* [BUG/PATCH] iphase ATM driver (Oops - Div0)
@ 2001-11-13 21:43 Till Immanuel Patzschke
0 siblings, 0 replies; only message in thread
From: Till Immanuel Patzschke @ 2001-11-13 21:43 UTC (permalink / raw)
To: pwang, acme; +Cc: linux-kernel@vger.kernel.org
[-- Attachment #1: Type: text/plain, Size: 870 bytes --]
Hi,
please find attached a fix for a nasty bug in the iphase ATM NIC. If te max_sdu
exceeds the limit the internal structure ia_vcc is freed BUT the pointer in the
vcc struct is NOT been reset. This results in a Div0 error in ia_close
(ctimeout = 300000 / ia_vcc->pcr) - clearing the referencing pointer fixes the
problem.
I've also diabled the "Misaligned SKB" check since it works fine w/o it
AND re-aligning is to expensive if it works (I know, jumping to conclusions, but
it works fine on my boxes)...
Could you please - at least - apply the pointer fix?
Thanks,
Immanuel
--
Till Immanuel Patzschke mailto: tip@internetwork-ag.de
interNetwork AG Phone: +49-(0)611-1731-121
Bierstadter Str. 7 Fax: +49-(0)611-1731-31
D-65189 Wiesbaden Web: http://www.internetwork-ag.de
[-- Attachment #2: qq2 --]
[-- Type: text/plain, Size: 1370 bytes --]
diff -Naur linux-2.4.10/drivers/atm/iphase.c linux-2.4.10-new/drivers/atm/iphase.c
--- linux-2.4.10/drivers/atm/iphase.c Tue Sep 18 07:52:34 2001
+++ linux-2.4.10-new/drivers/atm/iphase.c Tue Nov 13 18:40:49 2001
@@ -1777,8 +1777,9 @@
memset((caddr_t)ia_vcc, 0, sizeof(*ia_vcc));
if (vcc->qos.txtp.max_sdu >
(iadev->tx_buf_sz - sizeof(struct cpcs_trailer))){
- printk("IA: SDU size over the configured SDU size %d\n",
- iadev->tx_buf_sz);
+ printk("IA: SDU size over (%d) the configured SDU size %d\n",
+ vcc->qos.txtp.max_sdu,iadev->tx_buf_sz);
+ INPH_IA_VCC(vcc) = NULL;
kfree(ia_vcc);
return -EINVAL;
}
@@ -2896,14 +2897,16 @@
dev_kfree_skb_any(skb);
return 0;
}
+#if 0
if ((u32)skb->data & 3) {
- printk("Misaligned SKB\n");
+ printk("Misaligned SKB (0x%p)\n",skb->data);
if (vcc->pop)
vcc->pop(vcc, skb);
else
dev_kfree_skb_any(skb);
return 0;
}
+#endif
/* Get a descriptor number from our free descriptor queue
We get the descr number from the TCQ now, since I am using
the TCQ as a free buffer queue. Initially TCQ will be
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2001-11-13 21:34 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-11-13 21:43 [BUG/PATCH] iphase ATM driver (Oops - Div0) Till Immanuel Patzschke
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox