630fa2688634365c03edf2a189cf9225899fbcc5 |
|
19-Apr-2012 |
James Benton <jbenton@vmware.com> |
gallivm: llvm c-style for loops, allows us to create loops with conditions on entry, rather than condition check on loop
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
efc82aef35a2aac5d2ed9774f6d28f2626796416 |
|
01-Dec-2010 |
Brian Paul <brianp@vmware.com> |
gallivm/llvmpipe: squash merge of the llvm-context branch This branch defines a gallivm_state structure which contains the LLVMBuilderRef, LLVMContextRef, etc. All data structures built with this object can be periodically freed during a "garbage collection" operation. The gallivm_state object has to be passed to most of the builder functions where LLVMBuilderRef used to be used. Conflicts: src/gallium/auxiliary/gallivm/lp_bld_tgsi_soa.c src/gallium/drivers/llvmpipe/lp_state_setup.c
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
693667bf88abcb60332a6f94c8f06a9fe7d62613 |
|
10-Oct-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Use variables instead of Phis in loops. With this commit all explicit Phi emission is now gone.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
307df6a858dcab1bc10f3f52d9968acb3ea6d74f |
|
09-Oct-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Cleanup the rest of the flow module.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
d0ea4641597d23df2198fd76ed7430c06cef8c5d |
|
09-Oct-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Simplify if/then/else implementation. No need for for a flow stack anymore.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
d45c379027054e563c4f4379fb69fc9f68612f75 |
|
09-Oct-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Remove support for Phi generation. Simply rely on mem2reg pass. It's easier and more reliable.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
cc40abad519cc0f765c6d8f6fad4154bed8dd9c2 |
|
09-Oct-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Don't generate Phis for execution mask.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
aa4cb5e2d8d48c7dcc9653c61a9e25494e3e7b2a |
|
07-Oct-2010 |
Keith Whitwell <keithw@vmware.com> |
llvmpipe: try to be sensible about whether to branch after mask updates Don't branch more than once in quick succession. Don't branch at the end of the shader.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
a18c210a95794c79c6f26dbf4c66d4a85e29169d |
|
27-Apr-2010 |
José Fonseca <jfonseca@vmware.com> |
gallivm: Ensure all allocas are in the first block. Refactor the code to make this easier.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
9dd70e7b85ddbc73bd976c4dab81476aa36c557e |
|
06-Apr-2010 |
Zack Rusin <zackr@vmware.com> |
draw llvm: fix loop iteration and vertex header offsets the loop was doing a NE comparison which we could have skipped if the prim was triangles (3 verts) and our step was 4 verts. also fix offsets in conversion to aos.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
0d71ba46e613230c84165106c1fcc9027dec4cd3 |
|
15-Mar-2010 |
Brian Paul <brianp@vmware.com> |
gallivm/llvmpipe: rename os_llvm.h to lp_bld.h The llvm wrapper wasn't really an OS thing. Use lp_bld.h for now but we eventually should rename/re-prefix all the files/functions in the gallivm/ directory.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
489af2a3ba467e4341cb8504a0e59cf5828864d4 |
|
11-Mar-2010 |
Brian Paul <brianp@vmware.com> |
gallivm/llvmpipe: include os_llvm.h instead of llvm-c/Core.h
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
18d406e8a8a838c82ee4ec5dbf244ab8bba0855e |
|
09-Mar-2010 |
Zack Rusin <zackr@vmware.com> |
gallivm: implement loops
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|
c61bf363937f40624a5632745630d4f2b9907082 |
|
09-Feb-2010 |
Zack Rusin <zackr@vmware.com> |
llvmpipe: export the tgsi translation code to a common layer the llvmpipe tgsi translation is a lot more complete than what was in gallivm so replacing the latter with the former. this is needed since the draw llvm paths will use the same code. effectively the proven llvmpipe code becomes gallivm.
/external/mesa3d/src/gallium/auxiliary/gallivm/lp_bld_flow.h
|