| Message ID | 20231104105907.1365392-1-vincent.guittot@linaro.org |
|---|---|
| Headers | show |
| Series | consolidate and cleanup CPU capacity | expand |
On Sat, Nov 04, 2023 at 11:59:01AM +0100, Vincent Guittot wrote: > Create a new method to get a unique and fixed max frequency. Currently > cpuinfo.max_freq or the highest (or last) state of performance domain are > used as the max frequency when computing the frequency for a level of > utilization but: > - cpuinfo_max_freq can change at runtime. boost is one example of > such change. > - cpuinfo.max_freq and last item of the PD can be different leading to > different results between cpufreq and energy model. > > We need to save the reference frequency that has been used when computing > the CPUs capacity and use this fixed and coherent value to convert between > frequency and CPU's capacity. > > In fact, we already save the frequency that has been used when computing > the capacity of each CPU. We extend the precision to save kHz instead of > MHz currently and we modify the type to be aligned with other variables > used when converting frequency to capacity and the other way. > For all the topology related changes here and patch 6/7 and 7/7: Acked-by: Sudeep Holla <sudeep.holla@arm.com>
On Tue, 7 Nov 2023 at 13:01, Will Deacon <will@kernel.org> wrote: > > On Tue, Nov 07, 2023 at 12:18:20PM +0100, Vincent Guittot wrote: > > On Tue, 7 Nov 2023 at 11:38, Will Deacon <will@kernel.org> wrote: > > > > > > On Sat, Nov 04, 2023 at 11:59:07AM +0100, Vincent Guittot wrote: > > > > Use the new capacity_ref_freq to set the ratio that is used by AMU for > > > > computing the arch_scale_freq_capacity(). > > > > This helps to keep everything aligned using the same reference for > > > > computing CPUs capacity. > > > > > > > > The default value of the ratio (stored in per_cpu(arch_max_freq_scale)) > > > > ensures that arch_scale_freq_capacity() returns max capacity until it is > > > > set to its correct value with the cpu capacity and capacity_ref_freq. > > > > > > > > Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> > > > > --- > > > > arch/arm64/kernel/topology.c | 26 ++++++++++++++------------ > > > > drivers/base/arch_topology.c | 12 +++++++++++- > > > > include/linux/arch_topology.h | 1 + > > > > 3 files changed, 26 insertions(+), 13 deletions(-) > > > > > > > > diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c > > > > index 817d788cd866..615c1a20129f 100644 > > > > --- a/arch/arm64/kernel/topology.c > > > > +++ b/arch/arm64/kernel/topology.c > > > > @@ -82,7 +82,12 @@ int __init parse_acpi_topology(void) > > > > #undef pr_fmt > > > > #define pr_fmt(fmt) "AMU: " fmt > > > > > > > > -static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale); > > > > +/* > > > > + * Ensure that amu_scale_freq_tick() will return SCHED_CAPACITY_SCALE until > > > > + * the CPU capacity and its associated frequency have been correctly > > > > + * initialized. > > > > + */ > > > > +static DEFINE_PER_CPU_READ_MOSTLY(unsigned long, arch_max_freq_scale) = 1UL << (2 * SCHED_CAPACITY_SHIFT); > > > > > > This doesn't seem to match the comment? SCHED_CAPACITY_SCALE doesn't have > > > the '2 *' multiplier. > > > > The comment in freq_inv_set_max_ratio() says: > > > > * We use a factor of 2 * SCHED_CAPACITY_SHIFT -> SCHED_CAPACITY_SCALE² > > * in order to ensure a good resolution for arch_max_freq_scale for > > * very low reference frequencies (down to the KHz range which should > > * be unlikely). > > > > Then there is a " * arch_max_freq_scale) >> SCHED_CAPACITY_SHIFT" > > when computing the scale which brings the result back to > > SCHED_CAPACITY_SHIFT > > Ah, I see; I'd not spotted that amu_scale_freq_tick() is doing some > arithmetic on the value (it doesn't return anything because it's 'void'). > It's slightly confusing because the comment talks about SCHED_CAPACITY_SCALE > whereas all the code works on the shift, but I get it now. > > > > > static DEFINE_PER_CPU(u64, arch_const_cycles_prev); > > > > static DEFINE_PER_CPU(u64, arch_core_cycles_prev); > > > > static cpumask_var_t amu_fie_cpus; > > > > @@ -112,14 +117,14 @@ static inline bool freq_counters_valid(int cpu) > > > > return true; > > > > } > > > > > > > > -static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate) > > > > +void freq_inv_set_max_ratio(int cpu, u64 max_rate) > > > > { > > > > - u64 ratio; > > > > + u64 ratio, ref_rate = arch_timer_get_rate(); > > > > > > > > if (unlikely(!max_rate || !ref_rate)) { > > > > - pr_debug("CPU%d: invalid maximum or reference frequency.