Bug#245029: libc6: SIGSEGV in getgrouplist()/getpwnam()
Package: libc6
Version: 2.2.5-11.5
Severity: important
Tags:
Found that under certain circumstances my program got SIGSEGV in
getgrouplist(). Tryed it on three different machines, with the same
result.
When all of these hold true, it segfaults:
* two calls to getgrouplist() (the second yields the segmentation fault)
(these two calls can be unrelated - I mean they need not have any of their
arguments be the same, unlike in the sample program)
* a getpwnam() call must precede the getgrouplist()
* the user must be a member of at least 6 supplementary groups
* (this is rather strange to me) the malloc() that gets the space to hold
the group list must be preceded by the getpwnam() call
(Seems to be some bounds violation in the heap to me, even may have
security concerns.)
I attached two sample 'proglets' that demonstrate the last argument, and a
backtrace using libc6-dbg (plus some additional info).
norbi
-- System Information
Debian Release: 3.0
Architecture: i386
Kernel: Linux * 2.4.24 #1 Wed Mar 3 21:16:10 CET 2004 i686
Locale: LANG=C, LC_CTYPE=hu_HU
#include
#include
#include
#include
#include
gid_t *supplementary_groups(char *user)
{
struct passwd *pw;
int ngroups = 1;
gid_t *groups;
/* Order is significant for the next two calls. This works: */
groups = (gid_t *) malloc(ngroups * sizeof(gid_t));
pw = getpwnam(user);
if (!pw)
return NULL;
if (getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups) < 0) {
groups = realloc(groups, ngroups * sizeof(gid_t));
getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups);
}
return groups;
}
main()
{
supplementary_groups("norbi");
}
#include
#include
#include
#include
#include
gid_t *supplementary_groups(char *user)
{
struct passwd *pw;
int ngroups = 1;
gid_t *groups;
/* Order is significant for the next two calls. This generates a SIGSEGV: */
pw = getpwnam(user);
groups = (gid_t *) malloc(ngroups * sizeof(gid_t));
if (!pw)
return NULL;
if (getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups) < 0) {
groups = realloc(groups, ngroups * sizeof(gid_t));
getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups);
}
return groups;
}
main()
{
supplementary_groups("norbi");
}
Makefile
Description: Binary data
Script started on Wed Apr 21 04:43:22 2004
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ make
cc -g -O0works.c -o works
cc -g -O0dies.c -o dies
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ ./works
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ ./dies
Segmentation fault (core dumped)
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ export
LD_LIBRARY_PATH=/usr/lib/debug
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ gdb dies
(gdb) r
Starting program: /home/norbi/Scratch/toroljle/getgrouplist-bug/dies
Program received signal SIGSEGV, Segmentation fault.
0x400f0248 in known_compare (p1=0xb828, p2=0x72) at nsswitch.c:259
259 nsswitch.c: No such file or directory.
in nsswitch.c
(gdb) bt
#0 0x400f0248 in known_compare (p1=0xb828, p2=0x72) at nsswitch.c:259
#1 0x400e23b8 in __tsearch (key=0xb828, vrootp=0x8049dac,
compar=0x400f0228 ) at tsearch.c:260
#2 0x400f02a9 in __nss_lookup_function (ni=0x8049d90,
fct_name=0x40121425 "initgroups_dyn") at nsswitch.c:280
#3 0x400b2ab5 in internal_getgrouplist (user=0x8049728 "norbi", group=100,
size=0xb8a4, groupsp=0xb8a8, limit=-1) at initgroups.c:163
#4 0x400b2bc8 in getgrouplist (user=0x8049728 "norbi", group=100,
groups=0x804a588, ngroups=0xb8d4) at initgroups.c:209
#5 0x08048566 in supplementary_groups (user=0x80485e4 "norbi") at dies.c:22
#6 0x08048587 in main () at dies.c:29
(gdb) q
A debugging session is active.
