Commit Graph

8339 Commits

Author SHA1 Message Date
Michael Liao
9a508ef64a [PATCH] Fix VGATHER* operand constraints
Add earlyclobber constaints to prevent input register being allocated as
the output register because, according to Intel spec [1], "If any pair
of the index, mask, or destination registers are the same, this
instruction results a UD fault."

---
[1] http://software.intel.com/sites/default/files/319433-014.pdf



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183327 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-05 18:12:26 +00:00
Tom Stellard
ad7ecc65b1 R600: Make sure to schedule AR register uses and defs in the same clause
Reviewed-by: vljn at ovi.com

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183294 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-05 03:43:06 +00:00
Rafael Espindola
6afb65c2b7 Revert "R600: Add a pass that merge Vector Register"
This reverts commit r183279. CodeGen/R600/texture-input-merge.ll was failing.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183286 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-05 01:48:30 +00:00
Vincent Lejeune
bbbdba891b R600: Add a pass that merge Vector Register
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183279 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 23:17:26 +00:00
Vincent Lejeune
e67a4afb5d R600: Const/Neg/Abs can be folded to dot4
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183278 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 23:17:15 +00:00
Evan Cheng
00ed010d9e Cortex-R5 can issue Thumb2 integer division instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183275 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 22:52:09 +00:00
David Majnemer
35e7751af4 ARM: Fix crash in ARM backend inside of ARMConstantIslandPass
The ARM backend did not expect LDRBi12 to hold a constant pool operand.
Allow for LLVM to deal with the instruction similar to how it deals with
LDRi12.

This fixes PR16215.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183238 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 17:46:15 +00:00
Vincent Lejeune
98017a015b R600: Swizzle texture/export instructions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183229 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 15:04:53 +00:00
Vincent Lejeune
9328438329 R600: Add a test for r183108
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183228 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-04 15:03:35 +00:00
Tom Stellard
e5fcc0dee4 R600/SI: Add support for work item and work group intrinsics
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183138 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 17:40:18 +00:00
Tom Stellard
e7397ee81a R600/SI: Add a calling convention for compute shaders
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183137 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 17:40:11 +00:00
Tom Stellard
e86f9d70ca R600/SI: Custom lower i64 sign_extend
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183136 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 17:40:03 +00:00
Tom Stellard
132183510f R600/SI: Add support for global loads
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183131 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 17:39:43 +00:00
Vincent Lejeune
96fe0be43b R600: use capital letter for PV channel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183107 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 15:44:35 +00:00
Venkatraman Govindaraju
e7cbb792c9 Sparc: Add support for indirect branch and blockaddress in Sparc backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183094 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 05:58:33 +00:00
Venkatraman Govindaraju
85cc972a06 Sparc: When storing 0, use %g0 directly in the store instruction instead of
using two instructions (sethi and store).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183090 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-03 00:21:54 +00:00
Venkatraman Govindaraju
65ca7aa57d Sparc: Combine add/or/sethi instruction with restore if possible.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183088 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-02 21:48:17 +00:00
Venkatraman Govindaraju
dd48226b15 Sparc: Perform leaf procedure optimization by default
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183083 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-02 02:24:27 +00:00
Venkatraman Govindaraju
a0b34d6c4a Sparc: Mark functions calling llvm.vastart and llvm.returnaddress intrinsics as non-leaf functions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183079 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-01 20:42:48 +00:00
Tim Northover
3ba14fab1b Revert r183069: "TMP: LEA64_32r fixing"
Very sorry, it was committed from the wrong branch by mistake.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183070 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-01 10:23:46 +00:00
Tim Northover
4d3ace4da0 TMP: LEA64_32r fixing
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183069 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-01 10:21:54 +00:00
Tim Northover
85c622d6b6 X86: change MOV64ri64i32 into MOV32ri64
The MOV64ri64i32 instruction required hacky MCInst lowering because it
was allocated as setting a GR64, but the eventual instruction ("movl")
only set a GR32. This converts it into a so-called "MOV32ri64" which
still accepts a (appropriate) 64-bit immediate but defines a GR32.
This is then converted to the full GR64 by a SUBREG_TO_REG operation,
thus keeping everyone happy.

This fixes a typo in the opcode field of the original patch, which
should make the legact JIT work again (& adds test for that problem).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183068 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-01 09:55:14 +00:00
Venkatraman Govindaraju
72ad17c48c [Sparc] Generate correct code for leaf functions with stack objects
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183067 91177308-0d34-0410-b5e6-96231b3b80d8
2013-06-01 04:51:18 +00:00
Eric Christopher
34431085de Temporarily Revert "X86: change MOV64ri64i32 into MOV32ri64" as it
seems to have caused PR16192 and other JIT related failures.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183059 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 23:30:45 +00:00
Quentin Colombet
5b00f4edcb Modify how the formulae are rated in Loop Strength Reduce.
Namely, check if the target allows to fold more that one register in the
addressing mode and if yes, adjust the cost accordingly.