\n", > > > > + WARN_ONCE(1, "CPU%d: invalid maximum or reference frequency.\n", > > > > cpu); > > > > - return -EINVAL; > > > > + return; > > > > } > > > > > > > > /* > > > > @@ -139,12 +144,12 @@ static int freq_inv_set_max_ratio(int cpu, u64 max_rate, u64 ref_rate) > > > > ratio = div64_u64(ratio, max_rate); > > > > if (!ratio) { > > > > WARN_ONCE(1, "Reference frequency too low.\n"); > > > > - return -EINVAL; > > > > + return; > > > > } > > > > > > > > - per_cpu(arch_max_freq_scale, cpu) = (unsigned long)ratio; > > > > + WRITE_ONCE(per_cpu(arch_max_freq_scale, cpu), (unsigned long)ratio); > > > > > > Why is WRITE_ONCE() now needed? > > > > the tick can already use it. We want to make sure to use either the > > old or the new one but not an intermediate value > > Isn't that already the case without this patch? In other words, this should > be a separate change. Currently the ratio is set in the same notifier callback before registering the AMU function in topology_scale_freq_tick whereas there are now done in 2 separate notifier callbacks (still for the same event) but we don't know the order > > Will
This is the 1st part of consolidating how the max compute capacity is used in the scheduler and how we calculate the frequency for a level of utilization. Fix some unconsistancy when computing frequency for an utilization. There can be a mismatch between energy model and schedutil. Next step will be to make a difference between the original max compute capacity of a CPU and what is currently available when there is a capping applying forever (i.e. seconds or more). Changes since v4: - Capitalize the verb in subject - Remove usless parentheses in cppc_get_dmi_max_khz() - Use freq_ref pattern everywhere - Fix MHz / kHz units conversion for cppc_cpufreq - Move default definition of arch_scale_freq_ref() in include/linux/sched/topology.h beside arch_scale_cpu_capacity which faces similar default declaration behavior. This location covers all cases with arch and CONFIG_* which was not the case with previous attempts. Changes since v3: - Split patch 5 cpufreq/cppc - Fix topology_init_cpu_capacity_cppc() - Fix init if AMU ratio - Added some tags Changes since v2: - Remove the 1st patch which has been queued in tip - Rework how to initialize the reference frequency for cppc_cpufreq and change topology_init_cpu_capacity_cppc() to also set capacity_ref_freq - Add a RFC to convert AMU to use arch_scale_freq_ref and move the config of the AMU ratio to be done when intializing cpu capacity and capacity_ref_freq - Added some tags Changes since v1: - Fix typos - Added changes in cpufreq to use arch_scale_freq_ref() when calling arch_set_freq_scale (patch 3). - arch_scale_freq_ref() is always defined and returns 0 (as proposed by Ionela) when not defined by the arch. This simplifies the code with the addition of patch 3. - Simplify Energy Model which always uses arch_scale_freq_ref(). The latter returns 0 when not defined by arch instead of last item of the perf domain. This is not a problem because the function is only defined for compilation purpose in this case and we don't care about the returned value. (patch 5) - Added changes in cppc cpufreq to set capacity_ref_freq (patch 6) - Added reviewed tag for patch 1 which got a minor change but not for others as I did some changes which could make previous reviewed tag no more relevant. Vincent Guittot (7): topology: Add a new arch_scale_freq_reference cpufreq: Use the fixed and coherent frequency for scaling capacity cpufreq/schedutil: Use a fixed reference frequency energy_model: Use a fixed reference frequency cpufreq/cppc: Move and rename cppc_cpufreq_{perf_to_khz|khz_to_perf} cpufreq/cppc: Set the frequency used for computing the capacity arm64/amu: Use capacity_ref_freq to set AMU ratio arch/arm/include/asm/topology.h | 1 + arch/arm64/include/asm/topology.h | 1 + arch/arm64/kernel/topology.c | 26 +++--- arch/riscv/include/asm/topology.h | 1 + drivers/acpi/cppc_acpi.c | 104 ++++++++++++++++++++++ drivers/base/arch_topology.c | 56 ++++++++---- drivers/cpufreq/cppc_cpufreq.c | 139 ++++-------------------------- drivers/cpufreq/cpufreq.c | 4 +- include/acpi/cppc_acpi.h | 2 + include/linux/arch_topology.h | 8 ++ include/linux/cpufreq.h | 1 + include/linux/energy_model.h | 6 +- include/linux/sched/topology.h | 8 ++ kernel/sched/cpufreq_schedutil.c | 26 +++++- 14 files changed, 225 insertions(+), 158 deletions(-)