Hi, all,
With following code:
[CODE]
struct B {
int c;
int d;
};
#define X(a, b, c) \
do\
{\
if (a)\
printf(%d, %d\n, b.c, c);\
else\
printf(%d\n, c);\
}while(0);
[/CODE]
Why
int d = 24;
X(1, b, d);
can be compiled
On 2/6/07, Diego Novillo [EMAIL PROTECTED] wrote:
Paulo J. Matos wrote on 02/06/07 14:19:
Why before pass_build_ssa? (version 4.1.1)
It depends on the properties your pass requires. If you ask for
PROP_cfg and PROP_gimple_any then you should schedule it after the CFG
has been built, but if
[EMAIL PROTECTED] wrote:
With following code:
[CODE]
struct B {
int c;
int d;
};
#define X(a, b, c) \
do\
{\
if (a)\
printf(%d, %d\n, b.c, c);\
else\
printf(%d\n, c);\
}while(0);
[/CODE]
Why
int d = 24;
X(1, b, d);
Jan Hubicka [EMAIL PROTECTED] writes:
[...]
static inline void __attribute__((noreturn)) BUG(void)
{
__asm__ __volatile__(trap);
__builtin_unreached();
This is bit dificult to do in general since it introduces new kind of
control flow construct. It would be better to express such
Hello,
I'm planning to do some research on the optimization flags available
for GCC (currently, using 4.1.1). More in particular, we want to see
how we can come up with a set of combinations of flags which allow a
tradeoff between compilation time, execution time and code size (as
with
Kenneth Hoste wrote on 02/07/07 08:56:
[1] Almagor et al., Finding effective compilation sequences (LCES'04)
[2] Cooper et al., Optimizing for Reduced Code Space using Genetic
Algorithms (LCTES'99)
[3] Almagor et al., Compilation Order Matters: Exploring the
Structure of the Space of
Hi!
I create test to reproduce issue with cpu2006/454.calculix
See attached. File e_c3d.f contains cutted subroutine from calculix.
tr535.f main entry point of the test. you can use go-script as a
reference how i get these results. find_stall.pl script which find
problem instruction combinations.
There appears to be a bug in value-prof.c:visit_hist rev 121554.
This function always returns 0, which causes htab_traverse to exit
early. This means that only the first histogram that appears in cfun-
value_histograms-entries is ever checked, so verify_histograms
will only indicate an
This patch bootstraps and passes make check on x86_64.
Please do not cross-post. Patches should go to gcc-patches@ only.
--
Eric Botcazou
I am new to this list, so please excuse any obvious mistakes. I am
trying to check if two types are equal or one is derived from the
other within the compiler. One of the types is a struct that is
defined under the std namescope. How do I search for a node that is
a TYPE_DECL of the structure
Hi Ian,
sorry to bother again. I reduced the code (attached) that segfaults here
on Core 2 Duo [1]. If I add -fno-split-wide-types the code does not segfault.
That flag comes from your patchset [2].
execute:
# ./cc1 -quiet -m64 -O1 test.c -o test.o
Any ideas?
Regards,
Hanno
[1]
...
Ah, right... I wonder if we can keep the loop structure in place, even
after completely unrolling the loop - I mean the 'struct loop' in
'current_loops' (not the actual CFG), so that the SLP in loops would
have
a chance to at least consider vectorizing this loop.
Having a loop structure
Hanno Meyer-Thurow [EMAIL PROTECTED] writes:
Hi Ian,
sorry to bother again. I reduced the code (attached) that segfaults here
on Core 2 Duo [1]. If I add -fno-split-wide-types the code does not segfault.
That flag comes from your patchset [2].
execute:
# ./cc1 -quiet -m64 -O1 test.c -o
On 07 Feb 2007 13:46:43 -0800
Ian Lance Taylor [EMAIL PROTECTED] wrote:
* lower-subreg.c (resolve_clobber): Handle a subreg of a concatn.
Yes, that is there. I have revision 121690.
Hanno
Snapshot gcc-4.2-20070207 is now available on
ftp://gcc.gnu.org/pub/gcc/snapshots/4.2-20070207/
and on various mirrors, see http://gcc.gnu.org/mirrors.html for details.
This snapshot has been generated from the GCC 4.2 SVN branch
with the following options: svn://gcc.gnu.org/svn/gcc/branches
On Feb 7, 2007, at 1:05 PM, Prabhanjan Kambadur wrote:
I am new to this list, so please excuse any obvious mistakes. I am
trying to check if two types are equal or one is derived from the
other within the compiler. One of the types is a struct that is
defined under the std namescope. How do I
Yup, what you answered is indeed what I want
Thanks,
Anju
Hanno Meyer-Thurow [EMAIL PROTECTED] writes:
sorry to bother again. I reduced the code (attached) that segfaults here
on Core 2 Duo [1]. If I add -fno-split-wide-types the code does not segfault.
