Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | X86JITInfo::getLazyResolverFunction() should not read cpu id to determine ↵ | Evan Cheng | 2009-09-03 | 1 | -3/+4 | |
| | | | | | | | | whether sse is available. Just use consult subtarget. No functionality changes. llvm-svn: 80880 | |||||
* | Add support for modeling whether or not the processor has support for | Chris Lattner | 2009-09-02 | 1 | -0/+4 | |
| | | | | | | | conditional moves as a subtarget feature. This is the easy part of PR4841. llvm-svn: 80763 | |||||
* | change the -x86-asm-syntax=intel/att flag to be in X86TAI | Chris Lattner | 2009-08-11 | 1 | -17/+0 | |
| | | | | | | | instead of X86 Subtarget. This elimianates dependencies on X86Subtarget from X86TAI. llvm-svn: 78746 | |||||
* | Normalize Subtarget constructors to take a target triple string instead of | Daniel Dunbar | 2009-08-02 | 1 | -4/+3 | |
| | | | | | | | | | | Module*. Also, dropped uses of TargetMachine where unnecessary. The only target which still takes a TargetMachine& is Mips, I would appreciate it if someone would normalize this to match other targets. llvm-svn: 77918 | |||||
* | remove the now-dead TM argument to these methods. | Chris Lattner | 2009-07-10 | 1 | -2/+2 | |
| | | | | llvm-svn: 75276 | |||||
* | make PIC vs DynamicNoPIC be explicit in PICStyles. | Chris Lattner | 2009-07-10 | 1 | -7/+15 | |
| | | | | llvm-svn: 75275 | |||||
* | add a couple of predicates to test for "stub style pic in PIC mode" and ↵ | Chris Lattner | 2009-07-10 | 1 | -2/+6 | |
| | | | | | | "stub style pic in dynamic-no-pic" mode. llvm-svn: 75273 | |||||
* | simplify fast isel by using ClassifyGlobalReference. This | Chris Lattner | 2009-07-10 | 1 | -6/+0 | |
| | | | | | | elimiantes the last use of GVRequiresExtraLoad, so delete it. llvm-svn: 75244 | |||||
* | eliminate GVRequiresRegister, replacing it with predicates we | Chris Lattner | 2009-07-10 | 1 | -5/+0 | |
| | | | | | | need for other purposes. llvm-svn: 75243 | |||||
* | move some classification logic around. Now GVRequiresExtraLoad | Chris Lattner | 2009-07-10 | 1 | -14/+17 | |
| | | | | | | | is just a trivial wrapper around "ClassifyGlobalReference", which stole a ton of logic from LowerGlobalAddress. llvm-svn: 75237 | |||||
* | GVRequiresExtraLoad is now never used for calls, simplify it based on this. | Chris Lattner | 2009-07-10 | 1 | -2/+1 | |
| | | | | llvm-svn: 75232 | |||||
* | actually, just eliminate PCRelGVRequiresExtraLoad. It makes the code | Chris Lattner | 2009-07-10 | 1 | -6/+0 | |
| | | | | | | more complex and slow than just directly testing what we care about. llvm-svn: 75231 | |||||
* | There is only one case where GVRequiresExtraLoad returns true for calls: | Chris Lattner | 2009-07-10 | 1 | -1/+7 | |
| | | | | | | | split its handling out to PCRelGVRequiresExtraLoad, and simplify code based on this. llvm-svn: 75230 | |||||
* | the "isDirectCall" operand of GVRequiresRegister is always false, eliminate it. | Chris Lattner | 2009-07-10 | 1 | -3/+2 | |
| | | | | llvm-svn: 75229 | |||||
* | When in -static mode, force the PIC style to none. Doing this requires fixing | Chris Lattner | 2009-07-09 | 1 | -2/+6 | |
| | | | | | | | code which conflated RIPRel PIC with x86-64. Fix these to just check for X86-64 directly. llvm-svn: 75092 | |||||
* | Fix a subtarget feature bug. | David Greene | 2009-06-29 | 1 | -1/+1 | |
