Whoops. This little gem had the effect of making the output
operand (0) constraints disappear but not the input operand (1)
constraints for define_subst:ed patterns, probably because
there's another (match_dup 1) in the output template (not
investigated).
That went surprisingly unnoticed until I added a pass leaning
just a little bit harder on the define_subst:ed patterns and
then only by the libgfortran library generating assembly with
nominally incorrect syntax. (There was a move to a special
register from a general register, and it incorrectly matched a
pattern affecting condition codes.)
gcc:
* config/cris/cris.md ("setnz_subst", "setnz_subst", "setcc_subst"):
Use match_dup in output template, not match_operand.
---
gcc/config/cris/cris.md | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/gcc/config/cris/cris.md b/gcc/config/cris/cris.md
index c36a5402be3..074f5234402 100644
--- a/gcc/config/cris/cris.md
+++ b/gcc/config/cris/cris.md
@@ -275,7 +275,7 @@ (define_subst "setnz_subst"
"reload_completed"
[(set (reg:CC_NZ CRIS_CC0_REGNUM)
(compare:CC_NZ (match_dup 1) (const_int 0)))
- (set (match_operand 0) (match_operand 1))])
+ (set (match_dup 0) (match_dup 1))])
(define_subst_attr "setnzvc" "setnzvc_subst" "" "_setnzvc")
(define_subst_attr "ccnzvc" "setnzvc_subst" "" "_enabled")
@@ -288,7 +288,7 @@ (define_subst "setnzvc_subst"
"reload_completed"
[(set (reg:CC_NZVC CRIS_CC0_REGNUM)
(compare:CC_NZVC (match_dup 1) (const_int 0)))
- (set (match_operand 0) (match_operand 1))])
+ (set (match_dup 0) (match_dup 1))])
(define_subst_attr "setcc" "setcc_subst" "" "_setcc")
(define_subst_attr "cccc" "setcc_subst" "" "_enabled")
@@ -301,7 +301,7 @@ (define_subst "setcc_subst"
"reload_completed"
[(set (reg:CC CRIS_CC0_REGNUM)
(compare:CC (match_dup 1) (const_int 0)))
- (set (match_operand 0) (match_operand 1))])
+ (set (match_dup 0) (match_dup 1))])
;; Operand and operator predicates.
--
2.11.0
