RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-24 03:33:20 +01:00
Code Issues Projects Releases Wiki Activity
16,222 Commits 4 Branches 5 Tags 704 MiB
9e7ce53b82
Commit Graph

8582 Commits

Author SHA1 Message Date
Chris Lattner
9e7ce53b82 Simplify code and do not invalidate iterators. 
This fixes a crash compiling TimberWolfMC that was exposed due to recent
optimizer changes.

llvm-svn: 18831
 2004-12-12 18:23:20 +00:00
Chris Lattner
6131b06f73 Use the target triple to pick this target. 
llvm-svn: 18830
 2004-12-12 17:40:28 +00:00
Brian Gaeke
757a3aa9b9 Complete the list of MultiSource failures. 
llvm-svn: 18826
 2004-12-12 08:22:11 +00:00
Brian Gaeke
a440424596 hbd should be working now. 
llvm-svn: 18824
 2004-12-12 07:42:59 +00:00
Brian Gaeke
ee60e35a28 Finally enable the setcc-branch folding code. 
Also, fix a bug where ubyte 255 would sometimes be output as -1. This
was afflicting hbd.

llvm-svn: 18823
 2004-12-12 07:42:58 +00:00
Brian Gaeke
09c4a78ece Add (currently disabled) code for canFoldSetCC 
llvm-svn: 18820
 2004-12-12 06:22:30 +00:00
Chris Lattner
97adae1fa4 Though the previous xform applies to literally dozens (hundreds?) of variables 
in SPEC, the subsequent optimziations that we are after don't play with
with FP values, so disable this xform for them.  Really we just don't want
stuff like:

double G;   (always 0 or 412312.312)
  = G;

turning into:

bool G_b;
  = G_b ? 412312.312 : 0;

We'd rather just do the load.

-Chris

llvm-svn: 18819
 2004-12-12 06:03:06 +00:00
Brian Gaeke
55c163e41e Add stubs for setcc-branch folding support. 
llvm-svn: 18818
 2004-12-12 06:01:26 +00:00
Chris Lattner
f125dc0e49 If a variable can only hold two values, and is not already a bool, shrink it 
down to actually BE a bool.  This allows simple value range propagation
stuff work harder, deleting comparisons in bzip2 in some hot loops.

This implements GlobalOpt/integer-bool.ll, which is the essence of the
loop condition distilled into a testcase.

llvm-svn: 18817
 2004-12-12 05:53:50 +00:00
Chris Lattner
02c04bbf45 If one side of and/or is known to be 0/-1, it doesn't matter 
if the other side is overdefined.

This allows us to fold conditions like:  if (X < Y || Y > Z) in some cases.

llvm-svn: 18807
 2004-12-11 23:15:19 +00:00
Brian Gaeke
5d213ad8c3 Print llvm code one function at a time. 
llvm-svn: 18805
 2004-12-11 22:17:07 +00:00
Chris Lattner
5ba315dbbe Check in the file I forgot last night, to solve all of the crashes in every 
test in the suite.  :(

llvm-svn: 18804
 2004-12-11 22:10:29 +00:00
Brian Gaeke
80831ad19a JIT should print LLVM each function before selecting instructions for it. 
llvm-svn: 18803
 2004-12-11 18:41:09 +00:00
Reid Spencer
e43fb31a1f Rename Path::get -> Path::toString 
llvm-svn: 18802
 2004-12-11 17:37:01 +00:00
Chris Lattner
59ce936426 Only cound if we actually made a change. 
llvm-svn: 18800
 2004-12-11 17:00:14 +00:00
Chris Lattner
8378361f0c The split bb is really the exit of the old function 
llvm-svn: 18799
 2004-12-11 16:59:54 +00:00
Chris Lattner
06bfa390f6 Two bug fixes: 
1. Actually increment the Statistic for the GV elim optzn
 2. When resolving undef branches, only resolve branches in executable blocks,
    avoiding marking a bunch of completely dead blocks live.  This has a big
    impact on the quality of the generated code.

With this patch, we positively rip up vortex, compiling Ut_MoveBytes to a
single memcpy call. In vortex we get this:

     12 ipsccp           - Number of globals found to be constant
    986 ipsccp           - Number of arguments constant propagated
   1378 ipsccp           - Number of basic blocks unreachable
   8919 ipsccp           - Number of instructions removed

llvm-svn: 18796
 2004-12-11 06:05:53 +00:00
Chris Lattner
5af0ef5c44 Do not delete the entry block to a function. 
llvm-svn: 18795
 2004-12-11 05:32:19 +00:00
Brian Gaeke
220fb4f8cd Bools are *also* not ints. Sigh. Furthermore, most of the TargetMachine 
ctor parameters can be defaulted.

Print the transformed llvm code input to the instruction selector
when -print-machineinstrs is on, just like V9.

llvm-svn: 18794
 2004-12-11 05:19:04 +00:00
Brian Gaeke
24ea5d3dd7 Look for many more moves to fold (previously, we only 
*or g0, x      add g0, x          recognized * as a move)
 or  x, g0     add  x, g0
 or  0, x      add  0, x
 or  x, 0      add  x, 0

llvm-svn: 18793
 2004-12-11 05:19:03 +00:00
Brian Gaeke
948a8145bf Make GEPs not suck so much: 
* Don't emit the Index * ElementSize multiply if Index is a constant.
* Use a shift, not a multiply, if ElementSize is 1/2/4/8.
* If ElementSize fits in the immediate field of SMUL, then put it there.

