1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-29 23:12:55 +01:00
Commit Graph

24348 Commits

Author SHA1 Message Date
Evan Cheng
b7a2ab21a5 Add a new entry.
llvm-svn: 27963
2006-04-24 23:30:10 +00:00
Evan Cheng
3f495010ed Added a movq test case.
llvm-svn: 27962
2006-04-24 23:03:22 +00:00
Evan Cheng
0282b48ec2 Special case handling two wide build_vector(0, x).
llvm-svn: 27961
2006-04-24 22:58:52 +00:00
Evan Cheng
3306427d87 Some missing movlps, movhps, movlpd, and movhpd patterns.
llvm-svn: 27960
2006-04-24 21:58:20 +00:00
Evan Cheng
1eae7398a6 A little bit more build_vector enhancement for v8i16 cases.
llvm-svn: 27959
2006-04-24 18:01:45 +00:00
Evan Cheng
f74b046b06 Remove a completed entry.
llvm-svn: 27958
2006-04-24 17:38:16 +00:00
Chris Lattner
283f1992d4 Suggest looking at the SPARC backend. How weird is that?? :)
llvm-svn: 27957
2006-04-24 16:34:45 +00:00
Evan Cheng
679d1dd5cf Added addJumpTableIndex
llvm-svn: 27956
2006-04-24 06:42:15 +00:00
Evan Cheng
70237fcb5d MakeMIInst() should handle jump table index operands.
llvm-svn: 27955
2006-04-24 05:37:35 +00:00
Chris Lattner
86f1e02800 Add a note
llvm-svn: 27954
2006-04-23 19:47:09 +00:00
Evan Cheng
4812ce5035 MOVL shuffle (i.e. movd or movss / movsd from memory) of undef, V2 == V2
llvm-svn: 27953
2006-04-23 06:35:19 +00:00
Nate Begeman
7bb910dbe7 Fix the updating of the machine CFG when a PHI node was in a successor of
the jump table's range check block.  This re-enables 100% dense jump tables
by default on PPC & x86

llvm-svn: 27952
2006-04-23 06:26:20 +00:00
Jeff Cohen
6ae5da42b9 Keep Visual Studio informed.
llvm-svn: 27951
2006-04-23 04:37:08 +00:00
Nate Begeman
f86709e9e2 Code cleanup associated with jump tables, thanks to Chris for noticing
these.

llvm-svn: 27950
2006-04-22 23:52:35 +00:00
Nate Begeman
6cfb33a42c Turn of jump tables for a bit, there are still some issues to work out with
updating the machine CFG.

llvm-svn: 27949
2006-04-22 23:51:56 +00:00
Nate Begeman
0d74cbcb6b Optimized stores to the constant pool, while cool, are unnecessary.
llvm-svn: 27948
2006-04-22 22:31:45 +00:00
Nate Begeman
7ed816f900 JumpTable support! What this represents is working asm and jit support for
x86 and ppc for 100% dense switch statements when relocations are non-PIC.
This support will be extended and enhanced in the coming days to support
PIC, and less dense forms of jump tables.

llvm-svn: 27947
2006-04-22 18:53:45 +00:00
Evan Cheng
1c33e83af5 Don't do all the lowering stuff for 2-wide build_vector's. Also, minor optimization for shuffle of undef.
llvm-svn: 27946
2006-04-22 08:34:05 +00:00
Evan Cheng
ec33bd04fb Fix a performance regression. Use {p}shuf* when there are only two distinct elements in a build_vector.
llvm-svn: 27945
2006-04-22 06:21:46 +00:00
Evan Cheng
d0db6f1a7d Two more build_vector tests.
llvm-svn: 27944
2006-04-22 06:19:11 +00:00
Chris Lattner
de560fcaf7 Teach the JIT how to relocate LI, this fixes the JIT on Prolangs-C/TimberWolfMC
llvm-svn: 27943
2006-04-22 06:17:56 +00:00
Chris Lattner
59c6239ce7 This no longer needs the JIT.
llvm-svn: 27942
2006-04-22 05:04:23 +00:00
Chris Lattner
4f0240b577 Fix JIT support for static ctors, which was apparently completely broken!
This allows Prolangs-C++/city and probably a bunch of other stuff to work
well with the new front-end

llvm-svn: 27941
2006-04-22 05:02:46 +00:00
Evan Cheng
5cb5fdd8eb Revamp build_vector lowering to take advantage of movss and movd instructions.
movd always clear the top 96 bits and movss does so when it's loading the
value from memory.
The net result is codegen for 4-wide shuffles is much improved. It is near
optimal if one or more elements is a zero. e.g.

__m128i test(int a, int b) {
  return _mm_set_epi32(0, 0, b, a);
}

compiles to

_test:
	movd 8(%esp), %xmm1
	movd 4(%esp), %xmm0
	punpckldq %xmm1, %xmm0
	ret

compare to gcc:

_test:
	subl	$12, %esp
	movd	20(%esp), %xmm0
	movd	16(%esp), %xmm1
	punpckldq	%xmm0, %xmm1
	movq	%xmm1, %xmm0
	movhps	LC0, %xmm0
	addl	$12, %esp
	ret

or icc:

_test:
        movd      4(%esp), %xmm0                                #5.10
        movd      8(%esp), %xmm3                                #5.10
        xorl      %eax, %eax                                    #5.10
        movd      %eax, %xmm1                                   #5.10
        punpckldq %xmm1, %xmm0                                  #5.10
        movd      %eax, %xmm2                                   #5.10
        punpckldq %xmm2, %xmm3                                  #5.10
        punpckldq %xmm3, %xmm0                                  #5.10
        ret                                                     #5.10

