summaryrefslogtreecommitdiff
path: root/tests/gem_fence_thrash.c
AgeCommit message (Collapse)Author
2018-03-01lib: Provide an accelerated routine for readback from WCChris Wilson
Reading from WC is awfully slow as each access is uncached and so performed synchronously, stalling for the memory load. x86 did introduce some new instructions in SSE 4.1 to provide a small internal buffer to accelerate reading back a cacheline at a time from uncached memory, for this purpose. v2: Don't be lazy and handle misalignment. v3: Switch out of sse41 before emitting the generic memcpy routine v4: Replace opencoded memcpy_from_wc v5: Always flush the internal buffer before use (Eric) v6: Assume bulk moves, so check for dst alignment. v7: Use _mm_fence for _buitlin_ia32_mfence for consistency, remove superfluous defines (Ville) Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Eric Anholt <eric@anholt.net> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
2017-10-09igt/gem_fence_thrash: Use streaming reads for verifyChris Wilson
At the moment, the verify tests use an extremely brutal write-read of every dword, degrading performance to UC. If we break those up into cachelines, we can do a wcb write/read at a time instead, roughly 8x faster. We lose the accuracy of the forced wcb flushes around every dword, but we are retaining the overall behaviour of checking reads following writes instead. To compensate, we do check that a single dword write/read before using wcb aligned accesses. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2017-09-08build: Nuke #ifdef HAVE_CONFIG_H cargo-cultDaniel Vetter
We have it. Daniel Stone said the #ifdef HAVE_CONFIG_H comes from the X11 transition to the modular build, where in the imake -> modular build transition config.h wasn't universally available. Now we just make this a requirement (so yeah Android better generate one too). v2: Improve commit message a bit. Reviewed-by: Eric Anholt <eric@anholt.net> Acked-by: Jani Nikula <jani.nikula@intel.com> Acked-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Acked-by: Petri Latvala <petri.latvala@intel.com> Acked-by: Daniel Stone <daniels@collabora.com> Acked-by: Radoslaw Szwichtenberg <radoslaw.szwichtenberg@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2015-10-12Replace __gem_mmap__{cpu,gtt,wc}() + igt_assert() with gem_mmap__{cpu,gtt,wc}()Ville Syrjälä
gem_mmap__{cpu,gtt,wc}() already has the assert built in, so replace __gem_mmap__{cpu,gtt,wc}() + igt_assert() with it. Mostly done with coccinelle, with some manual help: @@ identifier I; expression E1, E2, E3, E4, E5, E6; @@ ( - I = __gem_mmap__gtt(E1, E2, E3, E4); + I = gem_mmap__gtt(E1, E2, E3, E4); ... - igt_assert(I); | - I = __gem_mmap__cpu(E1, E2, E3, E4, E5); + I = gem_mmap__cpu(E1, E2, E3, E4, E5); ... - igt_assert(I); | - I = __gem_mmap__wc(E1, E2, E3, E4, E5); + I = gem_mmap__wc(E1, E2, E3, E4, E5); ... - igt_assert(I); ) Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Stochastically-reviwewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2015-10-09Make gem_mmap__{cpu,gtt,wc}() assert on failureVille Syrjälä
Rename the current gem_mmap__{cpu,gtt,wc}() functions into __gem_mmap__{cpu,gtt,wc}(), and add back wrappers with the original name that assert that the pointer is valid. Most callers will expect a valid pointer and shouldn't have to bother with failures. To avoid changing anything (yet), sed 's/gem_mmap__/__gem_mmap__/g' over the entire codebase. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Stochastically-reviwewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2015-10-09s/gem_mmap/gem_mmap__gtt/Ville Syrjälä
Get rid of the gem_mmap() alias of gem_mmap__gtt(). I don't see any point in having it. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Stochastically-reviwewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2015-09-11convert drm_open_any*() calls to drm_open_driver*(DRIVER_INTEL) calls with cocciMicah Fedke
Apply the new API to all call sites within the test suite using the following semantic patch: // Semantic patch for replacing drm_open_any* with arch-specific drm_open_driver* calls @@ identifier i =~ "\bdrm_open_any\b"; @@ - i() + drm_open_driver(DRIVER_INTEL) @@ identifier i =~ "\bdrm_open_any_master\b"; @@ - i() + drm_open_driver_master(DRIVER_INTEL) @@ identifier i =~ "\bdrm_open_any_render\b"; @@ - i() + drm_open_driver_render(DRIVER_INTEL) @@ identifier i =~ "\b__drm_open_any\b"; @@ - i() + __drm_open_driver(DRIVER_INTEL) Signed-off-by: Micah Fedke <micah.fedke@collabora.co.uk> Signed-off-by: Thomas Wood <thomas.wood@intel.com>
