Port-sparc64 archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: esp vs. MSG_IGN_WIDE_RESIDUE



Hello,

On Tue, 13 Jun 2017 21:24:38 -0000 (UTC)
mlelstv%serpens.de@localhost (Michael van Elst) wrote:

> macallan%netbsd.org@localhost (Michael) writes:
> 
> >How would one correctly handle this - check if any read is followed by
> >one of these?  
> 
> The message should just correct the transfer length.
> 
> I guess you always need to have the I/O buffer padded so that the
> invalid bytes can be stored. While you could handle the last bytes
> with PIO, the likely consequence is that you then also want to
> handle unaligned buffers. Few drivers support such use-cases.

So, since esp(4)'s SCSI bus can only ever be 16bit wide at most, it
can't be more than one byte that needs to be ignored? Since things work
otherwise I would assume that in this case the transfer length is
rounded up to be even.
And how likely are we to ever see unaligned buffers? On SPARC no less?

have fun
Michael


Home | Main Index | Thread Index | Old Index