From: Rusty Russell <rusty@rustcorp.com.au>
To: Zoltan Boszormenyi <zboszor@freemail.hu>
Cc: lkml - Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: kernel/softirq.c issues under 2.6.5
Date: Tue, 27 Apr 2004 13:25:56 +1000 [thread overview]
Message-ID: <1083036355.2150.10.camel@bach> (raw)
In-Reply-To: <408A48BB.9040804@freemail.hu>
On Sat, 2004-04-24 at 21:00, Zoltan Boszormenyi wrote:
> Hi,
>
> > On Tue, 2004-04-20 at 17:58, Aivils wrote:
> >> Hi all!
> >>
> >> My 2.6.5 will not start until i applay patch bellow:
...
> This changes caused that the linuxconsole.sf.net patch against
> newer kernels (>2.6.3) did not boot. The .config is
> from 2.6.6-rc2-mm1 + changes from linuxconsole.sf.net patch.
> Do you want the whole patch to investigate?
No, but the bug is probably that patch submitting a softirq before it's
initialized. This patch should find it: look for the backtrace.
Name: Debug Too-Early Use of softirqs
Status: Untested
diff -urpN --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal .31025-linux-2.6.6-rc2-bk4/kernel/softirq.c .31025-linux-2.6.6-rc2-bk4.updated/kernel/softirq.c
--- .31025-linux-2.6.6-rc2-bk4/kernel/softirq.c 2004-04-22 08:04:01.000000000 +1000
+++ .31025-linux-2.6.6-rc2-bk4.updated/kernel/softirq.c 2004-04-27 13:21:57.000000000 +1000
@@ -190,14 +190,18 @@ struct tasklet_head
/* Some compilers disobey section attribute on statics when not
initialized -- RR */
-static DEFINE_PER_CPU(struct tasklet_head, tasklet_vec) = { NULL };
-static DEFINE_PER_CPU(struct tasklet_head, tasklet_hi_vec) = { NULL };
+static DEFINE_PER_CPU(struct tasklet_head, tasklet_vec) = { (void *)1 };
+static DEFINE_PER_CPU(struct tasklet_head, tasklet_hi_vec) = { (void *)1 };
void fastcall __tasklet_schedule(struct tasklet_struct *t)
{
unsigned long flags;
local_irq_save(flags);
+ if (__get_cpu_var(tasklet_vec).list == (void *)1) {
+ WARN_ON(1);
+ __get_cpu_var(tasklet_vec).list = NULL;
+ }
t->next = __get_cpu_var(tasklet_vec).list;
__get_cpu_var(tasklet_vec).list = t;
raise_softirq_irqoff(TASKLET_SOFTIRQ);
@@ -211,6 +215,10 @@ void fastcall __tasklet_hi_schedule(stru
unsigned long flags;
local_irq_save(flags);
+ if (__get_cpu_var(tasklet_hi_vec).list == (void *)1) {
+ WARN_ON(1);
+ __get_cpu_var(tasklet_hi_vec).list = NULL;
+ }
t->next = __get_cpu_var(tasklet_hi_vec).list;
__get_cpu_var(tasklet_hi_vec).list = t;
raise_softirq_irqoff(HI_SOFTIRQ);
@@ -423,8 +431,8 @@ static int __devinit cpu_callback(struct
switch (action) {
case CPU_UP_PREPARE:
- BUG_ON(per_cpu(tasklet_vec, hotcpu).list);
- BUG_ON(per_cpu(tasklet_hi_vec, hotcpu).list);
+ per_cpu(tasklet_vec, hotcpu).list = NULL;
+ per_cpu(tasklet_hi_vec, hotcpu).list = NULL;
p = kthread_create(ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu);
if (IS_ERR(p)) {
printk("ksoftirqd for %i failed\n", hotcpu);
--
Anyone who quotes me in their signature is an idiot -- Rusty Russell
next prev parent reply other threads:[~2004-04-27 3:26 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-04-24 11:00 kernel/softirq.c issues under 2.6.5 Zoltan Boszormenyi
2004-04-27 3:25 ` Rusty Russell [this message]
-- strict thread matches above, loose matches on Subject: below --
2004-04-20 7:58 Aivils
2004-04-20 22:21 ` Rusty Russell
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=1083036355.2150.10.camel@bach \
--to=rusty@rustcorp.com.au \
--cc=linux-kernel@vger.kernel.org \
--cc=zboszor@freemail.hu \
/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