* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-14 16:24 ` Will Deacon
0 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-02-14 16:24 UTC (permalink / raw)
To: Ivaylo Dimitrov
Cc: LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, kvalo, linville
On Fri, Feb 14, 2014 at 04:12:44PM +0000, Ivaylo Dimitrov wrote:
> On 13.02.2014 21:29, Will Deacon wrote:
> >
> > Can you try hacking crypto/memneq.c so that it doesn't use
> > CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS please? That would at least point the
> > finger at net/mac80211/rx.c or similar.
> >
>
> Well, I am lazy so I hacked net/mac80211/rx.c first:
No problem, thanks for having a go.
> index c24ca0d..6839c77 100644
> --- a/net/mac80211/rx.c
> +++ b/net/mac80211/rx.c
> @@ -1963,7 +1963,7 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx)
> }
> }
>
> -#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
> +//#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
> if (skb) {
> /* 'align' will only take the values 0 or 2 here since all
> * frames are required to be aligned to 2-byte boundaries
> @@ -1987,7 +1987,7 @@ ieee80211_deliver_skb(struct ieee80211_rx_data *rx)
> }
> }
> }
> -#endif
> +//#endif
>
> if (skb) {
> /* deliver to local stack */
>
>
> and that seems to fix the problem.
>
> I am not sure whom I should forward the problem.
Well, we probably need a bit more to go on, because I doubt that this code
is to blame. More likely, the issue is in the caller.
Looking at drivers/net/wireless/ti/wl1251/rx.c:182
/* The actual length doesn't include the target's alignment */
skb->len = desc->length - PLCP_HEADER_LENGTH;
fc = (u16 *)skb->data;
if ((*fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON)
beacon = 1;
wl1251_rx_status(wl, desc, &status, beacon);
wl1251_debug(DEBUG_RX, "rx skb 0x%p: %d B %s", skb, skb->len,
beacon ? "beacon" : "");
memcpy(IEEE80211_SKB_RXCB(skb), &status, sizeof(status));
ieee80211_rx_ni(wl->hw, skb);
I wonder whether that first line (with the comment about alignment) is
assuming some behaviour from the mac80211 layer.
You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
above.
Adding the original author (I think) and John Linville, since I'm well out
of my depth in this code!
Will
^ permalink raw reply [flat|nested] 30+ messages in thread* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-14 16:24 ` Will Deacon
@ 2014-02-14 17:11 ` Ivaylo Dimitrov
-1 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-14 17:11 UTC (permalink / raw)
To: linux-arm-kernel
On 14.02.2014 18:24, Will Deacon wrote:
>
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
This seems to fix the issue too, but I am afraid there is more to be
done, as I see
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
with 00:23:cd:17:86:d0
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
00:23:cd:17:86:d0 (try 1/3)
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
with AP with corrupt probe response
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
00:23:cd:17:86:d0 (try 1/3)
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
spit in dmesg log. AFAIR this is the first time I see "associating with
AP with corrupt probe response" in dmesg, no matter of the kernel
version. Despite of that error message, wlan seems to work as it should,
so far. I guess we need the driver's author saying on the issue.
Thanks,
Ivo
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-14 17:11 ` Ivaylo Dimitrov
0 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-14 17:11 UTC (permalink / raw)
To: Will Deacon
Cc: LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, kvalo, linville
On 14.02.2014 18:24, Will Deacon wrote:
>
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
This seems to fix the issue too, but I am afraid there is more to be
done, as I see
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
with 00:23:cd:17:86:d0
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
00:23:cd:17:86:d0 (try 1/3)
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
with AP with corrupt probe response
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
00:23:cd:17:86:d0 (try 1/3)
Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
spit in dmesg log. AFAIR this is the first time I see "associating with
AP with corrupt probe response" in dmesg, no matter of the kernel
version. Despite of that error message, wlan seems to work as it should,
so far. I guess we need the driver's author saying on the issue.