That flag comes from your patchset [2].
execute:
# ./cc1 -quiet -m64 -O1 test.c -o test.o
Hi,
I am working on gcc 4.1.1 and Itanium architecture. I want to modify
the machine description of ia64.md to add some checks before each ld
instruction. the
following is the original define_insn:
(define_insn *movqi_internal
[(set (match_operand:QI 0 destination_operand
--- Comment #4 from fxcoudert at gcc dot gnu dot org 2007-02-07 08:09
---
(In reply to comment #3)
I'm not yet sure how to deal with matmul, wether by
converting its arguments or by creating kind=1 and
kind=2 versions.
I think converting wil have a huge performance hit, so we'd
--- Comment #3 from fxcoudert at gcc dot gnu dot org 2007-02-07 08:11
---
Brooks, maybe it's time to backport them to 4.2?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30420
--- Comment #7 from fxcoudert at gcc dot gnu dot org 2007-02-07 08:14
---
Time for a 4.2 backport?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30389
--- Comment #2 from fxcoudert at gcc dot gnu dot org 2007-02-07 09:08
---
Created an attachment (id=13019)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13019action=view)
Patch implementing the -fbacktrace option
Here's an updated version of the patch I submitted some time ago. I
--- Comment #12 from brooks at gcc dot gnu dot org 2007-02-07 09:10 ---
(In reply to comment #7)
If I read the F2003 standrad correctly, then
your program conforms to F2003. You may want
to change this to an enhancement request.
This is incorrect -- the code does not conform to
--- Comment #13 from dominiq at lps dot ens dot fr 2007-02-07 09:24 ---
Subject: Re: recursive I/O hangs under OSX 10.3
Section 9.5.3.7.1, paragarph 2, defines a child I/O statement as one that's
occuring within a user-defined derived-type I/O function -- which is
definitely
not
--- Comment #3 from burnus at gcc dot gnu dot org 2007-02-07 09:47 ---
Patch implementing the -fbacktrace option
I think one should add also some userhandler
signal(SIGSEGV, my_segv_handler);
which calls show_backtrace and exits/coredumps then.
That way we calso get a backtrace for
--- Comment #2 from rguenth at gcc dot gnu dot org 2007-02-07 09:55 ---
note the size of class cls grows exponentially with its template parameter.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #1 from rguenth at gcc dot gnu dot org 2007-02-07 09:58 ---
Confirmed. Note we already NULLify the pointer in the caller for
_gfortran_deallocate (but I missed to fix the comment before that function as
well).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30723
--- Comment #10 from hp at gcc dot gnu dot org 2007-02-07 10:09 ---
Subject: Bug 28870
Author: hp
Date: Wed Feb 7 10:09:41 2007
New Revision: 121684
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=121684
Log:
PR testsuite/28870
*
--- Comment #11 from hp at gcc dot gnu dot org 2007-02-07 10:12 ---
Subject: Bug 28870
Author: hp
Date: Wed Feb 7 10:12:21 2007
New Revision: 121686
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=121686
Log:
PR testsuite/28870
*
--- Comment #12 from hp at gcc dot gnu dot org 2007-02-07 10:16 ---
The checked-in changes marked with this PR don't solve the timeout issue by
far,
but hint of a way to solve timeout problems for specific tests, for specific
(classes of) systems, in another way than changing the
--- Comment #3 from jakub at gcc dot gnu dot org 2007-02-07 12:16 ---
Subject: Bug 30703
Author: jakub
Date: Wed Feb 7 12:16:22 2007
New Revision: 121688
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=121688
Log:
PR c++/30703
* gimplify.c
--- Comment #4 from jakub at gcc dot gnu dot org 2007-02-07 13:35 ---
Subject: Bug 28468
Author: jakub
Date: Wed Feb 7 13:35:17 2007
New Revision: 121689
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=121689
Log:
2007-02-07 Bruno Haible [EMAIL PROTECTED]
config/
PR
--- Comment #4 from jakub at gcc dot gnu dot org 2007-02-07 13:37 ---
Subject: Bug 30703
Author: jakub
Date: Wed Feb 7 13:37:29 2007
New Revision: 121690
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=121690
Log:
PR c++/30703
* gimplify.c
--- Comment #5 from jakub at gcc dot gnu dot org 2007-02-07 13:39 ---
Approved offline by Alex Oliva, committed so far on the trunk.
