Hi Thomas & Doug, On Sun, 2013-10-20 at 21:05 +0200, Thomas Glanzmann wrote: > Hello Doug, > > * Douglas Gilbert <dgilbert@xxxxxxxxxxxx> [2013-10-20 17:18]: > > Another thing you might test for me with ddpt --xcopy > > is the ability to copy between 512 and 4096 byte logical > > block sized media. ddpt's design is more general than > > sg_xcopy (inherited from sg_dd) and ddpt should be > > able to handle that both with a normal (i.e. non xcopy) > > and using xcopy. Is is possible for you to set up > > such an environment to test that? > > I do, since the target in the Linux kernel does support setting a blocksize > other than 512 for file io backends. I tried and the target paniced. I'm taking > Nab and target-devel on CC. > > Howto reproduce: > > - I created two file_io volumes, one with 512 Byte blocksize > (default) and one with 4096 Byte blocksize. > Thanks for the bug-report, and apologies on the delayed response on this one.. I just sent out a series to address the OOPs this failure was triggering, along with two other related fixes. The series ends up explicitly checking + failing with CHECK_CONDITION status for the case where the block_sizes for the source + destination devices do not match. Thomas, I've been able to confirm on my end so please go ahead and double check at your earliest convenience. Btw, I'd like to include these into v3.12-rc7, so they will need to goto Linus in the next 2-3 days. --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