Thanks,
Ivo
^ permalink raw reply [flat|nested] 30+ messages in thread
* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-14 17:11 ` Ivaylo Dimitrov
@ 2014-02-23 19:06 ` Ivaylo Dimitrov
-1 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-23 19:06 UTC (permalink / raw)
To: linux-arm-kernel
On 14.02.2014 19:11, Ivaylo Dimitrov wrote:
>
>
> On 14.02.2014 18:24, Will Deacon wrote:
>>
>> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
>> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
>> above.
>
> This seems to fix the issue too, but I am afraid there is more to be
> done, as I see
>
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> with 00:23:cd:17:86:d0
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> 00:23:cd:17:86:d0 (try 1/3)
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> with AP with corrupt probe response
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> 00:23:cd:17:86:d0 (try 1/3)
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
>
> spit in dmesg log. AFAIR this is the first time I see "associating with
> AP with corrupt probe response" in dmesg, no matter of the kernel
> version. Despite of that error message, wlan seems to work as it should,
> so far. I guess we need the driver's author saying on the issue.
>
> Thanks,
> Ivo
ping?
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-23 19:06 ` Ivaylo Dimitrov
0 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-23 19:06 UTC (permalink / raw)
Cc: Will Deacon, LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, kvalo, linville
On 14.02.2014 19:11, Ivaylo Dimitrov wrote:
>
>
> On 14.02.2014 18:24, Will Deacon wrote:
>>
>> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
>> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
>> above.
>
> This seems to fix the issue too, but I am afraid there is more to be
> done, as I see
>
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> with 00:23:cd:17:86:d0
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> 00:23:cd:17:86:d0 (try 1/3)
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> with AP with corrupt probe response
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> 00:23:cd:17:86:d0 (try 1/3)
> Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
>
> spit in dmesg log. AFAIR this is the first time I see "associating with
> AP with corrupt probe response" in dmesg, no matter of the kernel
> version. Despite of that error message, wlan seems to work as it should,
> so far. I guess we need the driver's author saying on the issue.
>
> Thanks,
> Ivo
ping?
^ permalink raw reply [flat|nested] 30+ messages in thread
* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-23 19:06 ` Ivaylo Dimitrov
@ 2014-02-25 10:44 ` Pavel Machek
-1 siblings, 0 replies; 30+ messages in thread
From: Pavel Machek @ 2014-02-25 10:44 UTC (permalink / raw)
To: linux-arm-kernel
Hi!
> >>You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> >>commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> >>above.
> >
> >This seems to fix the issue too, but I am afraid there is more to be
> >done, as I see
> >
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> >with 00:23:cd:17:86:d0
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> >00:23:cd:17:86:d0 (try 1/3)
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> >with AP with corrupt probe response
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> >00:23:cd:17:86:d0 (try 1/3)
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> >from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
> >
> >spit in dmesg log. AFAIR this is the first time I see "associating with
> >AP with corrupt probe response" in dmesg, no matter of the kernel
> >version. Despite of that error message, wlan seems to work as it should,
> >so far. I guess we need the driver's author saying on the issue.
>
> ping?
If Will does not respond, we should simply revert
dce5c9e35bc4085bd33eccdb8c9ec5a643507a14 . Regressions are not
acceptable.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-25 10:44 ` Pavel Machek
0 siblings, 0 replies; 30+ messages in thread
From: Pavel Machek @ 2014-02-25 10:44 UTC (permalink / raw)
To: Ivaylo Dimitrov
Cc: Will Deacon, LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Sebastian Reichel,
Pali Rohár, kvalo, linville
Hi!
> >>You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> >>commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> >>above.
> >
> >This seems to fix the issue too, but I am afraid there is more to be
> >done, as I see
> >
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> >with 00:23:cd:17:86:d0
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> >00:23:cd:17:86:d0 (try 1/3)
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> >with AP with corrupt probe response
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> >00:23:cd:17:86:d0 (try 1/3)
> >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> >from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
> >
> >spit in dmesg log. AFAIR this is the first time I see "associating with
> >AP with corrupt probe response" in dmesg, no matter of the kernel
> >version. Despite of that error message, wlan seems to work as it should,
> >so far. I guess we need the driver's author saying on the issue.
