Summary:
This property is used to mark an intrinsic that only writes to memory, but
neither reads from memory nor has other side effects.
An example where this is useful is the llvm.amdgcn.buffer.store.format.*
intrinsic, which corresponds to a store instruction that goes through a special
buffer descriptor rather than through a plain pointer.
With this property, the intrinsic should still be handled as having side
effects at the LLVM IR level, but machine scheduling can make smarter
decisions.
Reviewers: tstellarAMD, arsenm, joker.eph, reames
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18291
llvm-svn: 266826
Summary:
The added testcase, which triggered this, was derived from a shader-db case
via bugpoint. A separate question is why scalar branching wasn't used.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19208
llvm-svn: 266825
Summary:
A shader stored the live mask (initial exec mask) in an SGPR which was then
spilled during register allocation. The allocator quite reasonably
optimized turned the spill into
v_writelane_b32 %vgpr, exec_lo, N
v_writelane_b32 %vgpr, exec_hi, N+1
at the beginning of the shader, confusing the SGPR accounting.
No test case, because si-sgpr-spill.ll together with an upcoming patch for
WQM handling exhibits the problem.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19199
llvm-svn: 266824
Also,
- Skip pass if machine module does not have debug info
- Minor comment changes
- Added test
Differential Revision: http://reviews.llvm.org/D19079
llvm-svn: 266626
Order should match the sp3 syntax, where destination (simm16 denoting the hwreg) is coming first.
Differential Revision: http://reviews.llvm.org/D19161
llvm-svn: 266617
Removed some unused headers, replaced some headers with forward class declarations.
Found using simple scripts like this one:
clear && ack --cpp -l '#include "llvm/ADT/IndexedMap.h"' | xargs grep -L 'IndexedMap[<]' | xargs grep -n --color=auto 'IndexedMap'
Patch by Eugene Kosov <claprix@yandex.ru>
Differential Revision: http://reviews.llvm.org/D19219
From: Mehdi Amini <mehdi.amini@apple.com>
llvm-svn: 266595
This resolves more frame indexes early and folds
the immediate offsets into the scratch mubuf instructions.
This cleans up a lot of the mess that's currently emitted,
such as emitting add 0s and repeatedly initializing the same
register to 0 when spilling.
llvm-svn: 266508
Perform store clustering just like load clustering. This change add
StoreClusterMutation in machine-scheduler. To control StoreClusterMutation,
added enableClusterStores() in TargetInstrInfo.h. This is enabled only on
AArch64 for now.
This change also add support for unscaled stores which were not handled in
getMemOpBaseRegImmOfs().
llvm-svn: 266437
Summary:
In the added test-case, the atomic instruction feeds into a non-machine
CopyToReg node which hasn't been selected yet, so guard against
non-machine opcodes here.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19043
llvm-svn: 266433
Currently what comes out of instruction selection is a
register initialized to -1, and then copied to m0.
MachineCSE doesn't consider copies, but we want these
to be CSEed. This isn't much of a problem currently,
because SIFoldOperands is run immediately after.
This avoids regressions when SIFoldOperands is run later
from leaving all copies to m0.
llvm-svn: 266377
Summary:
This adds the necessary target code to be able to run the ir translator.
Lowering function arguments and returns is a nop and there is no support
for RegBankSelect.
Reviewers: arsenm, qcolombet
Subscribers: arsenm, joker.eph, vkalintiris, llvm-commits
Differential Revision: http://reviews.llvm.org/D19077
llvm-svn: 266356
Summary:
This fully solves the problem where the StructurizeCFG pass does not
consider the same branches as uniform as the SIAnnotateControlFlow pass.
The patch in D19013 helps with this problem, but is not sufficient
(and, interestingly, causes a "regression" with one of the existing
test cases).
