We want to do this for 2 reasons:
1. Value tracking does not recognize the ashr variant, so it would fail to match for cases like D39766.
2. DAGCombiner does better at producing optimal codegen when we have the cmp+sel pattern.
More detail about what happens in the backend:
1. DAGCombiner has a generic transform for all targets to convert the scalar cmp+sel variant of abs
into the shift variant. That is the opposite of this IR canonicalization.
2. DAGCombiner has a generic transform for all targets to convert the vector cmp+sel variant of abs
into either an ABS node or the shift variant. That is again the opposite of this IR canonicalization.
3. DAGCombiner has a generic transform for all targets to convert the exact shift variants produced by #1 or #2
into an ISD::ABS node. Note: It would be an efficiency improvement if we had #1 go directly to an ABS node
when that's legal/custom.
4. The pattern matching above is incomplete, so it is possible to escape the intended/optimal codegen in a
variety of ways.
a. For #2, the vector path is missing the case for setlt with a '1' constant.
b. For #3, we are missing a match for commuted versions of the shift variants.
5. Therefore, this IR canonicalization can only help get us to the optimal codegen. The version of cmp+sel
produced by this patch will be recognized in the DAG and converted to an ABS node when possible or the
shift sequence when not.
6. In the following examples with this patch applied, we may get conditional moves rather than the shift
produced by the generic DAGCombiner transforms. The conditional move is created using a target-specific
decision for any given target. Whether it is optimal or not for a particular subtarget may be up for debate.
define i32 @abs_shifty(i32 %x) {
%signbit = ashr i32 %x, 31
%add = add i32 %signbit, %x
%abs = xor i32 %signbit, %add
ret i32 %abs
}
define i32 @abs_cmpsubsel(i32 %x) {
%cmp = icmp slt i32 %x, zeroinitializer
%sub = sub i32 zeroinitializer, %x
%abs = select i1 %cmp, i32 %sub, i32 %x
ret i32 %abs
}
define <4 x i32> @abs_shifty_vec(<4 x i32> %x) {
%signbit = ashr <4 x i32> %x, <i32 31, i32 31, i32 31, i32 31>
%add = add <4 x i32> %signbit, %x
%abs = xor <4 x i32> %signbit, %add
ret <4 x i32> %abs
}
define <4 x i32> @abs_cmpsubsel_vec(<4 x i32> %x) {
%cmp = icmp slt <4 x i32> %x, zeroinitializer
%sub = sub <4 x i32> zeroinitializer, %x
%abs = select <4 x i1> %cmp, <4 x i32> %sub, <4 x i32> %x
ret <4 x i32> %abs
}
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=x86_64 -mattr=avx
> abs_shifty:
> movl %edi, %eax
> negl %eax
> cmovll %edi, %eax
> retq
>
> abs_cmpsubsel:
> movl %edi, %eax
> negl %eax
> cmovll %edi, %eax
> retq
>
> abs_shifty_vec:
> vpabsd %xmm0, %xmm0
> retq
>
> abs_cmpsubsel_vec:
> vpabsd %xmm0, %xmm0
> retq
>
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=aarch64
> abs_shifty:
> cmp w0, #0 // =0
> cneg w0, w0, mi
> ret
>
> abs_cmpsubsel:
> cmp w0, #0 // =0
> cneg w0, w0, mi
> ret
>
> abs_shifty_vec:
> abs v0.4s, v0.4s
> ret
>
> abs_cmpsubsel_vec:
> abs v0.4s, v0.4s
> ret
>
> $ ./opt -instcombine shiftyabs.ll -S | ./llc -o - -mtriple=powerpc64le
> abs_shifty:
> srawi 4, 3, 31
> add 3, 3, 4
> xor 3, 3, 4
> blr
>
> abs_cmpsubsel:
> srawi 4, 3, 31
> add 3, 3, 4
> xor 3, 3, 4
> blr
>
> abs_shifty_vec:
> vspltisw 3, -16
> vspltisw 4, 15
> vsubuwm 3, 4, 3
> vsraw 3, 2, 3
> vadduwm 2, 2, 3
> xxlxor 34, 34, 35
> blr
>
> abs_cmpsubsel_vec:
> vspltisw 3, -16
> vspltisw 4, 15
> vsubuwm 3, 4, 3
> vsraw 3, 2, 3
> vadduwm 2, 2, 3
> xxlxor 34, 34, 35
> blr
>
Differential Revision: https://reviews.llvm.org/D40984
llvm-svn: 320921
There are 2 parts to getting the -fassociative-math command-line flag translated to LLVM FMF:
1. In the driver/frontend, we accept the flag and its 'no' inverse and deal with the
interactions with other flags like -ffast-math -fno-signed-zeros -fno-trapping-math.
This was mostly already done - we just need to translate the flag as a codegen option.
The test file is complicated because there are many potential combinations of flags here.
