Commit Graph

8206 Commits

Author SHA1 Message Date
Brian Gaeke
b31a828533 Update failing Benchmarks; point out that I'm skipping Shootout-C++.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17725 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-14 04:43:12 +00:00
Chris Lattner
f5990edc87 Rearrange some code, no functionality changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17724 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-14 04:24:28 +00:00
Brian Gaeke
1c745818ff Update expected UnitTests failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17723 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-14 03:22:08 +00:00
Brian Gaeke
24b90c3647 Rewrite outgoing arg handling to handle more weird corner cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17722 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-14 03:22:07 +00:00
Brian Gaeke
54799c2a51 Support UndefValue emission.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17721 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-14 03:22:05 +00:00
Chris Lattner
4c989b6bd7 Remove debugging code
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17719 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 23:32:53 +00:00
Chris Lattner
11a3d7b7dd Argument promotion transforms functions to unconditionally load their
argument pointers.  This is only valid to do if the function already
unconditionally loaded an argument or if the pointer passed in is known
to be valid.  Make sure to do the required checks.

This fixed ArgumentPromotion/control-flow.ll and the Burg program.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17718 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 23:31:34 +00:00
Chris Lattner
4b2c09fa22 Don't print unneeded labels
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17714 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 23:27:11 +00:00
Chris Lattner
4394d51467 Hack around stupidity in GCC, fixing Burg with the CBE and
CBackend/2004-11-13-FunctionPointerCast.llx


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17710 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 22:21:56 +00:00
Chris Lattner
ce7cafa960 shld is a very high latency operation. Instead of emitting it for shifts of
two or three, open code the equivalent operation which is faster on athlon
and P4 (by a substantial margin).

For example, instead of compiling this:

long long X2(long long Y) { return Y << 2; }

to:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        shldl $2, %eax, %edx
        shll $2, %eax
        ret

Compile it to:

X2:
        movl 4(%esp), %eax
        movl 8(%esp), %ecx
        movl %eax, %edx
        shrl $30, %edx
        leal (%edx,%ecx,4), %edx
        shll $2, %eax
        ret

Likewise, for << 3, compile to:

X3:
        movl 4(%esp), %eax
        movl 8(%esp), %ecx
        movl %eax, %edx
        shrl $29, %edx
        leal (%edx,%ecx,8), %edx
        shll $3, %eax
        ret

This matches icc, except that icc open codes the shifts as adds on the P4.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17707 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 20:48:57 +00:00
Chris Lattner
62f5a9402c Add missing check
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17706 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 20:04:38 +00:00
Chris Lattner
44205cadba Compile:
long long X3_2(long long Y) { return Y+Y; }
int X(int Y) { return Y+Y; }

into:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        addl %eax, %eax
        adcl %edx, %edx
        ret
X:
        movl 4(%esp), %eax
        addl %eax, %eax
        ret

instead of:

X3_2:
        movl 4(%esp), %eax
        movl 8(%esp), %edx
        shldl $1, %eax, %edx
        shll $1, %eax
        ret

X:
        movl 4(%esp), %eax
        shll $1, %eax
        ret


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17705 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 20:03:48 +00:00
Chris Lattner
50af16a97b Simplify handling of shifts to be the same as we do for adds. Add support
for (X * C1) + (X * C2) (where * can be mul or shl), allowing us to fold:

   Y+Y+Y+Y+Y+Y+Y+Y

into
         %tmp.8 = shl long %Y, ubyte 3           ; <long> [#uses=1]

instead of

        %tmp.4 = shl long %Y, ubyte 2           ; <long> [#uses=1]
        %tmp.12 = shl long %Y, ubyte 2          ; <long> [#uses=1]
        %tmp.8 = add long %tmp.4, %tmp.12               ; <long> [#uses=1]

This implements add.ll:test25

Also add support for (X*C1)-(X*C2) -> X*(C1-C2), implementing sub.ll:test18


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17704 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 19:50:12 +00:00
Chris Lattner
65aedc1a4e Fold:
(X + (X << C2)) --> X * ((1 << C2) + 1)
   ((X << C2) + X) --> X * ((1 << C2) + 1)

This means that we now canonicalize "Y+Y+Y" into:

        %tmp.2 = mul long %Y, 3         ; <long> [#uses=1]

instead of:

        %tmp.10 = shl long %Y, ubyte 1          ; <long> [#uses=1]
        %tmp.6 = add long %Y, %tmp.10               ; <long> [#uses=1]


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17701 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 19:31:40 +00:00
Chris Lattner
0c3b390a54 Lazily create the abort message, so only translation units that use unwind
will actually get it.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17700 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 19:07:32 +00:00
Chris Lattner
9ecc046184 Fix: CodeExtractor/2004-11-12-InvokeExtract.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17699 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-13 00:06:45 +00:00
Chris Lattner
68c311aac6 Fix a bug where the code extractor would get a bit confused handling invoke
instructions, setting DefBlock to a block it did not have dom info for.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17697 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-12 23:50:44 +00:00
Chris Lattner
44c94b9671 Simplify handling of constant initializers
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17696 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-12 22:42:57 +00:00
Reid Spencer
5c911ab6f1 Makefile for lib/Linker
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17695 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-12 20:38:45 +00:00
Reid Spencer
7f49602e5a This file originated in lib/VMCore/Linker.cpp but now lives in
lib/Linker/LinkModules.cpp


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17694 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-12 20:37:43 +00:00
Reid Spencer
c408c45900 This file originated in tools/gccld/Linker.cpp but now lives in
lib/Linker/LinkArchives.cpp


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17693 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-12 20:34:32 +00:00
Chris Lattner
04252fed78 Make this build in release mode
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17684 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-11 22:11:17 +00:00
Chris Lattner
d79c7b43f4 Actually, leave the check in. This prevents us from counting dead arguments
as IPCP opportunities.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17680 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-11 07:47:54 +00:00
Chris Lattner
3e062eacab Fix bug: IPConstantProp/deadarg.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17679 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-11 07:46:29 +00:00
Chris Lattner
ff1529b6cf Make IP Constant prop more aggressive about handling self recursive calls.
This implements IPConstantProp/recursion.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17666 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-10 19:43:59 +00:00
John Criswell
546faca4ef Correct the name of stosd for the AT&T syntax:
It's stosl (l for long == 32 bit).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17658 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-10 04:48:15 +00:00
Chris Lattner
c890bfe1da Do not let dead constant expressions hanging off of functions prevent IPCP.
This allows to elimination of a bunch of global pool descriptor args from
programs being pool allocated (and is also generally useful!)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17657 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 20:47:30 +00:00
Reid Spencer
4b82681c65 Fix isBytecodeFile to correctly recognized compressed bytecode too.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17655 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 20:27:23 +00:00
Reid Spencer
9195f37c5f * Implement getStatusInfo for getting stat(2) like information
* Implement createTemporaryFile for mkstemp(3) functionality
* Fix isBytecodeFile to accept llvc magic # (compressed) as bytecode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17654 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 20:26:31 +00:00
John Criswell
b5810282ee Recognize compressed LLVM bytecode files.
This should fix the problem of not being able to link compressed LLVM
bytecode files from LLVM libraries.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17648 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 19:37:07 +00:00
Reid Spencer
580a1164d8 Tune compression:
bzip2: block size 9 -> 5, reduces memory by 400Kbytes, doesn't affect speed
       or compression ratio on all but the largest bytecode files (>1MB)
zip:   level 9 -> 6, this speeds up compression time by ~30% but only
       degrades the compressed size by a few bytes per megabyte. Those few
       bytes aren't worth the effort.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17647 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 17:58:09 +00:00
Chris Lattner
a8e475093a Change this back so that I get stable numbers to reflect the change from the
nightly testers


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17646 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 08:05:23 +00:00
Chris Lattner
52f20f8308 Fix bug: 2004-11-08-FreeUseCrash.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17642 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 05:10:56 +00:00
Misha Brukman
710d1ce4f0 * Convert tabs to spaces
* Order #includes according to style guide
* Remove extraneous blank lines


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17639 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 04:27:19 +00:00
Misha Brukman
669b524399 Output the program name (in this case, gccld) with warning about invalid files
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17638 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 04:24:59 +00:00
Nate Begeman
a591457857 Allow hbd to be bugpointable on darwin by fixing common and linkonce codegen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17637 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-09 04:01:18 +00:00
Misha Brukman
336e56e67d Handle headers for compressed bytecode files
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17634 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 22:03:32 +00:00
Misha Brukman
eda20f933b Don't silently ignore invalid files: tell the user!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17633 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 22:03:10 +00:00
Chris Lattner
ac6d4854e1 Fix a bug that was preventing povray and namd from pool allocating correctly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17632 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 21:08:46 +00:00
Chris Lattner
beacefae61 Handle assert_fail special
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17631 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 21:08:28 +00:00
Reid Spencer
01a42e40ed Fix a typo: isvalid -> isValid
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17621 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 08:27:51 +00:00
Nate Begeman
676dee6ae9 Put int the getReg cast optimization from x86 so that we generate fewer
move instructions for the register allocator to coalesce.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17608 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-08 02:25:40 +00:00
Misha Brukman
895eddfad4 Implement ExecutionEngine::freeMachineCodeForFunction()
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17601 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 23:58:46 +00:00
Chris Lattner
ae18963a70 VERY large functions that are only called from one place are not really
exciting to inline.  Only inline medium or small sized functions with a
single call site.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17588 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 21:46:47 +00:00
Nate Begeman
075cdc655e Disable bogus cast elimination when the cast is used by a setcc instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17583 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 20:23:42 +00:00
Reid Spencer
f0c977cbac * Comments & cleanup per CL code review.
* Accept 'llvc' as signature for compressed bytecode


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17579 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 18:20:55 +00:00
Reid Spencer
36eea80616 Add comments per CL code review.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17578 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 18:19:00 +00:00
Reid Spencer
83296f5496 * Add comments and cleanup per CL code review
* Make signature for compressed bytecode llvc instead of unreadable
* Make the CompressionContext have a constructor and destructor.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17576 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 18:17:38 +00:00
Misha Brukman
fb4863ac55 Replace uses of llvm.org with llvm.cs.uiuc.edu
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17549 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 00:58:38 +00:00
Misha Brukman
d8b8912861 Consistently use llvm.cs.uiuc.edu instead of llvm.org
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@17548 91177308-0d34-0410-b5e6-96231b3b80d8
2004-11-07 00:58:20 +00:00