Doug Ledford wrote:
>
> Douglas Gilbert wrote:
>
> > So IOMETER could be ported now. The kiobuf issue is an internal
> > kernel matter that could be bypassed, if necessary. I would prefer
> > to wait for Stephen.
>
> Maybe, but very few big guys like Intel tend to work on the unstable branch
> when porting things like IOMETER. Likely they won't do anything until it's
> available to them in a standard, shipping kernel and also available via posix
> interfaces implemented in glibc as oppossed to hand writing sg based code.
Point taken. Various multi platform applications (e.g. SANE,
cdrecord and cdparanoia) have their own transport layer libraries
to give them a uniform interface over many scsi (and pseudo scsi)
pass through mechanisms. Joerg Schilling is actively promoting
his "libscg" separately from cdrecord (which uses libscg).
My guess would be that IOMETER does use NT's ASPI interface to
do calculations like IO/sec as described by Bob Frey's earlier
posting on this thread.
Well I didn't have to wait long for Stephen Tweedie who came out
with a major kiobuf patch shortly after my posting yesterday.
Doug Gilbert
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to [EMAIL PROTECTED]
