From: Grosjean Stephane <s.grosjean@peak-system.com>
To: Wolfgang Grandegger <wg@grandegger.com>
Cc: linux-can@vger.kernel.org
Subject: peak_pci: device channels chained list issue
Date: Fri, 23 Dec 2011 12:32:18 +0100 [thread overview]
Message-ID: <4EF466C2.9070204@peak-system.com> (raw)
In-Reply-To: <4EF2FA3F.3010308@grandegger.com>
Hi Wolfgang,
I think the way the device channels are chained in peak_pci_probe() is
buggy in the linux-can-next driver, and will lead to some Kernel Oops if
the sja1000 module is removed. I fixed it by handling that list in the
same manner than I handle channels in the PCAN-USB Pro adapter. Side
effect is that the channels net devices are removed from the last to the
first one...
Please have a look to the following patch.
Regards,
Stéphane
diff --git a/drivers/net/can/sja1000/peak_pci.c
b/drivers/net/can/sja1000/peak_pci.c
index 2c7f503..7b5a4ab 100644
--- a/drivers/net/can/sja1000/peak_pci.c
+++ b/drivers/net/can/sja1000/peak_pci.c
@@ -40,7 +40,7 @@ MODULE_LICENSE("GPL v2");
struct peak_pci_chan {
void __iomem *cfg_base; /* Common for all channels */
- struct net_device *next_dev; /* Chain of network devices */
+ struct net_device *prev_dev; /* Chain of network devices */
u16 icr_mask; /* Interrupt mask for fast ack */
};
@@ -98,7 +98,7 @@ static int __devinit peak_pci_probe(struct pci_dev *pdev,
{
struct sja1000_priv *priv;
struct peak_pci_chan *chan;
- struct net_device *dev, *dev0 = NULL;
+ struct net_device *dev, *prev_dev;
void __iomem *cfg_base, *reg_base;
u16 sub_sys_id, icr;
int i, err, channels;
@@ -196,18 +196,16 @@ static int __devinit peak_pci_probe(struct pci_dev
*pdev,
}
/* Create chain of SJA1000 devices */
- if (i == 0)
- dev0 = dev;
- else
- chan->next_dev = dev;
+ prev_dev = pci_get_drvdata(pdev);
+ if (prev_dev)
+ chan->prev_dev = prev_dev;
+ pci_set_drvdata(pdev, dev);
dev_info(&pdev->dev,
"%s at reg_base=0x%p cfg_base=0x%p irq=%d\n",
dev->name, priv->reg_base, chan->cfg_base, dev->irq);
}
- pci_set_drvdata(pdev, dev0);
-
/* Enable interrupts */
writew(icr, cfg_base + PITA_ICR + 2);
@@ -217,12 +215,12 @@ failure_remove_channels:
/* Disable interrupts */
writew(0x0, cfg_base + PITA_ICR + 2);
- for (dev = dev0; dev; dev = chan->next_dev) {
+ for (dev = pci_get_drvdata(pdev); dev; dev = chan->prev_dev) {
unregister_sja1000dev(dev);
free_sja1000dev(dev);
priv = netdev_priv(dev);
chan = priv->priv;
- dev = chan->next_dev;
+ dev = chan->prev_dev;
}
pci_iounmap(pdev, reg_base);
@@ -241,7 +239,7 @@ failure_disable_pci:
static void __devexit peak_pci_remove(struct pci_dev *pdev)
{
- struct net_device *dev = pci_get_drvdata(pdev); /* First device */
+ struct net_device *dev = pci_get_drvdata(pdev); /* Last device */
struct sja1000_priv *priv = netdev_priv(dev);
struct peak_pci_chan *chan = priv->priv;
void __iomem *cfg_base = chan->cfg_base;
@@ -255,7 +253,7 @@ static void __devexit peak_pci_remove(struct pci_dev
*pdev)
dev_info(&pdev->dev, "removing device %s\n", dev->name);
unregister_sja1000dev(dev);
free_sja1000dev(dev);
- dev = chan->next_dev;
+ dev = chan->prev_dev;
if (!dev)
break;
priv = netdev_priv(dev);
next prev parent reply other threads:[~2011-12-23 11:32 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-25 9:43 [PATCH net-next v2 0/4] can: cc770: add support for the Bosch CC770 and Intel AN82527 Wolfgang Grandegger
2011-11-25 9:43 ` [PATCH net-next v2 1/4] can: cc770: add driver core " Wolfgang Grandegger
[not found] ` <1322214204-1121-2-git-send-email-wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-11-26 15:11 ` Oliver Hartkopp
2011-11-28 11:28 ` [Socketcan-users] " Marc Kleine-Budde
2011-11-28 13:52 ` Wolfgang Grandegger
[not found] ` <4ED3922A.50704-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-11-28 14:01 ` Marc Kleine-Budde
2011-11-28 14:01 ` [Socketcan-users] " David Laight
2011-11-28 14:01 ` David Laight
[not found] ` <4ED3941D.3070302-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
2011-11-28 14:10 ` Wolfgang Grandegger
[not found] ` <4ED3966E.7080609-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-11-28 14:18 ` Marc Kleine-Budde
[not found] ` <4ED3704D.5020903-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
2011-11-29 9:20 ` Wolfgang Grandegger