| | | | | llvm-svn: 74428 | |||||
* | Add feature flags for AVX and FMA and fix some SSE4A feature flag | David Greene | 2009-06-26 | 1 | -3/+15 | |
| | | | | | | initialization problems. llvm-svn: 74350 | |||||
* | cosmetic changes. | Chris Lattner | 2009-06-21 | 1 | -1/+1 | |
| | | | | llvm-svn: 73836 | |||||
* | Update CPU capabilities for AMD machines | Stefanus Du Toit | 2009-05-26 | 1 | -0/+4 | |
| | | | | | | | | | | | | | | | | | - added processors k8-sse3, opteron-sse3, athlon64-sse3, amdfam10, and barcelona with appropriate sse3/4a levels - added FeatureSSE4A for amdfam10 processors in X86Subtarget: - added hasSSE4A - updated AutoDetectSubtargetFeatures to detect SSE4A - updated GetCurrentX86CPU to detect family 15 with sse3 as k8-sse3 and family 10h as amdfam10 New processor names match those used by gcc. Patch by Paul Redmond! llvm-svn: 72434 | |||||
* | Propagate CPU string out of SubtargetFeatures | Anton Korobeynikov | 2009-05-23 | 1 | -1/+2 | |
| | | | | llvm-svn: 72335 | |||||
* | Try again. Allow call to immediate address for ELF or when in static ↵ | Evan Cheng | 2009-05-20 | 1 | -0/+4 | |
| | | | | | | relocation mode. llvm-svn: 72160 | |||||
* | Tidy up #includes, deleting a bunch of unnecessary #includes. | Dan Gohman | 2009-01-05 | 1 | -1/+0 | |
| | | | | llvm-svn: 61715 | |||||
* | Do not isel load folding bt instructions for pentium m, core, core2, and AMD ↵ | Evan Cheng | 2009-01-02 | 1 | -0/+5 | |
| | | | | | | processors. These are significantly slower than a load followed by a bt of a register. llvm-svn: 61557 | |||||
* | Add initial support for back-scheduling address computations, | Dan Gohman | 2008-12-16 | 1 | -0/+6 | |
| | | | | | | | especially in the case of addresses computed from loop induction variables. llvm-svn: 61075 | |||||
* | Forgot a file. | Dale Johannesen | 2008-12-05 | 1 | -0/+6 | |
| | | | | llvm-svn: 60609 | |||||
* | Fix build with gcc-4.4: it doesn't like PICStyle | Duncan Sands | 2008-11-28 | 1 | -9/+9 | |
| | | | | | | being both a namespace and a variable name. llvm-svn: 60208 | |||||
* | Just don't transform this memset into "bzero" if no-builtin is specified. | Bill Wendling | 2008-09-30 | 1 | -1/+1 | |
| | | | | llvm-svn: 56888 | |||||
* | Add the new `-no-builtin' flag. This flag is meant to mimic the GCC | Bill Wendling | 2008-09-30 | 1 | -1/+1 | |
| | | | | | | | | | `-fno-builtin' flag. Currently, it's used to replace "memset" with "_bzero" instead of "__bzero" on Darwin10+. This arguably violates the meaning of this flag, but is currently sufficient. The meaning of this flag should become more specific over time. llvm-svn: 56885 | |||||
* | Use a dedicated IsLinux flag instead of an ELFLinux TargetType. | Dan Gohman | 2008-05-05 | 1 | -6/+9 | |
| | | | | llvm-svn: 50649 | |||||
* | Add AsmPrinter support for emitting a directive to declare that | Dan Gohman | 2008-05-05 | 1 | -2/+7 | |
| | | | | | | | | | the code being generated does not require an executable stack. Also, add target-specific code to make use of this on Linux on x86. llvm-svn: 50634 | |||||
* | Re-enable SSE4. | Evan Cheng | 2008-04-03 | 1 | -3/+2 | |
| | | | | llvm-svn: 49158 | |||||
* | Temporarily disabling SSE4 until we fix the encoding issues. | Evan Cheng | 2008-04-03 | 1 | -2/+3 | |
| | | | | llvm-svn: 49129 | |||||
