aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorHans-Peter Nilsson <hp@bitrange.com>2020-07-29 02:46:09 +0200
committerHans-Peter Nilsson <hp@bitrange.com>2020-07-29 02:46:09 +0200
commit5c180464b7b0827b3cc07a78e96dfe55352db33f (patch)
tree446856e47fb81d05e5b485d72245a99c1d55394e
parenteb08b5d1d79dca9d683d817806b338ad741da242 (diff)
downloadgcc-5c180464b7b0827b3cc07a78e96dfe55352db33f.zip
gcc-5c180464b7b0827b3cc07a78e96dfe55352db33f.tar.gz
gcc-5c180464b7b0827b3cc07a78e96dfe55352db33f.tar.bz2
config/mmix/mmix.h (NO_FUNCTION_CSE): Define to 1.
The tests gcc.dg/tree-ssa/loop-1.c and gcc.dg/weak/typeof-2.c assume this setting and are as a consequence riddled with exceptions for targets that actually do yield better code when calling through a register rather than repeatedly the same symbol. Nonetheless, defining it makes sense for MMIX. (Even better IMHO, this macro shouldn't exist and instead rtx costs be used or perhaps just a target-specific default of flag_no_function_cse.) gcc: * config/mmix/mmix.h (NO_FUNCTION_CSE): Define to 1.
-rw-r--r--gcc/config/mmix/mmix.h3
1 files changed, 3 insertions, 0 deletions
diff --git a/gcc/config/mmix/mmix.h b/gcc/config/mmix/mmix.h
index 9317c20..ac0be10 100644
--- a/gcc/config/mmix/mmix.h
+++ b/gcc/config/mmix/mmix.h
@@ -577,6 +577,9 @@ typedef struct { int regs; int lib; } CUMULATIVE_ARGS;
#define SLOW_BYTE_ACCESS 0
+/* A PUSHJ doesn't cost more than a PUSHGO, so don't needlessly create
+ the latter. */
+#define NO_FUNCTION_CSE 1
/* Node: Sections */