We should be blocking the operand while we are in the routine that tries to find commutable operand indices. Doing it later means we might have missed out on another valid set of operands we could have commuted.
The intrinsic case was the only case that could really prevent commuting in getFMA3OpcodeToCommuteOperands. All the other cases in getThreeSrcCommuteCase were not reachable conditions as they were protected by findThreeSrcCommutedOpIndices.
With that abort case pushed earlier, we can remove all the abort checks and replace with asserts.
llvm-svn: 335446
It's easy for domination numbers to get out-of-date, and this is no more
costly than any of the other verifiers we already have, so it seems nice
to have.
A stage3 build with this Works On My Machine, so this hasn't caught any
bugs... yet. :)
llvm-svn: 335444
Summary:
A WebAssemblyException object contains BBs that belong to a 'catch' part
of the try-catch-end structure. Because CFGSort requires all the BBs
within a catch part to be sorted together as it does for loops, this
pass calculates the nesting structure of catch part of exceptions in a
function. Now this assumes the use of Windows EH instructions.
Reviewers: dschuff, majnemer
Subscribers: jfb, mgorny, sbc100, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D44134
llvm-svn: 335439
Summary:
Add WebAssemblyLateEHPrepare pass that does several small jobs for
exception handling. This runs before CFGSort, and is different from
WasmEHPrepare pass that runs before ISel, even though the names are
similar.
Reviewers: dschuff, majnemer
Subscribers: sbc100, jgravelle-google, sunfish, mgorny, llvm-commits
Differential Revision: https://reviews.llvm.org/D46803
llvm-svn: 335438
They appear to be untested other than the test case for p37879.ll and I believe we should be using SimplifyDemandedElts here to handle these cases.
llvm-svn: 335436
This patch has the same motivating example as D48466:
define void @foo(i64 %x, i32 %c.0282.in, i32 %d.0280, i32* %ptr0, i32* %ptr1) {
%c.0282 = and i32 %c.0282.in, 268435455
%a16 = lshr i64 32508, %x
%a17 = and i64 %a16, 1
%tobool = icmp eq i64 %a17, 0
%. = select i1 %tobool, i32 1, i32 2
%.286 = select i1 %tobool, i32 27, i32 26
%shr97 = lshr i32 %c.0282, %.
%shl98 = shl i32 %c.0282.in, %.286
%or99 = or i32 %shr97, %shl98
%shr100 = lshr i32 %d.0280, %.
%shl101 = shl i32 %d.0280, %.286
%or102 = or i32 %shr100, %shl101
store i32 %or99, i32* %ptr0
store i32 %or102, i32* %ptr1
ret void
}
...but I'm trying to kill the setcc bool math sooner rather than later.
By matching a larger pattern that includes both the low-bit mask and the trailing add/sub,
we can create a universally good fold because we always eliminate the condition code
intermediate value.
Here are Alive proofs for these (currently instcombine folds the 'add' variants, but
misses the 'sub' patterns):
https://rise4fun.com/Alive/Gsyp
Name: sub of zext cmp mask
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%z = zext i1 %c to i32
%r = sub i32 C1, %z
=>
%optional_cast = zext i8 %a to i32
%r = add i32 %optional_cast, C1-1
Name: add of zext cmp mask
%a = and i32 %x, 1
%c = icmp eq i32 %a, 0
%z = zext i1 %c to i8
%r = add i8 %z, C1
=>
%optional_cast = trunc i32 %a to i8
%r = sub i8 C1+1, %optional_cast
All of the tests look like improvements or neutral to me. But it is possible that x86
test+set+bitop is better than what we now show here. I suspect we could do better by
adding another fold for the 'sub' variants.
We start with select-of-constant in IR in the larger motivating test, so that's why I
included tests with selects. Proofs for those variants:
https://rise4fun.com/Alive/Bx1
Name: true const is bigger
Pre: C2 == (C1 + 1)
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%r = select i1 %c, i64 C2, i64 C1
=>
%z = zext i8 %a to i64
%r = sub i64 C2, %z
Name: false const is bigger
Pre: C2 == (C1 + 1)
%a = and i8 %x, 1
%c = icmp eq i8 %a, 0
%r = select i1 %c, i64 C1, i64 C2
=>
%z = zext i8 %a to i64
%r = add i64 C1, %z
Differential Revision: https://reviews.llvm.org/D48466
llvm-svn: 335433
For some reason the 64-bit patterns were separated from their 8/16/32-bit friends, but only for add/sub/mul. For and/or/xor they were together.
llvm-svn: 335429
-Ensure EIP isn't used with an index reigster.
-Ensure EIP isn't used as index register.
-Ensure base register isn't a vector register.
-Ensure eiz/riz usage matches the size of their base register.
llvm-svn: 335412
FDiv is replaced with multiplication by reciprocal and invariant
reciprocal is hoisted out of the loop, while multiplication remains
even if invariant.
Switch checks for all invariant operands and only invariant
denominator to fix the issue.
Differential Revision: https://reviews.llvm.org/D48447
llvm-svn: 335411
Implements PR34259
Intrinsics.h is a very popular header. Most LLVM TUs care about things
like dbg_value, but they don't care how they are implemented. After I
split these out, IntrinsicImpl.inc is 1.7 MB, so this saves each LLVM TU
from scanning 1.7 MB of source that gets pre-processed away.
It also means we can modify intrinsic properties without triggering a
full rebuild, but that's probably less of a win.
I think the next best thing to do would be to split out the target
intrinsics into their own header. Very, very few TUs care about
target-specific intrinsics. It's very hard to split up the target
independent intrinsics like llvm.expect, assume, and dbg.value, though.
llvm-svn: 335407
Previously, to support (%dx) we left a wide open hole in our 16-bit memory address checking. This let this address value be used with any instruction without error in the parser. It would later fail in the encoder with an assertion failure on debug builds and who knows what on release builds.
This patch passes the mnemonic down to the memory operand parsing function so we can allow the (%dx) form only on specific instructions.
llvm-svn: 335403
This gets rid of a bunch of weird special cases; instead, just use SCEV
rewriting for everything. In addition to being simpler, this fixes a
bug where we would use the wrong stride in certain edge cases.
The one bit I'm not quite sure about is the trip count handling,
specifically the FIXME about overflow. In general, I think we need to
widen the exit condition, but that's probably not profitable if the new
type isn't legal, so we probably need a check somewhere. That said, I
don't think I'm making the existing problem any worse.
As a followup to this, a bunch of IV-related code in root-finding could
be cleaned up; with SCEV-based rewriting, there isn't any reason to
assume a loop will have exactly one or two PHI nodes.
Differential Revision: https://reviews.llvm.org/D45191
llvm-svn: 335400
This allows us to check these:
-16-bit addressing doesn't support scale so we should error if we find one there.
-Multiplying ESP/RSP by a scale even if the scale is 1 should be an error because ESP/RSP can't be an index.
llvm-svn: 335398
By default, the second register gets assigned to the index register slot. But ESP can't be an index register so we need to swap it with the other register.
There's still a slight bug that we allow [EAX+ESP*1]. The existence of the multiply even though its with 1 should force ESP to the index register and trigger an error, but it doesn't currently.
llvm-svn: 335394
Since we are now producing a summary also for regular LTO builds, we
need to run the NameAnonGlobals pass in those cases as well (the
summary cannot handle anonymous globals).
See https://reviews.llvm.org/D34156 for details on the original change.
This reverts commit 6c9ee4a4a438a8059aacc809b2dd57128fccd6b3.
llvm-svn: 335385
The second register is the index register and should only be %si or %di if used with a base register. And in that case the base register should be %bp or %bx.
This makes us compatible with gas.
We do still need to support both orders with Intel syntax which uses [bp+si] and [si+bp]
llvm-svn: 335384
(%bp) can't be encoded without a displacement. The encoding is instead used for displacement alone. So a 1 byte displacement of 0 must be used. But if there is an index register we can encode without a displacement.
llvm-svn: 335379
Summary:
In LoopUnswitch when replacing a branch Parent -> Succ with a conditional
branch Parent -> True & Parent->False, the DomTree updates should insert an edge for
each of True/False if True/False are different than Succ, and delete Parent->Succ edge
if both are different. The comparison with Succ appears to be incorect,
it's comparing with Parent instead.
There is no test failing either before or after this change, but it seems to me this is
the right way to do the update.
Reviewers: chandlerc, kuhar
Subscribers: sanjoy, jlebar, llvm-commits
Differential Revision: https://reviews.llvm.org/D48457
llvm-svn: 335369
Enable tryToVectorizeList to support InstructionsState alternate opcode patterns at a root (build vector etc.) as well as further down the vectorization tree.
NOTE: This patch reduces some of the debug reporting if there are opcode mismatches - I can try to add it back if it proves a problem. But it could get rather messy trying to provide equivalent verbose debug strings via getSameOpcode etc.
Differential Revision: https://reviews.llvm.org/D48488
llvm-svn: 335364
DWARF v5 explicitly represents file #0 in the line table. Prior
versions did not, so ".loc 0" is still an error in those cases.
Differential Revision: https://reviews.llvm.org/D48452
llvm-svn: 335350
SLP currently only accepts (F)Add/(F)Sub alternate counterpart ops to be merged into an alternate shuffle.
This patch relaxes this to accept any pair of BinaryOperator opcodes instead, assuming the target's cost model accepts the vectorization+shuffle.
Differential Revision: https://reviews.llvm.org/D48477
llvm-svn: 335349
With compilation fix.
Original commit message:
D39788 added a '.stack-size' section containing metadata on function stack sizes
to output ELF files behind the new -stack-size-section flag.
This change does following two things on top:
1) Imagine the case when there are -ffunction-sections flag given and there are text sections in COMDATs.
The patch adds a '.stack-size' section into corresponding COMDAT group, so that linker will be able to
eliminate them fast during resolving the COMDATs.
2) Patch sets a SHF_LINK_ORDER flag and links '.stack-size' with the corresponding .text.
With that linker will be able to do -gc-sections on dead stack sizes sections.
Differential revision: https://reviews.llvm.org/D46874
llvm-svn: 335336
D39788 added a '.stack-size' section containing metadata on function stack sizes
to output ELF files behind the new -stack-size-section flag.
This change does following two things on top:
1) Imagine the case when there are -ffunction-sections flag given and there are text sections in COMDATs.
The patch adds a '.stack-size' section into corresponding COMDAT group, so that linker will be able to
eliminate them fast during resolving the COMDATs.
2) Patch sets a SHF_LINK_ORDER flag and links '.stack-size' with the corresponding .text.
With that linker will be able to do -gc-sections on dead stack sizes sections.
Differential revision: https://reviews.llvm.org/D46874
llvm-svn: 335332
AArch64 was only setting costs for SK_Transpose, which meant that many of the simpler shuffles (e.g. SK_Select and SK_PermuteSingleSrc for larger vector elements) was being severely overestimated by the default shuffle expansion.
This patch adds costs to help improve SLP performance and avoid a regression in reductions introduced by D48174.
I'm not very knowledgeable about AArch64 shuffle lowering so I've kept the extra costs to a minimum - someone who knows this code can add extra costs which should improve vectorization a lot more.
Differential Revision: https://reviews.llvm.org/D48172
llvm-svn: 335329
This sets target feature FeatureStrictAlign for Armv6-m and Armv8-m.baseline,
because it has no support for unaligned accesses.
It looks like we always pass target feature "+strict-align" from
Clang, so this is not a user facing problem, but querying the subtarget
(in e.g. llc) for unaligned access support is incorrect.
Differential Revision: https://reviews.llvm.org/D48437
llvm-svn: 335326
Not sure why the 32/64 split is needed in the atomic_load
store hierarchies. The regular PatFrags do this, but we don't
do it for the existing handling for global.
llvm-svn: 335325
Changing the logic of scalar mask folding to check for valid input types rather
than against invalid ones, making it more robust and fixing PR37879.
Differential Revision: https://reviews.llvm.org/D48366
llvm-svn: 335323
This is the first pass in the main pipeline to use the legacy PM's
ability to run function analyses "on demand". Unfortunately, it turns
out there are bugs in that somewhat-hacky approach. At the very least,
it leaks memory and doesn't support -debug-pass=Structure. Unclear if
there are larger issues or not, but this should get the sanitizer bots
back to green by fixing the memory leaks.
llvm-svn: 335320
Summary:
We can select all instructions that are marked as legal in a full piglit run,
so now is a good time to make the TableGen'd instruction selector default
for all opcodes. This is NFC for a full piglit run, which is why there are
no tests.
Reviewers: arsenm, nhaehnle
Subscribers: kzhuravl, wdng, yaxunl, rovka, kristof.beyls, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48198
llvm-svn: 335319
clear out deleted loops from the current queue beyond just the current
loop.
This is important because SimpleLoopUnswitch will now enqueue the same
loop to be re-processed. When it does this with the legacy PM, we don't
have a way of canceling the rest of the pipeline and so we can end up
deleting the loop before we reprocess it. =/
This change also makes it easy to support deleting other loops in the
queue to process, although I don't have any use cases for that.
Differential Revision: https://reviews.llvm.org/D48470
llvm-svn: 335317
With non-commutative binops, we could be using the same
variable value as operand 0 in 1 binop and operand 1 in
the other, so we have to check for that possibility and
bail out.
llvm-svn: 335312
This patch adds support for generating a call graph profile from Branch Frequency Info.
The CGProfile module pass simply gets the block profile count for each BB and scans for call instructions. For each call instruction it adds an edge from the current function to the called function with the current BB block profile count as the weight.
After scanning all the functions, it generates an appending module flag containing the data. The format looks like:
!llvm.module.flags = !{!0}
!0 = !{i32 5, !"CG Profile", !1}
!1 = !{!2, !3, !4} ; List of edges
!2 = !{void ()* @a, void ()* @b, i64 32} ; Edge from a to b with a weight of 32
!3 = !{void (i1)* @freq, void ()* @a, i64 11}
!4 = !{void (i1)* @freq, void ()* @b, i64 20}
Differential Revision: https://reviews.llvm.org/D48105
llvm-svn: 335306
Summary:
The large code model allows code and data segments to exceed 2GB, which
means that some symbol references may require a displacement that cannot
be encoded as a displacement from RIP. The large PIC model even relaxes
the assumption that the GOT itself is within 2GB of all code. Therefore,
we need a special code sequence to materialize it:
.LtmpN:
leaq .LtmpN(%rip), %rbx
movabsq $_GLOBAL_OFFSET_TABLE_-.LtmpN, %rax # Scratch
addq %rax, %rbx # GOT base reg
From that, non-local references go through the GOT base register instead
of being PC-relative loads. Local references typically use GOTOFF
symbols, like this:
movq extern_gv@GOT(%rbx), %rax
movq local_gv@GOTOFF(%rbx), %rax
All calls end up being indirect:
movabsq $local_fn@GOTOFF, %rax
addq %rbx, %rax
callq *%rax
The medium code model retains the assumption that the code segment is
less than 2GB, so calls are once again direct, and the RIP-relative
loads can be used to access the GOT. Materializing the GOT is easy:
leaq _GLOBAL_OFFSET_TABLE_(%rip), %rbx # GOT base reg
DSO local data accesses will use it:
movq local_gv@GOTOFF(%rbx), %rax
Non-local data accesses will use RIP-relative addressing, which means we
may not always need to materialize the GOT base:
movq extern_gv@GOTPCREL(%rip), %rax
Direct calls are basically the same as they are in the small code model:
They use direct, PC-relative addressing, and the PLT is used for calls
to non-local functions.
This patch adds reasonably comprehensive testing of LEA, but there are
lots of interesting folding opportunities that are unimplemented.
Reviewers: chandlerc, echristo
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D47211
llvm-svn: 335297
Summary:
A reprise of D25849.
This crash was found through fuzzing some time ago and was documented in PR28879.
No check for load size has been added due to the following tests:
- Transforms/GVN/invariant.group.ll
- Transforms/GVN/pr10820.ll
These tests expect load sizes that are not a multiple of eight.
Thanks to @davide for the original patch.
Reviewers: nlopes, davide, RKSimon, reames, efriedma
Reviewed By: efriedma
Subscribers: davide, llvm-commits, Prazek
Differential Revision: https://reviews.llvm.org/D48330
llvm-svn: 335294
Summary:
This initiates a discussion on changing Polly accordingly while re-applying r335197 (D48338).
I have never worked on Polly. The proposed change to param_div_div_div_2.ll is not educated, but just patterns that match the output.
All LLVM files are already reviewed in D48338.
Reviewers: jdoerfert, bollu, efriedma
Subscribers: jlebar, sanjoy, hiraditya, llvm-commits, bixia
Differential Revision: https://reviews.llvm.org/D48453
llvm-svn: 335292
Summary:
GCC and the binutils COFF linker do comdats differently from MSVC.
If we want to be ABI compatible, we have to do what they do, which is to
emit unique section names like ".text$_Z3foov" instead of short section
names like ".text". Otherwise, the binutils linker gets confused and
reports multiple definition errors when two object files from GCC and
Clang containing the same inline function are linked together.
The best description of the issue is probably at
https://github.com/Alexpux/MINGW-packages/issues/1677, we don't seem to
have a good one in our tracker.
I fixed up the .pdata and .xdata sections needed everywhere other than
32-bit x86. GCC doesn't use associative comdats for those, it appears to
rely on the section name.
Reviewers: smeenai, compnerd, mstorsjo, martell, mati865
Subscribers: llvm-commits, hiraditya
Differential Revision: https://reviews.llvm.org/D48402
llvm-svn: 335286
This is the simplest case from PR37806:
https://bugs.llvm.org/show_bug.cgi?id=37806
If we have a common variable operand used in a pair of binops with vector constants
that are vector selected together, then we can constant shuffle the constant vectors
to eliminate the shuffle instruction.
This has some tricky parts that are hopefully addressed in the tests and their
respective comments:
1. If the shuffle mask contains an undef element, then that lane of the result is
undef:
http://llvm.org/docs/LangRef.html#shufflevector-instruction
Therefore, we can replace the constant in that lane with an undef value except
for div/rem. With div/rem, an undef in the divisor would cause the whole op to
be undef. So I'm using the same hack as in D47686 - replace the undefs with '1'.
2. Intersect the wrapping and FMF of the original binops for the new binop. There
should be no extra poison or fast-math potential in the new binop that wasn't
possible in the original code.
3. Disregard other uses. Given that we're eliminating uses (shortening the
dependency chain), I think that's always the right IR canonicalization. But
I purposely chose the udiv test to demonstrate the scenario where both
intermediate values have other uses because that seems likely worse for
codegen with an expensive math op. This seems like a very rare possibility to
me, so I don't think it requires a backend patch first.
Differential Revision: https://reviews.llvm.org/D48401
llvm-svn: 335283
Update AMDGPU assembler syntax behind the code-object-v3 feature:
* Replace/rename most AMDGPU assembler directives/symbols and document them.
* Provide more diagnostics (e.g. values out of range, missing values, repeated
values).
* Provide path for backwards compatibility, even with underlying descriptor
changes.
Differential Revision: https://reviews.llvm.org/D47736
llvm-svn: 335281
This reverts commit r335206.
As discussed here: https://reviews.llvm.org/rL333740, a fix will come
tomorrow. In the meanwhile, revert this to fix some bots.
llvm-svn: 335272
BlockWaitcntProcessedSet was not being cleared between calls, so it was
producing incorrect counts in cases where MBB addresses happened to coincide
across multiple calls.
Differential Revision: https://reviews.llvm.org/D48391
llvm-svn: 335268
and everything that comes with it from implementation
and v3 header files.
Leave definition in v2 header files for backwards
compatibility.
