[Bug c++/29883] i686-pc-cygwin/3.4.4 - internal compiler error: Segmentation fault
--- Comment #2 from mtrojanovich at doubleclick dot net 2006-11-18 06:31 --- My first error is in how I'm compiling the library file... Since that in itself is faulted I'm dropping severity. It's user error, but it is also user error that causes an internal compiler error. -- mtrojanovich at doubleclick dot net changed: What|Removed |Added Severity|normal |trivial http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29883
[Bug c/29884] gcc-4.1.1, gcc-4.0.1 generate segfaulting SSE code
--- Comment #3 from sergstesh at yahoo dot com 2006-11-18 06:18 --- Created an attachment (id=12639) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12639&action=view) .i source file not causing segfault This is .i source file with line #157..167 commented out, and there is no segfault in this case. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29884
[Bug c/29884] gcc-4.1.1, gcc-4.0.1 generate segfaulting SSE code
--- Comment #2 from sergstesh at yahoo dot com 2006-11-18 06:15 --- Created an attachment (id=12638) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12638&action=view) .i source causing segfault This is the .i file which causes segfault. Line #157..167 are not commented out. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29884
[Bug c/29884] gcc-4.1.1, gcc-4.0.1 generate segfaulting SSE code
--- Comment #1 from sergstesh at yahoo dot com 2006-11-18 06:10 --- Created an attachment (id=12637) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12637&action=view) the "C" file causing the failure If I run this file, it fails with: " checkpoint 1 &signal_spectrum_L[0]=80491e0 &signal_spectrum_R[0]=8049140 checkpoint 2 half_number_of_samples=16 bin_number=1 bin_number=1 &signal_spectrum_L[1]=80491e4 checkpoint 3 Segmentation fault ". Which means that this: 102 v_r_signal_L.v = *(vFloat *)&signal_spectrum_L[bin_number]; is the offending line. Please pay attention to these lines: 157 *(vFloat *)&signal_spectrum_L[bin_number] = v_r_signal_L.v; 158 *(vFloat *)&signal_spectrum_R[bin_number] = v_r_signal_R.v; 159 160 signal_spectrum_L[number_of_samples_minus_bin_number] = v_i_signal_L.f[0]; 161 signal_spectrum_R[number_of_samples_minus_bin_number] = v_i_signal_R.f[0]; 162 163 signal_spectrum_L[number_of_samples_minus_bin_number_minus_1] = v_i_signal_L.f[1]; 164 signal_spectrum_R[number_of_samples_minus_bin_number_minus_1] = v_i_signal_R.f[1]; 165 166 signal_spectrum_L[number_of_samples_minus_bin_number_minus_3] = v_i_signal_L.f[3]; 167 signal_spectrum_R[number_of_samples_minus_bin_number_minus_3] = v_i_signal_R.f[3]; - they are well below the offending line #102 in the same loop body, so at first site they shouldn't matter - segfaults occurs before them. Well, it's not the case, they do matter, if they are commented out, the segfault does not occur, the program completes normally. The above line #157..167 are the missing functional piece I discovered while debugging, it's writeback. I'll upload the .i files, as required, so you'll be able to compile the files and hopefully reproduce the bug. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29884
[Bug c++/29883] i686-pc-cygwin/3.4.4 - internal compiler error: Segmentation fault
--- Comment #1 from mtrojanovich at doubleclick dot net 2006-11-18 06:02 --- Created an attachment (id=12636) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12636&action=view) Employee base class that caused the error. No other files related. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29883
[Bug c/29884] New: gcc-4.1.1, gcc-4.0.1 generate segfaulting SSE code
Hello, this bug is most likely related to http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29818 , but now it's gcc-4.1.1, gcc-4.0.1, not gcc-3.4.6. Actually, it's the same code using which I discovered http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29818 , just this is the full version of the code, not the shortened one. This bug was discovered while I was debugging the code and added a missing functional part. -- Summary: gcc-4.1.1, gcc-4.0.1 generate segfaulting SSE code Product: gcc Version: 4.1.1 Status: UNCONFIRMED Severity: blocker Priority: P3 Component: c AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: sergstesh at yahoo dot com GCC build triplet: Linux comp.home.net 2.6.12-27mdk-i686-up-4GB #1 Tue Sep 26 12:41 GCC host triplet: Linux comp.home.net 2.6.12-27mdk-i686-up-4GB #1 Tue Sep 26 12:41 GCC target triplet: Linux comp.home.net 2.6.12-27mdk-i686-up-4GB #1 Tue Sep 26 12:41 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29884
