summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* nv50/ir: guess that the constant offset is the starting slot of arrayIlia Mirkin2015-05-221-2/+4
| | | | | | | | When we get something like IN[ADDR[0].x+5], we will now guess that we should look at IN[5] for the "base" information. Signed-off-by: Ilia Mirkin <[email protected]> Cc: "10.5 10.6" <[email protected]>
* nvc0/ir: set ftz when sources are floats, not just destinationsIlia Mirkin2015-05-221-3/+2
| | | | | | | | In the case of a compare, the destination might be a predicate, but we still want to flush denorms. Signed-off-by: Ilia Mirkin <[email protected]> Cc: "10.5 10.6" <[email protected]>
* nv50/ir: allow OP_SET to merge with OP_SET_AND/etc as well as a negIlia Mirkin2015-05-221-26/+55
| | | | | | | | | | | | | | | | | | | | This covers the pattern where a KILL_IF is used, which triggers a comparison of -x to 0. This can usually be folded into the comparison whose result is being compared to 0, however it may, itself, have already been combined with another comparison. That shouldn't impact the logic of this pass however. With this and the & 1.0 change, code like 00000020: 001c0001 80081df4 set b32 $r0 lt f32 $r0 0x3e800000 00000028: 001c0000 201fc000 and b32 $r0 $r0 0x3f800000 00000030: 7f9c001e dd885c00 set $p0 0x1 lt f32 neg $r0 0x0 00000038: 0000003c 19800000 $p0 discard becomes 00000020: 001c001d b5881df4 set $p0 0x1 lt f32 $r0 0x3e800000 00000028: 0000003c 19800000 $p0 discard Signed-off-by: Ilia Mirkin <[email protected]>
* nvc0/ir: optimize set & 1.0 to produce boolean-float setsIlia Mirkin2015-05-222-0/+29
| | | | | | | | This has started to happen more now that the backend is producing KILL_IF more often. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Tobias Klausmann <[email protected]>
* nvc0/ir: allow iset to produce a boolean floatIlia Mirkin2015-05-223-5/+16
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* nvc0/ir: avoid jumping to a sched instructionIlia Mirkin2015-05-223-2/+9
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* glx: fix Scons buildBrian Paul2015-05-221-1/+1
| | | | | | Replace -h with --header-tag as was done for the Makefile build. Reviewed-by: Dylan Baker <[email protected]>
* glapi: glX_proto_size.py: use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_proto_size.py: use argparse instead of getoptDylan Baker2015-05-222-46/+46
| | | | | | | | | | This is roughly equivalent to the original getopt, except that it removes the '-h' short option, which argparse reserves for auto-generated help messages. It does retain the long option specified by the getopt version, and changes the makefile to use that. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_proto_recv.py: Use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_proto_recv.py: use argparse instead of getoptDylan Baker2015-05-221-25/+30
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapy: gl_genexec.py: use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_genexec.py: use argparse instead of getoptDylan Baker2015-05-221-17/+12
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_proto_send.py: use a main function.Dylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_proto_send.py: use argparse instead of getoptDylan Baker2015-05-221-31/+28
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: glX_server_table.py: use argparse instead of getoptDylan Baker2015-05-221-23/+15
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_SPARC_asm.py: use main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_SPARC_asm.py use argparse instead of getoptDylan Baker2015-05-221-24/+14
| | | | | | | | Also drop -m switch, which only accepted a single value or raised an error, and was unused in the makefile. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_x86-64_asm.py: Use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_x86_64_asm.py: Use argparse instead of getoptDylan Baker2015-05-221-24/+16
| | | | | | | | | Also removes the redundant -m argument, which could only be set to 'generic', or it would raise an exception. This option wasn't used in the makefile. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_x86_asm.py: use a main functionDylan Baker2015-05-221-1/+5