Will backport to 4.2 branch after a week or so on the trunk.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28468
--- Comment #10 from jakub at gcc dot gnu dot org 2007-02-07 13:42 ---
Should be fixed on the trunk and 4.2 branch now, thanks Roger.
--
jakub at gcc dot gnu dot org changed:
What|Removed |Added
Got the following error message when compiling gcc itself:
---
TARGET_CPU_DEFAULT= \
HEADERS=auto-host.h ansidecl.h config/i386/xm-mingw32.h DEFINES= \
/bin/sh ../../gcc-4.2-20070131/gcc/mkconfig.sh config.h
TARGET_CPU_DEFAULT= \
--- Comment #4 from tvb at gnome dot org 2007-02-07 14:59 ---
(In reply to comment #3)
What is there to warn about?
If you do -Winitialize without -O, you get a warning so ...
I just tried the explicit -Wuninitialized, very good,
and I noticed at least its documented in the man
--- Comment #5 from pinskia at gcc dot gnu dot org 2007-02-07 16:39 ---
To me a warning should not be emitted while using -Wall -O0 as it will break
all the programs which use -Werror which is a lot of them. Also this is
already documented, if people don't read documentation, why would
--- Comment #12 from hhinnant at apple dot com 2007-02-07 16:46 ---
At the ad-hoc LWG meeting in Batavia, Jan 22-24, 2007, the LWG decided that
self referencing code in list::remove must work. Here is a preview of the
issue which is currently set to become official at the Spring '07
--- Comment #6 from tvb at gnome dot org 2007-02-07 16:48 ---
(In reply to comment #5)
To me a warning should not be emitted while using -Wall -O0 as it will break
all the programs which use -Werror which is a lot of them. Also this is
already documented, if people don't read
--- Comment #13 from hhinnant at apple dot com 2007-02-07 16:59 ---
(In reply to comment #12)
If we have a utility similar to boost::address_of, that might be better than
using operator to get the address of the value_type (to accommodate types
which overload operator).
Oops, I
--- Comment #14 from chris at bubblescope dot net 2007-02-07 17:12 ---
Unless __value comes from the list, does the standard require
__a.address(__value) to work?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17012
--- Comment #15 from hhinnant at apple dot com 2007-02-07 17:22 ---
(In reply to comment #14)
Unless __value comes from the list, does the standard require
__a.address(__value) to work?
That's a good question Chris. The way I read the current draft, I believe the
answer is no.
If you configure to build a 32-bit compiler on a 64-bit Linux system with:
CC='gcc -m32' /src/trunk/configure --{target,host,build}=i686-pc-linux-gnu ...
the compiler fails because it defaults to 32-bit code but the standard
assembler is 64 bit, and it fails in building libgcc. If you are
--- Comment #5 from tkoenig at gcc dot gnu dot org 2007-02-07 17:52 ---
Hi FX,
do you remember why always performing that check (ie, turn function to be
always true) is not the right thing to do?
When working on this, I hit numerous testsuite regressions
when always checking for
--- Comment #8 from tkoenig at gcc dot gnu dot org 2007-02-07 17:54 ---
The status of the other patch is: Waiting for review.
http://gcc.gnu.org/ml/gcc-patches/2007-02/msg00260.html
I can't do any regression-testing right now, because PR 30678 :-(
--
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-02-07 18:05 ---
It might be better to do something like powerpc64-linux-gnu does and define a
--with-cpu=default32 which compiles a 64bit capable compiler but defaults to
32bits.
--
pinskia at gcc dot gnu dot org changed:
--- Comment #6 from fxcoudert at gcc dot gnu dot org 2007-02-07 19:10
---
(In reply to comment #5)
If you find something that works without that argument (which is
a bit of a kudge), so much the better.
The patch I attached removes this argument, and it gives no regression on the
--- Comment #2 from fxcoudert at gcc dot gnu dot org 2007-02-07 19:22
---
Created an attachment (id=13020)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=13020action=view)
Patch to not generate calls to internal_free any more
--
--- Comment #14 from dominiq at lps dot ens dot fr 2007-02-07 19:55 ---
The test is known to fail on OSX 10.3 (gcc 4.3.0 20070202) and 10.4 (PPC with
gcc 4.2.0 20070124
and MacIntel gcc unknown).
