- Oct 26, 2010
-
-
Jakob Stoklund Olesen authored
llvm-svn: 117337
-
Bob Wilson authored
do not double-count the duplicate instructions by counting once from the beginning and again from the end. Keep track of where the duplicates from the beginning ended and don't go past that point when counting duplicates at the end. Radar 8589805. This change causes one of the MC/ARM/simple-fp-encoding tests to produce different (better!) code without the vmovne instruction being tested. I changed the test to produce vmovne and vmoveq instructions but moving between register files in the opposite direction. That's not quite the same but predicated versions of those instructions weren't being tested before, so at least the test coverage is not any worse, just different. llvm-svn: 117333
-
Bob Wilson authored
instructions separately from the count of non-predicated instructions. The instruction count is used in places to determine how many instructions to copy, predicate, etc. and things get confused if that count includes the extra cost for microcoded ops. llvm-svn: 117332
-
Bob Wilson authored
llvm-svn: 117331
-
Evan Cheng authored
llvm-svn: 117329
-
Rafael Espindola authored
the existing streamer methods that are endian safe. llvm-svn: 117323
-
Dale Johannesen authored
cannot use tailcall. PR 8461. llvm-svn: 117322
-
- Oct 25, 2010
-
-
Dan Gohman authored
llvm-svn: 117317
-
Devang Patel authored
Do not count use of sdisel for single call instruction. llvm-svn: 117316
-
Owen Anderson authored
llvm-svn: 117315
-
Dan Gohman authored
llvm-svn: 117314
-
Devang Patel authored
Add counters to count how many basic blocks are entirely selected by fastisel. llvm-svn: 117310
-
Owen Anderson authored
llvm-svn: 117309
-
Devang Patel authored
Add simple counter to count no. of basic blocks without any line number information. At -O0, these basic block coule cause less than optimial debugging experience. llvm-svn: 117307
-
Dan Gohman authored
bits open for future uses. llvm-svn: 117301
-
Daniel Dunbar authored
llvm-svn: 117299
-
Daniel Dunbar authored
llvm-svn: 117298
-
Owen Anderson authored
Attempt to provide correct encodings for NEON vbit and vbif, even though we can't test them at the moment. llvm-svn: 117294
-
Owen Anderson authored
llvm-svn: 117293
-
Jim Grosbach authored
llvm-svn: 117291
-
Dan Gohman authored
llvm-svn: 117288
-
Owen Anderson authored
llvm-svn: 117282
-
Rafael Espindola authored
llvm-svn: 117280
-
Owen Anderson authored
llvm-svn: 117279
-
Owen Anderson authored
llvm-svn: 117276
-
Rafael Espindola authored
llvm-svn: 117275
-
Owen Anderson authored
llvm-svn: 117274
-
Owen Anderson authored
llvm-svn: 117273
-
Jakob Stoklund Olesen authored
llvm-svn: 117272
-
Dan Gohman authored
llvm-svn: 117268
-
Dan Gohman authored
llvm-svn: 117266
-
Dan Gohman authored
from stores when folding in bitcasts. llvm-svn: 117265
-
Charles Davis authored
instruction at the beginning of each function that has the attribute, allowing the function to be easily hooked and/or patched. llvm-svn: 117264
-
Duncan Sands authored
Rather than rename this instance, use the cast-to-void idiom instead. This will hopefully fix the windows buildbots. llvm-svn: 117262
-
- Oct 24, 2010
-
-
Rafael Espindola authored
we check for _GLOBAL_OFFSET_TABLE_. llvm-svn: 117241
-
Duncan Sands authored
it isn't unreachable and should not be zapped. The check for the entry block was missing in one case: a block containing a unwind instruction. While there, do some small cleanups: "M" is not a great name for a Function* (it would be more appropriate for a Module*), change it to "Fn"; use Fn in more places. llvm-svn: 117224
-
- Oct 23, 2010
-
-
Benjamin Kramer authored
llvm-svn: 117213
-
Eric Christopher authored
note that we can actually handle some f64 arguments. llvm-svn: 117209
-
Benjamin Kramer authored
llvm-svn: 117206
-
Michael J. Spencer authored
llvm-svn: 117205
-