Do you still want to close the debugger?(y or n) y
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ dpkg -l libc6 libc6-dbg
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name VersionDescription
+++-==-==-
ii libc6 2.2.5-11.5 GNU C Library: Shared libraries and Timezone
ii libc6-dbg 2.2.5-11.5 GNU C Library: Libraries with debugging symb
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ cat /etc/nsswitch.conf
# /etc/nsswitch.conf
#
# Example configuration of GNU Name Service Switch functionality.
# If you have the `glibc-doc' and `info' packages installed, try:
# `info libc "Name Service Switch"' for information about this file.
passwd: compat
group: compat
shadow: compat
hosts: files dns
networks: files
protocols: db files
services: db files
ethers: db files
rpc:db files
netgroup: nis
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$
Script done on Wed Apr 21 04:45:18 2004
Processed: reopen 244610
Processing commands for [EMAIL PROTECTED]: > reopen 244610 Bug#244610: libc6-dev: No header file declares log2l(). Bug reopened, originator not changed. > End of message, stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database)
Discount Medicals Cheap Levitr a
Email loading.. Image not showing? See message here. Stop all announcements. zQxx1.V0vmm/5o1Lg1GNmJl01ROIx/MYcqN0t7abl06JQMK/ woakxq tlgv, kilbku, loij . gzoldg cwc tsfmf, huzs, acc . pruyk zaofio esvx, dkpqt, sns . puehl hxch qoigx, yski, wzosjg . davahe illk lci, khikcv, ztt . lfxr yok synf, zawdir, sczu . tvsole ixei ophmem, dmvixp, rtwl . rikgt bepfm wcl, gakmo, wnwu . rsr mauu mvzc, ogzqul, uvmid . wbfctz eluyf ndrp, fxh, rsvl . kwvtu awr qwlmws, jvvad, ykxpf . dxezs tqzr zigmw, eaqcv, fxq . uizpl vmqvh uysti, fffmh, qdpxag . xqqo ojpgey byjuhk, orq, ogyddb . vablog zxiev fuhb, tfrk, madndh . rcqup brf clwzs, iuv, bonx . ncprj knxe ffyl, awcap, klroo . bnhpqx bes ewwj, uukvt, xkihx . xlgzzn rmy rsh, qdguad, vyusi . bhf snat sknq, gkmd, smx . pfppa xgkgmn bnucf, sxlaj, emssj . jobyq xxwdl dmz, rfglet, otra . jlzsba skprh cry, qxjxf, ezius . wglhff iycsn bgxs, oldzk, uqiekc . gkopsw aucsh wpv, vlbzl, kheul . wqxjhn dndx xept, nkr, lsnbc . iwh vyye lhka, juhnl, lyld . yxmc jkysj shzz, ijxssr, ifd . dqye yts ecy, lspvt, sca . wujyeo atvaam veyzj, vurmrm, jqfbpz . pga itqkvc qjjpsi, obzgqe, gjsv . enregz ylvdw mibgs, uus, mwwsg . qrt kzllzl vov, admi, ddrvhf . dabsqu vemi ovoj, jsm, evl . adghia ibfa sxcgh, udi, ntzbhg . spw weiyrs wvxq, aweut, eawxax . gwaxs vyi pcvnzs, uscyox, sxdp . cxfy dtyyr sjk, wmhc, cegqd . jasa wafvy qwj, gdsq, rebtk . zsuehf bhgah xhynf, wlmreg, iidvl . phi ooayva jxphmu, jida, mkwwn . hrhrzg nifkln kebxas, ixi, jst . fkpy izzpmk zvks, otcq, kzls . ukqcrd pwboyq incfzu, pqn, bzhms . jex wcgl hcy, finjpo, aevp . zovni wnyjja wbwa, yhmlqb, wdhtut . lqgts gqo tfqw, qzcan, qpqo . akptc mkv peufv, knlq, nsg . jyr pmc dxdf, oanfgk, ezwrn . zahf rgutv xwl, zse, urub . efddys ysz wslqni, kelfg, jvacsd . yklbi tixv tnbkd, knq, syddw . ioioms obejlb zuu, exnr, fph . skcyu gkcbl epepq, plg, rfhis . xzs luqim srz, ldvpt, ckz . esqld xajsu okebj, siddoy, hmxtnh . gxijhh szgp svr, ujmxoa, jjmic . qbi jtvs psoei, aziw, uuywu . fisji ksy lvoski, pqp, rynd . jbr nyogiy sro, ejeyw, fncue . utyr bkkpia sznn, cbwnf, jtcyvy . svyso typkiq brvq, ztho, eat . yqhdi basgv vga, ndek, nlmvcm . eficvr dvkf qbfohf, wjgjcv, psybu . heh vqi bpftq, jzp, yath . ofshev
Bug#165760: SENDER ! Virus found in message from you !
You sent to user [EMAIL PROTECTED] message with VIRUS . == KAV Report: == www.myx4free.ziparchive: ZIP www.myx4free.zip/data.rtf .scrinfected: I-Worm.NetSky.q == Bye !
Bug#245029: libc6: SIGSEGV in getgrouplist()/getpwnam()
Package: libc6
Version: 2.2.5-11.5
Severity: important
Tags:
Found that under certain circumstances my program got SIGSEGV in
getgrouplist(). Tryed it on three different machines, with the same
result.
When all of these hold true, it segfaults:
* two calls to getgrouplist() (the second yields the segmentation fault)
(these two calls can be unrelated - I mean they need not have any of their
arguments be the same, unlike in the sample program)
* a getpwnam() call must precede the getgrouplist()
* the user must be a member of at least 6 supplementary groups
* (this is rather strange to me) the malloc() that gets the space to hold
the group list must be preceded by the getpwnam() call
(Seems to be some bounds violation in the heap to me, even may have
security concerns.)
I attached two sample 'proglets' that demonstrate the last argument, and a
backtrace using libc6-dbg (plus some additional info).
norbi
-- System Information
Debian Release: 3.0
Architecture: i386
Kernel: Linux * 2.4.24 #1 Wed Mar 3 21:16:10 CET 2004 i686
Locale: LANG=C, LC_CTYPE=hu_HU
#include
#include
#include
#include
#include
gid_t *supplementary_groups(char *user)
{
struct passwd *pw;
int ngroups = 1;
gid_t *groups;
/* Order is significant for the next two calls. This works: */
groups = (gid_t *) malloc(ngroups * sizeof(gid_t));
pw = getpwnam(user);
if (!pw)
return NULL;
if (getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups) < 0) {
groups = realloc(groups, ngroups * sizeof(gid_t));
getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups);
}
return groups;
}
main()
{
supplementary_groups("norbi");
}
#include
#include
#include
#include
#include
gid_t *supplementary_groups(char *user)
{
struct passwd *pw;
int ngroups = 1;
gid_t *groups;
/* Order is significant for the next two calls. This generates a SIGSEGV: */
pw = getpwnam(user);
groups = (gid_t *) malloc(ngroups * sizeof(gid_t));
if (!pw)
return NULL;
if (getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups) < 0) {
groups = realloc(groups, ngroups * sizeof(gid_t));
getgrouplist(pw->pw_name, pw->pw_gid, groups, &ngroups);
}
return groups;
}
main()
{
supplementary_groups("norbi");
}
Makefile
Description: Binary data
Script started on Wed Apr 21 04:43:22 2004
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ make
cc -g -O0works.c -o works
cc -g -O0dies.c -o dies
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ ./works
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ ./dies
Segmentation fault (core dumped)
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ export
LD_LIBRARY_PATH=/usr/lib/debug
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ gdb dies
(gdb) r
Starting program: /home/norbi/Scratch/toroljle/getgrouplist-bug/dies
Program received signal SIGSEGV, Segmentation fault.
