Since the child process doesn't call exec(), exit() attempted to run
atexit handlers registered by the parent process. This could result in
the child process hanging in exit() if there were still disk cache
threads alive when the parent process called fork(). (The CI runners
hit this multiple times when running tests in strace)
Fixes: 6a246f5c6d "aco/tests: Fix deadlock for too large test lists"
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Reviewed-by: Adam Jackson <ajax@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9472>
This prints the program with each instruction's contribution to it's
latency and various factors for the calculation of the Inverse Throughput
statistic.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8994>
Latency is estimanted duration of a single wave, ignoring others in the
CU. It is similar to the old cycles statistic except it it's more accurate
and considers memory operations.
The InvThroughput statistic is a combination of MaxWaves, Latency and the
portion of the wave's execution which does not use various resources.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8994>
This allows them to be scheduled properly and simplifies the assembler a
little.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8994>
Keep track of the current loop depth in Program and set the depth inside
Program::insert_block() instead of repeating it every time we insert one.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8994>
Obviously this didn't affect correctness. Not sure about performance.
It also changes enabled_channels to match radeonsi.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: f29c81f863 ("aco: use VOP2 for v_cvt_pkrtz_f16_f32 if possible")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9459>
The container is moved from before and hence returns size 0. To get the
correct value, the new instruction container must be used instead.
This was flagged by clang-tidy. The fixed call still triggers the
corresponding diagnostic, hence this change silences it by adding a
redundant clear() after move.
Fixes: 7f1b537304 ("aco: add new NOP insertion pass for GFX6-9")
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9432>
A Hitman 2 shader does: read64(local_invocation_index() * 4 - 4). This was
likely emitting a ds_read2_b32 on GFX6. For local_invocation_index()=0,
because the first dword was out-of-bounds, the second was likely also
considered out-of-bounds (even though it's not, at offset 0).
Likely fixes https://gitlab.freedesktop.org/mesa/mesa/-/issues/3882
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 57e6886f98 ("aco: refactor load_lds to use new helpers")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9332>
This reverts commit 1a0b0e8460.
The bounds checking behaviour of ds_read_b64, ds_read_b96 and ds_read_b128
make this feature very difficult to use safely.
This fixes a blocking artifact in Hitman 2. Previously, it contained:
ds_read_b64(local_invocation_index() * 4 - 4)
For local_invocation_index()=0, the second dword would be considered
out-of-bounds, even though it's at offset 0.
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9332>
In release builds, there should be no change, but in debug builds the
assert will help us catch undefined behavior resulting from using
util_cpu_caps before it is initialized.
With fix for u_half_test for MSVC from Jesse Natalie squashed in.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9266>
Since value-numbering no longer works across loops, we no longer need to
use v_readfirstlane_b32.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9288>
This becomes possible as long as we do
val = s_and_b32/64 exec, val
before any subgroup operations.
This precautional instruction can be removed by the
optimizer if 'val' was computed by a VOPC instruction
using the same exec mask.
Totals from 59 (0.04% of 146267) affected shaders (Navi10):
VGPRs: 2808 -> 2816 (+0.28%)
CodeSize: 340888 -> 340852 (-0.01%); split: -0.20%, +0.19%
Instrs: 61733 -> 61625 (-0.17%); split: -0.18%, +0.01%
Cycles: 470636 -> 469112 (-0.32%); split: -0.33%, +0.01%
VMEM: 8091 -> 7993 (-1.21%)
SMEM: 2736 -> 2719 (-0.62%); split: +0.29%, -0.91%
VClause: 1745 -> 1741 (-0.23%)
SClause: 2394 -> 2392 (-0.08%); split: -0.25%, +0.17%
Copies: 3249 -> 3253 (+0.12%); split: -0.62%, +0.74%
Branches: 1210 -> 1206 (-0.33%)
PreSGPRs: 3126 -> 3176 (+1.60%); split: -0.16%, +1.76%
Reviewed-by: Tony Wasserka <tony.wasserka@gmx.de>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9195>
Found in a Death Stranding shader with loop unrolling disabled.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Fixes: 9a089baff1 ("aco: optimize boolean phis with uniform selections")
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9193>
Fix defect reported by Coverity Scan.
Uninitialized scalar variable (UNINIT)
uninit_use: Using uninitialized value ds_state.front. Field ds_state.front.writeMask is uninitialized.
Fixes: d488d0fd7b ("aco: add framework for testing isel and integration tests")
Signed-off-by: Vinson Lee <vlee@freedesktop.org>
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9033>
The calculate_tess_lds_size function already returns the size in blocks
of the encoding granule, but we forgot to adjust config->lds_size.
This variable is not used to actually set the LDS size used for TCS,
but by ACO to make scheduling decisions.
Fossil DB stats on Sienna Cichlid:
Please note that the +3729.43% is NOT a regression.
The real LDS size used didn't change, it was just reported incorrectly.
Totals from 1342 (0.96% of 139391) affected shaders:
VGPRs: 60880 -> 80240 (+31.80%); split: -0.05%, +31.85%
CodeSize: 3378456 -> 3381224 (+0.08%); split: -0.23%, +0.31%
LDS: 687104 -> 26312192 (+3729.43%)
MaxWaves: 29794 -> 23962 (-19.57%)
Instrs: 644194 -> 644610 (+0.06%); split: -0.32%, +0.39%
Cycles: 2675068 -> 2676804 (+0.06%); split: -0.31%, +0.38%
VMEM: 428840 -> 517418 (+20.66%); split: +22.53%, -1.88%
SMEM: 91831 -> 88587 (-3.53%); split: +5.70%, -9.23%
VClause: 22740 -> 19384 (-14.76%); split: -16.18%, +1.42%
SClause: 19116 -> 18373 (-3.89%); split: -4.34%, +0.46%
Copies: 66662 -> 63448 (-4.82%); split: -5.55%, +0.73%
Fixes: cf89bdb9ba "radv: align the LDS size in calculate_tess_lds_size()"
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/9098>
Instead of assuming WGP mode on GFX10+ in different places, add a member
to Program that can be used instead.
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8761>
We used that to catch assembly errors in the past,
but now, there are too many hardware features we
use in ACO that are not supported by the LLVM disassembler,
that it is not really suited anymore as a debugging tool.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921>
This patch delays the calculation of GPR limits in order to
precisely incorporate extra registers (VCC etc.) and shared VGPRs.
Additionally, the allocation granularity is used to set the config.
This has some effect on the reported SGPR stats.
Totals (Navi10):
SGPRs: 6971787 -> 17753642 (+154.65%)
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921>
This also switches the alloc_granule of Tonga and Iceland
to 96, so that the calculation is consistent.
Also changes the granularity for RDNA to 16 to keep
better stats with the upcoming patch.
Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/8921>
