if_mvneta rx/tx byte counts wrong

Søren Schmidt soren.schmidt at gmail.com
Wed Nov 4 12:43:51 UTC 2020


Hi John

Good catch!, if I just use those as 32bit values it “just works” seems the Marvell version has a bug or config problem here..

> On 4 Nov 2020, at 13.26, John F Carr <jfc at mit.edu> wrote:
> 
> If you view in hex,
> 
> $ printf '%x %x %x\n' 443193380410133 3813930959736 274877907008
> 1931500019315 37800000378 4000000040
> 
> you see a 32 bit value duplicated into both halves of a 64 bit value.
> 
> On Nov 4, 2020, at 03:38 , Søren Schmidt <soren.schmidt at gmail.com> wrote:
>> 
>> Hi gang
>> 
>> While returning to my hacks on the Espressobin I noticed that netstat -w gives some interesting results:
>> 
>> root at EspressoBin:~ # netstat -w1
>>           input        (Total)           output
>>  packets  errs idrops      bytes    packets  errs      bytes colls
>>        0     0     0         60          0     0          0     0
>>        1     0     0 274877907008          0     0          0     0
>>        0     0     0         60          0     0          0     0
>>        1     0     0 274877907008          0     0          0     0
>>        1     0     0 274877907315          0     0          0     0
>>        3     0     1 1370094567743          0     0          0     0
>>        0     0     0         60          0     0          0     0
>>        1     0     0 274877909253          0     0          0     0
>> 
>> Looking at the sysctl’s its the same thing:
>> 
>> dev.mvneta.0.mib.tx_good_oct: 3813930959736
>> dev.mvneta.0.mib.rx_good_oct: 443193380410133
>> 
>> First though there might be byteordering involved but looking at the raw values that does not explain the behaviour, however the rx/tx count is the only values that are 64bit, all others are 32bit and are working as expected.
>> 
>> Any ideas / fixes ?
>> 
>> --
>> Søren Schmidt
>> sos at deepcore.dk / sos at freebsd.org
>> "So much code to hack, so little time"
>> 
>> 
>> 
>> _______________________________________________
>> freebsd-arm at freebsd.org mailing list
>> https://lists.freebsd.org/mailman/listinfo/freebsd-arm
>> To unsubscribe, send any mail to "freebsd-arm-unsubscribe at freebsd.org"
> 

--
Søren Schmidt
sos at deepcore.dk / sos at freebsd.org
"So much code to hack, so little time"





More information about the freebsd-arm mailing list