The conditional mod of these instructions determines the semantics of
the comparison itself (rather than being evaluated based on the result
of the instruction as is usually the case for most other instructions
that allow conditional mods), so it's in general not legal to
propagate a conditional mod into a CMP instruction. This prevents
cmod propagation from (mis)optimizing:
cmp.z.f0 tmp, ...
mov.z.f0 null, tmp
into:
cmp.z.f0 tmp, ...
which gives the negation of the flag result of the original sequence.
I could reproduce this easily with SIMD32 but I don't see any reason
why the problem would be SIMD32-specific, it was most likely working
by luck.
Cc: mesa-stable@lists.freedesktop.org
Reviewed-by: Jason Ekstrand <jason@jlekstrand.net>
break;
}
+ /* The conditional mod of the CMP/CMPN instructions behaves
+ * specially because the flag output is not calculated from the
+ * result of the instruction, but the other way around, which
+ * means that even if the condmod to propagate and the condmod
+ * from the CMP instruction are the same they will in general give
+ * different results because they are evaluated based on different
+ * inputs.
+ */
+ if (scan_inst->opcode == BRW_OPCODE_CMP ||
+ scan_inst->opcode == BRW_OPCODE_CMPN)
+ break;
+
/* Otherwise, try propagating the conditional. */
enum brw_conditional_mod cond =
inst->src[0].negate ? brw_swap_cmod(inst->conditional_mod)