Age | Commit message (Collapse) | Author | |
---|---|---|---|
2011-06-17 | Merge branch 'r300' into hhienvsa_r300 | PFO | |
2011-06-15 | [Radeon] r300: uses Android's mmap routine, as in r600 | Hugues Hiegel | |
2011-06-15 | [Radeon] r300: uses Android's mmap as done in r600 | PFO | |
2011-03-08 | r300g: decide whether a flush should be asynchronous when calling it | Marek Olšák | |
Thread offloading is not sometimes desirable, e.g. when mapping a buffer. | |||
2011-02-16 | r300g: fix a race between CS and SET_TILING ioctls | Marek Olšák | |
2011-02-15 | r300g: offload the CS ioctl to another thread | Marek Olšák | |
This is a multi-threading optimization which hides the kernel overhead behind a thread. It improves performance in CPU-limited apps by 2-15%. Of course you must have at least 2 cores for it to make any difference. It can be disabled with: export RADEON_THREAD=0 | |||
2011-02-15 | r300g: actually implement the is_buffer_busy hook the right way | Marek Olšák | |
Ooops. | |||
2011-02-15 | r300g: handle interaction between UNSYNCHRONIZED and DONTBLOCK flags in bo_map | Marek Olšák | |
The VBO module uses both, but they are somewhat opposite to each other. In this case, we pick UNSYNCHRONIZED and ignore DONTBLOCK. | |||
2011-02-15 | r300g: fix a possible race condition when mapping a buffer | Marek Olšák | |
This is the last one I think. | |||
2011-02-14 | r300g: flush CS in bo_map even if we get USAGE_DONTBLOCK | Marek Olšák | |
Because an app may do something like this: while (!(ptr = bo_map(..., DONT_BLOCK))) { /* Do some other work. */ } And it would be looping endlessly if we didn't flush. | |||
2011-02-14 | r300g: implement pb_manager::is_buffer_busy | Marek Olšák | |
2011-02-13 | r300g: fixup the handle_compare function | Marek Olšák | |
Accidentally negated in 685c3262b945a7f0e9f1f3a9409a12fdda08c828. | |||
2011-02-12 | r300g: typecast void* to unsigned correctly | Marek Olšák | |
2011-02-12 | r300g: improve function radeon_bo_is_referenced_by_cs | Marek Olšák | |
This should prevent calling into radeon_get_reloc when there's only one context. | |||
2011-02-11 | r300g: remove unused function prototypes, update copyright | Marek Olšák | |
2011-02-11 | r300g: plug a memory leak in winsys | Marek Olšák | |
2011-02-11 | r300g: remove unneeded code in winsys | Marek Olšák | |
We don't need the read/write flags. | |||
2011-02-11 | r300g: import the last bits of libdrm and cleanup the whole thing | Marek Olšák | |
Based on Dave's branch. The majority of this commit is a cleanup, mainly renaming things. There wasn't much code to import, just ioctl calls. Also done: - implemented unsynchronized bo_map (important optimization!) - radeon_bo_is_referenced_by_cs is no longer a refcount hack - dropped the libdrm_radeon dependency I'm surprised that this has resulted in less code in the end. |