Note that we are matching gcc's behavior that requires 'nsz' and no-trapping-math.
2. In codegen, we map the codegen option to FMF in the IR builder. This is simple code and
corresponding test.
For the motivating example from PR27372:
float foo(float a, float x) { return ((a + x) - x); }
$ ./clang -O2 27372.c -S -o - -ffast-math -fno-associative-math -emit-llvm | egrep 'fadd|fsub'
%add = fadd nnan ninf nsz arcp contract float %0, %1
%sub = fsub nnan ninf nsz arcp contract float %add, %2
So 'reassoc' is off as expected (and so is the new 'afn' but that's a different patch).
This case now works as expected end-to-end although the underlying logic is still wrong:
$ ./clang -O2 27372.c -S -o - -ffast-math -fno-associative-math | grep xmm
addss %xmm1, %xmm0
subss %xmm1, %xmm0
We're not done because the case where 'reassoc' is set is ignored by optimizer passes. Example:
$ ./clang -O2 27372.c -S -o - -fassociative-math -fno-signed-zeros -fno-trapping-math -emit-llvm | grep fadd
%add = fadd reassoc float %0, %1
$ ./clang -O2 27372.c -S -o - -fassociative-math -fno-signed-zeros -fno-trapping-math | grep xmm
addss %xmm1, %xmm0
subss %xmm1, %xmm0
Differential Revision: https://reviews.llvm.org/D39812
llvm-svn: 320920
Changes to the original scalar loop during LV code gen cause the return value
of Legal->isConsecutivePtr() to be inconsistent with the return value during
legal/cost phases (further analysis and information of the bug is in D39346).
This patch is an alternative fix to PR34965 following the CM_Widen approach
proposed by Ayal and Gil in D39346. It extends InstWidening enum with
CM_Widen_Reverse to properly record the widening decision for consecutive
reverse memory accesses and, consequently, get rid of the
Legal->isConsetuviePtr() call in LV code gen. I think this is a simpler/cleaner
solution to PR34965 than the one in D39346.
Fixes PR34965.
Patch by Diego Caballero, thanks!
Differential Revision: https://reviews.llvm.org/D40742
llvm-svn: 320913
r307148 added an assembly mnemonic spelling correction support and enabled it
on ARM. This enables that support on PowerPC as well.
Patch by Dmitry Venikov, thanks!
Differential Revision: https://reviews.llvm.org/D40552
llvm-svn: 320911
This mimics FileCheck's --check-prefixes option.
The default prefix is "expected". That is, "-verify" is equivalent to
"-verify=expected".
The goal is to permit exercising a single test suite source file with different
compiler options producing different sets of diagnostics. While cpp can be
combined with the existing -verify to accomplish the same goal, source is often
easier to maintain when it's not cluttered with preprocessor directives or
duplicate passages of code. For example, this patch also rewrites some existing
clang tests to demonstrate the benefit of this feature.
Patch by Joel E. Denny, thanks!
Differential Revision: https://reviews.llvm.org/D39694
llvm-svn: 320908
From working on lld I've learned this is generally the
preferred way for several reasons (e.g. more concise, improves
encapsulation).
Differential Revision: https://reviews.llvm.org/D41265
llvm-svn: 320906
Summary:
1. Use stream 0 only for combined module. Previously if combined module was not
processes ThinLTO used the stream for own output. However small changes in input,
could trigger combined module and shuffle outputs making life of llvm::LTO harder.
2. Always process combined module and write output to stream 0. Processing empty
combined module is cheap and allows llvm::LTO users to avoid implementing processing
which is already done in llvm::LTO.
Subscribers: mehdi_amini, inglorion, eraman, hiraditya
Differential Revision: https://reviews.llvm.org/D41267
llvm-svn: 320905
The frontend currently groups diagnostics from the command line according to
diagnostic level, but that places all notes last. Fix that by emitting such
diagnostics in the order they were generated.
Patch by Joel E. Denny, thanks!
Differential Revision: https://reviews.llvm.org/D40995
llvm-svn: 320904
r320895 modified a test so that it needs -enable-import-metadata which
is false by default for NDEBUG, found another place that needs this
added.
llvm-svn: 320903
This patch introduces a specialized way to lower overflow-checked
multiplications with mixed-sign operands. This fixes link failures and
ICEs on code like this:
void mul(int64_t a, uint64_t b) {
int64_t res;
__builtin_mul_overflow(a, b, &res);
}
The generic checked-binop irgen would use a 65-bit multiplication
intrinsic here, which requires runtime support for _muloti4 (128-bit
multiplication), and therefore fails to link on i386. To get an ICE
on x86_64, change the example to use __int128_t / __uint128_t.
Adding runtime and backend support for 65-bit or 129-bit checked
multiplication on all of our supported targets is infeasible.
