All of lore.kernel.org
 help / color / mirror / Atom feed
* [BUG] IBM440GX works too slow under nfsroot
@ 2004-10-21 23:57 Hiroshi DOYU
  2004-10-22  1:19 ` Eugene Surovegin
  2004-10-22  1:33 ` Eugene Surovegin
  0 siblings, 2 replies; 19+ messages in thread
From: Hiroshi DOYU @ 2004-10-21 23:57 UTC (permalink / raw)
  To: linuxppc-embedded

Hi all,

Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
I could succeed to boot it but, after nfs mounted, it's behavior 
seems too slow. I guess, it looks like the system is running 
half duplex. I have attached  the boot message below.

If anyone knows some information about such issue, please let me know.

Thanks,
-doyu

----------------
.....
Generic RTC Driver v1.07                                                                               
emac: IBM OCP EMAC Ethernet driver, version 2.0                                                        
Maintained by Benjamin Herrenschmidt <benh@kernel.crashing.org>                                        
mal0: Initialized, 4 tx channels, 2 rx channels                                                        
zmii0: bridge in SMII mode                                                                             
eth0: IBM emac, MAC 00:04:ac:e3:2a:f5                                                                  
eth0: Found Generic MII PHY (0x01)                                                                     
eth1: IBM emac, MAC 00:04:ac:e3:2a:f7                                                                  
eth1: Found Generic MII PHY (0x02)                                                                     
NET4: Linux TCP/IP 1.0 for NET4.0                                                                      
IP Protocols: ICMP, UDP, TCP, IGMP                                                                     
IP: routing cache hash table of 2048 buckets, 16Kbytes                                                 
TCP: Hash tables configured (established 16384 bind 32768)                                             
Sending BOOTP requests .<6>eth0: Link is Up                                                            
eth0: Speed: 10, Half duplex.                                                                          
. OK                                                                                                   
IP-Config: Got BOOTP answer from 192.168.1.10, my address is 192.168.1.2                               
IP-Config: Complete:                                                                                   
      device=eth0, addr=192.168.1.2, mask=255.255.255.0, gw=192.168.1.10,                              
     host=ibm440gx, domain=, nis-domain=(none),                                                        
     bootserver=192.168.1.10, rootserver=192.168.1.10, rootpath=/opt/montavista/pro/devkit/ppc/440/tart
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.                                                    
Looking up port of RPC 100003/2 on 192.168.1.10                                                        
Looking up port of RPC 100005/1 on 192.168.1.10                                                        
VFS: Mounted root (nfs filesystem).                                                                    
Mounted devfs on /dev                                                                                  
Freeing unused kernel memory: 76k init                                                                 
nfs: server 192.168.1.10 not responding, still trying                                                  
nfs: server 192.168.1.10 not responding, still trying                                                  
nfs: server 192.168.1.10 OK                                                                            
nfs: server 192.168.1.10 OK                                                                            
nfs: server 192.168.1.10 not responding, still trying
nfs: server 192.168.1.10 OK                                         
.....
----------------

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-21 23:57 [BUG] IBM440GX works too slow under nfsroot Hiroshi DOYU
@ 2004-10-22  1:19 ` Eugene Surovegin
  2004-10-22  1:26   ` Hiroshi DOYU
  2004-10-22  1:33 ` Eugene Surovegin
  1 sibling, 1 reply; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-22  1:19 UTC (permalink / raw)
  To: Hiroshi DOYU; +Cc: linuxppc-embedded

On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> Hi all,
> 
> Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> I could succeed to boot it but, after nfs mounted, it's behavior 
> seems too slow. I guess, it looks like the system is running 
> half duplex. I have attached  the boot message below.
> 
> If anyone knows some information about such issue, please let me know.

Is this 440GX ref platform (Ocotea) or your custom hardware?
If custom, please, try 440GX ref board and report the results.

Also, what _exact_ version are you using? 2.4.2x isn't very helpful.

