------------------------------------------------------------------------
r196588 | weimingz | 2013-12-06 09:56:48 -0800 (Fri, 06 Dec 2013) | 7 lines
Bug 18149: [AArch32] VSel instructions has no ARMCC field
The current peephole optimizing for compare inst assumes an instr that
uses CPSR has an MO for ARM Cond code.However, for VSEL instructions
(vseqeq, vselgt, vselgt, vselvs), there is no such operand nor do
they support the modification of Cond Code.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196704 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196456 | jiangning | 2013-12-04 18:12:01 -0800 (Wed, 04 Dec 2013) | 2 lines
For AArch64, add missing register cost calculation for big value types like v4i64 and v8i64.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196700 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196391 | hliao | 2013-12-04 09:44:22 -0800 (Wed, 04 Dec 2013) | 5 lines
[X86] Check YMM31/ZMM31 as well
- No test case as there's no calling convention preserve YMM31/ZMM31 only
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196653 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196261 | hliao | 2013-12-03 01:17:32 -0800 (Tue, 03 Dec 2013) | 13 lines
Enhance the fix of PR17631
- The fix to PR17631 fixes part of the cases where 'vzeroupper' should
not be issued before 'call' insn. There're other cases where helper
calls will be inserted not limited to epilog. These helper calls do
not follow the standard calling convention and won't clobber any YMM
registers. (So far, all call conventions will clobber any or part of
YMM registers.)
This patch enhances the previous fix to cover more cases 'vzerosupper' should
not be inserted by checking if that function call won't clobber any YMM
registers and skipping it if so.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196652 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196267 | rsandifo | 2013-12-03 11:01:54 +0000 (Tue, 03 Dec 2013) | 12 lines
[SystemZ] Fix choice of known-zero mask in insertion optimization
The backend converts 64-bit ORs into subreg moves if the upper 32 bits
of one operand and the low 32 bits of the other are known to be zero.
It then tries to peel away redundant ANDs from the upper 32 bits.
Since AND masks are canonicalized to exclude known-zero bits,
the test ORs the mask and the known-zero bits together before
checking for redundancy. The problem was that it was using the
wrong node when checking for known-zero bits, so could drop ANDs
that were still needed.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196268 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196046 | tnorthover | 2013-12-01 06:16:24 -0800 (Sun, 01 Dec 2013) | 8 lines
ARM: fix bug in -Oz stack adjustment folding
Previously, we clobbered callee-saved registers when folding an "add
sp, #N" into a "pop {rD, ...}" instruction. This change checks whether
a register we're going to add to the "pop" could actually be live
outside the function before doing so and should fix the issue.
This should fix PR18081.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196074 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195401 | lhames | 2013-11-21 16:46:32 -0800 (Thu, 21 Nov 2013) | 8 lines
Fix a typo where we were creating <def,kill> operands instead of
<def,dead> ones.
Add an assertion to make sure we catch this in the future.
Fixes <rdar://problem/15464559>.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196073 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r196044 | d0k | 2013-12-01 03:47:42 -0800 (Sun, 01 Dec 2013) | 6 lines
Revamp error checking in the ms inline asm parser.
- Actually abort when an error occurred.
- Check that the frontend lookup worked when parsing length/size/type operators.
Tested by a clang test. PR18096.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196070 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195973 | dsanders | 2013-11-30 13:47:57 +0000 (Sat, 30 Nov 2013) | 5 lines
[mips][msa] MSA loads and stores have a 10-bit offset. Account for this when lowering FrameIndex.
This prevents the compiler from emitting invalid ld.[bhwd]'s and st.[bhwd]'s
when the stack frame is between 512 and 32,768 bytes in size.
------------------------------------------------------------------------
Review of this commit by Matheus Almeida revealed that it is still possible to
emit invalid code (when the offset is not a multiple of the element size).
However, we agreed that this commit still represents an improvement since it
fixes many cases that previously emitted invalid code, and does not cause any
cases that previously emitted valid code to emit invalid code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196049 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195972 | dsanders | 2013-11-30 13:15:21 +0000 (Sat, 30 Nov 2013) | 5 lines
[mips][msa] A small refactor to reduce patch noise in my next commit
No functional change. An if-statement has been split into two nested if-statements.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196047 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195975 | zjovanovic | 2013-11-30 19:12:28 +0000 (Sat, 30 Nov 2013) | 1 line
Fixed issue with microMIPS long branch.
------------------------------------------------------------------------
r195976 | zjovanovic | 2013-11-30 19:13:15 +0000 (Sat, 30 Nov 2013) | 1 line
Test case for issue with microMIPS long branch.
------------------------------------------------------------------------
To expand on those commit messages:
The immediate in a MIPS branch is multiplied by the instruction size before use
as an offset. For many MIPS ISA's this is 4 bytes, but for microMIPS it is 2
bytes. This commit corrects the scale factor used for microMIPS so that attempts
to use large offsets result in a valid sequence of instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196043 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195941 | haoliu | 2013-11-28 18:11:22 -0800 (Thu, 28 Nov 2013) | 4 lines
AArch64: The pattern match should check the range of the immediate value.
Or we can generate some illegal instructions.
E.g. shrn2 v0.4s, v1.2d, #35. The legal range should be in [1, 16].
