Previously SHADER_OPCODE_MULH could only exist on Gen7+, so the
assertion assumed the Gen7+ accumulator rules. A future patch will
allow this instruction on at least Gen6, so update the assertion.
v2: Use get_lowered_simd_width instead of open coding it. Suggested by
Curro.
Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Reviewed-by: Matt Turner <mattst88@gmail.com> [v1]
using namespace brw;
+static unsigned get_lowered_simd_width(const struct brw_device_info *devinfo,
+ const fs_inst *inst);
+
void
fs_inst::init(enum opcode opcode, uint8_t exec_size, const fs_reg &dst,
const fs_reg *src, unsigned sources)
} else if (inst->opcode == SHADER_OPCODE_MULH) {
/* Should have been lowered to 8-wide. */
- assert(inst->exec_size <= 8);
+ assert(inst->exec_size <= get_lowered_simd_width(devinfo, inst));
const fs_reg acc = retype(brw_acc_reg(inst->exec_size),
inst->dst.type);
fs_inst *mul = ibld.MUL(acc, inst->src[0], inst->src[1]);