| Message ID | 032f2366a4cd57f781f2093d977b9cf9600c83b8.1585760495.git.segher@kernel.crashing.org |
|---|---|
| State | New |
| Headers | show |
| Series | rs6000: Make code questionably using r2 not ICE (PR94420) | expand |
diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md index dcccb03..11ab745 100644 --- a/gcc/config/rs6000/rs6000.md +++ b/gcc/config/rs6000/rs6000.md @@ -10311,7 +10311,8 @@ (define_insn "*largetoc_low_aix<mode>" (define_insn_and_split "*tocref<mode>" [(set (match_operand:P 0 "gpc_reg_operand" "=b") (match_operand:P 1 "small_toc_ref" "R"))] - "TARGET_TOC" + "TARGET_TOC + && legitimate_constant_pool_address_p (operands[1], QImode, false)" "la %0,%a1" "&& TARGET_CMODEL != CMODEL_SMALL && reload_completed" [(set (match_dup 0) (high:P (match_dup 1)))
The example code in the PR uses r2 (the TOC register) directly. In the RTL generated for that, r2 is copied to some pseudo, and then cprop propagates that into a "*tocref<mode>" insn, because nothing is preventing it from doing that. So, put the same condition in the insn condition for this as we will later encounter in the constraint anyway, fixing this. Tested on powerpc64-linux {-m32,-m64}. Committed to master. 2020-04-01 Segher Boessenkool <segher@kernel.crashing.org> PR target/94420 * config/rs6000/rs6000.md (*tocref<mode> for P): Add insn condition on operands[1]. --- gcc/config/rs6000/rs6000.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)