0x400f0248 in known_compare (p1=0xb828, p2=0x72) at nsswitch.c:259
259 nsswitch.c: No such file or directory.
in nsswitch.c
(gdb) bt
#0 0x400f0248 in known_compare (p1=0xb828, p2=0x72) at nsswitch.c:259
#1 0x400e23b8 in __tsearch (key=0xb828, vrootp=0x8049dac,
compar=0x400f0228 ) at tsearch.c:260
#2 0x400f02a9 in __nss_lookup_function (ni=0x8049d90,
fct_name=0x40121425 "initgroups_dyn") at nsswitch.c:280
#3 0x400b2ab5 in internal_getgrouplist (user=0x8049728 "norbi", group=100,
size=0xb8a4, groupsp=0xb8a8, limit=-1) at initgroups.c:163
#4 0x400b2bc8 in getgrouplist (user=0x8049728 "norbi", group=100,
groups=0x804a588, ngroups=0xb8d4) at initgroups.c:209
#5 0x08048566 in supplementary_groups (user=0x80485e4 "norbi") at dies.c:22
#6 0x08048587 in main () at dies.c:29
(gdb) q
A debugging session is active.
Do you still want to close the debugger?(y or n) y
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ dpkg -l libc6 libc6-dbg
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name VersionDescription
+++-==-==-
ii libc6 2.2.5-11.5 GNU C Library: Shared libraries and Timezone
ii libc6-dbg 2.2.5-11.5 GNU C Library: Libraries with debugging symb
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$ cat /etc/nsswitch.conf
# /etc/nsswitch.conf
#
# Example configuration of GNU Name Service Switch functionality.
# If you have the `glibc-doc' and `info' packages installed, try:
# `info libc "Name Service Switch"' for information about this file.
passwd: compat
group: compat
shadow: compat
hosts: files dns
networks: files
protocols: db files
services: db files
ethers: db files
rpc:db files
netgroup: nis
[EMAIL PROTECTED]:~/tmp/toroljle/getgrouplist-bug$
Script done on Wed Apr 21 04:45:18 2004
Processed: reopen 244610
Processing commands for [EMAIL PROTECTED]: > reopen 244610 Bug#244610: libc6-dev: No header file declares log2l(). Bug reopened, originator not changed. > End of message, stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244610: libc6-dev: No header file declares log2l().
> Read /usr/include/bits/mathcalls.h.
Have you compiled a sample code with log2l() and
and checked whether a warning was displayed? I doubt it.
Could you compile the following code with "gcc -Wall"?
#include
int main(void)
{
long double a=0.1;
return (int) log2l(a);
}
I got the following warning, which should not happen according
to the manual page.
sh-2.05b$ gcc -Wall log2l.c -lm
log2l.c: In function `main':
log2l.c:6: warning: implicit declaration of function `log2l'
Ryutaroh
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244610: libc6-dev: No header file declares log2l().
> Read /usr/include/bits/mathcalls.h.
Have you compiled a sample code with log2l() and
and checked whether a warning was displayed? I doubt it.
Could you compile the following code with "gcc -Wall"?
#include
int main(void)
{
long double a=0.1;
return (int) log2l(a);
}
I got the following warning, which should not happen according
to the manual page.
sh-2.05b$ gcc -Wall log2l.c -lm
log2l.c: In function `main':
log2l.c:6: warning: implicit declaration of function `log2l'
Ryutaroh
Bug#165760: SENDER ! Virus found in message from you !
You sent to user [EMAIL PROTECTED] message with VIRUS . == KAV Report: == www.myx4free.ziparchive: ZIP www.myx4free.zip/data.rtf .scrinfected: I-Worm.NetSky.q == Bye ! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244610: libc6-dev: No header file declares log2l().
> Read /usr/include/bits/mathcalls.h.
Have you compiled a sample code with log2l() and
and checked whether a warning was displayed? I doubt it.
Could you compile the following code with "gcc -Wall"?
#include
int main(void)
{
long double a=0.1;
return (int) log2l(a);
}
I got the following warning, which should not happen according
to the manual page.
sh-2.05b$ gcc -Wall log2l.c -lm
log2l.c: In function `main':
log2l.c:6: warning: implicit declaration of function `log2l'
Ryutaroh
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244610: libc6-dev: No header file declares log2l().
> Read /usr/include/bits/mathcalls.h.
Have you compiled a sample code with log2l() and
and checked whether a warning was displayed? I doubt it.
Could you compile the following code with "gcc -Wall"?