2011-11-25 9:43 ` [PATCH net-next v2 2/4] can: cc770: add legacy ISA bus driver for the CC770 and AN82527 Wolfgang Grandegger
[not found] ` <1322214204-1121-3-git-send-email-wg-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-11-26 14:59 ` Oliver Hartkopp
2011-11-28 8:56 ` Wolfgang Zarre
2011-11-28 9:17 ` Wolfgang Grandegger
2011-11-28 12:03 ` Wolfgang Zarre
[not found] ` <4ED37885.8080909-PyqsHJVlJN8AvxtiuMwx3w@public.gmane.org>
2011-11-28 16:06 ` Oliver Hartkopp
[not found] ` <4ED3B198.2040308-fJ+pQTUTwRTk1uMJSBkQmQ@public.gmane.org>
2011-11-29 9:16 ` Wolfgang Grandegger
[not found] ` <4ED4A2EC.40103-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-12-04 18:47 ` Wolfgang Zarre
2011-12-04 18:56 ` Wolfgang Grandegger
[not found] ` <4EDBC25D.50405-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-12-06 21:08 ` Wolfgang Zarre
[not found] ` <4EDE8435.5080100-PyqsHJVlJN8AvxtiuMwx3w@public.gmane.org>
2011-12-07 13:42 ` Wolfgang Grandegger
2011-12-09 10:26 ` Wolfgang Grandegger
2011-12-11 18:33 ` Wolfgang Zarre
[not found] ` <4EE4F76E.3000506-PyqsHJVlJN8AvxtiuMwx3w@public.gmane.org>
2011-12-12 9:23 ` Wolfgang Grandegger
2011-12-12 9:23 ` Wolfgang Grandegger
[not found] ` <4EE5C824.2050704-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-12-12 11:18 ` Wolfgang Zarre
2011-12-12 11:55 ` Wolfgang Grandegger
[not found] ` <4EE5EBBF.6080007-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-12-21 18:32 ` Wolfgang Zarre
2011-12-22 9:37 ` Wolfgang Grandegger
[not found] ` <4EF2FA3F.3010308-5Yr1BZd7O62+XT7JhA+gdA@public.gmane.org>
2011-12-22 13:20 ` Wolfgang Zarre
[not found] ` <4EF32E84.1080006-PyqsHJVlJN8AvxtiuMwx3w@public.gmane.org>
2011-12-31 9:39 ` Wolfgang Zarre
2012-01-04 13:10 ` Wolfgang Grandegger
2012-01-05 3:29 ` Wolfgang Zarre
[not found] ` <4F051927.8010600-PyqsHJVlJN8AvxtiuMwx3w@public.gmane.org>
2012-01-05 11:51 ` Wolfgang Grandegger
2012-01-05 12:00 ` David Laight
2012-01-05 12:00 ` David Laight
2012-01-09 21:47 ` Wolfgang Zarre
2012-01-09 23:11 ` Marc Kleine-Budde
2012-01-10 9:30 ` Wolfgang Grandegger
2012-01-10 12:30 ` Wolfgang Zarre
2012-01-10 14:20 ` Wolfgang Grandegger
2012-01-10 14:25 ` Wolfgang Grandegger
2012-01-11 9:00 ` Wolfgang Zarre
2012-01-11 9:37 ` David Laight
2012-01-11 9:37 ` David Laight
2012-01-11 14:37 ` Wolfgang Zarre
2012-01-11 9:38 ` Marc Kleine-Budde
2012-01-11 14:42 ` Wolfgang Zarre
2012-01-11 15:02 ` Marc Kleine-Budde
2012-01-10 10:00 ` David Laight
2012-01-10 10:00 ` David Laight
2012-01-10 12:41 ` Wolfgang Zarre
2012-01-10 14:43 ` Wolfgang Grandegger
2012-01-10 14:50 ` Oliver Hartkopp
2012-01-10 16:13 ` Wolfgang Zarre
2012-01-10 16:20 ` Marc Kleine-Budde
2012-01-10 16:23 ` Wolfgang Grandegger
2012-01-10 19:02 ` Wolfgang Zarre
2012-01-11 9:05 ` Wolfgang Zarre
2012-01-11 9:31 ` Marc Kleine-Budde
2012-01-10 11:41 ` Henrik Maier
2012-01-10 11:59 ` Wolfgang Grandegger
2012-01-10 12:43 ` Wolfgang Zarre
2012-01-05 19:45 ` Oliver Hartkopp
2012-01-05 20:48 ` Oliver Hartkopp
2011-12-23 11:32 ` Grosjean Stephane [this message]
2011-12-23 12:11 ` peak_pci: device channels chained list issue (cont.) Grosjean Stephane
2012-01-31 22:12 ` Marc Kleine-Budde
2011-11-28 12:09 ` [PATCH net-next v2 2/4] can: cc770: add legacy ISA bus driver for the CC770 and AN82527 Marc Kleine-Budde
[not found] ` <4ED379F3.1070206-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
2011-11-28 13:59 ` Wolfgang Grandegger
2011-11-28 14:03 ` David Laight
2011-11-28 14:03 ` David Laight
[not found] ` <AE90C24D6B3A694183C094C60CF0A2F6D8AEE9-CgBM+Bx2aUAnGFn1LkZF6NBPR1lH4CV8@public.gmane.org>
2011-11-28 14:09 ` Marc Kleine-Budde
[not found] ` <4ED3960F.4040508-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
2011-11-28 15:10 ` Wolfgang Grandegger
2011-11-25 9:43 ` [PATCH net-next v2 3/4] can: cc770: add platform " Wolfgang Grandegger
2011-11-25 9:43 ` Wolfgang Grandegger
2011-11-25 9:43 ` [PATCH net-next v2 4/4] powerpc: tqm8548/tqm8xx: add and update CAN device nodes Wolfgang Grandegger
2011-11-25 9:43 ` Wolfgang Grandegger
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=4EF466C2.9070204@peak-system.com \
--to=s.grosjean@peak-system.com \
--cc=linux-can@vger.kernel.org \
--cc=wg@grandegger.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.