On 08/30/2012 02:55 AM, Hu Tao wrote:
> if (ctrl->def->cpumask) {
> /* XXX why don't we keep 'cpumask' in the libvirt cpumap
> * format to start with ?!?! */
This comment now feels a bit out of date.
> for (i = 0 ; i < maxcpu && i < ctrl->def->cpumasklen ; i++)
> if (ctrl->def->cpumask[i])
> - VIR_USE_CPU(cpumap, i);
> + ignore_value(virBitmapSetBit(cpumap, i));
Why isn't ctrl->def->cpumask also stored as a virBitmap, at which point
this would be a copy operation instead of a hand-rolled loop?
>
> /* We are pressuming we are running between fork/exec of LXC
While you are here, s/pressuming/presuming/
> * so use '0' to indicate our own process ID. No threads are
> * running at this point
> */
> - if (virProcessInfoSetAffinity(0, /* Self */
> - cpumap, cpumaplen, maxcpu) < 0) {
> + if (virProcessInfoSetAffinity(0 /* Self */, cpumap) < 0) {
> VIR_FREE(cpumap);
Oops - this cannot be VIR_FREE(cpumap)...
> return -1;
> }
> - VIR_FREE(cpumap);
> + virBitmapFree(cpumap);
...but must match this change to virBitmapFree(cpumap).
> @@ -4044,10 +4046,10 @@ qemudDomainPinEmulator(virDomainPtr dom,
> virNodeInfo nodeinfo;
> int ret = -1;
> qemuDomainObjPrivatePtr priv;
> - bool canResetting = true;
> - int pcpu;
> + bool canResetting = false;
'canResetting' sounds odd. While you are touching this, can you rename
it to something saner, like 'doReset'?
> +++ b/src/util/processinfo.c
> @@ -59,8 +57,10 @@ realloc:
> }
>
> CPU_ZERO_S(masklen, mask);
> - for (i = 0 ; i < maxcpu ; i++) {
> - if (VIR_CPU_USABLE(map, maplen, 0, i))
> + for (i = 0 ; i < virBitmapSize(map); i++) {
> + if (virBitmapGetBit(map, i, &set) < 0)
> + return -1;
> + if (set)
> CPU_SET_S(i, masklen, mask);
Another case where virBitmap should do this in a single function call,
rather than you rolling the loop.
--
Eric Blake eblake@xxxxxxxxxx +1-919-301-3266
Libvirt virtualization library http://libvirt.org
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list