>
> ping?
If Will does not respond, we should simply revert
dce5c9e35bc4085bd33eccdb8c9ec5a643507a14 . Regressions are not
acceptable.
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
^ permalink raw reply [flat|nested] 30+ messages in thread
* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-25 10:44 ` Pavel Machek
@ 2014-02-25 10:55 ` Will Deacon
-1 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-02-25 10:55 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, Feb 25, 2014 at 10:44:44AM +0000, Pavel Machek wrote:
> > >>You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> > >>commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> > >>above.
> > >
> > >This seems to fix the issue too, but I am afraid there is more to be
> > >done, as I see
> > >
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> > >with 00:23:cd:17:86:d0
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> > >00:23:cd:17:86:d0 (try 1/3)
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> > >with AP with corrupt probe response
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> > >00:23:cd:17:86:d0 (try 1/3)
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> > >from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
> > >
> > >spit in dmesg log. AFAIR this is the first time I see "associating with
> > >AP with corrupt probe response" in dmesg, no matter of the kernel
> > >version. Despite of that error message, wlan seems to work as it should,
> > >so far. I guess we need the driver's author saying on the issue.
> >
> > ping?
>
> If Will does not respond, we should simply revert
> dce5c9e35bc4085bd33eccdb8c9ec5a643507a14 . Regressions are not
> acceptable.
Seriously Pavel, give me a break. I already responded pretty quickly to this
bug report, took a look at where the issue could be and proposed a fix
(which Ivaylo tested successfully). Now, wireless drivers are *waay* outside
my area of expertise, so I added some other folks to CC in the hope that
they could help turn my fix into something upstreamable. Meanwhile, Ivaylo
looks like he has something better than what I came up with already.
If you care about this driver, why don't you help us fix it rather than
making unhelpful suggestions from the sidelines?
Will
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-25 10:55 ` Will Deacon
0 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-02-25 10:55 UTC (permalink / raw)
To: Pavel Machek
Cc: Ivaylo Dimitrov, LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Sebastian Reichel,
Pali Rohár, kvalo@qca.qualcomm.com, linville@tuxdriver.com
On Tue, Feb 25, 2014 at 10:44:44AM +0000, Pavel Machek wrote:
> > >>You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> > >>commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> > >>above.
> > >
> > >This seems to fix the issue too, but I am afraid there is more to be
> > >done, as I see
> > >
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.599853] wlan0: authenticate
> > >with 00:23:cd:17:86:d0
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.781860] wlan0: send auth to
> > >00:23:cd:17:86:d0 (try 1/3)
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.799804] wlan0: authenticated
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.806243] wlan0: associating
> > >with AP with corrupt probe response
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.814544] wlan0: associate with
> > >00:23:cd:17:86:d0 (try 1/3)
> > >Feb 14 18:33:46 Nokia-N900 kernel: [ 88.826416] wlan0: RX AssocResp
> > >from 00:23:cd:17:86:d0 (capab=0x431 status=0 aid=2)
> > >
> > >spit in dmesg log. AFAIR this is the first time I see "associating with
> > >AP with corrupt probe response" in dmesg, no matter of the kernel
> > >version. Despite of that error message, wlan seems to work as it should,
> > >so far. I guess we need the driver's author saying on the issue.
> >
> > ping?
>
> If Will does not respond, we should simply revert
> dce5c9e35bc4085bd33eccdb8c9ec5a643507a14 . Regressions are not
> acceptable.
Seriously Pavel, give me a break. I already responded pretty quickly to this
bug report, took a look at where the issue could be and proposed a fix
(which Ivaylo tested successfully). Now, wireless drivers are *waay* outside
my area of expertise, so I added some other folks to CC in the hope that
they could help turn my fix into something upstreamable. Meanwhile, Ivaylo
looks like he has something better than what I came up with already.
