| Message ID | 161600382766.1780699.6787739229984093959.stgit@bahia.lan (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v2] target/ppc/kvm: Cache timebase frequency | expand |
On Wed, Mar 17, 2021 at 06:57:07PM +0100, Greg Kurz wrote: > Each vCPU core exposes its timebase frequency in the DT. When running > under KVM, this means parsing /proc/cpuinfo in order to get the timebase > frequency of the host CPU. > > The parsing appears to slow down the boot quite a bit with higher number > of cores: > > # of cores seconds spent in spapr_dt_cpus() > 8 0.550122 > 16 1.342375 > 32 2.850316 > 64 5.922505 > 96 9.109224 > 128 12.245504 > 256 24.957236 > 384 37.389113 > > The timebase frequency of the host CPU is identical for all > cores and it is an invariant for the VM lifetime. Cache it > instead of doing the same expensive parsing again and again. > > Rename kvmppc_get_tbfreq() to kvmppc_get_tbfreq_procfs() and > rename the 'retval' variable to make it clear it is used as > fallback only. Come up with a new version of kvmppc_get_tbfreq() > that calls kvmppc_get_tbfreq_procfs() only once and keep the > value in a static. > > Zero is certainly not a valid value for the timebase frequency. > Treat atoi() returning zero as another parsing error and return > the fallback value instead. This allows kvmppc_get_tbfreq() to > use zero as an indicator that kvmppc_get_tbfreq_procfs() hasn't > been called yet. > > With this patch applied: > > 384 0.518382 > > Signed-off-by: Greg Kurz <groug@kaod.org> Applied to ppc-for-6.0, thanks. > --- > v2: - do the caching in a distinct function for clarity (Philippe) > - rename 'retval' to 'tbfreq_fallback' > - expand the changelog a bit > --- > target/ppc/kvm.c | 25 +++++++++++++++++++------ > 1 file changed, 19 insertions(+), 6 deletions(-) > > diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c > index 298c1f882c67..104a308abb57 100644 > --- a/target/ppc/kvm.c > +++ b/target/ppc/kvm.c > @@ -1815,24 +1815,37 @@ static int read_cpuinfo(const char *field, char *value, int len) > return ret; > } > > -uint32_t kvmppc_get_tbfreq(void) > +static uint32_t kvmppc_get_tbfreq_procfs(void) > { > char line[512]; > char *ns; > - uint32_t retval = NANOSECONDS_PER_SECOND; > + uint32_t tbfreq_fallback = NANOSECONDS_PER_SECOND; > + uint32_t tbfreq_procfs; > > if (read_cpuinfo("timebase", line, sizeof(line))) { > - return retval; > + return tbfreq_fallback; > } > > ns = strchr(line, ':'); > if (!ns) { > - return retval; > + return tbfreq_fallback; > } > > - ns++; > + tbfreq_procfs = atoi(++ns); > + > + /* 0 is certainly not acceptable by the guest, return fallback value */ > + return tbfreq_procfs ? tbfreq_procfs : tbfreq_fallback; > +} > + > +uint32_t kvmppc_get_tbfreq(void) > +{ > + static uint32_t cached_tbfreq; > + > + if (!cached_tbfreq) { > + cached_tbfreq = kvmppc_get_tbfreq_procfs(); > + } > > - return atoi(ns); > + return cached_tbfreq; > } > > bool kvmppc_get_host_serial(char **value) > >
diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c index 298c1f882c67..104a308abb57 100644 --- a/target/ppc/kvm.c +++ b/target/ppc/kvm.c @@ -1815,24 +1815,37 @@ static int read_cpuinfo(const char *field, char *value, int len) return ret; } -uint32_t kvmppc_get_tbfreq(void) +static uint32_t kvmppc_get_tbfreq_procfs(void) { char line[512]; char *ns; - uint32_t retval = NANOSECONDS_PER_SECOND; + uint32_t tbfreq_fallback = NANOSECONDS_PER_SECOND; + uint32_t tbfreq_procfs; if (read_cpuinfo("timebase", line, sizeof(line))) { - return retval; + return tbfreq_fallback; } ns = strchr(line, ':'); if (!ns) { - return retval; + return tbfreq_fallback; } - ns++; + tbfreq_procfs = atoi(++ns); + + /* 0 is certainly not acceptable by the guest, return fallback value */ + return tbfreq_procfs ? tbfreq_procfs : tbfreq_fallback; +} + +uint32_t kvmppc_get_tbfreq(void) +{ + static uint32_t cached_tbfreq; + + if (!cached_tbfreq) { + cached_tbfreq = kvmppc_get_tbfreq_procfs(); + } - return atoi(ns); + return cached_tbfreq; } bool kvmppc_get_host_serial(char **value)
Each vCPU core exposes its timebase frequency in the DT. When running under KVM, this means parsing /proc/cpuinfo in order to get the timebase frequency of the host CPU. The parsing appears to slow down the boot quite a bit with higher number of cores: # of cores seconds spent in spapr_dt_cpus() 8 0.550122 16 1.342375 32 2.850316 64 5.922505 96 9.109224 128 12.245504 256 24.957236 384 37.389113 The timebase frequency of the host CPU is identical for all cores and it is an invariant for the VM lifetime. Cache it instead of doing the same expensive parsing again and again. Rename kvmppc_get_tbfreq() to kvmppc_get_tbfreq_procfs() and rename the 'retval' variable to make it clear it is used as fallback only. Come up with a new version of kvmppc_get_tbfreq() that calls kvmppc_get_tbfreq_procfs() only once and keep the value in a static. Zero is certainly not a valid value for the timebase frequency. Treat atoi() returning zero as another parsing error and return the fallback value instead. This allows kvmppc_get_tbfreq() to use zero as an indicator that kvmppc_get_tbfreq_procfs() hasn't been called yet. With this patch applied: 384 0.518382 Signed-off-by: Greg Kurz <groug@kaod.org> --- v2: - do the caching in a distinct function for clarity (Philippe) - rename 'retval' to 'tbfreq_fallback' - expand the changelog a bit --- target/ppc/kvm.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-)