From: Frederic Weisbecker <fweisbec@gmail.com>
To: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>, Damien Wyart <damien.wyart@free.fr>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Mike Galbraith <efault@gmx.de>, "Rafael J. Wysocki" <rjw@sisk.pl>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Kernel Testers List <kernel-testers@vger.kernel.org>
Subject: Re: [Bug #12650] Strange load average and ksoftirqd behavior with 2.6.29-rc2-git1
Date: Tue, 17 Feb 2009 05:34:23 +0100 [thread overview]
Message-ID: <20090217043422.GA5836@nowhere> (raw)
In-Reply-To: <20090216223944.GF6785@linux.vnet.ibm.com>
On Mon, Feb 16, 2009 at 02:39:44PM -0800, Paul E. McKenney wrote:
> On Mon, Feb 16, 2009 at 09:09:23PM +0100, Ingo Molnar wrote:
> >
> > * Paul E. McKenney <paulmck@linux.vnet.ibm.com> wrote:
> >
> > > Here the calls to rcu_process_callbacks() are only 75
> > > microseconds apart, so that this function is consuming more
> > > than 10% of a CPU. The strange thing is that I don't see a
> > > raise_softirq() in between, though perhaps it gets inlined or
> > > something that makes it invisible to ftrace.
> >
> > look at the latest trace please, that has even the most inline
> > raise-softirq method instrumented, so all the raising is
> > visible.
>
> Ah, my apologies! This time looking at:
>
> http://damien.wyart.free.fr/ksoftirqd_pb/trace_tip_2009.02.16_ksoftirqd_pb_abstime_proc.txt.gz
>
>
> 799.521187 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.521371 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.521555 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.521738 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.521934 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.522068 | 1) ksoftir-2324 | | rcu_check_callbacks() {
> 799.522208 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.522392 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.522575 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.522759 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.522956 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.523074 | 1) ksoftir-2324 | | rcu_check_callbacks() {
> 799.523214 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.523397 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.523579 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.523762 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.523960 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.524079 | 1) ksoftir-2324 | | rcu_check_callbacks() {
> 799.524220 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.524403 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.524587 | 1) <idle>-0 | | rcu_check_callbacks() {
> 799.524770 | 1) <idle>-0 | | rcu_check_callbacks() {
> [ . . . ]
>
> Yikes!!!
>
> Why is rcu_check_callbacks() being invoked so often? It should be called
> but once per jiffy, and here it is called no less than 22 times in about
> 3.5 milliseconds, meaning one call every 160 microseconds or so.
>
> Hmmm...
>
> Looks like we never return from:
>
> 799.521142 | 1) <idle>-0 | | tick_nohz_stop_sched_tick() {
>
> Perhaps we are taking an interrupt immediately after the
> local_irq_restore()? And at 799.521209 deciding to exit nohz mode.
> And then deciding to go back into nohz mode at 799.521326, 117
> microseconds later, after which we re-invoke rcu_check_callbacks(),
> which again raises RCU's softirq.
>
> And the reason we are invoking rcu_check_callbacks() so often appears
> to be in in arch/x86/kernel/process_32.c cpu_idle() near line 107,
> which explains my failure to reproduce on a 64-bit system:
>
> void cpu_idle(void)
> {
> int cpu = smp_processor_id();
>
> current_thread_info()->status |= TS_POLLING;
>
> /* endless idle loop with no priority at all */
> while (1) {
> tick_nohz_stop_sched_tick(1);
> while (!need_resched()) {
>
> check_pgt_cache();
> rmb();
>
> if (rcu_pending(cpu))
> rcu_check_callbacks(cpu, 0);
>
> if (cpu_is_offline(cpu))
> play_dead();
>
> local_irq_disable();
> __get_cpu_var(irq_stat).idle_timestamp = jiffies;
> /* Don't trace irqs off for idle */
> stop_critical_timings();
> pm_idle();
> start_critical_timings();
> }
> tick_nohz_restart_sched_tick();
> preempt_enable_no_resched();
> schedule();
> preempt_disable();
> }
> }
>
> If we go in and out of nohz mode quickly, we will invoke rcu_pending()
> each time. I would expect rcu_pending() to return 0 most of the time,
> but that apparently isn't the case with treercu...
