Announce: ksymoops 2.4.3 is available
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Content-Type: text/plain; charset=us-ascii
ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/v2.4
ksymoops-2.4.3.tar.gz Source tarball, includes RPM spec file
ksymoops-2.4.3-1.src.rpm As above, in SRPM format
ksymoops-2.4.3-1.i386.rpm Compiled with 2.96 20000731, glibc 2.2.2
ksymoops-2.4.3-1.ia64.rpm Compiled with gcc 2.96-ia64-20000731,
glibc-2.2.3.
patch-ksymoops-2.4.3.gz Patch from ksymoops 2.4.2 to 2.4.3.
No sparc binary, I do not have access to a sparc system with a decent version of binutils.
Changelog extract
* Add Pid:.
* Add -A "address list". Idea pinched from Randy Dunlap's ksysmap.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: Exmh version 2.1.1 10/15/1999
iD8DBQE7qwPli4UHNye0ZOoRAqtjAJ9JKaMGxlU8Bwqwmn7ShJ9OhlctmgCg2rPQ M4jW+vdQfuMSFPPRleKTj5Y=
=LC33
-----END PGP SIGNATURE-----
Recompiling ksymoops with gcc-3.0.2 produces the following errors:
(rpm --rebuild ksymoops-2.4.3-1.src.rpm):
ksymoops.c:114:1: directives may not be used inside a macro argument ksymoops.c:114:1: unterminated argument list invoking macro "printf" ksymoops.c: In function `usage':
ksymoops.c:117: parse error before string constant
gcc is using the flags "-march=i386 -mcpu=i686" when it tries to compile ksymoops.c
(rpm -bb ksymoops.spec after installing):
Reports the same errors as above and uses the same -march and -mcpu flags.
(make within the ksymoops directory):
Compile completes successfully. Although the following warnings are produce:
symbol.c:220:58: warning: trigraph ??> ignored
symbol.c:221:44: warning: trigraph ??> ignored
symbol.c:225:49: warning: trigraph ??> ignored
symbol.c:226:35: warning: trigraph ??> ignored
Stephen Torri
Quoted 37 lines · Keith Owens
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1Content-Type: text/plain; charset=us-ascii
ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/v2.4
ksymoops-2.4.3.tar.gz Source tarball, includes RPM spec file
ksymoops-2.4.3-1.src.rpm As above, in SRPM format
ksymoops-2.4.3-1.i386.rpm Compiled with 2.96 20000731, glibc 2.2.2
ksymoops-2.4.3-1.ia64.rpm Compiled with gcc 2.96-ia64-20000731,glibc-2.2.3.patch-ksymoops-2.4.3.gz Patch from ksymoops 2.4.2 to 2.4.3.
No sparc binary, I do not have access to a sparc system with a decent version of binutils.
Changelog extract
* Add Pid:.
* Add -A "address list". Idea pinched from Randy Dunlap's ksysmap.-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: Exmh version 2.1.1 10/15/1999iD8DBQE7qwPli4UHNye0ZOoRAqtjAJ9JKaMGxlU8Bwqwmn7ShJ9OhlctmgCg2rPQ M4jW+vdQfuMSFPPRleKTj5Y=
=LC33
-----END PGP SIGNATURE------
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Quoted 1 line · Stephen Torri
ksymoops.c:114:1: directives may not be used inside a macro argument
Some versions of glibc define prinf as a macro instead of a function. That is quite legal but it stops you using #ifdef inside printf calls. I will change ksymoops to work around that.
Quoted 4 lines
symbol.c:220:58: warning: trigraph ??> ignored symbol.c:221:44: warning: trigraph ??> ignored symbol.c:225:49: warning: trigraph ??> ignored symbol.c:226:35: warning: trigraph ??> ignored
I believe that is a gcc bug. The text is
snprintf(map, size,
options->hex ? "<END_OF_CODE+%llx/????>"
: "<END_OF_CODE+%lld/????>",
offset);gcc is complaining about trigraphs but they are inside a string constant, not in code. IMHO gcc should not flag trigraphs in string constants, report it as a gcc bug.
[email protected] (Keith Owens) wrote on 22.09.01 in <[email protected]>:
Quoted 5 lines
Quoted 4 lines
symbol.c:220:58: warning: trigraph ??> ignored symbol.c:221:44: warning: trigraph ??> ignored symbol.c:225:49: warning: trigraph ??> ignored symbol.c:226:35: warning: trigraph ??> ignored
Quoted 8 lines
I believe that is a gcc bug. The text is
snprintf(map, size, options->hex ? "<END_OF_CODE+%llx/????>" : "<END_OF_CODE+%lld/????>", offset);gcc is complaining about trigraphs but they are inside a string constant, not in code. IMHO gcc should not flag trigraphs in string constants, report it as a gcc bug.
Trigraphs are defined to work within string constants. In fact, the standard requires trigraphs to be replaced before the very first tokenizing of the input.
How else could the ??/ trigraph for \ possibly work?
Gcc is correct.
MfG Kai
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Content-Type: text/plain; charset=us-ascii
Some versions of glibc define printf as a macro which prevents the use of #ifdef inside printf. ksymoops-2.4.3-2.src.rpm contains a work around, no other changes were made.
ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/v2.4
ksymoops-2.4.3.tar.gz Source tarball, includes RPM spec file
ksymoops-2.4.3-2.src.rpm As above, in SRPM format
ksymoops-2.4.3-2.i386.rpm Compiled with 2.96 20000731, glibc 2.2.2
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: Exmh version 2.1.1 10/15/1999
iD8DBQE7q+47i4UHNye0ZOoRAra7AKDJU+24svbg+/TlYlMX4iaxH4X89wCfTrhg 4cA6JgRySssifgUePnSw4D4=
=i1U5
-----END PGP SIGNATURE-----