summaryrefslogtreecommitdiff
path: root/libdrm/intel/intel_bufmgr_gem.c
AgeCommit message (Collapse)Author
2009-10-08intel: Remove the asserts about the ignored alignment parameter.Eric Anholt
I slipped it in with the alloc_tiled changes, since we were explicitly throwing the parameter away. It caught some bogus released code, which we've now fixed, so remove the asserts to keep old drivers working.
2009-10-06intel: Add a bo_alloc function for tiled BOs.Jesse Barnes
This simplifies driver code in handling object allocation, and also gives us an opportunity to possibly cache tiled buffers if it turns out to be a win. [anholt: This is chopped out of the execbuf2 patch, as it seems to be useful separately and cleans up the execbuf2 changes to be more obvious]
2009-10-06intel: Reformat to the kernel coding style. Welcome to the 8-space future.Eric Anholt
This is done with: Lindent *.[ch] perl -pi -e 's|drm_intel_bo \* |drm_intel_bo *|g' *.[ch] perl -pi -e 's|drm_intel_bufmgr \* |drm_intel_bufmgr *|g' *.[ch] perl -pi -e 's|drm_intel_bo_gem \* |drm_intel_bo_gem *|g' *.[ch] perl -pi -e 's|drm_intel_bufmgr_gem \* |drm_intel_bufmgr_gem *|g' *.[ch] perl -pi -e 's|_fake \* |_fake *|g' *.[ch] hand-editing to whack indented comments into line and other touchups.
2009-10-06intel: Don't allocate more relocation entries than the BO could support.Eric Anholt
This saves 32k of relocation entry storage for many 965 state buffers. No noticeable impact on performance for cairo-gl firefox.
2009-10-03intel: report errnoChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2009-10-02intel: Use atomic refcountersChris Wilson
As the target architecture for Intel GPUs is the x86, we can presume to have reasonable compiler support for Intel atomic intrinsics, i.e. gcc, and so use those in preference to pulling in a complicated mess of fragile assembly. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> [anholt: hand-resolved against my previous commit. This brings cairo-gl firefox-talos-gfx time from 65 seconds back down to 62 seconds.] Signed-off-by: Eric Anholt <eric@anholt.net>
2009-10-02intel: Mark cached bo as purgeableChris Wilson
Set the DONTNEED flag on cached buffers so that the kernel is free to discard those when under memory pressure. [anholt: This takes firefox-talos-gfx time from ~62 seconds to ~65 seconds on my GM965, but it seems like a hit worth taking for the improved functionality from saving memory] Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-10-02intel: Don't free the reloc list when putting a freed BO in the cache.Eric Anholt
This takes firefox-talos-gfx from 74 seconds to 70 seconds on my GM965.
2009-10-01intel: Add a new function to check if a BO's reloc tree references some BO.Eric Anholt
There are a bunch of places in GL where if we can't do this we have to flush the batchbuffer, and the cost of lookups here is outweighed by flush savings.
2009-09-21intel: Remove the max_entries stuff that complicated bo reuse.Eric Anholt
I thought I was going to do all sorts of crazy experiments with it. I never did, and it turned out the free-after-a-few-seconds plan is working out fine.
2009-09-21intel: Remove the old swrast flag for reducing cache flushing.Eric Anholt
It hasn't been doing anything effective since 52e5d24fae4af6f2f4a5304a516c8c5ab347a11b, and we pretty much don't bo_map pinned buffers any more anyway.
2009-08-28Add drm_intel_bo_busy to query whether mapping a BO would block.Eric Anholt
2009-07-09intel: Free buffers in the BO cache that haven't been reused in a while.Eric Anholt
The goal of the BO cache is to keep buffers on hand for fast continuous use, as in every frame of a game or every batchbuffer of the X Server. Keeping older buffers on hand not only doesn't serve this purpose, it may hurt performance by resulting in disk cache getting kicked out, or even driving the system to swap. Bug #20766.
2009-07-06intel: Fix up math errors when allocating very large BOs.Eric Anholt
The logbase2 would overflow and wrap the size around to 0, making the code allocate a 4kb object instead. By simplifying the code to just walk the 14-entry bucket array comparing sizes instead of indexing on ffs(1 << logbase2(size)), we avoid silly math errors and have code of approximately the same speed. Many thanks to Simon Farnsworth for debugging and providing a working patch. Bug #27365.
2009-07-06intel: fix compiler warning about returning a void value.Edward O'Callaghan
bug #21999
2009-05-19intel: quiet a warning.Eric Anholt
2009-05-19intel: Only do BO caching up to 64MB objects.Eric Anholt
This avoids making objects significantly bigger than they would be otherwise, which would result in some failing at binding to the GTT. Found from firefox hanging on: http://upload.wikimedia.org/wikipedia/commons/b/b7/Singapore_port_panorama.jpg due to a software fallback trying to do a GTT-mapped copy between two 73MB BOs that were instead each 128MB, and failing because both couldn't fit simultaneously. The cost here is that we get no opportunity to cache these objects and avoid the mapping. But since the objects are a significant percentage of the aperture size, each mapped access is likely having to fault and rebind the object most of the time anyway. Bug #20152 (2/3)
2009-05-14libdrm/intel: Make get_pipe_from_crtc_id per-bufmgr. Return -1 on failure.Keith Packard
The convention is that all APIs are per-bufmgr, so make this one the same. Then, have it return -1 on failure so that the application can know what's going on and do something sensible. Signed-off-by: Keith Packard <keithp@keithp.com>
2009-05-14Add new drm_intel_get_pipe_from_crtc_id function.Carl Worth
This wraps the new DRM_IOCTL_I915_GET_PIPE_FROM_CRTC_ID ioctl, allowing applications to discover the pipe number corresponding to a given CRTC ID. This is necessary for doing pipe-specific operations such as waiting for vblank on a given CRTC.
2009-05-12libdrm/intel: add drm_intel_bo_disable_reuse apiKeith Packard
Scanout buffers need to be freed through the kernel as it holds a reference to them; exposing this API allows applications allocating scanout buffers to flag them as not reusable. Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Eric Anholt <eric@anholt.net>
2009-05-01libdrm/intel: assert that clients are using bo refcounting correctlyKeith Packard
Add assertions to drm_intel_gem_bo_reference, drm_intel_gem_bo_reference_locked and drm_intel_gem_bo_unreference_locked that the object has not been freed (refcount > 0). Mistakes in refcounting lead to attempts to insert a bo into a free list more than once which causes application failure as empty free lists are dereferenced as buffer objects. Signed-off-by: Keith Packard <keithp@keithp.com>
2009-03-26libdrm/intel: support GTT maps correctlyJesse Barnes
libdrm has some support for GTT mapping already, but there are bugs with it (no surprise since it hasn't been used much). In fixing 20803, I found that sharing bo_gem->virtual was a bad idea, since a previously mapped object might not end up getting GTT mapped, leading to corruption. So this patch splits the fields according to use, taking care to unmap both at free time (but preserving the map caching). There's still a risk we might run out of mappings (there's a sysctl tunable for max number of mappings per process, defaulted to 64k or so it looks like) but at least GTT maps will work with these changes (and some others for fixing PAT breakage in the kernel). Reviewed-by: Eric Anholt <eric@anholt.net> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2009-02-27intel: Update reloc_tree_size of the first buffer when we count the tree size.Eric Anholt
This helps avoid the n^2 performance cost of counting tree size when we get a lot of relocations into our batch buffer. rgb10text on keithp's laptop went from 136k glyphs/sec to 234k glyphs/sec.
2009-02-21intel: Add a new bufmgr alloc function to get BOs ready for rendering to.Eric Anholt
This avoids using the oldest BO in the BO cache and waiting for it to be idle before we turn around and render to it with the GPU. Thanks to Chris Wilson for pointing out how silly we were being.
2009-02-15intel: Clear ioctl parameter outvalues and pads that confuse valgrind.Eric Anholt
The minor CPU cost here is probably outweighed by bothering us with noise in the tool.
