World first Patch Technology for penis Enlargement
Pleasure your women - size does matter! http://www.gretan.com/ss/ The great discoveries are usually obvious. Everything has its beauty but not everyone sees it. Education is a better safeguard of liberty than a standing army. Honor does not have to be defended. Knowledge comes, but wisdom lingers. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Welcome to VIP Quality Software.
Cheap Software-Upgrades-educational bargains-oem-Software Buyer's Guide. http://hpb.el0thve7b6e3bxw.spurternj.com If my answers frighten you then you should cease asking scary questions. It is worse still to be ignorant of your ignorance. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
eDrugs Online
Pharmaceutical services for the modern men http://uhyw.5k8g9ong2fnvron.trophemahi.info You can not apply mathematics as long as words still becloud reality. If you obey all the rules, you miss all the fun. I am just going outside and may be some time. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#452108: what binary packages to update?
> Note that this issue is in theory pretty serious as in principle _any_ > binary package in the archive could be affected. Since the abs() is optimized away on miscompilations, searching for binaries that have the abs() symbol won't find the buggy ones. The only reliable way to find out what packages could be affected, is to grep for calls to abs() [I guess too many to hand-review]. Then the list could be filtered by dropping calls to abs() where parameteres are provably positive numbers (either constants, or unsigned types). Such a check could be implemented in static analysis tools, like CIL, but trying to compile the entire archive with CIL can be a challenge by itself. How do you intend to handle this bug for stable? Obviously updating all packages that call abs() is not a solution :( A quicker way to find out could be to compile all the archive with current gcc version. Then apply the patch that fixes the bug, recompile the entire archive, and see which .o,.so,.a, and executable files differ between the two. Is there a tool to automate such a task? --Edwin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#470557: smaller testcase for gcc misoptimization bug
Package: gcc-4.2
Version: 4.2.3-2
--- Please enter the report below this line. ---
I have manually reduced the previous testcase (it is 4K now, vs 32K).
The same testcase can be used for both x86_64 and x86(-32).
When I compile with -O3, or -O2 -finline-functions incorrect code is
generated.
$ gcc-4.2 -O2 -finline-functions gcc42_O3_bug.i
$ ./a.out
Aborted
$ gcc-4.2 -O2 gcc42_O3_bug.i
$ ./a.out
$ valgrind ./a.out
==11486== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 8 from 1)
.
$ sh -x ./test.sh gcc42_O3_bug.i
+ gcc-4.3 -fstack-protector-all -O1 -Wfatal-errors -Werror
-Wno-unused gcc42_O3_bug.i -o works3
+ test 0 == 0
+ ulimit -t 1
+ ./works3
+ test 0 == 0
+ gcc-4.2 -Wall -Wfatal-errors -Werror -Wno-unused -O3
gcc42_O3_bug.i -o fails
+ test 0 == 0
+ ulimit -t 1
+ ./fails
./test.sh: line 24: 11426 Aborted ./fails 2>/dev/null
+ test 134 == 134
+ exit 0
If you need more info, please ask.
--- System information. ---
Architecture: amd64
Kernel: Linux 2.6.25-rc4-00134-g84c6f60
Debian Release: lenny/sid
500 unstablewww.debian-multimedia.org
500 unstableftp.iasi.roedu.net
500 testing ftp.iasi.roedu.net
500 gutsy wine.budgetdedicated.com
1 experimentalftp.iasi.roedu.net
--- Package information. ---
Depends (Version) | Installed
=-+-=
binutils (>= 2.17cvs20070426) | 2.18.1~cvs20080103-1
cpp-4.2 (= 4.2.3-2) | 4.2.3-2
gcc-4.2-base (= 4.2.3-2) | 4.2.3-2
libc6 (>= 2.7-1) | 2.7-9
libgcc1(>= 1:4.2.3-2) | 1:4.3.0-1
typedef unsigned long int uint64_t;
void cli_dbgmsg_internal(const char *str,...) {
}
typedef long unsigned int size_t;
extern int memcmp (__const void *__s1, __const void *__s2, size_t __n)
__attribute__ ((__nothrow__)) __attribute__ ((__pure__)) __attribute__
((__nonnull__ (1, 2)));
extern void abort (void) __attribute__ ((__nothrow__)) __attribute__
((__noreturn__));
unsigned char data[] = {
0x50, 0x4d, 0x47, 0x4c, 0x38, 0x01, 0x00, 0x00, 0x63, 0x00, 0x00,
0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0x01, 0x2f, 0x00, 0x00,
0x00, 0x06, 0x2f, 0x23, 0x42, 0x53, 0x53, 0x43, 0x01, 0x88, 0x90, 0x79, 0x09,
0x08, 0x2f, 0x23, 0x49, 0x44, 0x58, 0x48, 0x44, 0x52, 0x01, 0xc0, };
typedef long int __off_t;
typedef __off_t off_t;
typedef unsigned char uint8_t;
typedef unsigned short int uint16_t;
typedef unsigned int uint32_t;
uint8_t cli_debug_flag=0, cli_leavetemps_flag;
typedef struct chm_itsf_header_tag {
uint32_t block_len __attribute__ ((packed));
}
chm_itsp_header_t;
typedef struct chm_metadata_tag {
char *m_area;
chm_itsp_header_t itsp_hdr;
off_t chunk_offset;
char *chunk_data;
char *chunk_current;
char *chunk_end;
uint16_t chunk_entries;
}
chm_metadata_t;
int cli_chm_open( chm_metadata_t *metadata);
int cli_chm_prepare_file(int fd, chm_metadata_t *metadata);
static uint64_t read_enc_int(char **start, char *end) {
uint64_t retval=0;
char *current;
current = *start;
do {
retval = (retval << 7) | (*current & 0x7f);
}
while (*current++ & 0x80);
*start = current;
return retval;
}
static int prepare_file(int fd, chm_metadata_t *metadata) {
uint64_t name_len, section;
while (metadata->chunk_entries != 0) {
if (metadata->chunk_current >= metadata->chunk_end) {
}
name_len = read_enc_int(&metadata->chunk_current,
metadata->chunk_end);
if (((metadata->chunk_current + name_len) >=
metadata->chunk_end) || ((metadata->chunk_current + name_len) <
metadata->chunk_data)) {
(!__builtin_expect(!!(cli_debug_flag),
0)) ? (void)0 : cli_dbgmsg_internal("Bad CHM name_len
detected\n");
}
metadata->chunk_current += name_len;
section = read_enc_int(&metadata->chunk_current,
metadata->chunk_end);
if(section == 47) {
abort();
}
if (section == 1) {
return 0;
}
}
return 0;
}
static int read_chunk(chm_metadata_t *metadata, int fd) {
if (metadata->m_area != ((void *)0)) {
metadata->chunk_data = metadata->m_area +
metadata->chunk_offset;
}
Bug#512686: libffi-dev: manpage #include confused
Package: libffi-dev Version: 3.0.7-1 Severity: normal man ffi_call synposis says , but example uses . Manpage should be consistent and use only , if that is correct for Debian. -- System Information: Debian Release: 5.0 APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.28 (SMP w/4 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages libffi-dev depends on: ii libffi5 3.0.7-1Foreign Function Interface library libffi-dev recommends no packages. libffi-dev suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#592664: [PR45260]: g++-4.5 -fprefetch-loop-arrays internal compiler error
Package: g++-4.5 Version: 4.5.1-1 Severity: normal Tags: upstream FYI I reported a gcc-4.5 bug here with full details here: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45260 It was initially reported against ClamAV here: https://wwws.clamav.net/bugzilla/show_bug.cgi?id=2190 GCC gives an internal compiler error when compiling ClamAV with -O2 -fprefetch-loop-arrays CXXFLAGS: llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp: In member function 'void llvm::TargetLowering::computeRegisterProperties()': llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp:608:6: internal compiler error: in expand_expr_addr_expr_1, at expr.c:6936 make[5]: *** [TargetLowering.lo] Error 1 -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.35-phenom (SMP w/6 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages g++-4.5 depends on: ii gcc-4.5 4.5.1-1The GNU C compiler ii gcc-4.5-base 4.5.1-1The GNU Compiler Collection (base ii libc6 2.11.2-2 Embedded GNU C Library: Shared lib ii libcloog-ppl0 0.15.9-1 the Chunky Loop Generator (runtime ii libelfg0 0.8.13-1 an ELF object file access library ii libgmp3c2 2:4.3.2+dfsg-1 Multiprecision arithmetic library ii libgmpxx4ldbl 2:4.3.2+dfsg-1 Multiprecision arithmetic library ii libmpc2 0.8.2-1+b1 multiple precision complex floatin ii libmpfr4 3.0.0-2multiple precision floating-point ii libppl-c2 0.10.2-7 Parma Polyhedra Library (C interfa ii libppl7 0.10.2-7 Parma Polyhedra Library (runtime l ii libstdc++6-4.5-dev4.5.1-1The GNU Standard C++ Library v3 (d ii zlib1g1:1.2.5.dfsg-1 compression library - runtime g++-4.5 recommends no packages. Versions of packages g++-4.5 suggests: ii g++-4.5-multilib 4.5.1-1The GNU C++ compiler (multilib fil pn gcc-4.5-doc(no description available) ii libstdc++6-4.5-dbg4.5.1-1The GNU Standard C++ Library v3 (d -- no debconf information -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100811203702.28112.43796.report...@deb0
