public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit
@ 2013-06-27 15:56 Martin.Jansa at gmail dot com
  2013-06-27 16:43 ` [Bug target/57735] " ktkachov at gcc dot gnu.org
                   ` (8 more replies)
  0 siblings, 9 replies; 10+ messages in thread
From: Martin.Jansa at gmail dot com @ 2013-06-27 15:56 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

            Bug ID: 57735
           Summary: ICE with -mtune=xscale (error: could not split insn)
                    when building webkit
           Product: gcc
           Version: 4.8.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: Martin.Jansa at gmail dot com

Created attachment 30391
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=30391&action=edit
JSGlobalObjectFunctions.cpp.tar.bz2

I'll try to attach smaller test-case later, but real-world reproducer follows:

/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xscalete-oe-linux-gnueabi/arm-oe-linux-gnueabi-g++
-v
Using built-in specs.
COLLECT_GCC=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xscalete-oe-linux-gnueabi/arm-oe-linux-gnueabi-g++
COLLECT_LTO_WRAPPER=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/libexec/xscalete-oe-linux-gnueabi/gcc/arm-oe-linux-gnueabi/4.8.1/lto-wrapper
Target: arm-oe-linux-gnueabi
Configured with:
/OE/shr-core/tmp-eglibc/work-shared/gcc-4.8.1-r0/gcc-4.8.1/configure
--build=x86_64-linux --host=x86_64-linux --target=arm-oe-linux-gnueabi
--prefix=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr
--exec_prefix=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr
--bindir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xscalete-oe-linux-gnueabi
--sbindir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/bin/xscalete-oe-linux-gnueabi
--libexecdir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/libexec/xscalete-oe-linux-gnueabi
--datadir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/share
--sysconfdir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/etc
--sharedstatedir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/com
--localstatedir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/var
--libdir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/lib/xscalete-oe-linux-gnueabi
--includedir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/include
--oldincludedir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/include
--infodir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/share/info
--mandir=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr/share/man
--disable-silent-rules --disable-dependency-tracking
--with-libtool-sysroot=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux
--with-gnu-ld --enable-shared --enable-languages=c,c++ --enable-threads=posix
--disable-multilib --enable-c99 --enable-long-long --enable-symvers=gnu
--enable-libstdcxx-pch --program-prefix=arm-oe-linux-gnueabi-
--without-local-prefix --enable-target-optspace --enable-lto --enable-libssp
--disable-bootstrap --disable-libmudflap --with-system-zlib
--with-linker-hash-style=gnu --enable-linker-build-id --with-ppl=no
--with-cloog=no --enable-checking=release --enable-cheaders=c_global
--with-float=soft
--with-gxx-include-dir=/OE/shr-core/tmp-eglibc/sysroots/spitz/usr/include/c++
--with-sysroot=/OE/shr-core/tmp-eglibc/sysroots/spitz
--with-build-sysroot=/OE/shr-core/tmp-eglibc/sysroots/spitz
--enable-poison-system-directories --disable-libunwind-exceptions
--with-mpfr=/OE/shr-core/tmp-eglibc/sysroots/x86_64-linux/usr
--with-system-zlib --disable-nls
Thread model: posix
gcc version 4.8.1 (GCC) 


/OE/shr-core/tmp-eglibc/work/xscalete-oe-linux-gnueabi/webkit-efl/1.9.3+svnr120144-r3/src/Source/JavaScriptCore/runtime/JSGlobalObjectFunctions.cpp:150:1:
error: could not split insn
 }
 ^
(insn 505 8094 507 (set (reg:SI 1 r1 [orig:494 D.142559 ] [494])
        (plus:SI (reg/v:SI 5 r5 [orig:157 character ] [157])
            (const_int -17179934720 [0xfffffffbffff0000])))
/OE/shr-core/tmp-eglibc/work/xscalete-oe-linux-gnueabi/webkit-efl/1.9.3+svnr120144-r3/src/Source/JavaScriptCore/runtime/JSGlobalObjectFunctions.cpp:108
5 {*arm_addsi3}
     (nil))
/OE/shr-core/tmp-eglibc/work/xscalete-oe-linux-gnueabi/webkit-efl/1.9.3+svnr120144-r3/src/Source/JavaScriptCore/runtime/JSGlobalObjectFunctions.cpp:150:1:
internal compiler error: in final_scan_insn, at final.c:2897
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
{standard input}: Assembler messages:
{standard input}: Error: open CFI at the end of file; missing .cfi_endproc
directive

Removing -mtune=xscale is enough to pass that (the same source builds fine for
other devices I'm using), using -mtune=arm926ej-s also works.

