The post for fixing the 1/2 baud rate problem on epia-m is here: http://www.clustermatic.org/pipermail/linuxbios/2003-December/006207.html
The earlier post mentioned probably had the setserial workaround but this is no good, you don't get 115200 baud or other baud rates that don't go into 57,600 exactly (like 38,400).
I'm on vacation this past week in Florida so haven't been able to assist any on the epia-m work.
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature + voltage monitoring). Other stuff seemed to be hacked and slashed out that I found necessary, but didn't appear to be replaced elsewhere.
When I get back to the office hopefully I can do a cvs checkout of latest version and give it a shot and see if it works for me. It sounds like it is working for everyone else right now, which is good news, so congrats all around.
So if it's this far along, it would be nice to get the SPD auto detection working. Maybe the path to that is integrate all these V1 changes into V2 and do it in V2.
-Dave
On Fri, 16 Apr 2004, Dave Ashley wrote:
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature + voltage monitoring). Other stuff seemed to be hacked and slashed out that I found necessary, but didn't appear to be replaced elsewhere.
don't worry. I did not apply the patch Dmitri sent. I just used it as a guide to the actual changes I made. All your stuff is still in there. Check the CVS and you should see your stuff intact. I really need that serial port patch though.
So if it's this far along, it would be nice to get the SPD auto detection working. Maybe the path to that is integrate all these V1 changes into V2 and do it in V2.
That's the next step. Once you bless this V1 port, I start the V2 port. That will become the port of choice for EPIA-M.
I'll need some help with that one ... this is a time-available thing for me right now, but I'm giving it all the time I can.
ron
On Fri, 16 Apr 2004, Dave Ashley wrote:
The post for fixing the 1/2 baud rate problem on epia-m is here: http://www.clustermatic.org/pipermail/linuxbios/2003-December/006207.html
OK, this fix is in the tree, build works, so I've committed it. I think it should work fine, but I'm still home, will test as soon as I get in.
ron
On Fri, 16 Apr 2004, ron minnich wrote:
On Fri, 16 Apr 2004, Dave Ashley wrote:
The post for fixing the 1/2 baud rate problem on epia-m is here: http://www.clustermatic.org/pipermail/linuxbios/2003-December/006207.html
OK, this fix is in the tree, build works, so I've committed it. I think it should work fine, but I'm still home, will test as soon as I get in.
The fix works fine.
Now if only Linux, once loaded, would show reasonable output on the serial port. Weird.
ron
* ron minnich rminnich@lanl.gov [040416 17:13]:
OK, this fix is in the tree, build works, so I've committed it. I think it should work fine, but I'm still home, will test as soon as I get in.
The fix works fine.
Now if only Linux, once loaded, would show reasonable output on the serial port. Weird.
What's the problem? I had something similar with my geode box at home. As soon as Linux 2.6 tries to initialize the serial port, the serial console breaks away, hanging the system. Could not resolve this until now
Stefan
On Fri, 16 Apr 2004, Stefan Reinauer wrote:
What's the problem? I had something similar with my geode box at home. As soon as Linux 2.6 tries to initialize the serial port, the serial console breaks away, hanging the system. Could not resolve this until now
I don't know, but with vga console working I hope to have a linux that will drive vga console, allowing me to debug serial port.
Monday.
ron
Dmitry, I can not not move the vga stuff before final_mainboard_fixup -- it won't work if I do. I think in general hardwaremain should wait to touch VGA until this point.
You can take a look at CVS if you want and see what to fix in epia-m to make this go.
Serial port now works fine at the right speeds.
Dave, see what you think of CVS.
Filo works fine, but I can not yet elfboot a kernel image from IDE-FLASH. The kernel just prints lots of ????????????? and eventually blows up. Weird. I can't get to a baud rate setting that works.
I gotta go do some paying work, and I want to see what you all think of the current state of things anyway, so please get back to me and let me know how it all looks.
ron
No problem, We might just do some temporary hack for EPIA-M only and leave it as a voluntary patch. The splash screen is a really nice thing for embedded devices. I can update HOWTO with this information. Dmitry/
----- Original Message ----- Subject: Re: epia-m 1/2 vga
Dmitry, I can not not move the vga stuff before final_mainboard_fixup -- it won't work if I do. I think in general hardwaremain should wait to touch VGA until this point.
On Fri, 16 Apr 2004, Dmitry Borisov wrote:
No problem, We might just do some temporary hack for EPIA-M only and leave it as a voluntary patch. The splash screen is a really nice thing for embedded devices. I can update HOWTO with this information.
send HOWTO, please look at CVS and see if you can get splash screen working and send me a patch.
ron
Hello LinuxBios fellows,
I know i'am a bit asking for trouble in what i like but if not trying to someone else will anyway (as i've read). I'am trying to bring this fix smbenablebus.inc into v2, following advices and notices of Dave
by build time i got : /mainboard/via/epia/smbus.inc: Assembler message: smbus.inc:49:Error: invalid character "_" in mnemonic smbus.inc:50:Error: invalid character "_" in mnemonic smbus.inc:51:Error: invalid character "_" in mnemonic crt0.S:155:Warning: inderect jmp without '*' make[1]: *** [crt0.o] Error 1 make[1]: *** [fallback-rom] Error 1
It stops in 51 in smbus.inc but there are other mnemonics with this char...