Bug#596965: gcc-4.4: crash on vector code with -m32 -msse -O2
Package: gcc-4.4 Version: 4.4.4-14 Severity: normal Tags: upstream Forwarded: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45678 I just reported this bug in gcc bugzilla: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45678 It happens when using gcc version 4.4.5 20100909 (prerelease) (Debian 4.4.4-14), it doesn't happen with gcc-4.3 and gcc-4.5. -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.36-rc4-phenom (SMP w/6 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages gcc-4.4 depends on: ii binutils 2.20.51.20100908-1 The GNU assembler, linker and bina ii cpp-4.4 4.4.4-14 The GNU C preprocessor ii gcc-4.4-base 4.4.4-14 The GNU Compiler Collection (base ii libc6 2.11.2-5 Embedded GNU C Library: Shared lib ii libgcc1 1:4.5.1-2 GCC support library ii libgomp1 4.5.1-2GCC OpenMP (GOMP) support library Versions of packages gcc-4.4 recommends: ii libc6-dev 2.11.2-5 Embedded GNU C Library: Developmen Versions of packages gcc-4.4 suggests: ii gcc-4.4-doc 4.4.4.nf1-1 documentation for the GNU compiler ii gcc-4.4-locales 4.4.4-14The GNU C compiler (native languag ii gcc-4.4-multilib 4.4.4-14The GNU C compiler (multilib files ii libcloog-ppl00.15.9-2the Chunky Loop Generator (runtime ii libgcc1-dbg 1:4.5.1-2 GCC support library (debug symbols ii libgomp1-dbg 4.5.1-2 GCC OpenMP (GOMP) support library ii libmudflap0-4.4-dev 4.4.4-14GCC mudflap support libraries (dev ii libmudflap0-dbg 4.5.1-2 GCC mudflap shared support librari ii libppl-c20.10.2-7Parma Polyhedra Library (C interfa ii libppl7 0.10.2-7Parma Polyhedra Library (runtime l -- debconf-show failed -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100915124726.4383.68851.report...@deb0
Bug#619963: testcase
Hi,
Attached is a simple testcase from VLC.
At -O0 everything is fine:
$ gcc-4.6 -S -O0 x.i
$ grep memmove x.s
call__memmove_chk
At -O1 it incorrectly it is incorrectly changed to memcpy:
$ gcc-4.6 -S -O1 x.i
$ grep memcpy x.s
callmemcpy
If I change the provided x.i, and remove the extern inline declaration
of memmove, then memmove is no longer changed to memcpy!
Also all is good with gcc-4.5:
$ gcc-4.5 -S -O2 x.i
$ grep memmove x.s
callmemmove
Poking around in gcc-4.6 sources this place in gcc/builtins.c seems to
be the place where memmove_chk -> memcpy conversion happens, I didn't
verify that though:
if (fcode == BUILT_IN_MEMMOVE_CHK)
{
unsigned int src_align
= get_pointer_alignment (src, BIGGEST_ALIGNMENT);
if (src_align == 0)
return NULL_RTX;
/* If src is categorized for a readonly section we can use
normal __memcpy_chk. */
if (readonly_data_expr (src))
{
tree fn = built_in_decls[BUILT_IN_MEMCPY_CHK];
Best regards,
--Edwin
typedef long unsigned int size_t;
extern __inline __attribute__ ((__always_inline__)) __attribute__
((__gnu_inline__, __artificial__)) void * __attribute__ ((__nothrow__)) memmove
(void *__restrict __dest, __const void *__restrict __src, size_t __len) {
return __builtin___memmove_chk (__dest, __src, __len,
__builtin_object_size (__dest, 0));
}
typedef struct aout_instance_t aout_instance_t;
typedef struct aout_input_t aout_input_t;
struct aout_instance_t {
aout_input_t * pp_inputs[5];
int i_nb_inputs;
};
int aout_DecDelete( aout_instance_t * p_aout, aout_input_t * p_input ) {
int i_input = p_aout->i_nb_inputs;
memmove( &p_aout->pp_inputs[i_input], &p_aout->pp_inputs[i_input + 1], (5 -
i_input - 1) * sizeof(aout_input_t *) );
}
Bug#619963: memmove
Marc Glisse wrote: > Hello, > > in the prototype of memmove (taken from glibc bits/string3.h), you will > notice __restrict in front of the arguments. This is not present in more > recent glibc versions and is a likely cause of your trouble. > Indeed if I remove __restrict the bug is gone, and if I upgrade to libc6 2.13-0exp5, VLC's dec.c seems to be compiled correctly too. Should gcc-4.6 depend on glibc >= 2.13? Or should gcc-4.6 provide a fixed prototype for memmove via fix-includes? Best regards, --Edwin -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4da175f3.2000...@gmail.com
Bug#541801: lib32mudflap0: circular dependency
Package: lib32mudflap0 Version: 4.4.1-1 Severity: normal I got this today after running aptitude dist-upgrade (after returning from a 2 week holiday): Writing extended state information... Done E: Couldn't configure pre-depend libc6-i386 for lib32mudflap0, probably a dependency cycle. A package failed to install. Trying to recover: Reading package lists... Done Building dependency tree Reading state information... Done Reading extended state information Initializing package states... Done Writing extended state information... Done Reading task descriptions... Done Ditto for lib32stdc++: Writing extended state information... Done E: Couldn't configure pre-depend libc6-i386 for lib32stdc++6, probably a dependency cycle. And lib32gcc1: Writing extended state information... Done E: Couldn't configure pre-depend libc6-i386 for lib32gcc1, probably a dependency cycle. After removing these 3 packages, I can do the upgrade. Am I doing something wrong here? -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (990, 'unstable'), (500, 'stable'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.30.1 (SMP w/4 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages lib32mudflap0 depends on: ii gcc-4.4-base 4.4.1-1The GNU Compiler Collection (base ii libc6-i3862.9-20 GNU C Library: 32-bit shared libra lib32mudflap0 recommends no packages. lib32mudflap0 suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#575448: please package gcc-4.5's gold LTO plugin!
Package: gcc-4.5 Version: 4.5-20100227-1 Severity: wishlist GCC-4.5 has a gold linker plugin to do LTO, see http://gcc.gnu.org/wiki/whopr/driver It should already be built during a normal gcc-4.5 build, the plugin should be called liblto-plugin.so. I can't find that in the Debian experimental package, please add it to the gcc-4.5 package (or create a new package for it). It should allow more LTO optimizations than just -flto alone. -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.33-00175-g2d1299a0 (SMP w/4 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages gcc-4.5 depends on: ii binutils2.20.1-3 The GNU assembler, linker and bina ii cpp-4.5 4.5-20100227-1 The GNU C preprocessor ii gcc-4.5-base4.5-20100227-1 The GNU Compiler Collection (base ii libc6 2.10.2-6 Embedded GNU C Library: Shared lib ii libcloog-ppl0 0.15.8-1 the Chunky Loop Generator (runtime ii libelfg00.8.12-0.1 an ELF object file access library ii libgcc1 1:4.5-20100227-1 GCC support library ii libgmp3c2 2:4.3.2+dfsg-1 Multiprecision arithmetic library ii libgmpxx4ldbl 2:4.3.2+dfsg-1 Multiprecision arithmetic library ii libgomp14.5-20100227-1 GCC OpenMP (GOMP) support library ii libmpc2 0.8.1-1 multiple precision complex floatin ii libmpfr1ldbl2.4.2-3 multiple precision floating-point ii libppl-c2 0.10.2-6 Parma Polyhedra Library (C interfa ii libppl7 0.10.2-6 Parma Polyhedra Library (runtime l ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime Versions of packages gcc-4.5 recommends: ii libc6-dev 2.10.2-6 Embedded GNU C Library: Developmen Versions of packages gcc-4.5 suggests: pn gcc-4.5-doc(no description available) pn gcc-4.5-locales(no description available) pn gcc-4.5-multilib (no description available) pn libgcc1-dbg(no description available) pn libgomp1-dbg (no description available) pn libmudflap0-4.5-dev(no description available) pn libmudflap0-dbg(no description available) -- no debconf information -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100325211915.30163.68846.report...@debian
Bug#575451: gcc-4.5: -flto doesn't remember -fPIC
Package: gcc-4.5
Version: 4.5-20100227-1
Severity: normal
Tags: experimental
gcc-4.5 with -flto doesn't work with -fPIC properly.
Simple testcase:
int entry(int a)
{
return bar(a)+1;
}
int bar(int a)
{
return a+4;
}
$ gcc-4.5 foo1.c -flto -fPIC -DPIC -c
$ gcc-4.5 foo2.c -flto -fPIC -DPIC -c
$ gcc-4.5 foo1.o foo2.o -flto -shared
/usr/bin/ld: /tmp/ccmA7RCK.lto.o: relocation R_X86_64_PC32 against symbol `bar'
can not be used when making a shared object; recompile with -fPIC
/usr/bin/ld: final link failed: Bad value
collect2: ld returned 1 exit status
This works however:
$ gcc-4.5 foo1.o foo2.o -flto -shared -fPIC
Now -fPIC is something libtool automatically adds, and I it doesn't add it at
linktime (perhaps other build systems don't either).
Could gcc's -flto see that all .o files involved in the link are -fPIC... and
make the resulting file -fPIC too?
Or at least it should see the -shared in the linker line, and automatically use
-fPIC when -flto is used.
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.33-00175-g2d1299a0 (SMP w/4 CPU cores; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Versions of packages gcc-4.5 depends on:
ii binutils2.20.1-3 The GNU assembler, linker and bina
ii cpp-4.5 4.5-20100227-1 The GNU C preprocessor
ii gcc-4.5-base4.5-20100227-1 The GNU Compiler Collection (base
ii libc6 2.10.2-6 Embedded GNU C Library: Shared lib
ii libcloog-ppl0 0.15.8-1 the Chunky Loop Generator (runtime
ii libelfg00.8.12-0.1 an ELF object file access library
ii libgcc1 1:4.5-20100227-1 GCC support library
ii libgmp3c2 2:4.3.2+dfsg-1 Multiprecision arithmetic library
ii libgmpxx4ldbl 2:4.3.2+dfsg-1 Multiprecision arithmetic library
ii libgomp14.5-20100227-1 GCC OpenMP (GOMP) support library
ii libmpc2 0.8.1-1 multiple precision complex floatin
ii libmpfr1ldbl2.4.2-3 multiple precision floating-point
ii libppl-c2 0.10.2-6 Parma Polyhedra Library (C interfa
ii libppl7 0.10.2-6 Parma Polyhedra Library (runtime l
ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime
Versions of packages gcc-4.5 recommends:
ii libc6-dev 2.10.2-6 Embedded GNU C Library: Developmen
Versions of packages gcc-4.5 suggests:
pn gcc-4.5-doc(no description available)
pn gcc-4.5-locales(no description available)
pn gcc-4.5-multilib (no description available)
pn libgcc1-dbg(no description available)
pn libgomp1-dbg (no description available)
pn libmudflap0-4.5-dev(no description available)
pn libmudflap0-dbg(no description available)
-- no debconf information
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100325213843.7735.41797.report...@debian
Bug#575448: please package gcc-4.5's gold LTO plugin!
On 2010-03-26 01:27, Matthias Klose wrote: > On 25.03.2010 22:19, Török Edwin wrote: >> Package: gcc-4.5 >> Version: 4.5-20100227-1 >> Severity: wishlist >> >> GCC-4.5 has a gold linker plugin to do LTO, see >> http://gcc.gnu.org/wiki/whopr/driver >> >> It should already be built during a normal gcc-4.5 build, >> the plugin should be called liblto-plugin.so. >> >> I can't find that in the Debian experimental package, please add it to >> the >> gcc-4.5 package (or create a new package for it). > > It was built before, you can enable it in a local build. Last time I > built using gold, the build went wrong: #571169. Seems to be working now, I am using this binutils-gold package though: 2.20.1-3, not the one from experimental. I commented out this line in debian/rules.defs # FIXME: stripping doesn't work with gold #with_gold := no I disabled the testsuite too (was taking way too long): # if you don't want to run the gcc testsuite, uncomment the next line with_check := disabled by hand Also I had some trouble with building the doc packages, so I just did a 'touch stamps/05-build-javadoc-stamp', didn't figure out how to turn off the build of the docs package completely. I installed these debs: ii cpp-4.5 4.5-20100321-1 The GNU C preprocessor ii gcc-4.5 4.5-20100321-1 The GNU C compiler ii gcc-4.5-base 4.5-20100321-1 The GNU Compiler Collection (base package) ii gcc-4.5-multilib 4.5-20100321-1 The GNU C compiler (multilib files) ii lib32gcc1 1:4.5-20100321-1 GCC support library (32 bit Version) ii lib32gomp14.5-20100321-1 GCC OpenMP (GOMP) support library (32bit) ii libgcc1 1:4.5-20100321-1 GCC support library ii libgomp1 4.5-20100321-1 GCC OpenMP (GOMP) support library ii libstdc++64.5-20100321-1 The GNU Standard C++ Library v3 apt-cache and everything else still seems to be working. I found the LTO plugin here (it was not installed by the package) ./debian/tmp/usr/lib/gcc/x86_64-linux-gnu/4.5/liblto_plugin.so So I copied it to /usr/lib/gcc/x86_64-linux-gnu/4.5/liblto_plugin.so manally, and -fuse-linker-plugin seems to be working now (well I get an ICE from it when trying to build ClamAV, will open a bug about that shortly). Can you please make the package install liblto_plugin.so when with_gold is yes? > If you retry, please > report the results in this report. I'm not sure I understand what 'this' refers to here (#575448 or #571169), so I'll just report it here. Best regards, --Edwin -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4bac9d29.8030...@gmail.com
Bug#575448: please package gcc-4.5's gold LTO plugin!
On 2010-03-26 13:40, Török Edwin wrote: > I found the LTO plugin here (it was not installed by the package) > ./debian/tmp/usr/lib/gcc/x86_64-linux-gnu/4.5/liblto_plugin.so > > So I copied it to /usr/lib/gcc/x86_64-linux-gnu/4.5/liblto_plugin.so > manally, and -fuse-linker-plugin seems to be working now > (well I get an ICE from it when trying to build ClamAV, will open a bug > about that shortly). ... and its not working, see http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42453 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=41584 I hit that bug while building libclamav with -flto -fuse-linker-plugin (after hacking libtool to pass the flags). Until those bugs are not fixed, I think you can leave gcc-4.5 package as is (without LTO plugin). Best regards, --Edwin -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4baca727.5030...@gmail.com
Bug#577621: gcc-4.4-plugin-dev: accepts non-GPL plugins
Package: gcc-4.4-plugin-dev
Version: 4.4.3-7
Severity: normal
According to the GCC plugin documentation plugins should declare a symbol
plugin_is_GPL_compatible.
However the gcc-4.4 in Debian loads the plugin regardless.
To reproduce:
$ cat >testcase.c <
int plugin_init(struct plugin_name_args *plugin_info,
struct plugin_gcc_version *version)
{ return 0; }
EOF
$ gcc -fPIC -DPIC -shared -o testcase.so testcase.c -I`gcc
-print-file-name=plugin`/include
$ gcc -fplugin=./testcase.so -c -x c /dev/null
$ echo $?
0
gcc-4.5 correctly rejects the plugin:
$ gcc-4.5 -fplugin=./testcase.so -c -x c /dev/null
cc1: fatal error: plugin ./testcase.so is not licensed under a GPL-compatible
license
./testcase.so: undefined symbol: plugin_is_GPL_compatible
compilation terminated.
I think this is a bug in Debian's backport of the plugin to gcc-4.4.
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.34-rc3-00138-gecb385a (SMP w/4 CPU cores; PREEMPT)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Versions of packages gcc-4.4-plugin-dev depends on:
ii gcc-4.4 4.4.3-7The GNU C compiler
ii gcc-4.4-base 4.4.3-7The GNU Compiler Collection (base
ii libgmp3-dev 2:4.3.2+dfsg-1 Multiprecision arithmetic library
gcc-4.4-plugin-dev recommends no packages.
gcc-4.4-plugin-dev suggests no packages.
-- no debconf information
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20100413073733.12104.49927.report...@debian
Bug#583141: gcc-doc: claims "Invariant Sections" in the manpage that are not actually present: isn't the manpage DFSG free?
Package: gcc-doc Version: 5:2 Severity: normal The gcc manpage says "with the Invariant Sections being "GNU General Public License" and "Funding Free Software"", however neither "GNU General Public License", nor "Funding Free Software" appears in the manpage itself, hence the manpage can be considered as not having any invariant sections. The DFSG FAQ says "Data licensed under the FDL with no invariant sections are considered DFSG-free as of GR 2006-001". Could the manpage be moved back to the main archive? (probably by putting it into the gcc package itself, or into another package?) -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (990, 'unstable'), (500, 'testing'), (1, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.34-rc5-00118-gf259493 (SMP w/4 CPU cores; PREEMPT) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages gcc-doc depends on: ii gcc-4.3-doc 4.3.4.nf1-1 documentation for the GNU compiler gcc-doc recommends no packages. gcc-doc suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100525175819.392.24757.report...@debian
Bug#336915: gcc-4.0: Incorrect "Warning statement has no effect" when using a GNU extension (possible regression)
Package: gcc-4.0
Version: 4.0.2-3
Severity: normal
When I compile the following program like this:
gcc -Wall t2.cpp
I get this warning:
t2.cpp: In function 'int x(T)':
t2.cpp:11: warning: statement has no effect
t2.cpp: In function 'int x3(T)':
t2.cpp:25: warning: statement has no effect
When I compile the program with gcc-3.4 I get no warning.
The statement DOES have an effect, its value is the value of the ({ ... })
parantheses.
Without it would give a syntax error.
