[linux-dvb] [Patch] Error in dvb_net.c on 64bit platforms

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Michael Krufky wrote:

> Ralph Metzler wrote:
>
>> Holger Kiehl writes:
>> > On Mon, 31 Oct 2005, Holger Kiehl wrote:
>> > > > Hello
>> > > > > Removing several dvb net interfaces as follows:
>> > > > >       /sbin/ifconfig dvb0_4 down
>> > >       /sbin/ifconfig dvb0_3 down
>> > >       /sbin/ifconfig dvb0_2 down
>> > >       /sbin/ifconfig dvb0_1 down
>> > >       /sbin/ifconfig dvb0_0 down
>> > > > > Always produces an oops on a 64 bit system. On a 32 bit 
>> system this is no
>> > > problem. Applying the attached patch makes it work on 64 bit 
>> systems.
>> > > > After booting the system I still get the oops:
>> > > Oct 31 14:51:43 sojus kernel: dvb_demux_feed_del: feed not in 
>> list (type=1 > state=0 pid=ffff)
>> > Oct 31 14:51:43 sojus kernel: Unable to handle kernel paging 
>> request at > 00000000000010f8 RIP:
>> > Oct 31 14:51:43 sojus kernel: 
>> <ffffffff8028dde0>{dvb_net_filter_sec_set+48}
>> > Oct 31 14:51:43 sojus kernel: PGD cce6a067 PUD 0
>> > Oct 31 14:51:43 sojus kernel: Oops: 0000 [1] SMP
>> > Oct 31 14:51:43 sojus kernel: CPU 1
>>
>>
>> There are some race conditions in the networking code which only show 
>> up on SMP. Please try the file below. It adds a mutex around the 
>> filter handling
>> code. I did not test this myself (lacking an SMP system for testing).
>> I did this blindly based on another bug report but did not hear back
>> from that person if this suffices to remove the race. So, if you or
>> anybody else with SMP can test it that would be great.
>> The could of course be something else showing up onl on 64 bit.
>> Does this ooops also happen on 32 bit SMP?
>>
>> Ralph
>>  
>>
> Ralph-
>
> Did you intend on having this committed?  If so, please generate a 
> patch, containing a sign-off and short description.  We only have 
> until next week if we want to get patches into kernel 2.6.15.
>
Mike,

Fixes can go into mainline at any point of time. Only new features have 
this freeze period. Moreover a completely new stuff also can go in at 
any point of time. The reason being these two category stuff do not 
break anything whereas others can cause broken stuff, while modifying 
existing code and hence the freeze period.

Manu



[Index of Archives]     [Linux Media]     [Video 4 Linux]     [Asterisk]     [Samba]     [Xorg]     [Xfree86]     [Linux USB]

  Powered by Linux