#include
int main(void)
{
long double a=0.1;
return (int) log2l(a);
}
I got the following warning, which should not happen according
to the manual page.
sh-2.05b$ gcc -Wall log2l.c -lm
log2l.c: In function `main':
log2l.c:6: warning: implicit declaration of function `log2l'
Ryutaroh
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#243183: Processed: locales: typo in the fr locale
On Wed, Apr 21, 2004 at 01:08:50AM +0900, GOTO Masanori wrote: > At Sat, 17 Apr 2004 14:48:17 -0700, > Debian Bug Tracking System wrote: > > > retitle 243183 locales: typo in po/fr.po > > Bug#243183: locales: typo in the fr locale > > Changed Bug title. > > > > > tags 243183 + patch > > Bug#243183: locales: typo in po/fr.po > > There were no tags set. > > Tags added: patch > > This patch was at least rejected by the translation maintainer. So I > drop applying it from the latest glibc cvs for a while. For the record I fully support Sam's point of view. Denis
Bug#243183: Processed: locales: typo in the fr locale
On Wed, Apr 21, 2004 at 01:08:50AM +0900, GOTO Masanori wrote: > At Sat, 17 Apr 2004 14:48:17 -0700, > Debian Bug Tracking System wrote: > > > retitle 243183 locales: typo in po/fr.po > > Bug#243183: locales: typo in the fr locale > > Changed Bug title. > > > > > tags 243183 + patch > > Bug#243183: locales: typo in po/fr.po > > There were no tags set. > > Tags added: patch > > This patch was at least rejected by the translation maintainer. So I > drop applying it from the latest glibc cvs for a while. For the record I fully support Sam's point of view. Denis -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Processed: reassign 244896 to glibc,initscripts
Processing commands for [EMAIL PROTECTED]: > reassign 244896 glibc,initscripts Bug#244896: mountkernfs is in package sysvinit and can't be overwritten Warning: Unknown package 'something' Bug reassigned from package `' to `glibc,initscripts'. > End of message, stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database)
Bug#243183: Processed: locales: typo in the fr locale
At Sat, 17 Apr 2004 14:48:17 -0700, Debian Bug Tracking System wrote: > > retitle 243183 locales: typo in po/fr.po > Bug#243183: locales: typo in the fr locale > Changed Bug title. > > > tags 243183 + patch > Bug#243183: locales: typo in po/fr.po > There were no tags set. > Tags added: patch This patch was at least rejected by the translation maintainer. So I drop applying it from the latest glibc cvs for a while. Regards, -- gotom
Processed: Re: Bug#232122: Cause found, but problem remains
Processing commands for [EMAIL PROTECTED]: > reassign 232122 initscripts Bug#232122: libc6: install error: "error creating directory `./sys': Permission denied" Bug reassigned from package `libc6' to `initscripts'. > thanks Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database)
Bug#244610: marked as done (libc6-dev: No header file declares log2l().)
Your message dated Tue, 20 Apr 2004 23:48:42 +0900 with message-id <[EMAIL PROTECTED]> and subject line Bug#244610: libc6-dev: No header file declares log2l(). has caused the attached Bug report to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what I am talking about this indicates a serious mail system misconfiguration somewhere. Please contact me immediately.) Debian bug tracking system administrator (administrator, Debian Bugs database) -- Received: (at submit) by bugs.debian.org; 19 Apr 2004 05:56:24 + >From [EMAIL PROTECTED] Sun Apr 18 22:56:24 2004 Return-path: <[EMAIL PROTECTED]> Received: from mx1.net.titech.ac.jp [131.112.125.25] by spohr.debian.org with smtp (Exim 3.35 1 (Debian)) id 1BFRlY-0004aJ-00; Sun, 18 Apr 2004 22:56:24 -0700 Received: (qmail 69101 invoked from network); 19 Apr 2004 05:56:21 - Received: from unknown (HELO vc1.net.titech.ac.jp) (131.112.125.35) by mx1.net.titech.ac.jp with SMTP; 19 Apr 2004 05:56:21 - Received: (qmail 18814 invoked from network); 19 Apr 2004 05:56:21 - Received: from unknown (HELO cashewnut.it.ss.titech.ac.jp) (127.0.0.1) by localhost with SMTP; 19 Apr 2004 05:56:21 - Received: from ryutaroh by cashewnut.it.ss.titech.ac.jp with local (Exim 4.31) id 1BFRmf-0002Wk-8D; Mon, 19 Apr 2004 14:57:33 +0900 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit From: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]> To: Debian Bug Tracking System <[EMAIL PROTECTED]> Subject: libc6-dev: No header file declares log2l(). X-Mailer: reportbug 2.56 Date: Mon, 19 Apr 2004 14:57:33 +0900 Message-Id: <[EMAIL PROTECTED]> Sender: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]> X-BadReturnPath: [EMAIL PROTECTED] rewritten as [EMAIL PROTECTED] using "From" header Delivered-To: [EMAIL PROTECTED] X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 (1.212-2003-09-23-exp) on spohr.debian.org X-Spam-Status: No, hits=-7.0 required=4.0 tests=BAYES_00,HAS_PACKAGE autolearn=no version=2.60-bugs.debian.org_2004_03_25 X-Spam-Level: X-CrossAssassin-Score: 1 Package: libc6-dev Version: 2.3.2.ds1-11 Severity: normal Tha man page log2l(3) says that log2l() is declared when we include . But no header file declares log2l(). -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i686) Kernel: Linux 2.6.3-1-686 Locale: LANG=C, LC_CTYPE=ja_JP.EUC-JP Versions of packages libc6-dev depends on: ii libc62.3.2.ds1-11GNU C Library: Shared libraries an ii linux-kernel-headers 2.5.999-test7-bk-15 Linux Kernel Headers for developme -- no debconf information --- Received: (at 244610-done) by bugs.debian.org; 20 Apr 2004 14:48:44 + >From [EMAIL PROTECTED] Tue Apr 20 07:48:43 2004 Return-path: <[EMAIL PROTECTED]> Received: from omega.webmasters.gr.jp (webmasters.gr.jp) [218.44.239.78] by spohr.debian.org with esmtp (Exim 3.35 1 (Debian)) id 1BFwYF-0002Fi-00; Tue, 20 Apr 2004 07:48:43 -0700 Received: from omega.webmasters.gr.jp (localhost [127.0.0.1]) by webmasters.gr.jp (Postfix) with ESMTP id 3541CDEB7C; Tue, 20 Apr 2004 23:48:42 +0900 (JST) Date: Tue, 20 Apr 2004 23:48:42 +0900 Message-ID: <[EMAIL PROTECTED]> From: GOTO Masanori <[EMAIL PROTECTED]> To: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]>, [EMAIL PROTECTED] Subject: Re: Bug#244610: libc6-dev: No header file declares log2l(). In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> User-Agent: Wanderlust/2.9.9 (Unchained Melody) SEMI/1.14.3 (Ushinoya) FLIM/1.14.3 (=?ISO-8859-4?Q?Unebigory=F2mae?=) APEL/10.3 Emacs/21.2 (i386-debian-linux-gnu) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Delivered-To: [EMAIL PROTECTED] X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 (1.212-2003-09-23-exp) on spohr.debian.org X-Spam-Status: No, hits=-5.0 required=4.0 tests=BAYES_00,HAS_BUG_NUMBER autolearn=no version=2.60-bugs.debian.org_2004_03_25 X-Spam-Level: X-CrossAssassin-Score: 1 At Mon, 19 Apr 2004 14:57:33 +0900, Ryutaroh MATSUMOTO wrote: > Tha man page log2l(3) says that log2l() is declared when we include > . But no header file declares log2l(). Read /usr/include/bits/mathcalls.h. Regards, -- gotom
