Custom Query (431 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (22 - 24 of 431)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Ticket Resolution Summary Owner Reporter
#1581 fixed fails to build from source (#endif without #if) Afif Elghraoui <afif@…> afif
Description

I attempted to backport your commit to drop use of union wait[1] to the Debian 8.1.9 package, but it causes the following build error (line numbers are relative to the 8.1.9 version):

../clients/qevent/qevent.c: In function ‘qevent_start_trigger_script’: ../clients/qevent/qevent.c:307:2: error: #endif without #if

#endif

../clients/qevent/Makefile:83: recipe for target 'qevent.o' failed make[2]: * [qevent.o] Error 1

The attached patch resolves the problem and also streamlines the surrounding ifdefs (otherwise essentially the same condition is being checked three times in a row)

  1. https://gitlab.com/loveshack/sge/commit/538640ffe09a44a14d9f0047708f9c87d90c29ac
#1579 fixed Re: [SGE-bugs] qconf - not an admin host dlove
Description

Chris Webb <christopher.webb@…> writes:

This information can be still be obtained from qmon on non-admin hosts so I assume this is not deliberate. I wonder if it was due to the fix for #1547?

Yes, thanks. I'll have to re-do that -- obviously not thinking straight.

#1578 fixed qtcsh compile failure opoplawski
Description

Trying to compile sge 8.1.9 on Fedora Rawhide (F26) I get:

cc -c -O2 -Wstrict-prototypes -DLINUX -DLINUXAMD64 -DLINUXAMD64 -D_GNU_SOURCE -DGETHOSTBYNAME_R6 -DGETHOSTBYADDR_R8 -DTARGET_64BIT -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -m64 -mtune=generic -DSGE_PQS_API -DSPOOLING_dynamic -DSECURE -DHAVE_HWLOC=1 -DCOMPILE_DC -D__SGE_COMPILE_WITH_GETTEXT__ -D__SGE_NO_USERMAPPING__ -U_GNU_SOURCE -Wno-error -DPROG_NAME='"qtcsh"' -DLINUXAMD64                 -I. -I.. -D_PATH_TCSHELL='"/usr/local/bin/tcsh"'    -I../../../libs/gdi -I../../../libs/gdi ../sh.proc.c
In file included from /usr/include/signal.h:28:0,
                 from ../sh.h:39,
                 from ../sh.proc.c:33:
/usr/include/features.h:177:3: warning: #warning "_BSD_SOURCE and _SVID_SOURCE are deprecated, use _DEFAULT_SOURCE" [-Wcpp]
 # warning "_BSD_SOURCE and _SVID_SOURCE are deprecated, use _DEFAULT_SOURCE"
   ^~~~~~~
../sh.proc.c: In function 'pchild':
../sh.proc.c:153:16: error: storage size of 'w' isn't known
     union wait w;
                ^
../sh.proc.c:153:16: warning: unused variable 'w' [-Wunused-variable]
Makefile:381: recipe for target 'sh.proc.o' failed

Looking sh.proc.c, I see:

#if defined(_BSD) || (defined(IRIS4D) && __STDC__) || defined(__lucid) || defined(linux) || defined(__GNU__) || defined(__GLIBC__)
# define BSDWAIT
#endif /* _BSD || (IRIS4D && __STDC__) || __lucid || glibc */
Which doesn't make much sense to me. waitpid() is preferred over wait3() according to the wait3(2) man page, so we should be avoiding it when we can. So I would think we would want to drop "defined(linux)
defined(GNU) defined(GLIBC)" from that conditional. That combined with dropping #define BSDTIMES from the linux config file seems to do the trick.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Note: See TracQuery for help on using queries.