RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-28 14:32:51 +01:00
Code Issues Projects Releases Wiki Activity
40,389 Commits 4 Branches 5 Tags 704 MiB
cc784f1662
Commit Graph

3519 Commits

Author SHA1 Message Date
Dan Gohman
cc784f1662 Re-introduce the 8-bit subreg zext-inreg patterns for x86-32, 
this time using MOV32to32_ and MOV16to16_. Thanks to Evan for
suggesting this.

llvm-svn: 54418
 2008-08-06 18:27:21 +00:00
Dan Gohman
99d70043f9 xchg does not modify FLAGS. 
llvm-svn: 54411
 2008-08-06 15:52:50 +00:00
Evan Cheng
f4d1119fbd Fix PR2620: Fix X86cmppd selection code so it expects operands to be v2f64. 
llvm-svn: 54376
 2008-08-05 22:19:15 +00:00
Dan Gohman
5d0df78ae0 Add an assert to catch invalid VECTOR_SHUFFLE mask indices. 
llvm-svn: 54329
 2008-08-04 23:09:15 +00:00
Andrew Lenharth
377c046675 Add atomic sub for other sizes 
llvm-svn: 54314
 2008-08-03 20:17:34 +00:00
Dan Gohman
efb5d2ce6e Reapply r54147 with a constraint to only use the 8-bit 
subreg form on x86-64, to avoid the problem with x86-32
having GPRs that don't have 8-bit subregs.

Also, change several 16-bit instructions to use 
equivalent 32-bit instructions. These have a smaller
encoding and avoid partial-register updates.

llvm-svn: 54223
 2008-07-30 18:09:17 +00:00
Dan Gohman
ebe629a4b2 Revert 54147. 
llvm-svn: 54148
 2008-07-29 01:02:18 +00:00
Dan Gohman
1816900fd1 Add x86 isel patterns to match what would be a ZERO_EXTEND_INREG operation, 
which is represented in codegen as an 'and' operation. This matches them
with movz instructions, instead of leaving them to be matched by and
instructions with an immediate field.

llvm-svn: 54147
 2008-07-28 22:18:25 +00:00
Dan Gohman
9742f7772d Rename SDOperand to SDValue. 
llvm-svn: 54128
 2008-07-27 21:46:04 +00:00
Dan Gohman
47c5cdbc34 Tidy SDNode::use_iterator, and complete the transition to have it 
parallel its analogue, Value::value_use_iterator. The operator* method
now returns the user, rather than the use.

llvm-svn: 54127
 2008-07-27 20:43:25 +00:00
Nate Begeman
5523d40e4b Disable mov{L, LP, HP, HLP, *DUP} shuffles for mmx 
mmx needs its own fancy shuffle logic based on unpack; for now we get correct but awful code.

Also commit Mon Ping's VSETCC patch

llvm-svn: 54039
 2008-07-25 19:05:58 +00:00
Nate Begeman
730880eec2 Fit in 80 cols 
llvm-svn: 54029
 2008-07-25 17:34:41 +00:00
Nate Begeman
73efed7a4c Remove dead PatLeaf; there are a number of issues around MMX movl that need to be fixed. 
llvm-svn: 54026
 2008-07-25 17:25:04 +00:00
Evan Cheng
20c9cdbe69 Fix PR2485: do all 4-element SSE shuffles in max. of 2 shuffle instructions. 
Based on patch by Nicolas Capens.

llvm-svn: 53939
 2008-07-23 00:22:17 +00:00
Evan Cheng
ff0bd19937 Factor out SSE 4 wide shuffle lowering code into its own function. No functionality changes. 
llvm-svn: 53933
 2008-07-22 21:13:36 +00:00
Evan Cheng
901d469e05 Fix PR2574: implement v2f32 scalar_to_vector. 
llvm-svn: 53927
 2008-07-22 18:39:19 +00:00
Anton Korobeynikov
f13fbd6879 Fix encoding of atomic compare and swap for i64 
llvm-svn: 53911
 2008-07-22 16:22:48 +00:00
Evan Cheng
a2bb31372d Eliminate a compilation warning. 
llvm-svn: 53873
 2008-07-21 20:02:45 +00:00
Dan Gohman
b91bef08a7 Add titles to the various SelectionDAG viewGraph calls 
that include useful information like the name of the
block being viewed and the current phase of compilation.

llvm-svn: 53872
 2008-07-21 20:00:07 +00:00
Duncan Sands
6e31474e71 Add VerifyNode, a place to put sanity checks on 
generic SDNode's (nodes with their own constructors
should do sanity checking in the constructor).  Add
sanity checks for BUILD_VECTOR and fix all the places
that were producing bogus BUILD_VECTORs, as found by
"make check".  My favorite is the BUILD_VECTOR with
only two operands that was being used to build a
vector with four elements!

llvm-svn: 53850
 2008-07-21 10:20:31 +00:00
Evan Cheng
ffd51ccf6b Use movaps instead of movups to spill 16-byte vector values when default alignment is >= 16. This fixes some massive performance regressions. 
llvm-svn: 53844
 2008-07-21 06:34:17 +00:00
Bill Wendling
98b6e63176 Fix for first part of PR2562. Generate the "pinsrw" instruction for inserts 
into v4i16 vectors.

llvm-svn: 53807
 2008-07-20 02:32:23 +00:00
Anton Korobeynikov
449fb584e4 Fix a FIXME :) 
llvm-svn: 53789
 2008-07-19 13:15:46 +00:00
Anton Korobeynikov
5c0eb7e991 Use generic ELFTargetAsmInfo and DarwinTargetAsmInfo for X86 code 
llvm-svn: 53788
 2008-07-19 13:15:21 +00:00
Anton Korobeynikov
6e00357dd6 Use aligned stack spills, where possible. This fixes PR2549. 
llvm-svn: 53784
 2008-07-19 06:30:51 +00:00
Dan Gohman
8981962672 Add a new function, ReplaceAllUsesOfValuesWith, which handles bulk 
replacement of multiple values. This is slightly more efficient
than doing multiple ReplaceAllUsesOfValueWith calls, and theoretically
could be optimized even further. However, an important property of this
new function is that it handles the case where the source value set and
destination value set overlap. This makes it feasible for isel to use
SelectNodeTo in many very common cases, which is advantageous because
SelectNodeTo avoids a temporary node and it doesn't require CSEMap
updates for users of values that don't change position.

Revamp MorphNodeTo, which is what does all the work of SelectNodeTo, to
handle operand lists more efficiently, and to correctly handle a number
of corner cases to which its new wider use exposes it.

This commit also includes a change to the encoding of post-isel opcodes
in SDNodes; now instead of being sandwiched between the target-independent
pre-isel opcodes and the target-dependent pre-isel opcodes, post-isel
opcodes are now represented as negative values. This makes it possible
to test if an opcode is pre-isel or post-isel without having to know
the size of the current target's post-isel instruction set.

These changes speed up llc overall by 3% and reduce memory usage by 10%
on the InstructionCombining.cpp testcase with -fast and -regalloc=local.

llvm-svn: 53728
 2008-07-17 19:10:17 +00:00
Nate Begeman
64f8f7f6bb Remove unnecessary readme entry 
llvm-svn: 53722
 2008-07-17 17:21:14 +00:00
Nate Begeman
61f6c21028 Fix a typo in last commit 
llvm-svn: 53720
 2008-07-17 17:04:58 +00:00
Nate Begeman
af01bfff99 SSE codegen for vsetcc nodes 
llvm-svn: 53719
 2008-07-17 16:51:19 +00:00
Mon P Wang
57cd9d6e5a When lowering certain atomics, we need to copy the memoperand from the old 
atomic operation to the new one.

llvm-svn: 53714
 2008-07-17 04:54:06 +00:00
Devang Patel
a6c5ff690a Mark function used by asm block as used, otherwise optimizer may not see the use and may delete the function. 
llvm-svn: 53692
 2008-07-16 17:54:34 +00:00
Dan Gohman
4c8c8e3aad Fix the result type of X86's truncate to i8. 
llvm-svn: 53688
 2008-07-16 16:20:48 +00:00
Evan Cheng
face16f9d8 x86-64 PIC JIT fixes: do not generate the extra load for external GV's. 
llvm-svn: 53661
 2008-07-16 01:34:02 +00:00
Evan Cheng
cabfd3f78c X86-64 PIC jump table values are different from x86-32 cases, they are dest - table base. 
llvm-svn: 53660
 2008-07-16 01:33:08 +00:00
Dan Gohman
bf47a27643 Add a utility function to MachineInstr for testing whether an instruction 
has exactly one MachineMemOperand, and change some X86 lowering code to
make use of it.

llvm-svn: 53498
 2008-07-12 00:10:52 +00:00
Dan Gohman
4c18394001 Include a frame index in the "fixed stack" pseudo source value 
instead of using the frame index for the SVOffset, which was
inconsistent.

llvm-svn: 53486
 2008-07-11 22:44:52 +00:00
Bill Wendling
9f17caa9a9 The frame address on an x86-64 box needs to be offset by -8, not -4. 
llvm-svn: 53450
 2008-07-11 07:18:52 +00:00
Evan Cheng
02a618dc56 Fix for PR2472. Use movss to set lower 32-bits of a zero XMM vector. 
llvm-svn: 53386
 2008-07-10 01:08:23 +00:00
Anton Korobeynikov
9eae9520a9 Remove a FIXME: we really need to use const_data section on darwin for 
constant pool, if relocation model is not static. This directly maps to
the way how GCC works.

llvm-svn: 53370
 2008-07-09 21:54:26 +00:00
Anton Korobeynikov
a5955dc461 Add FIXME for future checking. 
llvm-svn: 53368
 2008-07-09 21:38:28 +00:00
Dale Johannesen
36a38a5ba1 Emit debug info for data-only files. This version 
is X86 ATT only.

llvm-svn: 53355
 2008-07-09 20:55:35 +00:00
Anton Korobeynikov
61f4175d64 Add missed section 
llvm-svn: 53354
 2008-07-09 20:47:55 +00:00
Anton Korobeynikov
57e9182691 Distinguish .const and .const_data on Darwin, when needed. This is somehow crazy :) 
llvm-svn: 53350
 2008-07-09 20:01:42 +00:00
Anton Korobeynikov
67931c35bd Weak stuff always goes to coalesced sections on Darwin 
llvm-svn: 53340
 2008-07-09 19:06:02 +00:00
Dan Gohman
8a421248b9 Remove #include <iostream>. 
llvm-svn: 53333
 2008-07-09 18:08:48 +00:00
Anton Korobeynikov
2b2543166d Add FIXME needed to be resolved later 
llvm-svn: 53324
 2008-07-09 13:30:02 +00:00
Anton Korobeynikov
32e4256260 Typo 
llvm-svn: 53322
 2008-07-09 13:29:27 +00:00
Anton Korobeynikov
d31e7ad0cf Revert accidentially added stuff 
llvm-svn: 53321
 2008-07-09 13:29:08 +00:00
Anton Korobeynikov
03614f247c First sketch of special section objects 
llvm-svn: 53320
 2008-07-09 13:28:49 +00:00
Anton Korobeynikov
395dac000b Honour text sections 
llvm-svn: 53319
 2008-07-09 13:28:19 +00:00