9.2 ixgbe tx queue hang

Markus Gebert markus.gebert at hostpoint.ch
Tue Mar 25 23:07:23 UTC 2014 

On 25.03.2014, at 23:21, Rick Macklem <rmacklem at uoguelph.ca> wrote:

> Markus Gebert wrote:
>> 
>> On 25.03.2014, at 22:46, Rick Macklem <rmacklem at uoguelph.ca> wrote:
>> 
>>> Markus Gebert wrote:
>>>> 
>>>> On 25.03.2014, at 02:18, Rick Macklem <rmacklem at uoguelph.ca>
>>>> wrote:
>>>> 
>>>>> Christopher Forgeron wrote:
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> This is regarding the TSO patch that Rick suggested earlier.
>>>>>> (With
>>>>>> many thanks for his time and suggestion)
>>>>>> 
>>>>>> 
>>>>>> As I mentioned earlier, it did not fix the issue on a 10.0
>>>>>> system.
>>>>>> It
>>>>>> did make it less of a problem on 9.2, but either way, I think
>>>>>> it's
>>>>>> not needed, and shouldn't be considered as a patch for
>>>>>> testing/etc.
>>>>>> 
>>>>>> 
>>>>>> Patching TSO to anything other than a max value (and by default
>>>>>> the
>>>>>> code gives it IP_MAXPACKET) is confusing the matter, as the
>>>>>> packet
>>>>>> length ultimately needs to be adjusted for many things on the
>>>>>> fly
>>>>>> like TCP Options, etc. Using static header sizes won't be a good
>>>>>> idea.
>>>>>> 
>>>>> If you look at tcp_output(), you'll notice that it doesn't do TSO
>>>>> if
>>>>> there are any options. That way it knows that the TCP/IP header
>>>>> is
>>>>> just hdrlen.
>>>>> 
>>>>> If you don't limit the TSO packet (including TCP/IP and ethernet
>>>>> headers)
>>>>> to 64K, then the "ix" driver can't send them, which is the
>>>>> problem
>>>>> you guys are seeing.
>>>>> 
>>>>> There are other ways to fix this problem, but they all may
>>>>> introduce
>>>>> issues that reducing if_hw_tsomax by a small amount does not.
>>>>> For example, m_defrag() could be modified to use 4K pagesize
>>>>> clusters,
>>>>> but this might introduce memory fragmentation problems. (I
>>>>> observed
>>>>> what I think are memory fragmentation problems when I switched
>>>>> NFS
>>>>> to use 4K pagesize clusters for large I/O messages.)
>>>>> 
>>>>> If setting IP_MAXPACKET to 65518 fixes the problem (no more EFBIG
>>>>> error replies), then that is the size that if_hw_tsomax can be
>>>>> set
>>>>> to (just can't change IP_MAXPACKET, but that is defined for other
>>>>> things). (It just happens that IP_MAXPACKET is what if_hw_tsomax
>>>>> defaults to. It has no other effect w.r.t. TSO.)
>>>>> 
>>>>>> 
>>>>>> Additionally, it seems that setting nic TSO will/may be ignored
>>>>>> by
>>>>>> code like this in sys/netinet/tcp_output.c:
>>>>>> 
>>>> 
>>>> Is this confirmed or still a ‘it seems’? Have you actually seen a
>>>> tp->t_tsomax value in tcp_output() bigger than if_hw_tsomax or was
>>>> this just speculation because the values are stored in different
>>>> places? (Sorry, if you already stated this in another email, it’s
>>>> currently hard to keep track of all the information.)
>>>> 
>>>> Anyway, this dtrace one-liner should be a good test if other
>>>> values
>>>> appear in tp->t_tsomax:
>>>> 
>>>> # dtrace -n 'fbt::tcp_output:entry / args[0]->t_tsomax != 0 &&
>>>> args[0]->t_tsomax != 65518 / { printf("unexpected tp->t_tsomax:
>>>> %i\n", args[0]->t_tsomax); stack(); }'
>>>> 
>>>> Remember to adjust the value in the condition to whatever you’re
>>>> currently expecting. The value seems to be 0 for new connections,
>>>> probably when tcp_mss() has not been called yet. So that’s seems
>>>> normal and I have excluded that case too. This will also print a
>>>> kernel stack trace in case it sees an unexpected value.
>>>> 
>>>> 
>>>>> Yes, but I don't know why.
>>>>> The only conjecture I can come up with is that another net driver
>>>>> is
>>>>> stacked above "ix" and the setting for if_hw_tsomax doesn't
>>>>> propagate
>>>>> up. (If you look at the commit log message for r251296, the
>>>>> intent
>>>>> of adding if_hw_tsomax was to allow device drivers to set a
>>>>> smaller
>>>>> tsomax than IP_MAXPACKET.)
>>>>> 
>>>>> Are you using any of the "stacked" network device drivers like
>>>>> lagg? I don't even know what the others all are?
>>>>> Maybe someone else can list them?
>>>> 
>>>> I guess the most obvious are lagg and vlan (and probably carp on
>>>> FreeBSD 9.x or older).
>>>> 
>>>> On request from Jack, we’ve eliminated lagg and vlan from the
>>>> picture, which gives us plain ixgbe interfaces with no stacked
>>>> interfaces on top of it. And we can still reproduce the problem.
>>>> 
>>> This was related to the "did if_hw_tsomax set tp->t_tsomax to the
>>> same value?" question. Since you reported that my patch that set
>>> if_hw_tsomax in the driver didn't fix the problem, that suggests
>>> that tp->t_tsomax isn't being set to if_hw_tsomax from the driver,
>>> but we don't know why?
>> 
>> Jack asked us to remove lagg/vlans in the very beginning of this
>> thread, and when had done that, the problem was still there. So my
>> answer was not related to your recent patch. I wanted to clarify
>> that we have been testing with ixgbe only for quite some time and
>> that stacked interfaces could not be a source of problems in our
>> test scenario.
>> 
>> We have just started testing your patch that sets if_hw_tsomax
>> yesterday. So far I have it running on two systems along with some
>> printfs and the dtrace one-liner that watches over tp->t_tsomax in
>> tcp_output(). So far we’ve haven’t had any problems with these two
>> servers, and the dtrace probe never fired, so far it looks like
>> tp->t_tsomax always gets set from if_hw_tsomax. But it’s too soon to
>> make a conclusion, it may take days to trigger the problem again. It
>> might also be fixed with your patch.
>> 
> Righto. Setting if_hw_tsomax in the driver is supposed to set tp->t_tsomax
> and I could see it work in a trivial test (I hacked the code so the assignments
> are done for the non-tso case and it worked for the non-tso "re" driver I run.)
> { As an aside, one of these assignments does happen for non-tso cases, since
>  although it is indented, there are no {} for the block. In tcp_subr.c if I
>  recall. However, doing the assignment for the non-tso case seems harmless to me. }
> 
>> I’m booting more systems with the test kernel and I will be watching
>> all of them with dtrace to see I i find an occurence where
>> tp->t_tsomax is off. I hope that with more systems, I’ll have an
>> answer more quickly.
>> 
>> But digging around the code, I still don’t see a way how tp->tsomax
>> could not have been set from if_hw_tsomax when there are no stacked
>> interfaces…
>> 
> It seems to happen where you mentioned before. Since it only gets set
> from cap.tsomax and that gets set from if_hw_tsomax, it would be 0
> otherwise.

