| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Corrected Atom latencies for SSE SQRT instructions. | Preston Gurd | 2013-05-07 | 3 | -19/+36 |
| | | | | | llvm-svn: 181346 | ||||
| * | Re-enable AVX detection on x64 platforms. | Michael Kuperstein | 2013-05-07 | 1 | -1/+2 |
| | | | | | llvm-svn: 181313 | ||||
| * | Remove a recently redundant transform from X86ISelLowering. | David Majnemer | 2013-05-05 | 1 | -11/+0 |
| | | | | | | | | | | | | | X86ISelLowering has support to treat: (icmp ne (and (xor %flags, -1), (shl 1, flag)), 0) as if it were actually: (icmp eq (and %flags, (shl 1, flag)), 0) However, r179386 has code at the InstCombine level to handle this. llvm-svn: 181145 | ||||
| * | Add ArrayRef constructor from None, and do the cleanups that this ↵ | Dmitri Gribenko | 2013-05-05 | 1 | -2/+2 |
| | | | | | | | | | constructor enables Patch by Robert Wilhelm. llvm-svn: 181138 | ||||
| * | whitespace | Nadav Rotem | 2013-05-04 | 1 | -2/+2 |
| | | | | | llvm-svn: 181137 | ||||
| * | Fix an odd comment. | Nadav Rotem | 2013-05-04 | 1 | -2/+1 |
| | | | | | llvm-svn: 181136 | ||||
| * | X86: Add target description for btver2; make autodetection logic aware of AVX. | Benjamin Kramer | 2013-05-03 | 1 | -1/+6 |
| | | | | | llvm-svn: 181005 | ||||
| * | Unbreaking the non-x86 build bots by protecting the AVX test code properly. | Aaron Ballman | 2013-05-03 | 1 | -0/+4 |
| | | | | | llvm-svn: 180992 | ||||
| * | Correctly testing for AVX support in x86 based off code from Hosts.cpp. | Aaron Ballman | 2013-05-03 | 1 | -1/+18 |
| | | | | | llvm-svn: 180991 | ||||
| * | In MC asm parsing, account for the possibility of whitespace within | John McCall | 2013-05-03 | 1 | -18/+22 |
| | | | | | | | | | | | | | | | the "identifier" parsed by the frontend callback by skipping forward until we've consumed a token that ends at the point dictated by the callback. In addition, inform the callback when it's parsing an unevaluated operand (e.g. mov eax, LENGTH A::x) as opposed to an evaluated one (e.g. mov eax, [A::x]). This commit depends on a clang commit. llvm-svn: 180978 | ||||
| * | 80-col fixup. | Michael Liao | 2013-05-02 | 1 | -2/+3 |
| | | | | | llvm-svn: 180915 | ||||
| * | Avoid duplicating logic on frame register selecting when lowering eh_return | Michael Liao | 2013-05-02 | 1 | -9/+11 |
| | | | | | | | No functionality change llvm-svn: 180914 | ||||
| * | Avoid duplicating logic on frame register selecting when lowering frameaddr | Michael Liao | 2013-05-02 | 1 | -1/+3 |
| | | | | | | | No functionality change llvm-svn: 180912 | ||||
| * | Put VMOVPQIto64rr in the VRPDI class. | Rafael Espindola | 2013-05-01 | 1 | -3/+3 |
| | | | | | | | Patch by Joshua Magee. llvm-svn: 180842 | ||||
| * | X86: Now that we have a canonical form for vector integer abs, match it into ↵ | Benjamin Kramer | 2013-04-26 | 1 | -0/+51 |
| | | | | | | | pabs. llvm-svn: 180600 | ||||
| * | Make function documentation conform to llvm standards. | Preston Gurd | 2013-04-25 | 1 | -30/+32 |
| | | | | | | | Expunge all remaining traces and use of live variable information. llvm-svn: 180577 | ||||
| * | This patch adds the X86FixupLEAs pass, which will reduce instruction | Preston Gurd | 2013-04-25 | 7 | -0/+270 |
| | | | | | | | | | latency for certain models of the Intel Atom family, by converting instructions into their equivalent LEA instructions, when it is both useful and possible to do so. llvm-svn: 180573 | ||||