Attached archive contains 3 files:
-rw-r--r-- 1 bitbake bitbake 1932165 Jun 27 17:37
CMakeFiles/javascriptcore_efl.dir/runtime/JSGlobalObjectFunctions.cpp.E    #
preprocessed source
-rw-r--r-- 1 bitbake bitbake    3809 Jun 27 17:45
CMakeFiles/javascriptcore_efl.dir/runtime/JSGlobalObjectFunctions.cpp.cmd  #
command line used
-rw-r--r-- 1 bitbake bitbake   28308 Jun 27 17:46
CMakeFiles/javascriptcore_efl.dir/runtime/JSGlobalObjectFunctions.cpp.log  #
output with couple of warnings before that ICE

Sorry for packing it, I wasn't sure if you would prefer 2MB preprocessed
attachment.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
@ 2013-06-27 16:43 ` ktkachov at gcc dot gnu.org
  2013-06-28  8:20 ` ktkachov at gcc dot gnu.org
                   ` (7 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-06-27 16:43 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #1 from ktkachov at gcc dot gnu.org ---
Confirmed with 4.8.1
Works with 4.7 and current trunk 4.9 though.

Fails for -mcpu=xscale , iwmmxt and iwmmxt2

Trying to reduce...


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
  2013-06-27 16:43 ` [Bug target/57735] " ktkachov at gcc dot gnu.org
@ 2013-06-28  8:20 ` ktkachov at gcc dot gnu.org
  2013-06-28 10:27 ` ktkachov at gcc dot gnu.org
                   ` (6 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-06-28  8:20 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #2 from ktkachov at gcc dot gnu.org ---
Created attachment 30406
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=30406&action=edit
Reduced testcase

Reduced testcase, compile with -march=armv5te -marm  -mtune=xscale
-mfloat-abi=soft  -O1


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
  2013-06-27 16:43 ` [Bug target/57735] " ktkachov at gcc dot gnu.org
  2013-06-28  8:20 ` ktkachov at gcc dot gnu.org
@ 2013-06-28 10:27 ` ktkachov at gcc dot gnu.org
  2013-07-01 11:19 ` ktkachov at gcc dot gnu.org
                   ` (5 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-06-28 10:27 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

ktkachov at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-06-28
     Ever confirmed|0                           |1

--- Comment #3 from ktkachov at gcc dot gnu.org ---
Thus confirmed.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (2 preceding siblings ...)
  2013-06-28 10:27 ` ktkachov at gcc dot gnu.org
@ 2013-07-01 11:19 ` ktkachov at gcc dot gnu.org
  2013-07-01 16:16 ` ktkachov at gcc dot gnu.org
                   ` (4 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-07-01 11:19 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #5 from ktkachov at gcc dot gnu.org ---
(In reply to Mikael Pettersson from comment #4)
> The ICE on the reduced test case started with r186147 and stopped with
> r199188.
> They both touch the same code so I believe r199188 is a proper fix and not
> just coincidental.  Backporting r199188 to current 4.8 branch fixes the ICE
> there too,
> but I haven't done a full bootstrap and regression test run on anything
> except x86_64 yet.
> 
> r199188 did cause a regression (ICE on s390x), but the fix for that has
> already been backported to 4.8 branch (at r199813).

Thanks for investigating. I'll run a bootstrap on arm with the backported
patch.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (3 preceding siblings ...)
  2013-07-01 11:19 ` ktkachov at gcc dot gnu.org
@ 2013-07-01 16:16 ` ktkachov at gcc dot gnu.org
  2013-07-01 18:30 ` mikpe at it dot uu.se
                   ` (3 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-07-01 16:16 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #6 from ktkachov at gcc dot gnu.org ---
(In reply to Mikael Pettersson from comment #4)
> The ICE on the reduced test case started with r186147 and stopped with
> r199188.
> They both touch the same code so I believe r199188 is a proper fix and not
> just coincidental.  Backporting r199188 to current 4.8 branch fixes the ICE
> there too,
> but I haven't done a full bootstrap and regression test run on anything
> except x86_64 yet.
> 
> r199188 did cause a regression (ICE on s390x), but the fix for that has
> already been backported to 4.8 branch (at r199813).

After looking at it more closely: Mikael, are you sure the revisions are right?
It seems to me that r199813 is just the backport of r199188.

Can you please double-check the revision that fixes this current ICE?

Thanks,
Kyrill


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (4 preceding siblings ...)
  2013-07-01 16:16 ` ktkachov at gcc dot gnu.org
@ 2013-07-01 18:30 ` mikpe at it dot uu.se
  2013-09-09 13:11 ` ktkachov at gcc dot gnu.org
                   ` (2 subsequent siblings)
  8 siblings, 0 replies; 10+ messages in thread
From: mikpe at it dot uu.se @ 2013-07-01 18:30 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #7 from Mikael Pettersson <mikpe at it dot uu.se> ---
(In reply to ktkachov from comment #6)
> After looking at it more closely: Mikael, are you sure the revisions are
> right? It seems to me that r199813 is just the backport of r199188.
> 
> Can you please double-check the revision that fixes this current ICE?

Oops. The ICE was fixed by r198462, r199188 is the followup. Sorry about that.


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (5 preceding siblings ...)
  2013-07-01 18:30 ` mikpe at it dot uu.se
@ 2013-09-09 13:11 ` ktkachov at gcc dot gnu.org
  2013-09-09 13:16 ` ktkachov at gcc dot gnu.org
  2013-09-09 13:17 ` ktkachov at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-09-09 13:11 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #8 from ktkachov at gcc dot gnu.org ---
Author: ktkachov
Date: Mon Sep  9 13:11:45 2013
New Revision: 202397

URL: http://gcc.gnu.org/viewcvs?rev=202397&root=gcc&view=rev
Log:
[gcc/]
2013-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

    PR target/57735
    Backport from mainline
    2013-04-30  Richard Sandiford  <rsandifo@linux.vnet.ibm.com>

    * explow.c (plus_constant): Pass "mode" to immed_double_int_const.
    Use gen_int_mode rather than GEN_INT.

[gcc/testsuite]
2013-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

    PR target/57735
    * g++.dg/ext/pr57735.C: New test.

Added:
    branches/gcc-4_8-branch/gcc/testsuite/g++.dg/ext/pr57735.C
Modified:
    branches/gcc-4_8-branch/gcc/ChangeLog
    branches/gcc-4_8-branch/gcc/explow.c
    branches/gcc-4_8-branch/gcc/testsuite/ChangeLog


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (6 preceding siblings ...)
  2013-09-09 13:11 ` ktkachov at gcc dot gnu.org
@ 2013-09-09 13:16 ` ktkachov at gcc dot gnu.org
  2013-09-09 13:17 ` ktkachov at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-09-09 13:16 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

--- Comment #9 from ktkachov at gcc dot gnu.org ---
Author: ktkachov
Date: Mon Sep  9 13:16:44 2013
New Revision: 202398

URL: http://gcc.gnu.org/viewcvs?rev=202398&root=gcc&view=rev
Log:
2013-09-09  Kyrylo Tkachov  <kyrylo.tkachov@arm.com>

    PR target/57735
    * g++.dg/ext/pr57735.C: New test.

Added:
    trunk/gcc/testsuite/g++.dg/ext/pr57735.C
Modified:
    trunk/gcc/testsuite/ChangeLog


^ permalink raw reply	[flat|nested] 10+ messages in thread

* [Bug target/57735] ICE with -mtune=xscale (error: could not split insn) when building webkit
  2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
                   ` (7 preceding siblings ...)
  2013-09-09 13:16 ` ktkachov at gcc dot gnu.org
@ 2013-09-09 13:17 ` ktkachov at gcc dot gnu.org
  8 siblings, 0 replies; 10+ messages in thread
From: ktkachov at gcc dot gnu.org @ 2013-09-09 13:17 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57735

ktkachov at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED

--- Comment #10 from ktkachov at gcc dot gnu.org ---
Fixed in 4.8 branch. Testcase added in trunk.


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2013-09-09 13:17 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-06-27 15:56 [Bug c++/57735] New: ICE with -mtune=xscale (error: could not split insn) when building webkit Martin.Jansa at gmail dot com
2013-06-27 16:43 ` [Bug target/57735] " ktkachov at gcc dot gnu.org
2013-06-28  8:20 ` ktkachov at gcc dot gnu.org
2013-06-28 10:27 ` ktkachov at gcc dot gnu.org
2013-07-01 11:19 ` ktkachov at gcc dot gnu.org
2013-07-01 16:16 ` ktkachov at gcc dot gnu.org
2013-07-01 18:30 ` mikpe at it dot uu.se
2013-09-09 13:11 ` ktkachov at gcc dot gnu.org
2013-09-09 13:16 ` ktkachov at gcc dot gnu.org
2013-09-09 13:17 ` ktkachov at gcc dot gnu.org

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).