Prior to this commit, reg1 + scale * reg2 accesses were artificially preferred
to reg1 + reg2 accesses. Indeed, the cost model wrongly assumed that reg1 + reg2
needs a temporary register for the computation, whereas it was correctly
estimated for reg1 + scale * reg2.

<rdar://problem/13973908>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183021 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 17:20:29 +00:00
Richard Sandiford
b6606e46ab [SystemZ] Don't use LOAD and STORE REVERSED for volatile accesses
Unlike most -- hopefully "all other", but I'm still checking -- memory
instructions we support, LOAD REVERSED and STORE REVERSED may access
the memory location several times.  This means that they are not suitable
for volatile loads and stores.

This patch is a prerequisite for better atomic load and store support.
The same principle applies there: almost all memory instructions we
support are inherently atomic ("block concurrent"), but LOAD REVERSED
and STORE REVERSED are exceptions.

Other instructions continue to allow volatile operands.  I will add
positive "allows volatile" tests at the same time as the "allows atomic
load or store" tests.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@183002 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 13:25:22 +00:00
Justin Holewinski
5443e7d790 [NVPTX] Re-enable support for virtual registers in the final output
Now that 3.3 is branched, we are re-enabling virtual registers to help
iron out bugs before the next release. Some of the post-RA passes do
not play well with virtual registers, so we disable them for now. The
needed functionality of the PrologEpilogInserter pass is copied to a
new backend-specific NVPTXPrologEpilog pass.

The test for this commit is not breaking the existing tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182998 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 12:14:49 +00:00
Tim Northover
43887bf3e6 X86: change MOV64ri64i32 into MOV32ri64
The MOV64ri64i32 instruction required hacky MCInst lowering because it was
allocated as setting a GR64, but the eventual instruction ("movl") only set a
GR32. This converts it into a so-called "MOV32ri64" which still accepts a
(appropriate) 64-bit immediate but defines a GR32. This is then converted to
the full GR64 by a SUBREG_TO_REG operation, thus keeping everyone happy.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182991 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 09:57:13 +00:00
Akira Hatanaka
affed7e11d [mips] Big-endian code generation for atomic instructions.
Patch by Jyun-Yan You.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182984 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-31 03:25:44 +00:00
Rafael Espindola
9e3e730417 Revert r182937 and r182877.
r182877 broke MCJIT tests on ARM and r182937 was working around another failure
by r182877.

This should make the ARM bots green.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182960 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 20:37:52 +00:00
Benjamin Kramer
3a36cb0805 Force a triple so we don't get bitten by windows' different regalloc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182935 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 15:39:35 +00:00
Benjamin Kramer
051ed0ae2d Force fragile test to the atom scheduler model.
The pattern the test originally checked for doesn't occur on other -mcpu
settings. On atom it's still there though slightly differently scheduled.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182933 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 15:22:28 +00:00
Tim Northover
5b1552548a X86: allow registers 8-15 in test
This test was failing on some hosts when an unexpected register was used for a
variable. This just extends the regexp to allow the new x86-64 registers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182929 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 13:56:32 +00:00
Tim Northover
15983b80a0 X86: use sub-register sequences for MOV*r0 operations
Instead of having a bunch of separate MOV8r0, MOV16r0, ... pseudo-instructions,
it's better to use a single MOV32r0 (which will expand to "xorl %reg, %reg")
and obtain other sizes with EXTRACT_SUBREG and SUBREG_TO_REG. The encoding is
smaller and partial register updates can sometimes be avoided.

Until recently, this sequence was a barrier to rematerialization though. That
should now be fixed so it's an appropriate time to make the change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182928 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 13:19:42 +00:00
Justin Holewinski
d5c52f1d76 [NVPTX] Fix case where a sext load of an i1 type may produce an
ld.u1 instead of an ld.u8.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182924 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 12:22:39 +00:00
Richard Sandiford
14a926f13b [SystemZ] Enable unaligned accesses
The code to distinguish between unaligned and aligned addresses was
already there, so this is mostly just a switch-on-and-test process.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182920 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 09:45:42 +00:00
Rafael Espindola
7486d92a6c Change how we iterate over relocations on ELF.
For COFF and MachO, sections semantically have relocations that apply to them.
That is not the case on ELF.

In relocatable objects (.o), a section with relocations in ELF has offsets to
another section where the relocations should be applied.

In dynamic objects and executables, relocations don't have an offset, they have
a virtual address. The section sh_info may or may not point to another section,
but that is not actually used for resolving the relocations.

This patch exposes that in the ObjectFile API. It has the following advantages:

* Most (all?) clients can handle this more efficiently. They will normally walk
all relocations, so doing an effort to iterate in a particular order doesn't
save time.

* llvm-readobj now prints relocations in the same way the native readelf does.

* probably most important, relocations that don't point to any section are now
visible. This is the case of relocations in the rela.dyn section. See the
updated relocation-executable.test for example.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182908 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 03:05:14 +00:00
Bill Wendling
8a65e11c9f This testcase tests command line attributes which we don't yet support.
In fact, we're probably going to support these flags in completely different
ways. So this test is no longer valid.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182899 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-30 00:32:04 +00:00
Andrew Trick
6e0b2a0cb0 Order CALLSEQ_START and CALLSEQ_END nodes.
Fixes PR16146: gdb.base__call-ar-st.exp fails after
pre-RA-sched=source fixes.