This patch solves the problem by using simpler, specialized irgen for
the mixed-sign case.
llvm.org/PR34920, rdar://34963321
Testing: Apart from check-clang, I compared the output from this fairly
comprehensive test driver using unpatched & patched clangs:
https://gist.github.com/vedantk/3eb9c88f82e5c32f2e590555b4af5081
Differential Revision: https://reviews.llvm.org/D41149
llvm-svn: 320902
When unsafe algerbra is allowed calls to cabs(r) can be replaced by:
sqrt(creal(r)*creal(r) + cimag(r)*cimag(r))
Patch by Paul Walker, thanks!
Differential Revision: https://reviews.llvm.org/D40069
llvm-svn: 320901
This is a small step forward to move VPlan stuff to where it should belong (i.e., VPlan.*):
1. VP*Recipe classes in LoopVectorize.cpp are moved to VPlan.h.
2. Many of VP*Recipe::print() and execute() definitions are still left in
LoopVectorize.cpp since they refer to things declared in LoopVectorize.cpp. To
be moved to VPlan.cpp at a later time.
3. InterleaveGroup class is moved from anonymous namespace to llvm namespace.
Referencing it in anonymous namespace from VPlan.h ended up in warning.
Patch by Hideki Saito, thanks!
Differential Revision: https://reviews.llvm.org/D41045
llvm-svn: 320900
/debug and /debug:dwarf are orthogonal. An object file can contain both
CodeView and DWARF debug info, so the combination of /debug:dwarf and
/debug should generate both DWARF and a PDB, rather than /debug:dwarf
always suppressing PDB creation.
/nopdb is now redundant and can be removed. /debug /nopdb was previously
used to support DWARF, but specifying /debug:dwarf is entirely
equivalent to that combination now.
Differential Revision: https://reviews.llvm.org/D41310
llvm-svn: 320896
Summary:
This implements a missing feature to allow importing of aliases, which
was previously disabled because alias cannot be available_externally.
We instead import an alias as a copy of its aliasee.
Some additional work was required in the IndexBitcodeWriter for the
distributed build case, to ensure that the aliasee has a value id
in the distributed index file (i.e. even when it is not being
imported directly).
This is a performance win in codes that have many aliases, e.g. C++
applications that have many constructor and destructor aliases.
Reviewers: pcc
Subscribers: mehdi_amini, inglorion, eraman, llvm-commits
Differential Revision: https://reviews.llvm.org/D40747
llvm-svn: 320895
We can just pass multiple options to hasArgs (which will check for any
of those options being present) instead of calling it multiple times.
llvm-svn: 320892
Prior to this patch, a predicate wouldn't make sense outside of its
rule. Indeed, it was only during emitting a rule that a predicate would
be made aware of the IDs of the data it is checking. Because of that,
predicates could not be moved around or compared between each other.
NFC.
llvm-svn: 320887
Adds missing support for DW_FORM_data16.
Update of r320852, fixing the unittest to use a hand-coded struct
instead of std::array to guarantee data layout.
Differential Revision: https://reviews.llvm.org/D41090
llvm-svn: 320886
"Default" is a valid QoS for a thread on older versions of macOS,
like the one installed in the bot.
Thanks to Jason Molenda for helping me figuring out the problem.
<rdar://problem/28346273>
llvm-svn: 320883
Remove the unused setModule() function; it would be dangerous if someone
actually used it as it wouldn't reset/recompute various other module
related data.
llvm-svn: 320881
This is primarily to reduce stack usage, but ordering the use queue
according to the position in the code (earlier instructions visited
before later ones) reduces the number of unnecessary bottoms due to
visiting instructions out of order, e.g.
%reg1 = copy %reg0
%reg2 = copy %reg0
%reg3 = and %reg1, %reg2
Here, reg3 should be known to be same as reg0-2, but if reg3 is
evaluated after reg1 is updated, but before reg2 is updated, the two
inputs to the and will appear different, causing reg3 to become
bottom.
llvm-svn: 320866
This seemed to work due to a quirk in the X86 MC encoder that didn't emit a REX byte that the AND64ri8 implies when in 32-bit mode. This made the encoding the same as AND32ri8. I tried to add an assert to catch the dropped REX prefix that caught this.
llvm-svn: 320864
The target independent nodes will get legalized to the target specific nodes by their own legalization process. Someday I'd like to stop using a target specific for zero extends and truncates of legal types so the less places we reference the target specific opcode the better.
llvm-svn: 320863
When I wrote it I thought we were missing a potential optimization for KNL. But investigating further shows that for KNL we still do the optimal thing by widening to v4f32 and then using special isel patterns to widen again to zmm a register.
llvm-svn: 320862
The COFF linker automatically sets the IMAGE_DLL_CHARACTERISTICS_NO_SEH
when suitable, similarly to link.exe.
Differential Revision: https://reviews.llvm.org/D41275
llvm-svn: 320861
Sanjay Patel