No tests included here, because tests in D19013 already cover this.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19018
llvm-svn: 266346
Summary:
This pass is unnecessary and overly conservative. It was motivated by
situations like
def %vreg0:SGPR_32
...
if-block:
..
def %vreg1:SGPR_32
...
else-block:
...
use %vreg0:SGPR_32
...
and similar situations with uses after the non-uniform control flow, where
we are not allowed to assign %vreg0 and %vreg1 to the same physical register,
even though in the original, thread/workitem-based CFG, it looks like the
live ranges of these registers do not overlap.
However, by the time register allocation runs, we have moved to a wave-based
CFG that accurately represents the fact that the wave may run through both
the if- and the else-block. So the live ranges of %vreg0 and %vreg1 already
overlap even without the SIFixSGPRLiveRanges pass.
In addition to proving this change correct, I have tested it with Piglit
and a small number of other tests.
Reviewers: arsenm, tstellarAMD
Subscribers: MatzeB, arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19041
llvm-svn: 266345
Summary:
I've been carrying this change around with me for a while, because the if ()
managed to confuse me while following the code. All callers ensure that the
assertion holds.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19042
llvm-svn: 266344
Summary:
For GL_ARB_compute_shader we need to support workgroup sizes of at least 1024. However, if we want to allow large workgroup sizes, we may need to use less registers, as we have to run more waves per SIMD.
This patch adds an attribute to specify the maximum work group size the compiled program needs to support. It defaults, to 256, as that has no wave restrictions.
Reducing the number of registers available is done similarly to how the registers were reserved for chips with the sgpr init bug.
Reviewers: mareko, arsenm, tstellarAMD, nhaehnle
Subscribers: FireBurn, kerberizer, llvm-commits, arsenm
Differential Revision: http://reviews.llvm.org/D18340
Patch By: Bas Nieuwenhuizen
llvm-svn: 266337
Summary:
The code previously always used s1 as it was using the user + system SGPR
information for compute kernels. This is incorrect for Mesa shaders though,
The register should be the next SGPR after all user and system SGPR's.
We use that Mesa adds arguments for all input and system SGPR's and
take the next available SGPR for the scratch wave offset register.
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Reviewers: mareko, arsenm, nhaehnle, tstellarAMD
Subscribers: qcolombet, arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18941
Patch By: Bas Nieuwenhuizen
llvm-svn: 266336
Summary:
When we are spilling SGPRs to scratch memory, we usually don't have
free SGPRs to do the address calculation, so we need to re-use the
ScratchOffset register for the calculation.
Reviewers: arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18917
llvm-svn: 266244
Tests added along with implemented feature.
Note that there is a small leftover of unecessary MI sheduling issue
(more info in the review). CodeGen/AMDGPU/salu-to-valu.ll updated to fix
the false regression.
TODO: Support for TTMP quads, comma-separated syntax in "[]" and more.
Differential Revision: http://reviews.llvm.org/D17825
llvm-svn: 266205
Summary:
It seems like this was broken in r252327. I thought we had test cases
for this, but it's really hard to tirgger spills of this exact register
size since they aren't used very much.
Reviewers: arsenm, nhaehnle
Subscribers: nhaehnle, arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19021
llvm-svn: 266152
Summary:
They correspond to BUFFER_LOAD/STORE_DWORD[_X2,X3,X4] and mostly behave like
llvm.amdgcn.buffer.load/store.format. They will be used by Mesa for SSBO and
atomic counters at least when robust buffer access behavior is desired.
(These instructions perform no format conversion and do buffer range checking
per component.)
As a side effect of sharing patterns with llvm.amdgcn.buffer.store.format,
it has become trivial to add support for the f32 and v2f32 variants of that
intrinsic, so the patch does so.
Also DAG-ify (and fix) some tests that I noticed intermittent failures in
while developing this patch.
Some tests were (temporarily) adjusted for the required mayLoad/hasSideEffects
changes to the BUFFER_STORE_DWORD* instructions. See also
http://reviews.llvm.org/D18291.
Reviewers: arsenm, tstellarAMD, mareko
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18292
llvm-svn: 266126
Summary:
We will be able to handle this case much better once the hazard recognizer
is finished, but this conservative implementation fixes a hang with the piglit
test:
spec/arb_arrays_of_arrays/execution/sampler/fs-nested-struct-arrays-nonconst-nested-arra
Reviewers: arsenm, nhaehnle
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18988
llvm-svn: 266105
This helps clean up some of the mess when expanding unaligned 64-bit
loads when changed to be promote to v2i32, and fixes situations
where or x, 0 was emitted after splitting 64-bit ors during moveToVALU.
I think this could be a generic combine but I'm not sure.
llvm-svn: 266104
Summary:
Under certain circumstances, multi-level breaks (or what is understood by
the control flow passes as such) could be miscompiled in a way that causes
infinite loops, by emitting incorrect control flow intrinsics.
This fixes a hang in
dEQP-GLES3.functional.shaders.loops.while_dynamic_iterations.conditional_continue_vertex
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18967
llvm-svn: 266088
Standard load/store instructions with GLC bit set.
Reviewers: tstellardAMD, arsenm
Differential Revision: http://reviews.llvm.org/D18760
llvm-svn: 265709
Summary: This makes it possible to insert nops at the end of blocks.
Reviewers: arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18549
llvm-svn: 265678
For VGPR_32 operand disassembler expects a VGPR register encoded as 0..255 (enum8 src operand).
readfirstlane/readline actually has enum9 operand and this change fixes VGPR_32 to VS_32 (enum9 encoding).
Differential Revision: http://reviews.llvm.org/D18696
llvm-svn: 265670
This makes it possible to distinguish between mesa shaders
and other kernels even in the presence of compute shaders.
Patch By: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Differential Revision: http://reviews.llvm.org/D18559
llvm-svn: 265589
Summary:
Implement BUFFER_ATOMIC_CMPSWAP{,_X2} instructions on all GCN targets, and FLAT_ATOMIC_CMPSWAP{,_X2} on CI+.
32-bit instruction variants tested manually on Kabini and Bonaire. Tests and parts of code provided by Jan Veselý.
Patch by: Vedran Miletić
Reviewers: arsenm, tstellarAMD, nhaehnle
Subscribers: jvesely, scchan, kanarayan, arsenm
Differential Revision: http://reviews.llvm.org/D17280
llvm-svn: 265170
Summary:
This results in higher register usage, but should make it easier for
the compiler to hide latency.
This pass is a prerequisite for some more scheduler improvements, and I
think the increase register usage with this patch is acceptable, because
when combined with the scheduler improvements, the total register usage
will decrease.
shader-db stats:
2382 shaders in 478 tests
Totals:
SGPRS: 48672 -> 49088 (0.85 %)
VGPRS: 34148 -> 34847 (2.05 %)
Code Size: 1285816 -> 1289128 (0.26 %) bytes
LDS: 28 -> 28 (0.00 %) blocks
Scratch: 492544 -> 573440 (16.42 %) bytes per wave
Max Waves: 6856 -> 6846 (-0.15 %)
Wait states: 0 -> 0 (0.00 %)
Depends on D18451
Reviewers: nhaehnle, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18452
llvm-svn: 264876
Summary:
This helps prevent load clustering from drastically increasing register
pressure by trying to cluster 4 SMRDx8 loads together. The limit of 16
bytes was chosen, because it seems like that was the original intent
of setting the limit to 4 instructions, but more analysis could show
that a different limit is better.
This fixes yields small decreases in register usage with shader-db, but
also helps avoid a large increase in register usage when lane mask
tracking is enabled in the machine scheduler, because lane mask tracking
enables more opportunities for load clustering.
shader-db stats:
2379 shaders in 477 tests
Totals:
SGPRS: 49744 -> 48600 (-2.30 %)
VGPRS: 34120 -> 34076 (-0.13 %)
Code Size: 1282888 -> 1283184 (0.02 %) bytes
LDS: 28 -> 28 (0.00 %) blocks
Scratch: 495616 -> 492544 (-0.62 %) bytes per wave
Max Waves: 6843 -> 6853 (0.15 %)
Wait states: 0 -> 0 (0.00 %)
Reviewers: nhaehnle, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18451
llvm-svn: 264589
We're erasing MI here, but then immediately using it again inside the
`if`. This moves the erase after we're done using it.
Doing that reveals a second problem though - this case is missing a
break, so we fall through to the default and dereference MI again.
This is obviously a bug, though I don't know how to write a test that
triggers it - all we do in the error case is print some extra debug
output.
Both of these issue crash on lots of tests under ASAN with the
recycling allocator changes from PR26808 applied.
llvm-svn: 264442
Strengthen tests of storing frame indices.
Right now this just creates irrelevant scheduling changes.
We don't want to have multiple frame index operands
on an instruction. There seem to be various assumptions
that at least the same frame index will not appear twice
in the LocalStackSlotAllocation pass.
There's no reason to have this happen, and it just
makes it easy to introduce bugs where the immediate
offset is appplied to the storing instruction when it should
really be applied to the value being stored as a separate
add.
This might not be sufficient. It might still be problematic
to have an add fi, fi situation, but that's even less unlikely
to happen in real code.
llvm-svn: 264200
We can statically decide whether or not a register pressure set is for
SGPRs or VGPRs, so we don't need to re-compute this information in
SIRegisterInfo::getRegPressureSetLimit().
Differential Revision: http://reviews.llvm.org/D14805
llvm-svn: 264126
Summary:
Whole quad mode is already enabled for pixel shaders that compute
derivatives, but it must be suspended for instructions that cause a
shader to have side effects (i.e. stores and atomics).
This pass addresses the issue by storing the real (initial) live mask
in a register, masking EXEC before instructions that require exact
execution and (re-)enabling WQM where required.
This pass is run before register coalescing so that we can use
machine SSA for analysis.
The changes in this patch expose a problem with the second machine
scheduling pass: target independent instructions like COPY implicitly
use EXEC when they operate on VGPRs, but this fact is not encoded in
the MIR. This can lead to miscompilation because instructions are
moved past changes to EXEC.
This patch fixes the problem by adding use-implicit operands to
target independent instructions. Some general codegen passes are
relaxed to work with such implicit use operands.
Reviewers: arsenm, tstellarAMD, mareko
Subscribers: MatzeB, arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18162
llvm-svn: 263982
Summary:
When control flow is implemented using the exec mask, the compiler will
insert branch instructions to skip over the masked section when exec is
zero if the section contains more than a certain number of instructions.
The previous code would only count instructions in successor blocks,
and this patch modifies the code to start counting instructions in all
blocks between the start and end of the branch.
Reviewers: nhaehnle, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18282
llvm-svn: 263969
Summary:
Allow the selection of BUFFER_LOAD_FORMAT_x and _XY. Do this now before
the frontend patches land in Mesa. Eventually, we may want to automatically
reduce the size of loads at the LLVM IR level, which requires such overloads,
and in some cases Mesa can generate them directly.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18255
llvm-svn: 263792
Summary:
These intrinsics expose the BUFFER_ATOMIC_* instructions and will be used
by Mesa to implement atomics with buffer semantics. The intrinsic interface
matches that of buffer.load.format and buffer.store.format, except that the
GLC bit is not exposed (it is automatically deduced based on whether the
return value is used).
The change of hasSideEffects is required for TableGen to accept the pattern
that matches the intrinsic.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, rivanvx, llvm-commits
Differential Revision: http://reviews.llvm.org/D18151
llvm-svn: 263791
Summary:
We cannot easily deduce that an offset is in an SGPR, but the Mesa frontend
cannot easily make use of an explicit soffset parameter either. Furthermore,
it is likely that in the future, LLVM will be in a better position than the
frontend to choose an SGPR offset if possible.
Since there aren't any frontend uses of these intrinsics in upstream
repositories yet, I would like to take this opportunity to change the
intrinsic signatures to a single offset parameter, which is then selected
to immediate offsets or voffsets using a ComplexPattern.
Reviewers: arsenm, tstellarAMD, mareko
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18218
llvm-svn: 263790
Symmary:
ds_permute/ds_bpermute do not read memory so s_waitcnt is not needed.
Reviewers
arsenm, tstellarAMD
Subscribers
llvm-commits, arsenm
Differential Revision:
http://reviews.llvm.org/D18197
llvm-svn: 263720
Summary:
As explained by the comment, threads will typically see different values
returned by atomic instructions even if the arguments are equal.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18156
llvm-svn: 263719
Summary:
Uniform loops where the branch leaving the loop is predicated on VCCNZ
must be skipped if EXEC = 0, otherwise they will be infinite.
Reviewers: tstellarAMD, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18137
llvm-svn: 263658
And emit an error if it fails.
This prevents illegal instructions from getting sent to the GPU, which
would potentially result in a hang.
This is a candidate for the stable branch(es).
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
llvm-svn: 263627
Summary:
Static LDS size is saved in MachineFunctionInfo::LDSSize,
We define a pseudo instruction with usesCustomInserter bit set. Then, in EmitInstrWithCustomInserter,
we replace this pseudo instruction with a mov of MachineFunctionInfo::LDSSize.
Reviewers:
arsenm
tstellarAMD
Subscribers
llvm-commits, arsenm
Differential Revision:
http://reviews.llvm.org/D18064
llvm-svn: 263563
Summary:
When multiple threads perform an atomic op with the same arguments, they
will usually see different return values.
Reviewers: arsenm, tstellarAMD
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18101
llvm-svn: 263440
s_bitset0_b64, s_bitset1_b64 has 32-bit src0, not 64-bit.
s_rfe_b64 has just one destination operand and no source.
Uncomment S_BITCMP* and S_SETVSKIP, adjust SOPC_* classes for that.
Add s_memrealtime test and change comments in smem.s to follow common style.
Change test for s_memtime to use non-zero register to make it really test encoding.
Add tests for s_buffer_load*.
Add tests for SOPC instructions (same for SI and VI)
Differential Revision: http://reviews.llvm.org/D18040
llvm-svn: 263420
The constant is now at source operand 1 (previously at 2).
This is also how it is in legacy AMD sp3 assembler.
Update tests.
Differential Revision: http://reviews.llvm.org/D17984
llvm-svn: 263212
Frontend authors are strongly encouraged to keep allocas
in the entry block, so don't bother visiting every instruction
in the other blocks of the function.
llvm-svn: 263206
Summary:
They correspond to BUFFER_LOAD/STORE_FORMAT_XYZW and will be used by Mesa
to implement the GL_ARB_shader_image_load_store extension.
The intention is that for llvm.amdgcn.buffer.load.format, LLVM will decide
whether one of the _X/_XY/_XYZ opcodes can be used (similar to image sampling
and loads). However, this is not currently implemented.
For llvm.amdgcn.buffer.store, LLVM cannot decide to use one of the "smaller"
opcodes and therefore the intrinsic is overloaded. Currently, only the v4f32
is actually implemented since GLSL also only has a vec4 variant of the store
instructions, although it's conceivable that Mesa will want to be smarter
about this in the future.
BUFFER_LOAD_FORMAT_XYZW is already exposed via llvm.SI.vs.load.input, which
has a legacy name, pretends not to access memory, and does not capture the
full flexibility of the instruction.
Reviewers: arsenm, tstellarAMD, mareko
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D17277
llvm-svn: 263140
Supprot DPP syntax as used in SP3 (except several operands syntax).
Added dpp-specific operands in td-files.
Added DPP flag to TSFlags to determine if instruction is dpp in InstPrinter.
Support for VOP2 DPP instructions in td-files.
Some tests for DPP instructions.
ToDo:
- VOP2bInst:
- vcc is considered as operand
- AsmMatcher doesn't apply mnemonic aliases when parsing operands
- v_mac_f32
- v_nop
- disable instructions with 64-bit operands
- change dpp_ctrl assembler representation to conform sp3
Review: http://reviews.llvm.org/D17804
llvm-svn: 263008
Nicolai Haehnle