summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* Do not replace ostream << Module*, only ostream << Module&.Chris Lattner2005-02-131-5/+0
| | | | llvm-svn: 20157
* Print the module, not the pointer.Chris Lattner2005-02-132-2/+2
| | | | llvm-svn: 20156
* Make the check for global variables the same as the one for functions. InReid Spencer2005-02-131-1/+1
| | | | | | | | both cases they are looking for non-external variables/functions that do not have internal linkage. Using "!isExternal()" is a little more understandable than "hasInitializer()" llvm-svn: 20155
* Nuke blank line.Chris Lattner2005-02-131-1/+0
| | | | llvm-svn: 20154
* Minor cleanup. No need to explicitly tell the compiler the template arguments.Chris Lattner2005-02-131-3/+2
| | | | llvm-svn: 20153
* Make sure to clear the LazyFunctionLoadMap after we ParseAllFunctionBodies.Chris Lattner2005-02-131-1/+2
| | | | | | | | | Otherwise, clients who call ParseAllFunctionBodies will attempt to parse the function bodies twice, which is (uh) very very bad (tm). This fixes gccld on python. llvm-svn: 20152
* Do not put internal symbols into the symbol table. This shrinks the symbolChris Lattner2005-02-131-21/+11
| | | | | | | | table for archives in common cases, and prevents trying to resolve a external reference with an internal reference. This shrinks the libpython.a symbol table from 126302 to 19770 bytes. llvm-svn: 20151
* Include local time on the web page for start/end times.Chris Lattner2005-02-132-4/+5
| | | | llvm-svn: 20150
* Print something useful for gccld -v with an archive.Chris Lattner2005-02-131-0/+2
| | | | llvm-svn: 20148
* Fix PR506:Reid Spencer2005-02-131-4/+6
| | | | | | | * Use error returned from Archive::OpenAndLoad * Make sure only std::string is thrown so it gets caught and printed. llvm-svn: 20147
* Correct the recursive PHI node handling routines in a way that CANNOT induceChris Lattner2005-02-131-27/+83
| | | | | | | | | | | | | | | infinite loops (using the new replaceSymbolicValuesWithConcrete method). This patch reverts this patch: http://mail.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20050131/023830.html ... which was an attempted fix for this problem. Unfortunately, that patch caused test/Regression/Transforms/IndVarsSimplify/exit_value_tests.llx to fail and slightly castrated the entire analysis. This patch fixes it right. This patch is dedicated to jeffc, for making me deal with this. :) llvm-svn: 20146
* Add a new replaceSymbolicValuesWithConcrete method to the SCEV class,Chris Lattner2005-02-132-5/+59
| | | | | | adjust const'ness a bit to be more correct. llvm-svn: 20145
* non-ieee arith crashes passes on alphaAndrew Lenharth2005-02-131-0/+6
| | | | llvm-svn: 20144
* try to do better match for i32 addsAndrew Lenharth2005-02-121-1/+22
| | | | llvm-svn: 20143
* make FP conversion more conservative (matches gcc)Andrew Lenharth2005-02-121-3/+4
| | | | llvm-svn: 20142
* oops, I was sure this had already gond though the nightly testerAndrew Lenharth2005-02-121-1/+1
| | | | llvm-svn: 20141
* Add note about broken GCCChris Lattner2005-02-121-1/+6
| | | | llvm-svn: 20138
* added sign extend for booleanAndrew Lenharth2005-02-121-3/+5
| | | | llvm-svn: 20137
* Document the newly generalized model for constant globals.Chris Lattner2005-02-121-6/+16
| | | | llvm-svn: 20136
* Allow globals to be of different const'nesses when we link.Chris Lattner2005-02-121-8/+2
| | | | | | | This finally resolves PR502, PR450, and test/Regression/Linker/2005-02-12-ConstantGlobals{,-2}.ll correctly llvm-svn: 20135
* New test, adjust other test.Chris Lattner2005-02-122-1/+9
| | | | llvm-svn: 20134
* Reinstate this test, it is correct after all.Chris Lattner2005-02-121-0/+8
| | | | llvm-svn: 20133
* This test is invalidChris Lattner2005-02-121-12/+0
| | | | llvm-svn: 20131
* Fix for testcase Transforms/IndVarsSimplify/2005-02-11-InvokeCrash.llChris Lattner2005-02-121-0/+2
| | | | | | and PR504. llvm-svn: 20129
* new testcase for PR504Chris Lattner2005-02-121-0/+22
| | | | llvm-svn: 20128
