* Mainline regression - ICE c99-complit-2.c
@ 2004-03-19 22:12 Zack Weinberg
0 siblings, 0 replies; 2+ messages in thread
From: Zack Weinberg @ 2004-03-19 22:12 UTC (permalink / raw)
To: gcc, Richard Kenner
c99-complit-2.c ICEs on mainline since last afternoon, with this error
message:
gcc.dg/c99-complit-2.c:25: internal compiler error: tree check:
expected integer_type, enumeral_type, boolean_type, char_type or
real_type; have array_type in safe_from_p, at expr.c:5751
(gdb) frame 5
#5 0x08160897 in safe_from_p (x=0x401f42a0, exp=0x401f0af0, top_p=1)
at ../../gcc/expr.c:5751
5751 || TYPE_ARRAY_MAX_SIZE (TREE_TYPE (exp)) == NULL_TREE
exp here is a CONSTRUCTOR whose type is an ARRAY_TYPE. The problem is
in the definition of TYPE_ARRAY_MAX_SIZE:
#define TYPE_ARRAY_MAX_SIZE(ARRAY_TYPE) \
TYPE_MAX_VALUE (ARRAY_TYPE_CHECK (ARRAY_TYPE))
TYPE_MAX_VALUE applies NUMERICAL_TYPE_CHECK, which is mutually
exclusive with ARRAY_TYPE_CHECK. This ought to be
#define TYPE_ARRAY_MAX_SIZE(ARRAY_TYPE) \
(ARRAY_TYPE_CHECK (ARRAY_TYPE)->type.maxval)
I have far too many unrelated changes in my tree right now to fix
this myself, nor do I have the time to dig through tree.h for similar
errors. Please clean up your mess.
zw
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Mainline regression - ICE c99-complit-2.c
@ 2004-03-19 22:34 Richard Kenner
0 siblings, 0 replies; 2+ messages in thread
From: Richard Kenner @ 2004-03-19 22:34 UTC (permalink / raw)
To: zack; +Cc: gcc
c99-complit-2.c ICEs on mainline since last afternoon, with this error
message:
I wasn't getting that, but will investigate shortly.
#define TYPE_ARRAY_MAX_SIZE(ARRAY_TYPE) \
(ARRAY_TYPE_CHECK (ARRAY_TYPE)->type.maxval)
I believe that's right.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2004-03-19 21:02 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-03-19 22:12 Mainline regression - ICE c99-complit-2.c Zack Weinberg
2004-03-19 22:34 Richard Kenner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).