Differential Revision: https://reviews.llvm.org/D48191
llvm-svn: 335267
Summary:
The logic for handling the sinking of COPY instructions was generating
different code when building with debug flags.
The original code did not take into consideration debug instructions. This
resulted in the registers in the DBG_VALUE instructions being treated as used,
and prevented the COPY from being sunk. This patch avoids analyzing debug
instructions when trying to sink COPY instructions.
This patch also creates a routine from the code in MachineSinking::SinkInstruction to
perform the logic of sinking an instruction along with its debug instructions.
This functionality is used in multiple places, including the code for sinking COPY instrs.
Reviewers: junbuml, javed.absar, MatzeB, bjope
Reviewed By: bjope
Subscribers: aprantl, probinson, thegameg, jonpa, bjope, vsk, kristof.beyls, JDevlieghere, llvm-commits
Tags: #debug-info
Differential Revision: https://reviews.llvm.org/D45637
llvm-svn: 335264
The previous code worked with vectors, but it failed when the
vector constants contained undef elements.
The matchers handle those cases.
llvm-svn: 335262
This is outwardly NFC from what I can tell, but it should be more efficient
to simplify first (despite the name, SimplifyAssociativeOrCommutative does
not actually simplify as InstSimplify does - it creates/morphs instructions).
This should make it easier to refactor duplicated code that runs for all binops.
llvm-svn: 335258
This reverts commit d8f57105010cc7e78026e511d5def873fc91e0e7.
Original Commit:
Author: Haicheng Wu <haicheng@codeaurora.org>
Date: Sun Feb 18 13:51:33 2018 +0000
[AArch64] Coalesce Copy Zero during instruction selection
Add special case for copy of zero to avoid a double copy.
Differential Revision: https://reviews.llvm.org/D36104
Author's intention is to remove a BB that has one mov instruction. In
order to do that, d8f571050 pessmizes MachineSinking by introducing a
copy, such that mov instruction is NOT moved to the BB. Optimization
downstream gets rid of the BB with only mov instruction. This works well
if we have only one fall through branch as there is only one "extra"
mov instruction.
If we have multiple fall throughs, we will have a lot of redundant movs.
In such a case, it's better to have this BB which has one mov instruction.
This is causing degradation in jpeg, fft and other codebases. I believe
if we want to remove a BB with only one branch instruction, we should not
pessimize Machine Sinking at all, and find some other solution.
llvm-svn: 335251
Allowed folding for "and/or" binops with non-constant operand if
arguments of select are 0/-1 values.
Normally this code with "and" opcode does not get to a DAG combiner
and simplified yet in the InstCombine. However AMDGPU produces it
during lowering and InstCombine has no chance to optimize it out.
In turn the same pattern with "or" opcode can reach DAG.
Differential Revision: https://reviews.llvm.org/D48301
llvm-svn: 335250
This option allows codegen (such as DAGCombine or MI scheduling) to use alias
analysis information, which can help with the codegen on in-order cpu's,
especially machine scheduling. Here I have done things the same way as AArch64,
adding a subtarget feature to enable this for specific cores, and enabled it for
the R52 where we have a schedule to make use of it.
Differential Revision: https://reviews.llvm.org/D48074
llvm-svn: 335249
Summary:
When expanding the PseudoTail in expandFunctionCall() we were using X6
to save the return address. Since this is a tail call the return
address is not needed, this patch replaces it with X0 to be ignored.
This matches the behaviour listed in the ISA V2.2 document page 110.
tail offset -----> jalr x0, x6, offset
GCC exhibits the same behavior.
Reviewers: apazos, asb, mgrang
Reviewed By: asb
Subscribers: rbar, johnrusso, simoncook, niosHD, kito-cheng, shiva0217, zzheng, edward-jones, rogfer01
Differential Revision: https://reviews.llvm.org/D48343
llvm-svn: 335239
This should help in lowering the following four intrinsics:
_mm256_cvtepi32_epi8
_mm256_cvtepi64_epi16
_mm256_cvtepi64_epi8
_mm512_cvtepi64_epi8
Differential Revision: https://reviews.llvm.org/D46957
llvm-svn: 335238
Summary:
For sample and gather ops, we can accurately determine the set of
vaddr-size instruction variants that are required. This reduces
the size of instruction tables by ~5%.
