aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2016-08-22drm/amdgpu: link all shadow bo V2Chunming Zhou3-1/+19
V2: 1. use mutex instead of spinlock for shadow list, since its process could sleep. 2. move list_del to bo destroy phase. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-22drm/amdgpu: update pt shadow while updating pt V2Chunming Zhou1-1/+15
V2: move shadow parameter to amdgpu_pte_update_params. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Cc: [email protected] Signed-off-by: Alex Deucher <[email protected]>
2016-08-22drm/amdgpu: update pd shadow while updating pd V2Chunming Zhou2-19/+42
V2: Checking if shadow is valid. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-22drm/amdgpu: sync bo and shadow V3Chunming Zhou3-8/+76
Use shadow flag to judge which direction to sync. V2: Don't need bo pin, so remove it. V3: 1. Split to two functions, one is backup_to_shadow, another is restore_from_shadow. 2. Clean up previous shadow direction difinitions. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-22drm/amdgpu: add direct submision option for copy_bufferChunming Zhou4-11/+21
Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-22drm: Avoid calling dev_printk(.dev = NULL)Chris Wilson1-2/+5
Since dev_printk likes to print "(NULL device *):" when passed in a NULL pointer, we have to manually call printk() ourselves. Fixes: c4e68a583202 ("drm: Introduce DRM_DEV_* log messages") Signed-off-by: Chris Wilson <[email protected]> Cc: Eric Engestrom <[email protected]> Cc: Sean Paul <[email protected]> Signed-off-by: Sean Paul <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-22drm: avoid exposing kernel stack in compat_drm_getstatsHeinrich Schuchardt1-0/+1
The C standard does not specify the size of the integer used to store an enum. Hence in structure drm_stats32_t alignment bytes may exist. To avoid exposing bytes from the kernel stack it is necessary to initialize variable s32 completely. Signed-off-by: Heinrich Schuchardt <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-22reservation: fix small comment typoRob Clark1-1/+1
Signed-off-by: Rob Clark <[email protected]> [danvet: Mark up as function for proper cross-linking.] Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-22drm/i915: Update DRIVER_DATE to 20160822Daniel Vetter1-1/+1
Signed-off-by: Daniel Vetter <[email protected]>
2016-08-19drm/amdgpu: add need backup function V2Chunming Zhou3-1/+10
V2: add checking if need backup in amdgpu_bo_create. Signed-off-by: Chunming Zhou <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/amd/amdgpu: S3 resumed failed after 4-5 times loopjimqu1-52/+81
Phenomenon: software hang when device resume back, read UVD fence is 0xffffffff and read pcie pid is 0xffff. The issue is caused by VCE reset when update cg setting. according to HW programming guide, adjust update VCE cg sequence. The patch apply to VCE2.0. Signed-off-by: JimQu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/amdgpu: use domain's gpu_offset for start addrFlora Cui2-11/+3
Signed-off-by: Flora Cui <[email protected]> Reviewed-by: Ken Wang <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/amdgpu: update gart_pin_size only if the bo is pined to GTTFlora Cui1-1/+1
Signed-off-by: Flora Cui <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/amdgpu: check domain sanity in amdgpu_bo_pin_restricted()Flora Cui1-0/+5
abort if the bo is pined to other domain already Signed-off-by: Flora Cui <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19amdgpu: move ttm stuff to amdgpu_ttm.hFlora Cui3-51/+80
Signed-off-by: Flora Cui <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/amd/powerplay: enable power containment features for tonga.Rex Zhu5-9/+548
v2: fix build error introduced when fix code style problems. Signed-off-by: Rex Zhu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2016-08-19drm/i915: Use remap_io_mapping() to prefault all PTE in a single passChris Wilson5-45/+99
Very old numbers indicate this is a 66% improvement when remapping the entire object for fence contention - due to the elimination of track_pfn_insert and its strcmp. Signed-off-by: Chris Wilson <[email protected]> Testcase: igt/gem_fence_upload/performance Testcase: igt/gem_mmap_gtt Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm/i915: Embed the io-mapping struct inside drm_i915_privateChris Wilson6-14/+13