If you care about this driver, why don't you help us fix it rather than
making unhelpful suggestions from the sidelines?
Will
^ permalink raw reply [flat|nested] 30+ messages in thread
* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-14 16:24 ` Will Deacon
@ 2014-02-24 16:52 ` Kalle Valo
-1 siblings, 0 replies; 30+ messages in thread
From: Kalle Valo @ 2014-02-24 16:52 UTC (permalink / raw)
To: linux-arm-kernel
Will Deacon <will.deacon@arm.com> writes:
> Well, we probably need a bit more to go on, because I doubt that this code
> is to blame. More likely, the issue is in the caller.
> Looking at drivers/net/wireless/ti/wl1251/rx.c:182
>
> /* The actual length doesn't include the target's alignment */
> skb->len = desc->length - PLCP_HEADER_LENGTH;
>
> fc = (u16 *)skb->data;
>
> if ((*fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON)
> beacon = 1;
>
> wl1251_rx_status(wl, desc, &status, beacon);
>
> wl1251_debug(DEBUG_RX, "rx skb 0x%p: %d B %s", skb, skb->len,
> beacon ? "beacon" : "");
>
> memcpy(IEEE80211_SKB_RXCB(skb), &status, sizeof(status));
> ieee80211_rx_ni(wl->hw, skb);
>
> I wonder whether that first line (with the comment about alignment) is
> assuming some behaviour from the mac80211 layer.
>
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
>
> Adding the original author (I think) and John Linville, since I'm well out
> of my depth in this code!
I haven't touched wl1251 for years. You should send questions to
linux-wireless list and CC people who have worked on it recently.
--
Kalle Valo
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-24 16:52 ` Kalle Valo
0 siblings, 0 replies; 30+ messages in thread
From: Kalle Valo @ 2014-02-24 16:52 UTC (permalink / raw)
To: Will Deacon
Cc: Ivaylo Dimitrov, LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, linville
Will Deacon <will.deacon@arm.com> writes:
> Well, we probably need a bit more to go on, because I doubt that this code
> is to blame. More likely, the issue is in the caller.
> Looking at drivers/net/wireless/ti/wl1251/rx.c:182
>
> /* The actual length doesn't include the target's alignment */
> skb->len = desc->length - PLCP_HEADER_LENGTH;
>
> fc = (u16 *)skb->data;
>
> if ((*fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON)
> beacon = 1;
>
> wl1251_rx_status(wl, desc, &status, beacon);
>
> wl1251_debug(DEBUG_RX, "rx skb 0x%p: %d B %s", skb, skb->len,
> beacon ? "beacon" : "");
>
> memcpy(IEEE80211_SKB_RXCB(skb), &status, sizeof(status));
> ieee80211_rx_ni(wl->hw, skb);
>
> I wonder whether that first line (with the comment about alignment) is
> assuming some behaviour from the mac80211 layer.
>
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
>
> Adding the original author (I think) and John Linville, since I'm well out
> of my depth in this code!
I haven't touched wl1251 for years. You should send questions to
linux-wireless list and CC people who have worked on it recently.
--
Kalle Valo
^ permalink raw reply [flat|nested] 30+ messages in thread
* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-14 16:24 ` Will Deacon
@ 2014-02-25 7:56 ` Ivaylo Dimitrov
-1 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-25 7:56 UTC (permalink / raw)
To: linux-arm-kernel
Hi,
On 14.02.2014 18:24, Will Deacon wrote:
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
>
the following patch
diff --git a/drivers/net/wireless/ti/wl1251/rx.c
b/drivers/net/wireless/ti/wl1251/rx.c
index 123c4bb..cde0eaf 100644
--- a/drivers/net/wireless/ti/wl1251/rx.c
+++ b/drivers/net/wireless/ti/wl1251/rx.c
@@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
/* The actual length doesn't include the target's alignment */
- skb->len = desc->length - PLCP_HEADER_LENGTH;
+ skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
fc = (u16 *)skb->data;
seems to fix the issue, including those "corrupt probe response"
messages in dmesg log (I took that 'skb_trim' from the original Nokia
kernel). Will send a properly formatted patch shortly.