The number of machine instruction opcodes is reduced from 10002
to 9476.
Change-Id: Ie7fc65d3657b762c7816017fe70b2e9bec644a8a
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye
Differential Revision: https://reviews.llvm.org/D48168
llvm-svn: 335232
Summary:
This also removes the need for atomic pseudo instructions, since
we select the correct encoding directly in SITargetLowering::lowerImage
for dimension-aware image intrinsics.
Mesa uses dimension-aware image intrinsics since
commit a9a7993441.
Change-Id: I7473d20009476a4ed6d919cae4e6dca9ff42e77a
Reviewers: arsenm, rampitec, mareko, tpr, b-sumner
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48167
llvm-svn: 335231
Summary:
Use the expanded features of the TableGen generic tables to avoid manually
adding the combinatorially exploded set of intrinsics. The
getAMDGPUImageDimIntrinsic lookup function is early-out,
i.e. non-AMDGPU intrinsics will never look at the underlying table.
Use a generic approach for getting the new intrinsic overload to keep the
code simple, and make the image dmask handling more generic:
- handle non-sampler image loads
- handle the case where the set of demanded elements is not a prefix
There is some overlap between this code and an optimization that happens
in the backend during code generation. They currently complement each other:
- only the codegen optimization can generate vec3 loads
- only the InstCombine optimization can handle D16
The InstCombine optimization also likely covers more cases since the
codegen optimization is fairly ad-hoc. Ideally, we'll remove the optimization
in codegen once the infrastructure for vec3 is in place (which will probably
take a long time).
Modify the test cases to use dimension-aware intrinsics. This makes it
easier to see that the test coverage for the new intrinsics is equivalent,
and the old style intrinsics will be removed in a follow-up commit anyway.
Change-Id: I4b91ea661413d13004956fe4ef7d13d41b8ce3ad
Reviewers: arsenm, rampitec, majnemer
Subscribers: kzhuravl, wdng, mgorny, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48165
llvm-svn: 335230
Summary:
Having TableGen patterns for image intrinsics is hitting limitations:
for D16 we already have to manually pre-lower the packing of data
values, and we will have to do the same for A16 eventually.
Since there is already some custom C++ code anyway, it is arguably easier
to just do everything in C++, now that we can use the beefed-up generic
tables backend of TableGen to provide all the required metadata and map
intrinsics to corresponding opcodes. With this approach, all image
intrinsic lowering happens in SITargetLowering::lowerImage. That code is
dense due to all the cases that it handles, but it should still be easier
to follow than what we had before, by virtue of it all being done in a
single location, and by virtue of not relying on the TableGen pattern
magic that very few people really understand.
This means that we will have MachineSDNodes with MIMG instructions
during DAG combining, but that seems alright: previously we had
intrinsic nodes instead, but those are similarly opaque to the generic
CodeGen infrastructure, and the final pattern matching just did a 1:1
translation to machine instructions anyway. If anything, the fact that
we now merge the address words into a vector before DAG combine should
be an advantage.
Change-Id: I417f26bd88f54ce9781c1668acc01f3f99774de6
Reviewers: arsenm, rampitec, rtaylor, tstellar
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48017
llvm-svn: 335228
Summary:
This allows us to access rich information about MIMG opcodes from C++ code.
Simplifying the mapping between equivalent opcodes of different data size
becomes quite natural.
This also flattens the MIMG-related class and multiclass hierarchy a little,
and collapses together some of the scaffolding for sample and gather4 opcodes.
Change-Id: I1a2549fdc1e881ff100e5393d2d87e73729a0ccd
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48016
llvm-svn: 335227
Summary:
This will allows us to provide rich metadata about the instructions
in tables that are accessible by custom C++ code.