2015-08-21lib: add a single include headerThomas Wood
Add a header that includes all the headers for the library. This allows reorganisation of the library without affecting programs using it and also simplifies the headers that need to be included to use the library. Signed-off-by: Thomas Wood <thomas.wood@intel.com>
2015-06-23igt/gem_fence_thrash: Reduce memory usageDerek Morton
On android platforms with 1Gb RAM gem_fence_thrash was failing with an out of memory error. This patch causes gem_close() to be called when a handle is no longer required rather than relying on the cleanup when the fd is closed. This greatly improves the memory footprint of the test allowing it to run on 1Mb systems. Also fixed a leak of the 'threads' variable. v2: Simplified as per Chris Wilson's suggestion. Signed-off-by: Derek Morton <derek.j.morton@intel.com> [ickle: fix mmap leak from bo_copy()] Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2015-03-06igt.cocci: Replace igt_assert() with igt_assert_CMP() where possibleMatt Roper
The integer comparison macros give us better error output by including the actual values that failed the comparison. Signed-off-by: Matt Roper <matthew.d.roper@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
2014-11-04ioctl_wrappers: Pass in offset to CPU mmapsChris Wilson
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2014-05-14tests: sprinkle igt loggingDaniel Vetter
All the cases that simply dump some debug information and couldn't be converted to some of the fancier macros. Some information output removed when it's redundant with the subtest status. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-22lib: unnecessary header removal for drmtest.h, part 1Daniel Vetter
Brought a few missing headers to light in ioctl_wrappers.h, too. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-22tests|lib: remove assert.h includesDaniel Vetter
Only the igt core and non-test tools should have asserts to catch internal errors, tests and helper libraries should all user igt_asert instead. Fix things up where assert instead of igt_assert was used. One tiny step towards header sanity. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-11-01lib: add igt_main macroDaniel Vetter
In the past new testcases with subtest often forgot to add the call to igt_exit at the end of their main() function. That is now caught with a bit more obnoxious asserts, but it's still a nuissance. This little igt_main macro takes care of that (and also of calling the subtest machinery initialization code correctly). If no one objects I'll roll this out for all the simple cases (i.e. those tests that don't have additional argv parsing on top of the subtest machinery). v2: Roll it out across the board. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-09-25tests: use igt_assert/igt_require moreDaniel Vetter
With the new _f variants we can replace almost all of them. Also remove a ton of checks for argc != 1, they're a bit useless ... Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-09-03lib/drmtest: include sys/mman.h from drmtest.hDaniel Vetter
We need it for mmapping to get at PROT_READ|WRITE anyway. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-14tests: use igt_exit() consistently with subtestsDaniel Vetter
This is mostly important to get the SKIP reporting right, but I've found a few stragglers that wanted to get converted over to the igt result reporting completely. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-13tests: use igt_fail instead of exit(param != 0)Daniel Vetter
Mostly a sed job with too manual fixups: - one case of using _exit instead of exit - and one case which under some conditions use 77, so convert that check to an igt_skip. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-13tests: s/assert/igt_assertDaniel Vetter
Just a wholesale rollout for now, we can refine later on. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-12s/drmtest_/igt_/Daniel Vetter
Requested-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Damien Lespiau <damien.lespiau@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-12s/drmtest_subtest_block/drmtest_subtest/Daniel Vetter
The _block postfix meant to convey that a C statement/block must follow can be misread as the verb to block. So drop it. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-08-12lib/drmtest: Add drmtest_subtest_block macroDaniel Vetter
