[Ksummit-2012-discuss] [ATTEND] <ding> "Bring out your dead" <ding>...

Marc Zyngier marc.zyngier at arm.com
Wed Jun 27 16:39:30 UTC 2012


On 27/06/12 15:52, Arnd Bergmann wrote:
> On Tuesday 26 June 2012, James Bottomley wrote:
>>> Those are actually more likely to still in the hands of someone who
>>> cares about them than the x86 PCs are. Removing EISA support might
>>> imply killing a few platforms altogether.
>>
>> Well, but why?  EISA support seems to be nicely separated and the bus is
>> discoverable so it's not like you have to poke at random locations to do
>> discovery (unlike ISA); what actual problems is it causing?  I know I
>> build it for PARISC even if I can't use it, so it certainly compiles.
> 
> Agreed, I think removing EISA support by itself helps much, other than
> that it gets us closer to removing ISA card support eventually.
> 
> FWIW, this is a complete list of all device drivers supporting EISA:
> 
> drivers/net/ethernet/3com/3c509.c
> drivers/net/ethernet/3com/3c59x.c
> drivers/net/ethernet/8390/ac3200.c
> drivers/net/ethernet/8390/es3210.c
> drivers/net/ethernet/8390/lne390.c
> drivers/net/ethernet/8390/ne3210.c
> drivers/net/ethernet/8390/smc-ultra32.c
> drivers/net/ethernet/amd/depca.c
> drivers/net/ethernet/dec/tulip/de4x5.c
> drivers/net/ethernet/hp/hp100.c
> drivers/net/ethernet/ti/tlan.c
> drivers/net/fddi/defxx.c
> drivers/scsi/advansys.c
> drivers/scsi/aha1740.c
> drivers/scsi/aic7xxx/aic7770_osm.c
> drivers/scsi/sim710.c

These are mostly the devices that have been converted to the EISA bus
API (I did convert quite a few myself). Quite a few more are actually
not using it (drivers/scsi contains most of them), and have the bad
habit of directly poking into the IO space.

	M. (ex EISA maintainer, Alpha user... ;-)
-- 
Jazz is not dead. It just smells funny...



More information about the Ksummit-2012-discuss mailing list