* [9fans] Realtek 8110S/8169S driver.
@ 2004-12-16 21:29 jmk
2004-12-16 22:43 ` andrey mirtchovski
0 siblings, 1 reply; 6+ messages in thread
From: jmk @ 2004-12-16 21:29 UTC (permalink / raw)
To: 9fans
/sys/src/9/pc/ether8169.c on sources is a driver for the Realtek 8110S/8169S
Gigabit chips, we have a motherboard with one. I also did the mangling to make
it work in 9load, but it is inexplicably slow to load a kernel. If anyone
has one of these chips and would like to try to figure out the problem let me
know.
--jim
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Realtek 8110S/8169S driver.
2004-12-16 21:29 [9fans] Realtek 8110S/8169S driver jmk
@ 2004-12-16 22:43 ` andrey mirtchovski
2004-12-16 22:50 ` jmk
0 siblings, 1 reply; 6+ messages in thread
From: andrey mirtchovski @ 2004-12-16 22:43 UTC (permalink / raw)
To: 9fans
> /sys/src/9/pc/ether8169.c on sources is a driver for the Realtek 8110S/8169S
> Gigabit chips, we have a motherboard with one. I also did the mangling to make
> it work in 9load, but it is inexplicably slow to load a kernel. If anyone
> has one of these chips and would like to try to figure out the problem let me
> know.
>
> --jim
in the linuxbios case slowness to load the kernel from 9load has
always been caused by incorrect interrupt setup...
andrey
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Realtek 8110S/8169S driver.
2004-12-16 22:43 ` andrey mirtchovski
@ 2004-12-16 22:50 ` jmk
2004-12-16 22:59 ` Ronald G. Minnich
0 siblings, 1 reply; 6+ messages in thread
From: jmk @ 2004-12-16 22:50 UTC (permalink / raw)
To: 9fans
On Thu Dec 16 17:43:54 EST 2004, mirtchov@cpsc.ucalgary.ca wrote:
> > /sys/src/9/pc/ether8169.c on sources is a driver for the Realtek 8110S/8169S
> > Gigabit chips, we have a motherboard with one. I also did the mangling to make
> > it work in 9load, but it is inexplicably slow to load a kernel. If anyone
> > has one of these chips and would like to try to figure out the problem let me
> > know.
> >
> > --jim
>
> in the linuxbios case slowness to load the kernel from 9load has
> always been caused by incorrect interrupt setup...
>
> andrey
The card is getting real interrupts, and it looks to me like they are
correctly sequenced, i.e. when loading a kernel i see a loop of
send a request for a block from the tftp server
transmission complete interrupt
receive interrupt
plonk the received packet in the appropriate place
--jim
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Realtek 8110S/8169S driver.
2004-12-16 22:50 ` jmk
@ 2004-12-16 22:59 ` Ronald G. Minnich
2004-12-16 23:05 ` jmk
0 siblings, 1 reply; 6+ messages in thread
From: Ronald G. Minnich @ 2004-12-16 22:59 UTC (permalink / raw)
To: Fans of the OS Plan 9 from Bell Labs
any change it is half/full duplex nonsense? This is also a common source
of badness in performance.
ron
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Realtek 8110S/8169S driver.
2004-12-16 22:59 ` Ronald G. Minnich
@ 2004-12-16 23:05 ` jmk
2004-12-21 15:17 ` jmk
0 siblings, 1 reply; 6+ messages in thread
From: jmk @ 2004-12-16 23:05 UTC (permalink / raw)
To: 9fans
On Thu Dec 16 18:00:57 EST 2004, rminnich@lanl.gov wrote:
> any change it is half/full duplex nonsense? This is also a common source
> of badness in performance.
>
> ron
there are no transmission/reception errors reported. the phy says
it has negotiated correctly, e.g. it reports the correct speed and
duplex on both the 100 and 1000 links i've tried. the nic+phy require
no real management, they are very tightly bound and there's nothing
the driver has to do to the nic which depends on what the phy negotiates
as far as i can see.
--jim
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Realtek 8110S/8169S driver.
2004-12-16 23:05 ` jmk
@ 2004-12-21 15:17 ` jmk
0 siblings, 0 replies; 6+ messages in thread
From: jmk @ 2004-12-21 15:17 UTC (permalink / raw)
To: 9fans
I found the problem with the 9load version of ether8169.c
(stupid transcription error while manglig for the 9load driver nonsense).
Thanks to Sam <sah@softcardsystems.com> for a more up to date datasheet
for the part; although it didn't help with my bug it cleared up some other
inconsistencies I'd seen in the *n*x drivers.
There are updates on sources to /sys/src/boot/pc and /sys/src/9/pc/ether8169.c.
Comments, experiences, improvements welcome.
--jim
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2004-12-21 15:17 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-12-16 21:29 [9fans] Realtek 8110S/8169S driver jmk
2004-12-16 22:43 ` andrey mirtchovski
2004-12-16 22:50 ` jmk
2004-12-16 22:59 ` Ronald G. Minnich
2004-12-16 23:05 ` jmk
2004-12-21 15:17 ` jmk
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).