* | Speculatively micro-optimize memory-zeroing calls on Darwin 10. | Dan Gohman | 2008-04-01 | 1 | -0/+6 | |
| | | | | llvm-svn: 49048 | |||||
* | Add convenient helper for win64 check. Simplify things slightly. | Anton Korobeynikov | 2008-03-22 | 1 | -0/+3 | |
| | | | | llvm-svn: 48691 | |||||
* | Update comment. | Evan Cheng | 2008-02-12 | 1 | -2/+2 | |
| | | | | llvm-svn: 47002 | |||||
* | Fix a x86-64 codegen deficiency. Allow gv + offset when using rip addressing ↵ | Evan Cheng | 2008-02-07 | 1 | -8/+0 | |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mode. Before: _main: subq $8, %rsp leaq _X(%rip), %rax movsd 8(%rax), %xmm1 movss _X(%rip), %xmm0 call _t xorl %ecx, %ecx movl %ecx, %eax addq $8, %rsp ret Now: _main: subq $8, %rsp movsd _X+8(%rip), %xmm1 movss _X(%rip), %xmm0 call _t xorl %ecx, %ecx movl %ecx, %eax addq $8, %rsp ret Notice there is another idiotic codegen issue that needs to be fixed asap: xorl %ecx, %ecx movl %ecx, %eax llvm-svn: 46850 | |||||
* | SSE 4.1 Intrinsics and detection | Nate Begeman | 2008-02-03 | 1 | -1/+3 | |
| | | | | llvm-svn: 46681 | |||||
* | darwin9 and above support aligned common symbols. | Chris Lattner | 2008-01-02 | 1 | -1/+8 | |
| | | | | llvm-svn: 45494 | |||||
* | Remove attribution from file headers, per discussion on llvmdev. | Chris Lattner | 2007-12-29 | 1 | -2/+2 | |
| | | | | llvm-svn: 45418 | |||||
* | Make ARM an X86 memcpy expansion more similar to each other. | Rafael Espindola | 2007-10-31 | 1 | -7/+5 | |
| | | | | | | | | Now both subtarget define getMaxInlineSizeThreshold and the expansion uses it. This should not change generated code. llvm-svn: 43552 | |||||
* | Add support for having different alignment for objects on call frames. | Rafael Espindola | 2007-09-07 | 1 | -1/+1 | |
| | | | | | | | The x86-64 ABI states that objects passed on the stack have 8 byte alignment. Implement that. llvm-svn: 41768 | |||||
* | Mac OS X X86-64 ABI is same as the standard. | Evan Cheng | 2007-09-04 | 1 | -6/+2 | |
| | | | | llvm-svn: 41700 | |||||
* | Align i64 and f64 at 8 byte on x86-64. | Rafael Espindola | 2007-08-31 | 1 | -6/+12 | |
| | | | | | | | This is mandated table 3.1 at http://www.x86-64.org/documentation/abi.pdf llvm-svn: 41642 | |||||
* | Replace 4-line function with 10-line version per review comment. | Dale Johannesen | 2007-08-06 | 1 | -4/+10 | |
| | | | | llvm-svn: 40881 | |||||
* | Move lengthy conditional down 1 level per review comment. | Dale Johannesen | 2007-08-06 | 1 | -0/+5 | |
| | | | | llvm-svn: 40878 | |||||
* | Mac OS X X86-64 low 4G address not available. | Evan Cheng | 2007-08-01 | 1 | -0/+14 | |
| | | | | llvm-svn: 40701 | |||||
* | Add support for our first SSSE3 instruction "pmulhrsw". | Bill Wendling | 2007-04-10 | 1 | -2/+3 | |
| | | | | llvm-svn: 35869 | |||||
* | document some subtlety | Chris Lattner | 2007-01-16 | 1 | -1/+3 | |
| | | | | llvm-svn: 33257 | |||||
* | Instead of yet another enum indicating the "assembly language flavor", | Bill Wendling | 2007-01-16 | 1 | -0/+4 | |
| | | | | | | just use the one that's in the subtarget. llvm-svn: 33255 | |||||
* | * PIC codegen for X86/Linux has been implemented | Anton Korobeynikov | 2007-01-12 | 1 | -4/+21 | |
| | | | | | | | | * PIC-aware internal structures in X86 Codegen have been refactored * Visibility (default/weak) has been added * Docs fixes (external weak linkage, visibility, formatting) llvm-svn: 33136 |