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  1:19 ` Eugene Surovegin
@ 2004-10-22  1:26   ` Hiroshi DOYU
  0 siblings, 0 replies; 19+ messages in thread
From: Hiroshi DOYU @ 2004-10-22  1:26 UTC (permalink / raw)
  To: Eugene Surovegin; +Cc: linuxppc-embedded

Hi Eugene,

On Thu, 21 Oct 2004 18:19:07 -0700
Eugene Surovegin <ebs@ebshome.net> wrote:

> On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > Hi all,
> > 
> > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > I could succeed to boot it but, after nfs mounted, it's behavior 
> > seems too slow. I guess, it looks like the system is running 
> > half duplex. I have attached  the boot message below.
> > 
> > If anyone knows some information about such issue, please let me know.
> 
> Is this 440GX ref platform (Ocotea) or your custom hardware?
> If custom, please, try 440GX ref board and report the results.
> 
> Also, what _exact_ version are you using? 2.4.2x isn't very helpful.

Additonal information:

I am using Ocotea(REV3) and tried v2.4.27.

Thanks,
-doyu

> 
> --
> Eugene
> 
> 

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-21 23:57 [BUG] IBM440GX works too slow under nfsroot Hiroshi DOYU
  2004-10-22  1:19 ` Eugene Surovegin
@ 2004-10-22  1:33 ` Eugene Surovegin
  2004-10-22  1:38   ` Hiroshi DOYU
  1 sibling, 1 reply; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-22  1:33 UTC (permalink / raw)
  To: Hiroshi DOYU; +Cc: linuxppc-embedded

On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> I could succeed to boot it but, after nfs mounted, it's behavior 
> seems too slow. I guess, it looks like the system is running 
> half duplex. 

Yeah, driver thinks it's connected to 10Mb hub. How the board is 
actualy connected to the network?

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  1:33 ` Eugene Surovegin
@ 2004-10-22  1:38   ` Hiroshi DOYU
  2004-10-22  2:02     ` Eugene Surovegin
  2004-10-22 22:22     ` Matt Porter
  0 siblings, 2 replies; 19+ messages in thread
From: Hiroshi DOYU @ 2004-10-22  1:38 UTC (permalink / raw)
  To: Eugene Surovegin; +Cc: linuxppc-embedded

On Thu, 21 Oct 2004 18:33:20 -0700
Eugene Surovegin <ebs@ebshome.net> wrote:

> On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > I could succeed to boot it but, after nfs mounted, it's behavior 
> > seems too slow. I guess, it looks like the system is running 
> > half duplex. 
> 
> Yeah, driver thinks it's connected to 10Mb hub. How the board is 
> actualy connected to the network?

10Mb hub.

> 
> --
> Eugene
> 
> 

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  1:38   ` Hiroshi DOYU
@ 2004-10-22  2:02     ` Eugene Surovegin
  2004-10-22  6:26       ` Gerhard Jaeger
  2004-10-22 22:22     ` Matt Porter
  1 sibling, 1 reply; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-22  2:02 UTC (permalink / raw)
  To: Hiroshi DOYU; +Cc: linuxppc-embedded

On Fri, Oct 22, 2004 at 10:38:59AM +0900, Hiroshi DOYU wrote:
> On Thu, 21 Oct 2004 18:33:20 -0700
> Eugene Surovegin <ebs@ebshome.net> wrote:
> 
> > On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > > I could succeed to boot it but, after nfs mounted, it's behavior 
> > > seems too slow. I guess, it looks like the system is running 
> > > half duplex. 
> > 
> > Yeah, driver thinks it's connected to 10Mb hub. How the board is 
> > actualy connected to the network?
> 
> 10Mb hub.
> 

Wow, I really had to spend some time to find such ancient hw :). I'll 
try 2.4.27 with this hub during next couple of days (I'm a little busy 
right now) and let you know. I suspect nobody really tested EMAC driver 
in 10Mb HDX mode.

You can also bug Matt Porter, maybe he'll help you sooner :).

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  2:02     ` Eugene Surovegin
@ 2004-10-22  6:26       ` Gerhard Jaeger
  2004-10-22  8:29         ` Eugene Surovegin
  0 siblings, 1 reply; 19+ messages in thread
From: Gerhard Jaeger @ 2004-10-22  6:26 UTC (permalink / raw)
  To: linuxppc-embedded

On Friday 22 October 2004 04:02, Eugene Surovegin wrote:
> On Fri, Oct 22, 2004 at 10:38:59AM +0900, Hiroshi DOYU wrote:
> > On Thu, 21 Oct 2004 18:33:20 -0700
> >
> > Eugene Surovegin <ebs@ebshome.net> wrote:
> > > On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > > > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > > > I could succeed to boot it but, after nfs mounted, it's behavior
> > > > seems too slow. I guess, it looks like the system is running
> > > > half duplex.
> > >
> > > Yeah, driver thinks it's connected to 10Mb hub. How the board is
> > > actualy connected to the network?
> >
> > 10Mb hub.
>
> Wow, I really had to spend some time to find such ancient hw :). I'll
> try 2.4.27 with this hub during next couple of days (I'm a little busy
> right now) and let you know. I suspect nobody really tested EMAC driver
> in 10Mb HDX mode.
>
> You can also bug Matt Porter, maybe he'll help you sooner :).

Hi Eugene,

this problem has been discussed a while ago between my co-worker and
Matt-Porter - first on the list and later on in private. We also have this
problem on a custom board using the 440GX. It can be also seen on a Ebony
(440GP) as well and on our rather old Ocotea. Our kernel originally is based
on 2.4.26, but the EMAC driver has been backported from 2.6.9.
Fiddling around with some settings seems to help a bit, but in the end we see
excessive collisions. The reason is not clear yet. 
Matt could not confirm this behaviour 'til now.

Ciao,
  Gerhard

-- 
Gerhard Jaeger <gjaeger@sysgo.com>            
SYSGO AG                      Embedded and Real-Time Software
www.sysgo.com | www.elinos.com | www.osek.de | www.imerva.com

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  6:26       ` Gerhard Jaeger
@ 2004-10-22  8:29         ` Eugene Surovegin
  2004-10-22  9:39           ` Gerhard Jaeger
                             ` (2 more replies)
  0 siblings, 3 replies; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-22  8:29 UTC (permalink / raw)
  To: Gerhard Jaeger; +Cc: linuxppc-embedded

On Fri, Oct 22, 2004 at 08:26:40AM +0200, Gerhard Jaeger wrote:
> 
> this problem has been discussed a while ago between my co-worker and
> Matt-Porter - first on the list and later on in private. We also have this
> problem on a custom board using the 440GX. It can be also seen on a Ebony
> (440GP) as well and on our rather old Ocotea. Our kernel originally is based
> on 2.4.26, but the EMAC driver has been backported from 2.6.9.
> Fiddling around with some settings seems to help a bit, but in the end we see
> excessive collisions. The reason is not clear yet. 
> Matt could not confirm this behaviour 'til now.

Hmm, did you see problems on 100 HDX as well, or only on 10 HDX?

FWIW, I haven't seen any problems when testing my new EMAC driver with 
100Mb hub, maybe it's only 10Mb related.

Anyway, this is first time I'm hearing about half duplex problems, 
Matt just mentioned them on IRC to me.

I'll try to do some tests this weekend, although I have to admit half 
duplex doesn't seem very important these days, nevertheless, if this 
can be fixed, we better do this :).

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  8:29         ` Eugene Surovegin
@ 2004-10-22  9:39           ` Gerhard Jaeger
  2004-10-22 12:22           ` Mark Chambers
  2004-10-25  1:16           ` Eugene Surovegin
  2 siblings, 0 replies; 19+ messages in thread
From: Gerhard Jaeger @ 2004-10-22  9:39 UTC (permalink / raw)
  To: Eugene Surovegin; +Cc: linuxppc-embedded

On Friday 22 October 2004 10:29, Eugene Surovegin wrote:
> On Fri, Oct 22, 2004 at 08:26:40AM +0200, Gerhard Jaeger wrote:
> > this problem has been discussed a while ago between my co-worker and
> > Matt-Porter - first on the list and later on in private. We also have
> > this problem on a custom board using the 440GX. It can be also seen on a
> > Ebony (440GP) as well and on our rather old Ocotea. Our kernel originally
> > is based on 2.4.26, but the EMAC driver has been backported from 2.6.9.
> > Fiddling around with some settings seems to help a bit, but in the end we
> > see excessive collisions. The reason is not clear yet.
> > Matt could not confirm this behaviour 'til now.
>
> Hmm, did you see problems on 100 HDX as well, or only on 10 HDX?

I need to retest again, AFAIR. It was limited to the 10HDX.

