Re: NUMA API observations

From: Anton Blanchard (anton_at_samba.org)
Date: 06/14/04

  • Next message: David Eger: "pcibios_write_config_dword()? porting drivers to 2.6" 
    Date:	Tue, 15 Jun 2004 07:40:04 +1000
To: Andi Kleen <ak@muc.de>

     
    > interleave should always fall back to other nodes. Very weird.
    > Needs to be investigated. What were the actual arguments passed
    > to the syscalls?

    This one looks like a bug in my code. I wasnt setting numnodes high
    enough, so the node fallback lists werent being initialised for some
    nodes.

    > > My kernel is compiled with NR_CPUS=128, the setaffinity syscall must be
    > > called with a bitmap at least as big as the kernels cpumask_t. I will
    > > submit a patch for this shortly.
    >
    > Umm, what a misfeature. We size the buffer up to the biggest
    > running CPU. That should be enough.
    >
    > IMHO that's just a kernel bug. How should a user space
    > application sanely discover the cpumask_t size needed by the kernel?
    > Whoever designed that was on crack.

    glibc now uses a select style interface. Unfortunately the interface has
    changed about three times by now.

    > I will probably make it loop and double the buffer until EINVAL ends or it
    > passes a page and add a nasty comment.

    Perhaps we could use the new glibc interface and fall back to the loop
    on older glibcs.

    Anton
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

  • Next message: David Eger: "pcibios_write_config_dword()? porting drivers to 2.6"

    Relevant Pages

    • Re: patch to fix set_itimer() behaviour in boundary cases
      ... >>These are things we should fix. ... Rather hard to trap that in glibc. ... actually doing the proper wait using this interface is rather hard. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: NUMA API observations
      ... >> glibc now uses a select style interface. ... With the upcomming numactl version it should work. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Kernel includefile bug not fixed after a year :-(
      ... If the interface is going to change then the user should be informed about the ... glibc exports own interaces and for this reason needs to supply own include ... it is definitely true for the SCSI related ioctls. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: proper API for sched_setaffinity ?
      ... The sched_setaffinity interface really is very messy: ... API changes in same minor version of glibc should just not happen. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: 2.6.18-rc1-mm2: process `showconsole used the removed sysctl system call
      ... warning: process `touch' used the removed sysctl system call ... binary sysctl interface for some time time now". ... It is the glibc pthread library testing to see if you have an SMP ... kernel by greping for SMP in the UTS_VERSION string. ...
      (Linux-Kernel)