Lock granularity...

[Stephen C. Tweedie]

Hi,

On Thu, 27 Jan 2000 15:59:41 -0500, Michel W Zappe
<zapman en interlan.net> said:

> Right now i've mostly noticed the kernel lock's while perusing the
> code.  I haven't focused too much of my attention on those locks in
> particular, since my real bottleneck is in bdflush.  I was just
> wondering if there were any projects looking at that.  

You can enable kernel profiling and, from that, see exactly where in
the spinlock text segment we are stalling.  That was done fairly early
in 2.3, and at this point 2.3 has basically eliminated all the
contention points that show up significantly on the radar for up to 4
cpus.  There will always be special case workloads where we are
hitting contention on the big kernel lock, but things look pretty good
in general right now.

The one place where we still have a big bottleneck on SMP is the
network bottom half.  There is code to deal with that (the "softnet"
work), but it's not yet in the 2.3 tree.

--Stephen

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