| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2008-09-18 | pipebuffer: New callback to flush all temporary-held buffers. | José Fonseca | |
| Used mostly to aid debugging memory issues or to clean up resources when the drivers are long lived. | |||
| 2008-09-02 | pipebuffer: Add missing break statement to cache lookup logic. | José Fonseca | |
| Second loop was never run. Spotted by Keith. | |||
| 2008-09-02 | pipebuffer: Comment the slab code. Remove the freeSlabs list. | José Fonseca | |
| The freeSlabs list is not really needed as we free empty slabs immediately. Time based cached is done separately. | |||
| 2008-08-28 | pipebuffer: Fix/add detail to the under- overflow report messages. | José Fonseca | |
| 2008-08-28 | pipebuffer: Check buffer over- & underflows when mapping/unmapping too. | José Fonseca | |
| 2008-08-26 | gallium: thread wrapper clean-up | Brian Paul | |
| In p_thread.h replace _glthread_* functions with new pipe_* functions. Remove other old cruft. | |||
| 2008-08-25 | gallium: include <sched.h> on linux to get sched_yield() proto | Brian | |
| 2008-08-24 | gallium: include u_memory.h, u_math.h | Brian | |
| 2008-08-24 | gallium: refactor/replace p_util.h with util/u_memory.h and util/u_math.h | Brian Paul | |
| Also, rename p_tile.[ch] to u_tile.[ch] | |||
| 2008-08-07 | pipebuffer: Add an extra assertion to ensure buffers do not jump between lists. | José Fonseca | |
| 2008-07-20 | pipebuffer: More detailed description of bufer over-/undereflows. | José Fonseca | |
| 2008-07-13 | pb: buffer over/underflows are errors. | José Fonseca | |
| 2008-07-09 | gallium: fix logic in pb_check_usage() | Brian Paul | |
| 2008-07-03 | pipebuffer: Silent warnings. | José Fonseca | |
| Although rarely hit in normal apps, they are too noisy with test suites. | |||
| 2008-07-02 | pipebuffer: Debug buffer manager to detect buffer under- and overflows. | José Fonseca | |
| It should detect both cpu and gpu buffer overflows. | |||
| 2008-07-02 | pipebuffer: Verify usage flag consistency. Minor cleanups. | José Fonseca | |
| 2008-06-12 | pb: don't assert(0) on failure to allocate - this is a normal condition in ↵ | Keith Whitwell | |
| many drivers | |||
| 2008-06-12 | pipebuffer: Fix improper memory free. | José Fonseca | |
| 2008-06-10 | pipebuffer: Alternative buffer manager. | José Fonseca | |
| For situations where one has a reserve memory pool, or a faster/slower pool. | |||
| 2008-06-10 | pipebuffer: Be more lenient when matching cached buffer sizes. | José Fonseca | |
| Reuse cached buffers up to twice as big a requested. | |||
| 2008-05-27 | pipebuffer: Malloc buffer provider. | José Fonseca | |
| Simple wrapper around pb_malloc_buffer_create for convenience. | |||
| 2008-05-27 | pipebuffer: Allow slab allocations of buffers of inequal size. | José Fonseca | |
| 2008-05-22 | pipebuffer: More robust face null pointers. | José Fonseca | |
| It is really the caller responsibility not to call pipebuffer with null buffers, etc. But don't let the crash happen here, and still asserting early. | |||
| 2008-05-22 | pipebuffer: Don't retry allocating in slab suballocator. | José Fonseca | |
| In pipebuffer, fencing is done at on a level above sub-allocation, so no matter how many times slab allocator retries no buffer will be freed. The pipebuffer fencing implemention already retries allocating. | |||
| 2008-05-09 | gallium: Don't serialize GPU writes. | José Fonseca | |
| Only make sure the GPU is finished with a buffer before mapping. The opposite -- waiting for the CPU to be finished before handing to the CPU -- must be done before fencing. | |||
