Skip to contentSkip to conversation

IPMI driver updates for 2.4

DISCUSSION Marcelo Tosatti 2 participants 5 messages Aug 31, 2004 – Sep 13, 2004 3 messages not in the archive

lore
Share & tools
lore
#3 lore raw
In reply to a message not in the archive

Hi Corey,

You wrote a nice changelog entry - thanks for that.

I'm not sure if we should apply all of this in v2.4. It looks like a big driver revamp to me.

Is is all this critical for being merged in v2.4.x now?

Would it be very painful/unwanted to maintain only the bugfixes and not only new features? What you think about that?

Quoted 30 lines · Corey Minyard
Quoted 8 lines · Marcelo Tosatti

Corey,

Care to write a detailed changelog so we can apply this?

Thanks

I'm sorry this took so long, I have been dealing with disasters at work.

Ok, a detailed changelog:

* Add a new "system interface" driver that supports all the standard IPMI system interfaces (SMIC, BT, and KCS, see the IPMI spec for more details if you care). This driver will auto-detect the interface type and parameters via SMBIOS or ACPI tables.
* Deprecate the old KCS-only interface.
* Support non-contiguous registers for system interfaces.
* Add support for IPMI LAN bridging so messages can be received from and sent to system software connected to a LAN interface.
* Add support for powering off the system on a halt via the IPMI interface.
* Add support for storing panic logs in the IPMI event log.
* Allow control of message parameters (resends, timeouts)

I've also re-attached the patch.

Thanks,

#4 lore raw

There actually aren't any bug fixes. These are all new feature additions. It is a big driver revamp, but it has been around for a while

I guess it is not critical now, but it would be nicer for the users of the driver.

-Corey

Quoted 50 lines · Marcelo Tosatti

Hi Corey,

You wrote a nice changelog entry - thanks for that.

I'm not sure if we should apply all of this in v2.4. It looks like a big driver revamp to me.

Is is all this critical for being merged in v2.4.x now?

Would it be very painful/unwanted to maintain only the bugfixes and not only new features? What you think about that?

Quoted 36 lines · Corey Minyard
Quoted 10 lines · Marcelo Tosatti

Corey,

Care to write a detailed changelog so we can apply this?

Thanks

I'm sorry this took so long, I have been dealing with disasters at work.

Ok, a detailed changelog:

* Add a new "system interface" driver that supports all the standard IPMI system interfaces (SMIC, BT, and KCS, see the IPMI spec for more details if you care). This driver will auto-detect the interface type and parameters via SMBIOS or ACPI tables.
* Deprecate the old KCS-only interface.
* Support non-contiguous registers for system interfaces.
* Add support for IPMI LAN bridging so messages can be received from and sent to system software connected to a LAN interface.
* Add support for powering off the system on a halt via the IPMI interface.
* Add support for storing panic logs in the IPMI event log.
* Allow control of message parameters (resends, timeouts)

I've also re-attached the patch.

Thanks,

#5 lore raw
Quoted 5 lines · Corey Minyard

There actually aren't any bug fixes. These are all new feature additions. It is a big driver revamp, but it has been around for a while

I guess it is not critical now, but it would be nicer for the users of the driver.

Corey,

I dont have a clue about IPMI. But I would like to apply what is critical to users only at this point.

#6 lore raw

Ok, that's fine.

By "at this point" do you mean for the future of the 2.4 kernel or just this point in time and maybe in the next 2.4 kernel? A lot of the driver users want this in the 2.4 kernel, so I need to tell them what the state will be.

-Corey

Quoted 14 lines · Marcelo Tosatti
Quoted 5 lines · Corey Minyard

There actually aren't any bug fixes. These are all new feature additions. It is a big driver revamp, but it has been around for a while

I guess it is not critical now, but it would be nicer for the users of the driver.

Corey,

I dont have a clue about IPMI. But I would like to apply what is critical to users only at this point.

#7 lore raw
Quoted 6 lines · Corey Minyard

Ok, that's fine.

By "at this point" do you mean for the future of the 2.4 kernel or just this point in time and maybe in the next 2.4 kernel? A lot of the driver users want this in the 2.4 kernel, so I need to tell them what the state will be.

I mean in v2.4, not only v2.4.28.