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_x86_asm.py: use argparse instead of getoptDylan Baker2015-05-221-25/+14
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_gentable.py: use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_gentable.py: Replace getopt with argparseDylan Baker2015-05-221-14/+15
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_apitemp.py: Use a main functionDylan Baker2015-05-221-1/+6
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_apitemp.py: Convert to argparse instead of getoptDylan Baker2015-05-221-20/+19
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_enums.py: use main() function for if __name__ == "__main__"Dylan Baker2015-05-221-1/+5
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_enums.py: use argparse instead of getopt.Dylan Baker2015-05-221-16/+14
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_procs.py: Use argparse rather than getoptDylan Baker2015-05-221-23/+19
| | | | | Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_procs.py: Fix a few low hanging style thingsDylan Baker2015-05-221-7/+16
| | | | | | | Shuts up analysis tools to make them return actual problems. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: remap_helper.py: use argparse instead of optparseDylan Baker2015-05-221-24/+22
| | | | | | | Make the code simpler, cleaner, and easier to work with. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: remap_helper.py: Fix some low hanging style issuesDylan Baker2015-05-221-4/+13
| | | | | | | | This makes the tools shut up about a bunch of problems, making them more useful for catching actual problems. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_table.py: replace getopt with argparse.Dylan Baker2015-05-221-37/+33
| | | | | | | | | | | This results in slightly less code, but code that is much more readable. It has the advantage of putting everything together in one place, all of the code is self documenting, help messages are auto-generated, choices are automatically enforced, and the syntax is much less C like, taking advantage of python features and idioms. Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* glapi: gl_table.py: Fix some low hanging style issuesDylan Baker2015-05-221-16/+26
| | | | | | | | Making the tools shut up about worthless errors so you can see real ones is very useful Signed-off-by: Dylan Baker <[email protected]> Acked-by: Matt Turner <[email protected]>
* i965/disasm: Skip swizzle disassembly when using 3-src repctrl.Matt Turner2015-05-221-9/+12
| | | | | ... since it's always .x, and also always print the subreg offset when using repctrl.
* nir: Remove sRGB colorspace conversion round-trip.Matt Turner2015-05-221-0/+2
| | | | | | | | | | | | | | | | | | Some shaders in Civilization V and Beyond Earth do pow(pow(x, 2.2), 0.454545) which is converting to and from sRGB colorspace. A more general rule that replaces pow(pow(a, b), c) with pow(a, b * c) actually regresses two shaders in Sun Temple in which the result of the inner pow is used twice, once by another pow and once by another instruction. Also, since 2.2 * 0.454545 isn't exactly one, the more general pattern would have still left us with a pow, and I'm 2.2 * 0.454545 percent sure that's not what they want. instructions in affected programs: 934 -> 886 (-5.14%) helped: 16
* nv50: fix PIPELINE_STATISTICS with HUD, based on nvc0Samuel Pitoiset2015-05-221-1/+2
| | | | | | | Tested on NVA8. No regression for ARB_pipeline_statistics piglit tests. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* nv50: fix 64-bit queries with HUD, based on nvc0Samuel Pitoiset2015-05-221-1/+13
| | | | | | | | | A sequence number is written for 32-bits queries to make sure they are ready, but not for 64-bits queries. Instead, we have to use a fence in order to fix the HUD because it doesn't wait until the result is ready. Signed-off-by: Samuel Pitoiset <[email protected]> Reviewed-by: Ilia Mirkin <[email protected]>
* radeon/vce: adapt new firmware interface changesChristian König2015-05-224-2/+252
