Unless I'm missing something the effect of write caching will be nil from a
safety standpoint *if* the drive does *not* reorder writes. If cached
writes are written to the platter in order it seems that a loss of power
will simply mean that from the platter perspective the system will look like
the power was lost at "T-x" rather than "T" where "T" is actual time of the
outage and "x" is the age of the oldest piece of cached data. The net
effect is that the filesystem should be in no worse shape than if there were
no caching the power actually went out at "T-x". (Unless of course I am
missing something here.)
If the cached writes can be reordered then of course it stands that caching
would be unsafe. Does anyone know if IDE drives do this? I'm certainly no
expert in this area but I thought only SCSI drives reordered operations.
Also, do you happen to know if the barrier patch will apply cleanly to a RH
2.4.18-rc1-ac2 kernel and function properly?
Regards,
Scott Lee
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/