* [Bug c/28193] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
@ 2006-06-28 17:53 ` sjackman at gmail dot com
2006-06-28 20:40 ` [Bug target/28193] " sjackman at gmail dot com
` (6 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-28 17:53 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from sjackman at gmail dot com 2006-06-28 17:36 -------
Created an attachment (id=11772)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11772&action=view)
Backport thumb_find_work_register from 4.1.1
2005-03-01 Nick Clifton <nickc@redhat.com>
* config/arm/arm.c (thumb_find_work_register): Check all of the
argument registers to see if they are free, and a couple of
special cases where the last argument register but can be proved
to be available during the function's prologue.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
2006-06-28 17:53 ` [Bug c/28193] " sjackman at gmail dot com
@ 2006-06-28 20:40 ` sjackman at gmail dot com
2006-06-28 21:06 ` sjackman at gmail dot com
` (5 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-28 20:40 UTC (permalink / raw)
To: gcc-bugs
------- Comment #2 from sjackman at gmail dot com 2006-06-28 20:18 -------
This proposed patch does help. At the very least, it prevents the ICE on
compiling crtstuff.c while compiling the toolchain. However, with this patch
applied, I saw the same bug later while compiling newlib. As the comment in
thumb_find_work_register suggests, the real bug is probably in
thumb_compute_save_reg_mask.
make[8]: Entering directory
`/home/sjackman/src/toolchain/src/_build/arm-elf/thumb/pic/xip/newlib/libc/stdio'
arm-elf-gcc -B/home/sjackman/src/toolchain/src/_build/arm-elf/newlib/ -isystem
/home/sjackman/src/toolchain/src/_build/arm-elf/newlib/targ-include -isystem
/home/sjackman/src/toolchain/src/newlib/libc/include -DPACKAGE_NAME=\"newlib\"
-DPACKAGE_TARNAME=\"newlib\" -DPACKAGE_VERSION=\"1.14.0\"
-DPACKAGE_STRING=\"newlib\1.14.0\" -DPACKAGE_BUGREPORT=\"\" -I.
-I../../../../../../../../newlib/libc/stdio -O2 -D__NO_SYSCALLS__
-DNO_FLOATING_POINT -DHAVE_FCNTL -DHAVE_NANOSLEEP -fno-builtin -O2 -g -O2
-mthumb -fPIC -msingle-pic-base -c -o lib_a-findfp.o `test -f 'findfp.c' ||
echo '../../../../../../../../newlib/libc/stdio/'`findfp.c
../../../../../../../../newlib/libc/stdio/findfp.c: In function 'std':
../../../../../../../../newlib/libc/stdio/findfp.c:61: internal compiler
error:in thumb_find_work_register, at config/arm/arm.c:3173
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
make[8]: *** [lib_a-findfp.o] Error 1
make[8]: Leaving directory
`/home/sjackman/src/toolchain/src/_build/arm-elf/thumb/pic/xip/newlib/libc/stdio'
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
2006-06-28 17:53 ` [Bug c/28193] " sjackman at gmail dot com
2006-06-28 20:40 ` [Bug target/28193] " sjackman at gmail dot com
@ 2006-06-28 21:06 ` sjackman at gmail dot com
2006-06-28 23:31 ` sjackman at gmail dot com
` (4 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-28 21:06 UTC (permalink / raw)
To: gcc-bugs
------- Comment #3 from sjackman at gmail dot com 2006-06-28 20:51 -------
I tried backporting thumb_compute_save_reg_mask from GCC 4.1.1 to GCC 4.0.3
without success. I'll try backporting this entire patch from svn.
2005-03-01 Nick Clifton <nickc@redhat.com>
* config/arm/arm.c (thumb_find_work_register): Check all of the
argument registers to see if they are free, and a couple of
special cases where the last argument register but can be proved
to be available during the function's prologue.
(print_multi_reg, arm_compute_save_reg0_reg12_mask,
output_return_instruction, emit_multi_reg_push, thumb_pushpop,
thumb_unexpanded_epilogue): Use unsigned long as the type for the
register bit-mask.
(thumb_compute_save_reg_mask): Likewise. Also use
thumb_find_work_register() to ensure that there is agreement about
which work register is going to be used in the prologue.
(thumb_output_function_prologue): Use unsigned long as the type
for the register bit-mask. Also delay pushing the link register if
other high registers are going to be pushed.
(thumb_compute_save_reg_mask, emit_multi_reg_push,
print_multi-reg, number_of_first_bit_set, thumb_pushpop): Remove
redundant prototypes.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
` (2 preceding siblings ...)
2006-06-28 21:06 ` sjackman at gmail dot com
@ 2006-06-28 23:31 ` sjackman at gmail dot com
2006-06-28 23:32 ` [Bug target/28193] [4.0 only] " pinskia at gcc dot gnu dot org
` (3 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-28 23:31 UTC (permalink / raw)
To: gcc-bugs
------- Comment #4 from sjackman at gmail dot com 2006-06-28 22:30 -------
I tried applying r95736 (2005-03-01 Nick Clifton) and r103151 (2005-08-16
Richard Earnshaw) against GCC 4.0.3. Both these patches apply cleanly (with
offsets), but don't fix the problem compiling newlib.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] [4.0 only] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
` (3 preceding siblings ...)
2006-06-28 23:31 ` sjackman at gmail dot com
@ 2006-06-28 23:32 ` pinskia at gcc dot gnu dot org
2006-06-29 0:10 ` sjackman at gmail dot com
` (2 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2006-06-28 23:32 UTC (permalink / raw)
To: gcc-bugs
------- Comment #5 from pinskia at gcc dot gnu dot org 2006-06-28 23:31 -------
Is this a regression?
--
pinskia at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Keywords| |ice-on-valid-code
Summary|config/arm/arm.c:3140 ICE |[4.0 only]
| |config/arm/arm.c:3140 ICE
Target Milestone|--- |4.0.4
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] [4.0 only] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
` (4 preceding siblings ...)
2006-06-28 23:32 ` [Bug target/28193] [4.0 only] " pinskia at gcc dot gnu dot org
@ 2006-06-29 0:10 ` sjackman at gmail dot com
2006-06-29 0:30 ` sjackman at gmail dot com
2007-02-03 17:48 ` gdr at gcc dot gnu dot org
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-29 0:10 UTC (permalink / raw)
To: gcc-bugs
------- Comment #6 from sjackman at gmail dot com 2006-06-29 00:01 -------
Applying the patch from r103277 fixes this bug. I also had r95736 and r103151
applied against my 4.0.3 tree at the time. So, I don't know if r103277 alone is
sufficient. r103277 was meant to close PR target/23473, so this bug may be a
duplicate of #23473.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] [4.0 only] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
` (5 preceding siblings ...)
2006-06-29 0:10 ` sjackman at gmail dot com
@ 2006-06-29 0:30 ` sjackman at gmail dot com
2007-02-03 17:48 ` gdr at gcc dot gnu dot org
7 siblings, 0 replies; 9+ messages in thread
From: sjackman at gmail dot com @ 2006-06-29 0:30 UTC (permalink / raw)
To: gcc-bugs
------- Comment #7 from sjackman at gmail dot com 2006-06-29 00:10 -------
Subject: Re: [4.0 only] config/arm/arm.c:3140 ICE
On 28 Jun 2006 23:31:22 -0000, pinskia at gcc dot gnu dot org
<gcc-bugzilla@gcc.gnu.org> wrote:
> ------- Comment #5 from pinskia at gcc dot gnu dot org 2006-06-28 23:31 -------
> Is this a regression?
...
> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
I haven't tested any version of GCC earlier than 4.0.3 for this bug,
so I don't know if it's a regression. I was testing svn versions after
4.0.3 and before 4.1.0 to discover when this bug was fixed. It appears
it was fixed in r103277.
Cheers,
Shaun
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread
* [Bug target/28193] [4.0 only] config/arm/arm.c:3140 ICE
2006-06-28 16:23 [Bug c/28193] New: config/arm/arm.c:3140 ICE sjackman at gmail dot com
` (6 preceding siblings ...)
2006-06-29 0:30 ` sjackman at gmail dot com
@ 2007-02-03 17:48 ` gdr at gcc dot gnu dot org
7 siblings, 0 replies; 9+ messages in thread
From: gdr at gcc dot gnu dot org @ 2007-02-03 17:48 UTC (permalink / raw)
To: gcc-bugs
------- Comment #8 from gdr at gcc dot gnu dot org 2007-02-03 17:47 -------
Fixed in GCC-4.1.1
--
gdr at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |FIXED
Target Milestone|4.0.4 |4.1.1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28193
^ permalink raw reply [flat|nested] 9+ messages in thread