>
> FWIW, I haven't seen any problems when testing my new EMAC driver with
> 100Mb hub, maybe it's only 10Mb related.
>
> Anyway, this is first time I'm hearing about half duplex problems,
> Matt just mentioned them on IRC to me.
>
> I'll try to do some tests this weekend, although I have to admit half
> duplex doesn't seem very important these days, nevertheless, if this
> can be fixed, we better do this :).

Fully ACK ;)
In general you are right, 10HDX should not be that important, but
nevertheless, there are enough installations and setups, where it
is still needed. 
The strangest thing which can be observed is, that even when you have
a point-to-point connection and you force the connection to 10HDX you'll
see a lot of collisions. As I mentioned, with our settings (should be
the same as you'll find in current 2.6.9), it could happen, that there
are almost no collisions in the beginning, but from time to time you'll
get some collision firework, which stops after a few seconds...

Gerhard

-- 
Gerhard Jaeger <gjaeger@sysgo.com>            
SYSGO AG                      Embedded and Real-Time Software
www.sysgo.com | www.elinos.com | www.osek.de | www.imerva.com

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  8:29         ` Eugene Surovegin
  2004-10-22  9:39           ` Gerhard Jaeger
@ 2004-10-22 12:22           ` Mark Chambers
  2004-10-22 12:27             ` Gerhard Jaeger
  2004-10-25  1:16           ` Eugene Surovegin
  2 siblings, 1 reply; 19+ messages in thread
From: Mark Chambers @ 2004-10-22 12:22 UTC (permalink / raw)
  To: linuxppc-embedded

> >
> > this problem has been discussed a while ago between my co-worker and
> > Matt-Porter - first on the list and later on in private. We also have
this
> > problem on a custom board using the 440GX. It can be also seen on a
Ebony
> > (440GP) as well and on our rather old Ocotea. Our kernel originally is
based
etc.

I'm just curious here - I'm not familiar with IBM processors -
is there a separate PHY chip, and if so, what is it?

Mark C.

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22 12:22           ` Mark Chambers
@ 2004-10-22 12:27             ` Gerhard Jaeger
  0 siblings, 0 replies; 19+ messages in thread
From: Gerhard Jaeger @ 2004-10-22 12:27 UTC (permalink / raw)
  To: linuxppc-embedded

On Friday 22 October 2004 14:22, Mark Chambers wrote:
> > > this problem has been discussed a while ago between my co-worker and
> > > Matt-Porter - first on the list and later on in private. We also have this
> > > problem on a custom board using the 440GX. It can be also seen on a Ebony
> > > (440GP) as well and on our rather old Ocotea. Our kernel originally is
> > >based
> etc.
>
> I'm just curious here - I'm not familiar with IBM processors -
> is there a separate PHY chip, and if so, what is it?

Yes, on Ebony and Ocotea we have an AMD NetPHY AM79C875KC, on
Walnut, I'm not sure, on our custom boards we have also the AMD
and a NS DP83846A, as well as DP8385...

Gerhard

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  1:38   ` Hiroshi DOYU
  2004-10-22  2:02     ` Eugene Surovegin
@ 2004-10-22 22:22     ` Matt Porter
  1 sibling, 0 replies; 19+ messages in thread
From: Matt Porter @ 2004-10-22 22:22 UTC (permalink / raw)
  To: Hiroshi DOYU; +Cc: linuxppc-embedded

On Fri, Oct 22, 2004 at 10:38:59AM +0900, Hiroshi DOYU wrote:
> On Thu, 21 Oct 2004 18:33:20 -0700
> Eugene Surovegin <ebs@ebshome.net> wrote:
> 
> > On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > > I could succeed to boot it but, after nfs mounted, it's behavior 
> > > seems too slow. I guess, it looks like the system is running 
> > > half duplex. 
> > 
> > Yeah, driver thinks it's connected to 10Mb hub. How the board is 
> > actualy connected to the network?
> 
> 10Mb hub.

Well, I'm seeing awful half duplex performance on a 10/100Mbps hub
that has negotiated 100Mbps...this is on one hub at home.  At the
office, with a different hub, I seem not to have the problems.  I
will try a 10Mbps hub next.

This is on 440GX EMAC0 w/ current BK 2.6.

