AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

r600: double multiply can handle only one multiply at a time

Browse Source 
It seems clear that trying to multiply two pairs of doubles would result
in the temporary register getting overwritten by the second pair. So
make the code more explicit.

Tested-by: Glenn Kennard <glenn.kennard@gmail.com>
Tested-by: James Harvey <lothmordor@gmail.com>
Cc: 17.0 <mesa-stable@lists.freedesktop.org>
This commit is contained in:
Nicolai Hähnle
2017-01-19 14:38:54 +01:00
parent f3f9207786
commit 76b02d2fe1
1 changed files with 19 additions and 17 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/gallium/drivers/r600/r600_shader.c
+19 -17
View File
@@ -4326,25 +4326,27 @@ static int cayman_mul_double_instr(struct r600_shader_ctx *ctx)
int lasti = tgsi_last_instruction(inst->Dst[0].Register.WriteMask);
int t1 = ctx->temp_reg;
for (k = 0; k < 2; k++) {
if (!(inst->Dst[0].Register.WriteMask & (0x3 << (k * 2))))
continue;
/* t1 would get overwritten below if we actually tried to
* multiply two pairs of doubles at a time. */
assert(inst->Dst[0].Register.WriteMask == TGSI_WRITEMASK_XY ||
inst->Dst[0].Register.WriteMask == TGSI_WRITEMASK_ZW);
for (i = 0; i < 4; i++) {
memset(&alu, 0, sizeof(struct r600_bytecode_alu));
alu.op = ctx->inst_info->op;
for (j = 0; j < inst->Instruction.NumSrcRegs; j++) {
r600_bytecode_src(&alu.src[j], &ctx->src[j], k * 2 + ((i == 3) ? 0 : 1));
}
alu.dst.sel = t1;
alu.dst.chan = i;
alu.dst.write = 1;
if (i == 3)
alu.last = 1;
r = r600_bytecode_add_alu(ctx->bc, &alu);
if (r)
return r;
k = inst->Dst[0].Register.WriteMask == TGSI_WRITEMASK_XY ? 0 : 1;
for (i = 0; i < 4; i++) {
memset(&alu, 0, sizeof(struct r600_bytecode_alu));
alu.op = ctx->inst_info->op;
for (j = 0; j < inst->Instruction.NumSrcRegs; j++) {
r600_bytecode_src(&alu.src[j], &ctx->src[j], k * 2 + ((i == 3) ? 0 : 1));
}
alu.dst.sel = t1;
alu.dst.chan = i;
alu.dst.write = 1;
if (i == 3)
alu.last = 1;
r = r600_bytecode_add_alu(ctx->bc, &alu);
if (r)
return r;
}
for (i = 0; i <= lasti; i++) {