When I have filled the PR I have forgotten to say that the same bug was present
in
--- Comment #12 from tkoenig at gcc dot gnu dot org 2007-02-07 20:05
---
According to http://gcc.gnu.org/ml/gcc/2007-02/msg00067.html,
this was caused by
2007-02-03 Bruce Korb [EMAIL PROTECTED]
* inclhack.def (glibc_c99_inline_4): replace extern only if
surrounded by
--- Comment #5 from tkoenig at gcc dot gnu dot org 2007-02-07 20:08 ---
I think converting wil have a huge performance hit, so we'd better havec
kind=1
and kind=2 versions.
I agree, here's a patch to do this.
For speed reasons, we should also reverse the conversions
through the
#include stdarg.h
int f(int t, ...)
{
va_list a;
va_start (a, t);
va_arg(a, int);
int t1 = va_arg(a, int);
va_end(a);
return t1;
}
-
We get a warning on the line which just contains va_arg(a, int);
Even though the value is not used, a is still incremented so the result is not
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|--- |4.1.2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30729
$ cat warning.c
#if 0
gcc-4.1.1 -c -Os warning.c -Wunsafe-loop-optimizations
exit 0
#endif
void foo(unsigned int n)
{
while (n 10)
n -= 2;
}
^D
$ sh warning.c
warning.c: In function 'foo':
warning.c:8: warning: cannot optimize loop, the loop counter may overflow
--- Comment #7 from tkoenig at gcc dot gnu dot org 2007-02-07 20:49 ---
(In reply to comment #6)
The patch I attached removes this argument, and it gives no regression on the
testsuite. I also simplified the conditional expression by using a COND_EXPR
instead of generating different
--- Comment #13 from bkorb at gnu dot org 2007-02-07 21:02 ---
The problem was supposed to have been fixed:
http://gcc.gnu.org/ml/gcc-cvs/2007-02/msg00194.html
http://gcc.gnu.org/ml/gcc-cvs/2007-02/msg00172.html
If you are still having problems, please show the nature of the
--- Comment #3 from patchapp at dberlin dot org 2007-02-07 21:40 ---
Subject: Bug number PR target/29599
A patch for this bug has been added to the patch tracker.
The mailing list url for the patch is
http://gcc.gnu.org/ml/gcc-patches/2007-02/msg00646.html
--
--- Comment #3 from skottmckay at gmail dot com 2007-02-07 21:58 ---
Doesn't it need to be copy constructable for the rebinding to work for
_Vector_base::_Vector_impl::_Tp_alloc_type?
I agree that making it mt-safe doesn't quite fit with its intended usage.
Possibly the
--- Comment #8 from fxcoudert at gcc dot gnu dot org 2007-02-07 22:31
---
Reduced testcase:
enum, bind (c)
integer :: x
enumerator blue
end enum
end
It fails under valgrind for both i686-linux and x86_64-linux:
==5651== Invalid read of size 4
==5651==at 0x9DBD13:
/home/apinski/src/fsf/gcc/gcc/libobjc/exception.c: In function
'parse_lsda_header':
/home/apinski/src/fsf/gcc/gcc/libobjc/exception.c:94: warning: passing argument
2 of 'read_uleb128' from incompatible pointer type
/home/apinski/src/fsf/gcc/gcc/libobjc/exception.c:103: warning: passing
argument 2
--- Comment #1 from pinskia at gcc dot gnu dot org 2007-02-07 22:50 ---
Mine for both being the spu maintainer and the libobjc maintainer.
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|4.1.3 |4.1.2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28778
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|4.1.3 |4.1.2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25156
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|4.1.3 |---
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28125
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
Target Milestone|4.1.3 |4.1.2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26573
--- Comment #5 from dnovillo at gcc dot gnu dot org 2007-02-07 23:33
---
I cannot reproduce this bug with mainline as of 2007-02-06. The bug is still
latent though, so I will commit a variant of this patch to fix it.
Essentially, we should leave every TREE_ADDRESSABLE variable alone
--- Comment #9 from tobi at gcc dot gnu dot org 2007-02-08 00:44 ---
I reviewed the patch which introduced ENUM support, will take a look at this
tomorrow.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30478
--- Comment #15 from jvdelisle at gcc dot gnu dot org 2007-02-08 03:12
---
Try :
external fun
real fun
character(10) :: string
real a
a = fun()
print *, a,a
write(string,'(f10.6)') fun()
print *, string
end
real function fun()
print *, 'test'
fun = 1.0
end
Or increase the size of
--- Comment #2 from burnus at gcc dot gnu dot org 2007-02-08 07:32 ---
Seemingly fixed
--
burnus at gcc dot gnu dot org changed:
What|Removed |Added
71 matches
Mail list logo
