| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
| |
Adjust SimplifyDemandedVectorEltsForTargetNode to use the known elts masks instead of recomputing it locally.
llvm-svn: 374724
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
in combineSelect.
This seems to improve std::midpoint code where we have a min and
a max with the same condition. If we split the setcc we can end
up with two compares if the one of the operands is a constant.
Since we aggressively canonicalize compares with constants.
For non-constants it can interfere with our ability to share
control flow if we need to expand cmovs into control flow.
I'm also not sure I understand this min/max canonicalization code.
The motivating case talks about comparing with 0. But we don't
check for 0 explicitly.
Removes one instruction from the codegen for PR43658.
llvm-svn: 374706
|
|
|
|
|
|
| |
instructions with avx512.
llvm-svn: 374705
|
|
|
|
| |
llvm-svn: 374674
|
|
|
|
| |
llvm-svn: 374669
|
|
|
|
| |
llvm-svn: 374668
|
|
|
|
| |
llvm-svn: 374667
|
|
|
|
|
|
| |
This should go away once D66004 has landed and we can simplify shuffle chains using demanded elts.
llvm-svn: 374658
|
|
|
|
|
|
|
|
| |
I can't see any notable differences in costs between SSE2 and SSE42 arches for FADD/ADD reduction, so I've lowered the target to just SSE2.
I've also added vXi8 sum reduction costs in line with the PSADBW codegen and discussions on PR42674.
llvm-svn: 374655
|
|
|
|
|
|
|
|
|
|
|
| |
is the largest legal vector and the result type is at least 256 bits.
Since the input type is larger than 256-bits we'll need to some
concatenating to reassemble the results. The pack instructions
ability to concatenate while packing make this a shorter/faster
sequence.
llvm-svn: 374643
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
separately in loop-vectorize
In loop-vectorize, interleave count and vector factor depend on target register number. Currently, it does not
estimate different register pressure for different register class separately(especially for scalar type,
float type should not be on the same position with int type), so it's not accurate. Specifically,
it causes too many times interleaving/unrolling, result in too many register spills in loop body and hurting performance.
So we need classify the register classes in IR level, and importantly these are abstract register classes,
and are not the target register class of backend provided in td file. It's used to establish the mapping between
the types of IR values and the number of simultaneous live ranges to which we'd like to limit for some set of those types.
For example, POWER target, register num is special when VSX is enabled. When VSX is enabled, the number of int scalar register is 32(GPR),
float is 64(VSR), but for int and float vector register both are 64(VSR). So there should be 2 kinds of register class when vsx is enabled,
and 3 kinds of register class when VSX is NOT enabled.
It runs on POWER target, it makes big(+~30%) performance improvement in one specific bmk(503.bwaves_r) of spec2017 and no other obvious degressions.
Differential revision: https://reviews.llvm.org/D67148
llvm-svn: 374634
|
|
|
|
|
|
|
|
| |
We already did this for VTRUNCUS with a specific combination of
types. This extends this to VTRUNCS and handles any types where
a truncating store is legal.
llvm-svn: 374615
|
|
|
|
| |
llvm-svn: 374579
|
|
|
|
|
|
| |
Now that its used by isNegatibleForFree we should try to avoid costly deep recursion
llvm-svn: 374534
|
|
|
|
|
|
| |
saturating truncating store.
llvm-svn: 374509
|
|
|
|
|
|
|
|
|
|
| |
instructions in more situations.
If we don't have VLX we won't end up selecting a saturating
truncate for 256-bit or smaller vectors so we should just use
the pack lowering.
llvm-svn: 374487
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When handling the packus pattern for i32->i8 we do a two step
process using a packss to i16 followed by a packus to i8. If the
final i8 step is a type with less than 64-bits the packus step
will return SDValue(), but the i32->i16 step might have succeeded.
This leaves the nodes from the middle step dangling.
Guard against this by pre-checking that the number of elements is
at least 8 before doing the middle step.
With that check in place this should mean the only other
case the middle step itself can fail is when SSE2 is disabled. So
add an early SSE2 check then just assert that neither the middle
or final step ever fail.
llvm-svn: 374460
|
|
|
|
|
|
|
|
|
|
| |
inputs to be between 0 and 255 when zmm registers are disabled on SKX.
If we've disable zmm registers, the v16i32 will need to be split. This split will propagate through min/max the truncate. This creates two sequences that need to be concatenated back to v16i8. We can instead use packusdw to do part of the clamping, truncating, and concatenating all at once. Then we can use a vpmovuswb to finish off the clamp.
Differential Revision: https://reviews.llvm.org/D68763
llvm-svn: 374431
|
|
|
|
|
|
| |
Split off from D67557.
llvm-svn: 374356
|
|
|
|
|
|
| |
Split off from D67557, fixes the compile time regression mentioned in rL372756
llvm-svn: 374351
|
|
|
|
|
|
|
|
|
|
| |
placeholders. NFCI.
Continuing to undo the rL372756 reversion.
Differential Revision: https://reviews.llvm.org/D67557
llvm-svn: 374345
|
|
|
|
|
|
|
|
|
|
|
|
| |
As background, starting in D66309, I'm working on support unordered atomics analogous to volatile flags on normal LoadSDNode/StoreSDNodes for X86.
As part of that, I spent some time going through usages of LoadSDNode and StoreSDNode looking for cases where we might have missed a volatility check or need an atomic check. I couldn't find any cases that clearly miscompile - i.e. no test cases - but a couple of pieces in code loop suspicious though I can't figure out how to exercise them.
This patch adds defensive checks and asserts in the places my manual audit found. If anyone has any ideas on how to either a) disprove any of the checks, or b) hit the bug they might be fixing, I welcome suggestions.
Differential Revision: https://reviews.llvm.org/D68419
llvm-svn: 374261
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
separately in loop-vectorize"
Also Revert "[LoopVectorize] Fix non-debug builds after rL374017"
This reverts commit 9f41deccc0e648a006c9f38e11919f181b6c7e0a.
This reverts commit 18b6fe07bcf44294f200bd2b526cb737ed275c04.
The patch is breaking PowerPC internal build, checked with author, reverting
on behalf of him for now due to timezone.
llvm-svn: 374091
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
During the If-Converter optimization pay attention when copying or
deleting call instructions in order to keep call site information in
valid state.
Reviewers: aprantl, vsk, efriedma
Reviewed By: vsk, efriedma
Differential Revision: https://reviews.llvm.org/D66955
llvm-svn: 374068
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
in loop-vectorize
In loop-vectorize, interleave count and vector factor depend on target register number. Currently, it does not
estimate different register pressure for different register class separately(especially for scalar type,
float type should not be on the same position with int type), so it's not accurate. Specifically,
it causes too many times interleaving/unrolling, result in too many register spills in loop body and hurting performance.
So we need classify the register classes in IR level, and importantly these are abstract register classes,
and are not the target register class of backend provided in td file. It's used to establish the mapping between
the types of IR values and the number of simultaneous live ranges to which we'd like to limit for some set of those types.
For example, POWER target, register num is special when VSX is enabled. When VSX is enabled, the number of int scalar register is 32(GPR),
float is 64(VSR), but for int and float vector register both are 64(VSR). So there should be 2 kinds of register class when vsx is enabled,
and 3 kinds of register class when VSX is NOT enabled.
It runs on POWER target, it makes big(+~30%) performance improvement in one specific bmk(503.bwaves_r) of spec2017 and no other obvious degressions.
Differential revision: https://reviews.llvm.org/D67148
llvm-svn: 374017
|
|
|
|
|
|
|
|
|
|
| |
larger than i32.
Gather instructions can use i32 or i64 elements for indices. If
the index is zero extended from a type smaller than i32 to i64, we
can shrink the extend to just extend to i32.
llvm-svn: 373982
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When the target option GuaranteedTailCallOpt is specified, calls with
the fastcc calling convention will be transformed into tail calls if
they are in tail position. This diff adds a new calling convention,
tailcc, currently supported only on X86, which behaves the same way as
fastcc, except that the GuaranteedTailCallOpt flag does not need to
enabled in order to enable tail call optimization.
