Bug#639897: Please don't check /proc/mounts
Hi Aurelien,
thanks for your reply and sorry for mine being so late
On Fri, Sep 2, 2011 at 21:56, Aurelien Jarno <aurel32@debian.org> wrote:
> As always, when the libc doesn't have the information from a kernel
> syscall, it needs to get it another way, and it often ends up using
> /proc or /sys as the source of information. In some cases it can't take
> huge amount of time, but the problem is not that the libc doesn't parse
> the pseudo file inefficiently, just that the kernel has to generate
> insane amount of useless data.
Thanks for the explanation!
> In your case of statvfs(), the problem has been solved a bit more than
> one year ago, by exporting the missing data directly through the
> syscall. For that you need a libc >= 2.13 and a kernel >= 2.6.36. I
> guess in your case you are using Squeeze, which doesn't meet these
> requirements.
Yep, I'm on squeeze.
> The libc patch is trivial to backport, having it in squeeze is mostly
> about convincing the release managers that it's useful and safe. On the
Can you point me to that relevant patch? I can try to talk to RM but I
guess we won't get that far: given it won't solve the problem at hand
just by itself, because it requires an updated kernel, it would
probably get rejected - but worth a try.
> kernel side, it seems a bit more complicated, as the change has been
> followed by a lot of small bug fixes.
I see. What would you suggest as a better approach to solve our
problem? is backporting libc a feasable solution? bpo already have
2.6.38 (or now even later) kernel version, so we can probably use
that.
Thanks a lot for your input,
--
Sandro Tosi (aka morph, morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi
Reply to: