[Bug c++/26114] [4.2 Regression] g++.dg/init/ctor4.C and g++.old-deja/g++.jason/overload34.C and g++.old-deja/g++.mike/p11110.C fails
--- Comment #9 from janis at gcc dot gnu dot org 2006-04-17 19:36 --- Subject: Bug 26114 Author: janis Date: Mon Apr 17 19:35:56 2006 New Revision: 113010 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=113010 Log: PR c++/26114, c++/26115 * typeck.c (cxx_mark_addressable): Restore check for extra_warnings. * class.c (check_field_decls): Ditto. Modified: trunk/gcc/cp/ChangeLog trunk/gcc/cp/class.c trunk/gcc/cp/typeck.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26114
[Bug target/27187] New: gcc.target/powerpc/doloop-1.c doesn't get desired branch insn
Test gcc.target/powerpc/doloop-1.c has failed on mainline for powerpc64-linux
with -m64 and for powerpc-apple-darwin8.5.0 since this patch was added:
http://gcc.gnu.org/viewcvs?view=rev&rev=112126
r112126 | mkuvyrkov | 2006-03-16 05:20:39 + (Thu, 16 Mar 2006) | 9 lines
2006-03-16 Maxim Kuvyrkov <[EMAIL PROTECTED]>
* sched-rgn.c (extend_rgns): New static function.
(find_rgns): Use it.
(gather_region_statistics, print_region_statistics): New static
functions.
* params.def (PARAM_MAX_SCHED_EXTEND_REGIONS_ITERS): New parameter.
* doc/invoke.texi (max-sched-extend-regions-iters): Document.
The test fails because instruction bdnz is no longer generated. The loop in
queston went from:
.L7:
add 4,4,9
addi 9,9,1
bdnz .L7
to:
.L7:
addic. 0,0,-1
add 4,4,9
addi 9,9,1
bne 0,.L7
--
Summary: gcc.target/powerpc/doloop-1.c doesn't get desired branch
insn
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27187
[Bug tree-optimization/26304] [4.2 Regression] 25_algorithms/prev_permutation/1.cc on powerpc{64,}-linux and powerpc-darwin
--- Comment #16 from janis at gcc dot gnu dot org 2006-04-18 16:23 --- I verified that the failure starts with Jeff's patch: r110705 | law | 2006-02-07 10:31:27 -0800 (Tue, 07 Feb 2006) http://gcc.gnu.org/viewcvs?view=rev&rev=110705 -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26304
[Bug target/11594] testcase gcc.dg/20020103-1.c fails with "scan-assembler-not LC"
--- Comment #5 from janis at gcc dot gnu dot org 2006-04-18 16:52 --- As of mainline 20060417 the test is still failing on powerpc-linux, although it passes for AIX and Darwin. Until December 2005 when Alan fixed the target specifier, the test had not been run on powerpc64-linux with -m32. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11594
[Bug fortran/27229] New: char_transpose_1.f90 fails with ICE in gfc_conv_array_transpose
Test gfortran.dg/char_transpose_1.f90 fails on mainline for powerpc-unknown-linux-gnu with optimization: /home/gccbuild/gcc_trunk_anonsvn/gcc/gcc/testsuite/gfortran.dg/char_transpose_1.f90:9: internal compiler error: in gfc_conv_array_transpose, at fortran/trans-array.c:726^M Please submit a full bug report,^M with preprocessed source if appropriate.^M See http://gcc.gnu.org/bugs.html> for instructions.^M A regression hunt identified this patch: r112955 | kazu | 2006-04-14 14:31:32 + (Fri, 14 Apr 2006) | 8 lines * local-alloc.c (rest_of_handle_local_alloc): Use VEC instead of VARRAY. * reload1.c (reg_equiv_memory_loc_varray): Rename to reg_equiv_memory_loc_vec. Change the type to VEC(rtx,gc) *. (init_reload, reload): Use VEC instead of VARRAY. * reload.h: Update the prototype for reg_equiv_memory_loc_varray. http://gcc.gnu.org/viewcvs?view=rev&rev=112955 That seemed unlikely so I tried reverting that patch from today's mainline, and it did indeed allow the test to pass. The same ICE is reported in PR26119, filed 2006-02-06. -- Summary: char_transpose_1.f90 fails with ICE in gfc_conv_array_transpose Product: gcc Version: 4.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: fortran AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: janis at gcc dot gnu dot org GCC target triplet: powerpc-unknown-linux-gnu http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27229
[Bug target/27230] New: many obj-c++.dg tests ICE in rs6000_output_function_epilogue
On powerpc64-unknown-linux-gnu with -m64, 85 tests in obj-c++.dg fail with: elm3b11% /opt/gcc-nightly/trunk-20060420/bin/g++ -c -m64 basic.mm basic.mm: In function void __objc_gnu_init(): basic.mm:21: internal compiler error: in rs6000_output_function_epilogue, at config/rs6000/rs6000.c:15346 Please submit a full bug report, with preprocessed source if appropriate. See http://gcc.gnu.org/bugs.html> for instructions. This has apparently never worked on the FSF tree for this target. -- Summary: many obj-c++.dg tests ICE in rs6000_output_function_epilogue Product: gcc Version: 4.2.0 Status: UNCONFIRMED Keywords: ice-on-valid-code Severity: normal Priority: P3 Component: target AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: janis at gcc dot gnu dot org GCC target triplet: powerpc64-unknown-linux-gnu http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27230
[Bug objc++/23716] obj-c++.dg/comp-types-10.mm ICE with the GNU runtime
--- Comment #3 from janis at gcc dot gnu dot org 2006-04-20 23:31 --- This test and obj-c++.dg/try-catch-9.mm get the same ICE on powerpc-linux on trunk as of 2006-04-20, and have failed since I started including obj-c++ on 2005-11-02. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23716
[Bug objc++/23616] obj-c++.dg/try-catch-[29].mm (objc exceptions are broken) fails with the GNU Runtime
--- Comment #5 from janis at gcc dot gnu dot org 2006-04-21 00:10 --- try-catch-2.mm still fails the same way on powerpc-linux as of 2006-04-20. The other one currently gets an ICE. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23616
[Bug objc++/23614] obj-c++.dg/lookup-2.mm fails with the GNU runtime
--- Comment #4 from janis at gcc dot gnu dot org 2006-04-21 00:11 --- lookup-2.mm still fails the same way for powerpc-linux as of 2006-04-20. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23614
[Bug testsuite/23613] obj-c++.dg/isa-field-1.mm fails with the GNU runtime
--- Comment #4 from janis at gcc dot gnu dot org 2006-04-21 00:12 --- isa-field-1.mm still fails the same way for powerpc-linux as of 2006-04-20. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23613
[Bug objc++/27232] New: encode-3.mm fails with "deprecated conversion" warning
When run on powerpc-linux, test obj-c++.dg/encode-3.mm fails with an unexpected
warning:
elm3b11% /opt/gcc-nightly/trunk-20060420/bin/g++ -c encode-3.mm
encode-3.mm: In function int main():
encode-3.mm:18: warning: deprecated conversion from string constant to char*'
Test results for lots of other targets show the test failing.
--
Summary: encode-3.mm fails with "deprecated conversion" warning
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: objc++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27232
[Bug target/27282] [4.2 regression] ICE in final_scan_insn, at final.c:2448 - could not split insn
--- Comment #10 from janis at gcc dot gnu dot org 2006-04-25 21:16 --- A regression hunt of trunk on powerpc-linux using mini.c with -O2 identified: http://gcc.gnu.org/viewcvs?view=rev&rev=109016 r109016 | bonzini | 2005-12-23 16:07:53 + (Fri, 23 Dec 2005) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27282
[Bug rtl-optimization/26855] [4.2 Regression] ICE in add_deps_for_def with -fmodulo-sched -maltivec
--- Comment #2 from janis at gcc dot gnu dot org 2006-04-25 23:49 --- The patch that introduced this failure also causes the same ICE in several SPEC CPU2000 programs on powerpc64-linux for "-O2 -fmodulo-sched" with either -m32 or -m64. I haven't tried the patch that was supposed to fix the reported bug, http://gcc.gnu.org/ml/gcc-patches/2006-03/msg01642.html, or the follow-up smaller patches which don't seem to have been reviewed yet. I'm willing to test the proposed patch(es), including building SPEC CPU2000 with multiple sets of options and running the small, test input. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26855
[Bug rtl-optimization/26855] [4.2 Regression] ICE in add_deps_for_def with -fmodulo-sched -maltivec
--- Comment #4 from janis at gcc dot gnu dot org 2006-04-26 17:48 --- The patch doesn't apply cleanly now, which isn't surprising, but it also doesn't apply to mainline sources as of 2006-03-28, when it was submitted. What date or revision can I start with to try this patch, without porting it forward to today's sources? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26855
[Bug rtl-optimization/26855] [4.2 Regression] ICE in add_deps_for_def with -fmodulo-sched -maltivec
--- Comment #6 from janis at gcc dot gnu dot org 2006-04-26 20:54 --- Thanks, although I can wait if you have more pressing things to do. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26855
[Bug tree-optimization/27331] New: segfault in fold_convert with -ftree-vectorize -maltivec
The SPEC CPU2000 test 176.gcc fails to build on powerpc64-linux with "-O2
-ftree-vectorize -maltivec" for both -m32 and -m64. A regression hunt using a
minimized testcase (to be attached) identified this patch:
http://gcc.gnu.org/viewcvs?view=rev&rev=112621
r112621 | spop | 2006-04-02 04:27:40 + (Sun, 02 Apr 2006)
Output compiling the minimized testcase:
elm3b145% /opt/gcc-nightly/trunk-20060425/bin/gcc -O2 -ftree-vectorize
-maltivec -c recog.c
recog.c: In function constrain_operands:
recog.c:18: 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.
This segfaults in fold_convert in fold-const.c because argument "type" is 0.
--
Summary: segfault in fold_convert with -ftree-vectorize -maltivec
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27331
[Bug tree-optimization/27331] segfault in fold_convert with -ftree-vectorize -maltivec
--- Comment #1 from janis at gcc dot gnu dot org 2006-04-26 23:02 --- Created an attachment (id=11338) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11338&action=view) recog.c minimized testcase -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27331
[Bug tree-optimization/27332] New: ICE in try_interchange_loops with -ftree-loop-linear
Tests gap and twolf from SPEC CPU2000 ICE when built on powerpc64-linux with
"-O2 -ftree-loop-linear" and either -m32 or -m64. A third test, lucas, started
getting segfaults for those options at the same time. A regression hunt using
a minimized testcase (to be attached) identified this patch:
http://gcc.gnu.org/viewcvs?view=rev&rev=112437
r112437 | spop | 2006-03-28 04:19:26 + (Tue, 28 Mar 2006)
Output compiling the minimized testcase:
elm3b145% /opt/gcc-nightly/trunk/bin/gcc -O2 -ftree-loop-linear -c uloop.c
uloop.c: In function foo:
uloop.c:3: internal compiler error: vector VEC(ddr_p,base) index domain error,
in try_interchange_loops at tree-loop-linear.c:183
Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html> for instructions.
--
Summary: ICE in try_interchange_loops with -ftree-loop-linear
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27332
[Bug tree-optimization/27332] ICE in try_interchange_loops with -ftree-loop-linear
--- Comment #1 from janis at gcc dot gnu dot org 2006-04-26 23:05 --- Created an attachment (id=11339) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11339&action=view) test case -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27332
[Bug rtl-optimization/26855] [4.2 Regression] ICE in add_deps_for_def with -fmodulo-sched -maltivec
--- Comment #7 from janis at gcc dot gnu dot org 2006-04-29 00:02 --- I tried the patch at http://gcc.gnu.org/ml/gcc-patches/2006-04/msg01061.html on powerpc64-linux and used the resulting compilers with "-O2 -fmodulo-sched" to build SPEC CPU2000 and run with the small, test input, and also built and ran the special version of HMMER (which uses AltiVec macros) with those same options. I still get lots of failures: some tests ICE in the build, others get runtime failures. I got failures with different tests when I moved the compiler install tree to a different system, or when I ran it as a different user. There's something very flaky going on. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26855
[Bug c++/27370] [4.0 Regression] Bogus warning about ignoring function return value (__attribute__ ((warn_unused_result)))
--- Comment #3 from janis at gcc dot gnu dot org 2006-05-02 00:09 --- The warning for the original testcase went away with this patch: r81764 | dnovillo | 2004-05-13 06:41:07 + (Thu, 13 May 2004) | 3 lines Merge tree-ssa-20020619-branch into mainline. http://gcc.gnu.org/viewcvs?view=rev&rev=81764 Ouch! I've got another reghunt going to find out when the warning started being issued again between 4.0 and 4.1. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27370
[Bug c++/27370] [4.0 Regression] Bogus warning about ignoring function return value (__attribute__ ((warn_unused_result)))
--- Comment #4 from janis at gcc dot gnu dot org 2006-05-02 16:14 --- The warning was issued again with this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=101269 r101269 | jason | 2005-06-23 14:44:21 + (Thu, 23 Jun 2005) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27370
[Bug c++/27339] [4.1/4.2 Regression] out-of-class definition of value template parameter with private type
--- Comment #2 from janis at gcc dot gnu dot org 2006-05-02 16:16 --- A regression hunt on powerpc-linux identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=105682 r105682 | mmitchel | 2005-10-20 18:01:52 + (Thu, 20 Oct 2005) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27339
[Bug c++/27369] [4.2 Regression] tree check ICE when attribute externally_visible used
--- Comment #8 from janis at gcc dot gnu dot org 2006-05-02 18:18 --- A regression hunt on powerpc-linux using the test from comment #4 identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=101295 r101295 | hubicka | 2005-06-24 15:14:04 + (Fri, 24 Jun 2005) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||hubicka at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27369
[Bug tree-optimization/27093] [4.2 Regression] verify_ssa failed: definition does not dominate use
--- Comment #5 from janis at gcc dot gnu dot org 2006-05-02 23:04 --- A regression hunt on powerpc-linux using the test from comment #0 identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=111608 r111608 | dberlin | 2006-03-01 17:46:56 + (Wed, 01 Mar 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||dberlin at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27093
[Bug fortran/27229] char_transpose_1.f90 fails with ICE in gfc_conv_array_transpose
--- Comment #2 from janis at gcc dot gnu dot org 2006-05-08 15:08 --- Very odd; I do nightly builds of mainline on two different systems. On one of them this failure stopped on 20060430 and on the other it stopped on 20060503, but failed on 20060502. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27229
[Bug target/27266] [4.2 Regression] syncronization primitives cause ICE with -march=pentium or better
--- Comment #5 from janis at gcc dot gnu dot org 2006-05-09 23:57 --- A regression hunt using an i686-linux cross compiler on powerpc-linux with the testcase attached in comment #1 identified this large merge from gomp-branch: http://gcc.gnu.org/viewcvs?view=rev&rev=108997 r108997 | rth | 2005-12-23 00:43:34 + (Fri, 23 Dec 2005) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27266
[Bug c++/27210] [4.2 Regression] ICE on c++ template
--- Comment #7 from janis at gcc dot gnu dot org 2006-05-09 23:59 --- A regression hunt identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=112869 r112869 | mmitchel | 2006-04-11 22:59:57 + (Tue, 11 Apr 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27210
[Bug fortran/27229] char_transpose_1.f90 fails with ICE in gfc_conv_array_transpose
--- Comment #4 from janis at gcc dot gnu dot org 2006-05-10 16:03 --- The test passes on both of my nightly builds now, so I'm closing this. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27229
[Bug c++/27578] [4.2 Regression] ICE with attribute on a pointer in a function prototype
--- Comment #6 from janis at gcc dot gnu dot org 2006-05-12 20:38 --- A regression hunt on powerpc-linux using the testcase from comment #5 identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=113081 r113081 | mmitchel | 2006-04-19 16:58:23 + (Wed, 19 Apr 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27578
[Bug target/27619] New: wrong code for mixed-mode division with -mpowerpc64 -O1
The following code fails on powerpc64-linux compiled with -mpowerpc64 and -O1
or -O2; it passes with -O0, or without -mpowerpc64:
-
extern void abort (void);
int i = 274;
float f = 1.0f;
int
main ()
{
double d = i / f;
if (d != 274.0)
abort ();
return 0;
}
This fails with current trunk, with 4.1 and 4.0 branches, and with the last 3.4
branch build I did. This bug affects program vpr in SPEC CPU2000.