| * | Fix section relocation for SECTIONREL32 with immediate offset. | Rafael Espindola | 2013-04-25 | 1 | -2/+15 |
| | | | | | | | Patch by Kai Nacke. This matches the gnu as output. llvm-svn: 180568 | ||||
| * | [ms-inline asm] Removed this unnecessary check. In the current implementation, | Chad Rosier | 2013-04-22 | 1 | -1/+1 |
| | | | | | | | Disp will always be one of MCSymbolRefExpr or MCConstantExpr, and never NULL. llvm-svn: 180059 | ||||
| * | [ms-inline asm] Add the OpDecl to the InlineAsmIdentifierInfo struct and in turn | Chad Rosier | 2013-04-22 | 1 | -13/+18 |
| | | | | | | | | the MCParsedAsmOperand. Part of rdar://13663589 llvm-svn: 180054 | ||||
| * | Fix unused variable warning. | Chad Rosier | 2013-04-22 | 1 | -1/+1 |
| | | | | | llvm-svn: 180044 | ||||
| * | [ms-inline asm] Remove the identifier parsing logic from the AsmParser. This is | Chad Rosier | 2013-04-22 | 1 | -61/+55 |
| | | | | | | | | | now taken care of by the frontend, which allows us to parse arbitrary C/C++ variables. Part of rdar://13663589 llvm-svn: 180037 | ||||
| * | [ms-inline asm] Refactor/clean up the SemaLookup interface. No functional | Chad Rosier | 2013-04-22 | 1 | -34/+23 |
| | | | | | | | | change indended. Part of rdar://13663589 llvm-svn: 180028 | ||||
| * | Remove unused ShouldFoldAtomicFences flag. | Tim Northover | 2013-04-20 | 1 | -7/+0 |
| | | | | | | | | | I think it's almost impossible to fold atomic fences profitably under LLVM/C++11 semantics. As a result, this is now unused and just cluttering up the target interface. llvm-svn: 179940 | ||||
| * | Remove unused MEMBARRIER DAG node; it's been replaced by ATOMIC_FENCE. | Tim Northover | 2013-04-20 | 1 | -46/+0 |
| | | | | | llvm-svn: 179939 | ||||
| * | Move TryToFoldFastISelLoad to FastISel, where it belongs. In general, I'm | Eli Bendersky | 2013-04-19 | 1 | -9/+5 |
| | | | | | | | | trying to move as much FastISel logic as possible out of the main path in SelectionDAGISel - intermixing them just adds confusion. llvm-svn: 179902 | ||||
| * | ArrayRefize getMachineNode(). No functionality change. | Michael Liao | 2013-04-19 | 3 | -36/+18 |
| | | | | | llvm-svn: 179901 | ||||
| * | [ms-inline asm] Make code layout more canonical with iniline asm handled last. | Chad Rosier | 2013-04-19 | 1 | -13/+13 |
| | | | | | llvm-svn: 179875 | ||||
| * | [ms-inline asm] Refactor the parsing of identifiers. No functional change | Chad Rosier | 2013-04-19 | 1 | -39/+33 |
| | | | | | | | | indended. Part of rdar://13663589 llvm-svn: 179871 | ||||
| * | [ms-inline asm] Remove these asserts. C++ variables that use namespace | Chad Rosier | 2013-04-19 | 1 | -2/+0 |
| | | | | | | | qualifiers don't necessarily begin with an identifier (e.g., ::foo::bar). llvm-svn: 179867 | ||||
| * | [ms-inline asm] Move this variable into the scope in which it is used. | Chad Rosier | 2013-04-19 | 1 | -1/+1 |
| | | | | | llvm-svn: 179866 | ||||
| * | [ms-inline asm] Make this a hard error. | Chad Rosier | 2013-04-19 | 1 | -4/+4 |
| | | | | | llvm-svn: 179865 | ||||
| * | [ms-inline asm] Cleanup the dot operator implementation. | Chad Rosier | 2013-04-19 | 1 | -29/+15 |
| | | | | | llvm-svn: 179856 | ||||
| * | Use 'array_lengthof' as possible to avoid magic numbers | Michael Liao | 2013-04-19 | 1 | -23/+32 |
| | | | | | llvm-svn: 179833 | ||||
| * | Use an enum instead of magic constants to improve readability. | Bill Wendling | 2013-04-19 | 2 | -16/+44 |
| | | | | | llvm-svn: 179820 | ||||
| * | [ms-inline asm] Apply the condition code mnemonic aliases to both the Intel and | Chad Rosier | 2013-04-18 | 1 | -1/+1 |
| | | | | | | | | AT&T dialect. Test case for r179804 as well. rdar://13674398 and PR13340. llvm-svn: 179813 | ||||
| * | Set the compact unwind encoding to 'requires EH DWARF' if we cannot generate ↵ | Bill Wendling | 2013-04-18 | 1 | -9/+9 |
| | | | | | | | a CU encoding. llvm-svn: 179808 | ||||
| * | [asm parser] Add support for predicating MnemonicAlias based on the assembler | Chad Rosier | 2013-04-18 | 2 | -64/+70 |
| | | | | | | | | | variant/dialect. Addresses a FIXME in the emitMnemonicAliases function. Use and test case to come shortly. rdar://13688439 and part of PR13340. llvm-svn: 179804 | ||||
| * | X86: Add an SSE2 lowering for 64 bit compares when pcmpgtq (SSE4.2) isn't ↵ | Benjamin Kramer | 2013-04-18 | 1 | -15/+40 |
| | | | | | | | | | available. This pattern started popping up in vectorized min/max reductions. llvm-svn: 179797 | ||||
| * | Allow misaligned stores in x86 fast-isel. | Derek Schuff | 2013-04-18 | 1 | -5/+0 |
| | | | | | | | | | | | | | In X86FastISel::X86SelectStore(), improperly aligned stores are rejected and handled by the DAG-based ISel. However, X86FastISel::X86SelectLoad() makes no such requirement. There doesn't appear to be an x86 architectural correctness issue with allowing potentially unaligned store instructions. This patch removes this restriction. Patch by Jim Stichnot. llvm-svn: 179774 | ||||
| * | [ms-inline asm] Simplify some logic and add a FIXME for unhandled unary minus. | Chad Rosier | 2013-04-18 | 1 | -10/+13 |
| | | | | | llvm-svn: 179765 | ||||
| * | Make this private method. | Chad Rosier | 2013-04-18 | 1 | -4/+4 |
| | | | | | llvm-svn: 179764 | ||||
| * | [ms-inline asm] These should be int64_t, not uint64_t. | Chad Rosier | 2013-04-17 | 1 | -6/+6 |
| | | | | | llvm-svn: 179724 | ||||
| * | [ms-inline asm] Add support for the minus unary operator. Previously, we were | Chad Rosier | 2013-04-17 | 1 | -72/+106 |
| | | | | | | | | | | | | unable to handle cases such as __asm mov eax, 8*-8. This patch also attempts to simplify the state machine. Further, the error reporting has been improved. Test cases included, but more will be added to the clang side shortly. rdar://13668445 llvm-svn: 179719 | ||||
| * | This patch teaches x86 fast-isel to generate the native div/idiv instructions | Eli Bendersky | 2013-04-17 | 1 | -0/+125 |
| | | | | | | | | | | for the sdiv/srem/udiv/urem bitcode instructions. This is done for the i8, i16, and i32 types, as well as i64 for the x86_64 target. Patch by Jim Stichnoth llvm-svn: 179715 | ||||
| * | X86 cost model: Exit before calling getSimpleVT on non-simple VTs | Arnold Schwaighofer | 2013-04-17 | 1 | -0/+4 |
| | | | | | | | | | getSimpleVT can only handle simple value types. radar://13676022 llvm-svn: 179714 | ||||
| * | [ms-inline asm] Add support for parsing complex immediate expressions. Test | Chad Rosier | 2013-04-17 | 1 | -32/+51 |
| | | | | | | | | cases to be submitted on clang side shortly. rdar://13663768 and PR15760 llvm-svn: 179655 | ||||
| * | Remove unused variable from previous refactor. | Chad Rosier | 2013-04-16 | 1 | -3/+0 |
| | | | | | llvm-svn: 179611 | ||||
| * | [ms-inline asm] Refactor. No functional change intended. | Chad Rosier | 2013-04-16 | 1 | -405/+419 |
| | | | | | llvm-svn: 179610 | ||||
| * | [ms-inline asm] Remove some dead code. | Chad Rosier | 2013-04-16 | 1 | -8/+0 |
| | | | | | llvm-svn: 179607 | ||||

