[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 0/5] qdev: device capabilities
From: |
Gerd Hoffmann |
Subject: |
Re: [Qemu-devel] [PATCH 0/5] qdev: device capabilities |
Date: |
Mon, 07 Sep 2009 12:47:21 +0200 |
User-agent: |
Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.1) Gecko/20090814 Fedora/3.0-2.6.b3.fc11 Lightning/1.0pre Thunderbird/3.0b3 |
On 09/05/09 15:39, Anthony Liguori wrote:
Gerd Hoffmann wrote:
Look at patch #5 (watchdogs). What else do you suggest here?
There is nothing which can be used to identify the device as watchdog.
That's a problem with the watchdog timers. The WatchdogTimerModel stuff
really replicates qdev functionality.
Good point, I fully agree.
Patch #5 kills WatchdogTimerModel.
Ideally, the watchdog_list would go away and we would just make
-watchdog an alias for -device.
Patch #5 does just that. Almost. -watchdog continues to accept
watchdog devices only. And '-watchdog ?' lists watchdog devices only.
watchdog_list is gone, the qdev list is used instead. To identify the
watchdog devices in the qdev device model list the capability bit is
used. The patch description says so, doesn't it?
Instead of having a global watchdog
action, we should have a watchdog frontend/backend and allow a user to
specify the action for a watchdog backend. This is basically what
-watchdog-action does.
Sounds complicated. Isn't that over-engineering it a bit? The
-watchdog-action implementation is just two little functions in
watchdog.c right now ...
The current command line options assume one global watchdog. I don't see
why we shouldn't support two though. Even if we limited ourselves to
one, it should use the same infrastructure as everything else.
After applying patch #5 you can have two watchdogs, no problem ;)
cheers,
Gerd