Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | consolidate LinkLibraries into LinkItems | Chris Lattner | 2005-03-15 | 2 | -76/+57 |
| | | | | llvm-svn: 20629 | ||||
* | make sure to mark nodes in the globals graph incomplete after computing it | Chris Lattner | 2005-03-15 | 1 | -0/+1 |
| | | | | | | so that external globals (and whatever they point to) are marked incomplete. llvm-svn: 20628 | ||||
* | fix crashes when we only have a prototype for main. | Chris Lattner | 2005-03-15 | 3 | -3/+3 |
| | | | | llvm-svn: 20627 | ||||
* | Fix a crash that happens when mapping something like this: | Chris Lattner | 2005-03-15 | 1 | -6/+12 |
| | | | | | | | | | | | | | { short, short } to short where the second short maps onto the second field of the first struct. In this case, the struct index is not aligned, so we should avoid calling getLink(2), which asserts out. llvm-svn: 20626 | ||||
* | sure, I can set a flag, but if I never check it, why bother setting it? ↵ | Andrew Lenharth | 2005-03-15 | 2 | -3/+5 |
| | | | | | | Should fix 20 programs :) llvm-svn: 20623 | ||||
* | Make computeGGToGMapping compute an invnodemap | Chris Lattner | 2005-03-15 | 1 | -9/+11 |
| | | | | llvm-svn: 20622 | ||||
* | Finally fix (the right way) the problem where functions like this: | Chris Lattner | 2005-03-15 | 1 | -0/+24 |
| | | | | | | | | | | | void foo() { G = 1; } would have an empty DSGraph even though G (a global) is directly used in the function. llvm-svn: 20619 | ||||
* | Start using retnodes_* for iteration. | Chris Lattner | 2005-03-15 | 6 | -39/+34 |
| | | | | llvm-svn: 20618 | ||||
* | avoid varialbe name collisions | Chris Lattner | 2005-03-15 | 1 | -2/+3 |
| | | | | llvm-svn: 20606 | ||||
* | stop using method. | Chris Lattner | 2005-03-15 | 1 | -1/+1 |
| | | | | llvm-svn: 20603 | ||||
* | stop using arg_front | Chris Lattner | 2005-03-15 | 2 | -2/+2 |
| | | | | llvm-svn: 20599 | ||||
* | stop using arg_back | Chris Lattner | 2005-03-15 | 1 | -1/+1 |
| | | | | llvm-svn: 20598 | ||||
* | This mega patch converts us from using Function::a{iterator|begin|end} to | Chris Lattner | 2005-03-15 | 62 | -169/+169 |
| | | | | | | | | using Function::arg_{iterator|begin|end}. Likewise Module::g* -> Module::global_*. This patch is contributed by Gabor Greif, thanks! llvm-svn: 20597 | ||||
* | Don't crash if computing a mapping to a node with zero size | Chris Lattner | 2005-03-15 | 1 | -0/+2 |
| | | | | llvm-svn: 20595 | ||||
* | rename method, add counterpart | Chris Lattner | 2005-03-15 | 2 | -4/+16 |
| | | | | llvm-svn: 20593 | ||||
* | FP 0.0 setcc optimization, and generate short branch sequence for setcc(FP) ↵ | Andrew Lenharth | 2005-03-14 | 2 | -19/+37 |
| | | | | | | rather than stack usage llvm-svn: 20589 | ||||
* | add a method to compute a commonly used mapping. | Chris Lattner | 2005-03-14 | 2 | -5/+16 |
| | | | | llvm-svn: 20588 | ||||
* | regardless of whether or not the client things we should mark globals ↵ | Chris Lattner | 2005-03-13 | 1 | -7/+7 |
| | | | | | | | | incomplete, ALWAYS mark them incomplete if they are external! llvm-svn: 20586 | ||||
* | Make sure to remove incomplete markers before we add to them! :) | Chris Lattner | 2005-03-13 | 3 | -0/+3 |
| | | | | llvm-svn: 20585 | ||||
* | The incoming arguments to main (the argv list) are not complete! | Chris Lattner | 2005-03-13 | 1 | -4/+3 |
| | | | | llvm-svn: 20584 | ||||
* | After finishing BU analysis, move all global variables from the globals | Chris Lattner | 2005-03-13 | 3 | -6/+71 |
| | | | | | | graph into main and mark them complete. llvm-svn: 20583 | ||||
* | ADd support for printing eqgraphs. | Chris Lattner | 2005-03-13 | 2 | -2/+8 |
| | | | | llvm-svn: 20582 | ||||
* | Replace linear search with logrithmic one. | Chris Lattner | 2005-03-13 | 1 | -4/+1 |
| | | | | llvm-svn: 20580 | ||||
* | add a StructLayout::getElementContainingOffset method. | Chris Lattner | 2005-03-13 | 1 | -0/+17 |
| | | | | llvm-svn: 20579 | ||||
* | Should fix mesa | Andrew Lenharth | 2005-03-13 | 1 | -1/+1 |
| | | | | llvm-svn: 20577 | ||||
* | make sure to mark nodes returned from functions as incomplete | Chris Lattner | 2005-03-12 | 1 | -0/+1 |
| | | | | llvm-svn: 20576 | ||||
* | remove this from the PA namespace, leaving it in the llvm ns | Chris Lattner | 2005-03-12 | 1 | -8/+7 |
| | | | | llvm-svn: 20574 | ||||
* | Move this from the pool allocator project to here, where it logically belongs. | Chris Lattner | 2005-03-12 | 1 | -0/+442 |
| | | | | llvm-svn: 20570 | ||||
* | remove a pseudo instruction and improve inline constant generation | Andrew Lenharth | 2005-03-11 | 2 | -6/+29 |
| | | | | llvm-svn: 20563 | ||||
* | I didn't mean to check this in. :( | Chris Lattner | 2005-03-10 | 1 | -3/+1 |
| | | | | llvm-svn: 20555 | ||||
* | Fix a bug where we would incorrectly do a sign ext instead of a zero ext | Chris Lattner | 2005-03-10 | 1 | -1/+1 |
| | | | | | | | because we were checking the wrong thing. Thanks to andrew for pointing this out! llvm-svn: 20554 | ||||
* | yay for camel_cvs diff | Andrew Lenharth | 2005-03-10 | 1 | -1/+1 |
| | | | | llvm-svn: 20552 | ||||
* | some typoes and .bss isn't liked, at all | Andrew Lenharth | 2005-03-10 | 1 | -4/+4 |
| | | | | llvm-svn: 20542 | ||||
* | Allow the live interval analysis pass to be a bit more aggressive about | Chris Lattner | 2005-03-09 | 2 | -10/+62 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | numbering values in live ranges for physical registers. The alpha backend currently generates code that looks like this: vreg = preg ... preg = vreg use preg ... preg = vreg use preg etc. Because vreg contains the value of preg coming in, each of the copies back into preg contain that initial value as well. In the case of the Alpha, this allows this testcase: void "foo"(int %blah) { store int 5, int *%MyVar store int 12, int* %MyVar2 ret void } to compile to: foo: ldgp $29, 0($27) ldiq $0,5 stl $0,MyVar ldiq $0,12 stl $0,MyVar2 ret $31,($26),1 instead of: foo: ldgp $29, 0($27) bis $29,$29,$0 ldiq $1,5 bis $0,$0,$29 stl $1,MyVar ldiq $1,12 bis $0,$0,$29 stl $1,MyVar2 ret $31,($26),1 This does not seem to have any noticable effect on X86 code. This fixes PR535. llvm-svn: 20536 | ||||
* | minor correction for Register that isn't used | Andrew Lenharth | 2005-03-09 | 1 | -2/+2 |
| | | | | llvm-svn: 20535 | ||||
* | constant fold FP_ROUND_INREG, ZERO_EXTEND_INREG, and SIGN_EXTEND_INREG | Chris Lattner | 2005-03-09 | 1 | -0/+12 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This allows the alpha backend to compile: bool %test(uint %P) { %c = seteq uint %P, 0 ret bool %c } into: test: ldgp $29, 0($27) ZAP $16,240,$0 CMPEQ $0,0,$0 AND $0,1,$0 ret $31,($26),1 instead of: test: ldgp $29, 0($27) ZAP $16,240,$0 ldiq $1,0 ZAP $1,240,$1 CMPEQ $0,$1,$0 AND $0,1,$0 ret $31,($26),1 ... and fixes PR534. llvm-svn: 20534 | ||||
* | correct the computation of the isAbstract bit for types. | Chris Lattner | 2005-03-09 | 1 | -3/+8 |
| | | | | llvm-svn: 20533 | ||||
* | Fix Regression/Analysis/BasicAA/2005-03-09-BrokenBasicAA.ll, a miscompilation | Chris Lattner | 2005-03-09 | 1 | -2/+6 |
| | | | | | | that Alkis found with Java, thanks Alkis! llvm-svn: 20531 | ||||
* | Fix a typo in an assertion comment. | Reid Spencer | 2005-03-09 | 1 | -1/+1 |
| | | | | | | Patch contributed by Vladimir Merzliakov. llvm-svn: 20529 | ||||
* | Export two methods for getting -X and A-B. | Chris Lattner | 2005-03-09 | 1 | -10/+11 |
| | | | | llvm-svn: 20527 | ||||
* | get bugpoint working on ia64, by building .so's with -fpic. :) | Chris Lattner | 2005-03-09 | 1 | -0/+4 |
| | | | | llvm-svn: 20525 | ||||
* | Fix a crash handling 'undef bool', fixing an llc crash on 186.crafty | Chris Lattner | 2005-03-08 | 1 | -1/+1 |
| | | | | llvm-svn: 20523 | ||||
* | Patch to make assembly output compatible with mingw compilation (identical | Reid Spencer | 2005-03-08 | 1 | -2/+3 |
| | | | | | | to cygwin) llvm-svn: 20520 | ||||
* | Make sure to remove all dead type names from the symbol table, not just | Chris Lattner | 2005-03-08 | 1 | -9/+8 |
| | | | | | | | struct types. This fixes Regression/CodeGen/CBackend/2005-03-08-RecursiveTypeCrash.ll, a crash on Java output that Alkis reported. llvm-svn: 20519 | ||||
* | Single characters should be printed out as chars, not strings. | Misha Brukman | 2005-03-08 | 1 | -2/+2 |
| | | | | llvm-svn: 20515 | ||||
* | Fix an apparent ambiguity compiling on PPC | Chris Lattner | 2005-03-07 | 1 | -1/+1 |
| | | | | llvm-svn: 20507 | ||||
* | fix a bug where we thought arguments were constants :( | Chris Lattner | 2005-03-06 | 1 | -3/+7 |
| | | | | llvm-svn: 20506 | ||||
* | Fix Regression/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll, | Chris Lattner | 2005-03-06 | 1 | -1/+1 |
| | | | | | | hopefully not breaking too many other things. llvm-svn: 20505 | ||||
* | implement Transforms/LoopStrengthReduce/invariant_value_first_arg.ll | Chris Lattner | 2005-03-06 | 1 | -1/+1 |
| | | | | llvm-svn: 20501 | ||||
* | minor simplifications of the code. | Chris Lattner | 2005-03-06 | 1 | -9/+8 |
| | | | | llvm-svn: 20497 |