[Bug c++/29883] New: i686-pc-cygwin/3.4.4 - internal compiler error: Segmentation fault
29258 looks similar but is unconfirmed. Perhaps this will be straightforward to confirm. COMMAND THAT GENERATED THE ERROR. COMPILING WITH... $ g++ -Wall -o Employee.h Employee.h -mno-cygwin Employee.h:169: internal compiler error: Segmentation fault Please submit a full bug report, with preprocessed source if appropriate. See http://gcc.gnu.org/bugs.html> for instructions. CODE THAT GENERATED THE ERROR... (attached) MY g++ VERSION AND INSTALLATION DETAILS... $ g++ -v Reading specs from /usr/lib/gcc/i686-pc-cygwin/3.4.4/specs Configured with: /gcc/gcc-3.4.4/gcc-3.4.4-1/configure --verbose --prefix=/usr --exec-pre fix=/usr --sysconfdir=/etc --libdir=/usr/lib --libexecdir=/usr/lib --mandir=/usr/share/m an --infodir=/usr/share/info --enable-languages=c,ada,c++,d,f77,java,objc --enable-nls - -without-included-gettext --enable-version-specific-runtime-libs --without-x --enable-li bgcj --disable-java-awt --with-system-zlib --enable-interpreter --disable-libgcj-debug - -enable-threads=posix --enable-java-gc=boehm --disable-win32-registry --enable-sjlj-exce ptions --enable-hash-synchronization --enable-libstdcxx-debug : (reconfigured) Thread model: posix gcc version 3.4.4 (cygming special) (gdc 0.12, using dmd 0.125) MY ENVIRONMENT... Time of this report: 11/17/2006, 22:44:19 Machine name: x Operating System: Windows XP Professional (5.1, Build 2600) Service Pack 2 (2600.xpsp_sp2_gdr.050301-1519) Language: English (Regional Setting: English) System Manufacturer: IBM System Model: 2373KRU BIOS: Phoenix FirstBIOS(tm) Notebook Pro Version 2.0 for IBM ThinkPad Processor: Intel(R) Pentium(R) M processor 1.70GHz Memory: 1534MB RAM Page File: 436MB used, 2997MB available Windows Dir: C:\WINDOWS DirectX Version: DirectX 9.0c (4.09..0904) DX Setup Parameters: Not found DxDiag Version: 5.03.2600.2180 32bit Unicode -- Summary: i686-pc-cygwin/3.4.4 - internal compiler error: Segmentation fault Product: gcc Version: 3.4.4 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: mtrojanovich at doubleclick dot net http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29883
[Bug tree-optimization/29877] out of SSA (TER) extends variable life of variable beyond original declared
--- Comment #3 from shinwell at gcc dot gnu dot org 2006-11-18 04:57 --- Created an attachment (id=12635) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12635&action=view) Proposed patch 2006-11-17 Mark Shinwell <[EMAIL PROTECTED]> gcc/ * tree-outof-ssa.c (check_replaceable): Deem assignments with register variables on their RHS to not be replaceable. gcc/testsuite/ * gcc.target/arm/register-variables.c: New test. * gcc.dg/pr16194.c: Remove dg-error directive. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29877
[Bug libgcj/29151] [win32] Runtime.exec(String[] cmdarray, String[] envp) -> envp doesn't work
-- membar at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |membar at gcc dot gnu dot |dot org |org Status|UNCONFIRMED |ASSIGNED Component|java|libgcj Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2006-11-18 02:34:02 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29151
[Bug fortran/29882] gfortran testsuite doesn't recognize warnings and errors
--- Comment #3 from pinskia at gcc dot gnu dot org 2006-11-18 02:26 --- (In reply to comment #2) > There might be a bug about GCC testsuite also. And there is PR 25241. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added BugsThisDependsOn||25241 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29882
[Bug tree-optimization/9814] gcc fails to optimise if (l&2) l|=2 away
--- Comment #9 from pinskia at gcc dot gnu dot org 2006-11-18 01:34 --- (In reply to comment #8) > Shouldn't this be fixed by Roger Sayle's recent fold-const.c patch? No, in fact the generic (the one where 2 is turned into a variable) is not optimized either. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added Status|WAITING |NEW Last reconfirmed|2006-09-17 09:19:25 |2006-11-18 01:34:36 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9814
[Bug tree-optimization/9814] gcc fails to optimise if (l&2) l|=2 away
--- Comment #8 from steven at gcc dot gnu dot org 2006-11-18 01:27 --- Shouldn't this be fixed by Roger Sayle's recent fold-const.c patch? -- steven at gcc dot gnu dot org changed: What|Removed |Added CC||sayle at gcc dot gnu dot org Status|NEW |WAITING http://gcc.gnu.org/bugzilla/show_bug.cgi?id=9814
[Bug rtl-optimization/29881] union causes inefficient code
--- Comment #1 from pinskia at gcc dot gnu dot org 2006-11-18 00:14 ---
The problem here I think is unions.
Here is how I would have written this code (without using unions in fact):
void array_sample_fun(__m128 *dst, const __m128 *src, int length) {
__m128 af = _mm_set1_ps(1.20f);
__m128 bf = _mm_set1_ps(2.88f);
__m128 cf = _mm_set1_ps(-2.44f);
__m128 df = _mm_set1_ps(4.06f);
__m128 ef = _mm_set1_ps(-12.04f);
__m128i mask = _mm_set1_epi32(0xff << 23);
__m128i bias = _mm_set1_epi32(0x7f << 23);
__m128i t;
while (length-- != 0) {
__m128 vec;
vec = (*src++);
__m128 arg =
_mm_cvtepi32_ps(_mm_srai_epi32(_mm_sub_epi32(_mm_and_si128((__m128i)vec, mask),
bias),
23));
vec = (__m128)_mm_or_si128(_mm_andnot_si128(mask,
(__m128i)vec), bias);
*dst++ = _mm_add_ps(arg,
_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(
_mm_mul_ps(af, vec), bf), vec), cf), vec), df),
vec), ef));
}
}
--
The above gives good results for 32bit:
.L4:
movaps (%eax), %xmm0
movdqa %xmm4, %xmm1
addl$1, %ecx
addl$16, %eax
movdqa %xmm0, %xmm2
pandn %xmm0, %xmm1
movaps .LC0, %xmm0
por %xmm3, %xmm1
pand%xmm4, %xmm2
psubd %xmm3, %xmm2
mulps %xmm1, %xmm0
psrad $23, %xmm2
cvtdq2ps%xmm2, %xmm2
addps .LC1, %xmm0
mulps %xmm1, %xmm0
addps %xmm7, %xmm0
mulps %xmm1, %xmm0
addps %xmm6, %xmm0
mulps %xmm1, %xmm0
addps %xmm5, %xmm0
addps %xmm0, %xmm2
movaps %xmm2, (%edx)
addl$16, %edx
cmpl%ebx, %ecx
jne .L4
While your orginal example on 32bits has a store to the stack.
So this is just a case where union cause missed optimization
--
pinskia at gcc dot gnu dot org changed:
What|Removed |Added
GCC build triplet| x86_64-unknown-linux-gnu |x86_64-unknown-linux-gnu
GCC host triplet| x86_64-unknown-linux-gnu |x86_64-unknown-linux-gnu
GCC target triplet| x86_64-unknown-linux-gnu |x86_64-unknown-linux-gnu
Summary|inefficient/incorrect xmm |union causes inefficient
|registers usage |code
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29881
[Bug fortran/29882] gfortran testsuite doesn't recognize warnings and errors
--- Comment #2 from pinskia at gcc dot gnu dot org 2006-11-18 00:04 --- I think this is a known issue. There might be a bug about GCC testsuite also. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29882
[Bug fortran/29882] gfortran testsuite doesn't recognize warnings and errors
--- Comment #1 from jvdelisle at gcc dot gnu dot org 2006-11-17 23:45 --- Are you saying that it matches the message text emitted and thinks its OK, but does not recognize that it was an error vs a warning? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29882
[Bug fortran/29882] New: gfortran testsuite doesn't recognize warnings and errors
gfortran.dg/io_constraints_2.f90 issues an error on line 33 instead of a warning, but it still PASSES. -- Summary: gfortran testsuite doesn't recognize warnings and errors Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: fortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: fxcoudert at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29882
[Bug rtl-optimization/29881] New: inefficient/incorrect xmm registers usage
hello!
I'm writing fast approximations of some complex math functions. I'm trying do
them as low-level as possible, far from hand-written assembly though. I often
check the assembly code to verify the quality of it. During one of the checkups
i found that gcc (i checked 3.4.5/win32 and 4.0/4.1/fc5) doesn't make a good
use of the xmm registers when writing a mixed SSE/SSE2 code.
example code:
#include
typedef union {
__m128i i;
__m128 f;
} __vec128;
void array_sample_fun(__m128 *dst, const __m128 *src, int length) {
__m128 af = _mm_set1_ps(1.20f);
__m128 bf = _mm_set1_ps(2.88f);
__m128 cf = _mm_set1_ps(-2.44f);
__m128 df = _mm_set1_ps(4.06f);
__m128 ef = _mm_set1_ps(-12.04f);
__m128i mask = _mm_set1_epi32(0xff << 23);
__m128i bias = _mm_set1_epi32(0x7f << 23);
while (length-- != 0) {
__vec128 vec;
vec.f = *src++;
__m128 arg =
_mm_cvtepi32_ps(_mm_srai_epi32(_mm_sub_epi32(_mm_and_si128(vec.i, mask), bias),
23));
vec.i = _mm_or_si128(_mm_andnot_si128(mask, vec.i), bias);
*dst++ = _mm_add_ps(arg,
_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(_mm_mul_ps(_mm_add_ps(
_mm_mul_ps(af, vec.f), bf), vec.f), cf), vec.f), df),
vec.f), ef));
}
}
the main loop of the function looks like this:
.L4:
movaps (%rsi), %xmm0
addl$1, %eax
movdqa %xmm4, %xmm2
addq$16, %rsi
movaps %xmm0, -24(%rsp)
movdqa -24(%rsp), %xmm0
pandn %xmm0, %xmm2
movdqa %xmm0, %xmm1
movdqa %xmm2, %xmm0
pand%xmm4, %xmm1
por %xmm3, %xmm0
psubd %xmm3, %xmm1
psrad $23, %xmm1
cvtdq2ps%xmm1, %xmm1
movdqa %xmm0, -24(%rsp)
movaps %xmm9, %xmm0
movaps -24(%rsp), %xmm2
mulps %xmm2, %xmm0
addps %xmm8, %xmm0
mulps %xmm2, %xmm0
addps %xmm7, %xmm0
mulps %xmm2, %xmm0
addps %xmm6, %xmm0
mulps %xmm2, %xmm0
addps %xmm5, %xmm0
addps %xmm0, %xmm1
movaps %xmm1, (%rdi)
addq$16, %rdi
cmpl%edx, %eax
jne .L4
As you can see whenever i try to access a float vector from integer SSE2 unit
the compiler saves the register on the stack using movaps just to load it back
to same register using movdqa one instruction later - which is totally
unnecessary/inefficient! Same goes for accessing an integer value by SSE unit.
Is this behavior 'patchable'?
--
Summary: inefficient/incorrect xmm registers usage
Product: gcc
Version: 4.1.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: rtl-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: spd at poczta dot fm
GCC build triplet: x86_64-unknown-linux-gnu
GCC host triplet: x86_64-unknown-linux-gnu
GCC target triplet: x86_64-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29881
[Bug rtl-optimization/29879] [4.3 Regression] ICE: verify_flow_info: loop_father but no loops
-- pinskia at gcc dot gnu dot org changed: What|Removed |Added Summary|[Regression] ICE: |[4.3 Regression] ICE: |verify_flow_info: |verify_flow_info: |loop_father but no loops|loop_father but no loops Target Milestone|--- |4.3.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29879
[Bug driver/27845] Can't build 64-bit 4.2.0 with 4.1.1 on darwin-ppc
--- Comment #3 from lucier at math dot purdue dot edu 2006-11-17 22:00 --- Subject: Re: Can't build 64-bit 4.2.0 with 4.1.1 on darwin-ppc This problem no longer occurs. (This is not the correct way to build a 64-bit 4.2.0, either, but this specific problem no longer occurs.) Thanks. Brad -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27845
[Bug tree-optimization/29879] [Regression] ICE: verify_flow_info: loop_father but no loops
--- Comment #2 from fxcoudert at gcc dot gnu dot org 2006-11-17 21:15 --- *** Bug 29880 has been marked as a duplicate of this bug. *** -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added CC||3dw4rd at verizon dot net http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29879
[Bug libfortran/29880] ICE: Failed bootstrap on libgfortran.
--- Comment #1 from fxcoudert at gcc dot gnu dot org 2006-11-17 21:15 --- *** This bug has been marked as a duplicate of 29879 *** -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||DUPLICATE http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29880
[Bug libfortran/29880] New: ICE: Failed bootstrap on libgfortran.
Configure: ../gcc/configure --prefix=/home/ed/bin-4.3 --with-gmp=/usr/local --with-mpfr=/usr/local --enable-languages=c,c++,fortran This is mainline revision 118945. Here is the last stuff: /home/ed/obj-4.3/./gcc/xgcc -B/home/ed/obj-4.3/./gcc/ -B/home/ed/bin-4.3/x86_64-unknown-linux-gnu/bin/ -B/home/ed/bin-4.3/x86_64-unknown-linux-gnu/lib/ -isystem /home/ed/bin-4.3/x86_64-unknown-linux-gnu/include -isystem /home/ed/bin-4.3/x86_64-unknown-linux-gnu/sys-include -DHAVE_CONFIG_H -I. -I../../../gcc/libgfortran -I. -iquote../../../gcc/libgfortran/io -I../../../gcc/libgfortran/../gcc -I../../../gcc/libgfortran/../gcc/config -I../.././gcc -D_GNU_SOURCE -std=gnu99 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition -Wextra -Wwrite-strings -ftree-vectorize -funroll-loops -O2 -g -O2 -c ../../../gcc/libgfortran/generated/matmul_i4.c -fPIC -DPIC -o .libs/matmul_i4.o ../../../gcc/libgfortran/generated/matmul_i4.c: In function 'matmul_i4': ../../../gcc/libgfortran/generated/matmul_i4.c:337: error: verify_flow_info: Block 158 has loop_father, but there are no loops ../../../gcc/libgfortran/generated/matmul_i4.c:337: error: verify_flow_info: Block 157 has loop_father, but there are no loops ../../../gcc/libgfortran/generated/matmul_i4.c:337: error: verify_flow_info: Block 156 has loop_father, but there are no loops etc. ../../../gcc/libgfortran/generated/matmul_i4.c:337: error: verify_flow_info: Block 2 has loop_father, but there are no loops ../../../gcc/libgfortran/generated/matmul_i4.c:337: internal compiler error: verify_flow_info failed Please submit a full bug report, with preprocessed source if appropriate. See http://gcc.gnu.org/bugs.html> for instructions. make[3]: *** [matmul_i4.lo] Error 1 make[3]: Leaving directory `/home/ed/obj-4.3/x86_64-unknown-linux-gnu/libgfortran' make[2]: *** [all] Error 2 make[2]: Leaving directory `/home/ed/obj-4.3/x86_64-unknown-linux-gnu/libgfortran' make[1]: *** [all-target-libgfortran] Error 2 make[1]: Leaving directory `/home/ed/obj-4.3' make: *** [all] Error 2 -- Summary: ICE: Failed bootstrap on libgfortran. Product: gcc Version: 4.3.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: libfortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: 3dw4rd at verizon dot net GCC build triplet: x86_64-unknown-linux-gnu GCC host triplet: x86_64-unknown-linux-gnu GCC target triplet: x86_64-unknown-linux-gnu http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29880
[Bug fortran/24285] format(1000(a,$))
