maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
11,485 Commits 23 Branches 0 Tags 2.2 GiB
Commit Graph

11483 Commits

Author SHA1 Message Date
Chris Lattner 464e2ea567 Two changes: 
* In promote32, if we can just promote a constant value, do so instead of
    promoting a constant dynamically.
  * In visitReturn inst, actually USE the promote32 argument that takes a
    Value*

The end result of this is that we now generate this:

test:
        mov %EAX, 0
        ret

instead of...

test:
        mov %AX, 0
        movzx %EAX, %AX
        ret

for:

ushort %test() {
        ret ushort 0
}

llvm-svn: 12679
 2004-04-06 01:21:00 +00:00
Chris Lattner bf791614ed Merge the code generator miscompilation code into the optimizer miscompilation 
code.  This "instantly" gives us loop-extractor power to assist with the
debugment of our nasty codegen issues.  :)

llvm-svn: 12678
 2004-04-05 22:58:16 +00:00
Chris Lattner 95053a9f28 Make a method public 
llvm-svn: 12677
 2004-04-05 22:01:48 +00:00
Chris Lattner 9af52d12d9 Minor cleanups, remove some old debug code 
llvm-svn: 12676
 2004-04-05 21:37:55 +00:00
Chris Lattner 0434ba3ed0 Refactor and genericize code 
llvm-svn: 12675
 2004-04-05 21:37:38 +00:00
Chris Lattner b0d1e9d02e lli no longer takes the -quiet option! 
llvm-svn: 12674
 2004-04-05 20:28:41 +00:00
Chris Lattner c97b7b2285 Do not mangle intrinsics in any way! 
llvm-svn: 12673
 2004-04-05 20:17:53 +00:00
Chris Lattner 0f1df36bdd Make full use of the Mangler interface to simplify code 
llvm-svn: 12671
 2004-04-05 19:31:02 +00:00
Chris Lattner d4f78f270b Sparc don't got not "sqrtl", bum bum bum 
llvm-svn: 12670
 2004-04-05 19:05:15 +00:00
Misha Brukman 5ebc25c818 Kill warnings during an optimized compile where assert() disappears. 
llvm-svn: 12669
 2004-04-05 19:00:46 +00:00
Chris Lattner 29153fc2e5 Fix PR312 and IndVarsSimplify/2004-04-05-InvokeCastCrash.llx 
llvm-svn: 12668
 2004-04-05 18:46:55 +00:00
Chris Lattner 6f4fea937b New testcase for PR312 
llvm-svn: 12667
 2004-04-05 18:46:33 +00:00
Chris Lattner 4d1fcf1dcd Fix a bug in yesterdays checkins which broke siod. siod is a great testcase! :) 
llvm-svn: 12659
 2004-04-05 16:02:41 +00:00
Chris Lattner 8953b90aaa Fix InstCombine/2004-04-04-InstCombineReplaceAllUsesWith.ll 
llvm-svn: 12658
 2004-04-05 02:10:19 +00:00
Chris Lattner e79fd5c766 New testcase that crashes the instcombine pass. Dominance properties have 
no meaning if the code is not reachable.

llvm-svn: 12657
 2004-04-05 02:01:32 +00:00
Chris Lattner 677202b49e PR82 is finally fixed! 
llvm-svn: 12656
 2004-04-05 01:43:08 +00:00
Chris Lattner 6087034f3a Minor change 
llvm-svn: 12655
 2004-04-05 01:31:50 +00:00
Chris Lattner 33fd702590 Update getelementptr instruction description 
llvm-svn: 12654
 2004-04-05 01:30:49 +00:00
Chris Lattner 69193f93b6 Support getelementptr instructions which use uint's to index into structure 
types and can have arbitrary 32- and 64-bit integer types indexing into
sequential types.

llvm-svn: 12653
 2004-04-05 01:30:19 +00:00
Chris Lattner fd9fbe187d Support getelementptr instructions which use uint's to index into structure 
types and can have arbitrary 32- and 64-bit integer types indexing into
sequential types.

Auto-upgrade .ll files that use ubytes to index into structures to use uint's.

llvm-svn: 12652
 2004-04-05 01:30:04 +00:00
Chris Lattner 15701e84d1 Implement support for a new LLVM 1.3 bytecode format, which uses uint's 
to index into structure types and allows arbitrary 32- and 64-bit integer
types to index into sequential types.

llvm-svn: 12651
 2004-04-05 01:27:26 +00:00
Chris Lattner 60cf133a8e Be more restrictive with the index types we allow for sequential types 
llvm-svn: 12650
 2004-04-05 01:25:21 +00:00
Chris Lattner e916f16dbb PR305 is now fixed 
llvm-svn: 12649
 2004-04-05 00:40:55 +00:00
Chris Lattner dd28474610 Add ConstantExpr::get(Sign|Zero)Extend methods 
llvm-svn: 12648
 2004-04-04 23:20:30 +00:00
Chris Lattner dfcf8e34cf In the perhaps not-to-distant future, we might support gep instructions that 
have non-long indices for sequential types.  In order to avoid trying to figure
out how the v9 backend works, we'll just hack it in the preselection pass.