* Actually, this testcase is invalid, the bug is in the C++ FE.Chris Lattner2005-02-111-8/+0
| | | | llvm-svn: 20123
* New testcase that manyoso found compiling QT 4.0.Chris Lattner2005-02-111-0/+8
| | | | llvm-svn: 20122
* Slightly expand on the instructions.Chris Lattner2005-02-111-1/+3
| | | | llvm-svn: 20121
* Bug fixedChris Lattner2005-02-111-0/+2
| | | | llvm-svn: 20118
* New testcase for PR501Chris Lattner2005-02-111-0/+32
| | | | llvm-svn: 20117
* The pool allocator is now the llvm-poolalloc module in public CVSChris Lattner2005-02-112-4/+4
| | | | llvm-svn: 20116
* fix a bunch of regressions due to call behaviorAndrew Lenharth2005-02-101-4/+4
| | | | llvm-svn: 20110
* Localize globals if they are only used in main(). This replaces theAlkis Evlogimenos2005-02-101-0/+40
| | | | | | | global with an alloca, which eventually gets promoted into a register. This enables a lot of other optimizations later on. llvm-svn: 20109
* Added new circuit finding alogrithm.Tanya Lattner2005-02-104-51/+487
| | | | | | Fixed bug in graph so that phi ite diff edges are added. llvm-svn: 20108
* Allow modsched and local scheduling to both be run.Tanya Lattner2005-02-101-4/+3
| | | | llvm-svn: 20107
* so, if you beat on it, you too can talk emacs into having a sane indenting ↵Andrew Lenharth2005-02-103-549/+575
| | | | | | policy... Also, optimize many function calls with pc-relative calls (partial prologue skipping for that case coming soon), try to fix the random jumps to strange places problem by pesimizing div et. al. register usage and fixing up GP before using, some calling convention tweaks, and make frame pointer unallocatable (not strickly necessary, but let's go for correctness first) llvm-svn: 20106
* fix fp branchAndrew Lenharth2005-02-101-6/+6
| | | | llvm-svn: 20105
* Give props to Brian for V8 backendMisha Brukman2005-02-101-0/+1
| | | | llvm-svn: 20104
* * Fix spelling of `volatile'Misha Brukman2005-02-101-9/+9
| | | | | | * Align comments with tablegen elements llvm-svn: 20103
* This will be moved elsewhere.Chris Lattner2005-02-091-8/+1
| | | | llvm-svn: 20101
* * Use doc_code style for blocks of codeMisha Brukman2005-02-091-14/+16
| | | | | | | | * Convert < to &lt; in <pre> block * Use doc_hilite instead of doc_red style * Put some classes and options in <tt> blocks llvm-svn: 20100
* Do not mention colors in style names, we might want to change styles themselvesMisha Brukman2005-02-091-1/+1
| | | | llvm-svn: 20099
* * Use an HTML list <ol> instead of custom superscripts in the platform notesMisha Brukman2005-02-091-45/+44
| | | | | | | | * Reorder software requirements list to show relative importance * Use superscripts for footnotes in software list rather than "Note 1" * Use new doc_notes class to separate footnotes from main doc text llvm-svn: 20098
* Add style to display notes on a different background than the main textMisha Brukman2005-02-091-0/+1
| | | | llvm-svn: 20097
* Turn footnotes into links, make the page validate.Chris Lattner2005-02-091-18/+23
| | | | llvm-svn: 20096
* If needed, this should be in the 'how to compile the cfe' doc.Chris Lattner2005-02-091-3/+1
| | | | llvm-svn: 20095
* this should be incorporated into the buildChris Lattner2005-02-091-1/+0
| | | | llvm-svn: 20094
* * Separate superscript footnotes with a comma, otherwise it's a 2-digit numberMisha Brukman2005-02-091-14/+18
| | | | | | | | | | * Make a URL into an HTML link using anchor tag * Put code in fixed-width font * Since the MinGW footnote is significantly more than the rest of the footnotes combined, it needs to have a border and background color to separate it from the regular flow of text for readability purposes. llvm-svn: 20093
* Add instructions for MINGW, contributed by Henrik BachChris Lattner2005-02-091-1/+17
| | | | llvm-svn: 20092
* Don't print a 'Total Execution Time' line for the 'Miscellaneous UngroupedChris Lattner2005-02-091-6/+14
| | | | | | | Timers' section. Since these are random timers in the program it doesn't make sense to sum them up. llvm-svn: 20090
OpenPOWER on IntegriCloud