Re: Wrong patches

Riley Williams (rhw@bigfoot.com)
Wed, 28 Oct 1998 21:11:07 +0000 (GMT)


Hi Keith.

>>> you should read a little more closely; if patch detects a reversed
>>> patch, it will ask you if you want to apply -R to this hunk.

> [My reply that the patch won't *look* reversed if the mailing list
> got a backwards patch, but 2.1.127 got the patch applied as
> intended, was sent via private email in hopes that this thread
> would go away.]

>> To my mind, there's one obvious solution to this whole argument that
>> shouldn't be hard to implement: Why not get diff to check the
>> timestamps of the files being compared, and issuing a warning if it
>> sees it's being asked to create a patch from a new file to an older
>> one?

> patch would have to apply the timestamps in the patch file itself,
> which would make those timestamps unreliable==useless for
> dependencies. It's fundamentally a synchronization problem, and
> which humans implement semaphores, monitors, or conditional
> critical sections?

Can I suggest you re-read my comment as quoted above - I said to make
the change to DIFF - not to PATCH as you appear to have assumed. The
ONLY time anything can be done about this problem is when the diff is
first produced at the original site, not when it's being applied at
other sites...

Best wishes from Riley.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/