cvs commit: src/sys/dev/usb umass.c

[M. Warner Losh]

In message: <42088.1172642925 at critter.freebsd.dk>
            "Poul-Henning Kamp" <phk at phk.freebsd.dk> writes:
: In message <200702272233.l1RMXocb004983 at repoman.freebsd.org>, Warner Losh write
: s:
: 
: >  Some USB mass storage devices return the number of sectors in response
: >  to a READ_CAPACITY request rather than the maximum sector (off by one
: >  problem).  This causes a huge cascade of errors as the geom tasting
: >  code tries to read the last sector (which isn't really there in the
: >  face of this error).  automated tools that manipulate disk labels and
: >  such also have issues.
: >  
: >  Create a new quirk READ_CAPACITY_OFFBY1 
: 
: A better idea would be to have scsi_da.c try to read the 
: last sector and chop it if it fails.

Why is that a better idea?  There are only a few known bad bridges out
there that do this (Linux has about a dozen in its quirk list).  CAM
errors are rather verbose, and in this case there are 4 retries each
giving about 5 lines of output (I know this because something in the
geom tasting reads the last sector, or tries).

After careful consideration on scsi@, this was agreed to be the least
painful solution to the broken bridges.  There's already a lot of code
in umass to cope with the quirky umass devices, and a little more
wouldn't hurt.

Warner