Muli Ben-Yehuda wrote:
> On Tue, Feb 06, 2007 at 02:59:48PM -0600, Anthony Liguori wrote:
>
>   
>> Propagate it up and let QEMU decide what to do.
>>
>> There is at least one other user of libkvmctl that I know of.
>>     
>
> Ok, I'll rework it this way (and fix the rest of exit()'s in libkvm
> while I'm at it). Do we agree the right thing for qemu to do on ioctl
> failures at this time is exit()?
>   

I don't think it's the best thing, but it's acceptable IMHO.

Regards,

Anthony Liguori

>> QEMU uses signals quite a bit so I think being defensive on syscalls
>> is best.  It's stung me a few times.
>>     
>
> Syscalls don't return EINTR unless they're written specifically to
> check for sigpending() and return EINTR. I don't think there's any
> point to check for EINTR here when we know that the ioctl() will never
> return it. If that changes, we'll have to audit all of the callers
> anyway and change them.
>
> Cheers,
> Muli
>   


-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier.
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
kvm-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Reply via email to