| 2008-05-09 | pipebuffer: Temporarily reimplement validation as growable array. | José Fonseca | |
| Jose | |||
| 2008-05-08 | gallium: Make sure functions have proper prototypes and remove unused variable. | Michel Dänzer | |
| 2008-05-08 | pipebuffer: Preliminary buffer validation. | José Fonseca | |
| Use table to store a list of buffers to validate. Unfortunately cso_hash shrinks/regrows the hash every time, so still has to be addressed. Multi-thread validation is still WIP. | |||
| 2008-05-08 | pipebuffer: Don't include standard headers directly. | José Fonseca | |
| 2008-04-28 | pb: remove unused variable, squash warning | Keith Whitwell | |
| 2008-04-27 | pipebuffer: Be extra cautious with the incoming buffers. | José Fonseca | |
| A common mistake is trying to fence user or malloc buffers. So don't let the crash happen inside pipebuffer lib. | |||
| 2008-04-25 | gallium: Windows user mode portability fixes. | José Fonseca | |
| 2008-04-24 | pipebuffer: New function to flush the buffer cache. | José Fonseca | |
| 2008-04-22 | pipebuffer: Temporarily allow simultaneous CPU writes. | José Fonseca | |
| Also, fast path for re-fencing the same buffer multiple times with the same fence. | |||
| 2008-04-15 | gallium: Fix seg fault (James Vogt). | José Fonseca | |
| 2008-04-15 | gallium: Several fixes to buffer caching. | José Fonseca | |
| 2008-04-15 | gallium: Allow to use a single slab. | José Fonseca | |
| We often want to use a pool of equally sized buffers, so this makes the slab suballocator a drop-in replacement. | |||
| 2008-04-15 | gallium: Serialize buffers writes. | José Fonseca | |
| Allow concurrent reads from buffers by the CPU/GPU, but serialize all writes. | |||
| 2008-04-13 | gallium: Buffer cache. | José Fonseca | |
| 2008-04-13 | gallium: Initial port of Thomas slab suballocator to pipebuffer. | José Fonseca | |
| Not tested yet -- just compiles. This includes only the slab algorithm. Fencing is already implemented in pb_bufmgr_fence and time-based caching will be commited in a separate module shortly. | |||
| 2008-04-08 | gallium: Fix overzealous assert. | José Fonseca | |
| 2008-04-05 | gallium: Keep fenced buffers list ordered. | José Fonseca | |
| This allows to keep the list small without the overhead of full walks. | |||
| 2008-04-05 | gallium: Fix typo. | José Fonseca | |
| 2008-04-01 | gallium: Do not assume that buffers are freed in the same order they are fenced. | José Fonseca | |
| Also free buffers as soon as possible. This short term fix corrects the fenced list behavior but it will impact on performance. The long term fix is probably replace the linked list (legacy from the bufpool code) by a binary tree. | |||
| 2008-02-27 | gallium: Make headers C++ friendly. | José Fonseca | |
| 2008-02-24 | gallium: MSVC fixes. | José Fonseca | |
| 2008-02-23 | Bring in several forgotten MSVC fixes. | José Fonseca | |
| 2008-02-19 | Remove src/mesa and src/mesa/main from gallium source include paths. | José Fonseca | |
| 2008-02-19 | Simplify makefile boilerplate code. | José Fonseca | |
| Don't define ASM_SOURCES variable globally -- reserve that variable to be defined locally by makefiles, together with C_SOURCES and CPP_SOURCES. | |||
| 2008-02-19 | Move mm.c code into util module. | José Fonseca | |
| Using the u_ prefix to distingish the c source files that support gallium interfaces and those that have really no relation with gallium itself. | |||
 |
index : android-x86-mesa.git | |
| Androïd/x86 port of Mesa drivers | Hugues Hiegel |
| summaryrefslogtreecommitdiff |