On Wed, Jul 22, 2009 at 04:23:45PM +0100, Daniel P. Berrange wrote:
* src/qemu_driver.c: Add driver methods qemuGetSchedulerType, qemuGetSchedulerParameters, qemuSetSchedulerParameters * src/lxc_driver.c: Fix to use unsigned long long consistently for schedular parameters * src/cgroup.h, src/cgroup.c: Fix cpu_shares to take unsigned long long * src/util.c, src/util.h, src/libvirt_private.syms: Add a virStrToDouble helper * src/virsh.c: Fix handling of --set arg to schedinfo command to honour the designated data type of each schedular tunable as declared by the driver
Signed-off-by: Daniel P. Berrange <berrange@redhat.com> --- src/cgroup.c | 10 +- src/cgroup.h | 4 +- src/libvirt_private.syms | 1 + src/lxc_driver.c | 9 ++- src/qemu_driver.c | 151 +++++++++++++++++++++++++++++- src/util.c | 20 ++++ src/util.h | 3 + [...] @@ -1355,9 +1355,14 @@ static int lxcSetSchedulerParameters(virDomainPtr domain,
for (i = 0; i < nparams; i++) { virSchedParameterPtr param = ¶ms[i]; + if (param->type != VIR_DOMAIN_SCHED_FIELD_ULLONG) { + lxcError(NULL, domain, VIR_ERR_INVALID_ARG, + _("invalid type for cpu_shares tunable, expected a 'ullong'")); + goto cleanup; + }
Humpf, we are actually breaking the API in some ways there, what is the argument ? Consistency with qemu scheduling parameters ?
if (STREQ(param->field, "cpu_shares")) {
[...]
+ if (STREQ(param->field, "cpu_shares")) { + int rc; + if (param->type != VIR_DOMAIN_SCHED_FIELD_ULLONG) {
Apparently, but is that worth it ? [...]
--- a/src/virsh.c +++ b/src/virsh.c @@ -1181,111 +1181,118 @@ static const vshCmdOptDef opts_schedinfo[] = { };
static int -cmdSchedinfo(vshControl *ctl, const vshCmd *cmd) +cmdSchedInfoUpdate(vshControl *ctl, const vshCmd *cmd, + virSchedParameterPtr param) { [...] + /* Legacy 'weight' parameter */ + if (STREQ(param->field, "weight") && + param->type == VIR_DOMAIN_SCHED_FIELD_UINT && + vshCommandOptBool(cmd, "weight")) { + int val; + val = vshCommandOptInt(cmd, "weight", &found); + if (!found) { vshError(ctl, FALSE, "%s", _("Invalid value of weight")); - goto cleanup; + return -1; } else { - nr_inputparams++; + param->value.ui = val; } + return 1;
This API is and remains a pain, looking at the ESX driver there is just yet another completely different set ... ACK, though the lxc breakage is a bit nasty IMHO Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel@veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/