>
> What is the easiest way for me to make it easy to trace the return path
> from __rcu_pending()? Make each return path call an empty function
> located off where the compiler cannot see it, I guess... Diagnostic
> patch along these lines below. Frederic, Damien, could you please give
> it a go? (And of course please let me know if something else is
> needed.)
No, you don't need that, you can use ftrace_printk, it will generate a C-comment like
inside the functions, ie:
__rcu_pending() {
/* pending_qs */
}
I've converted your below patch with ftrace_printks and tested it under an old P2
with rcu_tree and 1000 Hz. I made a trace during an idle state, and well, looks like I'm
lucky :-)
I guess I successfully reproduced the softirq/rcu overhead.
Please find the below patch to trace the rcu_pending return path, as well as the trace I made.
Sorry, the trace is a bit buggy with sometimes flying orphans C like comments.
When I will have more time, I will fix that.
The trace is here http://dl.free.fr/uyWGgCbx4
It looks like it mostly returns 1 because of the waiting for quiescent state:
$ cat rcutrace | grep "/* pending_none" | wc -l
221
$ cat rcutrace | grep "/* pending_qs" | wc -l
248
$ cat rcutrace | grep "/* pending" | wc -l
469
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index b2fd602..c9e78f6 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -45,6 +45,7 @@
#include <linux/cpu.h>
#include <linux/mutex.h>
#include <linux/time.h>
+#include <linux/ftrace.h>
#ifdef CONFIG_DEBUG_LOCK_ALLOC
static struct lock_class_key rcu_lock_key;
@@ -1249,31 +1250,44 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp)
check_cpu_stall(rsp, rdp);
/* Is the RCU core waiting for a quiescent state from this CPU? */
- if (rdp->qs_pending)
+ if (rdp->qs_pending) {
+ ftrace_printk("pending_qs\n");
return 1;
+ }
/* Does this CPU have callbacks ready to invoke? */
- if (cpu_has_callbacks_ready_to_invoke(rdp))
+ if (cpu_has_callbacks_ready_to_invoke(rdp)) {
+ ftrace_printk("pending_ready_invoke\n");
return 1;
+ }
/* Has RCU gone idle with this CPU needing another grace period? */
- if (cpu_needs_another_gp(rsp, rdp))
+ if (cpu_needs_another_gp(rsp, rdp)) {
+ ftrace_printk("pending_gp\n");
return 1;
+ }
/* Has another RCU grace period completed? */
- if (ACCESS_ONCE(rsp->completed) != rdp->completed) /* outside of lock */
+ if (ACCESS_ONCE(rsp->completed) != rdp->completed) {/* outside of lock */
+ ftrace_printk("pending_gp_completed\n");
return 1;
+ }
/* Has a new RCU grace period started? */
- if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) /* outside of lock */
+ if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) { /* outside of lock */
+ ftrace_printk("pending_gp_new_started\n");
return 1;
+ }
/* Has an RCU GP gone long enough to send resched IPIs &c? */
if (ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum) &&
((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0 ||
- (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0))
+ (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0)) {
+ ftrace_printk("pending_ipi\n");
return 1;
+ }
+ ftrace_printk("pending_none\n");
/* nothing to do */
return 0;
}
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> ---
>
> rcupdate.c | 23 +++++++++++++++++++++++
> rcutree.c | 31 +++++++++++++++++++++++++------
> 2 files changed, 48 insertions(+), 6 deletions(-)
>
> diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
> index d92a76a..42bbf03 100644
> --- a/kernel/rcupdate.c
> +++ b/kernel/rcupdate.c
> @@ -175,3 +175,26 @@ void __init rcu_init(void)
> __rcu_init();
> }
>
> +void __rcu_pending_qs_pending(void)
> +{
> +}
> +
> +void __rcu_pending_callbacks_ready(void)
> +{
> +}
> +
> +void __rcu_pending_needs_gp(void)
> +{
> +}
> +
> +void __rcu_pending_new_completed(void)
> +{
> +}
> +
> +void __rcu_pending_new_gp(void)
> +{
> +}
> +
> +void __rcu_pending_fqs(void)
> +{
> +}
> diff --git a/kernel/rcutree.c b/kernel/rcutree.c
> index b2fd602..e2d72c3 100644
> --- a/kernel/rcutree.c
> +++ b/kernel/rcutree.c
> @@ -1234,6 +1234,13 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
> }
> EXPORT_SYMBOL_GPL(call_rcu_bh);
>
> +extern void __rcu_pending_qs_pending(void);
> +extern void __rcu_pending_callbacks_ready(void);
> +extern void __rcu_pending_needs_gp(void);
> +extern void __rcu_pending_new_completed(void);
> +extern void __rcu_pending_new_gp(void);
> +extern void __rcu_pending_fqs(void);
> +
> /*
> * Check to see if there is any immediate RCU-related work to be done
> * by the current CPU, for the specified type of RCU, returning 1 if so.