--- Comment #6 from fxcoudert at gcc dot gnu dot org 2006-11-17 21:09 --- Created an attachment (id=12633) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=12633&action=view) Patch fixing this PR This patch allows $ at other places than the end of the format, and makes gfortran issue a warning in that case. -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |fxcoudert at gcc dot gnu dot |dot org |org Status|NEW |ASSIGNED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24285
[Bug tree-optimization/29879] [Regression] ICE: verify_flow_info: loop_father but no loops
--- Comment #1 from rguenth at gcc dot gnu dot org 2006-11-17 21:07 --- Confirmed. -- rguenth at gcc dot gnu dot org changed: What|Removed |Added CC||rakdver at gcc dot gnu dot ||org Status|UNCONFIRMED |NEW Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2006-11-17 21:07:18 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29879
[Bug tree-optimization/29879] New: [Regression] ICE: verify_flow_info: loop_father but no loops
The following program gives with current gcc SVN 4.3.0 20061117 an ICE when
compiled with "-O2 -funroll-loops":
ice.c: In function 'main':
ice.c:14: error: verify_flow_info: Block 2 has loop_father, but there are no
loops
ice.c:14: internal compiler error: verify_flow_info failed
The same options work with gcc 4.2.0 20061117.
typedef struct mytype
{
int i;
}
mytype;
int main ()
{
int i8;
struct mytype y;
y.i = 2;
i8 = 8;
return 0;
}
--
Summary: [Regression] ICE: verify_flow_info: loop_father but no
loops
Product: gcc
Version: 4.3.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: burnus at gcc dot gnu dot org
GCC target triplet: x86_64-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29879
[Bug other/29878] Clarify gcc jc1 cc1 cc1plus installation problem No such file or directory error
--- Comment #1 from pinskia at gcc dot gnu dot org 2006-11-17 19:55 --- gcc: installation problem, cannot exec 'jc1': Check GNU Java Compiler 'gcj' is installed. Except this is not really a GCC issue because when java is not enabled, you get an error about java not being installed. What the issue is that distros split up GCC into different packages which means it is an installation issue but not a GCC issue. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||INVALID http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29878
[Bug tree-optimization/29877] out of SSA (TER) extends variable life of variable beyond original declared
--- Comment #2 from amacleod at redhat dot com 2006-11-17 19:28 --- A .i file would have been handy. IN any case thats easily worked around. The issue here isn't that the lifetoime of something is being extended since the tree optimizers already made __res function scope. The problem is that it is defined to be a specific register: register long long unsigned int __res __asm__ (*r2); and when its copied to t_10 = __res that is OK as well. t_10 is then used a single time a few stmts away, and this triggers TER to substitute __res at t_10's use location. The simplest and obvious fix I think would be to disable expansion of an expression if the base variable is an __asm__ like that. So it would leave the copy of __res to t_10 and disallow t_10 to be subsituted. I think this will resolve the problem, and I will look at it shortly. -- amacleod at redhat dot com changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |amacleod at redhat dot com |dot org | Status|UNCONFIRMED |ASSIGNED Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2006-11-17 19:28:11 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29877
[Bug rtl-optimization/20070] If-conversion can't match equivalent code, and cross-jumping only works for literal matches
--- Comment #27 from amylaar at gcc dot gnu dot org 2006-11-17 18:58 --- We've been using this patch: http://gcc.gnu.org/ml/gcc-patches/2006-02/msg01488.html successfully in a 4.1.1 based compiler. It is not optimal, though: the check for equality of the regsets in struct_equiv_init is unnecessary, and it can also lead skipping possible optimizations. The frequent live updates are also not necessary. This optimization can work with register life info that is inexact, as long as it is conservative, i.e. it may flag registers as live that are not, but it most not pretend that live registers are dead. Note also that re-integration of the if-conversion improvements was not within the scope of this patch, i.e. this is still an outstanding issue. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20070
[Bug other/29878] New: Clarify gcc jc1 cc1 cc1plus installation problem No such file or directory error