-Matt

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  8:29         ` Eugene Surovegin
  2004-10-22  9:39           ` Gerhard Jaeger
  2004-10-22 12:22           ` Mark Chambers
@ 2004-10-25  1:16           ` Eugene Surovegin
  2004-10-25  6:47             ` Gerhard Jaeger
  2 siblings, 1 reply; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-25  1:16 UTC (permalink / raw)
  To: Gerhard Jaeger, linuxppc-embedded, Hiroshi DOYU, Matt Porter

On Fri, Oct 22, 2004 at 01:29:55AM -0700, Eugene Surovegin wrote:
> FWIW, I haven't seen any problems when testing my new EMAC driver with 
> 100Mb hub, maybe it's only 10Mb related.
> 
> Anyway, this is first time I'm hearing about half duplex problems, 
> Matt just mentioned them on IRC to me.
> 
> I'll try to do some tests this weekend, although I have to admit half 
> duplex doesn't seem very important these days, nevertheless, if this 
> can be fixed, we better do this :).

OK, I did some tests with 2.4.28-pre4, 2.6.9 & 2.6.9-emac (2.6.9 with 
my new EMAC driver). I used Ocotea rev.3 with 440GX rev.C (533MHz).

I can confirm I see some problems in 10HDX mode - there is visible 
slow down during system boot up (NFS root). Just after logging into 
system, nfsstat shows around 350 retransmits.

My 2.6.9-emac tree doesn't show this behavior, I see no visible 
slow-down and nfsstat shows 0 retransmits.

I tried some quick hacks for 2.6.9 driver but was unable to find 
settings which could help (my new driver was almost re-written from 
scratch and I did a lot of things differently).

For *adventurous*, this new EMAC driver is available as a patch against 
2.6.9 BK tree at http://kernel.ebshome.net/. Please, note, it's still 
project in progress and it lacks support for all new 440GX features 
(GigE, TAH, jumbo frames, EMAC2/EMAC3). This is NAPI version with 
correct (I hope) error handling, chip access etc. On Ebony I was able 
to route around 144Kpps. I plan to implement support for _some_ 440GX 
new features soon (during next month, hopefully).

If there is an interest, I can make a patch against current 2.4 BK 
tree, but this will take some time (I'm a little swamped right now).

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-25  1:16           ` Eugene Surovegin
@ 2004-10-25  6:47             ` Gerhard Jaeger
  2004-10-25  7:02               ` Eugene Surovegin
  2004-11-08  2:55               ` Eugene Surovegin
  0 siblings, 2 replies; 19+ messages in thread
From: Gerhard Jaeger @ 2004-10-25  6:47 UTC (permalink / raw)
  To: Eugene Surovegin; +Cc: linuxppc-embedded

On Monday 25 October 2004 03:16, Eugene Surovegin wrote:
> On Fri, Oct 22, 2004 at 01:29:55AM -0700, Eugene Surovegin wrote:
> > FWIW, I haven't seen any problems when testing my new EMAC driver with
> > 100Mb hub, maybe it's only 10Mb related.
> >
> > Anyway, this is first time I'm hearing about half duplex problems,
> > Matt just mentioned them on IRC to me.
> >
> > I'll try to do some tests this weekend, although I have to admit half
> > duplex doesn't seem very important these days, nevertheless, if this
> > can be fixed, we better do this :).
>
> OK, I did some tests with 2.4.28-pre4, 2.6.9 & 2.6.9-emac (2.6.9 with
> my new EMAC driver). I used Ocotea rev.3 with 440GX rev.C (533MHz).
>
> I can confirm I see some problems in 10HDX mode - there is visible
> slow down during system boot up (NFS root). Just after logging into
> system, nfsstat shows around 350 retransmits.
>
> My 2.6.9-emac tree doesn't show this behavior, I see no visible
> slow-down and nfsstat shows 0 retransmits.
>
> I tried some quick hacks for 2.6.9 driver but was unable to find
> settings which could help (my new driver was almost re-written from
> scratch and I did a lot of things differently).

Hmmm, that's what we also found out, nothing really helped.

> For *adventurous*, this new EMAC driver is available as a patch against
> 2.6.9 BK tree at http://kernel.ebshome.net/. Please, note, it's still
> project in progress and it lacks support for all new 440GX features
> (GigE, TAH, jumbo frames, EMAC2/EMAC3). This is NAPI version with
> correct (I hope) error handling, chip access etc. On Ebony I was able
> to route around 144Kpps. I plan to implement support for _some_ 440GX
> new features soon (during next month, hopefully).

I will give it a try ASAP.

