Re: direct-IO writes strange behavior

[Alberto Nava <beto@xxxxxxxxxxx>]

Alberto Nava wrote:

> I'll try this again on test9 or test11.
I tried this with the 2.6.0-test11 and it works fine. The 512k
writes to a new file are propagaded all the way to the drives. It
work without having to use '-d unwritten' or preallocation.

Here is a snip of the proc file I used to verify it.

(8,16)@85886 b=f24f0d80 r=ef9c49b0 s=140288(size=524288) prio=0 rw 1
(8,32)@85886 b=ef891680 r=ef9c47b8 s=140288(size=524288) prio=0 rw 1
(8,48)@85895 b=ef891680 r=ef9c49b0 s=140288(size=524288) prio=0 rw 1
(8,64)@85895 b=f24f0d80 r=ef9c47b8 s=140288(size=524288) prio=0 rw 1
(8,80)@85903 b=ef891680 r=ef9c47b8 s=140288(size=524288) prio=0 rw 1


I'll use stock Linus until the direct-io rework is completed/tested.

Thanks a lot for your help
beto