There are still room for improvement, for example the FP variant of the above example:

__m128 test(float a, float b) {
  return _mm_set_ps(0.0, 0.0, b, a);
}

_test:
	movss 8(%esp), %xmm1
	movss 4(%esp), %xmm0
	unpcklps %xmm1, %xmm0
	xorps %xmm1, %xmm1
	movlhps %xmm1, %xmm0
	ret

The xorps and movlhps are unnecessary. This will require post legalizer optimization to handle.

llvm-svn: 27939
2006-04-21 23:03:30 +00:00
Nate Begeman
dc60393018 Fix the comment
llvm-svn: 27938
2006-04-21 22:11:27 +00:00
Nate Begeman
67b3094f27 Change the PPC JIT to use a Static relocation model
llvm-svn: 27937
2006-04-21 22:04:15 +00:00
Chris Lattner
e6f4065bca typo
llvm-svn: 27936
2006-04-21 21:37:40 +00:00
Chris Lattner
d81dcf9da4 fix thinko
llvm-svn: 27935
2006-04-21 21:05:22 +00:00
Chris Lattner
84a811d57e add some low-prio notes
llvm-svn: 27934
2006-04-21 21:03:21 +00:00
Chris Lattner
8f7c394f3a The BFS scheduler is apparently nondeterminstic (causes many llvmgcc bootstrap
miscompares).  Switch RISC targets to use the list-td scheduler, which isn't.

llvm-svn: 27933
2006-04-21 17:16:16 +00:00
Evan Cheng
27a8d019e0 movddup is a SSE3 instruction.
llvm-svn: 27932
2006-04-21 16:42:47 +00:00
Chris Lattner
7cc9363937 Remove a hack required by V9.
llvm-svn: 27931
2006-04-21 15:33:35 +00:00
Chris Lattner
3b9a7570cb Fix a couple more memory issues
llvm-svn: 27930
2006-04-21 15:32:26 +00:00
Reid Spencer
5e33caae17 Remove the extraneous --defined-only option to nm. This is the default and
some versions of nm don't recognize it (its a gnu option).

llvm-svn: 27928
2006-04-21 05:29:25 +00:00
Evan Cheng
19394ba09a Add && to each RUN: line (except the last one).
llvm-svn: 27927
2006-04-21 04:58:23 +00:00
Evan Cheng
b940192ce1 More build_vector tests.
llvm-svn: 27926
2006-04-21 01:22:41 +00:00
Evan Cheng
0365ba09c8 Check for llc crash.
llvm-svn: 27925
2006-04-21 01:21:23 +00:00
Evan Cheng
5cae86002c To be replaced with another test.
llvm-svn: 27924
2006-04-21 01:12:32 +00:00
Evan Cheng
e0289de5ab Now generating perfect (I think) code for "vector set" with a single non-zero
scalar value.

e.g.
        _mm_set_epi32(0, a, 0, 0);
==>
	movd 4(%esp), %xmm0
	pshufd $69, %xmm0, %xmm0

        _mm_set_epi8(0, 0, 0, 0, 0, a, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
==>
	movzbw 4(%esp), %ax
	movzwl %ax, %eax
	pxor %xmm0, %xmm0
	pinsrw $5, %eax, %xmm0

llvm-svn: 27923
2006-04-21 01:05:10 +00:00
Chris Lattner
2ae3ed5e1a Fix a really subtle and obnoxious memory bug that caused issues with an
llvm-gcc4 boostrap.  Whenever a node is deleted by the dag combiner, it
*must* be returned by the visit function, or the dag combiner will not
know that the node has been processed (and will, e.g., send it to the
target dag combine xforms).

llvm-svn: 27922
2006-04-20 23:55:59 +00:00
Reid Spencer
05e3e8fdc9 Don't require the "dot" program if -flat option is given, and don't produce
any of the fancy graphs or other output. Just produce the flat, makefile
style output on stdout.

llvm-svn: 27921
2006-04-20 23:09:57 +00:00
Jeff Cohen
933edad31c Keep Visual Studio informed.
llvm-svn: 27920
2006-04-20 22:19:06 +00:00
Reid Spencer
9c2a65de0e Configure llvm-config in tools, not utils.
llvm-svn: 27919
2006-04-20 22:15:30 +00:00
Reid Spencer
ff132fb460 Remove the llvm-config directory. Its now in tools.
llvm-svn: 27918
2006-04-20 21:16:32 +00:00
Reid Spencer
d9b9561f39 llvm-config doesn't live here anymore ==> tools
llvm-svn: 27917
2006-04-20 21:15:41 +00:00
Reid Spencer
aea3695ef8 Okay, llvm-config is good to go now.
llvm-svn: 27916
2006-04-20 21:14:39 +00:00
Reid Spencer
ab12fd6b0d Several Changes To Support Building llvm-config:
1. If Perl is not available, don't run Perl dependent targets.
2. Check in FinalLibDeps.txt for build environments that can't build it
   because there's no Perl processor.
3. Ensure that FinalLibDeps.txt depends on LibDeps.txt so it is
   automatically regenerated.
4. Support objdir != srcdir building.

llvm-svn: 27915
2006-04-20 21:13:58 +00:00
Reid Spencer
bcfe43138e Don't build llvm-config until issues are resolved.
llvm-svn: 27914
2006-04-20 21:00:24 +00:00
Reid Spencer
0b4370f18b Build the llvm-config directory as a tool.
llvm-svn: 27913
2006-04-20 20:53:23 +00:00
Chris Lattner
01afcd337a Fix Transforms/ScalarRepl/2006-04-20-PromoteCrash.ll
llvm-svn: 27912
2006-04-20 20:48:50 +00:00