Hello, I'd like to submit a patch to clarify the error message below, as "jc1" would never be a directory. I'll also do the same for cc1 and cc1plus and a Fortran equivalent if there is one. As part of that patch I'd like to make the error message give a clearer indication that the GNU Java Compiler might not be installed, or other backend. So for example the error message could be changed too: gcc: installation problem, cannot exec 'jc1': Check GNU Java Compiler 'gcj' is installed. Would a patch to change this be supported for review? Kind regards Jon $ gcc -o hello hello.java gcc: installation problem, cannot exec 'jc1': No such file or directory [EMAIL PROTECTED]:~$ gcc --version gcc (GCC) 4.0.3 (Ubuntu 4.0.3-1ubuntu5) Copyright (C) 2006 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. $ uname -a Linux now2g 2.6.15-27-k7 #1 SMP PREEMPT Sat Sep 16 02:35:20 UTC 2006 i686 GNU/Linux I've filed this in the "other" component, as there isn't a "general" component. -- Summary: Clarify gcc jc1 cc1 cc1plus installation problem No such file or directory error Product: gcc Version: 4.0.3 Status: UNCONFIRMED Severity: trivial Priority: P3 Component: other AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: jg at jguk dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29878
[Bug tree-optimization/29877] out of SSA (TER) extends variable life of variable beyond original declared
--- Comment #1 from pinskia at gcc dot gnu dot org 2006-11-17 18:31 --- TER causes this ... -- pinskia at gcc dot gnu dot org changed: What|Removed |Added Summary|out of SSA extends variable |out of SSA (TER) extends |life of variable beyond |variable life of variable |original declared |beyond original declared http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29877
[Bug tree-optimization/29877] New: Tree optimization extends variable life of variable beyond original declared
There is a bug on the tree optimization code generated by gcc that produces corruption of variables on some corner cases of code generated by Linux kernel for ARM. I originally found the bug on codesourcery releases but was able to reproduced it on gcc 4.1 that I builded. Detailed explanation and some insights from codesourcery guys can be found at this mail thread: http://www.codesourcery.com/archives/arm-gnu/msg00961.html The example code that triggers the bug is found there too. -- Summary: Tree optimization extends variable life of variable beyond original declared Product: gcc Version: 4.1.1 Status: UNCONFIRMED Severity: major Priority: P3 Component: tree-optimization AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: diego dot dompe at ridgerun dot com GCC build triplet: x86_64-pc-linux-gnu GCC host triplet: x86_64-pc-linux-gnu GCC target triplet: arm-linux-uclibcgnueabi http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29877
[Bug fortran/29876] New: ICE on bad operator in ONLY clause of USE statement
The first error below is expected, but the second one should not be printed out that way :) $ cat translation_tests.f90 module foo end module foo program test use foo, only : bar, operator(.none.) end program test $ gfortran translation_tests.f90 translation_tests.f90:5.17: use foo, only : bar, operator(.none.) 1 Error: Symbol 'bar' referenced at (1) not found in module 'foo' translation_tests.f90:5.39: use foo, only : bar, operator(.none.) 1 Internal Error at (1): gfc_code2string(): Bad code -- Summary: ICE on bad operator in ONLY clause of USE statement Product: gcc Version: 4.3.0 Status: UNCONFIRMED Keywords: ice-on-invalid-code, diagnostic Severity: normal Priority: P3 Component: fortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: fxcoudert at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29876
[Bug fortran/29821] ICE in gfc_typenode_for_spec, at fortran/trans-types.c:666ans-types.c:666
--- Comment #6 from pault at gcc dot gnu dot org 2006-11-17 16:22 --- A patch is on its way - the error is in expr.c(find_array_section); the code presently returns FAILURE if the elements of the array section are not constant. This is not correct for any of the simplify functions, which should return SUCCESS even if simplification has not occurred - FAILURE is reserved for errors. Paul -- pault at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |pault at gcc dot gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2006-11-14 06:31:49 |2006-11-17 16:22:31 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29821
[Bug ada/29802] [4.2 Regression] wrong directory in makefile for ada and libada when srcdir=.
--- Comment #12 from bonzini at gnu dot org 2006-11-17 15:30 --- now fixed on all affected branches. -- bonzini at gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Known to fail|4.0.4 4.2.0 |4.0.4 Known to work|4.1.2 4.3.0 |4.1.2 4.2.0 4.3.0 Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29802
[Bug ada/29802] [4.2 Regression] wrong directory in makefile for ada and libada when srcdir=.
