Re: [patchset] Lockfree fd lookup 0 of 5

[William Lee Irwin III]

On Mon, 2 Aug 2004 17:56:07 +0100 viro@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx wrote:
>> How about this for comparison?  That's just a dumb "convert to rwlock"
>> patch; we can be smarter in e.g. close_on_exec handling, but that's a
>> separate story.

On Mon, Aug 02, 2004 at 01:07:29PM -0700, David S. Miller wrote:
> Compares to plain spinlocks, rwlock's don't buy you much,
> if anything, these days.
> Especially for short sequences of code.

I've found unusual results in this area. e.g. it does appear to matter
for mapping->tree_lock for database workloads that heavily share a
given file and access it in parallel. The radix tree walk, though
intuitively short, is long enough to make the rwlock a win in the
database-oriented uses and microbenchmarks starting around 4x.


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