If you want to do actually use interrupts you can, but most drivers don't. See amd756_do_pause() for an example of the easy way... Also, I checked in the patch to mkpatch/FILES, but for full mkpatch support there is more to be done, see the bottom of doc/developers/new_drivers for details. If you would do that too I would be most appreciative... thanks mds Vojtech Pavlik wrote: > > On Sun, Nov 17, 2002 at 01:36:32PM -0500, Mark D. Studebaker wrote: > > > your patch looks good and I'm going to check it in now. > > I'm going to add some text to the docs to say that each driver supports > > one of the two interfaces on the chip. > > > > The i2c-8111 driver busy waits with udelay(). This shouldn't be > > necessary. > > Would you please change to an interruptible call (see the other drivers > > for examples) and send me a patch against the patch below. > > Ok, I will. I might even implement support for interrupt-based > completion notification in the driver. > > > thanks > > mds > > > > Vojtech Pavlik wrote: > > > > > > On Sun, Nov 10, 2002 at 02:35:00PM -0500, Mark D. Studebaker wrote: > > > > > > > OK. > > > > I did the SMBus 2.0 support in i2c-core and i2c-i801, you're the > > > > first person I know of to use it besides me, I'll look at what you did > > > > and I welcome any comments. Have you tested the 2.0 support? > > > > > > Since the only devices so far connected to the 2.0 bus of my 8111 do not > > > support any of the 2.0 features, I don't think I can really test the PEC > > > or ARP functionality. > > > > > > > Nobody working on 6650 or 1026; if you start, let us know, > > > > and we will update our new drivers page to say that. > > > > > > I haven't started yet, but I will implement those, since I need them. > > > > > > > Vojtech Pavlik wrote: > > > > > > > > > > On Thu, Nov 07, 2002 at 05:48:53PM -0500, Mark Studebaker wrote: > > > > > > > > > > > can you send me the 8111 datasheet please? > > > > > > > > > > I'm sorry, I can not. I thought it was public, but apparently it is not > > > > > yet. It'll appear on the AMD website sooner or later, though. > > > > > > > > > > Anyway, the new driver together with the ACPI 2.0 spec should be enough > > > > > self-explaining, and I can tell you that the 8111 spec is mostly a > > > > > cut-and-paste from the ACPI 2.0 spec in the SMBus 2.0 controller > > > > > chapter. > > > > > > > > > > Another issue - I'll be needing drivers for the max6650 and adm1026 > > > > > monitoring chips - are there any plans for supporting those, or do I > > > > > have to write the support myself? > > > > > > > > > > Vojtech > > > > > > > > > > > Vojtech Pavlik wrote: > > > > > > > > > > > > > > Hi! > > > > > > > > > > > > > > The AMD-8111 IO-Hub (which you likely haven't seen yet, but definitely > > > > > > > will at some point in time), has two SMBus interfaces, one SMBus 1.0 > > > > > > > interface, same as in other AMD chipsets (AMD-756, etc), and one SMBus > > > > > > > 2.0 interface modelled after the ACPI 2.0 SMBus interface specification. > > > > > > > > > > > > > > The attached patches add support for both the interfaces, one by > > > > > > > modifying the i2c-amd756 driver, and the second by creating a new > > > > > > > i2c-amd8111 driver. > > > > > > > > > > > > > > I tried to write the new driver as cleanly as possible, so I hope you'll > > > > > > > be able to include the patches into the next lm_sensors release without > > > > > > > any major problems. > > > > > > > > > > > > > > Thanks, > > > > > > > -- > > > > > > > Vojtech Pavlik > > > > > > > SuSE Labs > > > > > > > > > > > > > > -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- > > > > > > > > > > > > > > i2c-2.6.5-8111.diffName: i2c-2.6.5-8111.diff > > > > > > > Type: Plain Text (text/plain) > > > > > > > > > > > > > > lm_sensors-2.6.5-8111.diffName: lm_sensors-2.6.5-8111.diff > > > > > > > Type: Plain Text (text/plain) > > > > > > > > > > -- > > > > > Vojtech Pavlik > > > > > SuSE Labs > > > > > > -- > > > Vojtech Pavlik > > > SuSE Labs > > -- > Vojtech Pavlik > SuSE Labs