llvm-svn: 12647
 2004-04-04 20:44:05 +00:00
Chris Lattner ca76d11a81 Adjust to new interface 
llvm-svn: 12646
 2004-04-04 19:47:06 +00:00
Chris Lattner 5453b2f376 Support iteration over constant instructions 
llvm-svn: 12645
 2004-04-04 19:46:54 +00:00
Chris Lattner 092d260fc1 Adjust to new gep_type_iterator prototypes. 
llvm-svn: 12644
 2004-04-04 17:30:06 +00:00
Chris Lattner bc8ba73cf1 Remove a bunch of cruft that was used to be backwards compatible with the last 
prerelease format for LLVM bytecode files.  Now we only are compatible with
LLVM 1.0+.

llvm-svn: 12643
 2004-04-03 23:43:42 +00:00
Chris Lattner 476f71e30b Allow for use of arbitrary iterator types... 
llvm-svn: 12642
 2004-04-03 23:29:11 +00:00
Chris Lattner 8ed3c8aa13 Implement test/Regression/Transforms/GCSE/undefined_load.ll 
llvm-svn: 12641
 2004-04-03 00:45:16 +00:00
Chris Lattner db033fa425 New testcase 
llvm-svn: 12640
 2004-04-03 00:44:56 +00:00
Chris Lattner 0defaa1cbc Add a break in the default case 
llvm-svn: 12639
 2004-04-03 00:43:03 +00:00
Brian Gaeke 5ade501501 Add autoconf support for isStandardOutAConsole (). 
llvm-svn: 12638
 2004-04-02 21:26:04 +00:00
Brian Gaeke 0d372ee696 Regenerated using autoheader-2.57. 
llvm-svn: 12637
 2004-04-02 21:26:03 +00:00
Brian Gaeke 8bdbb3d7eb Regenerated using autoconf-2.57. 
llvm-svn: 12636
 2004-04-02 21:26:02 +00:00
Brian Gaeke c6dba95bbc check for isatty function 
llvm-svn: 12635
 2004-04-02 21:06:44 +00:00
Chris Lattner 9dcd639e5c Remove obsolete analyses 
llvm-svn: 12634
 2004-04-02 20:56:33 +00:00
Chris Lattner 6748cca268 Remove obsolete files 
llvm-svn: 12633
 2004-04-02 20:56:24 +00:00
Brian Gaeke d0bdad38f3 Add support for many of the MRegisterInfo callbacks. 
Eliminating call-frame pseudo instrs and frame indices are still stubs.
Flesh out the emitPrologue method based on better ABI knowledge.

llvm-svn: 12632
 2004-04-02 20:53:37 +00:00
Brian Gaeke d4869e4107 Add load, store, and NOP instructions. 
Fix up comments.

llvm-svn: 12631
 2004-04-02 20:53:37 +00:00
Brian Gaeke b65254a34d Add support for printing pc-relative displacements of functions (as used in 
the CALL instruction).

llvm-svn: 12630
 2004-04-02 20:53:35 +00:00
Brian Gaeke 2fd46b6e43 Add support for call instructions (0-ary only for now). 
llvm-svn: 12629
 2004-04-02 20:53:33 +00:00
Chris Lattner 6f29c89597 cleanup some long-dead code 
llvm-svn: 12628
 2004-04-02 20:46:26 +00:00
Chris Lattner d4f122044a Ignore configure produced files 
llvm-svn: 12627
 2004-04-02 20:34:30 +00:00
Chris Lattner 64cdd1ef55 Tweak libraries for scev changes 
llvm-svn: 12625
 2004-04-02 20:32:46 +00:00
Chris Lattner d8d803adb9 new testcase 
llvm-svn: 12624
 2004-04-02 20:27:47 +00:00
Chris Lattner 0916921332 Comment out debugging printouts 
llvm-svn: 12623
 2004-04-02 20:26:46 +00:00
Chris Lattner 76378746c5 New testcases for the indvars pass 
llvm-svn: 12622
 2004-04-02 20:26:04 +00:00
Chris Lattner 2c416ae11f Tweak testcase to work with new indvars pass 
llvm-svn: 12621
 2004-04-02 20:25:26 +00:00
Chris Lattner e61b67d7d5 Rewrite the indvars pass to use the ScalarEvolution analysis. 
This also implements some new features for the indvars pass, including
linear function test replacement, exit value substitution, and it works with
a much more general class of induction variables and loops.

llvm-svn: 12620
 2004-04-02 20:24:31 +00:00
Chris Lattner d934c70cf8 Add a new analysis 
llvm-svn: 12619
 2004-04-02 20:23:17 +00:00
Chris Lattner eed034bcd3 Fix the obvious bug in my previous checkin 
llvm-svn: 12618
 2004-04-02 18:15:10 +00:00
Chris Lattner 9f0db32625 Implement Transforms/SimplifyCFG/return-merge.ll 
This actually causes us to turn code like:

  return C ? A : B;

into a select instruction.

llvm-svn: 12617
 2004-04-02 18:13:43 +00:00
Chris Lattner 99c9b9c02f New testcase 
llvm-svn: 12616
 2004-04-02 18:12:49 +00:00
Alkis Evlogimenos d64e904e27 Clean up code a bit. 
llvm-svn: 12615
 2004-04-02 18:11:32 +00:00
Brian Gaeke 2845936a34 Only strip symbols if emitting bytecode to the assembly file. 
Move lowerselect pass to come after preselection.  Move machine
code construction and stack slots pass to come right before instruction
selection. This is to help fix perlbmk.