As io_mapping.h now always allocates the struct, we can avoid that allocation and extra pointer dance by embedding the struct inside drm_i915_private Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19io-mapping: Always create a struct to hold metadata about the io-mappingChris Wilson1-35/+57
Currently, we only allocate a structure to hold metadata if we need to allocate an ioremap for every access, such as on x86-32. However, it would be useful to store basic information about the io-mapping, such as its page protection, on all platforms. Signed-off-by: Chris Wilson <[email protected]> Cc: [email protected] Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm/i915/fbc: Allow on unfenced surfaces, for recent genChris Wilson1-2/+4
Only fbc1 is tied to using a fence. Later iterations of fbc are more flexible and allow operation on unfenced frontbuffers. Signed-off-by: Chris Wilson <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: "Zanoni, Paulo R" <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm/i915/fbc: Don't set an illegal fence if unfencedChris Wilson1-16/+32
If the frontbuffer doesn't have an associated fence, it will have a fence reg of -1. If we attempt to OR in this register into the FBC control register we end up setting all control bits, oops! Signed-off-by: Chris Wilson <[email protected]> Cc: Joonas Lahtinen <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: "Zanoni, Paulo R" <[email protected]> Reviwed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm/i915: Flush delayed fence releases after resetChris Wilson1-1/+5
What I never hit in testing, but Mika immediately did, was a GPU hang with a pending fence release (where a tiled object has been changed by the user to be untiled, and the update has not yet been committed to the fence register). As the stride/tiling is 0, this causes a divide-by-zero error when trying to write the new fence parameters: [ 28.784518] drm/i915: Resetting chip after gpu hang [ 28.784551] divide error: 0000 [#1] PREEMPT SMP [ 28.784565] Modules linked in: nls_iso8859_1 nls_cp437 vfat fat mxm_wmi x86_pkg_temp_thermal snd_hda_codec_hdmi kvm irqbypass snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_codec serio_raw snd_hwdep snd_hda_core snd_pcm snd_seq_midi snd_seq_midi_event snd_rawmidi snd_seq snd_timer snd_seq_device snd soundcore mac_hid wmi efivarfs autofs4 raid10 raid456 libcrc32c async_raid6_recov async_memcpy async_pq raid6_pq async_xor xor async_tx raid0 multipath linear psmouse e1000e ptp pps_core nvme nvme_core i915 i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm video [ 28.784738] CPU: 0 PID: 1692 Comm: kworker/0:2 Not tainted 4.8.0-rc2+ #895 [ 28.784752] Hardware name: System manufacturer System Product Name/Z170M-PLUS, BIOS 1803 05/09/2016 [ 28.784786] Workqueue: events_long i915_hangcheck_elapsed [i915] [ 28.784814] task: ffff923c18f59d40 task.stack: ffff923c1b7e4000 [ 28.784827] RIP: 0010:[<ffffffffc0475b5f>] [<ffffffffc0475b5f>] fence_write+0x9f/0x3b0 [i915] [ 28.784854] RSP: 0018:ffff923c1b7e7b30 EFLAGS: 00010246 [ 28.784866] RAX: 00000000008ca000 RBX: ffff923c18540000 RCX: 0000000000000020 [ 28.784880] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000000596d000 [ 28.784894] RBP: ffff923c1b7e7b68 R08: 0000000000000000 R09: 0000000000000000 [ 28.784908] R10: 0000000000000000 R11: 00000000008ca000 R12: ffff923c1ef9d600 [ 28.784921] R13: 0000000000100040 R14: 0000000000100044 R15: ffff923c18549908 [ 28.784935] FS: 0000000000000000(0000) GS:ffff923c36c00000(0000) knlGS:0000000000000000 [ 28.784951] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 28.784962] CR2: 00007f193373c893 CR3: 0000000419c78000 CR4: 00000000003406f0 [ 28.784976] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 28.784990] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 28.785004] Stack: [ 28.785009] 000000000596c03b ffff923c1b7e7b68 ffff923c18549938 0000000000000009 [ 28.785026] ffff923c18540000 ffff923c18549280 ffff923c18547ce8 ffff923c1b7e7b90 [ 28.785044] ffffffffc04761f9 ffff923c18540000 ffff923c18547d00 ffff923c18548ff8 [ 28.785062] Call Trace: [ 28.785078] [<ffffffffc04761f9>] i915_gem_restore_fences+0x39/0x50 [i915] [ 28.785102] [<ffffffffc047fe89>] i915_gem_reset+0x179/0x300 [i915] Reported-by: Mika Kuoppala <[email protected]> Fixes: 49ef5294cda2 ("drm/i915: Move fence tracking from object to vma") Signed-off-by: Chris Wilson <[email protected]> Cc: Mika Kuoppala <[email protected]> Cc: Joonas Lahtinen <[email protected]> Reviewed-by: Mika Kuoppala <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm/i915: Reattach comment, complete type specificationDave Gordon1-2/+3
