PTRACE_GETSIGINFO definitely exists on SLES10 SP3 64-bit: suse10$ cat /etc/issue
Welcome to SUSE Linux Enterprise Server 10 SP3 (x86_64) - Kernel \r (\l). suse10$ find /usr/include -type f | xargs grep PTRACE_GETSIGINFO /usr/include/linux/ptrace.h:#define PTRACE_GETSIGINFO 0x4202 suse10$ rpm -qf /usr/include/linux/ptrace.h glibc-devel-2.4-31.77.76.1 This is on a testing system I have lying around... No time to try compiling Valgrind there at the moment, though. Does Valgrind #include <linux/ptrace.h> ? A quick scan of the headers on SLES10 suggests <sys/ptrace.h> is not sufficient to pick up this #define... - Pat On Mon, Sep 15, 2014 at 12:59 PM, Philippe Waroquiers <philippe.waroqui...@skynet.be> wrote: > On Mon, 2014-09-15 at 16:44 +0200, Matthias Apitz wrote: > >> vgdb-invoker-ptrace.c:319: error: 'PTRACE_GETSIGINFO' undeclared (first use >> in this function) > A similar compilation error was encountered on s390x/RHEL5/glibc 2.3.4, > as this glibc version does not define PTRACE_GETSIGINFO. > Ptrace manual describes that PTRACE_GETSIGINFO is available since Linux > 2.3.99-pre6) so your kernel (2.6.16) should support PTRACE_GETSIGINFO. > > Which glibc version do you have ? > The easy fix to try is to define PTRACE_GETSIGINFO as for s390x, > e.g. by just removing the #if defined(VGA_s390x) in the following > extract of vgdb-invoker-ptrace.c: > #if defined(VGA_s390x) > /* RHEL 5 uses glibc 2.3.4 which does not define PTRACE_GETSIGINFO */ > # ifndef PTRACE_GETSIGINFO > # define PTRACE_GETSIGINFO 0x4202 > # endif > #endif > >> ... >> /usr/lib64/gcc/x86_64-suse-linux/4.1.2/../../../../x86_64-suse-linux/bin/ld: >> warning: i386 architecture of input file `valgrind-m_debuglog.o' is >> incompatible with i386:x86-64 output >> valgrind-m_debuglog.o: In function `myvprintf_int64': >> /home/sisis/guru/valgrind-3.10.0/coregrind/m_debuglog.c:685: undefined >> reference to `__umoddi3' >> /home/sisis/guru/valgrind-3.10.0/coregrind/m_debuglog.c:686: undefined >> reference to `__udivdi3' > This is strange. Lines 685 and 686 have not changed since 2005. > But seeing the first error msg, it looks like you might have a leftover > of the '32 bits only' causing a problem. > To test with only 64 bits, it might be better to start from a clean > source tree. > > Philippe > > > > ------------------------------------------------------------------------------ > Want excitement? > Manually upgrade your production database. > When you want reliability, choose Perforce > Perforce version control. Predictably reliable. > http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk > _______________________________________________ > Valgrind-users mailing list > Valgrind-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/valgrind-users ------------------------------------------------------------------------------ Want excitement? Manually upgrade your production database. When you want reliability, choose Perforce Perforce version control. Predictably reliable. http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk _______________________________________________ Valgrind-users mailing list Valgrind-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/valgrind-users
