EEPro100 bug in 2.4.6pre5

EEPro100 bug in 2.4.6pre5

[Alan Cox]

Looking over the merges Im doing from 2.4.6pre I found a couple of suprises
the most dubious of which is the rather bogus hackery on eepro100.c

Someone has done S/CONFIG_EEPRO100_PM/CONFIG_PM/ on the driver and in doing
so permanently enabled the eepro100 pm code which to say the least doesnt work
for a lot of people but gives them weird eepro100 hangs

Since the drivers/net/Config option wasnt removed I can only conclude this was
someones internal hack that leaked out in error

Linus - I think you need to revert that eepro100 change.

Alan

Re: EEPro100 bug in 2.4.6pre5

[Jeff Garzik]

Alan Cox wrote:
> Someone has done S/CONFIG_EEPRO100_PM/CONFIG_PM/ on the driver and in doing
> so permanently enabled the eepro100 pm code which to say the least doesnt work
> for a lot of people but gives them weird eepro100 hangs

Do you have a bug report of this actually breaking?

eepro100 is doing standard PCI PM.  The only reason AFAICS why it was
breaking for people was that the previous PCI PM code did not do all the
stuff it needed to do.  PCI PM should cover the various cases correctly,
now, ditto eepro100.

-- 
Jeff Garzik      | Andre the Giant has a posse.
Building 1024    |
MandrakeSoft     |

Re: EEPro100 bug in 2.4.6pre5

[Alan Cox]

> Do you have a bug report of this actually breaking?

I've not been able to make 2.4.6pre5 stay up long enough to do any real
testing on this. It just keeps hanging all the time anyway

> eepro100 is doing standard PCI PM.  The only reason AFAICS why it was
> breaking for people was that the previous PCI PM code did not do all the
> stuff it needed to do.  PCI PM should cover the various cases correctly,
> now, ditto eepro100.

The they should fix Config.in