2009-01-27intel: don't count fences on 965 and later, as they don't use them.Eric Anholt
2009-01-27intel: Fix under-counting of fences registers required in check_aperture.Eric Anholt
2009-01-27intel: libdrm support for fence management in execbufJesse Barnes
This patch tries to use the available fence count to figure out whether a given batch can succeed or not (just like the aperture check). Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Eric Anholt <eric@anholt.net>
2009-01-14intel: Retry pin ioctl on -EINTR.Eric Anholt
2008-12-17intel: Cache tiling/swizzle state in user mode. Reset tiling on reuse.Keith Packard
Remember tiling mode values provided by appplications, and record tiling mode when creating a buffer from another application. This eliminates any need to ask the kernel for tiling values and also makes reused buffers get the right tiling. Signed-off-by: Keith Packard <keithp@keithp.com>
2008-12-17intel: Debug output %u vs uint64_t warning fixKeith Packard
Signed-off-by: Keith Packard <keithp@keithp.com>
2008-12-17intel: return error status from drm_intel_gem_bo_mapKeith Packard
Applications may actually care if the mapping operation failed, so when it happens, return an error indication. errno is probably trashed by fprintf though. Signed-off-by: Keith Packard <keithp@keithp.com>
2008-12-17intel: Dump out memory usage information when the kernel fails to pinKeith Packard
The execbuffer ioctl returns ENOMEM when it fails to pin all of the buffers in the GTT. This is usually caused by the DRM client attempting to use too much memory in a single request. Dumping out the requested and available memory values should help point out failures in the DRM code to catch over commitments of this form. Signed-off-by: Keith Packard <keithp@keithp.com>
2008-12-17libdrm: add mode setting filesJesse Barnes
Add mode setting files to libdrm, including xf86drmMode.* and the new drm_mode.h header. Also add a couple of tests to sanity check the kernel interfaces and update code to support them.
2008-12-14intel: don't skip set_domain on mapping of shared buffers.Eric Anholt
2008-12-14intel: don't let named buffers into the BO cache.Eric Anholt
We wouldn't want some remaining 3D rendering to scribble on our batchbuffer.
2008-12-14intel: Remove the mapped flag, which is adequately covered by bo_gem->virtual.Eric Anholt
2008-12-10Revert "Merge branch 'modesetting-gem'"Jesse Barnes
This reverts commit 6656db10551bbb8770dd945b6d81d5138521f208. We really just want the libdrm and ioctl bits, not all the driver stuff.
2008-12-03Merge branch 'master' into modesetting-gemJesse Barnes
2008-12-02intel: Add a function for setting (GTT,GTT) domain, for use by UXA.Eric Anholt
This function can also serve the role that the bo_wait_rendering did, when write_enable is unset.
2008-11-19libdrm_intel: fix merge errorJesse Barnes
don't take the lock twice
2008-11-13Merge branch 'master' into modesetting-gemJesse Barnes
Conflicts: libdrm/Makefile.am libdrm/intel/intel_bufmgr.h libdrm/intel/intel_bufmgr_fake.c libdrm/intel/intel_bufmgr_gem.c shared-core/drm.h shared-core/i915_dma.c shared-core/i915_irq.c shared-core/radeon_cp.c shared-core/radeon_drv.h
2008-11-13libdrm_intel: fix warnings on 64 bitJesse Barnes
Cast a couple of %llx args to unsigned long long.
2008-11-13libdrm: add support for i915 GTT mapping ioctlJesse Barnes
Add a drm_intel_gem_bo_map_gtt() function for mapping a buffer object through the aperture rather than directly to its CPU cacheable memory.
2008-11-07intel: Restart on interrupt of bo_wait_rendering instead of complaining.Eric Anholt
2008-10-30intel: Rename dri_ and intel_ symbols to drm_intel_.Eric Anholt
I wanted to avoid doing this, as it's a bunch of churn, but there was a conflict between the dri_ symbols in libdrm and the symbols that were in Mesa in 7.2, which broke Mesa 7.2 AIGLX when the 2D driver had loaded new libdrm symbols. The new naming was recommended by cworth for giving the code a unique prefix identifying where the code lives. Additionally, take the opportunity to fix up two API mistakes: emit_reloc's arguments were in a nonsensical order, and set_tiling lacked the stride argument that the kernel will want to use soon. API compatibility with released code is maintained using #defines.
2008-10-28intel: Add dri_bufmgr_check_aperture support for bufmgr_gem.Eric Anholt
This relies on a new kernel ioctl to get the available aperture size. In order to provide reasonable performance from dri_bufmgr_check_aperture, we now require that once a buffer has been used as the target of a relocation, it gets no further relocations added to it. This cuts the cost of check_aperture from 10% to 1% in the 3D driver with no code changes, but slightly complicates our plans for the 2D driver.
2008-10-16intel: ioctl is not defined to return -errnoKeith Packard
Don't count on ioctl returning -errno; use errno directly. Signed-off-by: Keith Packard <keithp@keithp.com> Signed-off-by: Eric Anholt <eric@anholt.net>
2008-10-17intel: Protect bufmgr objects with a pthread mutex.Eric Anholt
We want to be able to use the bufmgr from multiple threads for GL, and thus we need to protect the internal structures. The pthread-stubs package is used so that programs not linked against pthreads get weak symbols to stubs and don't eat most of the cost.
2008-10-14intel: Add interface for getting tiling mode of a bo.Eric Anholt