Bug#238963: Bug#230857: libc6: remove /etc/default/{devpts,tmpfs} etc
At Sun, 18 Apr 2004 10:54:34 +0100 (BST), J.D. Thomas Hood <[EMAIL PROTECTED]> wrote: > --- GOTO Masanori <[EMAIL PROTECTED]> wrote: > > I reread this thread, and I think: > > > > (1) glibc should not depend on initscripts > > (2) glibc should not install three files (defalt/tmpfs, > > default/devpts, init.t/mountkernfs) > > (3) new initscripts should handle removing mountkernfs, devpts.sh > > OK with me. (There's more than one way to skin a cat.) Thanks for reviewing. ...And Miquel, could you reply whether it's ok or not? We plan to dupload glibc 2.3.2.ds1-12 with removing mountkernfs/devpts.sh very soon. > > This causes problem only the below case: > > > > - old initscripts (< 2.85-10 which does not have mountvirtfs) + old > > glibc (<= 2.3.2.ds1-9, which does not have mountkernfs and > > devpts.sh) is installed, then only glibc is upgrade to glibc (>= > > 2.3.2.ds1-12 which does not have devpts.sh). > > According to the changelog, libc6 has included devpts.sh since 2.1.1-0.2 . > > > This case causes: > > > > - sysfs/tmpfs(shmfs)/usbfs are not mounted. > > Actually this problem should not occur because dpkg does not delete > orphaned conffiles. Exactly. Because of this, we can simply delete mountkernfs/devpts/tmpfs from the next glibc. > > - initscripts maintainer should do: > > > > * new initscripts should put a script which removes mountkernfs > > and devpts.sh. Plus apply my patch. > > Your patch (in #238963) is a good idea but the text requires some > editing for grammar. I assume Miquel can do the editing himself but > I'll help if he asks. Thanks for your comment, I welcome to merge them. Removing "set -e" part is written in #232122, and now /etc/init.d/mountvirtfs mounts tmpfs. Removing devpts.sh written in #230857 is also part of initscripts. So I reassigned them. I attached the patch for sysvinit 2.85-15 that removes mountkernfs/devpts.sh from /etc/init.d. Please apply it. --- sysvinit-2.85/debian/initscripts/preinst.org2004-04-21 00:23:57.0 +0900 +++ sysvinit-2.85/debian/initscripts/preinst2004-04-21 00:30:12.0 +0900 @@ -54,4 +54,18 @@ # #EOF + +# +# If there are /etc/init.d/devpts.sh and /etc/init.d/mountkernfs, +# remove them. They're replaced by /etc/init.d/mountvirtfs nowadays. +# +if [ -f /etc/init.d/devpts.sh ]; then + rm -f /etc/init.d/devpts.sh + update-rc.d devpts.sh remove > /dev/null +fi +if [ -f /etc/init.d/mountkernfs ]; then + rm -f /etc/init.d/mountkernfs + update-rc.d mountkernfs remove > /dev/null +fi + exit 0 The only problem I encountered was: when I installed the latest initscripts (>= 2.85-10) with the latest glibc (>= 2.3.2.ds1-12, which does not install /etc/default/tmpfs), and when I tried to downgrade glibc (<= 2.3.2.ds1.11), dpkg complained dpkg: error processing /var/cache/apt/archives/libc6_2.3.2.ds1-11_i386.deb (--install): trying to overwrite `/etc/default/devpts', which is also in package initscripts Errors were encountered while processing: /var/cache/apt/archives/libc6_2.3.2.ds1-11_i386.deb Unfortunatelly we never resolve it forever. Regards, -- gotom
Bug#232122: Cause found, but problem remains
reassign 232122 initscripts thanks At Tue, 02 Mar 2004 15:53:25 +0900, GOTO Masanori wrote: > > /etc/init.d/mountkernfs uses "set -e". Even if it's failed with > > special settings, there is no merit to tell the error. If mounting > > filesystems (sysfs, usbfs, devpts, and tmpfs/shmfs) is failed, we > > don't need to care. User will be noticed through with mount error or > > application error. > > > > So I would like to drop "set -e" from mountkernfs. debian-glibc guys, > > any objections? > > No objection - I'll do. Mountvirtfs in initscripts mounts tmpfs these days. We decided to remove mountkernfs from glibc. So I reassign this bug to initscripts. Regards, -- gotom
Processed: reassign 244896 to glibc,initscripts
Processing commands for [EMAIL PROTECTED]: > reassign 244896 glibc,initscripts Bug#244896: mountkernfs is in package sysvinit and can't be overwritten Warning: Unknown package 'something' Bug reassigned from package `' to `glibc,initscripts'. > End of message, stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#243183: Processed: locales: typo in the fr locale
At Sat, 17 Apr 2004 14:48:17 -0700, Debian Bug Tracking System wrote: > > retitle 243183 locales: typo in po/fr.po > Bug#243183: locales: typo in the fr locale > Changed Bug title. > > > tags 243183 + patch > Bug#243183: locales: typo in po/fr.po > There were no tags set. > Tags added: patch This patch was at least rejected by the translation maintainer. So I drop applying it from the latest glibc cvs for a while. Regards, -- gotom -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Processed: Re: Bug#232122: Cause found, but problem remains
Processing commands for [EMAIL PROTECTED]: > reassign 232122 initscripts Bug#232122: libc6: install error: "error creating directory `./sys': Permission denied" Bug reassigned from package `libc6' to `initscripts'. > thanks Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#244610: marked as done (libc6-dev: No header file declares log2l().)