The code in the return statement was taken from standard header files
(bits/byteswap.h)
Here it is the file t2.cpp:
template
int x(T t)
{
return (__extension__ (
{
register unsigned short int __v, __x = (t);
if (__builtin_constant_p (__x))
__v = __x) >> 8) & 0xff) | (((__x) & 0xff) <<
8));
else
__asm__ ("rorw $8, %w0" : "=r" (__v) : "0" (__x) :
"cc");
__v;
}
));
}
template
int x3(T t)
{
return (
{
register unsigned short int __v, __x = (t);
if (__builtin_constant_p (__x))
__v = __x) >> 8) & 0xff) | (((__x) & 0xff) <<
8));
else
__asm__ ("rorw $8, %w0" : "=r" (__v) : "0" (__x) :
"cc");
__v;
});
}
template
int x2(T t)
{
register unsigned short int __v, __x = (t);
if (__builtin_constant_p (__x))
__v = __x) >> 8) & 0xff) | (((__x) & 0xff) << 8));
else
__asm__ ("rorw $8, %w0" : "=r" (__v) : "0" (__x) : "cc");
return __v;
}
int main()
{
int z=4,y,y2,y3;
y=x(z);
y2=x2(z);
y3=x3(z);
return 0;
}
Some info about gcc: it is from debian unstable, latest.
Output of gcc -v --save-temp -Wall t2.cpp:
Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,java,f95,objc,ada,treelang --prefix=/usr
--enable-shared --with-system-zlib --libexecdir=/usr/lib
--without-included-gettext --enable-threads=posix --enable-nls
--program-suffix=-4.0 --enable-__cxa_atexit --enable-libstdcxx-allocator=mt
--enable-clocale=gnu --enable-libstdcxx-debug --enable-java-gc=boehm
--enable-java-awt=gtk --enable-gtk-cairo
--with-java-home=/usr/lib/jvm/java-1.4.2-gcj-4.0-1.4.2.0/jre --enable-mpfr
--disable-werror --enable-checking=release i486-linux-gnu
Thread model: posix
gcc version 4.0.3 20051023 (prerelease) (Debian 4.0.2-3)
/usr/lib/gcc/i486-linux-gnu/4.0.3/cc1plus -E -quiet -v -D_GNU_SOURCE t2.cpp
-mtune=i486 -Wall -fpch-preprocess -o t2.ii
ignoring nonexistent directory "/usr/local/include/i486-linux-gnu"
ignoring nonexistent directory "/usr/include/i486-linux-gnu"
#include "..." search starts here:
#include <...> search starts here:
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../include/c++/4.0.3
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../include/c++/4.0.3/i486-linux-gnu
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../include/c++/4.0.3/backward
/usr/local/include
/usr/lib/gcc/i486-linux-gnu/4.0.3/include
/usr/include
End of search list.
/usr/lib/gcc/i486-linux-gnu/4.0.3/cc1plus -fpreprocessed t2.ii -quiet
-dumpbase t2.cpp -mtune=i486 -auxbase t2 -Wall -version -o t2.s
GNU C++ version 4.0.3 20051023 (prerelease) (Debian 4.0.2-3) (i486-linux-gnu)
compiled by GNU C version 4.0.3 20051023 (prerelease) (Debian 4.0.2-3).
GGC heuristics: --param ggc-min-expand=62 --param ggc-min-heapsize=60528
t2.cpp: In function 'int x(T)':
t2.cpp:11: warning: statement has no effect
t2.cpp: In function 'int x3(T)':
t2.cpp:25: warning: statement has no effect
as -V -Qy --32 -o t2.o t2.s
GNU assembler version 2.16.91 (i486-linux-gnu) using BFD version 2.16.91
20050902 Debian GNU/Linux
/usr/lib/gcc/i486-linux-gnu/4.0.3/collect2 --eh-frame-hdr -m elf_i386
-dynamic-linker /lib/ld-linux.so.2
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../lib/crt1.o
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../lib/crti.o
/usr/lib/gcc/i486-linux-gnu/4.0.3/crtbegin.o
-L/usr/lib/gcc/i486-linux-gnu/4.0.3 -L/usr/lib/gcc/i486-linux-gnu/4.0.3
-L/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../lib
-L/usr/lib/gcc/i486-linux-gnu/4.0.3/../../.. -L/lib/../lib -L/usr/lib/../lib
t2.o -lgcc --as-needed -lgcc_s --no-as-needed -lc -lgcc --as-needed -lgcc_s
--no-as-needed /usr/lib/gcc/i486-linux-gnu/4.0.3/crtend.o
/usr/lib/gcc/i486-linux-gnu/4.0.3/../../../../lib/crtn.o
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: i386 (i686)
Shell: /bin/sh linked to /bin/dash
Kernel: Linux 2.6.14-1-k7
Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1)
Versions of packages gcc-4.0 depends on:
ii binutils
Bug#703229: [libasan0] libasan0: /usr/bin/ld.bfd.real: cannot find libasan_preinit.o: No such file or directory
