From: "Alex Züpke" <alexander.zuepke@hs-rm.de>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] QEMU ARM SMP: IPI delivery delayed until next main loop event // how to improve IPI latency?
Date: Fri, 12 Jun 2015 18:38:57 +0200 [thread overview]
Message-ID: <557B0B21.2030009@hs-rm.de> (raw)
Hi,
I'm benchmarking some IPI (== inter-processor-interrupt) synchronization stuff of my custom kernel on QEMU ARM (qemu-system-arm -M vexpress-a15 -smp 2) and ran into the following problem: pending IPIs are delayed until the QEMU main loop receives an event (for example the timer interrupt expires or I press a key on the console).
The following timing diagram tries to show this:
CPU #0 CPU #1
====== ======
... other stuff ... WFI (wait for interrupt, like x86 "HLT")
send SGI in MPCore
polls for completeness
<time passes ...>
polls ...
<... and passes ...>
still polls ...
<... and passes ...>
still polls ...
<... and passes ...>
<timer interrupt expires>
<now QEMU switches to CPU #1>
receives IPI
signals completeness
WFI
<QEMU switches to CPU #0>
polling done
process timer interrupt
...
My timer is setup to generate an interrupt once a second, so I only get 1 IPI interrupt per second on QEMU. When I run the test on real hardware (i.MX6Q), I get millions of IPIs instead.
I tried to "fix" this by forcing QEMU back into the main loop and added a call to qemu_notify_event() in the IPI-sending path of the ARM interrupt controller:
diff --git a/hw/intc/arm_gic.c b/hw/intc/arm_gic.c
index c1d2e70..20dba75 100644
--- a/hw/intc/arm_gic.c
+++ b/hw/intc/arm_gic.c
@@ -21,6 +21,7 @@
#include "hw/sysbus.h"
#include "gic_internal.h"
#include "qom/cpu.h"
+#include "qemu/main-loop.h"
//#define DEBUG_GIC
@@ -898,6 +899,7 @@ static void gic_dist_writel(void *opaque, hwaddr offset,
target_cpu = ctz32(mask);
}
gic_update(s);
+ qemu_notify_event();
return;
}
gic_dist_writew(opaque, offset, value & 0xffff, attrs);
It works as expects (I get thousands of IPIs per second now), but it does not "feel right", so is there a better way to improve the responsiveness of IPI handling in QEMU?
Best regards
Alex
next reply other threads:[~2015-06-12 16:39 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-12 16:38 Alex Züpke [this message]
2015-06-12 18:03 ` [Qemu-devel] QEMU ARM SMP: IPI delivery delayed until next main loop event // how to improve IPI latency? Peter Maydell
2015-06-15 14:44 ` Alex Züpke
2015-06-15 14:51 ` Peter Maydell
2015-06-15 15:05 ` Alex Züpke
2015-06-15 18:41 ` Peter Maydell
2015-06-15 18:58 ` Peter Maydell
2015-06-15 20:03 ` Alex Zuepke
2015-06-16 10:33 ` Peter Maydell
2015-06-16 10:59 ` Peter Maydell
2015-06-16 11:11 ` Alex Züpke
2015-06-16 11:53 ` Peter Maydell
2015-06-16 12:21 ` Alex Züpke
2015-06-19 15:53 ` Peter Maydell
2015-06-23 7:31 ` Frederic Konrad
2015-06-23 8:09 ` Peter Maydell
2015-06-23 8:33 ` Frederic Konrad
2015-06-23 18:15 ` Peter Maydell
2015-06-25 17:13 ` Peter Maydell
2015-06-15 15:04 ` Peter Maydell
2015-06-15 15:07 ` Alex Züpke
2015-06-15 15:18 ` Peter Maydell
2015-06-15 15:36 ` Alex Züpke
2015-06-15 15:49 ` Peter Maydell
2015-06-15 16:12 ` Alex Züpke
2015-06-15 21:39 ` Peter Crosthwaite
2015-06-19 16:57 ` Paolo Bonzini
2015-06-19 17:25 ` Peter Maydell
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=557B0B21.2030009@hs-rm.de \
--to=alexander.zuepke@hs-rm.de \
--cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).