emacs-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Build problem: Error: symbol `pread64' is already defined


From: Burton Samograd
Subject: Re: Build problem: Error: symbol `pread64' is already defined
Date: Sat, 10 Nov 2012 13:17:36 -0700
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.2.50 (gnu/linux)

Burton Samograd <address@hidden> writes:

> On 12-11-04 10:44 AM, Burton Samograd wrote:
>> Hello,
>>
>> I'm trying to build the current emacs git HEAD on one of my machines
>> (Centos 5.8 on a VM host so I can't update it to a newer version)
>> and I keep getting the following error:
>>
>> gcc -std=gnu99 -DHAVE_CONFIG_H -I. -I/root/src/emacs-git/lib
>> -I../src -I/root/src/emacs-git/src -g3 -O2 -MT careadlinkat.o -MD
>> -MP -MF .deps/careadlinkat.Tpo -c -o careadlinkat.o careadlinkat.c
>> /tmp/cc2jQqUg.s: Assembler messages:
>> /tmp/cc2jQqUg.s:122: Error: symbol `pread64' is already defined
>>
>> From what I can see this looks to be an issue with my libc headers
>> (unistd.h most directly I think) .  The glibc on this system is
>> 2.5.81.el5_8.7 and I've recently done a 'yum update' to the current
>> release for this Centos version.
>>
>> I've done a side by side comparison of the section of unistd.h on
>> two machines, one a current Ubuntu and the Centos one that is having
>> problems; they are almost close to identical with only a 'if
>> defined' guard that should have nothing to do with with the systems
>> I am running on (I hope).
>>
>> I've also updated my gcc to 4.7.2, the current release.
>>
>> Any ideas on what might be causing this build problem on this
>> specific machine and not others?  This isn't just with the current
>> HEAD, but it's been going on for a few weeks now; I'm just finally
>> getting around to reporting it.
>>
>> --
>> Burton Samograd
>>
> I'd like to add that I have been able to build emacs git HEAD on this
> machine in the past, maybe about a month ago.

This looks to be a problem with newer versions of gcc combined with the
libc of Centos 5.8, and something to do with definitions in my generated
emacs config.h. If I compile using the system compiler (/usr/bin/gcc) I
have no problems building, but when I use my upgraded gcc I get the
message about the duplicate decleration of 'pread64'.  If I comment out
config.h and use the upgraded compiler tom compile careadlink.c I get no
errors.

The following is the contents of unistd.h that contains any references
to pread64:

------------------------- unistd.h ---------------------------------
#ifdef __USE_UNIX98
# ifndef __USE_FILE_OFFSET64
/* Read NBYTES into BUF from FD at the given position OFFSET without
   changing the file pointer.  Return the number read, -1 for errors
   or 0 for EOF.

   This function is a cancellation point and therefore not marked with
   __THROW.  */
extern ssize_t pread (int __fd, void *__buf, size_t __nbytes,
                      __off_t __offset) __wur;

/* Write N bytes of BUF to FD at the given position OFFSET without
   changing the file pointer.  Return the number written, or -1.

   This function is a cancellation point and therefore not marked with
   __THROW.  */
extern ssize_t pwrite (int __fd, __const void *__buf, size_t __n,
                       __off_t __offset) __wur;
# else
#  ifdef __REDIRECT
extern ssize_t __REDIRECT (pread, (int __fd, void *__buf, size_t __nbytes,
                                   __off64_t __offset),
                           pread64) __wur;
extern ssize_t __REDIRECT (pwrite, (int __fd, __const void *__buf,
                                    size_t __nbytes, __off64_t __offset),
                           pwrite64) __wur;
#  else
#   define pread pread64
#   define pwrite pwrite64
#  endif
# endif

# ifdef __USE_LARGEFILE64
/* Read NBYTES into BUF from FD at the given position OFFSET without
   changing the file pointer.  Return the number read, -1 for errors
   or 0 for EOF.  */
extern ssize_t pread64 (int __fd, void *__buf, size_t __nbytes,
                        __off64_t __offset) __wur;
/* Write N bytes of BUF to FD at the given position OFFSET without
   changing the file pointer.  Return the number written, or -1.  */
extern ssize_t pwrite64 (int __fd, __const void *__buf, size_t __n,
                         __off64_t __offset) __wur;
# endif
#endif
------------------------- unistd.h ---------------------------------

None of __USE_UNIX98, __USE_FILE_OFFSET64 nor __USE_LARGEFILE64 are
defined by config.h, although _FILE_OFFSET_BITS is defined to 64 so this
will probably cause __USE_LARGEFILE64 to be defined later.

For now I'll just use the system compiler.  This is a tricky bug to
track down what exactly is going wrong and who's bug it actually is
given the age of the CentOS distro and the newness of the compiler. But
for now things are working and I'll just stick with using the system
compiler.

--
Burton Samograd




reply via email to

[Prev in Thread] Current Thread [Next in Thread]