In the recent patch bc3d674 drm/i915: Allow userspace to request no-error-capture upon ... the final version moved the flags and the associated #defines around so they were adjacent; unfortunately, they ended up between a comment and the thing (hw_id) to which the comment applies :( So this patch reshuffles the comment and subject back together. Also, as we're touching 'hw_id', let's change it from just 'unsigned' to a fully-specified 'unsigned int', because some code checking tools (including checkpatch) object to plain 'unsigned'. Fixes: bc3d674462e5 ("drm/i915: Allow userspace to request no-error-capture...") Signed-off-by: Dave Gordon <[email protected]> Cc: Chris Wilson <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Chris Wilson <[email protected]> Signed-off-by: Chris Wilson <[email protected]>
2016-08-19drm: Don't implement empty prepare_fb()/cleanup_fb()Laurent Pinchart5-51/+0
The plane .prepare_fb() and .cleanup_fb() helpers are optional, there's no need to implement empty stubs, and no need to explicitly set the function pointers to NULL either. Signed-off-by: Laurent Pinchart <[email protected]> [danvet: Resolved conflicts with Chris' patch.] Signed-off-by: Daniel Vetter <[email protected]>
2016-08-19virtio-gpu: Use memdup_user() rather than duplicating its implementationMarkus Elfring1-9/+4
Reuse existing functionality from memdup_user() instead of keeping duplicate source code. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19GPU-DRM-Savage: Use memdup_user() rather than duplicatingMarkus Elfring1-9/+3
Reuse existing functionality from memdup_user() instead of keeping duplicate source code. This issue was detected by using the Coccinelle software. Signed-off-by: Markus Elfring <[email protected]> Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-19drm: Allow drivers to modify plane_state in prepare_fb/cleanup_fbChris Wilson11-22/+22
The drivers have to modify the atomic plane state during the prepare_fb callback so they track allocations, reservations and dependencies for this atomic operation involving this fb. In particular, how else do we set the plane->fence from the framebuffer! Signed-off-by: Chris Wilson <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: [email protected] Signed-off-by: Daniel Vetter <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Accelerate copies from WC memoryChris Wilson1-27/+43
If we need to use clflush to prepare our batch for reads from memory, we can bypass the cache instead by using non-temporal copies. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Use binary search for faster register lookupChris Wilson1-22/+20
A significant proportion of the cmdparsing time for some batches is the cost to find the register in the mmiotable. We ensure that those tables are in ascending order such that we could do a binary search if it was ever merited. It is. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Check for SKIP descriptors firstChris Wilson1-0/+3
If the command descriptor says to skip it, ignore checking for anyother other conflict. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Compare against the previous command descriptorChris Wilson1-7/+14
On the blitter (and in test code), we see long sequences of repeated commands, e.g. XY_PIXEL_BLT, XY_SCANLINE_BLT, or XY_SRC_COPY. For these, we can skip the hashtable lookup by remembering the previous command descriptor and doing a straightforward compare of the command header. The corollary is that we need to do one extra comparison before lookup up new commands. v2: Less magic mask (ok, it is still magic, but now you cannot see!) Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Improve hash functionChris Wilson1-20/+31
The existing code's hashfunction is very suboptimal (most 3D commands use the same bucket degrading the hash to a long list). The code even acknowledge that the issue was known and the fix simple: /* * If we attempt to generate a perfect hash, we should be able to look at bits * 31:29 of a command from a batch buffer and use the full mask for that * client. The existing INSTR_CLIENT_MASK/SHIFT defines can be used for this. */ Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Only cache the dst vmapChris Wilson1-14/+19
For simplicity, we want to continue using a contiguous mapping of the command buffer, but we can reduce the number of vmappings we hold by switching over to a page-by-page copy from the user batch buffer to the shadow. The cost for saving one linear mapping is about 5% in trivial workloads - which is more or less the overhead in calling kmap_atomic(). Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Use cached vmappingsChris Wilson2-81/+54
The single largest factor in the overhead of parsing the commands is the setup of the virtual mapping to provide a continuous block for the batch buffer. If we keep those vmappings around (against the better judgement of mm/vmalloc.c, which we offset by handwaving and looking suggestively at the shrinker) we can dramatically improve the performance of the parser for small batches (such as media workloads). Furthermore, we can use the prepare shmem read/write functions to determine how best we need to clflush the range (rather than every page of the object). The impact of caching both src/dst vmaps is +80% on ivb and +140% on byt for the throughput on small batches. (Caching just the dst vmap and iterating over the src, doing a page by page copy is roughly 5% slower on both platforms. That may be an acceptable trade-off to eliminate one cached vmapping, and we may be able to reduce the per-page copying overhead further.) For *this* simple test case, the cmdparser is now within a factor of 2 of ideal performance. Signed-off-by: Chris Wilson <[email protected]> Cc: Matthew Auld <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Add the TIMESTAMP register for the other enginesChris Wilson1-0/+5
Since I have been using the BCS_TIMESTAMP to measure latency of execution upon the blitter ring, allow regular userspace to also read from that register. They are already allowed RCS_TIMESTAMP! Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/cmdparser: Make initialisation failure non-fatalChris Wilson3-16/+19
If the developer adds a register in the wrong order, we BUG during boot. That makes development and testing very difficult. Let's be a bit more friendly and disable the command parser with a big warning if the tables are invalid. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Matthew Auld <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Stop discarding GTT cache-domain on unbind vmaChris Wilson1-23/+3
Since commit 43566dedde54 ("drm/i915: Broaden application of set-domain(GTT)") we allowed objects to be in the GTT domain, but unbound. Therefore removing the GTT cache domain when removing the GGTT vma is no longer semantically correct. An unfortunate side-effect is we lose the wondrously named i915_gem_object_finish_gtt(), not to be confused with i915_gem_gtt_finish_object()! Signed-off-by: Chris Wilson <[email protected]> Cc: Akash Goel <[email protected]> Cc: Joonas Lahtinen <[email protected]> Cc: Tvrtko Ursulin <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Bump the inactive tracking for all VMA accessedChris Wilson1-6/+23
We track the LRU access for eviction and bump the last access for the user GGTT on set-to-gtt. When we do so we need to not only bump the primary GGTT VMA but all partials as well. Similarly we want to bump the last access tracking for when unpinning an object from the scanout so that they do not get promptly evicted and hopefully remain available for reuse on the next frame. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Track display alignment on VMAChris Wilson2-13/+9
When using the aliasing ppgtt and pageflipping with the shrinker/eviction active, we note that we often have to rebind the backbuffer before flipping onto the scanout because it has an invalid alignment. If we store the worst-case alignment required for a VMA, we can avoid having to rebind at critical junctures. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Fallback to using unmappable memory for scanoutChris Wilson2-4/+14
The existing ABI says that scanouts are pinned into the mappable region so that legacy clients (e.g. old Xorg or plymouthd) can write directly into the scanout through a GTT mapping. However if the surface does not fit into the mappable region, we are better off just trying to fit it anywhere and hoping for the best. (Any userspace that is capable of using ginormous scanouts is also likely not to rely on pure GTT updates.) With the partial vma fault support, we are no longer restricted to only using scanouts that we can pin (though it is still preferred for performance reasons and for powersaving features like FBC). v2: Skip fence pinning when not mappable. v3: Add a comment to explain the possible ramifications of not being able to use fences for unmappable scanouts. v4: Rebase to skip over some local patches v5: Rebase to defer until after we have unmappable GTT fault support Signed-off-by: Chris Wilson <[email protected]> Cc: Deepak S <[email protected]> Cc: Damien Lespiau <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: Jani Nikula <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Choose not to evict faultable objects from the GGTTChris Wilson3-4/+16
