| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Wrap to 80 cols | Chris Lattner | 2006-06-09 | 1 | -4/+5 |
| | | | | | llvm-svn: 28743 | ||||
| * | Don't build tblgen with -pedantic or -Wno-long-long | Reid Spencer | 2006-06-01 | 1 | -0/+4 |
| | | | | | llvm-svn: 28638 | ||||
| * | Can't trust NodeDepth when checking for possibility of load folding creating | Evan Cheng | 2006-05-25 | 1 | -7/+5 |
| | | | | | | | | | a cycle. This increase the search space and will increase compile time (in practice it appears to be small, e.g. 176.gcc goes from 62 sec to 65 sec) that will be addressed later. llvm-svn: 28476 | ||||
| * | Fixed a really ugly bug. The TableGen'd isel is not freeing the "inflight set" | Evan Cheng | 2006-05-25 | 1 | -9/+21 |
| | | | | | | | | correctly. That is causing non-deterministic behavior (and possibly preventing some load folding from happening). llvm-svn: 28458 | ||||
| * | Don't make zero-sized static arrays | Chris Lattner | 2006-05-24 | 1 | -1/+1 |
| | | | | | llvm-svn: 28448 | ||||
| * | Patches to make the LLVM sources more -pedantic clean. Patch provided | Chris Lattner | 2006-05-24 | 1 | -1/+1 |
| | | | | | | | by Anton Korobeynikov! This is a step towards closing PR786. llvm-svn: 28447 | ||||
| * | Now that iPTR is a fully resolved type. We end up losing the type check for | Evan Cheng | 2006-05-19 | 1 | -1/+1 |
| | | | | | | | | | | | | patterns that look like this: def : Pat<(i32 (X86Wrapper tconstpool :$dst)), (MOV32ri tconstpool :$dst)>; InsertOneTypeCheck should copy the type from the resolved pattern to the unresolved one as long as there types are different. llvm-svn: 28389 | ||||
| * | lib/Target/Target.td | Evan Cheng | 2006-05-18 | 1 | -5/+7 |
| | | | | | llvm-svn: 28386 | ||||
| * | Don't generate getCalleeSaveReg and getCalleeSaveRegClasses anymore. | Evan Cheng | 2006-05-18 | 3 | -40/+0 |
| | | | | | llvm-svn: 28376 | ||||
| * | Typo | Evan Cheng | 2006-05-17 | 1 | -2/+2 |
| | | | | | llvm-svn: 28366 | ||||
| * | Remove PointerType from target definition. Use abstract type MVT::iPTR to | Evan Cheng | 2006-05-17 | 4 | -65/+93 |
| | | | | | | | represent pointer type. llvm-svn: 28363 | ||||
| * | Allow patterns to refer to physical registers that belong to multiple | Evan Cheng | 2006-05-16 | 3 | -6/+26 |
| | | | | | | | register classes. llvm-svn: 28323 | ||||
| * | Noop instruction | Evan Cheng | 2006-05-12 | 1 | -2/+0 |
| | | | | | llvm-svn: 28241 | ||||
| * | Unused instruction | Evan Cheng | 2006-05-12 | 1 | -1/+0 |
| | | | | | llvm-svn: 28240 | ||||
| * | Also add super- register classes info. | Evan Cheng | 2006-05-11 | 1 | -0/+38 |
| | | | | | llvm-svn: 28221 | ||||
| * | Watch out for the following case: | Evan Cheng | 2006-05-10 | 1 | -11/+37 |
| | | | | | | | | | 1. Use expects a chain output. 2. Node is expanded into multiple target ops. 3. One of the inner node produces a chain, the outer most one doesn't. llvm-svn: 28209 | ||||
| * | Fix a load folding bug. It is exposed by a multi- resulting instructions | Evan Cheng | 2006-05-10 | 1 | -59/+63 |
| | | | | | | | def : Pat<> pattern. llvm-svn: 28208 | ||||
| * | Add sub-register class information. | Evan Cheng | 2006-05-09 | 1 | -3/+49 |
