status of ppc64 patches
Linas Vepstas
linas at austin.ibm.com
Thu Oct 21 04:45:01 EST 2004
On Tue, Oct 19, 2004 at 08:03:01PM -0500, Olof Johansson was heard to remark:
>
> Also: Regarding re-basing patches: It has to be the duty of the developer
> of the patch to re-base it to current trees if it will no longer apply
> cleanly.
I think this misses the point. I've re-based some of my patches more
than half-a-dozen times, and this has gotten so tedious that I've
just sort of stopped bothering sending in patches. Excessive delays
in moving patches upstream just kills the development process.
Patches need to be handled in a timely manner, while they are still
'fresh', so that they don't need to be rebased.
Put it another way: it is, at this time, impossible for me to rebase,
because I know that my patches will conflict with others in the
un-applied patch queue. So all I can do is wait for the patch queue to
shrink, wait till the others get into the Torvalds tree, then bk pull,
then hurry, hurry, rebase, test, submit, and hope I get in before
someone else does and wrecks it again. The turn-around time for
"getting lucky" like this is over a month, and if one doesn't get
lucky the first month, one has to wait a whole 'nother month for
one's next shot.
--linas
More information about the Linuxppc64-dev
mailing list