Update comments.

Make the sequence of passes in addPassesToJITCompile look more like
the sequence of passes in addPassesToEmitAssembly, including support
for -print-machineinstrs.

llvm-svn: 12614
 2004-04-02 17:52:40 +00:00
Brian Gaeke cc24411c0a Add support for constant select expressions. Clarify the assertion failure msg. 
llvm-svn: 12613
 2004-04-02 17:52:29 +00:00
Chris Lattner 3838243d04 Minor speedup 
llvm-svn: 12612
 2004-04-02 16:28:32 +00:00
Alkis Evlogimenos fe66caa9a0 Fix type in comments 
llvm-svn: 12611
 2004-04-02 16:02:50 +00:00
Alkis Evlogimenos 5fc4772d5e Fix type in instruction builder instantiation 
llvm-svn: 12610
 2004-04-02 15:51:03 +00:00
Chris Lattner a45a216ff6 Make the verifier API more complete and useful. 
Patch contributed by Reid Spencer

llvm-svn: 12609
 2004-04-02 15:45:08 +00:00
Chris Lattner b753dd1cb4 Make the verifier API more complete and useful 
llvm-svn: 12608
 2004-04-02 15:44:33 +00:00
Alkis Evlogimenos d186ed02e4 Add more ADC and SBB variants 
llvm-svn: 12607
 2004-04-02 07:11:10 +00:00
Chris Lattner bcf2623cab minor formatting change 
llvm-svn: 12606
 2004-04-02 06:32:45 +00:00
Chris Lattner a413b08580 Fix two pretty serious bugs: 
1. Each time the loop extractor extracted a loop, we would leak a module.
  2. When we extracted a loop, we didn't add the new function to the list of
     miscompiled functions.  Thus if the bug was in a loop nest and we
     extracted it, we could actually *LOSE THE BUG*, which is very bad.

With these patches, bugpoint has successfully found a bug for me in a function
with several nested loops, and cut it down to just one of them. :) :)

llvm-svn: 12605
 2004-04-02 06:32:17 +00:00
Chris Lattner af67dac7f4 Fix a fairly nasty bug that prevented bugpoint from working quite right when 
hacking on programs with two functions that have the same name.

llvm-svn: 12604
 2004-04-02 06:30:33 +00:00
Chris Lattner 73c141ac71 If the program returns a non-zero exit value, don't leave files laying 
around

llvm-svn: 12603
 2004-04-02 05:33:06 +00:00
Chris Lattner 46e18c7f87 Fix wonky header 
Address PR305: LLVM tools will happily spew bytecode onto your terminal

llvm-svn: 12602
 2004-04-02 05:06:57 +00:00
Chris Lattner 3af1fffd3c Add new function 
llvm-svn: 12601
 2004-04-02 05:04:12 +00:00
Chris Lattner 454e18317d Add new function, autoconf support required tho 
llvm-svn: 12600
 2004-04-02 05:04:03 +00:00
Chris Lattner fd8f46b1cd Bug fixed 
llvm-svn: 12598
 2004-04-01 20:31:29 +00:00
Chris Lattner c24019c825 Fix PR310 and TailDup/2004-04-01-DemoteRegToStack.llx 
llvm-svn: 12597
 2004-04-01 20:28:45 +00:00
Chris Lattner 914a7d0d90 New testcase for PR310 
llvm-svn: 12596
 2004-04-01 20:28:35 +00:00
Chris Lattner 59fdf74968 Remove some assertions that are now bogus with the last patch I put in 
llvm-svn: 12595
 2004-04-01 19:21:46 +00:00
Chris Lattner ff03fe3b13 Bug fixed 
llvm-svn: 12593
 2004-04-01 19:09:49 +00:00
Chris Lattner 146d0df5e4 Fix PR306: Loop simplify incorrectly updates dominator information 
Testcase: LoopSimplify/2004-04-01-IncorrectDomUpdate.ll

llvm-svn: 12592
 2004-04-01 19:06:07 +00:00
Chris Lattner f475a12626 New testcase for PR306 
llvm-svn: 12591
 2004-04-01 19:05:54 +00:00
Misha Brukman b78259a57d Fix grammar. 
llvm-svn: 12590
 2004-04-01 17:15:42 +00:00
Chris Lattner b28dd11e33 Add support for select constant expressions to the CBE, fixing SIOD 
llvm-svn: 12589
 2004-04-01 05:28:26 +00:00
Chris Lattner bc7e35b3b1 Simplify code by using the more powerful BuildMI forms. 
Implement a small optimization.  In test/Regression/CodeGen/X86/select.ll,
we now generate this for foldSel3:

foldSel3:
        mov %AL, BYTE PTR [%ESP + 4]
        fld DWORD PTR [%ESP + 8]
        fld DWORD PTR [%ESP + 12]
        mov %EAX, DWORD PTR [%ESP + 16]
        mov %ECX, DWORD PTR [%ESP + 20]
        cmp %EAX, %ECX
        fxch %ST(1)
        fcmovae %ST(0), %ST(1)
***     fstp %ST(1)
        ret