| | | | | | | | | v2: make this also compatible with original released firmware v3 (chk): switch to original idea of separate files for fw versions Signed-off-by: Leo Liu <[email protected]> Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> (v2)
* radeon/vce: move CPB handling function into common codeChristian König2015-05-223-29/+48
| | | | | | They are not firmware version dependent. Signed-off-by: Christian König <[email protected]>
* u_math: uses assert, include assert.hDave Airlie2015-05-221-0/+1
| | | | | | | | | | | this fixes a build problem found on RHEL s390. not sure what configure options caused it, I couldn't get it on x86 here. Reviewed-by: Brian Paul <[email protected]> Cc: "10.6" [email protected] Signed-off-by: Dave Airlie <[email protected]>
* glsl: remove element_type() helperTimothy Arceri2015-05-2211-36/+23
| | | | | | | | | | | | | | We now have is_array() and without_array() that make the code much clearer and remove the need for this. For all remaining calls to this we already knew that the type was an array so returning a null wasn't adding any value. v2: use without_array() in _mesa_ast_array_index_to_hir() and don't use without_array() in lower_clip_distance_visitor() as we want to make sure the array is 2D. Reviewed-by: Matt Turner <[email protected]>
* glsl: Use AM_V_GEN/AM_V_at in NIR rules.Matt Turner2015-05-211-10/+10
|
* freedreno/a3xx: set .zw of sprite coords to .01Ilia Mirkin2015-05-201-3/+6
| | | | | | | | Fixes non-determinism in bin/point-sprite rendering, and the stars on the intro screen to neverball. Cc: "10.6" <[email protected]> Signed-off-by: Ilia Mirkin <[email protected]>
* freedreno/ir3: fix immediate usage in tgsi tex feIlia Mirkin2015-05-201-5/+9
| | | | | | | | get_immediate will return a const reference, the requested immediate isn't necessarily in the x slot. Make sure to use the swizzle. Signed-off-by: Ilia Mirkin <[email protected]> Cc: [email protected]
* targets/osmesa: drop the -module tag from LDFLAGSEmil Velikov2015-05-201-1/+0
| | | | | | | | Gallium equivalent of commit 06ff751f97f(darwin: Fix install name of libOSMesa) Cc: "10.5 10.6" <[email protected]> Signed-off-by: Emil Velikov <[email protected]>
* darwin: Fix install name of libOSMesaJeremy Huddleston Sequoia2015-05-201-1/+0
| | | | | | | | | | | | | | Passing -module to glibtool causes the resulting library to be called libSomething.so rather than libSomething.dylib on darwin. Regardless if libOSMesa is a library or a module, it has been used as the former for quite some time. Update the build to reflect that and resolve the naming issue. Cc: "10.5 10.6" <[email protected]> Signed-off-by: Jeremy Huddleston Sequoia <[email protected]> [Emil Velikov: Tweak the commit message.] Reviewed-by: Emil Velikov <[email protected]>
* swrast: Build fix for SolarisAlan Coopersmith2015-05-202-1/+4
| | | | | | | | Fixes regression from commit 5b2d3480f57168d50ad24cf0b8c9244414bd3701 Cc: "10.5 10.6" <[email protected]> Signed-off-by: Alan Coopersmith <[email protected]> Reviewed-by: Jeremy Huddleston Sequoia <[email protected]>
* nir: Get rid of the array elements parameter on load/store intrinsicsJason Ekstrand2015-05-206-56/+34
| | | | | | | | | | | | | Previously, we used intrinsic->const_index[1] to represent "the number of array elements to load" for load/store intrinsics. However, this set to 1 by every pass that ever creates a load/store intrinsic. Also, while it might make some sense for registers, it makes no sense whatsoever in SSA. On top of that, the i965 backend was the only backend to ever support it; freedreno and vc4 just assert that it's always 1. Let's just delete it. Signed-off-by: Jason Ekstrand <[email protected]> Reviewed-by: Connor Abbott <[email protected]> Reviewed-by: Rob Clark <[email protected]>
* gallium: remove TGSI_SAT_MINUS_PLUS_ONEMarek Olšák2015-05-2020-188/+47
| | | | | | | | It's a remnant of some old NV extension. Unused. I also have a patch that removes predicates if anyone is interested. Reviewed-by: Roland Scheidegger <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 11:50:46 +0000