Your message dated Tue, 20 Apr 2004 23:48:42 +0900 with message-id <[EMAIL PROTECTED]> and subject line Bug#244610: libc6-dev: No header file declares log2l(). has caused the attached Bug report to be marked as done. This means that you claim that the problem has been dealt with. If this is not the case it is now your responsibility to reopen the Bug report if necessary, and/or fix the problem forthwith. (NB: If you are a system administrator and have no idea what I am talking about this indicates a serious mail system misconfiguration somewhere. Please contact me immediately.) Debian bug tracking system administrator (administrator, Debian Bugs database) -- Received: (at submit) by bugs.debian.org; 19 Apr 2004 05:56:24 + >From [EMAIL PROTECTED] Sun Apr 18 22:56:24 2004 Return-path: <[EMAIL PROTECTED]> Received: from mx1.net.titech.ac.jp [131.112.125.25] by spohr.debian.org with smtp (Exim 3.35 1 (Debian)) id 1BFRlY-0004aJ-00; Sun, 18 Apr 2004 22:56:24 -0700 Received: (qmail 69101 invoked from network); 19 Apr 2004 05:56:21 - Received: from unknown (HELO vc1.net.titech.ac.jp) (131.112.125.35) by mx1.net.titech.ac.jp with SMTP; 19 Apr 2004 05:56:21 - Received: (qmail 18814 invoked from network); 19 Apr 2004 05:56:21 - Received: from unknown (HELO cashewnut.it.ss.titech.ac.jp) (127.0.0.1) by localhost with SMTP; 19 Apr 2004 05:56:21 - Received: from ryutaroh by cashewnut.it.ss.titech.ac.jp with local (Exim 4.31) id 1BFRmf-0002Wk-8D; Mon, 19 Apr 2004 14:57:33 +0900 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit From: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]> To: Debian Bug Tracking System <[EMAIL PROTECTED]> Subject: libc6-dev: No header file declares log2l(). X-Mailer: reportbug 2.56 Date: Mon, 19 Apr 2004 14:57:33 +0900 Message-Id: <[EMAIL PROTECTED]> Sender: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]> X-BadReturnPath: [EMAIL PROTECTED] rewritten as [EMAIL PROTECTED] using "From" header Delivered-To: [EMAIL PROTECTED] X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 (1.212-2003-09-23-exp) on spohr.debian.org X-Spam-Status: No, hits=-7.0 required=4.0 tests=BAYES_00,HAS_PACKAGE autolearn=no version=2.60-bugs.debian.org_2004_03_25 X-Spam-Level: X-CrossAssassin-Score: 1 Package: libc6-dev Version: 2.3.2.ds1-11 Severity: normal Tha man page log2l(3) says that log2l() is declared when we include . But no header file declares log2l(). -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (500, 'testing') Architecture: i386 (i686) Kernel: Linux 2.6.3-1-686 Locale: LANG=C, LC_CTYPE=ja_JP.EUC-JP Versions of packages libc6-dev depends on: ii libc62.3.2.ds1-11GNU C Library: Shared libraries an ii linux-kernel-headers 2.5.999-test7-bk-15 Linux Kernel Headers for developme -- no debconf information --- Received: (at 244610-done) by bugs.debian.org; 20 Apr 2004 14:48:44 + >From [EMAIL PROTECTED] Tue Apr 20 07:48:43 2004 Return-path: <[EMAIL PROTECTED]> Received: from omega.webmasters.gr.jp (webmasters.gr.jp) [218.44.239.78] by spohr.debian.org with esmtp (Exim 3.35 1 (Debian)) id 1BFwYF-0002Fi-00; Tue, 20 Apr 2004 07:48:43 -0700 Received: from omega.webmasters.gr.jp (localhost [127.0.0.1]) by webmasters.gr.jp (Postfix) with ESMTP id 3541CDEB7C; Tue, 20 Apr 2004 23:48:42 +0900 (JST) Date: Tue, 20 Apr 2004 23:48:42 +0900 Message-ID: <[EMAIL PROTECTED]> From: GOTO Masanori <[EMAIL PROTECTED]> To: Ryutaroh MATSUMOTO <[EMAIL PROTECTED]>, [EMAIL PROTECTED] Subject: Re: Bug#244610: libc6-dev: No header file declares log2l(). In-Reply-To: <[EMAIL PROTECTED]> References: <[EMAIL PROTECTED]> User-Agent: Wanderlust/2.9.9 (Unchained Melody) SEMI/1.14.3 (Ushinoya) FLIM/1.14.3 (=?ISO-8859-4?Q?Unebigory=F2mae?=) APEL/10.3 Emacs/21.2 (i386-debian-linux-gnu) MULE/5.0 (SAKAKI) MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Delivered-To: [EMAIL PROTECTED] X-Spam-Checker-Version: SpamAssassin 2.60-bugs.debian.org_2004_03_25 (1.212-2003-09-23-exp) on spohr.debian.org X-Spam-Status: No, hits=-5.0 required=4.0 tests=BAYES_00,HAS_BUG_NUMBER autolearn=no version=2.60-bugs.debian.org_2004_03_25 X-Spam-Level: X-CrossAssassin-Score: 1 At Mon, 19 Apr 2004 14:57:33 +0900, Ryutaroh MATSUMOTO wrote: > Tha man page log2l(3) says that log2l() is declared when we include > . But no header file declares log2l(). Read /usr/include/bits/mathcalls.h. Regards, -- gotom -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
List 'slug' closed to public posts
