diff options
author | Sergei Shtylyov <sshtylyov@ru.mvista.com> | 2007-02-07 18:18:20 +0100 |
---|---|---|
committer | Bartlomiej Zolnierkiewicz <bzolnier@gmail.com> | 2007-02-07 18:18:20 +0100 |
commit | 6273d26a5b280cb96b804424de323560b301ca51 (patch) | |
tree | 69c79073c85bab2767db45fe33014fbda8572e7b /fs/eventpoll.c | |
parent | 7b73ee05d0acb926923d43d78b61add776ea4bb1 (diff) |
hpt366: HPT36x PCI clock detection fix
Fix minor coding mistake in the HPT36x PCI clock detection code noticed by
Bartlomiej Zolnierkiewicz -- it always reported 33 MHz due to the missing
'break' statements. This, however, most probably never mattered -- in fact, I
was thinking of removing the 25/40 MHz cases completely since HPT36x BIOSes
didn't seem to set any other value than 7 into the 'cmd_high_time' field, i.e.
supported only 33 MHz PCI.
Note that in the original driver there was another bug: 25 and 40 MHz cases
were interchanged. Since the 'cmd_high_time' field is in units of PCI clocks,
a lower clock count just *cannot* correspond to a higher frequency, i. e. it
should be 5 for 25 MHz PCI and 9 for 40 MHz PCI, not the other way around.
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Diffstat (limited to 'fs/eventpoll.c')
0 files changed, 0 insertions, 0 deletions