From: Al Viro <viro@zeniv.linux.org.uk>
To: Bob Gill <gillb5@telus.net>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Can we get a bootable 6.11-rc kernel?
Date: Mon, 9 Sep 2024 05:36:45 +0100 [thread overview]
Message-ID: <20240909043645.GK1049718@ZenIV> (raw)
In-Reply-To: <96030dde-6c02-4308-b41b-48aeeba670f3@telus.net>
On Sun, Sep 08, 2024 at 10:04:27PM -0600, Bob Gill wrote:
> Hi. It would be nice to have a 6.11-rc kernel that doesn't crash. I have
> an AMD video card. I can boot into recovery, but any time I try to run
> "service lightdm start" or startx, I get a nice black screen, and a within a
> second or 2 my keyboard doesn't even let me type the caps-lock key.
git bisect? If it's a couple of minutes per test, should be doable within
reasonable time (depends upon your .config - if it's allmodconfig, each
build will take quite a while on early stages of bisection; if it's trimmed
down to your box, that's a couple of minutes per build). In any case, there
shouldn't be all that many bisection steps; if 6.10 works and 6.11-rc1 doesn't,
you are looking at something under 15 builds, at a guess...
Clone the mainline tree, then
; git bisect start
status: waiting for both good and bad commits
; git bisect good v6.10
status: waiting for bad commit, 1 good commit known
; git bisect bad v6.11-rc1
Bisecting: 7446 revisions left to test after this (roughly 13 steps)
[280e36f0d5b997173d014c07484c03a7f7750668] nsfs: use cleanup guard
;
then keep building the trees and testing them. E.g.
; cp config-I-want .config
; yes ""|make oldconfig
; make -j8 bindeb-pkg
then install the resulting .deb - probably ought to work on ubuntu boxen same
as on debian ones; you might need to install some stuff, depending upon your
config, but something like apt build-dep linux would probably be a usable
starting point.
No matter how you build them, build and test, then say either
; git bisect bad
or
; git bisect good
depending upon the test result. As long as the breakage is reliably
caught by your test, you should arrive at the commit that introduced it.
That would make things much easier to deal with - at least you'll know
who's the likely suspect.
Of course, if the breakage is _not_ reliably caught, you risk going in the
wrong direction and arriving at something that clearly has nothing to do
with your problem, but it sounds like the breakage is not of that sort...
next prev parent reply other threads:[~2024-09-09 4:36 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-09 4:04 Can we get a bootable 6.11-rc kernel? Bob Gill
2024-09-09 4:36 ` Al Viro [this message]
2024-09-09 4:48 ` Linux regression tracking (Thorsten Leemhuis)
2024-09-15 20:59 ` 6.11.0 crash on X startup Bob Gill
2024-09-16 2:43 ` Bob Gill
2024-09-16 11:12 ` Dr. David Alan Gilbert
2024-09-16 16:00 ` Bob Gill
2024-09-16 17:20 ` Dr. David Alan Gilbert
2024-09-16 20:30 ` Bob Gill
2024-09-16 20:36 ` Dr. David Alan Gilbert
2024-09-16 21:25 ` Bob Gill
2024-09-16 23:03 ` Dr. David Alan Gilbert
2024-09-17 2:01 ` Bob Gill
2024-09-17 15:55 ` Dr. David Alan Gilbert
2024-09-17 16:39 ` Bob Gill
2024-09-17 18:02 ` Dr. David Alan Gilbert
2024-09-17 17:59 ` Bob Gill
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240909043645.GK1049718@ZenIV \
--to=viro@zeniv.linux.org.uk \
--cc=gillb5@telus.net \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox