Bug#402318: libc6-dev: /usr/include/semaphore.h can't be used without pthread.h
Hi Aurelien,
Florian Le Goff a écrit :
Package: libc6-dev
Version: 2.3.6.ds1-9
Severity: wishlist
Hi,
There is an issue with the /usr/include/semaphore.h file and what
the manpages are saying for sem_open(3) and POSIX specifications
for the semaphore.h file.
In a nutshell, with the libc6-dev package, one can't do
the (basic) following :
#include
Without having included pthread.h before. That breaks a few
.c files that would have compiled without any error on other POSIX
systems with the following message :
In file included from test.c:10:
/usr/include/semaphore.h:34: error: field '__sem_lock' has incomplete
type
(where line 10 of test.c is "#include ").
Maybe should we inlcude pthread.h in /usr/include/semaphore.h ?
I don't have this problem here. /usr/include/semaphore.h includes
sys/types.h which includes bits/pthreadtypes.h, which defines
_pthread_fastlock.
You're right. I missed an -ansi in my $CFLAGS (see below).
Could you please send me a testcase and the compilation options you used?
Here it is :
> ble ~/bl $ cat test.c
> #include
>
> int main()
> {
>return 0;
> }
With -ansi, it fails :
--
> ble ~/bl $ cc -W -ansi -Wall -Werror -o test test.c
> In file included from test.c:1:
> /usr/include/semaphore.h:34: error: field '__sem_lock' has
> incomplete type
> ble ~/bl $
But, if I include before , the -ansi flag does
not cause any error nor warning.
Without, it works flawlessly :
---
ble ~/bl $ cc -W -Wall -Werror -o test test.c
ble ~/bl $
Cheers,
--
Florian Le Goff
r1906 - glibc-package/tags
Author: aurel32 Date: 2007-01-18 22:45:33 +0100 (Thu, 18 Jan 2007) New Revision: 1906 Added: glibc-package/tags/2.3.6.ds1-10/ Log: Tagging 2.3.6.ds1-10 Copied: glibc-package/tags/2.3.6.ds1-10 (from rev 1904, glibc-package/trunk) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
r1905 - glibc-package/branches/glibc-2.5/debian/patches/any
Author: aurel32
Date: 2007-01-18 22:26:27 +0100 (Thu, 18 Jan 2007)
New Revision: 1905
Modified:
glibc-package/branches/glibc-2.5/debian/patches/any/local-no-pagesize.diff
Log:
Revert i386 part of the PAGE_SIZE patch
Modified:
glibc-package/branches/glibc-2.5/debian/patches/any/local-no-pagesize.diff
===
--- glibc-package/branches/glibc-2.5/debian/patches/any/local-no-pagesize.diff
2007-01-09 10:10:34 UTC (rev 1904)
+++ glibc-package/branches/glibc-2.5/debian/patches/any/local-no-pagesize.diff
2007-01-18 21:26:27 UTC (rev 1905)
@@ -17,27 +17,6 @@
#define UPAGES1
#define HOST_TEXT_START_ADDR (u.start_code)
#define HOST_DATA_START_ADDR (u.start_data)
glibc-2.3.6.orig/sysdeps/unix/sysv/linux/i386/sys/user.h 2001-07-06
06:56:17.0 +0200
-+++ glibc-2.3.6/sysdeps/unix/sysv/linux/i386/sys/user.h2006-10-21
22:38:40.748151944 +0200
-@@ -23,6 +23,8 @@
-too much into it. Don't use it for anything other than GDB unless
-you know what you are doing. */
-
-+#include
-+
- struct user_fpregs_struct
- {
- long int cwd;
-@@ -92,8 +94,7 @@
- int u_debugreg [8];
- };
-
--#define PAGE_SHIFT12
--#define PAGE_SIZE (1UL << PAGE_SHIFT)
-+#define PAGE_SIZE sysconf(_SC_PAGESIZE)
- #define PAGE_MASK (~(PAGE_SIZE-1))
- #define NBPG PAGE_SIZE
- #define UPAGES1
--- glibc-2.3.6.orig/sysdeps/unix/sysv/linux/ia64/sys/user.h 2002-08-25
05:57:38.0 +0200
+++ glibc-2.3.6/sysdeps/unix/sysv/linux/ia64/sys/user.h2006-10-21
22:38:50.554268504 +0200
@@ -21,6 +21,7 @@
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#402776: still exists in 2.3.6.ds1-10
On Thu, Jan 18, 2007 at 07:38:45PM +0100, Aurelien Jarno wrote: > Looking more closely to the manpage it seems there is actually no bug, > the program behaves as expected: > > "For each zonename on the command line, print the time at the lowest > possible time value, the time one day after the lowest possible time > value, the times both one second before..." > > So that's normal that the output is different on architectures where > time_t has a 64-bit type. > > -9223372036854775808 and 9223372036854775807 are not really > representable in our current gregorian calendar, so those date are > displayed as "NULL". makes sense. -- dann frazier -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: PAGE_SIZE export
Martin Mares a écrit :
> Hello!
>
>> False, the glibc has nevered exported PAGE_SIZE, it has always been
>> provided by kernel headers.
>
> Sorry, but that's not true. If I download
> /debian/pool/main/g/glibc/glibc_2.3.6.ds1.orig.tar.gz, extract it, extract
> glibc-2.3.6.ds1.tar.bz2 inside and look at
> glibc-2.3.6/sysdeps/unix/sysv/linux/i386/sys/user.h, I see:
>
> | #define PAGE_SHIFT 12
> | #define PAGE_SIZE (1UL << PAGE_SHIFT)
> | #define PAGE_MASK (~(PAGE_SIZE-1))
>
> This file is normally installed as /usr/include/sys/user.h.