Doesn't do more than an if (drmtest_run_test(name)) right now, but as soon as we get a bit of infrastructure to handle test failures and skipping, this will get more interesting. Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-07-18tests: Black list tests we don't want to run on simulationDamien Lespiau
Let's start by a small set of tests, to eventually consider running more. The current list should then be: gem_mmap gem_pread_after_blit gem_ring_sync_loop gem_ctx_basic gem_pipe_control_store_loop gem_storedw_loop_render gem_storedw_loop_blt gem_storedw_loop_bsd gem_render_linear_blits gem_tiled_blits gem_cpu_reloc gem_exec_nop gem_mmap_gtt v2 add (Daniel Vetter) gem_exec_bad_domains gem_exec_faulting_reloc gem_flink gem_reg_read gem_reloc_overflow gem_tiling_max_stride prime_* Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
2013-04-09tests: Use gem_available_fences()Ville Syrjälä
lib/drmtest.c provides gem_available_fences(). Use it where appropriate. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-04-08tests/gem_fence_trash: make threaded tests more through-roughDaniel Vetter
With this at least the y-tiled test reliably fails on my machines, but x-tiled still passes on some. More ideas to tune this highly welcome. v2: Fill cpu caches with data for each newly allocated bo. This seems to do the trick on my snb here _really_ reliably. So apparently the backsnoop for llc gtt writes is the crucial ingredient here to make the test fail. While at it, also stop leaking mmap space. v3: Fixup commit message. Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-03-27gem_fence_thrash: Fix array allocation size for LP64 systemsChris Wilson
2013-03-27tests: add write-verify test to gem_fence_thrashMika Kuoppala
Add write-verify test to gem_fence_thrash. Test will create multiple threads per fence then verify the write into fenced region. v2: non-threaded, non-tiled tests added. suggested by Chris Wilson. Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2013-03-04build: Guard the inclusions of config.h with HAVE_CONFIG_HDamien Lespiau
autoconf can be configured to not generate a config.h but to give the defines with command line arguments instead. In this case, there's no config.h to include. To work in both cases autoconf adds a HAVE_CONFIG_H define on the command line to signal there's a config.h to include. Signed-off-by: Damien Lespiau <damien.lespiau@intel.com>
2012-03-24tests/gem_fence_thrash: make some good use of drmtest helpersDaniel Vetter
Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-03-14configure: Add AC_SYS_LARGEFILE for mmap64 support on 64-bitChris Wilson
Without enabling largefile support the prototype for mmap64() is broken on Linux/x86_64 with the result being a 32-bit integer sign-extended to fill a 64-bit pointer. Hilarity ensues.
2012-01-10lib/drmtest: extract gem_set_domain and gem_syncDaniel Vetter
gem_sync just does a gtt sync by using set_domain(GTT, GTT). Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-01-08Fix pthread compiler flags to work on Solaris and with Studio compilerAlan Coopersmith
AC_USE_SYSTEM_EXTENSIONS is used to let autoconf take care of setting the right system defintions for POSIX threads & similar extensions to the base POSIX API. Since it will also #define _GNU_SOURCE 1, remove the extra setting to avoid compiler warnings about redefined macros. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-01-08Use sched_yield instead of pthread_yieldAlan Coopersmith
Despite the name implying it's a standard part of the POSIX threads API, pthread_yield is actually non-standard and less portable than sched_yield. For instance, Solaris only has sched_yield, and not pthread_yield. Since even the Linux man page suggests using sched_yield, just call that. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2011-10-15tests/gem_fence_trash: some retuningDaniel Vetter
We actually want to test lru behaviour, so do a bit of work with the fence before yielding to the next thread (we use twice as many fences as there are, so yielding always is pretty bad, no matter how clever our fence stealing). Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2009-09-01Add gem_fence_thrash testChris Wilson
This is a test case that overcommits fence registers between threads, which are copying from one fenced bo to another. In earlier versions of the driver this would cause excessive spinning as the first inactive (i.e. not in use by the GPU) would be used to service the next page. After all the fence registers had been allocated, in effect only the very first fence would then be used for all subsequent faults.
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-11 13:26:35 +0000