diff options
| author | cinap_lenrek <cinap_lenrek@felloff.net> | 2015-07-13 18:37:56 +0200 |
|---|---|---|
| committer | cinap_lenrek <cinap_lenrek@felloff.net> | 2015-07-13 18:37:56 +0200 |
| commit | f5688dd6c9bfb6cee00f2c5e5f190d7a2ffaa6c6 (patch) | |
| tree | 665cea3d24df8d6143db6c24a804b0237c9deb00 /sys/src/cmd/9660srv/iobuf.c | |
| parent | 2559f92a83d2a04000fb1a0d392c6e2597758b11 (diff) | |
9660srv: go back to 128k clusters
doing tests taring up 9front.iso shows the following:
lowering the cluster size back to 128k avoids over half the
reads. 837888 sectors read for 512k vs. 347712 sectors with
128k cluster size.
Diffstat (limited to 'sys/src/cmd/9660srv/iobuf.c')
| -rw-r--r-- | sys/src/cmd/9660srv/iobuf.c | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/sys/src/cmd/9660srv/iobuf.c b/sys/src/cmd/9660srv/iobuf.c index 30eafb3a4..6bba533e8 100644 --- a/sys/src/cmd/9660srv/iobuf.c +++ b/sys/src/cmd/9660srv/iobuf.c @@ -23,9 +23,8 @@ * it's worth worrying about. -rsc */ -/* trying a larger value to get greater throughput - geoff */ -#define BUFPERCLUST 256 /* sectors/cluster; was 64, 64*Sectorsize = 128kb */ -#define NCLUST 16 +#define BUFPERCLUST 64 /* sectors/cluster; 64*Sectorsize = 128kb */ +#define NCLUST 64 int nclust = NCLUST; |