RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-23 13:02:52 +02:00
Code Issues Projects Releases Wiki Activity
111,345 Commits 4 Branches 5 Tags 704 MiB
576ceb22ba
Commit Graph

14 Commits

Author SHA1 Message Date
Benjamin Kramer
600e24a1cb SDAG: Legalize vector BSWAP into a shuffle if the shuffle is legal but the bswap not. 
- On ARM/ARM64 we get a vrev because the shuffle matching code is really smart. We still unroll anything that's not v4i32 though.
- On X86 we get a pshufb with SSSE3. Required more cleverness in isShuffleMaskLegal.
- On PPC we get a vperm for v8i16 and v4i32. v2i64 is unrolled.

llvm-svn: 209123
 2014-05-19 13:12:38 +00:00
Saleem Abdulrasool
53e8c5a4af ARM: fixup more tests to specify the target more explicitly 
This changes the tests that were targeting ARM EABI to explicitly specify the
environment rather than relying on the default.  This breaks with the new
Windows on ARM support when running the tests on Windows where the default
environment is no longer EABI.

Take the opportunity to avoid a pointless redirect (helps when trying to debug
with providing a command line invocation which can be copy and pasted) and
removing a few greps in favour of FileCheck.

llvm-svn: 205541
 2014-04-03 16:01:44 +00:00
Stephen Lin
3ae734a60c Convert CodeGen/*/*.ll tests to use the new CHECK-LABEL for easier debugging. No functionality change and all tests pass after conversion. 
This was done with the following sed invocation to catch label lines demarking function boundaries:
    sed -i '' "s/^;\( *\)\([A-Z0-9_]*\):\( *\)test\([A-Za-z0-9_-]*\):\( *\)$/;\1\2-LABEL:\3test\4:\5/g" test/CodeGen/*/*.ll
which was written conservatively to avoid false positives rather than false negatives. I scanned through all the changes and everything looks correct.

llvm-svn: 186258
 2013-07-13 20:38:47 +00:00
Chad Rosier
a588421976 When performing a truncating store, it's possible to rearrange the data 
in-register, such that we can use a single vector store rather then a 
series of scalar stores.

For func_4_8 the generated code

	vldr	d16, LCPI0_0
	vmov	d17, r0, r1
	vadd.i16	d16, d17, d16
	vmov.u16	r0, d16[3]
	strb	r0, [r2, #3]
	vmov.u16	r0, d16[2]
	strb	r0, [r2, #2]
	vmov.u16	r0, d16[1]
	strb	r0, [r2, #1]
	vmov.u16	r0, d16[0]
	strb	r0, [r2]
	bx	lr

becomes

	vldr	d16, LCPI0_0
	vmov	d17, r0, r1
	vadd.i16	d16, d17, d16
	vuzp.8	d16, d17
	vst1.32	{d16[0]}, [r2, :32]
	bx	lr

I'm not fond of how this combine pessimizes 2012-03-13-DAGCombineBug.ll,
but I couldn't think of a way to judiciously apply this combine.

This

	ldrh	r0, [r0, #4]
	strh	r0, [r1]

becomes

	vldr	d16, [r0]
	vmov.u16	r0, d16[2]
	vmov.32	d16[0], r0
	vuzp.16	d16, d17
	vst1.32	{d16[0]}, [r1, :32]

PR11158
rdar://10703339

llvm-svn: 154340
 2012-04-09 20:32:02 +00:00
Nadav Rotem
b03ef2cd0b Add CHECKs and document PR11158. 
llvm-svn: 142240
 2011-10-17 20:23:23 +00:00
Nadav Rotem
8ed6f090ef Enable element promotion type legalization by deafault. 
Changed tests which assumed that vectors are legalized by widening them.

llvm-svn: 142152
 2011-10-16 20:31:33 +00:00
Tanya Lattner
6814933ea6 Handle perfect shuffle case that generates a vrev for vectors of floats. 
Add test case.

llvm-svn: 131582
 2011-05-18 21:44:54 +00:00
Tanya Lattner
06cb9cbf98 In r131488 I misunderstood how VREV works. It splits the vector in half and splits each half. Therefore, the real problem was that we were using a VREV64 for a 4xi16, when we should have been using a VREV32. 
Updated test case and reverted change to the PerfectShuffle Table.

llvm-svn: 131529
 2011-05-18 06:42:21 +00:00
Tanya Lattner
7145d69427 vrev is incorrectly defined in the perfect shuffle table. The ordering is backwards (should be 0x3210 versus 0x1032) which exposed a bug when doing a shuffle on a 4xi16. I've attached a test case. 
llvm-svn: 131488
 2011-05-17 20:48:40 +00:00
Bob Wilson
cdc8dff3ac SelectionDAG shuffle nodes do not allow operands with different numbers of 
elements than the result vector type.  So, when an instruction like:

%8 = shufflevector <2 x float> %4, <2 x float> %7, <4 x i32> <i32 1, i32 0, i32 3, i32 2>

is translated to a DAG, each operand is changed to a concat_vectors node that appends 2 undef elements.  That is:

shuffle [a,b], [c,d] is changed to:
shuffle [a,b,u,u], [c,d,u,u]

That's probably the right thing for x86 but for NEON, we'd much rather have:

shuffle [a,b,c,d], undef

Teach the DAG combiner how to do that transformation for ARM.  Radar 8597007.

llvm-svn: 117482
 2010-10-27 20:38:28 +00:00
Bob Wilson
6239dc42c6 Allow more cases of undef shuffle indices and add tests for them. 
llvm-svn: 111226
 2010-08-17 05:54:34 +00:00
Rafael Espindola
d7a63bead9 Remove arm_apcscc from the test files. It is the default and doing this 
matches what llvm-gcc and clang now produce.

llvm-svn: 106221
 2010-06-17 15:18:27 +00:00
Dan Gohman
142428ce64 Eliminate more uses of llvm-as and llvm-dis. 
llvm-svn: 81293
 2009-09-09 00:09:15 +00:00
Bob Wilson
ec256c8938 Add support for ARM Neon VREV instructions. 
Patch by Anton Korzh, with some modifications from me.

llvm-svn: 77101
 2009-07-26 00:39:34 +00:00