Patch by Xiaoyi Guo!

This also fixes an unsupported dbg.value test case. Codegen was
previously incorrect but the test was passing by luck.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182885 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 22:03:55 +00:00
JF Bastien
f567a6d39b Enable FastISel on ARM for Linux and NaCl
FastISel was only enabled for iOS ARM and Thumb2, this patch enables it
for ARM (not Thumb2) on Linux and NaCl.

Thumb2 support needs a bit more work, mainly around register class
restrictions.

The patch punts to SelectionDAG when doing TLS relocation on non-Darwin
targets. I will fix this and other FastISel-to-SelectionDAG failures in
a separate patch.

The patch also forces FastISel to retain frame pointers: iOS always
keeps them for backtracking (so emitted code won't change because of
this), but Linux was getting much worse code that was incorrect when
using big frames (such as test-suite's lencod). I'll also fix this in a
later patch, it will probably require a peephole so that FastISel
doesn't rematerialize frame pointers back-to-back.

The test changes are straightforward, similar to:
  http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130513/174279.html
They also add a vararg test that got dropped in that change.

I ran all of test-suite on A15 hardware with --optimize-option=-O0 and
all the tests pass.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182877 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 20:38:10 +00:00
Tim Northover
aae0fa998a Teach ReMaterialization to be more cunning about subregisters
This allows rematerialization during register coalescing to handle
more cases involving operations like SUBREG_TO_REG which might need to
be rematerialized using sub-register indices.

For example, code like:
    v1(GPR64):sub_32 = MOVZ something
    v2(GPR64) = COPY v1(GPR64)
should be convertable to:
    v2(GPR64):sub_32 = MOVZ something

but previously we just gave up in places like this

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182872 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 19:32:06 +00:00
Richard Sandiford
598377060d [SystemZ] Two tests missing from previous commit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182847 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 11:59:26 +00:00
Richard Sandiford
2d664abbfc [SystemZ] Immediate compare-and-branch support
This patch adds support for the CIJ and CGIJ instructions.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182846 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 11:58:52 +00:00
Venkatraman Govindaraju
5300869256 [Sparc] Add support for leaf functions in sparc backend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182822 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-29 04:46:31 +00:00
Richard Sandiford
d50bcb2162 [SystemZ] Register compare-and-branch support
This patch adds support for the CRJ and CGRJ instructions.  Support for
the immediate forms will be a separate patch.

The architecture has a large number of comparison instructions.  I think
it's generally better to concentrate on using the "best" comparison
instruction first and foremost, then only use something like CRJ if
CR really was the natual choice of comparison instruction.  The patch
therefore opportunistically converts separate CR and BRC instructions
into a single CRJ while emitting instructions in ISelLowering.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182764 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-28 10:41:11 +00:00
Preston Gurd
b704d23062 Convert sqrt functions into sqrt instructions when -ffast-math is in effect.
When -ffast-math is in effect (on Linux, at least), clang defines
__FINITE_MATH_ONLY__ > 0 when including <math.h>. This causes the
preprocessor to include <bits/math-finite.h>, which renames the sqrt functions.
For instance, "sqrt" is renamed as "__sqrt_finite". 

This patch adds the 3 new names in such a way that they will be treated
as equivalent to their respective original names.





git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182739 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-27 15:44:35 +00:00
Rafael Espindola
f594e41ae9 Add a cpu to try to bring back the atom bots.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182734 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-27 13:22:52 +00:00
Hal Finkel
1907cad7c8 Prefer to duplicate PPC Altivec loads when expanding unaligned loads
When expanding unaligned Altivec loads, we use the decremented offset trick to
prevent page faults. Unfortunately, if we have a sequence of consecutive
unaligned loads, this leads to suboptimal code generation because the 'extra'
load from the first unaligned load can be combined with the base load from the
second (but only if the decremented offset trick is not used for the first).
Search up and down the chain, through loads and token factors, looking for
consecutive loads, and if one is found, don't use the offset reduction trick.
These duplicate loads are later combined to yield the desired sequence (in the
future, we might want a more-powerful chain search, but that will require some
changes to allow the combiner routines to access the AA object).

This should complete the initial implementation of the optimized unaligned
Altivec load expansion. There is some refactoring that should be done, but
that will happen when the unaligned store expansion is added.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182719 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-26 18:08:30 +00:00
Andrew Trick
9edb37feb5 Fix PR16143: Insert DEBUG_VALUE before terminator.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182717 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-26 08:58:50 +00:00
Hal Finkel
5a0e60425f PPC: Combine duplicate (offset) lvsl Altivec intrinsics
The lvsl permutation control instruction is a function only of the alignment of
the pointer operand (relative to the 16-byte natural alignment of Altivec
vectors). As a result, multiple lvsl intrinsics where the operands differ by a
multiple of 16 can be combined.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182708 91177308-0d34-0410-b5e6-96231b3b80d8
2013-05-25 04:05:05 +00:00