Fix a bug where struct offsets might be truncated (ConstantSInt::get is
now used instead of ConstantInt::get).

llvm-svn: 18792
 2004-12-11 05:19:02 +00:00
Brian Gaeke
e0643b792b Update lists of failing benchmarks, including info on which 
ones are failing in cbe.

llvm-svn: 18791
 2004-12-11 05:19:01 +00:00
Chris Lattner
943f94d2b3 Implement Transforms/SCCP/ipsccp-gvar.ll, by tracking values stored to 
non-address-taken global variables.

llvm-svn: 18790
 2004-12-11 05:15:59 +00:00
Chris Lattner
beeab3a124 Fix a bug where we could delete dead invoke instructions with uses. 
In functions where we fully constant prop the return value, replace all
ret instructions with 'ret undef'.

llvm-svn: 18786
 2004-12-11 02:53:57 +00:00
Reid Spencer
e2af9dc813 Path::get -> Path::toString 
llvm-svn: 18785
 2004-12-11 00:14:15 +00:00
Chris Lattner
d1d00e017b Implement SCCP/ipsccp-conditional.ll, by totally deleting dead blocks. 
llvm-svn: 18781
 2004-12-10 22:29:08 +00:00
Chris Lattner
5be2c2e299 Fix SCCP/2004-12-10-UndefBranchBug.ll 
llvm-svn: 18776
 2004-12-10 20:41:50 +00:00
Chris Lattner
3c31189583 Make sure to link the target-triple as well, so it ends up in the .llvm.bc file 
llvm-svn: 18774
 2004-12-10 20:26:15 +00:00
Chris Lattner
40e97884f6 Fix Regression/Transforms/SimplifyCFG/2004-12-10-SimplifyCFGCrash.ll, 
and the failure on make_dparser last night.

llvm-svn: 18766
 2004-12-10 17:42:31 +00:00
Brian Gaeke
dc916ae40f Move -lowerselect later in the chain; some select instructions were 
slipping through into the instruction selector, which can't deal with
them yet.

llvm-svn: 18758
 2004-12-10 08:39:30 +00:00
Brian Gaeke
7ec3883e1a Add the rest of the multiply instructions. 
llvm-svn: 18757
 2004-12-10 08:39:29 +00:00
Brian Gaeke
2a9ecc433f Support binary operations with immediates for <= cInt. 
llvm-svn: 18756
 2004-12-10 08:39:28 +00:00
Brian Gaeke
45f3af8d88 Update lists of failing benchmarks (except C++...something is the 
matter with my sparcv8 libstdc++.a) and to-do list.

llvm-svn: 18755
 2004-12-10 08:39:27 +00:00
Chris Lattner
ead42a768e This is the initial implementation of IPSCCP, as requested by Brian. 
This implements SCCP/ipsccp-basic.ll, rips apart Olden/mst (as described in
PR415), and does other nice things.

There is still more to come with this, but it's a start.

llvm-svn: 18752
 2004-12-10 08:02:06 +00:00
Brian Gaeke
1da3720799 Emit correct prototype for __builtin_alloca on V8. 
llvm-svn: 18745
 2004-12-10 05:44:45 +00:00
Alkis Evlogimenos
9b02192468 Fix writer to properly quote label names when they don't contain 
simple characters.

llvm-svn: 18744
 2004-12-10 05:41:10 +00:00
Alkis Evlogimenos
189fdf11db Do not allow empty label names. 
llvm-svn: 18743
 2004-12-10 05:40:19 +00:00
Chris Lattner
e0936a0bd3 Implement test/Feature/escaped_label.ll 
llvm-svn: 18741
 2004-12-10 05:27:29 +00:00
Brian Gaeke
91cf4fe1ca Add SparcV8 target back into the build 
llvm-svn: 18738
 2004-12-10 04:54:21 +00:00
Brian Gaeke
5c8cefdf9a Adjust paths: Sparc/V8 --> SparcV8 
llvm-svn: 18737
 2004-12-10 04:48:57 +00:00
Brian Gaeke
262fe40da0 Make this file self-contained. 
llvm-svn: 18736
 2004-12-10 04:46:30 +00:00
Brian Gaeke
ef7289195a Update list of failing MultiSource benchmarks. It works out to +5 -5, but I 
think some of these might be the CFE's fault; a rebuild should come soon.

llvm-svn: 18735
 2004-12-10 04:42:46 +00:00
Brian Gaeke
999a5ba9ba When FpMOVDs appeared in pairs, we were mistakenly skipping over the latter of 
each pair. I think this fixes that.

One of these days, I swear I'm going to get the hang of C++ iterators.
Really.

llvm-svn: 18734
 2004-12-10 04:42:45 +00:00
Brian Gaeke
a196f80d71 We're continuing to make progress on MultiSource. 
llvm-svn: 18714
 2004-12-09 18:54:31 +00:00
Brian Gaeke
706a0c3988 Bytes and shorts are aligned differently from words. 
llvm-svn: 18713
 2004-12-09 18:51:02 +00:00
Brian Gaeke
43cb9ee8a4 Fix asm-printing directives (how did we not see this before...apparently, 
everything was an int!)

llvm-svn: 18712
 2004-12-09 18:51:01 +00:00
Chris Lattner
f5962a7d6c note to self: Do not check in debugging code! 
llvm-svn: 18693
 2004-12-09 07:15:52 +00:00
Chris Lattner
1e2b3fd343 Implement trivial sinking for load instructions. This causes us to sink 567 loads in spec 
llvm-svn: 18692
 2004-12-09 07:14:34 +00:00
Chris Lattner
c5f6c3fd48 Remove a dead field, make the map go to integer type ID to hash better and 
avoid a getType.

llvm-svn: 18691
 2004-12-09 06:19:44 +00:00
Jeff Cohen
2d53514c83 Fix residual Visual Studio build problems 
llvm-svn: 18688
 2004-12-09 05:51:11 +00:00