--
Summary: wrong code for mixed-mode division with -mpowerpc64 -O1
Product: gcc
Version: 4.0.4
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc64-unknown-linux-gnu
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27619
[Bug target/27619] wrong code for mixed-mode division with -mpowerpc64 -O1
--- Comment #9 from janis at gcc dot gnu dot org 2006-05-16 17:55 --- Good grief, if it "might not work with Linux" then it shouldn't be available for GNU/Linux targets. -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||dje at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27619
[Bug target/27619] wrong code for mixed-mode division with -mpowerpc64 -O1
--- Comment #10 from janis at gcc dot gnu dot org 2006-05-16 18:26 --- By the way, I found this by running SPEC CPU2000, FreePOOMA, FTensor, and Blitz++ with several sets of options plus either "-m32", "-m64", or "-m32 -mpowerpc64" and this was the only failure I saw. This failure happens consistently. It seems as if a kernel issue would affect additional tests. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27619
[Bug target/27619] wrong code for mixed-mode division with -mpowerpc64 -O1
--- Comment #11 from janis at gcc dot gnu dot org 2006-05-16 22:42 --- As mentioned above, the Linux kernel does not provide context switching support needed for "-m32 -mpowerpc64". I'm looking into disabling it for powerpc64-linux. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27619
[Bug target/27682] New: float to int conversion doesn't raise invalid exception
GCC claims to follow C99 Annex F when converting a floating value to an integer
type when the integral part exceeds the range of the integer type (C99 6.3.1.4
and F.4). Annex F says that in this case the conversion raises the "invalid"
floating-point exception. On powerpc64-linux and x86_64-linux, however, the
instruction that is used to convert double to int converts to a 64-bit integer,
not a 32-bit register, so the exception is not raised. On powerpc64-linux with
-m64 the instruction is fctidz (floating convert to integer doubleword with
round toward zero). If I replace that instruction with fctiwz (convert to
integer word) in the .s file and compile that, the code behaves as expected.
Here's a small test case:
#include
extern void abort (void);
double x = 4294967296.0;
unsigned int i;
int
main ()
{
#pragma STDC FENV_ACCESS ON
feclearexcept (FE_ALL_EXCEPT);
i = x;/* value exceeds the range of 32-bit int */
if (! fetestexcept (FE_INVALID))
abort ();
return 0;
}
I see this behavior in all powerpc64-linux compilers I've checked from 3.2
through mainline. The exception is raised for tests compiled with -m32.
This is a problem for applications that try to detect such invalid conversions
after checking for __STDC_IEC_559__, which claims that the implementation
conforms to Annex F.
A possibly-related issue is that this macro is defined in a header file
provided by glibc, not by GCC.
--
Summary: float to int conversion doesn't raise invalid exception
Product: gcc
Version: 4.0.4
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc64-linux, x86_64-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27682
[Bug target/27682] float to int conversion doesn't raise invalid exception
--- Comment #1 from janis at gcc dot gnu dot org 2006-05-19 16:13 --- This is related to PR21360, although that one is about the result of an out-of-range conversion. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27682
[Bug target/27682] float to int conversion doesn't raise invalid exception
--- Comment #4 from janis at gcc dot gnu dot org 2006-05-19 17:25 --- This has been brought up to the glibc developers in the past: http://sourceware.org/ml/libc-alpha/2005-03/msg00196.html It's unfortunate that they don't agree that "an implementation" of C consists of a combination of products, and that unless they all support Annex F, the "implementation" doesn't support it and should not define __STDC_IEC_559__. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27682
[Bug c++/27648] [4.0/4.1/4.2 Regression] ICE on attribute on pointers in static_cast
--- Comment #3 from janis at gcc dot gnu dot org 2006-05-22 18:40 ---
I get errors for both the submitter's testcase and the testcase in comment #1.
With the comment #1 testcase on powerpc-linux:
elm3b11% /opt/gcc-nightly/3.4/bin/g++ -c 27648.cc
27648.cc: In function `int main(int, char**)':
27648.cc:4: error: expected `>' before "__attribute__"
27648.cc:4: error: expected `(' before "__attribute__"
27648.cc:4: error: expected primary-expression before "__attribute__"
27648.cc:4: error: expected `)' before "__attribute__"
27648.cc:4: error: expected `;' before "__attribute__"
elm3b11% /home/janis/tools/gcc-3.3.5-ppc32/bin/g++ -c 27648.cc
27648.cc: In function `int main(int, char**)':
27648.cc:4: error: invalid type modifier within pointer declarator
elm3b11% /home/janis/tools/gcc-3.2.3-ppc32/bin/g++ -c 27648.cc
27648.cc: In function `int main(int, char**)':
27648.cc:4: invalid type modifier within pointer declarator
Is the testcase valid code? I can do a regression hunt for when the compiler
started getting the ICE instead of errors, but is that helpful?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27648
[Bug c++/27640] [4.2 Regression] segfault, related to constructor instantiation and virtual base case and templates and new operator
--- Comment #6 from janis at gcc dot gnu dot org 2006-05-22 19:48 --- A regression hunt on powerpc-linux using the testcase from comment #4 identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=112869 r112869 | mmitchel | 2006-04-11 22:59:57 + (Tue, 11 Apr 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27640
[Bug c++/27693] [4.0/4.1/4.2 Regression] strange interaction with const and sizeof and variable declarations in g++-4.0
--- Comment #2 from janis at gcc dot gnu dot org 2006-05-22 21:18 --- A regression hunt of mainline on powerpc-linux identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=111229 r111229 | mmitchel | 2006-02-18 08:36:11 + (Sat, 18 Feb 2006) The patch was applied to gcc-4_1-branch and gcc-4_0-branch on the same day. I did not verify that it caused the regressions there. -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27693
[Bug c++/27648] [4.0/4.1/4.2 Regression] ICE on attribute on pointers in static_cast
--- Comment #5 from janis at gcc dot gnu dot org 2006-05-22 23:02 --- A regression hunt for mainline on powerpc-linux identified this patch as the one for which the testcase starts getting an ICE instead of the error messages shown for GCC 3.4 (branch) in comment #3: http://gcc.gnu.org/viewcvs?view=rev&rev=83482 r83482 | mmitchel | 2004-06-22 07:18:38 + (Tue, 22 Jun 2004) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27648
[Bug tree-optimization/27745] New: ICE in execute_todo with -O2 -ftree-loop-linear
GNU Fortran fails on powerpc-linux with "-O2 -ftree-loop-linear" with an ICE
for this testcase, minimized from swim.f in SPEC CPU2000:
SUBROUTINE BUG
INTEGER I, J, M
REAL V
COMMON A(100,100), B(100,100), M, V
DO 200 I = 1, M
DO 100 J = 1, M
V = V + A(I,J)
100CONTINUE
B(I,I) = B(I,I) * I
200 CONTINUE
STOP
END
The output is:
elm3b11% /opt/gcc-nightly/trunk/bin/gfortran -c -O2 -ftree-loop-linear bug.f
bug.f: In function bug:
bug.f:1: internal compiler error: in execute_todo, at passes.c:714
Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html> for instructions.
A regression hunt identified this patch:
http://gcc.gnu.org/viewcvs?view=rev&rev=113862
r113862 | spop | 2006-05-17 14:25:59 + (Wed, 17 May 2006)
The same patch also introduces memory exhaustion failures when compiling equake
or ammp from SPEC CPU2000 on powerpc64-linux with "-O2 -ftree-loop-linear" and
either -m32 or -m64, and for applu with those options and -m64. I don't have a
minimized testcase for any of those.
--
Summary: ICE in execute_todo with -O2 -ftree-loop-linear
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27745
[Bug c/27747] [4.2 Regression] ice _Pragma and macros
--- Comment #3 from janis at gcc dot gnu dot org 2006-05-24 18:23 --- A regression hunt on powerpc-linux identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=109336 r109336 | rth | 2006-01-04 16:33:38 + (Wed, 04 Jan 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||rth at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27747
[Bug c++/27748] [4.2 Regression] rejects _Pragma with macros
--- Comment #3 from janis at gcc dot gnu dot org 2006-05-24 21:17 --- A regression hunt on powerpc-linux identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=109336 r109336 | rth | 2006-01-04 16:33:38 + (Wed, 04 Jan 2006) This is the same one that started the ICE for the corresponding C test in 27747. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27748
[Bug target/27770] New: wrong code in spec tests for -ftree-vectorize -maltivec
Two tests from SPEC CPU2000, parser and gzip, fail at runtime when built on
powerpc64-linux with "-O2 -ftree-vectorize -maltivec -mabi=altivec". parser
fails with both -m32 and -m64, gzip with -m64. A regression hunt identified
this patch as the start of the failures:
http://gcc.gnu.org/viewcvs?view=rev&rev=113395
r113395 | dje | 2006-04-30 19:23:13 + (Sun, 30 Apr 2006)
I've been able to duplicate the failure with 32-bit parser by compiling only
post-process.c with "-m32 -O2 -ftree-vectorize -maltivec -mabi=altivec" and
everything else with "-m32 -O2". In addition, if all static variables in
post-process.c are made global, the test passes, but with only word_links and
visited left static it fails. I can continue to work on coming up with a
self-contained test case, but perhaps this is enough information; let me know.
--
Summary: wrong code in spec tests for -ftree-vectorize -maltivec
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: wrong-code
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27770
[Bug target/27770] wrong code in spec tests for -ftree-vectorize -maltivec
--- Comment #1 from janis at gcc dot gnu dot org 2006-05-25 16:30 --- There's also a problem in the parser file analyze-linkage.c. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27770
[Bug middle-end/27773] [4.2 regression] ICE: in find_lattice_value, at tree-complex.c:133
--- Comment #7 from janis at gcc dot gnu dot org 2006-05-26 18:36 --- A regression hunt on powerpc-linux using the testcase from comment #5 with --ffast-math identified the following patch: http://gcc.gnu.org/viewcvs?view=rev&rev=107218 r107218 | rguenth | 2005-11-19 11:29:10 + (Sat, 19 Nov 2005) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||rguenth at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27773
[Bug target/27682] float to int conversion doesn't raise invalid exception
--- Comment #7 from janis at gcc dot gnu dot org 2006-05-30 16:59 --- If the value doesn't fit then the result is undefined and doesn't matter; getting the invalid flag is what's important. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27682
[Bug tree-optimization/27865] New: tree check failure building FreePOOMA
FreePOOMA 2.4.1 fails to build on powerpc-linux with beginning with this patch:
http://gcc.gnu.org/viewcvs?view=rev&rev=114057
r114057 | rakdver | 2006-05-24 22:55:15 + (Wed, 24 May 2006)
Output with a minimized testcase:
elm3b145% /opt/gcc-nightly/trunk/bin/g++ -c -O2 poomabug.cc
poomabug.cc: In constructor GLD::GLD(const II&, const PP&, const
CM&) [with PP = GP<2> ()(), int Dim = 2]:
poomabug.cc:117: internal compiler error: tree check: expected integer_type or
enumeral_type or boolean_type or real_type, have pointer_type in
adjust_range_with_scev, at tree-vrp.c:2079
Please submit a full bug report,
with preprocessed source if appropriate.
See http://gcc.gnu.org/bugs.html> for instructions.
--
Summary: tree check failure building FreePOOMA
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Keywords: ice-on-valid-code
Severity: normal
Priority: P3
Component: tree-optimization
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27865
[Bug tree-optimization/27865] tree check failure building FreePOOMA
--- Comment #1 from janis at gcc dot gnu dot org 2006-06-01 17:45 --- Created an attachment (id=11574) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11574&action=view) minimized testcase -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27865
[Bug target/27858] [4.2 Regression] ICE in spill_failure, at reload1.c:1911while bootstrapping 4.2 on alpha
--- Comment #5 from janis at gcc dot gnu dot org 2006-06-01 17:50 --- A regression hunt using an alpha-linux cross compiler on powerpc64-linux with the testcase mini.c identified the following patch: http://gcc.gnu.org/viewcvs?view=rev&rev=113632 r113632 | sayle | 2006-05-08 21:09:49 + (Mon, 08 May 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||sayle at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27858
[Bug tree-optimization/27882] [4.2 regression] segfault in ipa-inline.c, if (e->callee->local.disregard_inline_limits
--- Comment #7 from janis at gcc dot gnu dot org 2006-06-05 19:45 --- A regression hunt on powerpc-linux using the less-reduced test case with the options "-O2 --param ggc-min-expand=0 --param ggc-min-heapsize=0" identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=112753 r112753 | hubicka | 2006-04-07 15:24:39 + (Fri, 07 Apr 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||hubicka at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27882
[Bug target/27770] [4.2 Regression] wrong code in spec tests for -ftree-vectorize -maltivec
--- Comment #5 from janis at gcc dot gnu dot org 2006-06-05 20:58 --- I haven't yet had time to continue trying to come up with a minimized testcase but hope to get to that soon. I had a vague hope that someone who understands the section anchor support and has access to SPEC CPU2000 could take a quick look at the generated code for post-process.c and notice something obvious. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27770
[Bug testsuite/26813] Testsuite does not check for ICEs
--- Comment #3 from janis at gcc dot gnu dot org 2006-06-06 20:48 --- Gosh, I don't remember ever seeing this stuff about new exit codes for ICEs. I can rethink how the testsuite detects them, now that we know where to look. Does the ICE message get translated for different locales? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26813
[Bug tree-optimization/27830] [4.1/4.2 regression] ICE: verify_stmts failed (invalid operand to unary operator)
--- Comment #5 from janis at gcc dot gnu dot org 2006-06-07 18:23 --- A regression hunt on powerpc-linux with the testcase in comment #3 identified this mainline patch: http://gcc.gnu.org/viewcvs?view=rev&rev=112709 r112709 | rguenth | 2006-04-05 15:41:18 + (Wed, 05 Apr 2006) The same patch was added to the 4.1 branch a couple of days later, so I assume it's also the cause of the failure there. GCC 4.1.0 does not fail. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27830
[Bug target/27907] [4.2 regression] ICE in expand_simple_unop, at optabs.c:2307
--- Comment #2 from janis at gcc dot gnu dot org 2006-06-07 21:52 --- A regression hunt using an alpha-linux cross compiler on powerpc-linux with the first testcase from comment #1 identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=110912 r110912 | rakdver | 2006-02-13 11:21:23 + (Mon, 13 Feb 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||rakdver at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27907
[Bug middle-end/27881] [4.1 Regression] Memory exhausted with -finline-functions on testsuite file alias3.C
--- Comment #7 from janis at gcc dot gnu dot org 2006-06-07 22:44 --- A regression hunt on powerpc-linux using the testcase in the description with "ulimit -v 50" identified this patch as the start of the failures: http://gcc.gnu.org/viewcvs?view=rev&rev=102521 r102521 | hubicka | 2005-07-28 21:45:27 + (Thu, 28 Jul 2005) The test still fails on mainline with r109576 and passes with r109587, so something on 2006-01-11 in between those revisions fixes it. Builds are broken in that range, but I'm adapting my reghunt setup to handle that so I might come up with an answer. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27881
[Bug middle-end/27881] [4.1 Regression] Memory exhausted with -finline-functions on testsuite file alias3.C
--- Comment #8 from janis at gcc dot gnu dot org 2006-06-07 23:19 --- The failures stop on mainline with this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=109580 r109580 | hubicka | 2006-01-11 13:13:37 + (Wed, 11 Jan 2006) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||hubicka at gcc dot gnu dot ||org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27881
[Bug c++/27826] [4.0/4.1 Regression] ICE in copy_to_mode_reg
--- Comment #12 from janis at gcc dot gnu dot org 2006-06-09 18:10 --- The reduced testcase from comment #7 doesn't fail with my i686 cross compilers on powerpc-linux, so I did a regression hunt using the testcase from comment #4. It identified this patch as the fix on mainline: http://gcc.gnu.org/viewcvs?view=rev&rev=113199 r113199 | mmitchel | 2006-04-23 18:04:33 + (Sun, 23 Apr 2006) This is indeed Mark's bitfield patch, Andrew guessed correctly. I've got another regression hunt running to find the break; it's currently between revisions 83668 and 83876 in late May 2004; there are broken builds in there so it might take awhile to keep the hunt progressing. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27826
[Bug c++/27826] [4.0/4.1 Regression] ICE in copy_to_mode_reg
--- Comment #13 from janis at gcc dot gnu dot org 2006-06-09 18:44 --- The test starts failing on mainline with this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=83858 r83858 | rth | 2004-06-29 16:25:28 + (Tue, 29 Jun 2004) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27826
[Bug other/28002] New: decNumber sources need GPL+exception for use in libgcc
The decNumber sources files in libdecnumber were contributed by IBM and are currently covered by the GPL. When GCC is configured with --enable-decimal-float, some of these functions are included in the static version of libgcc. The license needs to change from GPL to GPL plus exception. I've assigned this to myself as a reminder to contact the appropriate people within IBM and then the FSF. -- Summary: decNumber sources need GPL+exception for use in libgcc Product: gcc Version: 4.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: other AssignedTo: janis at gcc dot gnu dot org ReportedBy: janis at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28002
[Bug target/28034] New: section anchors break -fprofile-generate
Section anchors have broken profile data on powerpc*-linux; the problems
described here started with this mainline patch:
http://gcc.gnu.org/viewcvs?view=rev&rev=113395
r113395 | dje | 2006-04-30 19:23:13 + (Sun, 30 Apr 2006)
Tests mesa, gap, and perlbmk in SPEC CPU2000 fail when compiled with -O1 and
-fprofile-generate/-fprofile-use, although all source files that contain static
variables have extra space generated after the profile counts. Here's what I
get for file dlist.c in mesa:
.section".bss"
.align 3
.set.LANCHOR0,. + 0
.type .LPBX1, @object
.size .LPBX1, 8496
.LPBX1:
.zero 8496
.zero 18446744073709551128
.type init_flag.9563, @object
.size init_flag.9563, 4
In this case the assembler complains that ".space repeat count is zero" for the
line with the second .zero, and the -fprofile-use compile fails because of
corrupted profile data. gcov shows bogus numbers in this case, although when
the size of the second .zero is small, tests run correctly and gcov data looks
fine.
I don't have a small test case that fails for -fprofile-use or has bogus gcov
results, but here's one where the second .zero is clearly bogus:
void
foo (int *p, int n)
{
int i;
for (i = 0; i < n; i++)
p[i] = i;
}
static int tmp [1000];
int
main ()
{
foo (tmp, 1000);
return 0;
}
Part of the corresponding .s file:
.section".bss"
.align 3
.set.LANCHOR0,. + 0
.type .LPBX1, @object
.size .LPBX1, 32
.LPBX1:
.zero 32
.zero 7976
.type tmp, @object
.size tmp, 4000
tmp:
.zero 4000
--
Summary: section anchors break -fprofile-generate
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28034
[Bug libgomp/28046] New: libgomp test pr27337.C fails intermittently
My mainline testing gets intermittent failures for test libgomp/testsuite/libgomp.g++/pr27337.c. Some of the failures are due to wrong counts from the constructor and destructor after the call to foo, and others are for wrong values of x.i (with values in the range 19-27) after the call to bar. The percentage of failures depends on which system I test on, but I haven't figured out the significant factors: CPU type, number of processors, binutils version, kernel version, load on the system I haven't studied omp programs; the failures could be due to errors in the test itself. The failures occur at all levels of optimization. I get this failure on powerpc*-linux (with both -m32 and -m64) and on x86_64-linux. -- Summary: libgomp test pr27337.C fails intermittently Product: gcc Version: 4.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: libgomp AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: janis at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28046
[Bug middle-end/28042] optimizer (-O2) changes the semantics of my programs
--- Comment #10 from janis at gcc dot gnu dot org 2006-06-16 18:59 --- The regression hunt to find when the testcase starts passing identified this mainline patch: http://gcc.gnu.org/viewcvs?view=rev&rev=106400 r106400 | rth | 2005-11-02 21:44:17 + (Wed, 02 Nov 2005) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28042
[Bug c++/27693] [4.0/4.1 Regression] strange interaction with const and sizeof and variable declarations in g++-4.0
--- Comment #4 from janis at gcc dot gnu dot org 2006-06-16 21:12 --- A regression hunt confirmed that this was fixed by: http://gcc.gnu.org/viewcvs?view=rev&rev=114119 r114119 | mmitchel | 2006-05-25 20:18:26 + (Thu, 25 May 2006) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27693
[Bug middle-end/28034] [4.2 Regression] section anchors break -fprofile-generate
--- Comment #4 from janis at gcc dot gnu dot org 2006-06-19 21:08 --- I tried the patch with a C-only bootstrap for biarch powerpc64-linux and ran the three CPU2000 tests that had failed with profile generate/use; with the patch they work. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28034
[Bug c++/28113] New: vectors initialized in ctors, not at compile time, cause altivec-3.C failure
Test g++.dg/ext/altivec-3.C has been failing for mainline on powerpc64-linux
systems without VMX hardware since this patch, a fix for PR20103, was added:
http://gcc.gnu.org/viewcvs?view=rev&rev=114119
r114119 | mmitchel | 2006-05-25 20:18:26 + (Thu, 25 May 2006)
The test calls a function at runtime to determine if VMX hardware is available
and if not, skips the rest of the test. It now fails because vector global
variables are initialized by constructors instead of at compile time. Here's a
minimal example:
__vector signed int vsi = (__vector signed int) {40, 41, 42, 43};
int main () { return 0; }
If the new behavior is expected then I'll change the test to be compile-only on
systems without VMX hardware.
--
Summary: vectors initialized in ctors, not at compile time,
cause altivec-3.C failure
Product: gcc
Version: 4.2.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28113
[Bug c++/28113] [4.2 Regression] vectors initialized in ctors, not at compile time, cause altivec-3.C failure
--- Comment #5 from janis at gcc dot gnu dot org 2006-06-20 23:34 --- Subject: Bug 28113 Author: janis Date: Tue Jun 20 23:33:58 2006 New Revision: 114832 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=114832 Log: PR c++/28113 * g++.dg/ext/altivec-3.C: Remove cast from vector initializers. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/g++.dg/ext/altivec-3.C -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28113
[Bug c++/28088] [4.1 Regression] Internal compiler error on boost mpl test/apply.cpp
--- Comment #7 from janis at gcc dot gnu dot org 2006-07-05 20:35 --- Regression tests using the reduced testcase from comment #3 on mainline for powerpc-linux identified this patch where the test starts failing: http://gcc.gnu.org/viewcvs?view=rev&rev=103604 r103604 | mmitchel | 2005-08-29 14:08:50 + (Mon, 29 Aug 2005) and this one where it starts passing again: http://gcc.gnu.org/viewcvs?view=rev&rev=110693 r110693 | mmitchel | 2006-02-07 11:11:30 + (Tue, 07 Feb 2006) The regression hunts didn't use any special options and didn't ICE, but got these messages: elm3b11% /opt/gcc-nightly/trunk-20060131/bin/g++ -c 28088.cc 28088.cc: In instantiation of const int template_arity >::value: 28088.cc:14: instantiated from here 28088.cc:8: error: template_arity::value is not a valid template argument for type int because it is a non-constant expression 28088.cc:8: error: template_arity::value is not a valid template argument for type int because it is a non-constant expression -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||mmitchel at gcc dot gnu dot ||org AssignedTo|unassigned at gcc dot gnu |janis at gcc dot gnu dot org |dot org | Status|NEW |ASSIGNED Last reconfirmed|2006-06-21 04:21:48 |2006-07-05 20:35:23 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28088
[Bug middle-end/27889] [4.1/4.2 Regression] ICE on complex assignment in nested function
--- Comment #18 from janis at gcc dot gnu dot org 2006-07-13 18:18 --- A regression hunt on powerpc-linux using the C testcase from comment #6 identified the following patch: http://gcc.gnu.org/viewcvs?view=rev&rev=100793 r100793 | rth | 2005-06-09 07:43:40 + (Thu, 09 Jun 2005) -- janis at gcc dot gnu dot org changed: What|Removed |Added CC||rth at gcc dot gnu dot org http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27889
[Bug rtl-optimization/27616] [4.1/4.2 Regression] Internal error with -O1 (CSE)
--- Comment #8 from janis at gcc dot gnu dot org 2006-07-14 00:20 --- A regression hunt on powerpc-linux using an i686-linux cross compiler identified this patch: http://gcc.gnu.org/viewcvs?view=rev&rev=102570 r102570 | hubicka | 2005-07-29 22:22:41 + (Fri, 29 Jul 2005) The compiler didn't segfault on the test system, but it did run for a very long time so the test script used by the regression hunt timed out after 5 seconds. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27616
[Bug testsuite/37202] FAIL: gcc.dg/visibility-1[4-9].c
--- Comment #14 from janis at gcc dot gnu dot org 2008-11-12 17:48 --- Subject: Bug 37202 Author: janis Date: Wed Nov 12 17:47:13 2008 New Revision: 141794 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141794 Log: 2008-11-12 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/37202 * gcc.dg/visibility-14.c: Skip on *-*-darwin*. * gcc.dg/visibility-15.c: Same. * gcc.dg/visibility-16.c: Same. * gcc.dg/visibility-17.c: Same. * gcc.dg/visibility-18.c: Same. * gcc.dg/visibility-19.c: Same. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/gcc.dg/visibility-14.c trunk/gcc/testsuite/gcc.dg/visibility-15.c trunk/gcc/testsuite/gcc.dg/visibility-16.c trunk/gcc/testsuite/gcc.dg/visibility-17.c trunk/gcc/testsuite/gcc.dg/visibility-18.c trunk/gcc/testsuite/gcc.dg/visibility-19.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=37202
[Bug testsuite/38008] gcc/testsuite/gcc.c-torture/execute/builtins/lib/sprintf.c unportable
--- Comment #2 from janis at gcc dot gnu dot org 2008-11-12 17:53 --- Subject: Bug 38008 Author: janis Date: Wed Nov 12 17:52:24 2008 New Revision: 141795 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141795 Log: 2008-11-12 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/38008 * gcc.c-torture/execute/builtins/lib/sprintf.c: Make sprintf call more portable for darwin10. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/gcc.c-torture/execute/builtins/lib/sprintf.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38008
[Bug bootstrap/38010] gcc/config.gcc needs adjustment for darwin10
--- Comment #2 from janis at gcc dot gnu dot org 2008-11-12 21:35 --- Subject: Bug 38010 Author: janis Date: Wed Nov 12 21:33:34 2008 New Revision: 141803 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141803 Log: 2008-11-12 Jack Howarth <[EMAIL PROTECTED]> PR bootstrap/38010 * gcc/config.gcc: Use darwin9.h on darwin10 as well. Modified: trunk/gcc/ChangeLog trunk/gcc/config.gcc -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38010
[Bug testsuite/38098] gcc.dg/cpp/subframework1.c -fno-show-column fails on darwin9
--- Comment #1 from janis at gcc dot gnu dot org 2008-11-14 18:38 --- Subject: Bug 38098 Author: janis Date: Fri Nov 14 18:36:41 2008 New Revision: 141862 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141862 Log: 2008-11-14 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/38098 * gcc.dg/cpp/subframework1.c: Use -Wno-deprecated. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/gcc.dg/cpp/subframework1.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38098
[Bug testsuite/38099] tmpdir-gcc.dg-struct-layout-1/t027 c_compat_x_tst.o-c_compat_y_tst.o execute failure
--- Comment #24 from janis at gcc dot gnu dot org 2008-11-17 19:41 --- Subject: Bug 38099 Author: janis Date: Mon Nov 17 19:39:54 2008 New Revision: 141948 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=141948 Log: 2008-11-14 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/38099 * gcc.dg/compat/struct-layout-1_generate.c: Also use -no-mmx on i?86/x86_64 darwin. * g++.dg/compat/struct-layout-1_generate.c: Same. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/g++.dg/compat/struct-layout-1_generate.c trunk/gcc/testsuite/gcc.dg/compat/struct-layout-1_generate.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38099
[Bug testsuite/34575] gcc.target/powerpc/parity-1.c and popcount-1.c scan-assembler popcntb on darwin9
--- Comment #4 from janis at gcc dot gnu dot org 2008-11-17 20:19 --- Fixed by Jack's patch. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|NEW |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34575
[Bug testsuite/38008] gcc/testsuite/gcc.c-torture/execute/builtins/lib/sprintf.c unportable
--- Comment #3 from janis at gcc dot gnu dot org 2008-11-17 20:20 --- Fixed by Jack's patch. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38008
[Bug testsuite/38098] gcc.dg/cpp/subframework1.c -fno-show-column fails on darwin9
--- Comment #2 from janis at gcc dot gnu dot org 2008-11-17 20:21 --- Fixed by Jack's patch. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38098
[Bug testsuite/38099] tmpdir-gcc.dg-struct-layout-1/t027 c_compat_x_tst.o-c_compat_y_tst.o execute failure
--- Comment #25 from janis at gcc dot gnu dot org 2008-11-17 20:23 --- Fixed by Jack's patch. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|NEW |RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38099
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #21 from janis at gcc dot gnu dot org 2008-11-18 22:48 --- Interesting that you should ask, I modified the patch yesterday and intend to submit it as soon as I've done some more testing. The current version adds dg-timeout, which sets the timeout for running the compiler in gcc and libstdc++ tests to a specified number of seconds, or dg-timeout-factor, which multiplies the usual timeout by the specified factor. The "usual timeout" is the value for the target board, which can be overridden by gcc_timeout or v3_timeout defined in $HOME/.dejagnurc. Let me know if you've got suggestions for improvement. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #23 from janis at gcc dot gnu dot org 2008-11-21 00:58 --- I posted a patch for compiler tests at http://gcc.gnu.org/ml/gcc-patches/2008-11/msg01066.html but after playing around more realized that it shouldn't be necessary to allow setting a default in .dejagnurc, since the tester should be able to specify a target board that has a timeout. I tried that, adding "set_board_info gcc,timeout 100" to the baseboards/unix.exp file that's being used, and nothing happened; the tests still use the 300-second default from DejaGnu's remote_exec proc in remote.exp. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #24 from janis at gcc dot gnu dot org 2008-11-21 01:28 --- For the libstdc++ tests, which are the original focus of this PR, is it enough to provide dg-timeout and dg-timeout-factor and either leave the 600 default, or else take the larger of that and [target_info gcc,timeout]? By the way, dg-timeout-factor from my latest patch takes a floating-point value so it can be used to limit the time used for a known hanging test as well as enabling long-running tests to complete. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug c++/38076] FAIL: g++.dg/other/anon5.C
--- Comment #2 from janis at gcc dot gnu dot org 2008-11-24 18:07 --- Subject: Bug 38076 Author: janis Date: Mon Nov 24 18:05:50 2008 New Revision: 142163 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142163 Log: 2008-11-24 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/38076 * g++.dg/other/anon5.C: Skip on Darwin. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/g++.dg/other/anon5.C -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38076
[Bug testsuite/38241] g++.old-deja/g++.pt/const2.C fails on *-apple-darwin9
--- Comment #3 from janis at gcc dot gnu dot org 2008-11-24 18:12 --- Subject: Bug 38241 Author: janis Date: Mon Nov 24 18:11:12 2008 New Revision: 142164 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142164 Log: 2008-11-24 Jack Howarth <[EMAIL PROTECTED]> PR testsuite/38241 * g++.old-deja/g++.pt/const2.C: Prune output for Darwin. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/g++.old-deja/g++.pt/const2.C -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38241
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #25 from janis at gcc dot gnu dot org 2008-11-24 22:31 --- I'm still tweaking this, to support dg-timeout and dg-timeout-factor plus tool-specific default timeouts for gcc, libstdc++-v3, libgomp, and libmudflap. As for picking up gcc,timeout for the target board I discovered that it helps immensely if the unix.exp file I modify is the one on the machine where I'm running the tests; d'oh. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug libgomp/38270] New: libgomp test failures due to missing memory barrier
Libgomp test libgomp.c/atomic-3.c started failing with wrong answers on power5 and power6 systems for -m32, and tests nqueens-1.c and sort-1.c started hanging intermittently for both -m32 and -m64 on those systems, with the addition of r139969, which removes the memory barrier from rs6000_split_lock_test_and_set. Adding back the memory barrier allows those tests to pass again. David Edelsohn told me that sync_lock_test_and_set does not require the stricter semantics of a memory barrier and if libgomp needs the memory barrier then it must emit one itself. The patch submission for r139969, http://gcc.gnu.org/ml/gcc/2008-09/msg00038.html, has followups from David and from Richard Henderson briefly discussing the semantics of sync_lock_test_and_set. -- Summary: libgomp test failures due to missing memory barrier Product: gcc Version: 4.4.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: libgomp AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: janis at gcc dot gnu dot org GCC target triplet: powerpc64-unknown-linux-gnu http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38270
[Bug libgomp/38270] libgomp test failures due to missing memory barrier
--- Comment #2 from janis at gcc dot gnu dot org 2008-11-26 17:39 --- This is the same problem as in PR37938, as H.J. said, but for powerpc*-linux. I'll try copying the ia64 solution. I see that my objections to having a separate libgomp mutex.h for powerpc have already been raised by Steve Ellcey and addressed in the mailing list. -- janis at gcc dot gnu dot org changed: What|Removed |Added AssignedTo|unassigned at gcc dot gnu |janis at gcc dot gnu dot org |dot org | Status|UNCONFIRMED |ASSIGNED Ever Confirmed|0 |1 Last reconfirmed|-00-00 00:00:00 |2008-11-26 17:39:34 date|| http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38270
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #26 from janis at gcc dot gnu dot org 2008-11-26 18:52 --- Subject: Bug 28870 Author: janis Date: Wed Nov 26 18:51:07 2008 New Revision: 142225 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142225 Log: gcc/ PR testsuite/28870 * doc/sourcebuild.texi (Test Directives): Add dg-timeout and dg-timeout-factor. gcc/testsuite/ PR testsuite/28870 * lib/timeout.exp: New. * lib/timeout-dg.exp: New. * lib/gcc-dg.exp: Include new timeout library files. (dg-test): Unset timeout variables. * lib/gcc.exp (gcc_target_compile): Set timeout value from new proc. * lib/g++.exp (g++_target_compile): Ditto. * lib/gfortran.exp (gfortran_target_compile): Ditto. * lib/objc.exp (objc_target_compile): Ditto. * lib/obj-c++.exp (obj-c++_target_compile): Ditto. * lib/obj-c++.exp (obj-c++_target_compile): Ditto. * lib/gnat.exp (gnat_target_compile): Ditto. libstdc++-v3/ PR testsuite/28870 * testsuite/lib/libstdc++.exp: Include new timeout library files. (libstdc++_init): Define tool_timeout. (dg-test): Override DejaGnu proc. (v3_target_compile): Set timeout value from new proc. (v3_target_compile_as_c): Ditto. libmudflap/ PR testsuite/28870 * testsuite/lib/mfdg.exp (dg-test): Use new timeout support. (dg-timeout): Remove. (standard-wait): Remove. * testsuite/lib/libmudflap.exp: Include new timeout library files. (libmudflap_target_compile): Set timeout value from new proc. libgomp/ PR testsuite/28870 * testsuite/lib/libgomp.exp: Include new timeout library files. (libgomp_target_compile): Set timeout value from new proc. Added: trunk/gcc/testsuite/lib/timeout-dg.exp trunk/gcc/testsuite/lib/timeout.exp Modified: trunk/gcc/ChangeLog trunk/gcc/doc/sourcebuild.texi trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/lib/g++.exp trunk/gcc/testsuite/lib/gcc-dg.exp trunk/gcc/testsuite/lib/gcc.exp trunk/gcc/testsuite/lib/gfortran.exp trunk/gcc/testsuite/lib/gnat.exp trunk/gcc/testsuite/lib/obj-c++.exp trunk/gcc/testsuite/lib/objc.exp trunk/libgomp/ChangeLog trunk/libgomp/testsuite/lib/libgomp.exp trunk/libmudflap/ChangeLog trunk/libmudflap/testsuite/lib/libmudflap.exp trunk/libmudflap/testsuite/lib/mfdg.exp trunk/libstdc++-v3/ChangeLog trunk/libstdc++-v3/testsuite/lib/libstdc++.exp -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug libgomp/38270] libgomp test failures due to missing memory barrier
--- Comment #3 from janis at gcc dot gnu dot org 2008-11-26 22:08 --- I'm about to take off for a long holiday weekand and haven't finished testing, but it fixes the test failures to copy mutex.h from libgomp/config/linux/ia64 to libgomp/config/linux/powerpc and update the comments. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38270
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #27 from janis at gcc dot gnu dot org 2008-11-26 22:16 --- Subject: Bug 28870 Author: janis Date: Wed Nov 26 22:15:07 2008 New Revision: 142230 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142230 Log: PR testsuite/28870 * lib/objc.exp (objc_target_compile): Fix typo from previous change. * lib/gnat.exp (gnat_target_compile): Fix typo from previous change. Modified: trunk/gcc/testsuite/ChangeLog trunk/gcc/testsuite/lib/gnat.exp trunk/gcc/testsuite/lib/objc.exp -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug c++/38253] g++.dg/ipa/iinline-1.C scan-ipa-dump inline fails on powerpc-apple-darwin9
--- Comment #4 from janis at gcc dot gnu dot org 2008-12-02 00:00 --- This test also fails on powerpc64-unknown-linux-gnu for both -m32 and -m64. It passed on 2008-08-24 (r138089) and failed on 2008-08-25 (r138121). Let me know if you'd like me to run a regression hunt and/or provide dump files. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38253
[Bug middle-end/38219] gcc.dg/tree-ssa/vrp47.c fails on powerpc-apple-darwin9
--- Comment #3 from janis at gcc dot gnu dot org 2008-12-02 00:05 --- This test also fails on powerpc64-unknown-linux-gnu for both -m32 and -m64. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38219
[Bug libgomp/38270] libgomp test failures due to missing memory barrier
--- Comment #4 from janis at gcc dot gnu dot org 2008-12-02 00:20 --- Subject: Bug 38270 Author: janis Date: Tue Dec 2 00:18:56 2008 New Revision: 142341 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142341 Log: PR libgomp/38270 * config/linux/powerpc/mutex.h: New. Added: trunk/libgomp/config/linux/powerpc/mutex.h Modified: trunk/libgomp/ChangeLog -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38270
[Bug libgomp/38270] libgomp test failures due to missing memory barrier
--- Comment #5 from janis at gcc dot gnu dot org 2008-12-02 00:50 --- Fixed. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38270
[Bug c++/38362] New: AltiVec context-sensitive keywords broke test for __vector bool
When the AltiVec context-sensitive keyword changes went in Andrew Pinski
reported in http://gcc.gnu.org/ml/gcc-patches/2008-07/msg01116.html that the
changes broke test g++.dg/ext/altivec-types-1.C. There was a brief discussion
between Ben Elliston and Ulrich Weigand about whether "__vector bool" should be
recognized as a valid AltiVec type (with "bool" having the AltiVec meaning
rather than the normal C++ meaning) when has not been included.
Please figure it out and fix either the compiler or the test.
--
Summary: AltiVec context-sensitive keywords broke test for
__vector bool
Product: gcc
Version: 4.4.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: janis at gcc dot gnu dot org
GCC host triplet: powerpc*-*-*
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38362
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #28 from janis at gcc dot gnu dot org 2008-12-02 17:45 --- Subject: Bug 28870 Author: janis Date: Tue Dec 2 17:44:08 2008 New Revision: 142366 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142366 Log: Backport from mainline: 2008-11-26 Janis Johnson <[EMAIL PROTECTED]> gcc/ PR testsuite/28870 * doc/sourcebuild.texi (Test Directives): Add dg-timeout and dg-timeout-factor. gcc/testsuite/ PR testsuite/28870 * lib/timeout.exp: New. * lib/timeout-dg.exp: New. * lib/gcc-dg.exp: Include new timeout library files. (dg-test): Unset timeout variables. * lib/gcc.exp (gcc_target_compile): Set timeout value from new proc. * lib/g++.exp (g++_target_compile): Ditto. * lib/gfortran.exp (gfortran_target_compile): Ditto. * lib/objc.exp (objc_target_compile): Ditto. * lib/obj-c++.exp (obj-c++_target_compile): Ditto. * lib/obj-c++.exp (obj-c++_target_compile): Ditto. * lib/gnat.exp (gnat_target_compile): Ditto. libstdc++-v3/ PR testsuite/28870 * testsuite/lib/libstdc++.exp: Include new timeout library files. (libstdc++_init): Define tool_timeout. (dg-test): Override DejaGnu proc. (v3_target_compile): Set timeout value from new proc. (v3_target_compile_as_c): Ditto. libmudflap/ PR testsuite/28870 * testsuite/lib/mfdg.exp (dg-test): Use new timeout support. (dg-timeout): Remove. (standard-wait): Remove. * testsuite/lib/libmudflap.exp: Include new timeout library files. (libmudflap_target_compile): Set timeout value from new proc. libgomp/ PR testsuite/28870 * testsuite/lib/libgomp.exp: Include new timeout library files. (libgomp_target_compile): Set timeout value from new proc. Added: branches/gcc-4_3-branch/gcc/testsuite/lib/timeout-dg.exp branches/gcc-4_3-branch/gcc/testsuite/lib/timeout.exp Modified: branches/gcc-4_3-branch/gcc/ChangeLog branches/gcc-4_3-branch/gcc/doc/sourcebuild.texi branches/gcc-4_3-branch/gcc/testsuite/ChangeLog branches/gcc-4_3-branch/gcc/testsuite/lib/g++.exp branches/gcc-4_3-branch/gcc/testsuite/lib/gcc-dg.exp branches/gcc-4_3-branch/gcc/testsuite/lib/gcc.exp branches/gcc-4_3-branch/gcc/testsuite/lib/gfortran.exp branches/gcc-4_3-branch/gcc/testsuite/lib/gnat.exp branches/gcc-4_3-branch/gcc/testsuite/lib/obj-c++.exp branches/gcc-4_3-branch/gcc/testsuite/lib/objc.exp branches/gcc-4_3-branch/libgomp/ChangeLog branches/gcc-4_3-branch/libgomp/testsuite/lib/libgomp.exp branches/gcc-4_3-branch/libmudflap/ChangeLog branches/gcc-4_3-branch/libmudflap/testsuite/lib/libmudflap.exp branches/gcc-4_3-branch/libmudflap/testsuite/lib/mfdg.exp branches/gcc-4_3-branch/libstdc++-v3/ChangeLog branches/gcc-4_3-branch/libstdc++-v3/testsuite/lib/libstdc++.exp -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug testsuite/28870] [4.2/4.3/4.4 Regression] configuring, over-riding timeout values in testsuite
--- Comment #29 from janis at gcc dot gnu dot org 2008-12-02 17:49 --- Fixed in trunk (4.4) and 4.3; 4.2 isn't expected to have additional releases so I haven't backported it there. The patch solves the original reported problem, a hard-coded timeout for libstdc++ tests, by allowing the timeout value for a testsuite to be overridden by "gcc,,timeout" in the target board, and for individual tests to specify a timeout either by a number of seconds or by a multiple of the current default timeout. The same mechanism is also used now by the compiler tests and tests for libmudflap and libgomp; it can be extended to libffi and libjava if there is interest. -- janis at gcc dot gnu dot org changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution||FIXED http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28870
[Bug testsuite/29056] gcc.target/powerpc/ppc-negeq0-1.c fails on powerpc64
--- Comment #10 from janis at gcc dot gnu dot org 2008-12-02 22:13 --- Subject: Bug 29056 Author: janis Date: Tue Dec 2 22:11:55 2008 New Revision: 142374 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142374 Log: Backport from mainline: 2008-08-26 Janis Johnson <[EMAIL PROTECTED]> * gcc.dg/Wstrict-aliasing-bogus-ref-all-2.c: Ignore a warning. * g++.dg/ext/altivec-3.C: Move AltiVec code out of main. Backport from mainline: 2008-07-10 2008-07-10 Joseph Myers <[EMAIL PROTECTED]> PR middle-end/29056 * gcc.target/powerpc/ppc-negeq0-1.c: Use long instead of int. Adjust shift and scan-assembler-not pattern to allow for 64-bit case. Backport from mainline: 2008-06-09 Joseph Myers <[EMAIL PROTECTED]> * gcc.dg/pr34856.c: Condition use of -maltivec on powerpc_altivec_ok. Use -w on other powerpc*-*-linux*. Backport from mainline: 2008-05-20 Janis Johnson <[EMAIL PROTECTED]> * g++.dg/ext/vector14.C: Ignore a possible warning. Backport from mainline: 2008-04-21 Steve Ellcey <[EMAIL PROTECTED]> * gcc.dg/vect/pr33953.c: XFAIL if vect_no_align. Backport from mainline: 2008-04-04 Janis Johnson <[EMAIL PROTECTED]> * gcc.dg/var-expand3.c: Skip for powerpc-linux if not on AltiVec HW. * gcc.dg/torture/builtin-modf-1.c: Use special options for powerpc*-*-linux*. Backport from mainline: 2008-04-04 Janis Johnson <[EMAIL PROTECTED]> * g++.dg/other/anon5.C: Don't depend on line number for error message. Modified: branches/gcc-4_3-branch/gcc/testsuite/ChangeLog branches/gcc-4_3-branch/gcc/testsuite/g++.dg/ext/altivec-3.C branches/gcc-4_3-branch/gcc/testsuite/g++.dg/ext/vector14.C branches/gcc-4_3-branch/gcc/testsuite/g++.dg/other/anon5.C branches/gcc-4_3-branch/gcc/testsuite/gcc.dg/Wstrict-aliasing-bogus-ref-all-2.c branches/gcc-4_3-branch/gcc/testsuite/gcc.dg/pr34856.c branches/gcc-4_3-branch/gcc/testsuite/gcc.dg/torture/builtin-modf-1.c branches/gcc-4_3-branch/gcc/testsuite/gcc.dg/var-expand3.c branches/gcc-4_3-branch/gcc/testsuite/gcc.dg/vect/pr33953.c branches/gcc-4_3-branch/gcc/testsuite/gcc.target/powerpc/ppc-negeq0-1.c -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=29056
[Bug tree-optimization/31677] gcc.dg/memcpy-1.c tree dump scan fails on several targets
--- Comment #2 from janis at gcc dot gnu dot org 2008-12-02 22:28 --- Adding "--param sra-max-structure-size=32" as suggested in comment #1 makes the test pass on powerpc64-linux with -m32 and -m64. Jan, is that an appropriate fix/workaround? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31677
