Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Separate compatibility declarations | Chris Robinson | 2013-10-28 | 1 | -0/+1 |
| | |||||
* | Move SetThreadName to threads.c | Chris Robinson | 2013-10-27 | 1 | -38/+0 |
| | |||||
* | Add a method to set the running thread's name | Chris Robinson | 2013-10-26 | 1 | -0/+38 |
| | |||||
* | Add a CMake option to not define the IDs used on Windows | Chris Robinson | 2013-10-07 | 1 | -0/+3 |
| | | | | | This includes the GUIDs, IIDs, CLSID, and PropertyKeys. It is up to the user to ensure the appropriate IDs are defined when linked. | ||||
* | Use explicit bit offsets when checking for SSE and SSE2 | Chris Robinson | 2013-06-06 | 1 | -4/+2 |
| | | | | Clang's cpuid.h doesn't contain the bit_* macros. | ||||
* | Check the right flag for tracing SSE2 support | Chris Robinson | 2013-05-29 | 1 | -1/+1 |
| | |||||
* | Use fegetenv/fesetenv to backup and restore the FPU state | Chris Robinson | 2013-05-22 | 1 | -4/+4 |
| | |||||
* | Merge the fesetround FPU handler with the assembly one | Chris Robinson | 2013-05-22 | 1 | -22/+23 |
| | | | | | | The assembly is only needed for SSE since we can't access the intrinsics without the -msse switch, which itself would cause SSE code to be generated elsewhere automtically. | ||||
* | Check for SSE2 and set the denormals-are-zero bit for mixing if available | Chris Robinson | 2013-05-22 | 1 | -3/+14 |
| | |||||
* | Set the correct SSE bits for round-to-zero | Chris Robinson | 2013-05-22 | 1 | -1/+1 |
| | |||||
* | Avoid using a temp buffer for al_print | Chris Robinson | 2013-05-22 | 1 | -12/+5 |
| | | | | | | | | It's now using two *printf calls, which unfortuantely means there could be a race between the two and cause the message to break up if something else tries to print to the same file. This shouldn't really be a big deal since al_print isn't used that often, and it now allows for lines of practically unlimited length. | ||||
* | Explicitly check for malloc.h | Chris Robinson | 2012-11-25 | 1 | -1/+1 |
| | |||||
* | malloc.h isn't standard, and is only needed for _aligned_malloc | Chris Robinson | 2012-11-25 | 1 | -1/+3 |
| | |||||
* | Avoid storing the SSE state if it's not supported | Chris Robinson | 2012-11-13 | 1 | -1/+4 |
| | |||||
* | Include standard headers before conditional ones | Chris Robinson | 2012-11-04 | 1 | -7/+6 |
| | |||||
* | Don't assume x86 or x64 to use IsProcessorFeaturePresent | Chris Robinson | 2012-10-08 | 1 | -3/+2 |
| | |||||
* | Fix a type conversion warning | Chris Robinson | 2012-10-08 | 1 | -2/+3 |
| | |||||
* | Use IsProcessorFeaturePresent to detect SSE on Windows | Chris Robinson | 2012-10-07 | 1 | -0/+10 |
| | |||||
* | Revert "Use __cpuid from intrin.h in Windows when available" | Chris Robinson | 2012-10-07 | 1 | -33/+0 |
| | | | | This reverts commit 6b870714a91c9607acc4501234a00c7c94fa2b85. | ||||
* | Use __cpuid from intrin.h in Windows when available | Chris Robinson | 2012-10-07 | 1 | -0/+33 |
| | |||||
* | Add trace, warn, and error markers to logged output | Chris Robinson | 2012-10-07 | 1 | -3/+3 |
| | |||||
* | Include windows.h before cpuid.h to avoid __cpuid clashes | Chris Robinson | 2012-10-06 | 1 | -17/+17 |
| | |||||
* | Precision control bits don't exist with SSE | Chris Robinson | 2012-09-18 | 1 | -5/+4 |
| | |||||
* | Win64 doesn't allow _controlfp or __control87_2 to set the precision control ↵ | Chris Robinson | 2012-09-18 | 1 | -0/+6 |
| | | | | bits | ||||
* | Properly restore the SSE control word with __control87_2 | Chris Robinson | 2012-09-16 | 1 | -1/+1 |
| | |||||
* | Properly handle the SSE control word | Chris Robinson | 2012-09-16 | 1 | -16/+37 |
| | |||||
* | Use __control87_2 when available | Chris Robinson | 2012-09-16 | 1 | -0/+7 |
| | |||||
* | Use a struct to store the FPU mode | Chris Robinson | 2012-09-16 | 1 | -12/+9 |
| | |||||
* | Don't include alu.h in alMain.h | Chris Robinson | 2012-09-14 | 1 | -0/+6 |
| | |||||
* | Enable flush-to-zero mode when possible | Chris Robinson | 2012-09-14 | 1 | -0/+40 |
| | |||||
* | Use _aligned_free for pointers returned by _aligned_malloc | Chris Robinson | 2012-08-24 | 1 | -1/+3 |
| | |||||
* | Fix aligned allocator fallback | Chris Robinson | 2012-08-16 | 1 | -3/+2 |
| | |||||
* | Check the max CPUID functions before calling them | Chris Robinson | 2012-08-15 | 1 | -4/+11 |
| | |||||
* | Add wrapper methods to ensure aligned allocations | Chris Robinson | 2012-08-15 | 1 | -0/+48 |
| | |||||
* | Check the correct macro for ARM Neon support | Chris Robinson | 2012-08-15 | 1 | -1/+1 |
| | |||||
* | Don't bother handling MMX since we don't use it | Chris Robinson | 2012-08-15 | 1 | -8/+3 |
| | |||||
* | Add a config option to disable use of CPU extensions | Chris Robinson | 2012-08-13 | 1 | -8/+11 |
| | |||||
* | Check for some CPU extensions | Chris Robinson | 2012-08-13 | 1 | -0/+55 |
| | |||||
* | Add some includes | Chris Robinson | 2012-06-20 | 1 | -0/+9 |
| | |||||
* | Remove an unused property key definition | Chris Robinson | 2012-03-12 | 1 | -2/+0 |
| | |||||
* | Don't use GUIDs to ID mmdevapi devices, and don't enumerate if not needed | Chris Robinson | 2012-03-01 | 1 | -0/+16 |
| | |||||
* | Support device enumeration with mmdevapi | Chris Robinson | 2012-02-16 | 1 | -0/+8 |
| | |||||
* | Don't fail to insert a map entry when the key exists and the limit is reached | Chris Robinson | 2011-11-26 | 1 | -9/+9 |
| | |||||
* | Return the key's value from the map when it's removed | Chris Robinson | 2011-10-06 | 1 | -30/+4 |
| | |||||
* | Only warn when a symbol fails to load | Chris Robinson | 2011-09-24 | 1 | -1/+1 |
| | |||||
* | Move Sleep implementation into helper.c and emulate sched_yield for Windows | Chris Robinson | 2011-09-20 | 1 | -2/+12 |
| | |||||
* | Fix indentation | Chris Robinson | 2011-09-20 | 1 | -1/+1 |
| | |||||
* | Log the function name instead of the source file and line number | Chris Robinson | 2011-09-02 | 1 | -8/+2 |
| | |||||
* | Don't set a thread priority if real-time isn't requested | Chris Robinson | 2011-08-31 | 1 | -10/+2 |
| | |||||
* | Add a function to initialize a RWLock | Chris Robinson | 2011-08-31 | 1 | -5/+10 |
| |