From 90027f85786406133a5180998a75fb612b6a221e Mon Sep 17 00:00:00 2001 From: Robert Ellison Date: Tue, 11 Nov 2008 13:57:10 -0700 Subject: CELL: two-sided stencil fixes With these changes, the tests/stencil_twoside test now works. - Eliminate blending from the stencil_twoside test, as it produces an unneeded dependency on having blending working - The spe_splat() function will now work if the register being splatted and the destination register are the same - Separate fragment code generated for front-facing and back-facing fragments. Often these are the same; if two-sided stenciling is on, they can be different. This is easier and faster than generating code that does both tests and merges the results. - Fixed a cut/paste bug where if the back Z-pass stencil operation were different from all the other operations, the back Z-fail results were incorrect. --- src/gallium/auxiliary/rtasm/rtasm_ppc_spe.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'src/gallium/auxiliary/rtasm') diff --git a/src/gallium/auxiliary/rtasm/rtasm_ppc_spe.c b/src/gallium/auxiliary/rtasm/rtasm_ppc_spe.c index f8568f690b..1bd9f1c8dd 100644 --- a/src/gallium/auxiliary/rtasm/rtasm_ppc_spe.c +++ b/src/gallium/auxiliary/rtasm/rtasm_ppc_spe.c @@ -958,9 +958,12 @@ spe_compare_greater_uint(struct spe_function *p, unsigned rT, unsigned rA, unsig void spe_splat(struct spe_function *p, unsigned rT, unsigned rA) { + /* Use a temporary, just in case rT == rA */ + unsigned int tmp_reg = spe_allocate_available_register(p); /* Duplicate bytes 0, 1, 2, and 3 across the whole register */ - spe_ila(p, rT, 0x00010203); - spe_shufb(p, rT, rA, rA, rT); + spe_ila(p, tmp_reg, 0x00010203); + spe_shufb(p, rT, rA, rA, tmp_reg); + spe_release_register(p, tmp_reg); } -- cgit v1.2.3