Sorry, my sentence was probably a bit misleading. What you’re saying is what I meant. There’s the tcp_mss() -> tcp_mss_update() -> tcp_maxmtu() call chain that ultimately sets tp->t_tsomax from if_hw_tsomax (via the cap struct). tp->t_tsomax is indeed 0 on fresh connections when tcp_mss() has not been called yet, I could confirm that with dtrace. As soon as the connection gets running, it’s set to whatever the interface’s if_hw_tsomax is.

What I have _not_ found is another place that alters tp->t_tsomax, so I really don’t get how Christopher can see different values for tp->t_tsomax.


> Christopher sees in change when he changes IP_MAXPACET, so
> the default setting works, but for him setting it in the driver didn't,
> for some reason?

Christopher, can you run your tests again with default IP_MAXPACKET and just Rick's if_hw_maxtso patch? I think it’s important to confirm that tp->t_tsomax is really off in that case, which you can easily test with my dtrace one-liner. I’m running this test too, but I will take me much more time until I can make a statement.

> Thanks for doing the testing, rick

No problem. Thank you guys!


>> 
>> Markus
>> 
>> 
>>> rick
>>> 
>>>> 
>>>> Markus
>>>> 
>>>> 
>>>>> 
>>>>> rick
>>>>>> 
>>>>>> 10.0 Code:
>>>>>> 
>>>>>> 780 if (len > tp->t_tsomax - hdrlen) { !!
>>>>>> 781 len = tp->t_tsomax - hdrlen; !!
>>>>>> 782 sendalot = 1;
>>>>>> 783 }
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> I've put debugging here, set the nic's max TSO as per Rick's
>>>>>> patch
>>>>>> (
>>>>>> set to say 32k), and have seen that tp->t_tsomax ==
>>>>>> IP_MAXPACKET.
>>>>>> It's being set someplace else, and thus our attempts to set TSO
>>>>>> on
>>>>>> the nic may be in vain.
>>>>>> 
>>>>>> 
>>>>>> It may have mattered more in 9.2, as I see the code doesn't use
>>>>>> tp->t_tsomax in some locations, and may actually default to what
>>>>>> the
>>>>>> nic is set to.
>>>>>> 
>>>>>> The NIC may still win, I didn't walk through the code to
>>>>>> confirm,
>>>>>> it
>>>>>> was enough to suggest to me that setting TSO wouldn't fix this
>>>>>> issue.
>>>>>> 
>>>>>> 
>>>>>> However, this is still a TSO related issue, it's just not one
>>>>>> related
>>>>>> to the setting of TSO's max size.
>>>>>> 
>>>>>> A 10.0-STABLE system with tso disabled on ix0 doesn't have a
>>>>>> single
>>>>>> packet over IP_MAXPACKET in 1 hour of runtime. I'll let it go a
>>>>>> bit
>>>>>> longer to increase confidence in this assertion, but I don't
>>>>>> want
>>>>>> to
>>>>>> waste time on this when I could be logging problem packets on a
>>>>>> system with TSO enabled.
>>>>>> 
>>>>>> 
>>>>>> Comments are very welcome..
>>>>>> 
>>>>>> 
>>>>>> 
>>>>> _______________________________________________
>>>>> freebsd-net at freebsd.org mailing list
>>>>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>>>>> To unsubscribe, send any mail to
>>>>> "freebsd-net-unsubscribe at freebsd.org"
>>>>> 
>>>> 
>>>> 
>>> _______________________________________________
>>> freebsd-net at freebsd.org mailing list
>>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>>> To unsubscribe, send any mail to
>>> "freebsd-net-unsubscribe at freebsd.org"
>> 
>> _______________________________________________
>> freebsd-net at freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> To unsubscribe, send any mail to
>> "freebsd-net-unsubscribe at freebsd.org"
>> 
> _______________________________________________
> freebsd-net at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe at freebsd.org"

More information about the freebsd-net mailing list