Re: PATCH 2.3.23 pre 2 compile fixes

Donald Becker (becker@cesdis1.gsfc.nasa.gov)
Tue, 19 Oct 1999 18:46:10 -0400 (EDT)


On Tue, 19 Oct 1999, Linus Torvalds wrote:
> > But I wonder why I put in the effort -- all of my work is thrown away when
> > you accept random, tested-only-on-one-machine patches from anywhere, and
> > then refuse to put in the tested, updated drivers.
>
> Donald, your latest "tested" drivers have been a QA disaster, and you
> should know it.

In August the story was "I won't accept drivers with kernel-compatibility
defines" and "put the very latest versions into the 2.3 kernels, or they
won't get into 2.4". I did that with the "2.3" series drivers, which were
the very latest (test, not "release" versions) at the time.

> - you don't send me patches in a timely manner. There are drivers that

>From my side, I send an updated driver, you report "it didn't work for
someone" (and that person never sent a problem report to the driver mailing
list), and it doesn't go in.

I'm trying to save you work, and have a broader early-test base, by having a
separate test process that doesn't require people to update to the latest,
probably-broken development kernel. You should *expect* that a driver has a
three or six month old date on it.

> - very occasionally your new "stable" drivers won't even work on the two
> or three machines I test on. This has sometimes meant that the eepro100
> driver in particular has lagged behind a _lot_, for the simple reason
> that newer drivers did not work at all for me.

I'm running 50 eepro100 interfaces in a cluster with the new driver. The
old driver in the kernel has known bugs. The mailing list is active.

[[ Intel is a bad case. They claim to be chummy with RedHat and other, yet
won't give me the time of day let alone errata lists and programming info on
the chips. ]]

> The "not send timely patches" thing applies to the PCI code in question. I
> have never EVER seen a patch from you to implement the PCI search
> function. Not now, not a year ago, not EVER.
>
> And then you complain when I apply patches that ARE sent to me? Get a
> grip, Donald.

I'm the one that has to sort out the bug reports of "version 1.23" with "do
you mean v1.23 of two years ago with five random patches that aren't in my
CVS tree, or with only four random patches".

Donald Becker
Scyld Computing Corporation, and
USRA-CESDIS, becker@cesdis.gsfc.nasa.gov

-
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/