabel <[email protected]>
2017-08-10drm/i915/cnl: Add slice and subslice information to debugfs.Rodrigo Vivi1-1/+1
A missing part to EU slice power gating is the debugfs interface. This patch actually should have been squashed to the initial EU slice power gating one. v2: Initial patch was merged without this part. Fixes: c7ae7e9ab207 ("drm/i915/cnl: Configure EU slice power gating.") Cc: Joonas Lahtinen <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915/gen10: fix WM latency printingPaulo Zanoni1-1/+1
Gen 10 is just like Gen 9, so let's consider that all the future platforms are going to be like gen 9 instead of being like gen8-. Cc: Mahesh Kumar <[email protected]> Cc: Maarten Lankhorst <[email protected]> Signed-off-by: Paulo Zanoni <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Maarten Lankhorst <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915/gen10: fix the gen 10 SAGV block timePaulo Zanoni1-7/+6
A previous commit added CNL to intel_has_sagv(), but forgot to adjust the SAGV block time to gen 10 platforms. Cc: Mahesh Kumar <[email protected]> Cc: Maarten Lankhorst <[email protected]> Signed-off-by: Paulo Zanoni <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Maarten Lankhorst <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915/cnl: Enable SAGV for Cannonlake.Rodrigo Vivi1-1/+2
For now inherit from previous platforms. v2: Rebase on top of CFL. Cc: Mahesh Kumar <[email protected]> Cc: Maarten Lankhorst <[email protected]> Cc: Paulo Zanoni <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Maarten Lankhorst <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915/gen10+: use the SKL code for reading WM latenciesPaulo Zanoni1-2/+4
Gen 10 should use the exact same code as Gen 9, so change the check to take this into consideration, and also assume that future platforms will run this code. Also add a MISSING_CASE(), just in case we do something wrong, instead of silently failing. Cc: Mahesh Kumar <[email protected]> Cc: Maarten Lankhorst <[email protected]> Signed-off-by: Paulo Zanoni <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Maarten Lankhorst <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915: Avoid null dereference if mst_port is unset.Rodrigo Vivi1-0/+3
I'm not sure if this is really the case and I don't believe this is the real fix for the bug mentioned here, but since I don't see a reliable path when mst_port is set and when mode_valid is requested I believe it is worth to have this protection here. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102022 Cc: Elizabeth <[email protected]> Cc: Stefan Assmann <[email protected]> Cc: Dhinakaran Pandiyan <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Reviewed-by: Chris Wilson <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915/perf: Drop redundant check for perf.initialised on resetChris Wilson1-5/+2
As we cannot have an exclusive stream set if the perf has not been initialized, we only need to check for that exclusive stream. Signed-off-by: Chris Wilson <[email protected]> Cc: Lionel Landwerlin <[email protected]> Cc: Matthew Auld <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Matthew Auld <[email protected]>
2017-08-10drm/i915/perf: Drop lockdep assert for i915_oa_init_reg_state()Chris Wilson1-2/+0
This is called from execlist context init which we need to be unlocked. Commit f89823c21224 ("drm/i915/perf: Implement I915_PERF_ADD/REMOVE_CONFIG interface") added a lockdep assert to this path for unclear reasons, remove it again! Fixes: f89823c21224 ("drm/i915/perf: Implement I915_PERF_ADD/REMOVE_CONFIG interface") Signed-off-by: Chris Wilson <[email protected]> Cc: Lionel Landwerlin <[email protected]> Cc: Matthew Auld <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Matthew Auld <[email protected]>
2017-08-10drm/i915/perf: Initialise dynamic sysfs group before creationChris Wilson1-2/+1
Another case where we need to call sysfs_attr_init() to setup the internal lockdep class prior to use: [ 9.325229] BUG: key ffff880168bc7bb0 not in .data! [ 9.325240] DEBUG_LOCKS_WARN_ON(1) [ 9.325250] ------------[ cut here ]------------ [ 9.325280] WARNING: CPU: 1 PID: 275 at kernel/locking/lockdep.c:3156 lockdep_init_map+0x1b2/0x1c0 [ 9.325301] Modules linked in: intel_powerclamp(+) coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel i915(+) snd_hda_intel snd_hda_codec snd_hwdep r8169 mii snd_hda_core snd_pcm prime_numbers i2c_hid pinctrl_geminilake pinctrl_intel [ 9.325375] CPU: 1 PID: 275 Comm: modprobe Not tainted 4.13.0-rc4-CI-Trybot_1040+ #1 [ 9.325395] Hardware name: Intel Corp. Geminilake/GLK RVP2 LP4SD (07), BIOS GELKRVPA.X64.0045.B51.1704281422 04/28/2017 [ 9.325422] task: ffff8801721a4ec0 task.stack: ffffc900001dc000 [ 9.325440] RIP: 0010:lockdep_init_map+0x1b2/0x1c0 [ 9.325456] RSP: 0018:ffffc900001dfa10 EFLAGS: 00010282 [ 9.325473] RAX: 0000000000000016 RBX: ffff880168d54b80 RCX: 0000000000000000 [ 9.325488] RDX: 0000000080000001 RSI: 0000000000000001 RDI: ffffffff810f0800 [ 9.325505] RBP: ffffc900001dfa30 R08: 0000000000000001 R09: 0000000000000000 [ 9.325521] R10: 0000000000000000 R11: 0000000000000000 R12: ffff880168bc7bb0 [ 9.325537] R13: 0000000000000000 R14: ffff880168bc7b98 R15: ffffffff81a263a0 [ 9.325554] FS: 00007fb60c3fd700(0000) GS:ffff88017fc80000(0000) knlGS:0000000000000000 [ 9.325574] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 9.325588] CR2: 0000006582777d80 CR3: 000000016d818000 CR4: 00000000003406e0 [ 9.325604] Call Trace: [ 9.325618] __kernfs_create_file+0x76/0xe0 [ 9.325632] sysfs_add_file_mode_ns+0x8a/0x1a0 [ 9.325646] internal_create_group+0xea/0x2c0 [ 9.325660] sysfs_create_group+0x13/0x20 [ 9.325737] i915_perf_register+0xde/0x220 [i915] [ 9.325800] i915_driver_load+0xa77/0x16c0 [i915] [ 9.325863] i915_pci_probe+0x37/0x90 [i915] [ 9.325880] pci_device_probe+0xa8/0x130 [ 9.325894] driver_probe_device+0x29c/0x450 [ 9.325908] __driver_attach+0xe3/0xf0 [ 9.325922] ? driver_probe_device+0x450/0x450 [ 9.325935] bus_for_each_dev+0x62/0xa0 [ 9.325948] driver_attach+0x1e/0x20 [ 9.325960] bus_add_driver+0x173/0x270 [ 9.325974] driver_register+0x60/0xe0 [ 9.325986] __pci_register_driver+0x60/0x70 [ 9.326044] i915_init+0x6f/0x78 [i915] [ 9.326066] ? 0xffffffffa024e000 [ 9.326079] do_one_initcall+0x43/0x170 [ 9.326094] ? rcu_read_lock_sched_held+0x7a/0x90 [ 9.326109] ? kmem_cache_alloc_trace+0x261/0x2d0 [ 9.326124] do_init_module+0x5f/0x206 [ 9.326137] load_module+0x2561/0x2da0 [ 9.326150] ? show_coresize+0x30/0x30 [ 9.326165] ? kernel_read_file+0x105/0x190 [ 9.326180] SyS_finit_module+0xc1/0x100 [ 9.326192] ? SyS_finit_module+0xc1/0x100 [ 9.326210] entry_SYSCALL_64_fastpath+0x1c/0xb1 [ 9.326223] RIP: 0033:0x7fb60bf359f9 [ 9.326234] RSP: 002b:00007fff92b47c48 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 9.326255] RAX: ffffffffffffffda RBX: ffffffff814898a3 RCX: 00007fb60bf359f9 [ 9.326271] RDX: 0000000000000000 RSI: 00000028a9ceef8b RDI: 0000000000000000 [ 9.326287] RBP: ffffc900001dff88 R08: 0000000000000000 R09: 0000000000000000 [ 9.326303] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000040000 [ 9.326319] R13: 00000028aaef2a70 R14: 0000000000000000 R15: 00000028aaeee5d0 [ 9.326339] ? __this_cpu_preempt_check+0x13/0x20 [ 9.326353] Code: f1 39 00 85 c0 0f 84 38 ff ff ff 83 3d 9f 44 ce 01 00 0f 85 2b ff ff ff 48 c7 c6 b2 a2 c7 81 48 c7 c7 53 40 c5 81 e8 3f 82 01 00 <0f> ff e9 11 ff ff ff 0f 1f 80 00 00 00 00 55 31 c9 31 d2 31 f6 Fixes: 701f8231a2fe ("drm/i915/perf: prune OA configs") Signed-off-by: Chris Wilson <[email protected]> Cc: Lionel Landwerlin <[email protected]> Cc: Matthew Auld <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Lionel Landwerlin <[email protected]>
2017-08-10drm/i915: add register macro definition style guideJani Nikula1-0/+91
