Date: Sun, 3 Jan 2010 15:54:31 +0100 From: =?iso-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de> To: =?iso-8859-1?Q?Michael_T=FCxen?= <Michael.Tuexen@lurchi.franken.de> Cc: Barney Cordoba <barney_cordoba@yahoo.com>, jfvogel@gmail.com, freebsd-net@freebsd.org Subject: Re: igb interrupt moderation Message-ID: <133FB4C8-AF96-4CBC-8291-2325592AB628@lurchi.franken.de> In-Reply-To: <B91C36D0-4D52-485D-A1A7-1477705E758E@lurchi.franken.de> References: <969646.48748.qm@web63905.mail.re1.yahoo.com> <B91C36D0-4D52-485D-A1A7-1477705E758E@lurchi.franken.de>
next in thread | previous in thread | raw e-mail | index | archive | help
Dear all,
I just figured out that there is a newer version of the spec
available: 2.45. Some of the issues as indicated inline
are already resolved.=20
Best regards
Michael
On Jan 3, 2010, at 2:55 PM, Michael T=FCxen wrote:
> Hi Barney, Hi Jack,
>=20
> some comments and some more questions inside...
>=20
> Best regards
> Michael
>=20
> On Jan 2, 2010, at 8:42 PM, Barney Cordoba wrote:
>=20
>> Jack,
>>=20
>> I'm trying to get some clarification on differences I'm finding =
between
>> the 82575 and 82576 parts with respect to interrupt moderation. The =
spec
>> I have for the 82576 (82576_Datasheet_v2p1.pdf) indicates that the=20
> I'm only commenting 82576. You can get rev 2.41 from intels website...
It is really 2.45 now
>>=20
>> ITR algorithm is different than the one used (I don't have one of the
>> secret copies of the 82575 spec). The algorithm shown is
>>=20
>> interrupts/sec =3D 1/(2 * 10-6sec x interval) (page 295, Section =
7.3.4)
>>=20
>> which is clearly wrong from practice. I have an 82576 (device id =
10C9)
> If you look at section 8.8.12, you find other formulas...
> Jack: Which ones are correct?
The formulas is 8.8.12 are gone. Issue resolved.
>> if I use the 125d setting in the example get just under 32000 =
interrupts
>> per second. Clearly your code doesnt implement this, nor do you have
>> different settings for the 82575 and 82576 parts. So I assume that =
the=20
>> same formula for the em parts hold for the igb parts, and that the=20
>> datasheet is wrong?
>>=20
>> There does seem to be a slight difference. The setting that gets 1000
>> ints/second on the 82575 generates about 1020 on the 82576. Not a big
>> deal but I wonder why there's a difference? Is the reference clock =
for
>> these something that may not be fixed and could vary from board to=20
>> board? Note that both devices are on the same MB.
>>=20
>> Also, it seems that settings to EITR over 32767 wrap on the 82576 =
(for
>> example writing 32768 to EITR is the same as writing a 1). So the =
minimum setting on the 82576 is around 125 ints/second. The 82575 can =
accept=20
>> values up the 65535 before wrapping.=20
> Hmm, looking at the table in 8.8.12 would suggest:
> Setting it to one sets a reserved bit, but does not change the =
interval.
> Setting it to 2^15 should set the LLI_EN bit, but does not change in =
interval.
>=20
> Jack is setting the register to
> igb_low_latency: 128
> igb_ave_latency: 450
> igb_bulk_latency: 1200
>=20
> This would result in intervals of:
> igb_low_latency: 32
> igb_ave_latency: 112
> igb_bulk_latency: 300
> Jack: What are the corresponding interrupt rates? The spec provides =
different
> formulas and talks about a 1us, 2us or 8us counter. Not sure what =
is right...
The interrupt rates are according to the formula in the spec
igb_low_latency: 31250
igb_ave_latency: 8929
igb_bulk_latency: 3333
Jack: Is this right?
> Jack: Why are you setting bit1 (which is reserved) in the case =
igb_ave_latency?
Still valid.
>=20
> And another question for Jack:
> In igb_update_aim() you do
> if (olditr !=3D newitr) {
> /* Change interrupt rate */
> rxr->eitr_setting =3D newitr;
> E1000_WRITE_REG(&adapter->hw, E1000_EITR(rxr->me),
> newitr | (newitr << 16));
> }
> So why are setting the higher bits of the EITR? You are setting
> igb_low_latency: the LL Counter becomes 0, the moderation counter =
becomes 16
> igb_ave_latency: the LL Counter becomes 2, the moderation counter =
becomes 56
> igb_bulk_latency: the LL Counter becomes 16, the moderation counter =
becomes 148
Still valid.
>=20
> I really do not understand these settings. Maybe the spec is wrong? Or =
you do mean
> if (olditr !=3D newitr) {
> /* Change interrupt rate */
> rxr->eitr_setting =3D newitr;
> E1000_WRITE_REG(&adapter->hw, E1000_EITR(rxr->me), =
newitr);
> }
> Or do you want to preserve the counters, set the CNT_INGR bit and mean
> if (olditr !=3D newitr) {
> /* Change interrupt rate */
> rxr->eitr_setting =3D newitr;
> E1000_WRITE_REG(&adapter->hw, E1000_EITR(rxr->me), =
0x80000000 | newitr);
> }
>=20
> Could you clarify that?
Still valid.
>>=20
>> The 82576 document doesn't have a map of the register that I can =
find, so
>> Im curious as to whether these observations are something I can =
assume is
>> true across all parts and motherboards/cards, or is there some
>> implementation variance that will cause these to only apply to the =
ones
>> I happen to be testing?
>>=20
>> Thanks,
>>=20
>> Barney
>>=20
>>=20
>>=20
>>=20
>> _______________________________________________
>> freebsd-net@freebsd.org mailing list
>> http://lists.freebsd.org/mailman/listinfo/freebsd-net
>> To unsubscribe, send any mail to =
"freebsd-net-unsubscribe@freebsd.org"
>>=20
>=20
> _______________________________________________
> freebsd-net@freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-net
> To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"
>=20
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?133FB4C8-AF96-4CBC-8291-2325592AB628>