Instead of:

foldSel3:
        mov %AL, BYTE PTR [%ESP + 4]
        fld DWORD PTR [%ESP + 8]
        fld DWORD PTR [%ESP + 12]
        mov %EAX, DWORD PTR [%ESP + 16]
        mov %ECX, DWORD PTR [%ESP + 20]
        cmp %EAX, %ECX
        fxch %ST(1)
        fcmovae %ST(0), %ST(1)
***     fxch %ST(1)
***     fstp %ST(0)
        ret

In practice, this only effects code size: performance should be basically
unaffected.

llvm-svn: 12588
 2004-04-01 04:06:09 +00:00
Chris Lattner 3c8561442c Wrap at 80 cols 
llvm-svn: 12587
 2004-04-01 04:03:27 +00:00
Chris Lattner fb893edbe3 Allow converting a builder to an iterator 
llvm-svn: 12586
 2004-04-01 04:03:10 +00:00
Chris Lattner cfbc023153 Tests for fp cmov's that I forgot to check in earlier 
llvm-svn: 12585
 2004-04-01 03:47:56 +00:00
Chris Lattner 1a7e43c5cf Clear out all of the changes. Reset version numbers to 1.3 
add note about select

llvm-svn: 12584
 2004-04-01 00:41:31 +00:00
Brian Gaeke 8f177d9171 Add clear() forwarding method. 
llvm-svn: 12580
 2004-03-31 22:43:12 +00:00
Chris Lattner d55509c281 Generate slightly smaller code, "test R, R" instead of "cmp R, 0" 
llvm-svn: 12579
 2004-03-31 22:22:36 +00:00
Chris Lattner a4b15f04c6 The X86 backend no longer needs the select lowering pass. 
llvm-svn: 12578
 2004-03-31 22:03:46 +00:00
Chris Lattner 37a7f09d80 Codegen FP select instructions into X86 conditional moves. Annoyingly enough 
the X86 does not support a full set of fp cmove instructions, so we can't always
fold the condition into the select.  :(  Yuck.

llvm-svn: 12577
 2004-03-31 22:03:35 +00:00
Chris Lattner c07c95816a Add support for floating point conditional move instructions 
llvm-svn: 12576
 2004-03-31 22:02:36 +00:00
Chris Lattner cbb4ed9ef2 Add support for FP cmoves 
llvm-svn: 12575
 2004-03-31 22:02:21 +00:00
Chris Lattner 9fe1646804 Add FP conditional move instructions, which annoyingly have special properties 
that require the asmwriter to be extended (printing implicit uses before the
explicit operands)

llvm-svn: 12574
 2004-03-31 22:02:13 +00:00
Chris Lattner 61fab1409d Add warning 
llvm-svn: 12573
 2004-03-31 22:00:30 +00:00
Chris Lattner afbafb55ea MBB::remove should not modify the iterator passed in 
llvm-svn: 12572
 2004-03-31 21:59:59 +00:00
Chris Lattner acccf32aa3 MachineBasicBlock::remove should not modify the iterator passed in 
llvm-svn: 12571
 2004-03-31 21:59:29 +00:00
Chris Lattner 7c92c296ba Improve description, add warning 
llvm-svn: 12570
 2004-03-31 21:59:07 +00:00
Chris Lattner 49ed8ad580 MachineBasicBlock::remove should not change the iterator passed into it 
llvm-svn: 12569
 2004-03-31 21:58:50 +00:00
Brian Gaeke 623710bf2f Factor out getStaticStackSize from InsertPrologCode(), so that I can more 
easily steal it for a separate use in the reoptimizer.

llvm-svn: 12568
 2004-03-31 20:58:37 +00:00
Brian Gaeke 77e78a4531 Use the true, decoded name of the archive member in getObjectType. 
In ReadArchiveBuffer, make sure that MemberName is set in the case where
getObjectType would want to return SVR4LongFilename.

llvm-svn: 12567
 2004-03-31 19:51:00 +00:00
Chris Lattner c0ba90e35c Avoid TRUE and FALSE which apparently conflict with some macros on OSX 
llvm-svn: 12566
 2004-03-31 03:49:47 +00:00
Chris Lattner 709f03e2dd Fix linking of constant expr casts due to type resolution changes. With 
this and the other patches 253.perlbmk links again.

llvm-svn: 12565
 2004-03-31 02:58:28 +00:00
Chris Lattner 467cb2bc5d Add support for constant expr casts 
llvm-svn: 12564
 2004-03-31 02:56:11 +00:00
Chris Lattner 974efc72be Add support for reading constantexpr select instructions 
llvm-svn: 12563
 2004-03-31 02:53:59 +00:00
Chris Lattner b7897ce431 Fix a latent bug in select constantexpr handling that was broke 253.perlbmk 
llvm-svn: 12562
 2004-03-30 22:51:03 +00:00
Alkis Evlogimenos a333b1382f Correctly update LiveVariables when an instruction changes 
llvm-svn: 12561
 2004-03-30 22:44:39 +00:00
Chris Lattner 32817f5985 Fold comparisons into select instructions, making much better code and 
using our broad selection of movcc instructions.  :)

llvm-svn: 12560
 2004-03-30 22:39:09 +00:00
Chris Lattner 2dfdad631f Test folding comparisons into select instructions 
llvm-svn: 12559
 2004-03-30 22:37:04 +00:00
Chris Lattner f78f843adf New testcase. This now codegens to: 
max:
        mov %EAX, DWORD PTR [%ESP + 4]
        mov %ECX, DWORD PTR [%ESP + 8]
        cmp %EAX, %ECX
        cmovle %EAX, %ECX
        ret

Someone should really implement passing arguments through registers for
known-internal functions.  :)

llvm-svn: 12558
 2004-03-30 22:36:52 +00:00
Chris Lattner 358c97e7ae Implement spill code folding for all of the conditional move instructions 
llvm-svn: 12554
 2004-03-30 21:29:47 +00:00
Chris Lattner 53b58cb854 Add direct support for integer select instructions, though we still don't support 
folding compares into the select yet.

llvm-svn: 12553
 2004-03-30 21:22:00 +00:00
Chris Lattner b5d47a9420 New testcase for select instructions 
llvm-svn: 12552
 2004-03-30 21:21:14 +00:00
Chris Lattner deab5e212d Fix bug: Assembler/2004-03-30-UnclosedFunctionCrash.llx 
llvm-svn: 12551
 2004-03-30 20:58:25 +00:00
Chris Lattner 090516fe93 New testcase that crashes the assembler 
llvm-svn: 12550
 2004-03-30 20:58:00 +00:00
Chris Lattner 1563983d81 Fix some serious bugs in the cmov descriptions, which didn't cause a problem because 
we never generated them

Make indentation a bit more consistent

llvm-svn: 12549
 2004-03-30 20:18:02 +00:00
Brian Gaeke ef327be6ed Start cleaning up this pass so that I can debug it. 
llvm-svn: 12548
 2004-03-30 19:53:46 +00:00
Chris Lattner f17d2d33fb Oops, actually USE the previously computed value 
llvm-svn: 12547
 2004-03-30 19:45:39 +00:00
Chris Lattner fdf3202d69 Test general value/value selection which we can do now that we use the 
select instruction

llvm-svn: 12546
 2004-03-30 19:45:11 +00:00
Chris Lattner 81bdcb90ce Now that all the code generators support the select instruction, and the instcombine 
pass can eliminate many nasty cases of them, start generating them in the optimizers

llvm-svn: 12545
 2004-03-30 19:44:05 +00:00
Chris Lattner 533bc49775 Implement select.ll:test[3-6] 
llvm-svn: 12544
 2004-03-30 19:37:13 +00:00
Chris Lattner b785d280ec Add some testcases for select simplification 
llvm-svn: 12543
 2004-03-30 19:36:54 +00:00
Chris Lattner 0048e574fb Fix a fairly major performance problem. If a PHI node had a constant as 
an incoming value from a block, the selector would evaluate the constant
at the TOP of the block instead of at the end of the block.  This made the
live range for the constant span the entire block, increasing register
pressure needlessly.

llvm-svn: 12542
 2004-03-30 19:10:12 +00:00
Chris Lattner 1a0e9ac2f5 Add the select lowering pass to get initial support for select instructions 
llvm-svn: 12541
 2004-03-30 18:41:59 +00:00
Chris Lattner 059f390257 Add a simple select instruction lowering pass 
llvm-svn: 12540
 2004-03-30 18:41:10 +00:00
Chris Lattner 55481f78d3 Add some new methods 
llvm-svn: 12539
 2004-03-30 00:20:08 +00:00
Brian Gaeke fd02378d3e Don't warn about a null live range if the Value is a ConstantIntegral. 
Otherwise, if you're in debugging mode, you get warnings for (apparently)
every immediate constant in the function during reg. allocation.

llvm-svn: 12538
 2004-03-29 21:58:41 +00:00
Brian Gaeke 2fe0ac9ade Change how the beginnings and ends of MachineFunctions are printed. Get 
rid of the funny cast.

llvm-svn: 12537
 2004-03-29 21:58:31 +00:00
Chris Lattner 5152b9ed34 Adjust to new itf 
llvm-svn: 12534
 2004-03-29 20:42:49 +00:00
Chris Lattner 8582975b62 Relax the interface a bit 
llvm-svn: 12533
 2004-03-29 20:42:38 +00:00
John Criswell fa2f3686a3 Merged in RELEASE_12. 
llvm-svn: 12532
 2004-03-29 20:23:11 +00:00
John Criswell d2ff5cf7e8 Merged in RELEASE_12. 
This fixes the zlib linking problems.

llvm-svn: 12531
 2004-03-29 20:22:30 +00:00
Chris Lattner 3cdc27c868 Handle -0.0 correctly 
llvm-svn: 12530
 2004-03-29 19:51:24 +00:00
Chris Lattner 87d6b404b5 Add a faq entry 
llvm-svn: 12529
 2004-03-29 19:14:35 +00:00
Chris Lattner 817175f883 Add a bunch of methods that should have been added a long time ago. 
llvm-svn: 12526
 2004-03-29 02:37:53 +00:00
Chris Lattner 5f126b74d3 Add another check 
llvm-svn: 12525
 2004-03-29 00:29:36 +00:00
Chris Lattner 587a9f1416 Fix an assertion 
Contributed by Reid Spencer

llvm-svn: 12524
 2004-03-29 00:17:20 +00:00
Chris Lattner dd7ce6c31a Make error message a bit nicer. 
Contributed by Reid Spencer

llvm-svn: 12523
 2004-03-29 00:16:01 +00:00
Chris Lattner 04d6d9d606 Add two methods which have been needed for a long time: Type::get(Un)signedVersion 
llvm-svn: 12522
 2004-03-26 21:43:22 +00:00
Chris Lattner 3aa7b9d08c Make sure to get the headers from zlib correctly 
llvm-svn: 12521
 2004-03-26 17:04:53 +00:00
Chris Lattner 56b5051428 X % -1 == X % 1 == 0 
llvm-svn: 12520
 2004-03-26 16:11:24 +00:00
Chris Lattner 57c67b06e9 Two changes: 
#1 is to unconditionally strip constantpointerrefs out of
instruction operands where they are absolutely pointless and inhibit
optimization.  GRRR!

#2 is to implement InstCombine/getelementptr_const.ll

llvm-svn: 12519
 2004-03-25 22:59:29 +00:00
Chris Lattner 3932cab427 New testcase 
llvm-svn: 12518
 2004-03-25 22:59:06 +00:00
Chris Lattner 82d34eb470 Minor efficiency improvement, finegrainify namespacification 
llvm-svn: 12517
 2004-03-25 22:56:03 +00:00
Chris Lattner 998c0674be Add note about setting up linux systems to directly execute LLVM bytecode files 
Fixes PR303

llvm-svn: 12516
 2004-03-25 20:38:40 +00:00
Chris Lattner 3922d607a5 Hrm, we were leaking ~1M of garbage that valgrind never told us about because 
it was "reachable".  Cute.

llvm-svn: 12515
 2004-03-19 23:34:33 +00:00
Chris Lattner 9e0b81c3e0 Add a dose of reality into the release notes 
llvm-svn: 12510
 2004-03-19 19:22:28 +00:00
Chris Lattner abb77c9959 Teach the optimizer to delete zero sized alloca's (but not mallocs!) 
llvm-svn: 12507
 2004-03-19 06:08:10 +00:00
Chris Lattner 597f1da506 New testcase, the optimizer can delete zero sized allocas 
llvm-svn: 12506
 2004-03-19 06:07:43 +00:00
Chris Lattner 91f297d4fe Perhaps the last bugfix before the release? HOPEFULLY! :) 
llvm-svn: 12505
 2004-03-19 04:26:04 +00:00
Chris Lattner ba6641d598 Fix PR299: [Sparc] Code generator asserts on alloc of zero size type 
llvm-svn: 12504
 2004-03-19 04:21:43 +00:00
Chris Lattner 6ca9b89abb Malloc doesn't kill a load. This patch need not go into 1.2 though. 
llvm-svn: 12500
 2004-03-18 17:01:26 +00:00
Chris Lattner 0b41ebe65c Fix a minor wording bug in the manual 
llvm-svn: 12496
 2004-03-18 14:58:55 +00:00
Chris Lattner 211ae0bf05 This is the last remaining bug that I know of in the loop extractor. The loop 
extractor has extracted hundreds of loops from the SPEC benchmarks without crashing
and without misoptimizing the programs.

This testcase could be reduced substantially more by hand, but I don't have time
to work on it right now.

llvm-svn: 12495
 2004-03-18 06:41:59 +00:00
Chris Lattner dc47e27188 Fix a really nasty bug that was breaking ijpeg in LLC mode. We were incorrectly 
folding load instructions into other instructions across free instruction
boundaries.  Perhaps this will also fix the other strange failures?

llvm-svn: 12494
 2004-03-18 06:29:54 +00:00
Chris Lattner 232155dc1b Fix bug: CodeExtractor/2004-03-17-MissedLiveIns.ll 
With this fix we now successfully extract all 149 loops from 256.bzip2 without
crashing or miscompiling the program!

llvm-svn: 12493
 2004-03-18 05:56:32 +00:00
Chris Lattner 2bcfcc90c8 New testcase, reduced from 256.bzip2 
llvm-svn: 12492
 2004-03-18 05:55:29 +00:00
Chris Lattner e83693560a Add statistics to the loop extractor. The loop extractor has successfully 
extracted all 63 loops for Olden/bh without crashing and without
miscompiling the program!!!

