This is the mail archive of the gcc@gcc.gnu.org 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]

[tree-ssa] Merge as of 2004-01-26


Thanks to everyone who helped out with the merge.  There should not be
new regressions except:

On x86-64, we are failing gcc.c-torture/unsorted/opout.c and
gcc.dg/i386-sse-6.c.

Neither the branch nor mainline (as of 2004-01-26) bootstrap on ia64. 
The failure is GGC related, stage1/cc1 dies in ia64_reorg trying to
access a poisoned page.  Mainline is broken for x86 today, so I think
I'll wait for the next merge, or maybe somebody already knows about this
failure?

This merge has been horrible enough, so I decided not to wait any
longer.  We can deal with this IA64 failure later.

(gdb) run
Starting program: /notnfs/dnovillo/BLD-merge-tree-ssa-native.tonic2.ia64/gcc/stage1/cc1 -fpreprocessed c-decl.i -quiet -dumpbase c-decl.c -auxbase-strip c-decl.o -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long -Wold-style-definition -Werror -version -fno-common -o c-decl.s
GNU C version 3.5-tree-ssa 20040130 (merged 20040126) (ia64-unknown-linux-gnu)
        compiled by GNU C version 3.2.3 20030502 (Red Hat Linux 3.2.3-20).
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096

Program received signal SIGSEGV, Segmentation fault.
get_addr (x=0x6000000000762268)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/alias.c:1506
1506        if (CONSTANT_P (l->loc))
(gdb) p l
$1 = (struct elt_loc_list *) 0xafafafafafafafaf
(gdb) bt
#0  get_addr (x=0x6000000000762268)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/alias.c:1506
#1  0x40000000003df080 in write_dependence_p (mem=0x2000000002e16a40,
    x=0x2000000002e17980, writep=0, constp=1)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/alias.c:2253
#2  0x40000000003df680 in anti_dependence (mem=0x2000000002e16a40,
    x=0x2000000002e17980)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/alias.c:2288
#3  0x4000000000ef85a0 in sched_analyze_1 (deps=0x60000fffffffac20,
    x=0x20000000035e9660, insn=0x2000000002cf6bc0)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/sched-deps.c:567
#4  0x4000000000efa260 in sched_analyze_insn (deps=0x60000fffffffac20,
    x=0x20000000035e9660, insn=0x2000000002cf6bc0, loop_notes=0x0)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/sched-deps.c:820
#5  0x4000000000eff2d0 in sched_analyze (deps=0x60000fffffffac20,
    head=0x2000000002d7f390, tail=0x2000000003270c80)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/sched-deps.c:1181
#6  0x4000000000f042e0 in schedule_ebb (head=0x2000000002d7f390,
    tail=0x2000000003270c80)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/sched-ebb.c:493
#7  0x4000000000f05040 in schedule_ebbs (dump_file=0x0)
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/sched-ebb.c:622
#8  0x4000000000d29f60 in ia64_reorg ()
    at /home/cygnus/dnovillo/merge-tree-ssa/src.ia64/gcc/config/ia64/ia64.c:7500


Bootstrapped and tested on x86 and x86-64.


Diego.


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