| | | | | | llvm-svn: 28195 | ||||
| * | Set isStore of instructions with ISD::TRUNCSTORE root node. | Evan Cheng | 2006-05-03 | 1 | -3/+5 |
| | | | | | llvm-svn: 28075 | ||||
| * | Put instruction names into the first non TargetInstrInfo namespace found. | Chris Lattner | 2006-05-01 | 1 | -10/+18 |
| | | | | | llvm-svn: 28043 | ||||
| * | instructions can be in different namespaces. Make sure to use the right | Chris Lattner | 2006-05-01 | 1 | -8/+4 |
| | | | | | | | one for each instruction. llvm-svn: 28038 | ||||
| * | Formating | Evan Cheng | 2006-05-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 28036 | ||||
| * | Mark instructions whose pattern is (store ...) isStore. | Evan Cheng | 2006-05-01 | 1 | -1/+17 |
| | | | | | llvm-svn: 28032 | ||||
| * | Remove the temporary option: -no-isel-fold-inflight | Evan Cheng | 2006-04-28 | 1 | -2/+1 |
| | | | | | llvm-svn: 28012 | ||||
| * | When isel'ing a node, mark its operands "InFlight" before selecting them. These | Evan Cheng | 2006-04-28 | 1 | -2/+34 |
| | | | | | | | | | nodes should not be folded into other nodes. This fixes the miscompilation of PR 749. Temporarily under flag control. llvm-svn: 28002 | ||||
| * | JumpTable support! What this represents is working asm and jit support for | Nate Begeman | 2006-04-22 | 1 | -0/+1 |
| | | | | | | | | | x86 and ppc for 100% dense switch statements when relocations are non-PIC. This support will be extended and enhanced in the coming days to support PIC, and less dense forms of jump tables. llvm-svn: 27947 | ||||
| * | Don't fill in fields that no longer exist. | Chris Lattner | 2006-04-20 | 1 | -2/+1 |
| | | | | | llvm-svn: 27898 | ||||
| * | Rename AddedCost to AddedComplexity. | Evan Cheng | 2006-04-19 | 2 | -13/+15 |
| | | | | | llvm-svn: 27841 | ||||
| * | Allow "let AddedCost = n in" to increase pattern complexity. | Evan Cheng | 2006-04-19 | 2 | -7/+21 |
| | | | | | llvm-svn: 27834 | ||||
| * | Add missing things to the distribution. | Reid Spencer | 2006-04-13 | 1 | -0/+2 |
| | | | | | llvm-svn: 27650 | ||||
| * | Fix a typo: Instr* -> Intr* | Chris Lattner | 2006-04-10 | 1 | -3/+3 |
| | | | | | llvm-svn: 27568 | ||||
| * | Infer element types for shuffle masks | Chris Lattner | 2006-04-06 | 1 | -0/+20 |
| | | | | | llvm-svn: 27456 | ||||
| * | rename a method, to avoid confusion with llvm intrinsics. | Chris Lattner | 2006-04-06 | 1 | -4/+3 |
| | | | | | llvm-svn: 27455 | ||||
| * | Adjust the Intrinsics.gen interface a little bit | Chris Lattner | 2006-04-02 | 1 | -2/+2 |
| | | | | | llvm-svn: 27345 | ||||
| * | regenerate | Chris Lattner | 2006-03-31 | 5 | -423/+421 |
| | | | | | llvm-svn: 27313 | ||||
| * | Generalize the previous binary operator support and add a string concatenation | Chris Lattner | 2006-03-31 | 4 | -38/+109 |
| | | | | | | | operation. This implements Regression/TableGen/strconcat.td. llvm-svn: 27312 | ||||
| * | Allow bits init values to be used in patterns, turn them into ints. | Chris Lattner | 2006-03-31 | 1 | -0/+19 |
| | | | | | llvm-svn: 27286 | ||||
| * | Final bugfix for PR724. GCC won't inline varargs functions, so use one to | Chris Lattner | 2006-03-31 | 1 | -17/+9 |
| | | | | | | | | validate the prototype of intrinsic functions. This prevents GCC from going crazy and inlining too much stuff, eventually running out of memory. llvm-svn: 27283 | ||||
| * | When emitting code for the verifier, instead of emitting each case statement | Chris Lattner | 2006-03-31 | 1 | -9/+42 |
| | | | | | | | | | independently, batch up checks so that identically typed intrinsics share verifier code. This dramatically reduces the size of the verifier function, which should help avoid GCC running out of memory compiling Verifier.cpp. llvm-svn: 27281 | ||||
| * | regenerate | Chris Lattner | 2006-03-30 | 3 | -1663/+1208 |
| | | | | | llvm-svn: 27264 | ||||
| * | Implement Regression/TableGen/DagDefSubst.ll | Chris Lattner | 2006-03-30 | 4 | -41/+45 |
| | | | | | llvm-svn: 27263 | ||||
| * | Don't sort the names before outputing the intrinsic name table. It causes a | Evan Cheng | 2006-03-28 | 1 | -7/+2 |
| | | | | | | | | | mismatch against the enum table. This is a part of Sabre's master plan to drive me nuts with subtle bugs that happens to only affect x86 be. :-) llvm-svn: 27237 | ||||
| * | Tblgen doesn't like multiple SDNode<> definitions that map to the sameenum ↵ | Chris Lattner | 2006-03-28 | 1 | -13/+17 |
| | | | | | | | | | value. Split them into separate enums. Also, don't emit dynamic checks when we can compute them statically llvm-svn: 27202 | ||||
| * | Only compute intrinsic valuetypes when in a target .td file. | Chris Lattner | 2006-03-28 | 2 | -3/+16 |
| | | | | | llvm-svn: 27197 | ||||
| * | revert this, it breaks things. | Chris Lattner | 2006-03-28 | 2 | -5/+4 |
| | | | | | llvm-svn: 27196 | ||||
| * | Add support for decoding iPTR to the right pointer type. | Chris Lattner | 2006-03-27 | 3 | -7/+19 |
| | | | | | llvm-svn: 27188 | ||||
| * | Print error messages like this: | Chris Lattner | 2006-03-27 | 1 | -1/+1 |
| | | | | | | | | | | tblgen: In STVEBX: Intrinsic 'llvm.ppc.altivec.stvebx' expects 3 operands, not 2 operands! instead of like this: tblgen: In STVEBX: Intrinsic 'intrinsic_void expects 3 operands, not 2 operands! llvm-svn: 27185 | ||||
| * | Add a missing check which cause an invalid .td file to crash tblgen | Chris Lattner | 2006-03-25 | 1 | -0/+3 |
| | | | | | llvm-svn: 27126 | ||||
| * | When failing selection for an intrinsic, print this: | Chris Lattner | 2006-03-25 | 1 | -2/+16 |
| | | | | | | | | | | | Cannot yet select: intrinsic %llvm.ppc.altivec.lvx instead of this: Cannot yet select: 0x9b047e0: v4i32,ch = INTRINSIC 0x9b04540:1, 0x9b04710, 0x9b04790, 0x9b04540 llvm-svn: 27110 | ||||
| * | Change approach so that we get codegen for free for intrinsics. With this, | Chris Lattner | 2006-03-24 | 2 | -38/+94 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | intrinsics that don't take pointer arguments now work. For example, we can compile this: int test3( __m128d *A) { return _mm_movemask_pd(*A); } int test4( __m128 *A) { return _mm_movemask_ps(*A); } to this: _test3: movl 4(%esp), %eax movapd (%eax), %xmm0 movmskpd %xmm0, %eax ret _test4: movl 4(%esp), %eax movaps (%eax), %xmm0 movmskps %xmm0, %eax ret llvm-svn: 27090 | ||||