Patch by Dwight Guth <dwight.guth@runtimeverification.com>!
Reviewed By: lebedev.ri, paquette, rnk
Differential Revision: https://reviews.llvm.org/D67855
llvm-svn: 373976
|
|
|
|
|
|
|
|
| |
NFCI.
Stop all the callers from having to check the value type before calling getTargetShuffleInputs.
llvm-svn: 373915
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
possible.
Move the erasing and iterator updating inside to match the
other slow LEA function.
I've adapted code from optTwoAddrLEA and basically rebuilt the
implementation here. We do lose the kill flags now just like
optTwoAddrLEA. This runs late enough in the pipeline that
shouldn't really be a problem.
llvm-svn: 373877
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
load (PR43217)
If a fp scalar is loaded and then used as both a scalar and a vector broadcast, perform the load as a broadcast and then extract the scalar for 'free' from the 0th element.
This involved switching the order of the X86ISD::BROADCAST combines so we only convert to X86ISD::BROADCAST_LOAD once all other canonicalizations have been attempted.
Adds a DAGCombinerInfo::recursivelyDeleteUnusedNodes wrapper.
Fixes PR43217
Differential Revision: https://reviews.llvm.org/D68544
llvm-svn: 373871
|
|
|
|
| |
llvm-svn: 373870
|
|
|
|
|
|
|
|
| |
directly.
Move the resolveTargetShuffleInputsAndMask call to after the shuffle mask combine before the undef/zero constant fold instead.
llvm-svn: 373868
|
|
|
|
|
|
|
|
| |
Replaces setTargetShuffleZeroElements with getTargetShuffleAndZeroables which reports the Zeroable elements but doesn't merge them into the decoded target shuffle mask (the merging has been moved up into getTargetShuffleInputs until we can get rid of it entirely).
This is part of the work to fix PR43024 and allow us to use SimplifyDemandedElts to simplify shuffle chains - we need to get to a point where the target shuffle mask isn't adjusted by its source inputs but instead we cache them in a parallel Zeroable mask.
llvm-svn: 373867
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
v8i64->v8i8 truncate when v8i64 isn't legal
Summary:
The default legalization for v16i64->v16i8 tries to create a multiple stage truncate concatenating after each stage and truncating again. But avx512 implements truncates with multiple uops. So it should be better to truncate all the way to the desired element size and then concatenate the pieces using unpckl instructions. This minimizes the number of 2 uop truncates. The unpcks are all single uop instructions.
I tried to handle this by just custom splitting the v16i64->v16i8 shuffle. And hoped that the DAG combiner would leave the two halves in the state needed to make D68374 do the job for each half. This worked for the first half, but the second half got messed up. So I've implemented custom handling for v8i64->v8i8 when v8i64 needs to be split to produce the VTRUNCs directly.
Reviewers: RKSimon, spatel
Reviewed By: RKSimon
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D68428
llvm-svn: 373864
|
|
|
|
|
|
| |
of using setTargetShuffleZeroElements directly. NFCI.
llvm-svn: 373855
|
|
|
|
| |
llvm-svn: 373849
|
|
|
|
|
|
|
|
|
|
| |
We can make use of the Zeroable mask to indicate which elements we can safely set to zero instead of creating a target shuffle mask on the fly.
This allows us to remove createTargetShuffleMask.
This is part of the work to fix PR43024 and allow us to use SimplifyDemandedElts to simplify shuffle chains - we need to get to a point where the target shuffle masks isn't adjusted by its source inputs in setTargetShuffleZeroElements but instead we cache them in a parallel Zeroable mask.
llvm-svn: 373846
|
|
|
|
| |
llvm-svn: 373845
|
|
|
|
|
|
|
|
|
|
|
|
| |
(PR42025)
As discussed on PR42025, with more complex boolean math we can end up with many truncations/extensions of the comparison results through each bitop.