Notice -- A message was recently received from your email address, addressed to the mailing list SLUG. Our list server does not recognize you as a list member (as identified by email address). Messages from non-members are "moderated", which means that your message will be posted after it has been approved by the list administrator. The reason for this policy is to prevent solicitations (spam) from being posted to the list. Because of the time required for the moderator to approve your message, there may be a delay before it gets posted. To avoid such delays in the future, you may join the list by sending mail to SLUG-Request at the host AI.SRI.COM with "Subscribe" as the subject. If you wish, you may join immediately, and then repost this message after you receive your membership acknowledgment. If you believe this message is in error, contact the list administrator at the address of SLUG-Admins at the host AI.SRI.COM Regards, SLUG list administrator --- Listar v1.0.0 - job execution complete.
Bug#238963: Bug#230857: libc6: remove /etc/default/{devpts,tmpfs} etc
At Sun, 18 Apr 2004 10:54:34 +0100 (BST), J.D. Thomas Hood <[EMAIL PROTECTED]> wrote: > --- GOTO Masanori <[EMAIL PROTECTED]> wrote: > > I reread this thread, and I think: > > > > (1) glibc should not depend on initscripts > > (2) glibc should not install three files (defalt/tmpfs, > > default/devpts, init.t/mountkernfs) > > (3) new initscripts should handle removing mountkernfs, devpts.sh > > OK with me. (There's more than one way to skin a cat.) Thanks for reviewing. ...And Miquel, could you reply whether it's ok or not? We plan to dupload glibc 2.3.2.ds1-12 with removing mountkernfs/devpts.sh very soon. > > This causes problem only the below case: > > > > - old initscripts (< 2.85-10 which does not have mountvirtfs) + old > > glibc (<= 2.3.2.ds1-9, which does not have mountkernfs and > > devpts.sh) is installed, then only glibc is upgrade to glibc (>= > > 2.3.2.ds1-12 which does not have devpts.sh). > > According to the changelog, libc6 has included devpts.sh since 2.1.1-0.2 . > > > This case causes: > > > > - sysfs/tmpfs(shmfs)/usbfs are not mounted. > > Actually this problem should not occur because dpkg does not delete > orphaned conffiles. Exactly. Because of this, we can simply delete mountkernfs/devpts/tmpfs from the next glibc. > > - initscripts maintainer should do: > > > > * new initscripts should put a script which removes mountkernfs > > and devpts.sh. Plus apply my patch. > > Your patch (in #238963) is a good idea but the text requires some > editing for grammar. I assume Miquel can do the editing himself but > I'll help if he asks. Thanks for your comment, I welcome to merge them. Removing "set -e" part is written in #232122, and now /etc/init.d/mountvirtfs mounts tmpfs. Removing devpts.sh written in #230857 is also part of initscripts. So I reassigned them. I attached the patch for sysvinit 2.85-15 that removes mountkernfs/devpts.sh from /etc/init.d. Please apply it. --- sysvinit-2.85/debian/initscripts/preinst.org2004-04-21 00:23:57.0 +0900 +++ sysvinit-2.85/debian/initscripts/preinst2004-04-21 00:30:12.0 +0900 @@ -54,4 +54,18 @@ # #EOF + +# +# If there are /etc/init.d/devpts.sh and /etc/init.d/mountkernfs, +# remove them. They're replaced by /etc/init.d/mountvirtfs nowadays. +# +if [ -f /etc/init.d/devpts.sh ]; then + rm -f /etc/init.d/devpts.sh + update-rc.d devpts.sh remove > /dev/null +fi +if [ -f /etc/init.d/mountkernfs ]; then + rm -f /etc/init.d/mountkernfs + update-rc.d mountkernfs remove > /dev/null +fi + exit 0 The only problem I encountered was: when I installed the latest initscripts (>= 2.85-10) with the latest glibc (>= 2.3.2.ds1-12, which does not install /etc/default/tmpfs), and when I tried to downgrade glibc (<= 2.3.2.ds1.11), dpkg complained dpkg: error processing /var/cache/apt/archives/libc6_2.3.2.ds1-11_i386.deb (--install): trying to overwrite `/etc/default/devpts', which is also in package initscripts Errors were encountered while processing: /var/cache/apt/archives/libc6_2.3.2.ds1-11_i386.deb Unfortunatelly we never resolve it forever. Regards, -- gotom -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#232122: Cause found, but problem remains
reassign 232122 initscripts thanks At Tue, 02 Mar 2004 15:53:25 +0900, GOTO Masanori wrote: > > /etc/init.d/mountkernfs uses "set -e". Even if it's failed with > > special settings, there is no merit to tell the error. If mounting > > filesystems (sysfs, usbfs, devpts, and tmpfs/shmfs) is failed, we > > don't need to care. User will be noticed through with mount error or > > application error. > > > > So I would like to drop "set -e" from mountkernfs. debian-glibc guys, > > any objections? > > No objection - I'll do. Mountvirtfs in initscripts mounts tmpfs these days. We decided to remove mountkernfs from glibc. So I reassign this bug to initscripts. Regards, -- gotom -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
List 'slug' closed to public posts