This is not to try to force a new style; this is my interpretation of what the most common existing style is. With hopes I don't need to answer so many questions about style going forward. Start a new style section in the i915 document to bolt the register style guide into. v2: vertical alignment, incorporate to kernel-doc, and more Cc: Daniel Vetter <[email protected]> Reviewed-by: Rodrigo Vivi <[email protected]> Reviewed-by: Daniel Vetter <[email protected]> Reviewed-by: Dhinakaran Pandiyan <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/9de4a5b1bea4e76461c70a1dd66751581de0124f.1502368010.git.jani.nikula@intel.com
2017-08-10drm/i915: enum i915_power_well_id is not proper kernel-docJani Nikula1-1/+1
Revert to a normal comment, as the enum isn't properly documented anyway. Signed-off-by: Jani Nikula <[email protected]> Reviewed-by: Rodrigo Vivi <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/f2c44a0aa00ea7d9b71e7a3183a7507f98811146.1502368010.git.jani.nikula@intel.com
2017-08-10drm/i915: Add format modifiers for IntelBen Widawsky3-12/+292
This was based on a patch originally by Kristian. It has been modified pretty heavily to use the new callbacks from the previous patch. v2: - Add LINEAR and Yf modifiers to list (Ville) - Combine i8xx and i965 into one list of formats (Ville) - Allow 1010102 formats for Y/Yf tiled (Ville) v3: - Handle cursor formats (Ville) - Put handling for LINEAR in the mod_support functions (Ville) v4: - List each modifier explicitly in supported modifiers (Ville) - Handle the CURSOR plane (Ville) v5: - Split out cursor and sprite handling (Ville) v6: - Actually use the sprite funcs (Emil) - Use unreachable (Emil) v7: - Only allow Intel modifiers and LINEAR (Ben) v8 - Fix spite assert introduced in v6 (Daniel) v9 - Change vendor check logic to avoid magic 56 (Emil) - Reorder skl_mod_support (Ville) - make intel_plane_funcs static, could be done as of v5 (Ville) - rename local variable intel_format_modifiers to modifiers (Ville) - actually use sprite modifiers - split out modifier/formats by platform (Ville) v10: - Undo vendor check from v9 v11: - Squash CCS advertisement into this patch (daniels) - Don't advertise CCS on higher sprite planes (daniels) v12: - Don't advertise Y-tiled or CCS on any sprite planes, since we don't allocate enough DDB space for it to work. (daniels) Cc: Ville Syrjälä <[email protected]> Reviewed-by: Emil Velikov <[email protected]> (v8) Signed-off-by: Ben Widawsky <[email protected]> Signed-off-by: Daniel Stone <[email protected]>
2017-08-10drm/i915: Add render decompression supportVille Syrjälä4-18/+272
SKL+ display engine can scan out certain kinds of compressed surfaces produced by the render engine. This involved telling the display engine the location of the color control surfae (CCS) which describes which parts of the main surface are compressed and which are not. The location of CCS is provided by userspace as just another plane with its own offset. Add the required stuff to validate the user provided AUX plane metadata and convert the user provided linear offset into something the hardware can consume. Due to hardware limitations we require that the main surface and the AUX surface (CCS) be part of the same bo. The hardware also makes life hard by not allowing you to provide separate x/y offsets for the main and AUX surfaces (excpet with NV12), so finding suitable offsets for both requires a bit of work. Assuming we still want keep playing tricks with the offsets. I've just gone with a dumb "search backward for suitable offsets" approach, which is far from optimal, but it works. Also not all planes will be capable of scanning out compressed surfaces, and eg. 90/270 degree rotation is not supported in combination with decompression either. This patch may contain work from at least the following people: * Vandana Kannan <[email protected]> * Daniel Vetter <[email protected]> * Ben Widawsky <[email protected]> v2: Deal with display workarounds 0390, 0531, 1125 (Paulo) v3: Pretend CCS tiles are regular 128 byte wide Y tiles (Jason) Put the AUX register defines to the correct place Fix up the slightly bogus rotation check v4: Use I915_WRITE_FW() due to plane update locking changes s/return -EINVAL/goto err/ in intel_framebuffer_init() Eliminate a bunch hardcoded numbers in CCS code v5: (By Ben) conflict resolution + - res_blocks += fixed_16_16_to_u32_round_up(y_tile_minimum); + res_blocks += fixed16_to_u32_round_up(y_tile_minimum); v6: (daniels) Fix botched commit message. Cc: Paulo Zanoni <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: Ben Widawsky <[email protected]> Cc: Jason Ekstrand <[email protected]> Signed-off-by: Ville Syrjä <[email protected]> Reviewed-by: Ben Widawsky <[email protected]> (v1) Reviewed-by: Daniel Stone <[email protected]> Signed-off-by: Ben Widawsky <[email protected]> Signed-off-by: Daniel Stone <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-10drm/i915: Implement .get_format_info() hook for CCSVille Syrjälä1-0/+36
SKL+ display engine can scan out certain kinds of compressed surfaces produced by the render engine. This involved telling the display engine the location of the color control surfae (CCS) which describes which parts of the main surface are compressed and which are not. The location of CCS is provided by userspace as just another plane with its own offset. By providing our own format information for the CCS formats, we should be able to make framebuffer_check() do the right thing for the CCS surface as well. Note that we'll return the same format info for both Y and Yf tiled format as that's what happens with the non-CCS Y vs. Yf as well. If desired, we could potentially return a unique pointer for each pixel_format+tiling+ccs combination, in which case we immediately be able to tell if any of that stuff changed by just comparing the pointers. But that does sound a bit wasteful space wise. v2: Drop the 'dev' argument from the hook v3: Include the description of the CCS surface layout v4: Pretend CCS tiles are regular 128 byte wide Y tiles (Jason) v5: Re-drop 'dev', fix commit message, add missing drm_fourcc.h description of CCS layout. (daniels) Cc: Daniel Vetter <[email protected]> Cc: Ben Widawsky <[email protected]> Cc: Jason Ekstrand <[email protected]> Acked-by: Jason Ekstrand <[email protected]> Reviewed-by: Ben Widawsky <[email protected]> (v3) Reviewed-by: Daniel Stone <[email protected]> Signed-off-by: Ville Syrjä <[email protected]> Signed-off-by: Ben Widawsky <[email protected]> Signed-off-by: Daniel Stone <[email protected]>
2017-08-10Merge airlied/drm-next into drm-intel-next-queuedDaniel Vetter320-4021/+8235
Ben Widawsky/Daniel Stone need the extended modifier support from drm-misc to be able to merge CCS support for i915.ko Signed-off-by: Daniel Vetter <[email protected]>
2017-08-10drm/i915: Supply the engine-id for our mock_engine()Chris Wilson3-5/+8
In the original selftest, we didn't care what the engine->id was, just that it could uniquely identify it. Later though, we started tracking the mock engines in the fixed size arrays around the drm_i915_private and so we now require their indices to be correct. This becomes an issue when using the standalone harness which runs all available tests at module load, and so we quickly assign an out-of-bounds index to an engine as we reallocate the mock GEM device between tests. It doesn't show up in igt/drv_selftest as that runs each subtest individually. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102045 Signed-off-by: Chris Wilson <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Mika Kuoppala <[email protected]>
2017-08-10locking/lockdep: Rework FS_RECLAIM annotationPeter Zijlstra1-2/+3
A while ago someone, and I cannot find the email just now, asked if we could not implement the RECLAIM_FS inversion stuff with a 'fake' lock like we use for other things like workqueues etc. I think this should be possible which allows reducing the 'irq' states and will reduce the amount of __bfs() lookups we do. Removing the 1 IRQ state results in 4 less __bfs() walks per dependency, improving lockdep performance. And by moving this annotation out of the lockdep code it becomes easier for the mm people to extend. Signed-off-by: Peter Zijlstra (Intel) <[email protected]> Cc: Byungchul Park <[email protected]> Cc: Linus Torvalds <[email protected]> Cc: Mel Gorman <[email protected]> Cc: Michal Hocko <[email protected]> Cc: Nikolay Borisov <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Signed-off-by: Ingo Molnar <[email protected]>