Often times we do not want to evict mapped objects from the GGTT as these are quite expensive to teardown and frequently reused (causing an equally, if not more so, expensive setup). In particular, when faulting in a new object we want to avoid evicting an active object, or else we may trigger a page-fault-of-doom as we ping-pong between evicting two objects. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Drop ORIGIN_GTT for untracked GTT writesChris Wilson2-4/+10
If FBC is set on a framebuffer that is unmapped, all GTT faults will be from a partial mapping. Writes by the user through the partial VMA are then untracked by the FBC and so we must use the ORIGIN_CPU when flushing the I915_GEM_DOMAIN_GTT. v2: Keep ORIGIN_CPU for set-to-domain(.write=CPU) Signed-off-by: Chris Wilson <[email protected]> Cc: Joonas Lahtinen <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: "Zanoni, Paulo R" <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Convert partial ggtt vma to full ggtt if it spans the entire objectChris Wilson1-0/+6
If we want to create a partial vma from a chunk that is the same size as the object, create a normal ggtt vma instead. The benefit is that it will match future requests for the normal ggtt. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Fix partial GGTT faultingChris Wilson1-31/+30
We want to always use the partial VMA as a fallback for a failure to bind the object into the GGTT. This extends the support partial objects in the GGTT to cover everything, not just objects too large. v2: Call the partial view, view not partial. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Choose partial chunksize based on tile row sizeChris Wilson1-1/+16
In order to support setting up fences for partial mappings of an object, we have to align those mappings with the fence. The minimum chunksize we choose is at least the size of a single tile row. v2: Make minimum chunk size a define for later use Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Move fence tracking from object to vmaChris Wilson12-388/+330
In order to handle tiled partial GTT mmappings, we need to associate the fence with an individual vma. v2: A couple of silly drops replaced spotted by Joonas Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Rename fence.lru_list to linkChris Wilson4-8/+7
Our current practice is to only name the actual list (here dev_priv->fence_list) using "list", and elements upon that list are referred to as "link". Further, the lru nature is of the list and not of the node and including in the name does not disambiguate the link from anything else. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915/userptr: Make gup errors stickierChris Wilson1-10/+7
Keep any error reported by the gup_worker until we are notified that the arena has changed (via the mmu-notifier). This has the importance of making two consecutive calls to i915_gem_object_get_pages() reporting the same error, and curtailing a loop of detecting a fault and requeueing a gup_worker. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Mika Kuoppala <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Allocate rings from stolenChris Wilson1-4/+2
If we have stolen available, make use of it for ringbuffer allocation. Previously this was restricted to !llc platforms, as writing to stolen requires a GGTT mapping - but now that we have partial mappable support, the mappable aperture isn't quite so precious so we can use it more freely and ringbuffers are a good user for the otherwise wasted stolen. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
2016-08-18drm/i915: Allow ringbuffers to be bound anywhereChris Wilson2-8/+8
Now that we have WC vmapping available, we can bind our rings anywhere in the GGTT and do not need to restrict them to the mappable region. Except for stolen objects, for which direct access is verbatim and we must use the mappable aperture. Signed-off-by: Chris Wilson <[email protected]> Reviewed-by: Joonas Lahtinen <[email protected]> Link: http://patchwork.freedesktop.org/patch/msgid/[email protected]
Powered by cgit, Gruvboxed by Blaster4385.