This patch handles the cases introduced in combineBitcastvxi1 by pushing the sign extension through the AND/OR/XOR ops so its just the original SETCC ops that gets extended.
Differential Revision: https://reviews.llvm.org/D68226
llvm-svn: 373834
|
|
|
|
|
|
| |
Rename some variables to match lowerShuffleAsRepeatedMaskAndLanePermute - prep work toward adding some equivalent sublane functionality.
llvm-svn: 373832
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
based on the immediate in MCInstLower
The immediate form of VPCMP can represent these completely. The
vpcmpgt/eq are just shorter encodings.
This patch removes the isel patterns and just swaps the opcodes
and removes the immediate in MCInstLower. This matches where we do
some other encodings tricks.
Removes over 10K bytes from the isel table.
Differential Revision: https://reviews.llvm.org/D68446
llvm-svn: 373766
|
|
|
|
|
|
|
|
| |
We already do this for ISD::TRUNCATE, but we can do the same for X86ISD::VTRUNC
Differential Revision: https://reviews.llvm.org/D68432
llvm-svn: 373765
|
|
|
|
| |
llvm-svn: 373706
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
truncated to v4i8 and concatenated into the lower 8 bytes with undef/zero upper bytes.
This patch recognizes the shuffle pattern we get from a
v8i64->v8i8 truncate when v8i64 isn't a legal type.
With VLX we can use two VTRUNCs, unpckldq, and a insert_subvector.
Diffrential Revision: https://reviews.llvm.org/D68374
llvm-svn: 373645
|
|
|
|
|
|
|
|
|
|
| |
We can make use of the Zeroable mask to indicate which elements we can safely set to zero instead of creating a target shuffle mask on the fly.
This only leaves one user of createTargetShuffleMask which we can hopefully get rid of in a similar manner.
This is part of the work to fix PR43024 and allow us to use SimplifyDemandedElts to simplify shuffle chains - we need to get to a point where the target shuffle masks isn't adjusted by its source inputs in setTargetShuffleZeroElements but instead we cache them in a parallel Zeroable mask.
llvm-svn: 373641
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
vbroadcast_load if the scalar size is the same.
This improves broadcast load folding of i64 elements on 32-bit
targets where i64 isn't legal.
Previously we had to represent these as vXf64 vbroadcast_loads and
a bitcast to vXi64. But we didn't have any isel patterns
looking for that.
This also allows us to remove or simplify some isel patterns that
were looking for bitcasted vbroadcast_loads.
llvm-svn: 373566
|
|
|
|
|
|
|
|
| |
VPMULLQ/VPMAXSQ/VPMAXUQ/VPMINSQ/VPMINUQ patterns.
More fixes for PR36191.
llvm-svn: 373560
|
|
|
|
|
|
| |
copy/pasted from right above them. NFC
llvm-svn: 373559
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
of bits that might be undef
The previous code tried to do a trick where we would extract the subvector from the location we were inserting. Then xor that with the new value. Take the xored value and clear out the bits above the subvector size. Then shift that xored subvector to the insert location. And finally xor that with the original vector. Since the old subvector was used in both xors, this would leave just the new subvector at the inserted location. Since the surrounding bits had been zeroed no other bits of the original vector would be modified.
Unfortunately, if the old subvector came from undef we might aggressively propagate the undef. Then we end up with the XORs not cancelling because they aren't using the same value for the two uses of the old subvector. @bkramer gave me a case that demonstrated this, but we haven't reduced it enough to make it easily readable to see what's happening.
This patch uses a safer, but more costly approach. It isolate the bits above the insertion and bits below the insert point and ORs those together leaving 0 for the insertion location. Then widens the subvector with 0s in the upper bits, shifts it into position with 0s in the lower bits. Then we do another OR.
Differential Revision: https://reviews.llvm.org/D68311
llvm-svn: 373495
|
|
|
|
|
|
|
|
|
| |
These patterns use zmm registers for 128/256-bit compares when
the VLX instructions aren't available. Previously we only
supported registers, but as PR36191 notes we can fold broadcast
loads, but not regular loads.
llvm-svn: 373423
|