> @@ -1249,30 +1256,42 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp)
> check_cpu_stall(rsp, rdp);
>
> /* Is the RCU core waiting for a quiescent state from this CPU? */
> - if (rdp->qs_pending)
> + if (rdp->qs_pending) {
> + __rcu_pending_qs_pending();
> return 1;
> + }
>
> /* Does this CPU have callbacks ready to invoke? */
> - if (cpu_has_callbacks_ready_to_invoke(rdp))
> + if (cpu_has_callbacks_ready_to_invoke(rdp)) {
> + __rcu_pending_callbacks_ready();
> return 1;
> + }
>
> /* Has RCU gone idle with this CPU needing another grace period? */
> - if (cpu_needs_another_gp(rsp, rdp))
> + if (cpu_needs_another_gp(rsp, rdp)) {
> + __rcu_pending_needs_gp();
> return 1;
> + }
>
> /* Has another RCU grace period completed? */
> - if (ACCESS_ONCE(rsp->completed) != rdp->completed) /* outside of lock */
> + if (ACCESS_ONCE(rsp->completed) != rdp->completed) /* outside of lock */ {
> + __rcu_pending_new_completed();
> return 1;
> + }
>
> /* Has a new RCU grace period started? */
> - if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) /* outside of lock */
> + if (ACCESS_ONCE(rsp->gpnum) != rdp->gpnum) /* outside of lock */ {
> + __rcu_pending_new_gp();
> return 1;
> + }
>
> /* Has an RCU GP gone long enough to send resched IPIs &c? */
> if (ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum) &&
> ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0 ||
> - (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0))
> + (rdp->n_rcu_pending_force_qs - rdp->n_rcu_pending) < 0)) {
> + __rcu_pending_fqs();
> return 1;
> + }
>
> /* nothing to do */
> return 0;
next prev parent reply other threads:[~2009-02-17 4:34 UTC|newest]
Thread overview: 131+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-02-14 20:35 2.6.29-rc5: Reported regressions from 2.6.28 Rafael J. Wysocki
2009-02-14 20:35 ` [Bug #12414] iwl4965 cannot use "ap auto" on latest 2.6.28/29? Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12444] X hangs following switch from radeonfb console - Bisected Rafael J. Wysocki
2009-02-16 17:52 ` Graham Murray
2009-02-16 21:52 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12418] Repeated ioctl(4, 0x40046445, ..) loop in glxgears Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12490] ath5k related kernel panic in 2.6.29-rc1 Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12419] possible circular locking dependency on i915 dma Rafael J. Wysocki
2009-02-16 3:50 ` Wang Chen
2009-02-14 20:38 ` [Bug #12497] new barrier warnings in 2.6.29-rc1 Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12496] swsusp cannot find resume device (sometimes) Rafael J. Wysocki
2009-02-15 0:05 ` Arjan van de Ven
2009-02-15 14:23 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12491] i915 lockdep warning Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12494] Sony backlight regression from 2.6.28 to 29-rc Rafael J. Wysocki
2009-02-17 10:51 ` Norbert Preining
2009-02-14 20:38 ` [Bug #12499] Problem with using bluetooth adaper connected to usb port Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12501] build bug in eeepc-laptop.c Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12502] pipe_read oops on sh Rafael J. Wysocki
2009-02-15 0:23 ` Adrian McMenamin
2009-02-15 14:27 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12510] 2.6.29-rc2 dies on startup Rafael J. Wysocki
2009-02-16 21:02 ` Ferenc Wagner
2009-02-16 21:12 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12551] end_request: I/O error, dev cciss/c0d0, sector 87435720 Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12610] sync-Regression in 2.6.28.2? Rafael J. Wysocki
2009-02-21 17:56 ` Theodore Tso
2009-02-22 10:02 ` Rafael J. Wysocki
2009-02-23 4:35 ` Greg KH
2009-02-23 5:37 ` Theodore Tso
2009-02-23 16:54 ` [stable] " Greg KH
2009-02-14 20:38 ` [Bug #12609] v2.6.29-rc2 libata sff 32bit PIO regression Rafael J. Wysocki
2009-02-15 4:20 ` Larry Finger
2009-02-15 8:10 ` Jeff Garzik
2009-02-15 12:05 ` Sergei Shtylyov
2009-02-15 16:48 ` Hugh Dickins
2009-02-14 20:38 ` [Bug #12574] possible circular locking dependency detected Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12571] Suspend-resume on Dell Latitude D410 newly broken in 2.6.29-rc* Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12615] boot hangs while bringing up gianfar ethernet Rafael J. Wysocki
2009-02-15 14:42 ` Peter Korsgaard
2009-02-15 21:08 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12613] [Suspend regression][DRM, RADEON] Rafael J. Wysocki
[not found] ` <4997E7D7.60205@numericable.fr>
2009-02-15 10:20 ` etienne
2009-02-14 20:38 ` [Bug #12659] Failure to resume two Sandisk USB flash drives attached to a Belkin USB Busport Mobile (F5U022) Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12617] unable to compile e100 firmware into kernel Rafael J. Wysocki
2009-02-15 17:38 ` David Woodhouse
2009-02-15 19:58 ` Andrey Borzenkov
2009-02-15 21:09 ` Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12618] hackbench [pthread mode] regression with 2.6.29-rc3 Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12650] Strange load average and ksoftirqd behavior with 2.6.29-rc2-git1 Rafael J. Wysocki
2009-02-15 8:09 ` Damien Wyart
2009-02-15 9:00 ` Ingo Molnar
2009-02-15 9:51 ` Damien Wyart
2009-02-15 10:13 ` Ingo Molnar
2009-02-15 10:34 ` Damien Wyart
2009-02-15 10:41 ` Damien Wyart
2009-02-15 10:42 ` Damien Wyart
2009-02-15 10:43 ` Damien Wyart
2009-02-15 11:01 ` Ingo Molnar
2009-02-15 14:06 ` Frederic Weisbecker
2009-02-15 18:03 ` Damien Wyart
2009-02-15 19:18 ` Damien Wyart
2009-02-15 19:31 ` Ingo Molnar
2009-02-16 8:42 ` Damien Wyart
2009-02-16 9:21 ` Ingo Molnar
2009-02-16 10:49 ` Damien Wyart
2009-02-16 9:25 ` Ingo Molnar
2009-02-16 9:27 ` Ingo Molnar
2009-02-16 9:32 ` Ingo Molnar
2009-02-16 9:50 ` Ingo Molnar
2009-02-16 11:56 ` Damien Wyart
2009-02-16 12:26 ` Ingo Molnar
2009-02-16 13:02 ` Damien Wyart
2009-02-16 13:21 ` Ingo Molnar
2009-02-16 16:06 ` Paul E. McKenney
2009-02-16 18:56 ` Paul E. McKenney
2009-02-16 19:08 ` Frederic Weisbecker
2009-02-16 20:02 ` Frederic Weisbecker
2009-02-16 21:31 ` Paul E. McKenney
2009-02-16 20:09 ` Ingo Molnar
2009-02-16 22:39 ` Paul E. McKenney
2009-02-16 22:51 ` Paul E. McKenney
2009-02-17 9:46 ` Ingo Molnar
2009-02-17 14:01 ` Paul E. McKenney
2009-02-17 15:39 ` Damien Wyart
2009-02-17 16:05 ` Paul E. McKenney
2009-02-17 21:48 ` Ingo Molnar
2009-02-17 4:34 ` Frederic Weisbecker [this message]
2009-02-17 15:10 ` Paul E. McKenney
2009-02-17 16:00 ` Frederic Weisbecker
2009-02-17 22:37 ` Frederic Weisbecker
2009-02-17 22:48 ` Paul E. McKenney
2009-02-18 0:38 ` Ingo Molnar
2009-02-18 1:02 ` Paul E. McKenney
2009-02-17 6:11 ` Damien Wyart
2009-02-17 15:11 ` Paul E. McKenney
2009-02-16 20:44 ` Damien Wyart
2009-02-15 10:12 ` Christian Kujau
2009-02-15 10:54 ` Ingo Molnar
2009-02-14 20:38 ` [Bug #12668] USB flash disk surprise disconnect Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12663] Commit 8c7e58e690ae60ab4215b025f433ed4af261e103 breaks resume Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12660] Linux 2.6.28.3 freezing on a 32-bits x86 Thinkpad T43p Rafael J. Wysocki
2009-02-14 23:29 ` Mathieu Desnoyers
2009-02-14 20:38 ` [Bug #12680] Not having a VIA PadLock hardware incurs a long delay in probing on modules insertion attempt Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12681] s2ram: fails to wake up on Acer Extensa 4220 (SMP disabled) Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12670] BUG: unable to handle kernel paging request at pin_to_kill+0x21 Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12671] uvc_status_cleanup(): undefined reference to `input_unregister_device' Rafael J. Wysocki
2009-02-14 20:38 ` [Bug #12705] X200: Brightness broken since 2.6.29-rc4-58-g4c098bc Rafael J. Wysocki
2009-02-15 13:43 ` Matthew Garrett
2009-02-15 14:37 ` Rafael J. Wysocki
2009-02-17 23:05 ` Eric Anholt
2009-02-17 23:13 ` Matthew Garrett
2009-02-17 23:23 ` Jesse Barnes
2009-02-18 9:36 ` Nico Schottelius
2009-02-13 9:33 ` Nico Schottelius
2009-02-13 9:40 ` Nico Schottelius
2009-02-13 13:43 ` Matthew Garrett
2009-03-10 2:28 ` Eric Anholt
2009-03-10 5:38 ` Nico Schottelius
2009-02-13 9:42 ` Ingo Molnar
2009-02-13 18:05 ` Len Brown
2009-02-16 9:06 ` ZhangRui
2009-02-16 10:58 ` Nico Schottelius
2009-02-16 13:13 ` Nico Schottelius
2009-02-16 21:40 ` Norbert Preining
2009-02-16 15:54 ` Nico Schottelius
2009-02-19 9:01 ` Nico Schottelius
2009-02-14 20:38 ` [Bug #12706] Oopses and ACPI problems (Linus 2.6.29-rc4) Rafael J. Wysocki
2009-02-16 7:29 ` 2.6.29-rc5: Reported regressions from 2.6.28 Jarek Poplawski
2009-02-16 21:11 ` Rafael J. Wysocki
-- strict thread matches above, loose matches on Subject: below --
2009-02-08 19:05 2.6.29-rc4: " Rafael J. Wysocki
2009-02-08 19:21 ` [Bug #12650] Strange load average and ksoftirqd behavior with 2.6.29-rc2-git1 Rafael J. Wysocki
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=20090217043422.GA5836@nowhere \
--to=fweisbec@gmail.com \
--cc=a.p.zijlstra@chello.nl \
--cc=damien.wyart@free.fr \
--cc=efault@gmx.de \
--cc=kernel-testers@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=paulmck@linux.vnet.ibm.com \
--cc=rjw@sisk.pl \
/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;
as well as URLs for NNTP newsgroup(s).