Package: libasan0 Version: 4.8-20130315-1 Severity: normal --- Please enter the report below this line. --- GCC-4.8's(from experimental) -fsanitize=address flag is not working: $ gcc-4.8 -fsanitize=address test.c /usr/bin/ld.bfd.real: cannot find libasan_preinit.o: No such file or directory collect2: error: ld returned 1 exit status $ apt-cache policy libasan0 libasan0: Installed: 4.8-20130315-1 Candidate: 4.8-20130315-1 Version table: *** 4.8-20130315-1 0 1 http://ftp.ro.debian.org/debian/ experimental/main amd64 Packages 100 /var/lib/dpkg/status $ dpkg -L libasan0 /. /usr /usr/lib /usr/lib/x86_64-linux-gnu /usr/lib/x86_64-linux-gnu/libasan.so.0.0.0 /usr/share /usr/share/doc /usr/lib/x86_64-linux-gnu/libasan.so.0 /usr/share/doc/libasan0 --- System information. --- Architecture: amd64 Kernel: Linux 3.8.3 Debian Release: 7.0 500 unstableftp.ro.debian.org 500 stable ftp.ro.debian.org 1 experimentalftp.ro.debian.org --- Package information. --- Depends (Version) | Installed ===-+-=== gcc-4.8-base (= 4.8-20130315-1) | libc6(>= 2.2.5) | libgcc1(>= 1:4.1.1) | libstdc++6 (>= 4.1.1) | Package's Recommends field is empty. Package's Suggests field is empty. -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/5145a2a3.5050...@etorok.net
Bug#703230: [gcc-4.8] gcc-4.8: -fsanitize=thread requires -fPIC too
Package: gcc-4.8 Version: 4.8-20130315-1 Severity: normal --- Please enter the report below this line. --- GCC says that -pie or -shared is needed to use -fsanitize=thread, but that is not enough on amd64: $ /usr/bin/gcc-4.8 -fsanitize=thread test.c -pie /usr/bin/ld.bfd.real: /tmp/ccBEgPXP.o: relocation R_X86_64_PC32 against undefined symbol `__tsan_func_entry' can not be used when making a shared object; recompile with -fPIC /usr/bin/ld.bfd.real: final link failed: Bad value collect2: error: ld returned 1 exit status -fPIC is needed too: $ /usr/bin/gcc-4.8 -fsanitize=thread test.c -pie -fPIC --- System information. --- Architecture: amd64 Kernel: Linux 3.8.3 Debian Release: 7.0 500 unstableftp.ro.debian.org 500 stable ftp.ro.debian.org 1 experimentalftp.ro.debian.org --- Package information. --- Depends (Version) | Installed =-+-=== gcc-4.8-base (= 4.8-20130315-1) | cpp-4.8(= 4.8-20130315-1) | binutils(>= 2.22) | libgcc-4.8-dev (= 4.8-20130315-1) | libc6 (>= 2.14) | libcloog-isl4 (>= 0.17) | libgmp10 | libisl10(>= 0.10) | libmpc2 | libmpfr4 (>= 3.1.1) | zlib1g (>= 1:1.1.4) | Recommends (Version) | Installed ==-+- libc6-dev (>= 2.13-5) | Suggests(Version) | Installed =-+-== gcc-4.8-multilib | libmudflap0-4.8-dev(= 4.8-20130315-1) | gcc-4.8-doc (>= 4.8) | gcc-4.8-locales (>= 4.8) | libgcc1-dbg (>= 1:4.8-20130315-1) | 1:4.8-20130315-1 libgomp1-dbg (>= 4.8-20130315-1) | libitm1-dbg (>= 4.8-20130315-1) | libatomic1-dbg(>= 4.8-20130315-1) | libasan0-dbg (>= 4.8-20130315-1) | libtsan0-dbg (>= 4.8-20130315-1) | libbacktrace1-dbg (>= 4.8-20130315-1) | libquadmath0-dbg (>= 4.8-20130315-1) | libmudflap0-dbg (>= 4.8-20130315-1) | binutils-gold (>= 2.22) | 2.21.52.20110707-1 -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/5145a30d.7000...@etorok.net
Bug#703230: [gcc-4.8] gcc-4.8: -fsanitize=thread requires -fPIC too
On 03/19/2013 12:07 AM, Matthias Klose wrote: > Am 17.03.2013 04:03, schrieb Török Edwin: >> GCC says that -pie or -shared is needed to use -fsanitize=thread > > where? > > In the error message when you use just -fsanitize=thread: $ gcc-4.8 -fsanitize=thread x.c gcc-4.8.real: error: -fsanitize=thread linking must be done with -pie or -shared It should probably mention -fPIC on amd64, as it doesn't work with just -pie. The manual doesn't mention this either, but thats an upstream problem as Debian didn't package that yet. Best regards, --Edwin -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/51479420.7000...@etorok.net
Bug#703230: [gcc-4.8] gcc-4.8: -fsanitize=thread requires -fPIC too
forwarded 703230 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56682 thanks -- To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/514b3d46.6030...@etorok.net
