This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[ARM] Fix build failure due to movsi_compare0


A recent change somewhere exposed a latent bug between LRA and the definition
of the movsi_compare0 pattern.

This pattern ties the source and destination register of a set together
a (match_dup) and register constraints:

   [(set (reg:CC CC_REGNUM)
	(compare:CC (match_operand:SI 1 "s_register_operand" "0,r")
		    (const_int 0)))
    (set (match_operand:SI 0 "s_register_operand" "=r,r")
	(match_dup 1))]

This confuses LRA which expects the source and destination register of
a set to be different.

reduced.c: In function '_IO_vfscanf_internal':
reduced.c:104:1: internal compiler error: in lra_create_copy, at lra.c:1512
0x8c3f9a lra_create_copy(int, int, int)
0x8e4ab0 process_bb_lives
0x8e4ab0 lra_create_live_ranges(bool)
0x8c5a39 lra(_IO_FILE*)
0x873a96 do_reload
0x873a96 execute
Please submit a full bug report,

We can fix the pattern by moving away from match_dup and using register
tying with constraints consistently.

I'm not entirely convinced that this is legitimate (my vague recollection is
that register tying should only be used to tie inputs to outputs).

This has passed testing on a bunch of ARM targets, and fixes the build
issues I've been seeing.

OK for trunk?



2014-06-11  James Greenhalgh  <>

	* config/arm/ (movsi_compare0): Clarify intentions using
	register tying.
diff --git a/gcc/config/arm/ b/gcc/config/arm/
index f58a79b..a01333b 100644
--- a/gcc/config/arm/
+++ b/gcc/config/arm/
@@ -6582,10 +6582,10 @@
 (define_insn "*movsi_compare0"
   [(set (reg:CC CC_REGNUM)
-	(compare:CC (match_operand:SI 1 "s_register_operand" "0,r")
-		    (const_int 0)))
+	(compare:CC (match_operand:SI 2 "s_register_operand" "0,1")
+	    (const_int 0)))
    (set (match_operand:SI 0 "s_register_operand" "=r,r")
-	(match_dup 1))]
+	(match_operand:SI 1 "s_register_operand" "0,r"))]
    cmp%?\\t%0, #0

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]