--- Comment #11 from bonzini at gnu dot org 2006-11-17 15:30 --- Subject: Bug 29802 Author: bonzini Date: Fri Nov 17 15:29:54 2006 New Revision: 118941 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118941 Log: 2006-11-16 Paolo Bonzini <[EMAIL PROTECTED]> * Makefile.tpl (clean-target-libgcc): Test for gcc Makefile presence. (unstage): Test for stage_last presence. * Makefile.tpl (clean-stage*): Test separately for package/Makefile and stageN-package/Makefile. PR bootstrap/29802 * Makefile.tpl (POSTSTAGE1_FLAGS_TO_PASS): Add HOST_SUBDIR in STAGE_PREFIX. * Makefile.in: Regenerate. 2006-11-16 Paolo Bonzini <[EMAIL PROTECTED]> PR bootstrap/29802 * configure.ac: Call GCC_TOPLEV_SUBDIRS. * configure: Regenerate. * Makefile.in: Replace host_subdir. Modified: branches/gcc-4_2-branch/ChangeLog branches/gcc-4_2-branch/Makefile.in branches/gcc-4_2-branch/Makefile.tpl branches/gcc-4_2-branch/libada/ChangeLog branches/gcc-4_2-branch/libada/Makefile.in branches/gcc-4_2-branch/libada/configure branches/gcc-4_2-branch/libada/configure.ac -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29802
[Bug ada/27936] [4.2/4.3 regression] gnatbind fails to link
--- Comment #13 from ebotcazou at gcc dot gnu dot org 2006-11-17 15:14 --- Sorry for long delay. -- ebotcazou at gcc dot gnu dot org changed: What|Removed |Added URL||http://gcc.gnu.org/ml/gcc- ||patches/2006- ||11/msg01247.html Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27936
[Bug ada/27936] [4.2/4.3 regression] gnatbind fails to link
--- Comment #12 from ebotcazou at gcc dot gnu dot org 2006-11-17 15:11 --- Subject: Bug 27936 Author: ebotcazou Date: Fri Nov 17 15:11:17 2006 New Revision: 118940 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118940 Log: PR ada/27936 * ada-tree.h (DECL_READONLY_ONCE_ELAB): New macro. * decl.c (elaborate_expression_1): Test the DECL_READONLY_ONCE_ELAB flag in addition to TREE_READONLY to assert the constantness of variables for elaboration purposes. * trans.c (add_decl_expr): Do not dynamically elaborate padded objects if the initializer takes into account the padding. Set DECL_READONLY_ONCE_ELAB flag on variables originally TREE_READONLY but whose elaboration cannot be performed statically. Modified: branches/gcc-4_2-branch/gcc/ada/ChangeLog branches/gcc-4_2-branch/gcc/ada/ada-tree.h branches/gcc-4_2-branch/gcc/ada/decl.c branches/gcc-4_2-branch/gcc/ada/trans.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27936
[Bug fortran/29779] [4.3 Regression] vectorizer fortran testcases failing
--- Comment #8 from pinskia at gcc dot gnu dot org 2006-11-17 15:11 --- (In reply to comment #7) > indeed it does, thanks! > are you going to submit it to mainline? Should be submitted this weekend, today is a party day for me, launch and all :). -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29779
[Bug ada/27936] [4.2/4.3 regression] gnatbind fails to link
--- Comment #11 from ebotcazou at gcc dot gnu dot org 2006-11-17 15:10 --- Subject: Bug 27936 Author: ebotcazou Date: Fri Nov 17 15:10:28 2006 New Revision: 118939 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118939 Log: PR ada/27936 * trans.c (add_decl_expr): Do not dynamically elaborate padded objects if the initializer takes into account the padding. Modified: trunk/gcc/ada/ChangeLog trunk/gcc/ada/trans.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27936
[Bug tree-optimization/29788] [4.1 Regression] ICE in var_ann, at tree-flow-inline.h:130
--- Comment #17 from pinskia at gcc dot gnu dot org 2006-11-17 15:10 --- Fixed. -- pinskia at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29788
[Bug tree-optimization/29788] [4.1 Regression] ICE in var_ann, at tree-flow-inline.h:130
--- Comment #16 from pinskia at gcc dot gnu dot org 2006-11-17 15:09 --- Subject: Bug 29788 Author: pinskia Date: Fri Nov 17 15:09:12 2006 New Revision: 118938 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118938 Log: 2006-11-17 Andrew Pinski <[EMAIL PROTECTED]> PR tree-opt/29788 * fold-const.c (fold_indirect_ref_1): Fold *&CONST_DECL down to what is the const decl is a place holder for. 2006-11-17 Andrew Pinski <[EMAIL PROTECTED]> PR tree-opt/29788 * gfortran.fortran-torture/compile/inline_1.f90: New testcase. Added: branches/gcc-4_1-branch/gcc/testsuite/gfortran.fortran-torture/compile/inline_1.f90 - copied unchanged from r118861, trunk/gcc/testsuite/gfortran.fortran-torture/compile/inline_1.f90 Modified: branches/gcc-4_1-branch/gcc/ChangeLog branches/gcc-4_1-branch/gcc/fold-const.c branches/gcc-4_1-branch/gcc/testsuite/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29788
[Bug fortran/25707] support for Fortran 2003 USE statements, INTRINSIC and NONINTRINSIC
--- Comment #6 from fxcoudert at gcc dot gnu dot org 2006-11-17 11:16 --- Fixed on mainline (rev. 118930) -- fxcoudert at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED Target Milestone|--- |4.3.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25707
[Bug ada/29802] [4.2/4.3 Regression] wrong directory in makefile for ada and libada when srcdir=.
