From: William Lee Irwin III <wli@holomorphy.com>
To: Jesse Barnes <jbarnes@engr.sgi.com>,
Andrew Morton <akpm@osdl.org>,
linux-kernel@vger.kernel.org, Ingo Molnar <mingo@elte.hu>,
Nick Piggin <nickpiggin@yahoo.com.au>
Subject: Re: 2.6.8-rc3-mm2
Date: Mon, 9 Aug 2004 14:10:42 -0700 [thread overview]
Message-ID: <20040809211042.GY11200@holomorphy.com> (raw)
In-Reply-To: <20040809204357.GX11200@holomorphy.com>
On Mon, Aug 09, 2004 at 12:17:50PM -0700, Jesse Barnes wrote:
>>> If I apply everything up to and including schedstat-v10.patch, it
>>> boots fine. So it might be sched-init_idle-fork_by_hand-consolidation.patch
>>> or something nearby... Just reverting the sched-single-array.patch
>>> wasn't enough.
On Mon, Aug 09, 2004 at 12:53:23PM -0700, William Lee Irwin III wrote:
>> I can reproduce here (ia64/Altix). Fixing.
On Mon, Aug 09, 2004 at 01:43:57PM -0700, William Lee Irwin III wrote:
> It comes up with the following applied. Now distilling into a bugfix.
The following does *NOT* come up. The difference appears to be the delay
from the printk()'s.
-- wli
--- mm2-2.6.8-rc3/arch/ia64/kernel/smpboot.c.orig 2004-08-10 05:24:20.000000000 -0700
+++ mm2-2.6.8-rc3/arch/ia64/kernel/smpboot.c 2004-08-10 06:50:21.774710311 -0700
@@ -356,44 +356,17 @@
return cpu_idle();
}
-struct create_idle {
- struct task_struct *idle;
- struct completion done;
- int cpu;
-};
-
-void
-do_fork_idle(void *_c_idle)
-{
- struct create_idle *c_idle = _c_idle;
-
- c_idle->idle = fork_idle(c_idle->cpu);
- complete(&c_idle->done);
-}
-
static int __devinit
do_boot_cpu (int sapicid, int cpu)
{
int timeout;
- struct create_idle c_idle = {
- .cpu = cpu,
- .done = COMPLETION_INITIALIZER(c_idle.done),
- };
- DECLARE_WORK(work, do_fork_idle, &c_idle);
/*
* We can't use kernel_thread since we must avoid to reschedule the child.
*/
- if (!keventd_up() || current_is_keventd())
- work.func(work.data);
- else {
- schedule_work(&work);
- wait_for_completion(&c_idle.done);
- }
-
- if (IS_ERR(c_idle.idle))
+ task_for_booting_cpu = fork_idle(cpu);
+ if (IS_ERR(task_for_booting_cpu))
panic("failed fork for CPU %d", cpu);
- task_for_booting_cpu = c_idle.idle;
-
+ mb();
Dprintk("Sending wakeup vector %lu to AP 0x%x/0x%x.\n", ap_wakeup_vector, cpu, sapicid);
platform_send_ipi(cpu, ap_wakeup_vector, IA64_IPI_DM_INT, 0);
--- mm2-2.6.8-rc3/kernel/fork.c.orig 2004-08-10 06:02:27.000000000 -0700
+++ mm2-2.6.8-rc3/kernel/fork.c 2004-08-10 06:03:11.000000000 -0700
@@ -1190,7 +1190,7 @@
goto fork_out;
}
-task_t * __init fork_idle(int cpu)
+task_t *fork_idle(int cpu)
{
task_t *task;
struct pt_regs regs;
next prev parent reply other threads:[~2004-08-09 21:11 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-08-08 22:29 2.6.8-rc3-mm2 Andrew Morton
2004-08-08 22:55 ` 2.6.8-rc3-mm2 Brice Goglin
2004-08-08 23:04 ` 2.6.8-rc3-mm2 Andrew Morton
2004-08-08 23:17 ` 2.6.8-rc3-mm2 Andrew Morton
2004-08-09 2:17 ` [patch] 2.6.8-rc3-mm2: sk98lin/skge.c doesn't compile with PROC_FS=n Adrian Bunk
2004-08-09 5:14 ` 2.6.8-rc3-mm2 (compile stats) John Cherry
2004-08-09 8:20 ` 2.6.8-rc3-mm2 Rick Lindsley
2004-08-09 8:25 ` 2.6.8-rc3-mm2 Andrew Morton
2004-08-09 8:29 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 8:33 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 8:43 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 8:49 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 10:36 ` 2.6.8-rc3-mm2 Espen Fjellvær Olsen
2004-08-09 12:14 ` 2.6.8-rc3-mm2 Vladimir V. Saveliev
2004-08-09 17:51 ` 2.6.8-rc3-mm2 Hans Reiser
2004-08-09 17:50 ` 2.6.8-rc3-mm2 Johannes Stezenbach
2004-08-09 18:22 ` 2.6.8-rc3-mm2 Jesse Barnes
2004-08-09 18:25 ` 2.6.8-rc3-mm2 Andrew Morton
2004-08-09 18:32 ` 2.6.8-rc3-mm2 Jesse Barnes
2004-08-09 19:17 ` 2.6.8-rc3-mm2 Jesse Barnes
2004-08-09 19:53 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 20:43 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-09 21:10 ` William Lee Irwin III [this message]
2004-08-09 22:45 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 6:34 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 8:02 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 8:30 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 8:56 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 9:00 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 9:06 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 8:04 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 8:08 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 8:17 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 8:23 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 9:00 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 9:38 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 10:02 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 11:53 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 12:52 ` 2.6.8-rc3-mm2 V13
2004-08-10 12:51 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 12:55 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 12:56 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 13:01 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 13:10 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 13:12 ` 2.6.8-rc3-mm2 Russell King
2004-08-10 13:16 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 13:39 ` 2.6.8-rc3-mm2 Russell King
2004-08-10 13:48 ` 2.6.8-rc3-mm2 Ingo Molnar
2004-08-10 6:32 ` 2.6.8-rc3-mm2 William Lee Irwin III
2004-08-10 8:34 ` 2.6.8-rc3-mm2 Rick Lindsley
2004-08-10 14:55 ` 2.6.8-rc3-mm2 Martin J. Bligh
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=20040809211042.GY11200@holomorphy.com \
--to=wli@holomorphy.com \
--cc=akpm@osdl.org \
--cc=jbarnes@engr.sgi.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=nickpiggin@yahoo.com.au \
/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