2017-08-10drm/i915/gvt: Add shadow context descriptor updatingKechen Lu3-0/+26
The current context logic only updates the descriptor of context when it's being pinned to graphics memory space. But this cannot satisfy the requirement of shadow context. The addressing mode of the pinned shadow context descriptor may be changed according to the guest addressing mode. And this won't be updated, as the already pinned shadow context has no chance to update its descriptor. And this will lead to GPU hang issue, as shadow context is used with wrong descriptor. This patch fixes this issue by letting the pinned shadow context descriptor update its addressing mode on demand. This patch fixes GPU HANG issue which happends after changing the grub parameter i915.enable_ppgtt form 0x01 to 0x03 or vice versa and then rebooting the guest. Signed-off-by: Tina Zhang <[email protected]> Signed-off-by: Kechen Lu <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: expose vGPU context hw idZhenyu Wang1-0/+17
This exposes vGPU context hw id in mdev sysfs which is used to do vGPU based profiling. Retrieved vGPU context hw id can be set through i915 perf ioctl to set profiling for target vGPU. Cc: Jiao Pengyuan <[email protected]> Cc: Niu Bing <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Refine the intel_vgpu_reset_gtt reset functionChuanxiao Dong3-9/+5
When doing the VGPU reset, we don't need to do the gtt/ppgtt reset. This will make the GVT to do the ppgtt shadow every time for a workload and caused really bad performance after a VGPU reset. This patch will make sure ppgtt clean only happen at device module level reset to fix this. Signed-off-by: Chuanxiao Dong <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Add carefully checking in GTT walker pathsChangbin Du2-37/+64
When debugging the gtt code, found the intel_vgpu_gma_to_gpa() can translate any given GMA though the GMA is not valid. This because the GTT ops suppress the possible errors, which may result in an invalid PT entry is retrieved by upper caller. This patch changed the prototype of pte ops to propagate status to callers. Then we make sure the GTT walker stop as early as when a error is detected to prevent undefined behavior. Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Remove duplicated MMIO entriesJian Jun Chen1-8/+10
Remove duplicated MMIO entries in the tracked MMIO list. -EEXIST is returned if duplicated MMIO entries are found when new MMIO entry is added. v2: - Use WARN(1, ...) for more verbose message. (Zhenyu) Signed-off-by: Jian Jun Chen <[email protected]> Cc: Zhi Wang <[email protected]> Cc: Changbin Du <[email protected]> Cc: Zhenyu Wang <[email protected]> Reviewed-by: Yulei Zhang <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: take runtime pm when do early scan and shadowZhenyu Wang1-0/+2
Need to take runtime pm when do early scan/shadow of workload for request operations. Fixes: 7fa56bd159bc ("drm/i915/gvt: Audit and shadow workload during ELSP writing") Cc: Ping Gao <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Replace duplicated code with exist functionPing Gao1-3/+3
Use the exist function intel_gvt_ggtt_validate_range to replace these duplicated code that do the same thing. Signed-off-by: Ping Gao <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: To check whether workload scan and shadow has mutex holdPing Gao1-0/+2
The function workload scan and shadow have to hold the drm.struct_mutex before called. To avoid misusing of this function, add a lockdep assert in it. Signed-off-by: Ping Gao <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Audit and shadow workload during ELSP writingPing Gao3-0/+19
Let the workload audit and shadow ahead of vGPU scheduling, that will eliminate GPU idle time and improve performance for multi-VM. The performance of Heaven running simultaneously in 3VMs has improved 20% after this patch. v2:Remove condition current->vgpu==vgpu when shadow during ELSP writing. Signed-off-by: Ping Gao <[email protected]> Reviewed-by: Zhi Wang <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Factor out scan and shadow from workload dispatchPing Gao4-35/+54
To perform the workload scan and shadow in ELSP writing stage for performance consideration, the workload scan and shadow stuffs should be factored out from dispatch_workload(). v2:Put context pin before i915_add_request; Refine the comments; Rename some APIs; v3:workload->status should set only when error happens. v4:i915_add_request is must to have after i915_gem_request_alloc. Signed-off-by: Ping Gao <[email protected]> Reviewed-by: Zhi Wang <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Optimize ring siwtch 2x faster again by light weight mmio ↵Changbin Du1-13/+26
access wrapper The I915_READ/WRITE is not only a mmio read/write, it also contains debug checking and Forcewake domain lookup. This is too heavy for GVT ring switch case which access batch of mmio registers on ring switch. We can handle Forcewake manually and use the raw i915_read/write instead. The benefit from this is 2x faster mmio switch performance. Before After cycles ~550000 ~250000 v2: Use existing I915_READ_FW/I915_WRITE_FW macro. (zhenyu) Signed-off-by: Changbin Du <[email protected]> Reviewed-by: Zhenyu Wang <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Optimize ring siwtch 2x faster by removing unnecessary ↵Changbin Du1-6/+13
POSTING_READ There are lots of POSTING_READ alongside each mmio write Op. While actually this is not necessary. It just bring too much latency since PCIe read Op is very slow which is of non-posted transaction. For PCIe device, the mem transaction for strong ordering rules are: o PCIe mmio write sequence is FIFO. Posted request cannot pass previous posted request. o PCIe mmio read will not go ahead of previous write. Intel graphics doesn't support RO, so we can apply above rules. In our case, we only need one POSTING_READ at last. This can remove half of mmio read Op and then the average ring switch performance is nearly doubled. Before After cycles ~970000 ~550000 Signed-off-by: Changbin Du <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10drm/i915/gvt: Use gvt_err to print the resource not enough errorChuanxiao Dong1-2/+2
It is better to use gvt_err when the gvt resource is not enough so the user can be notified from the kernel dmesg. And this kind of error message is gvt related. Suggested-by: Bing Niu <[email protected]> Signed-off-by: Chuanxiao Dong <[email protected]> Cc: Bing Niu <[email protected]> Signed-off-by: Zhenyu Wang <[email protected]>
2017-08-10Merge branch 'linux-4.13' of git://github.com/skeggsb/linux into drm-fixesDave Airlie1-0/+2
single nouveau regression fix. * 'linux-4.13' of git://github.com/skeggsb/linux: drm/nouveau/disp/nv04: avoid creation of output paths
2017-08-10drm/nouveau/disp/nv04: avoid creation of output pathsBen Skeggs1-0/+2
Fixes hitting WARN_ON() during initialisation of pre-NV50 GPUs, caused by the recent changes to support pad macro routing on GM20x. We currently don't use them here for older GPUs anyway. Signed-off-by: Ben Skeggs <[email protected]>
2017-08-10drm: make DRM_STM default nMichał Mirosław1-1/+0
Default config value for all other drivers is N. Signed-off-by: Michał Mirosław <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
2017-08-10Merge tag 'drm-misc-next-2017-08-08' of ↵Dave Airlie209-2037/+3450
git://anongit.freedesktop.org/git/drm-misc into drm-next UAPI Changes: - vc4: Add ioctl to allow attaching a label to a bo (Eric) - Add new format/modifier blob plane property (Ben) - armada: Use __u32/__u64 instead of uint32_t/uint64_t (Mikko) - [kinda uapi] fb_helper: Expose display_info size via fb_info (David) Core Changes: - Default gem_dumb_[map_offset|destroy] as mmap/destroy implementations (Noralf) - Simplify atomic properties by removing the helpers and handling in core (Daniel) Driver Changes: - stm: Add STM32 DSI controller driver (Phillipe) - vc4: Add HDMI CEC support (Hans) - rockchip: Refactor register init & soc version handling (Mark) - misc: Remove .load_lut, .gamma_set, .gamma_get dead code (Peter) - dw-hdmi: Add HDMI CEC support (Russell) Cc: Philippe CORNU <[email protected]> Cc: Hans Verkuil <[email protected]> Cc: Eric Anholt <[email protected]> Cc: Noralf Trønnes <[email protected]> Cc: Ben Widawsky <[email protected]> Cc: Mark yao <[email protected]> Cc: Peter Rosin <[email protected]> Cc: Russell King <[email protected]> Cc: Mikko Rapeli <[email protected]> Cc: David Lechner <[email protected]> Cc: Daniel Vetter <[email protected]> * tag 'drm-misc-next-2017-08-08' of git://anongit.freedesktop.org/git/drm-misc: (107 commits) drm: Nuke drm_atomic_legacy_backoff drm: Nuke drm_atomic_helper_connector_dpms drm: Nuke drm_atomic_helper_connector_set_property drm: Nuke drm_atomic_helper_plane_set_property drm: Nuke drm_atomic_helper_crtc_set_property drm: Handle properties in the core for atomic drivers drm: Don't update property values for atomic drivers drm/omap: Rework the rotation-on-crtc hack drm/radeon: Use the drm_driver.dumb_destroy default drm/i915: Use the drm_driver.dumb_destroy default drm/sti: Use .dumb_map_offset and .dumb_destroy defaults drm: bridge: synopsys/dw-hdmi: Provide default configuration function for HDMI 2.0 PHY drm/fb-helper: pass physical dimensions to fbdev uapi drm/armada_drm.h: use __u32 and __u64 instead of uint32_t and uint64_t drm/bridge: dw-hdmi: remove CEC engine register definitions drm/bridge: dw-hdmi: add cec driver drm/bridge: dw-hdmi: add missing cec_notifier_put drm: remove unused and redundant callbacks staging: vboxvideo: remove dead gamma lut code drm: dw-hdmi-i2s: add missing company name on Copyright ...
2017-08-10Merge branch 'etnaviv/fixes' of https://git.pengutronix.de/git/lst/linux ↵Dave Airlie1-2/+2
into drm-fixes single etnaviv fix. * 'etnaviv/fixes' of https://git.pengutronix.de/git/lst/linux: drm/etnaviv: Fix off-by-one error in reloc checking
2017-08-10Merge tag 'drm-intel-fixes-2017-08-09-1' of ↵Dave Airlie10-37/+82
git://anongit.freedesktop.org/git/drm-intel into drm-fixes drm/i915 fixes for v4.13-rc5 * tag 'drm-intel-fixes-2017-08-09-1' of git://anongit.freedesktop.org/git/drm-intel: drm/i915: fix backlight invert for non-zero minimum brightness drm/i915/shrinker: Wrap need_resched() inside preempt-disable drm/i915/perf: fix flex eu registers programming drm/i915: Fix out-of-bounds array access in bdw_load_gamma_lut drm/i915/gvt: Change the max length of mmio_reg_rw from 4 to 8 drm/i915/gvt: Initialize MMIO Block with HW state drm/i915/gvt: clean workload queue if error happened drm/i915/gvt: change resetting to resetting_eng
2017-08-10Merge tag 'drm-misc-fixes-2017-08-08' of ↵Dave Airlie3-22/+24
git://anongit.freedesktop.org/git/drm-misc into drm-fixes Core Changes: - dma-buf: Allow multiple sync_files to wrap a single dma-fence (Chris) Driver Changes: - rockchip: misc fixes to vop driver from the downstream rockchip tree (Mark) - Error path cleanups to tc358767 & host1x (Lucas & Paul, respectively) * tag 'drm-misc-fixes-2017-08-08' of git://anongit.freedesktop.org/git/drm-misc: drm/rockchip: vop: report error when check resource error drm/rockchip: vop: round_up pitches to word align drm/rockchip: vop: fix NV12 video display error drm/rockchip: vop: fix iommu page fault when resume dma-buf/sync_file: Allow multiple sync_files to wrap a single dma-fence drm/bridge: tc358767: fix probe without attached output node
2017-08-10Merge tag 'exynos-drm-fixes-for-v4.13-rc4' of ↵Dave Airlie1-1/+13
