Messages in this thread | | | Date | Wed, 17 Sep 2003 12:31:27 +0200 | From | Jens Axboe <> | Subject | Re: ide-scsi oops was: 2.6.0-test4-mm3 |
| |
On Tue, Sep 16 2003, Ian Hastie wrote: > On Tuesday 16 Sep 2003 10:20, Jens Axboe wrote: > > On Tue, Sep 16 2003, Ian Hastie wrote: > > > On Thursday 11 Sep 2003 22:52, Jens Axboe wrote: > > > > Surely the pro version supports open-by-device as well? And then it > > > > should work fine. > > > > > > It does. However it also produces the same error message as cdrecord > > > when doing so, ie > > > > > > Warning: Open by 'devname' is unintentional and not supported. > > > > > > The implication being that it could go away or become broken at any time. > > > > I wouldn't read anything in to that if I were you. Joerg has some mis > > guided ideas about ATAPI addressing, but he would be a fool to remove > > open by devname at this point. > > What about this version of the argument then? There are a number if > pieces of software, eg cdrdao, that don't support open by devname. > The kernel developers would be foolish to remove support for them at > this time. Works both ways doesn't it.
(cc me if you want me to read the mails, thanks)
That's a different discussion - they don't work with SG_IO typically either, so they await the block sg driver anyways. It doesn't change the fact that trying to pretend devices are hanging off a SCSI bus with bus and device ids when they are not is just horrible.
-- Jens Axboe
- 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/
| |