--- Comment #10 from bonzini at gnu dot org 2006-11-17 10:32 --- Subject: Bug 29802 Author: bonzini Date: Fri Nov 17 10:31:47 2006 New Revision: 118927 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118927 Log: 2006-11-16 Paolo Bonzini <[EMAIL PROTECTED]> * Makefile.tpl (clean-target-libgcc): Test for gcc Makefile presence. (unstage): Test for stage_last presence. PR bootstrap/29802 * Makefile.tpl (POSTSTAGE1_FLAGS_TO_PASS): Add HOST_SUBDIR in STAGE_PREFIX. * Makefile.in: Regenerate. libada: 2006-11-16 Paolo Bonzini <[EMAIL PROTECTED]> PR bootstrap/29802 * configure.ac: Call GCC_TOPLEV_SUBDIRS. * configure: Regenerate. * Makefile.in: Replace host_subdir. Modified: trunk/ChangeLog trunk/Makefile.in trunk/Makefile.tpl trunk/libada/ChangeLog trunk/libada/Makefile.in trunk/libada/configure trunk/libada/configure.ac -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29802
[Bug tree-optimization/29801] [4.3 regression] ICE in set_lattice_value (tree-ssa-ccp.c)
--- Comment #7 from rakdver at gcc dot gnu dot org 2006-11-17 10:29 --- Subject: Bug 29801 Author: rakdver Date: Fri Nov 17 10:29:07 2006 New Revision: 118926 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118926 Log: PR tree-optimization/29801 * tree-ssa-ccp.c (get_symbol_constant_value): New function. (get_default_value): Use get_symbol_constant_value. (set_lattice_value): ICE when the value of the constant is changed. (visit_assignment): Ignore VDEFs of read-only variables. * gcc.dg/pr29801.c: New test. Added: trunk/gcc/testsuite/gcc.dg/pr29801.c Modified: trunk/gcc/ChangeLog trunk/gcc/testsuite/ChangeLog trunk/gcc/tree-ssa-ccp.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29801
[Bug middle-end/29584] [4.0/4.1 Regression] internal compiler error on optimization
--- Comment #10 from jakub at gcc dot gnu dot org 2006-11-17 09:01 --- Subject: Bug 29584 Author: jakub Date: Fri Nov 17 09:01:31 2006 New Revision: 118923 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118923 Log: PR middle-end/29584 * tree-ssa-dom.c (simplify_switch_and_lookup_avail_expr): Don't optimize if DEF doesn't have integral type. * gcc.dg/torture/pr29584.c: New test. Added: branches/gcc-4_1-branch/gcc/testsuite/gcc.dg/torture/pr29584.c Modified: branches/gcc-4_1-branch/gcc/ChangeLog branches/gcc-4_1-branch/gcc/testsuite/ChangeLog branches/gcc-4_1-branch/gcc/tree-ssa-dom.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29584
[Bug middle-end/29584] [4.0/4.1 Regression] internal compiler error on optimization
--- Comment #9 from jakub at gcc dot gnu dot org 2006-11-17 08:59 --- Subject: Bug 29584 Author: jakub Date: Fri Nov 17 08:59:08 2006 New Revision: 118922 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118922 Log: PR middle-end/29584 * tree-ssa-forwprop.c (simplify_switch_expr): Don't optimize if DEF doesn't have integral type. * gcc.dg/torture/pr29584.c: New test. Added: branches/gcc-4_2-branch/gcc/testsuite/gcc.dg/torture/pr29584.c Modified: branches/gcc-4_2-branch/gcc/ChangeLog branches/gcc-4_2-branch/gcc/testsuite/ChangeLog branches/gcc-4_2-branch/gcc/tree-ssa-forwprop.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29584
[Bug middle-end/29584] [4.0/4.1 Regression] internal compiler error on optimization
--- Comment #8 from jakub at gcc dot gnu dot org 2006-11-17 08:58 --- Subject: Bug 29584 Author: jakub Date: Fri Nov 17 08:57:45 2006 New Revision: 118921 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=118921 Log: PR middle-end/29584 * tree-ssa-forwprop.c (simplify_switch_expr): Don't optimize if DEF doesn't have integral type. * gcc.dg/torture/pr29584.c: New test. Added: trunk/gcc/testsuite/gcc.dg/torture/pr29584.c Modified: trunk/gcc/ChangeLog trunk/gcc/testsuite/ChangeLog trunk/gcc/tree-ssa-forwprop.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29584