git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-fixes Fix a issue to display system memory region outside a gem buffer. * tag 'exynos-drm-fixes-for-v4.13-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos: drm/exynos: forbid creating framebuffers from too small GEM buffers
2017-08-10Merge branch 'msm-fixes-4.13-rc3' of ↵Dave Airlie12-145/+116
git://people.freedesktop.org/~robclark/linux into drm-fixes Bunch of msm fixes for 4.13 * 'msm-fixes-4.13-rc3' of git://people.freedesktop.org/~robclark/linux: drm/msm: gpu: don't abuse dma_alloc for non-DMA allocations drm/msm: gpu: call qcom_mdt interfaces only for ARCH_QCOM drm/msm/adreno: Prevent unclocked access when retrieving timestamps drm/msm: Remove __user from __u64 data types drm/msm: args->fence should be args->flags drm/msm: Turn off hardware clock gating before reading A5XX registers drm/msm: Allow hardware clock gating to be toggled drm/msm: Remove some potentially blocked register ranges drm/msm/mdp5: Drop clock names with "_clk" suffix drm/msm/mdp5: Fix typo in encoder_enable path drm/msm: NULL pointer dereference in drivers/gpu/drm/msm/msm_gem_vma.c drm/msm: fix WARN_ON in add_vma() with no iommu drm/msm/dsi: Calculate link clock rates with updated dsi->lanes drm/msm/mdp5: fix unclocked register access in _cursor_set() drm/msm: unlock on error in msm_gem_get_iova() drm/msm: fix an integer overflow test drm/msm/mdp5: Fix compilation warnings
2017-08-09drm/rockchip: Fix suspend crash when drm is not boundJeffy Chen1-2/+10
Currently we are allocating drm_device in rockchip_drm_bind, so if the suspend/resume code access it when drm is not bound, we would hit this crash: [ 253.402836] Unable to handle kernel NULL pointer dereference at virtual address 00000028 [ 253.402837] pgd = ffffffc06c9b0000 [ 253.402841] [00000028] *pgd=0000000000000000, *pud=0000000000000000 [ 253.402844] Internal error: Oops: 96000005 [#1] PREEMPT SMP [ 253.402859] Modules linked in: btusb btrtl btbcm btintel bluetooth ath10k_pci ath10k_core ar10k_ath ar10k_mac80211 cfg80211 ip6table_filter asix usbnet mii [ 253.402864] CPU: 4 PID: 1331 Comm: cat Not tainted 4.4.70 #15 [ 253.402865] Hardware name: Google Scarlet (DT) [ 253.402867] task: ffffffc076c0ce00 ti: ffffffc06c2c8000 task.ti: ffffffc06c2c8000 [ 253.402871] PC is at rockchip_drm_sys_suspend+0x20/0x5c Add sanity checks to prevent that. Reported-by: Brian Norris <[email protected]> Signed-off-by: Jeffy Chen <[email protected]> Signed-off-by: Sean Paul <[email protected]> Link: https://patchwork.kernel.org/patch/9890297/
2017-08-09drm/i915/psr: Preserve SRD_CTL bit 29 on PSR initJim Bride2-0/+2
Bit 29 of SRD_CTL needs to have its value preserved according to the B-Spec, so right before we write out the register we go ahead and read the register and preserve the value of that bit before we write out the configured register value. v2: Spaces => tabs, minor name change, and commit message wording (Rodrigo) Cc: Rodrigo Vivi <[email protected]> Cc: Chris Wilson <[email protected]> Cc: Jani Nikula <[email protected]> Signed-off-by: Jim Bride <[email protected]> Reviewed-by: Rodrigo Vivi <[email protected]> Signed-off-by: Rodrigo Vivi <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2017-08-09drm: make drm_mode_config_func constBhumika Goyal2-2/+2
Make these structures const as they are only stored in the funcs field of a drm_mode_config structure which is of type const. Done using Coccinelle: @match disable optional_qualifier@ identifier s; @@ static struct drm_mode_config_funcs s = {...}; @ref@ position p; identifier match.s; @@ s@p @good1@ identifier y; position ref.p; identifier match.s; @@ struct drm_mode_config y = {...,.funcs=&s@p,...}; @good2@ struct drm_mode_config y; identifier match.s; position ref.p; @@ y.funcs = &s@p @bad depends on !good1 && !good2@ position ref.p; identifier match.s; @@ s@p @depends on forall !bad disable optional_qualifier@ identifier match.s; @@ static + const struct drm_mode_config_funcs s; Signed-off-by: Bhumika Goyal <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Powered by cgit, Gruvboxed by Blaster4385.