>
> If there is an interest, I can make a patch against current 2.4 BK
> tree, but this will take some time (I'm a little swamped right now).

of course there's some interest, at least by me ;) Would be great if you'll
find some time for that.

Thank you,
  Gerhard

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-25  6:47             ` Gerhard Jaeger
@ 2004-10-25  7:02               ` Eugene Surovegin
  2004-11-08  2:55               ` Eugene Surovegin
  1 sibling, 0 replies; 19+ messages in thread
From: Eugene Surovegin @ 2004-10-25  7:02 UTC (permalink / raw)
  To: Gerhard Jaeger; +Cc: linuxppc-embedded

On Mon, Oct 25, 2004 at 08:47:00AM +0200, Gerhard Jaeger wrote:
> On Monday 25 October 2004 03:16, Eugene Surovegin wrote:

[snip]

> > For *adventurous*, this new EMAC driver is available as a patch against
> > 2.6.9 BK tree at http://kernel.ebshome.net/. Please, note, it's still
> > project in progress and it lacks support for all new 440GX features
> > (GigE, TAH, jumbo frames, EMAC2/EMAC3). This is NAPI version with
> > correct (I hope) error handling, chip access etc. On Ebony I was able
> > to route around 144Kpps. I plan to implement support for _some_ 440GX
> > new features soon (during next month, hopefully).
> 
> I will give it a try ASAP.
> 

I've just put Ethtool patch which adds EMAC specific reg dump at the 
same URL.

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-25  6:47             ` Gerhard Jaeger
  2004-10-25  7:02               ` Eugene Surovegin
@ 2004-11-08  2:55               ` Eugene Surovegin
  1 sibling, 0 replies; 19+ messages in thread
From: Eugene Surovegin @ 2004-11-08  2:55 UTC (permalink / raw)
  To: Gerhard Jaeger; +Cc: linuxppc-embedded

On Mon, Oct 25, 2004 at 08:47:00AM +0200, Gerhard Jaeger wrote:
> > If there is an interest, I can make a patch against current 2.4 BK
> > tree, but this will take some time (I'm a little swamped right now).
> 
> of course there's some interest, at least by me ;) Would be great if you'll
> find some time for that.

Patch against current 2.4 BK is available at http://kernel.ebshome.net. 
Quick tested on Ebony & Ocotea.

--
Eugene

^ permalink raw reply	[flat|nested] 19+ messages in thread

* RE: [BUG] IBM440GX works too slow under nfsroot
@ 2004-10-22  7:36 Neil Wilson
  0 siblings, 0 replies; 19+ messages in thread
From: Neil Wilson @ 2004-10-22  7:36 UTC (permalink / raw)
  To: Hiroshi DOYU, linuxppc-embedded

Hi,

I too had a similar problem with our custom 440GX board - this week in
fact.  It would DHCP, mount NFS then keep on reporting that the server
was not responding.  I eventually tracked it down to the fact that we
are running a single MII phy and the phy map was not being setup
correctly by my port.  As soon as I changed this in the arch_setup it
has worked ever since. BTW it is linuxppc 2.6.8.1,RevC
440,100BaseTX,busybox root.  This may not be Hiroshis problem but
thought I would add to the conversation...

Neil

> -----Original Message-----
> From: linuxppc-embedded-bounces@ozlabs.org=20
> [mailto:linuxppc-embedded-bounces@ozlabs.org] On Behalf Of=20
> Hiroshi DOYU
> Sent: Friday, October 22, 2004 12:58 AM
> To: linuxppc-embedded@ozlabs.org
> Subject: [BUG] IBM440GX works too slow under nfsroot
>=20
> Hi all,
>=20
> Now I am strugging with booting vanilla(2.4.2x) on ibm440gx=20
> and I could succeed to boot it but, after nfs mounted, it's=20
> behavior seems too slow. I guess, it looks like the system is=20
> running half duplex. I have attached  the boot message below.
>=20
> If anyone knows some information about such issue, please let me know.
>=20
> Thanks,
> -doyu
>=20
> ----------------
> .....
> Generic RTC Driver v1.07                                     =20
>                                         =20
> emac: IBM OCP EMAC Ethernet driver, version 2.0              =20
>                                         =20
> Maintained by Benjamin Herrenschmidt=20
> <benh@kernel.crashing.org>                                       =20
> mal0: Initialized, 4 tx channels, 2 rx channels              =20
>                                         =20
> zmii0: bridge in SMII mode                                   =20
>                                         =20
> eth0: IBM emac, MAC 00:04:ac:e3:2a:f5                        =20
>                                         =20
> eth0: Found Generic MII PHY (0x01)                           =20
>                                         =20
> eth1: IBM emac, MAC 00:04:ac:e3:2a:f7                        =20
>                                         =20
> eth1: Found Generic MII PHY (0x02)                           =20
>                                         =20
> NET4: Linux TCP/IP 1.0 for NET4.0                            =20
>                                         =20
> IP Protocols: ICMP, UDP, TCP, IGMP                           =20
>                                         =20
> IP: routing cache hash table of 2048 buckets, 16Kbytes       =20
>                                         =20
> TCP: Hash tables configured (established 16384 bind 32768)   =20
>                                         =20
> Sending BOOTP requests .<6>eth0: Link is Up                  =20
>                                         =20
> eth0: Speed: 10, Half duplex.                                =20
>                                         =20
> . OK                                                         =20
>                                         =20
> IP-Config: Got BOOTP answer from 192.168.1.10, my address is=20
> 192.168.1.2                              =20
> IP-Config: Complete:                                         =20
>                                         =20
>       device=3Deth0, addr=3D192.168.1.2, mask=3D255.255.255.0,=20
> gw=3D192.168.1.10,                             =20
>      host=3Dibm440gx, domain=3D, nis-domain=3D(none),              =20
>                                         =20
>      bootserver=3D192.168.1.10, rootserver=3D192.168.1.10,=20
> rootpath=3D/opt/montavista/pro/devkit/ppc/440/tart
> NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.          =20
>                                         =20
> Looking up port of RPC 100003/2 on 192.168.1.10              =20
>                                         =20
> Looking up port of RPC 100005/1 on 192.168.1.10              =20
>                                         =20
> VFS: Mounted root (nfs filesystem).                          =20
>                                         =20
> Mounted devfs on /dev                                        =20
>                                         =20
> Freeing unused kernel memory: 76k init                       =20
>                                         =20
> nfs: server 192.168.1.10 not responding, still trying        =20
>                                         =20
> nfs: server 192.168.1.10 not responding, still trying        =20
>                                         =20
> nfs: server 192.168.1.10 OK                                  =20
>                                         =20
> nfs: server 192.168.1.10 OK                                  =20
>                                         =20
> nfs: server 192.168.1.10 not responding, still trying
> nfs: server 192.168.1.10 OK                                        =20
> .....
> ----------------
>=20
> _______________________________________________
> Linuxppc-embedded mailing list
> Linuxppc-embedded@ozlabs.org
> https://ozlabs.org/mailman/listinfo/linuxppc-embedded
>=20

^ permalink raw reply	[flat|nested] 19+ messages in thread

* Re: [BUG] IBM440GX works too slow under nfsroot
  2004-10-22  5:57 Daren Hayward
@ 2004-10-22  6:30 ` Gerhard Jaeger
  0 siblings, 0 replies; 19+ messages in thread
From: Gerhard Jaeger @ 2004-10-22  6:30 UTC (permalink / raw)
  To: linuxppc-embedded; +Cc: Daren Hayward

On Friday 22 October 2004 07:57, Daren Hayward wrote:
> What revision of the CPU do you have?? I know that upto rev C half duplex
> has a problem with the cpu and only REV C silicon works if I am not
> mistaken, there is an errata on IBM's website about this
>
> Daren
>
> -----Original Message-----
> From: Eugene Surovegin [mailto:ebs@ebshome.net]
> Sent: 22 October 2004 03:02
> To: Hiroshi DOYU
> Cc: Linux PPC Embedded
> Subject: Re: [BUG] IBM440GX works too slow under nfsroot
>
> On Fri, Oct 22, 2004 at 10:38:59AM +0900, Hiroshi DOYU wrote:
> > On Thu, 21 Oct 2004 18:33:20 -0700
> >
> > Eugene Surovegin <ebs@ebshome.net> wrote:
> > > On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > > > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > > > I could succeed to boot it but, after nfs mounted, it's behavior
> > > > seems too slow. I guess, it looks like the system is running
> > > > half duplex.
> > >
> > > Yeah, driver thinks it's connected to 10Mb hub. How the board is
> > > actualy connected to the network?
> >
> > 10Mb hub.
>
> Wow, I really had to spend some time to find such ancient hw :). I'll
> try 2.4.27 with this hub during next couple of days (I'm a little busy
> right now) and let you know. I suspect nobody really tested EMAC driver
> in 10Mb HDX mode.
>
> You can also bug Matt Porter, maybe he'll help you sooner :).