A out-of-the-bit question : I've noticed that we're building a fallbackrom, and i found the concept unavoidable and very interesting, but does/should it needs all the capabilities a normal rom has ? (for example half rate would be enough, you know thinking let's try not to give it too much reasons to break down:-) Indeed i thought fallback rom was some kind of in-case-of-emergency-boot-anyway-please.. just enough abilities some we can repair the normal rom to boot up back from. Well it just a something that passed by me...
Have nice week-end. mathieu
Le ven 16/04/2004 à 15:32, Dave Ashley a écrit :
The post for fixing the 1/2 baud rate problem on epia-m is here: http://www.clustermatic.org/pipermail/linuxbios/2003-December/006207.html
The earlier post mentioned probably had the setserial workaround but this is no good, you don't get 115200 baud or other baud rates that don't go into 57,600 exactly (like 38,400).
I'm on vacation this past week in Florida so haven't been able to assist any on the epia-m work.
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature + voltage monitoring). Other stuff seemed to be hacked and slashed out that I found necessary, but didn't appear to be replaced elsewhere.
When I get back to the office hopefully I can do a cvs checkout of latest version and give it a shot and see if it works for me. It sounds like it is working for everyone else right now, which is good news, so congrats all around.
So if it's this far along, it would be nice to get the SPD auto detection working. Maybe the path to that is integrate all these V1 changes into V2 and do it in V2.
-Dave _______________________________________________ Linuxbios mailing list Linuxbios@clustermatic.org http://www.clustermatic.org/mailman/listinfo/linuxbios
On Fri, 16 Apr 2004, Dave Ashley wrote:
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature
+
voltageOn Fri, 16 Apr 2004, Dave Ashley wrote:
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature
+
voltage monitoring). Other stuff seemed to be hacked and slashed out
that
I found necessary, but didn't appear to be replaced elsewhere.
don't worry. I did not apply the patch Dmitri sent. I just used it as a guide to the actual changes I made. All your stuff is still in there. Check the CVS and you should see your stuff intact. I really need that serial port patch though.
So if it's this far along, it would be nice to get the SPD auto
detection
working. Maybe the path to that is integrate all these V1 changes into
V2
and do it in V2.
That's the next step. Once you bless this V1 port, I start the V2 port. That will become the port of choice for EPIA-M.
I'll needOn Fri, 16 Apr 2004, Dave Ashley wrote:
One thing I noticed in Dmitri's patch is it seemed to cut out all my code to initialize the VT1211's hardware monitor (used for temperature
+
voltage monitoring). Other stuff seemed to be hacked and slashed out
that
I found necessary, but didn't appear to be replaced elsewhere.
don't worry. I did not apply the patch Dmitri sent. I just used it as a guide to the actual changes I made. All your stuff is still in there. Check the CVS and you should see your stuff intact. I really need that serial port patch though.
So if it's this far along, it would be nice to get the SPD auto
detection
working. Maybe the path to that is integrate all these V1 changes into
V2
and do it in V2.
That's the next step. Once you bless this V1 port, I start the V2 port. That will become the port of choice for EPIA-M.
I'll need some help with that one ... this is a time-available thing for me right now, but I'm giving it all the time I can.
ron some help with that one ... this is a time-available thing for me right now, but I'm giving it all the time I can.
ron monitoring). Other stuff seemed to be hacked and slashed out that
I found necessary, but didn't appear to be replaced elsewhere.
don't worry. I did not apply the patch Dmitri sent. I just used it as a guide to the actual changes I made. All your stuff is still in there. Check the CVS and you should see your stuff intact. I really need that serial port patch though.
So if it's this far along, it would be nice to get the SPD auto
detection
working. Maybe the path to that is integrate all these V1 changes into
V2
and do it in V2.
That's the next step. Once you bless this V1 port, I start the V2 port. That will become the port of choice for EPIA-M.
I'll need some help with that one ... this is a time-available thing
for
me right now, but I'm giving it all the time I can.
ron
On 16 Apr 2004, Mathieu Deschamps wrote:
I'am trying to bring this fix smbenablebus.inc into v2, following advices and notices of Dave
don't do this.
Just wait until we're all happy with V1, I will build a V2 version. If you are using .inc files you're going the WRONG direction.
ron
-
Dave Ashley -
Dmitry Borisov -
Mathieu Deschamps -
ron minnich -
Stefan Reinauer