Change-Id: Id9305a26304ab6a6cceb6c65c8cd49141cc0101d
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D48011
llvm-svn: 335224
Summary:
Kill instructions sometimes do use SCC in unusual circumstances, when
v_cmpx cannot be used due to the operands that are involved.
Additionally, even if SCC was never defined by the expansion, kill pseudos
could previously occur between an s_cmp and an s_cbranch_scc, which breaks
the SCC liveness tracking when the pseudo is expanded to split the basic
block. While it would be possible to explicitly mark the SCC as live-in for
the successor basic block, it's simpler to just mark the pseudo as using SCC,
so that such a sequence is never emitted by instruction selection in the
first place.
A similar issue affects indirect source/dest pseudos in principle, although
I haven't been able to come up with a test case where it actually matters
(this affects instruction selection, so a MIR test can't be used).
Fixes: dEQP-GLES3.functional.shaders.discard.dynamic_loop_always
Change-Id: Ica8d82ecff1a763b892a1112cf1b06c948863a4f
Reviewers: arsenm, rampitec
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D47761
llvm-svn: 335223
Summary:
This allows us to reduce the number of different machine instruction
opcodes, which reduces the table sizes and helps flatten the TableGen
multiclass hierarchies.
We can do this because for each hardware MIMG opcode, we have a full set
of IMAGE_xxx_Vn_Vm machine instructions for all required sizes of vdata
and vaddr registers. Instead of having separate D16 machine instructions,
a packed D16 instructions loading e.g. 4 components can simply use the
same V2 opcode variant that non-D16 instructions use.
We still require a TSFlag for D16 buffer instructions, because the
D16-ness of buffer instructions is part of the opcode. Renaming the flag
should help avoid future confusion.
The one non-obvious code change is that for gather4 instructions, the
disassembler can no longer automatically decide whether to use a V2 or
a V4 variant. The existing logic which choose the correct variant for
other MIMG instruction is extended to cover gather4 as well.
As a bonus, some of the assembler error messages are now more helpful
(e.g., complaining about a wrong data size instead of a non-existing
instruction).
While we're at it, delete a whole bunch of dead legacy TableGen code.
Change-Id: I89b02c2841c06f95e662541433e597f5d4553978
Reviewers: arsenm, rampitec, kzhuravl, artem.tamazov, dp, rtaylor
Subscribers: wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D47434
llvm-svn: 335222
Summary:
This also allows inner foreach loops to have a list that depends on
the iteration variable of an outer foreach loop. The test cases show
some very simple examples of how this can be used.
This was perhaps the last remaining major non-orthogonality in the
TableGen frontend.
Change-Id: I79b92d41a5c0e7c03cc8af4000c5e1bda5ef464d
Reviewers: tra, simon_tatham, craig.topper, MartinO, arsenm
Subscribers: wdng, llvm-commits
Differential Revision: https://reviews.llvm.org/D47431
llvm-svn: 335221
This enables da-delinearize in Dependence Analysis for delinearizing array
accesses into multiple dimensions. This can help to increase the power of
Dependence analysis on multi-dimensional arrays and prevent having to fall
back to the slower and less accurate MIV tests. It adds static checks on the
bounds of the arrays to ensure that one dimension doesn't overflow into
another, and brings our code in line with our tests.
Differential Revision: https://reviews.llvm.org/D45872
llvm-svn: 335217
These were being over cautious for costs for one/two op general shuffles - VSHUFPD doesn't have to replicate the same shuffle in both lanes like VSHUFPS does.
llvm-svn: 335216
Summary:
In some cases, these operands lacked the IsDebug property, which is meant to signal that
they should not affect codegen. This patch adds a check for this property in the
MachineVerifier and adds it where it was missing.
This includes refactorings to use MachineInstrBuilder construction functions instead of
manually setting up the intrinsic everywhere.
Patch by: JesperAntonsson
Reviewers: aprantl, rnk, echristo, javed.absar
Reviewed By: aprantl
Subscribers: qcolombet, sdardis, nemanjai, JDevlieghere, atanasyan, llvm-commits
Differential Revision: https://reviews.llvm.org/D48319
llvm-svn: 335214
Simon Pilgrim