>
> The Debian patch glibc_2.3.6.ds1-9.diff.gz (namely
> glibc-2.3.6.ds1/debian/patches/any/local-no-pagesize.diff)
> contains:
>
> | +--- glibc-2.3.6.orig/sysdeps/unix/sysv/linux/i386/sys/user.h 2001-07-06
> 06:56:17.0 +0200
> | glibc-2.3.6/sysdeps/unix/sysv/linux/i386/sys/user.h2006-10-21
> 22:38:40.748151944 +0200
> | +@@ -23,6 +23,8 @@
> | +too much into it. Don't use it for anything other than GDB unless
> | +you know what you are doing. */
> | +
> | ++#include
> | ++
> | + struct user_fpregs_struct
> | + {
> | + long int cwd;
> | +@@ -92,8 +94,7 @@
> | + int u_debugreg [8];
> | + };
> | +
> | +-#define PAGE_SHIFT12
> | +-#define PAGE_SIZE (1UL << PAGE_SHIFT)
> | ++#define PAGE_SIZE sysconf(_SC_PAGESIZE)
> | + #define PAGE_MASK (~(PAGE_SIZE-1))
> | + #define NBPG PAGE_SIZE
> | + #define UPAGES1
>
> What's the purpose of this change?
>
>> Again false, it has been part of the kernel API since ages. The kernel
>> API has changed, please blame the kernel developers, not us.
>
> I am not talking about kernel API, but about glibc API, which is changed
> by the patch.
Oops sorry. So basically this only affects i386. We will probably change
that for glibc 2.5.
--
.''`. Aurelien Jarno | GPG: 1024D/F1BCDB73
: :' : Debian developer | Electrical Engineer
`. `' [EMAIL PROTECTED] | [EMAIL PROTECTED]
`-people.debian.org/~aurel32 | www.aurel32.net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#402776: still exists in 2.3.6.ds1-10
dann frazier a écrit : > reopen 402776 > found 402776 2.3.6.ds1-10 > thanks > > Reopening, I can still reproduce this on ia64. > > [EMAIL PROTECTED]:~$ dpkg -l libc6.1 > 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 libc6.12.3.6.ds1-10 GNU C Library: Shared libraries > > [EMAIL PROTECTED]:~$ zdump -v America/New_York | head -4 > America/New_York -9223372036854775808 = NULL > America/New_York -9223372036854689408 = NULL > America/New_York Sun Mar 31 06:59:59 1918 UTC = Sun Mar 31 01:59:59 1918 EST > isdst=0 gmtoff=-18000 > America/New_York Sun Mar 31 07:00:00 1918 UTC = Sun Mar 31 03:00:00 1918 EDT > isdst=1 gmtoff=-14400 > [EMAIL PROTECTED]:~$ zdump -v America/New_York | tail -4 > America/New_York Sun Nov 1 05:59:59 2037 UTC = Sun Nov 1 01:59:59 2037 EDT > isdst=1 gmtoff=-14400 > America/New_York Sun Nov 1 06:00:00 2037 UTC = Sun Nov 1 01:00:00 2037 EST > isdst=0 gmtoff=-18000 > America/New_York 9223372036854689407 = NULL > America/New_York 9223372036854775807 = NULL > Looking more closely to the manpage it seems there is actually no bug, the program behaves as expected: "For each zonename on the command line, print the time at the lowest possible time value, the time one day after the lowest possible time value, the times both one second before..." So that's normal that the output is different on architectures where time_t has a 64-bit type. -9223372036854775808 and 9223372036854775807 are not really representable in our current gregorian calendar, so those date are displayed as "NULL". -- .''`. Aurelien Jarno | GPG: 1024D/F1BCDB73 : :' : Debian developer | Electrical Engineer `. `' [EMAIL PROTECTED] | [EMAIL PROTECTED] `-people.debian.org/~aurel32 | www.aurel32.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#402776: still exists in 2.3.6.ds1-10
reopen 402776 found 402776 2.3.6.ds1-10 thanks Reopening, I can still reproduce this on ia64. [EMAIL PROTECTED]:~$ dpkg -l libc6.1 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 libc6.12.3.6.ds1-10 GNU C Library: Shared libraries [EMAIL PROTECTED]:~$ zdump -v America/New_York | head -4 America/New_York -9223372036854775808 = NULL America/New_York -9223372036854689408 = NULL America/New_York Sun Mar 31 06:59:59 1918 UTC = Sun Mar 31 01:59:59 1918 EST isdst=0 gmtoff=-18000 America/New_York Sun Mar 31 07:00:00 1918 UTC = Sun Mar 31 03:00:00 1918 EDT isdst=1 gmtoff=-14400 [EMAIL PROTECTED]:~$ zdump -v America/New_York | tail -4 America/New_York Sun Nov 1 05:59:59 2037 UTC = Sun Nov 1 01:59:59 2037 EDT isdst=1 gmtoff=-14400 America/New_York Sun Nov 1 06:00:00 2037 UTC = Sun Nov 1 01:00:00 2037 EST isdst=0 gmtoff=-18000 America/New_York 9223372036854689407 = NULL America/New_York 9223372036854775807 = NULL -- dann frazier | HP Open Source and Linux Organization -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Processed: still exists in 2.3.6.ds1-10
Processing commands for [EMAIL PROTECTED]: > reopen 402776 Bug#402776: zdump -v errors on 64 bit architectures 'reopen' is deprecated when a bug has been closed with a version; use 'found' or 'submitter' as appropriate instead. Bug reopened, originator not changed. > found 402776 2.3.6.ds1-10 Bug#402776: zdump -v errors on 64 bit architectures Bug marked as found in version 2.3.6.ds1-10. > 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]
