lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

[PATCH] pipe: enable atomic copying of pipe data to/from user space

Browse Source 
The pipe ->map() method uses kmap() to virtually map the pages, which
is both slow and has known scalability issues on SMP. This patch enables
atomic copying of pipe pages, by pre-faulting data and using kmap_atomic()
instead.

lmbench bw_pipe and lat_pipe measurements agree this is a Good Thing. Here
are results from that on a UP machine with highmem (1.5GiB of RAM), running
first a UP kernel, SMP kernel, and SMP kernel patched.

Vanilla-UP:
Pipe bandwidth: 1622.28 MB/sec
Pipe bandwidth: 1610.59 MB/sec
Pipe bandwidth: 1608.30 MB/sec
Pipe latency: 7.3275 microseconds
Pipe latency: 7.2995 microseconds
Pipe latency: 7.3097 microseconds

Vanilla-SMP:
Pipe bandwidth: 1382.19 MB/sec
Pipe bandwidth: 1317.27 MB/sec
Pipe bandwidth: 1355.61 MB/sec
Pipe latency: 9.6402 microseconds
Pipe latency: 9.6696 microseconds
Pipe latency: 9.6153 microseconds

Patched-SMP:
Pipe bandwidth: 1578.70 MB/sec
Pipe bandwidth: 1579.95 MB/sec
Pipe bandwidth: 1578.63 MB/sec
Pipe latency: 9.1654 microseconds
Pipe latency: 9.2266 microseconds
Pipe latency: 9.1527 microseconds

Signed-off-by: Jens Axboe <axboe@suse.de>
This commit is contained in:
Jens Axboe
2006-05-01 20:02:05 +02:00
parent e27dedd84c
commit f6762b7ad8
3 changed files with 126 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fs/splice.c
View File

@@ -640,13 +640,13 @@ find_page:
/*
* Careful, ->map() uses KM_USER0!
*/
char *src = buf->ops->map(info, buf);
char *src = buf->ops->map(info, buf, 1);
char *dst = kmap_atomic(page, KM_USER1);
memcpy(dst + offset, src + buf->offset, this_len);
flush_dcache_page(page);
kunmap_atomic(dst, KM_USER1);
buf->ops->unmap(info, buf);
buf->ops->unmap(info, buf, src);
}
ret = mapping->a_ops->commit_write(file, page, offset, offset+this_len);