Thanks,
Ivo
^ permalink raw reply related [flat|nested] 30+ messages in thread* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-25 7:56 ` Ivaylo Dimitrov
0 siblings, 0 replies; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-02-25 7:56 UTC (permalink / raw)
To: Will Deacon
Cc: LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, kvalo, linville
Hi,
On 14.02.2014 18:24, Will Deacon wrote:
> You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> above.
>
the following patch
diff --git a/drivers/net/wireless/ti/wl1251/rx.c
b/drivers/net/wireless/ti/wl1251/rx.c
index 123c4bb..cde0eaf 100644
--- a/drivers/net/wireless/ti/wl1251/rx.c
+++ b/drivers/net/wireless/ti/wl1251/rx.c
@@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
/* The actual length doesn't include the target's alignment */
- skb->len = desc->length - PLCP_HEADER_LENGTH;
+ skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
fc = (u16 *)skb->data;
seems to fix the issue, including those "corrupt probe response"
messages in dmesg log (I took that 'skb_trim' from the original Nokia
kernel). Will send a properly formatted patch shortly.
Thanks,
Ivo
^ permalink raw reply related [flat|nested] 30+ messages in thread* [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
2014-02-25 7:56 ` Ivaylo Dimitrov
@ 2014-02-25 10:57 ` Will Deacon
-1 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-02-25 10:57 UTC (permalink / raw)
To: linux-arm-kernel
Hi Ivaylo,
On Tue, Feb 25, 2014 at 07:56:11AM +0000, Ivaylo Dimitrov wrote:
> On 14.02.2014 18:24, Will Deacon wrote:
> > You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> > commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> > above.
> >
>
> the following patch
>
> diff --git a/drivers/net/wireless/ti/wl1251/rx.c
> b/drivers/net/wireless/ti/wl1251/rx.c
> index 123c4bb..cde0eaf 100644
> --- a/drivers/net/wireless/ti/wl1251/rx.c
> +++ b/drivers/net/wireless/ti/wl1251/rx.c
> @@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
> wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
>
> /* The actual length doesn't include the target's alignment */
> - skb->len = desc->length - PLCP_HEADER_LENGTH;
> + skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
>
> fc = (u16 *)skb->data;
>
> seems to fix the issue, including those "corrupt probe response"
> messages in dmesg log (I took that 'skb_trim' from the original Nokia
> kernel). Will send a properly formatted patch shortly.
Good idea -- don't forget to CC linux-wireless, as suggested by Kalle.
Will
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [BISECTED] ssh - Received disconnect from x.x.x.x: 2: Bad packet length 3149594624
@ 2014-02-25 10:57 ` Will Deacon
0 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-02-25 10:57 UTC (permalink / raw)
To: Ivaylo Dimitrov
Cc: LKML, linux@arm.linux.org.uk,
linux-arm-kernel@lists.infradead.org, Pavel Machek,
Sebastian Reichel, Pali Rohár, kvalo@qca.qualcomm.com,
linville@tuxdriver.com
Hi Ivaylo,
On Tue, Feb 25, 2014 at 07:56:11AM +0000, Ivaylo Dimitrov wrote:
> On 14.02.2014 18:24, Will Deacon wrote:
> > You could try putting back the UNALIGNED_ACCESS in net/mac80211/rx.c and
> > commenting out the skb->len = desc->length - PLCP_HEADER_LENGTH; line
> > above.
> >
>
> the following patch
>
> diff --git a/drivers/net/wireless/ti/wl1251/rx.c
> b/drivers/net/wireless/ti/wl1251/rx.c
> index 123c4bb..cde0eaf 100644
> --- a/drivers/net/wireless/ti/wl1251/rx.c
> +++ b/drivers/net/wireless/ti/wl1251/rx.c
> @@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
> wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
>
> /* The actual length doesn't include the target's alignment */
> - skb->len = desc->length - PLCP_HEADER_LENGTH;
> + skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
>
> fc = (u16 *)skb->data;
>
> seems to fix the issue, including those "corrupt probe response"
> messages in dmesg log (I took that 'skb_trim' from the original Nokia
> kernel). Will send a properly formatted patch shortly.