llvm-svn: 12491
 2004-03-18 05:46:10 +00:00
Chris Lattner 5bce0c807d Fix problem with PHI nodes having multiple predecessors from different 
exit nodes

llvm-svn: 12490
 2004-03-18 05:43:18 +00:00
Chris Lattner acd75986ee Fix CodeExtractor/2004-03-17-UpdatePHIsOutsideRegion.ll 
llvm-svn: 12489
 2004-03-18 05:38:31 +00:00
Chris Lattner 4c7fcbf77b New testcase 
llvm-svn: 12488
 2004-03-18 05:33:39 +00:00
Chris Lattner 320d59f4cd Seriously simplify and correct the PHI node handling code. 
llvm-svn: 12487
 2004-03-18 05:28:49 +00:00
Chris Lattner d8017a340d Fix CodeExtractor/2004-03-17-OutputMismatch.ll 
llvm-svn: 12486
 2004-03-18 04:12:05 +00:00
Chris Lattner b7b099d569 New testcase that crashes the code extractor 
llvm-svn: 12485
 2004-03-18 04:10:02 +00:00
Chris Lattner 37de257ef0 Fix several bugs in the extractor: 
1. Names were not put on the new arguments created (ok, this just helps sanity :)
2. Fix outgoing pointer values
3. Do not insert stores for values that had not been computed
4. Fix some wierd problems with the outset calculation

This fixes CodeExtractor/2004-03-14-DominanceProblem.ll, making the extractor
work on at least one simple case!

llvm-svn: 12484
 2004-03-18 03:49:40 +00:00
Chris Lattner e9235d2dde The code extractor needs dominator info. Provide it 
llvm-svn: 12483
 2004-03-18 03:48:06 +00:00
Chris Lattner 835a759686 The code extractor needs dominator information, so we provide it. 
llvm-svn: 12482
 2004-03-18 03:47:46 +00:00
Chris Lattner cee3404d0a Prune #includes, moving the module interface to the front. Note that this 
exposed the fact that the header was not self-contained.  There is a reason
we do things :)

llvm-svn: 12481
 2004-03-18 03:15:29 +00:00
Chris Lattner 5b460d2ad7 Make this header file self-contained 
llvm-svn: 12480
 2004-03-18 03:14:56 +00:00
Chris Lattner 3022b1bc2c Fix a HORRIBLY NASTY bug that caused siod to stop working last night. 
llvm-svn: 12479
 2004-03-17 23:22:04 +00:00
Chris Lattner baf3f62819 Fix a tag-o 
llvm-svn: 12477
 2004-03-17 21:33:32 +00:00
Chris Lattner 8e8716518e Rewrite the second on AnalysisUsage usage. This documents the new 
addRequiredTransitive member that Misha added, and explains the whole
concept a lot better.  Also, the document used incorrect "subsubsection"
tags instead of "doc_subsubsection" which this fixes.

llvm-svn: 12474
 2004-03-17 21:09:55 +00:00
Chris Lattner b38cc9c026 When loop extraction succeeds, make sure to map the function pointers over 
to avoid dangling references.

llvm-svn: 12470
 2004-03-17 17:42:09 +00:00
Chris Lattner 7e51a502bc Fix an inverted condition that causes us to think that loop extraction 
accomplished something when it really did not.  This does not fix the bigger problem tho.

llvm-svn: 12469
 2004-03-17 17:37:18 +00:00
Chris Lattner 425726d9c2 Fix an iterator invalidation problem in a "buggy" pass 
llvm-svn: 12468
 2004-03-17 17:29:08 +00:00
Chris Lattner 2820235689 update release notes for 1.2 
llvm-svn: 12467
 2004-03-17 03:54:41 +00:00
John Criswell d064f2124e Modify test to use the %t substition (temporary name). This is available 
in the QMTest Testrunner tests.
Please note that putting output files in the Output directory no longer
works, as QMTest does not build Output directories anymore (nor does the
test run in a separate subdirectory, anyway).

llvm-svn: 12466
 2004-03-17 02:31:29 +00:00
Chris Lattner a078f47b39 Fix compilation of mesa, which I broke earlier today 
llvm-svn: 12465
 2004-03-17 02:02:47 +00:00
Chris Lattner 684fa5ac64 Be more accurate 
llvm-svn: 12464
 2004-03-17 01:59:27 +00:00
Chris Lattner 10a8d735c8 cleanup comment 
llvm-svn: 12463
 2004-03-17 01:29:36 +00:00
Alkis Evlogimenos 02a5354d9b Make the set of fixed (preallocated) intervals be a fixed superset of 
unhandled + handled. So unhandled is now including all fixed intervals
and fixed intervals never changes when processing a function.

llvm-svn: 12462
 2004-03-17 00:48:59 +00:00
Chris Lattner a3783a577e Fix bug in previous checkin 
llvm-svn: 12458
 2004-03-16 23:36:49 +00:00
Chris Lattner 95057f6ad1 Okay, so there is no reasonable way for tail duplication to update SSA form, 
as it is making effectively arbitrary modifications to the CFG and we don't
have a domset/domfrontier implementations that can handle the dynamic updates.
Instead of having a bunch of code that doesn't actually work in practice,
just demote any potentially tricky values to the stack (causing the problem
to go away entirely).  Later invocations of mem2reg will rebuild SSA for us.

This fixes all of the major performance regressions with tail duplication
from LLVM 1.1.  For example, this loop:

---
int popcount(int x) {
  int result = 0;
  while (x != 0) {
    result = result + (x & 0x1);
    x = x >> 1;
  }
  return result;
}
---
Used to be compiled into:

int %popcount(int %X) {
entry:
	br label %loopentry

loopentry:		; preds = %entry, %no_exit
	%x.0 = phi int [ %X, %entry ], [ %tmp.9, %no_exit ]		; <int> [#uses=3]
	%result.1.0 = phi int [ 0, %entry ], [ %tmp.6, %no_exit ]		; <int> [#uses=2]
	%tmp.1 = seteq int %x.0, 0		; <bool> [#uses=1]
	br bool %tmp.1, label %loopexit, label %no_exit

no_exit:		; preds = %loopentry
	%tmp.4 = and int %x.0, 1		; <int> [#uses=1]
	%tmp.6 = add int %tmp.4, %result.1.0		; <int> [#uses=1]
	%tmp.9 = shr int %x.0, ubyte 1		; <int> [#uses=1]
	br label %loopentry

loopexit:		; preds = %loopentry
	ret int %result.1.0
}

And is now compiled into:

int %popcount(int %X) {
entry:
        br label %no_exit

no_exit:                ; preds = %entry, %no_exit
        %x.0.0 = phi int [ %X, %entry ], [ %tmp.9, %no_exit ]          ; <int> [#uses=2]
        %result.1.0.0 = phi int [ 0, %entry ], [ %tmp.6, %no_exit ]             ; <int> [#uses=1]
        %tmp.4 = and int %x.0.0, 1              ; <int> [#uses=1]
        %tmp.6 = add int %tmp.4, %result.1.0.0          ; <int> [#uses=2]
        %tmp.9 = shr int %x.0.0, ubyte 1                ; <int> [#uses=2]
        %tmp.1 = seteq int %tmp.9, 0            ; <bool> [#uses=1]
        br bool %tmp.1, label %loopexit, label %no_exit

loopexit:               ; preds = %no_exit
        ret int %tmp.6
}

llvm-svn: 12457
 2004-03-16 23:29:09 +00:00
Chris Lattner bb1a2cc7ab This code was both incredibly complex and incredibly broken. Fix it. 
llvm-svn: 12456
 2004-03-16 23:23:11 +00:00
Chris Lattner 0e64dda7ca Another simple testcase 
llvm-svn: 12455
 2004-03-16 23:07:52 +00:00
Brian Gaeke d962dd5528 The .type directive on Solaris uses the # character instead of @. 
llvm-svn: 12454
 2004-03-16 22:52:04 +00:00
Brian Gaeke 3fb0e203cd Fix bug in zero-extending of shorts. 
llvm-svn: 12453
 2004-03-16 22:45:42 +00:00
Brian Gaeke ae22ce5370 Add UMULrr and SMULrr instructions. 
llvm-svn: 12452
 2004-03-16 22:37:13 +00:00
Brian Gaeke ae0ca8baa6 Use ! for comment char; it works in both Solaris as and GAS. 
llvm-svn: 12451
 2004-03-16 22:37:12 +00:00
Brian Gaeke 67fcefbfef Make getClass more robust by adding cLong. 
Add handling for Mul instruction.

llvm-svn: 12450
 2004-03-16 22:37:11 +00:00
Chris Lattner 64f2517c07 Bug fixed 
llvm-svn: 12449
 2004-03-16 21:55:23 +00:00
Chris Lattner bd10066d97 New testcase for PR297 
llvm-svn: 12448
 2004-03-16 21:52:24 +00:00
Brian Gaeke 33e83b64b4 Add a better head-of-file comment. 
llvm-svn: 12447
 2004-03-16 21:47:20 +00:00
Chris Lattner 1ab58dd802 Bug not fixed after all 
llvm-svn: 12446
 2004-03-16 21:39:04 +00:00
Chris Lattner 618c3990c7 Fix test 
llvm-svn: 12445
 2004-03-16 21:10:15 +00:00
Chris Lattner a0fb904d18 Fix testcase 
llvm-svn: 12444
 2004-03-16 20:04:55 +00:00
Chris Lattner fa48edfb7d Punt if we see gigantic PHI nodes. This improves a huge interpreter loop 
testcase from 32.5s in -raise to take .3s

llvm-svn: 12443
 2004-03-16 19:52:53 +00:00
Chris Lattner 7a7b114871 Do not try to optimize PHI nodes with incredibly high degree. This reduces SCCP 
time from 615s to 1.49s on a large testcase that has a gigantic switch statement
that all of the blocks in the function go to (an intepreter).

llvm-svn: 12442
 2004-03-16 19:49:59 +00:00
Chris Lattner a64923ad26 Do not copy gigantic switch instructions 
llvm-svn: 12441
 2004-03-16 19:45:22 +00:00
Chris Lattner 0ea6017b87 Fix bug in bug name. It must be the painkillers that I haven't been taking 
llvm-svn: 12440
 2004-03-16 09:01:10 +00:00
Chris Lattner d3dc850c6c Implement a new feature in the CFE, moving a GCC extension from the unsupported 
to the supported list

llvm-svn: 12439
 2004-03-16 08:54:07 +00:00
Chris Lattner 91d855f150 New testcase 
llvm-svn: 12437
 2004-03-16 08:49:48 +00:00