Daren,
 
you are right about that, but this seems to be not true, even the latest REV C
seems to have problems - that's what we can observe here :(

Gerhard

-- 
Gerhard Jaeger <gjaeger@sysgo.com>            
SYSGO AG                      Embedded and Real-Time Software
www.sysgo.com | www.elinos.com | www.osek.de | www.imerva.com

^ permalink raw reply	[flat|nested] 19+ messages in thread

* RE: [BUG] IBM440GX works too slow under nfsroot
@ 2004-10-22  5:57 Daren Hayward
  2004-10-22  6:30 ` Gerhard Jaeger
  0 siblings, 1 reply; 19+ messages in thread
From: Daren Hayward @ 2004-10-22  5:57 UTC (permalink / raw)
  To: 'Eugene Surovegin', Hiroshi DOYU; +Cc: Linux PPC Embedded

What revision of the CPU do you have?? I know that upto rev C half duplex
has a problem with the cpu and only REV C silicon works if I am not
mistaken, there is an errata on IBM's website about this

Daren

-----Original Message-----
From: Eugene Surovegin [mailto:ebs@ebshome.net]
Sent: 22 October 2004 03:02
To: Hiroshi DOYU
Cc: Linux PPC Embedded
Subject: Re: [BUG] IBM440GX works too slow under nfsroot


On Fri, Oct 22, 2004 at 10:38:59AM +0900, Hiroshi DOYU wrote:
> On Thu, 21 Oct 2004 18:33:20 -0700
> Eugene Surovegin <ebs@ebshome.net> wrote:
> 
> > On Fri, Oct 22, 2004 at 08:57:33AM +0900, Hiroshi DOYU wrote:
> > > Now I am strugging with booting vanilla(2.4.2x) on ibm440gx and
> > > I could succeed to boot it but, after nfs mounted, it's behavior 
> > > seems too slow. I guess, it looks like the system is running 
> > > half duplex. 
> > 
> > Yeah, driver thinks it's connected to 10Mb hub. How the board is 
> > actualy connected to the network?
> 
> 10Mb hub.
> 

Wow, I really had to spend some time to find such ancient hw :). I'll 
try 2.4.27 with this hub during next couple of days (I'm a little busy 
right now) and let you know. I suspect nobody really tested EMAC driver 
in 10Mb HDX mode.

You can also bug Matt Porter, maybe he'll help you sooner :).

--
Eugene
_______________________________________________
Linuxppc-embedded mailing list
Linuxppc-embedded@ozlabs.org
https://ozlabs.org/mailman/listinfo/linuxppc-embedded

^ permalink raw reply	[flat|nested] 19+ messages in thread

end of thread, other threads:[~2004-11-08  2:55 UTC | newest]

Thread overview: 19+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-10-21 23:57 [BUG] IBM440GX works too slow under nfsroot Hiroshi DOYU
2004-10-22  1:19 ` Eugene Surovegin
2004-10-22  1:26   ` Hiroshi DOYU
2004-10-22  1:33 ` Eugene Surovegin
2004-10-22  1:38   ` Hiroshi DOYU
2004-10-22  2:02     ` Eugene Surovegin
2004-10-22  6:26       ` Gerhard Jaeger
2004-10-22  8:29         ` Eugene Surovegin
2004-10-22  9:39           ` Gerhard Jaeger
2004-10-22 12:22           ` Mark Chambers
2004-10-22 12:27             ` Gerhard Jaeger
2004-10-25  1:16           ` Eugene Surovegin
2004-10-25  6:47             ` Gerhard Jaeger
2004-10-25  7:02               ` Eugene Surovegin
2004-11-08  2:55               ` Eugene Surovegin
2004-10-22 22:22     ` Matt Porter
2004-10-22  5:57 Daren Hayward
2004-10-22  6:30 ` Gerhard Jaeger
2004-10-22  7:36 Neil Wilson

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.