Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | Fix malloc.h is deprecated warning on DragonFly BSD. | Edward O'Callaghan | 2009-11-02 | 1 | -1/+3 | |
| | | | | llvm-svn: 85782 | |||||
* | Most stack straces don't need 3 digits worth of levels. | Dan Gohman | 2009-10-30 | 1 | -1/+1 | |
| | | | | llvm-svn: 85575 | |||||
* | Move DataTypes.h to include/llvm/System, update all users. This breaks the last | Chandler Carruth | 2009-10-26 | 2 | -2/+2 | |
| | | | | | | direct inclusion edge from System to Support. llvm-svn: 85086 | |||||
* | Remove unused includes. | Chandler Carruth | 2009-10-25 | 2 | -2/+0 | |
| | | | | llvm-svn: 85074 | |||||
* | Use 'waitpid' instead of 'wait'. Basing Program::Wait() on 'wait()' ↵ | Ted Kremenek | 2009-10-22 | 1 | -1/+1 | |
| | | | | | | | | | | | prevents it being correct within a multithreaded context. This address: PR 5277 (Program::Wait is unsafe to call from multiple threads). Note: If waitpid() turns out to be non-portable, we can add more autoconf magic, or look into another solution. llvm-svn: 84903 | |||||
* | Haiku porting patches, Credit to Paul Davey. | Edward O'Callaghan | 2009-10-12 | 1 | -1/+1 | |
| | | | | llvm-svn: 83823 | |||||
* | MingW build fixes | Benjamin Kramer | 2009-10-02 | 1 | -0/+6 | |
| | | | | | | | | | | - MingW needs -lpsapi (in ${LIBS}) linked after -lLLVMSystem. Noticed by Ronald Pijnacker! - Some parts of the System library must be build with exceptions on windows. Based on a patch by Jay Foad! llvm-svn: 83251 | |||||
* | ... missed hiding a variable for MSVC only. | Daniel Dunbar | 2009-09-22 | 1 | -0/+2 | |
| | | | | llvm-svn: 82552 | |||||
* | Hide MSVC specific CRT interaction behind _MSC_VER. | Daniel Dunbar | 2009-09-22 | 1 | -1/+7 | |
| | | | | llvm-svn: 82551 | |||||
* | Remove the GetProcessId() call from Win32/Program.inc, take 2. | Mikhail Glushenkov | 2009-09-22 | 1 | -9/+23 | |
| | | | | | | | GetProcessId() was introduced only in Windows XP, and we want to support earlier versions. llvm-svn: 82548 | |||||
* | Add a magic LLVM_DISABLE_CRT_DEBUG environment variable which we check in ↵ | Daniel Dunbar | 2009-09-22 | 1 | -1/+39 | |
| | | | | | | | | RegisterHandler and use to disable the Win32 crash dialogs. These are a major blocker to any kind of automated testing. Also, tweak the 'lit' test runner to set this variable unconditionally. llvm-svn: 82537 | |||||
* | Revert "Get rid of GetProcessId in Win32/Program.inc.", this breaks | Daniel Dunbar | 2009-09-22 | 2 | -22/+43 | |
| | | | | | | ExecuteAndWait. llvm-svn: 82522 | |||||
* | Get rid of GetProcessId in Win32/Program.inc. | Mikhail Glushenkov | 2009-09-15 | 2 | -43/+22 | |
| | | | | | | | GetProcessId was introduced only in XP. As a bonus, this change makes Program objects copyable, since Program is now basically a PID. llvm-svn: 81826 | |||||
* | Experimental fix for PR4960. | Daniel Dunbar | 2009-09-12 | 1 | -5/+14 | |
| | | | | | | - Could we just always implement this as __clear_cache for __GNUC__? llvm-svn: 81655 | |||||
* | Fix llvm-extract's "writing bitcode to a terminal" warning, which wasn't | Dan Gohman | 2009-09-11 | 2 | -16/+14 | |
| | | | | | | | | | | | | | | | working. To support this, add an is_displayed() function to raw_ostream, and generalize Process::StandardOutIsDisplayed and friends in order to support it. Also, call RemoveFileOnSignal before creating a file instead of after, so that the file isn't left behind if the program is interrupted between when the file is created and RemoveFileOnSignal is called. While here, add a -S to llvm-extract and port it to IRReader so that it supports assembly input. llvm-svn: 81568 | |||||
* | Add a shortcut for OS X to Path::GetMainExecutable. This gives a nice speedup on | Benjamin Kramer | 2009-09-09 | 1 | -1/+15 | |
| | | | | | | clang's testsuite. llvm-svn: 81333 | |||||
* | Check that the 'kill' call succeeded. | Mikhail Glushenkov | 2009-09-09 | 2 | -2/+13 | |
| | | | | | | Thanks to Duncan Sands for spotting this. llvm-svn: 81328 | |||||
* | This should unbreak the build on 64-bit Linux. | Mikhail Glushenkov | 2009-09-08 | 1 | -3/+6 | |
| | | | | llvm-svn: 81252 | |||||
* | Const-correctness. | Mikhail Glushenkov | 2009-09-08 | 2 | -2/+2 | |
| | | | | llvm-svn: 81249 | |||||
* | Get rid of the Pid_ member in the Program class. | Mikhail Glushenkov | 2009-09-08 | 2 | -20/+29 | |
| | | | | llvm-svn: 81247 | |||||
* | Add a Kill() function to the Program class. | Mikhail Glushenkov | 2009-09-08 | 2 | -9/+30 | |
| | | | | llvm-svn: 81246 | |||||
* | Using a signal handler that does nothing should be | Duncan Sands | 2009-09-07 | 1 | -6/+2 | |
| | | | | | | equivalent to SIG_IGN. llvm-svn: 81144 | |||||
* | Remove some not-really-used variables, as warned | Duncan Sands | 2009-09-06 | 1 | -3/+0 | |
| | | | | | | about by icc (#593, partial). Patch by Erick Tryzelaar. llvm-svn: 81115 | |||||
* | Tweak code into an equivalent form for which icc | Duncan Sands | 2009-09-06 | 3 | -17/+13 | |
| | | | | | | | doesn't warn about unreachable instructions. Patch by Erick Tryzelaar (#111). llvm-svn: 81110 | |||||
* | Improve llvm::getHostTriple for some cases where the LLVM_HOSTTRIPLE is not | Daniel Dunbar | 2009-09-03 | 1 | -4/+42 | |
| | | | | | | reliable. llvm-svn: 80863 | |||||
* | Fix ExplicitSymbols leak. | Torok Edwin | 2009-08-31 | 1 | -0/+7 | |
| | | | | llvm-svn: 80589 | |||||
* | remove the last uses of Config/alloca.h | Chris Lattner | 2009-08-23 | 1 | -3/+7 | |
| | | | | llvm-svn: 79873 | |||||
* | Prune #includes from llvm/Linker.h and llvm/System/Path.h, | Chris Lattner | 2009-08-23 | 7 | -26/+16 | |
| | | | | | | | | | | | | | | | | | | forcing them down into various .cpp files. This change also: 1. Renames TimeValue::toString() and Path::toString() to ::str() for similarity with the STL. 2. Removes all stream insertion support for sys::Path, forcing clients to call .str(). 3. Removes a use of Config/alloca.h from bugpoint, using smallvector instead. 4. Weans llvm-db off <iostream> sys::Path really needs to be gutted, but I don't have the desire to do it at this point. llvm-svn: 79869 | |||||
* | Get rid of a helgrind warning. If this is _actually_ a performance problem, | Owen Anderson | 2009-08-19 | 1 | -3/+3 | |
| | | | | | | we can find a way to cache the answer that isn't racy. llvm-svn: 79472 | |||||
* | Add locking around signal handler registration. | Owen Anderson | 2009-08-17 | 1 | -0/+14 | |
| | | | | llvm-svn: 79254 | |||||
* | This void is implicit in C++. | Dan Gohman | 2009-08-12 | 1 | -1/+1 | |
| | | | | llvm-svn: 78848 | |||||
* | improve win32 path support, patch by Baptiste Lepilleur! | Chris Lattner | 2009-08-12 | 1 | -1/+2 | |