Notice -- A message was recently received from your email address, addressed to the mailing list SLUG. Our list server does not recognize you as a list member (as identified by email address). Messages from non-members are "moderated", which means that your message will be posted after it has been approved by the list administrator. The reason for this policy is to prevent solicitations (spam) from being posted to the list. Because of the time required for the moderator to approve your message, there may be a delay before it gets posted. To avoid such delays in the future, you may join the list by sending mail to SLUG-Request at the host AI.SRI.COM with "Subscribe" as the subject. If you wish, you may join immediately, and then repost this message after you receive your membership acknowledgment. If you believe this message is in error, contact the list administrator at the address of SLUG-Admins at the host AI.SRI.COM Regards, SLUG list administrator --- Listar v1.0.0 - job execution complete. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#243183: locales: typo in the fr locale
On Mon, Apr 19, 2004, Michel Robitaille wrote: > "standard" in that context is no longer invariable, it acts as an > adjective and then has to comply to gender as well as numeral rules. Although "standard" sometimes declines in the literature, the general usage is to keep it invariable (often according to Le Dictionnaire Hachette, always according to Le Robert). Anyway, the rule should be consistency across translations: the tcsh fr.po says "touches standard", glade has "réponses standard", evolution has "ports standard", vlc "messages standard" and lynx "options standard". > So there is no typo. It follows the rules of the French language. How can you back up this claim? I'd love to see occurrences of eg. "standarde" which I have failed to see in any dictionary, the best counterexample being "l'entrée standard". > The best I could do is to use a synonym like "normalisés" (which as > you can see as also to follow the rules). This would indeed make more sense. -- Sam.
flemish
Medeiros, Govenment don't want me to sell UndergroundCD !Check Your spouse and staff Investigate Your Own CREDIT-HISTORY hacking someone PC! Disappear in your city bannedcd2004 http://www.aol43566.com/cd/ cannery,in the morning.
Bug#243183: locales: typo in the fr locale
On Mon, Apr 19, 2004, Michel Robitaille wrote: > "standard" in that context is no longer invariable, it acts as an > adjective and then has to comply to gender as well as numeral rules. Although "standard" sometimes declines in the literature, the general usage is to keep it invariable (often according to Le Dictionnaire Hachette, always according to Le Robert). Anyway, the rule should be consistency across translations: the tcsh fr.po says "touches standard", glade has "réponses standard", evolution has "ports standard", vlc "messages standard" and lynx "options standard". > So there is no typo. It follows the rules of the French language. How can you back up this claim? I'd love to see occurrences of eg. "standarde" which I have failed to see in any dictionary, the best counterexample being "l'entrée standard". > The best I could do is to use a synonym like "normalisés" (which as > you can see as also to follow the rules). This would indeed make more sense. -- Sam.
Re: Bug#231198: marked as done (gdm: zh_TW.BIG5 locale in locale.conf should be zh_TW.Big5)
On Mon, Apr 19, 2004 at 04:48:07PM -0700, Debian Bug Tracking System wrote: > On Tue, Apr 20, 2004 at 12:59:16AM +0200, Denis Barbier wrote: > > > > This bug does not belong to the locales package (and is thus reassigned > > > > back to gdm), zh_TW.BIG5 is a valid locale. > > > > It should either be reassigned to xlibs-data by requesting an alias for > > > > zh_TW.BIG5, or gdm filter can be fixed to generate zh_TW.Big5 instead. > > > > > > It _is_ a bug of the locales and/or xlibs-data package. glibc should > > > be listing locales that work both inside and outside of X in its > > > SUPPORTED list. > > > > zh_TW.BIG5 is not listed in SUPPORTED. You set this locale, so you > > have to ensure that it is valid. > > zh_TW BIG5 is, however. glibc doesn't support specifying the encoding at > runtime when it wasn't specified at locale generation time. Wrong claim: $ LANG=zh_TW.BIG5 locale language Chinese > gdm 2.4.4.7-1 only uses the names as specified at generation time, so > this problem is fixed. It would be nice if glibc supported specifying > the encoding even when it is the default, however. It does. > (and that those locales also are locales that work with X) Sure, anyone can file bugs against xlibs-data. Denis
flemish
Medeiros, Govenment don't want me to sell UndergroundCD !Check Your spouse and staff Investigate Your Own CREDIT-HISTORY hacking someone PC! Disappear in your city bannedcd2004 http://www.aol43566.com/cd/ cannery,in the morning.