Good idea -- don't forget to CC linux-wireless, as suggested by Kalle.
Will
^ permalink raw reply [flat|nested] 30+ messages in thread
* [PATCH] wl1251: use skb_trim to make skb shorter
2014-02-25 10:57 ` Will Deacon
(?)
@ 2014-03-01 12:52 ` Ivaylo Dimitrov
2014-03-04 14:26 ` Will Deacon
-1 siblings, 1 reply; 30+ messages in thread
From: Ivaylo Dimitrov @ 2014-03-01 12:52 UTC (permalink / raw)
To: linville
Cc: will.deacon, pali.rohar, pavel, linux-wireless, linux-kernel,
Ivaylo Dimitrov
the current code is directly setting skb->len, which is not correct and
brings problems with HAVE_EFFICIENT_UNALIGNED_ACCESS enabled in config
Signed-off-by: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com>
---
drivers/net/wireless/ti/wl1251/rx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ti/wl1251/rx.c b/drivers/net/wireless/ti/wl1251/rx.c
index 123c4bb..cde0eaf 100644
--- a/drivers/net/wireless/ti/wl1251/rx.c
+++ b/drivers/net/wireless/ti/wl1251/rx.c
@@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
/* The actual length doesn't include the target's alignment */
- skb->len = desc->length - PLCP_HEADER_LENGTH;
+ skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
fc = (u16 *)skb->data;
--
1.7.9.5
^ permalink raw reply related [flat|nested] 30+ messages in thread
* Re: [PATCH] wl1251: use skb_trim to make skb shorter
2014-03-01 12:52 ` [PATCH] wl1251: use skb_trim to make skb shorter Ivaylo Dimitrov
@ 2014-03-04 14:26 ` Will Deacon
2014-03-04 18:33 ` John W. Linville
0 siblings, 1 reply; 30+ messages in thread
From: Will Deacon @ 2014-03-04 14:26 UTC (permalink / raw)
To: Ivaylo Dimitrov
Cc: linville@tuxdriver.com, pali.rohar@gmail.com, pavel@ucw.cz,
linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org
Hi Jonn,
On Sat, Mar 01, 2014 at 12:52:06PM +0000, Ivaylo Dimitrov wrote:
> the current code is directly setting skb->len, which is not correct and
> brings problems with HAVE_EFFICIENT_UNALIGNED_ACCESS enabled in config
>
> Signed-off-by: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com>
> ---
Can you pick this patch up please? It addresses a regression introduced in
the merge window (ARM started selecting HAVE_EFFICIENT_UNALIGNED_ACCESS).
Cheers,
Will
> drivers/net/wireless/ti/wl1251/rx.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/wireless/ti/wl1251/rx.c b/drivers/net/wireless/ti/wl1251/rx.c
> index 123c4bb..cde0eaf 100644
> --- a/drivers/net/wireless/ti/wl1251/rx.c
> +++ b/drivers/net/wireless/ti/wl1251/rx.c
> @@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
> wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
>
> /* The actual length doesn't include the target's alignment */
> - skb->len = desc->length - PLCP_HEADER_LENGTH;
> + skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
>
> fc = (u16 *)skb->data;
>
> --
> 1.7.9.5
>
>
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [PATCH] wl1251: use skb_trim to make skb shorter
2014-03-04 14:26 ` Will Deacon
@ 2014-03-04 18:33 ` John W. Linville
2014-03-05 10:06 ` Will Deacon
0 siblings, 1 reply; 30+ messages in thread
From: John W. Linville @ 2014-03-04 18:33 UTC (permalink / raw)
To: Will Deacon
Cc: Ivaylo Dimitrov, pali.rohar@gmail.com, pavel@ucw.cz,
linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org
OK, I'll queue it as a fix...
On Tue, Mar 04, 2014 at 02:26:58PM +0000, Will Deacon wrote:
> Hi Jonn,
>
> On Sat, Mar 01, 2014 at 12:52:06PM +0000, Ivaylo Dimitrov wrote:
> > the current code is directly setting skb->len, which is not correct and
> > brings problems with HAVE_EFFICIENT_UNALIGNED_ACCESS enabled in config
> >
> > Signed-off-by: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com>
> > ---
>
> Can you pick this patch up please? It addresses a regression introduced in
> the merge window (ARM started selecting HAVE_EFFICIENT_UNALIGNED_ACCESS).
>
> Cheers,
>
> Will
>
> > drivers/net/wireless/ti/wl1251/rx.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/wireless/ti/wl1251/rx.c b/drivers/net/wireless/ti/wl1251/rx.c
> > index 123c4bb..cde0eaf 100644
> > --- a/drivers/net/wireless/ti/wl1251/rx.c
> > +++ b/drivers/net/wireless/ti/wl1251/rx.c
> > @@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
> > wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
> >
> > /* The actual length doesn't include the target's alignment */
> > - skb->len = desc->length - PLCP_HEADER_LENGTH;
> > + skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
> >
> > fc = (u16 *)skb->data;
> >
> > --
> > 1.7.9.5
> >
> >
>
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [PATCH] wl1251: use skb_trim to make skb shorter
2014-03-04 18:33 ` John W. Linville
@ 2014-03-05 10:06 ` Will Deacon
0 siblings, 0 replies; 30+ messages in thread
From: Will Deacon @ 2014-03-05 10:06 UTC (permalink / raw)
To: John W. Linville
Cc: Ivaylo Dimitrov, pali.rohar@gmail.com, pavel@ucw.cz,
linux-wireless@vger.kernel.org, linux-kernel@vger.kernel.org
On Tue, Mar 04, 2014 at 06:33:48PM +0000, John W. Linville wrote:
> OK, I'll queue it as a fix...
Thanks, John.
Will
> On Tue, Mar 04, 2014 at 02:26:58PM +0000, Will Deacon wrote:
> > Hi Jonn,
> >
> > On Sat, Mar 01, 2014 at 12:52:06PM +0000, Ivaylo Dimitrov wrote:
> > > the current code is directly setting skb->len, which is not correct and
> > > brings problems with HAVE_EFFICIENT_UNALIGNED_ACCESS enabled in config
> > >
> > > Signed-off-by: Ivaylo Dimitrov <ivo.g.dimitrov.75@gmail.com>
> > > ---
> >
> > Can you pick this patch up please? It addresses a regression introduced in
> > the merge window (ARM started selecting HAVE_EFFICIENT_UNALIGNED_ACCESS).
> >
> > Cheers,
> >
> > Will
> >
> > > drivers/net/wireless/ti/wl1251/rx.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/net/wireless/ti/wl1251/rx.c b/drivers/net/wireless/ti/wl1251/rx.c
> > > index 123c4bb..cde0eaf 100644
> > > --- a/drivers/net/wireless/ti/wl1251/rx.c
> > > +++ b/drivers/net/wireless/ti/wl1251/rx.c
> > > @@ -180,7 +180,7 @@ static void wl1251_rx_body(struct wl1251 *wl,
> > > wl1251_mem_read(wl, rx_packet_ring_addr, rx_buffer, length);
> > >
> > > /* The actual length doesn't include the target's alignment */
> > > - skb->len = desc->length - PLCP_HEADER_LENGTH;
> > > + skb_trim(skb, desc->length - PLCP_HEADER_LENGTH);
> > >
> > > fc = (u16 *)skb->data;
> > >
> > > --
> > > 1.7.9.5
> > >
> > >
> >
>
> --
> John W. Linville Someday the world will need a hero, and you
> linville@tuxdriver.com might be all we have. Be ready.
>
^ permalink raw reply [flat|nested] 30+ messages in thread