| | | | | llvm-svn: 78823 | |||||
* | Fix a bunch of namespace pollution. | Dan Gohman | 2009-08-07 | 1 | -1/+1 | |
| | | | | llvm-svn: 78363 | |||||
* | Minor code simplification. | Dan Gohman | 2009-08-05 | 1 | -4/+2 | |
| | | | | llvm-svn: 78239 | |||||
* | Update a comment to reflect the current code. | Dan Gohman | 2009-08-05 | 1 | -1/+2 | |
| | | | | llvm-svn: 78215 | |||||
* | Remove needless uses of std::flush in the parent process after a | Dan Gohman | 2009-08-05 | 1 | -7/+0 | |
| | | | | | | | fork call. This eliminates a need for <iostream>. Also remove needless fsync calls. llvm-svn: 78131 | |||||
* | Use _exit rather than exit in the child process after a failed exec. | Dan Gohman | 2009-08-05 | 1 | -3/+7 | |
| | | | | | | Add a comment explaining why. llvm-svn: 78128 | |||||
* | Follow Unix behavior and return 127 if the command is not found, | Dan Gohman | 2009-08-04 | 1 | -1/+1 | |
| | | | | | | and 126 if it is not executable. llvm-svn: 78120 | |||||
* | When exec() fails, return 127 instead of errno; the parent process has no way to | Daniel Dunbar | 2009-08-04 | 1 | -3/+3 | |
| | | | | | | | distinguish that the result is errno, so it can't use it to provide more information about the error (it also exposes the numeric value of errno). llvm-svn: 78098 | |||||
* | Fix a race condition in getting the process exit code on Win32. | Daniel Dunbar | 2009-08-03 | 2 | -5/+25 | |
| | | | | llvm-svn: 77953 | |||||
* | Empty arguments need to be quoted on Win32. | Daniel Dunbar | 2009-08-02 | 1 | -3/+9 | |
| | | | | llvm-svn: 77913 | |||||
* | eXtensible Systems doesn't exist any more and Reid donated this to the | Chris Lattner | 2009-08-01 | 1 | -6/+0 | |
| | | | | | | project under the normal license, yay. :) llvm-svn: 77823 | |||||
* | Remove another F_OK. | Dan Gohman | 2009-07-29 | 1 | -1/+1 | |
| | | | | llvm-svn: 77405 | |||||
* | On "Unix", if Program::FindProgramByName is given a name containing | Dan Gohman | 2009-07-28 | 1 | -1/+1 | |
| | | | | | | | slashes, just go with it, regardless of whether it looks like it will be executable. This follows the behavior of sh(1) more closely. llvm-svn: 77396 | |||||
* | It isn't necessary to use F_OK when using R_OK or similar. | Dan Gohman | 2009-07-28 | 1 | -2/+2 | |
| | | | | llvm-svn: 77392 | |||||
* | Fix a fixme; don't take binaries from the working directory. | Benjamin Kramer | 2009-07-28 | 1 | -3/+3 | |
| | | | | | | | This fixes clang on non-darwin platforms if a file called 'ld' or 'as' is in the working directory. Based on patch by Pawel Worach! llvm-svn: 77379 | |||||
* | CMake: make sure that the *.inc files for libSystem show up in the resulting | Douglas Gregor | 2009-07-27 | 1 | -0/+23 | |
| | | | | | | project. llvm-svn: 77213 | |||||
* | Don't use llvm_report_error in libSystem, this is a layering violation. | Daniel Dunbar | 2009-07-26 | 1 | -2/+2 | |
| | | | | llvm-svn: 77169 | |||||
* | Re-committing r76828 with the JIT memory manager changes now that the build | Reid Kleckner | 2009-07-23 | 2 | -6/+8 | |
| | | | | | | bots like the BumpPtrAllocator changes. llvm-svn: 76902 | |||||
* | Reverting r76825 and r76828, since they caused clang runtime errors and some ↵ | Reid Kleckner | 2009-07-23 | 2 | -8/+6 | |
| | | | | | | build failure involving memset. llvm-svn: 76838 |