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196033 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195905 | jiangning | 2013-11-27 17:34:55 -0800 (Wed, 27 Nov 2013) | 3 lines
Remove the variable only used by assert to avoid the build failure
caused by build options [-Werror,-Wunused-variable].
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@196026 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195879 | tstellar | 2013-11-27 13:23:29 -0800 (Wed, 27 Nov 2013) | 6 lines
R600/SI: Use SGPR_32 register class for 32-bit SMRD outputs
Writing to the M0 register from an SMRD instruction hangs the GPU, so
we need to use the SGPR_32 register class, which does not include M0.
NOTE: This is a candidate for the 3.4 branch.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195999 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195878 | tstellar | 2013-11-27 13:23:20 -0800 (Wed, 27 Nov 2013) | 3 lines
R600: Add support for ISD::FROUND
NOTE: This is a candidate for the 3.4 branch.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195998 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195576 | venkatra | 2013-11-24 12:23:25 -0800 (Sun, 24 Nov 2013) | 2 lines
[Sparc] Emit large negative adjustments to SP/FP with sethi+xor instead of sethi+or. This generates correct code for both sparc32 and sparc64.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195870 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195574 | venkatra | 2013-11-24 10:41:49 -0800 (Sun, 24 Nov 2013) | 2 lines
[SparcV9]: Do not emit .register directives for global registers that are clobbered by calls but not used in the function itself.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195868 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195779 | hliao | 2013-11-26 12:31:31 -0800 (Tue, 26 Nov 2013) | 7 lines
Fix PR18054
- Fix bug in (vsext (vzext x)) -> (vsext x) in SIGN_EXTEND_IN_REG
lowering where we need to check whether x is a vector type (in-reg
type) of i8, i16 or i32; otherwise, that optimization is not valid.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195821 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195424 | haoliu | 2013-11-22 00:47:22 -0800 (Fri, 22 Nov 2013) | 4 lines
Fix the bugs about AArch64 Load/Store vector types and bitcast between i64 and vector types.
e.g. "%tmp = load <2 x i64>* %ptr" can't be selected.
"%tmp = bitcast i64 %in to <2 x i32>" can't be selected.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195764 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195684 | rafael | 2013-11-25 12:50:03 -0800 (Mon, 25 Nov 2013) | 3 lines
Do the string comparison in the constructor instead of once per nop.
Thanks to Roman Divacky for the suggestion.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195746 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195456 | dsanders | 2013-11-22 05:22:52 -0800 (Fri, 22 Nov 2013) | 4 lines
Fix typo in a comment added in r195455.
Credit to Matheus Almeida for spotting it.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195743 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195364 | dsanders | 2013-11-21 08:11:31 -0800 (Thu, 21 Nov 2013) | 12 lines
[mips][msa] Fix a corner case in performORCombine() when combining nodes into VSELECT.
Mask == ~InvMask asserts if the width of Mask and InvMask differ.
The combine isn't valid (with two exceptions, see below) if the widths differ
so test for this before testing Mask == ~InvMask.
In the specific cases of Mask=~0 and InvMask=0, as well as Mask=0 and
InvMask=~0, the combine is still valid. However, there are more appropriate
combines that could be used in these cases such as folding x & 0 to 0, or
x & ~0 to x.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195741 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195343 | dsanders | 2013-11-21 03:40:14 -0800 (Thu, 21 Nov 2013) | 5 lines
[mips][msa/dsp] Only do DSP combines if DSP is enabled.
Fixes a crash (null pointer dereferenced) when MSA is enabled.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195740 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195731 | rsandifo | 2013-11-26 10:53:16 +0000 (Tue, 26 Nov 2013) | 7 lines
[SystemZ] Fix incorrect use of RISBG for a zero-extended right shift
We would wrongly transform the testcase into the equivalent of an AND with 1.
The problem was that, when testing whether the shifted-in bits of the right
shift were significant, we used the width of the final zero-extended result
rather than the width of the shifted value.
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195736 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195716 | kevinqin | 2013-11-25 19:26:47 -0800 (Mon, 25 Nov 2013) | 3 lines
Refactored the implementation of AArch64 NEON instruction ZIP, UZP
and TRN.
Fix a bug when mixed use of vget_high_u8() and vuzp_u8().
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195735 91177308-0d34-0410-b5e6-96231b3b80d8
------------------------------------------------------------------------
r195679 | rafael | 2013-11-25 12:15:14 -0800 (Mon, 25 Nov 2013) | 12 lines
Don't use nopl in cpus that don't support it.
Patch by Mikulas Patocka. I added the test. I checked that for cpu names that
gas knows about, it also doesn't generate nopl.
The modified cpus:
i686 - there are i686-class CPUs that don't have nopl: Via c3, Transmeta
Crusoe, Microsoft VirtualBox - see
https://bbs.archlinux.org/viewtopic.php?pid=775414
k6, k6-2, k6-3, winchip-c6, winchip2 - these are 586-class CPUs
via c3 c3-2 - see https://bugs.archlinux.org/task/19733 as a proof that
Via c3 and c3-Nehemiah don't have nopl
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195730 91177308-0d34-0410-b5e6-96231b3b80d8
