From: Paul Fulghum <paulkf@microgate.com>
To: scott.feldman@intel.com
Cc: linux-kernel <linux-kernel@vger.kernel.org>
Subject: 2.6.9-rc2-mm4 e100 enable_irq unbalanced from
Date: Mon, 27 Sep 2004 14:24:55 -0500 [thread overview]
Message-ID: <1096313095.2601.20.camel@deimos.microgate.com> (raw)
The e100 module is generating a warning:
Sep 27 13:30:29 deimos kernel: e100: Intel(R) PRO/100 Network Driver, 3.1.4-NAPI
Sep 27 13:30:29 deimos kernel: e100: Copyright(c) 1999-2004 Intel Corporation
Sep 27 13:30:29 deimos kernel: e100: eth0: e100_probe: addr 0xfecfc000, irq 16, MAC addr 00:90:27:3A:C5:E3
Sep 27 13:30:29 deimos kernel: enable_irq(16) unbalanced from ec83ff33
Sep 27 13:30:29 deimos kernel: [<c010923f>] enable_irq+0xcf/0xe0
Sep 27 13:30:29 deimos kernel: [<ec83ff33>] e100_up+0xf3/0x1f0 [e100]
Sep 27 13:30:29 deimos kernel: [<ec83ff33>] e100_up+0xf3/0x1f0 [e100]
Sep 27 13:30:29 deimos kernel: [<ec83f410>] e100_intr+0x0/0x140 [e100]
Sep 27 13:30:29 deimos kernel: [<ec841131>] e100_open+0x31/0x80 [e100]
Sep 27 13:30:29 deimos kernel: [<c0318d4c>] dev_open+0x8c/0xa0
Sep 27 13:30:29 deimos kernel: [<c031cc74>] dev_mc_upload+0x24/0x40
Sep 27 13:30:29 deimos kernel: [<c031a4ea>] dev_change_flags+0x12a/0x150
Sep 27 13:30:29 deimos kernel: [<c0318c0d>] dev_load+0x2d/0x80
Sep 27 13:30:29 deimos kernel: [<c0355b37>] devinet_ioctl+0x277/0x730
e100_up calls disable_irq, request_irq, then enable_irq
as shown below.
static int e100_up(struct nic *nic)
{
...
disable_irq(nic->pdev->irq);
...
if((err = request_irq(nic->pdev->irq, e100_intr, SA_SHIRQ,
nic->netdev->name, nic->netdev)))
goto err_no_irq;
e100_enable_irq(nic);
enable_irq(nic->pdev->irq);
netif_wake_queue(nic->netdev);
return 0;
...
}
On this machine, the e100 is the only device on that IRQ.
request_irq calls setup_irq which clears the irq descriptor
depth member to 0 and enables the interrupt because this
is the first device to use that interrupt.
This results in the warning on the next enable_irq().
I'm not sure why the driver is calling disable_irq
IRQ before calling request_irq. You can't get that
interrupt until you call request_irq, and once you
call request_irq you can (at least when this is
the first device on that IRQ) even before the
call to enable_irq.
I suspect the correct thing is to remove
disable_irq/enable_irq from e100_up.
I don't see any purpose for these calls in e100_up.
--
Paul Fulghum
paulkf@microgate.com
next reply other threads:[~2004-09-27 19:25 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-09-27 19:24 Paul Fulghum [this message]
2004-09-27 21:12 ` 2.6.9-rc2-mm4 e100 enable_irq unbalanced from Paul Fulghum
2004-09-30 18:21 ` Jeremy Fitzhardinge
2004-09-27 23:00 ` J.A. Magallon
2004-09-27 23:09 ` Paul Fulghum
2004-09-28 22:13 ` J.A. Magallon
-- strict thread matches above, loose matches on Subject: below --
2004-09-28 21:03 Feldman, Scott
2004-09-29 3:06 ` Paul Fulghum
2004-09-29 12:13 ` Alan Cox
2004-09-29 13:31 ` Russell King
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=1096313095.2601.20.camel@deimos.microgate.com \
--to=paulkf@microgate.com \
--cc=linux-kernel@vger.kernel.org \
--cc=scott.feldman@intel.com \
/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