From 9b02c54bc951fca884ba5719f42a27e8240965bf Mon Sep 17 00:00:00 2001 From: Markus Mayer Date: Mon, 19 Dec 2016 12:10:27 -0800 Subject: cpufreq: brcmstb-avs-cpufreq: extend sysfs entry brcm_avs_pmap We extend the brcm_avs_pmap sysfs entry (which issues the GET_PMAP command to AVS) to include all fields from struct pmap. This means adding mode (AVS, DVS, DVFS) and state (the P-state) to the output. Signed-off-by: Markus Mayer Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/brcmstb-avs-cpufreq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/cpufreq/brcmstb-avs-cpufreq.c b/drivers/cpufreq/brcmstb-avs-cpufreq.c index 4fda623e55bb..2c6e3253ba64 100644 --- a/drivers/cpufreq/brcmstb-avs-cpufreq.c +++ b/drivers/cpufreq/brcmstb-avs-cpufreq.c @@ -954,9 +954,9 @@ static ssize_t show_brcm_avs_pmap(struct cpufreq_policy *policy, char *buf) brcm_avs_parse_p1(pmap.p1, &mdiv_p0, &pdiv, &ndiv); brcm_avs_parse_p2(pmap.p2, &mdiv_p1, &mdiv_p2, &mdiv_p3, &mdiv_p4); - return sprintf(buf, "0x%08x 0x%08x %u %u %u %u %u %u %u\n", + return sprintf(buf, "0x%08x 0x%08x %u %u %u %u %u %u %u %u %u\n", pmap.p1, pmap.p2, ndiv, pdiv, mdiv_p0, mdiv_p1, mdiv_p2, - mdiv_p3, mdiv_p4); + mdiv_p3, mdiv_p4, pmap.mode, pmap.state); } static ssize_t show_brcm_avs_voltage(struct cpufreq_policy *policy, char *buf) -- cgit v1.2.3-59-g8ed1b From 3c223c19aea85d3dda1416c187915f4a30b04b1f Mon Sep 17 00:00:00 2001 From: Markus Mayer Date: Mon, 19 Dec 2016 12:10:28 -0800 Subject: cpufreq: brcmstb-avs-cpufreq: properly retrieve P-state upon suspend The AVS GET_PMAP command does return a P-state along with the P-map information. However, that P-state is the initial P-state when the P-map was first downloaded to AVS. It is *not* the current P-state. Therefore, we explicitly retrieve the P-state using the GET_PSTATE command. Signed-off-by: Markus Mayer Acked-by: Viresh Kumar Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/brcmstb-avs-cpufreq.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/cpufreq/brcmstb-avs-cpufreq.c b/drivers/cpufreq/brcmstb-avs-cpufreq.c index 2c6e3253ba64..c94360671f41 100644 --- a/drivers/cpufreq/brcmstb-avs-cpufreq.c +++ b/drivers/cpufreq/brcmstb-avs-cpufreq.c @@ -784,8 +784,19 @@ static int brcm_avs_target_index(struct cpufreq_policy *policy, static int brcm_avs_suspend(struct cpufreq_policy *policy) { struct private_data *priv = policy->driver_data; + int ret; + + ret = brcm_avs_get_pmap(priv, &priv->pmap); + if (ret) + return ret; - return brcm_avs_get_pmap(priv, &priv->pmap); + /* + * We can't use the P-state returned by brcm_avs_get_pmap(), since + * that's the initial P-state from when the P-map was downloaded to the + * AVS co-processor, not necessarily the P-state we are running at now. + * So, we get the current P-state explicitly. + */ + return brcm_avs_get_pstate(priv, &priv->pmap.state); } static int brcm_avs_resume(struct cpufreq_policy *policy) -- cgit v1.2.3-59-g8ed1b From 6e978b22efa1db9f6e71b24440b5f1d93e968ee3 Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Fri, 3 Feb 2017 14:18:39 -0800 Subject: cpufreq: intel_pstate: Disable energy efficiency optimization Some Kabylake desktop processors may not reach max turbo when running in HWP mode, even if running under sustained 100% utilization. This occurs when the HWP.EPP (Energy Performance Preference) is set to "balance_power" (0x80) -- the default on most systems. It occurs because the platform BIOS may erroneously enable an energy-efficiency setting -- MSR_IA32_POWER_CTL BIT-EE, which is not recommended to be enabled on this SKU. On the failing systems, this BIOS issue was not discovered when the desktop motherboard was tested with Windows, because the BIOS also neglects to provide the ACPI/CPPC table, that Windows requires to enable HWP, and so Windows runs in legacy P-state mode, where this setting has no effect. Linux' intel_pstate driver does not require ACPI/CPPC to enable HWP, and so it runs in HWP mode, exposing this incorrect BIOS configuration. There are several ways to address this problem. First, Linux can also run in legacy P-state mode on this system. As intel_pstate is how Linux enables HWP, booting with "intel_pstate=disable" will run in acpi-cpufreq/ondemand legacy p-state mode. Or second, the "performance" governor can be used with intel_pstate, which will modify HWP.EPP to 0. Or third, starting in 4.10, the /sys/devices/system/cpu/cpufreq/policy*/energy_performance_preference attribute in can be updated from "balance_power" to "performance". Or fourth, apply this patch, which fixes the erroneous setting of MSR_IA32_POWER_CTL BIT_EE on this model, allowing the default configuration to function as designed. Signed-off-by: Srinivas Pandruvada Reviewed-by: Len Brown Cc: 4.6+ # 4.6+ Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/intel_pstate.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index f91c25718d16..86e36544925f 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -1235,6 +1235,25 @@ static void intel_pstate_hwp_enable(struct cpudata *cpudata) cpudata->epp_default = intel_pstate_get_epp(cpudata, 0); } +#define MSR_IA32_POWER_CTL_BIT_EE 19 + +/* Disable energy efficiency optimization */ +static void intel_pstate_disable_ee(int cpu) +{ + u64 power_ctl; + int ret; + + ret = rdmsrl_on_cpu(cpu, MSR_IA32_POWER_CTL, &power_ctl); + if (ret) + return; + + if (!(power_ctl & BIT(MSR_IA32_POWER_CTL_BIT_EE))) { + pr_info("Disabling energy efficiency optimization\n"); + power_ctl |= BIT(MSR_IA32_POWER_CTL_BIT_EE); + wrmsrl_on_cpu(cpu, MSR_IA32_POWER_CTL, power_ctl); + } +} + static int atom_get_min_pstate(void) { u64 value; @@ -1845,6 +1864,11 @@ static const struct x86_cpu_id intel_pstate_cpu_oob_ids[] __initconst = { {} }; +static const struct x86_cpu_id intel_pstate_cpu_ee_disable_ids[] = { + ICPU(INTEL_FAM6_KABYLAKE_DESKTOP, core_params), + {} +}; + static int intel_pstate_init_cpu(unsigned int cpunum) { struct cpudata *cpu; @@ -1875,6 +1899,12 @@ static int intel_pstate_init_cpu(unsigned int cpunum) cpu->cpu = cpunum; if (hwp_active) { + const struct x86_cpu_id *id; + + id = x86_match_cpu(intel_pstate_cpu_ee_disable_ids); + if (id) + intel_pstate_disable_ee(cpunum); + intel_pstate_hwp_enable(cpu); pid_params.sample_rate_ms = 50; pid_params.sample_rate_ns = 50 * NSEC_PER_MSEC; -- cgit v1.2.3-59-g8ed1b From a9306a63631493afc75893a4ac405d4e1cbae6aa Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 4 Feb 2017 00:44:36 +0100 Subject: PM / runtime: Avoid false-positive warnings from might_sleep_if() The might_sleep_if() assertions in __pm_runtime_idle(), __pm_runtime_suspend() and __pm_runtime_resume() may generate false-positive warnings in some situations. For example, that happens if a nested pm_runtime_get_sync()/pm_runtime_put() pair is executed with disabled interrupts within an outer pm_runtime_get_sync()/pm_runtime_put() section for the same device. [Generally, pm_runtime_get_sync() may sleep, so it should not be called with disabled interrupts, but in this particular case the previous pm_runtime_get_sync() guarantees that the device will not be suspended, so the inner pm_runtime_get_sync() will return immediately after incrementing the device's usage counter.] That started to happen in the i915 driver in 4.10-rc, leading to the following splat: BUG: sleeping function called from invalid context at drivers/base/power/runtime.c:1032 in_atomic(): 1, irqs_disabled(): 0, pid: 1500, name: Xorg 1 lock held by Xorg/1500: #0: (&dev->struct_mutex){+.+.+.}, at: [] i915_mutex_lock_interruptible+0x43/0x140 [i915] CPU: 0 PID: 1500 Comm: Xorg Not tainted Call Trace: dump_stack+0x85/0xc2 ___might_sleep+0x196/0x260 __might_sleep+0x53/0xb0 __pm_runtime_resume+0x7a/0x90 intel_runtime_pm_get+0x25/0x90 [i915] aliasing_gtt_bind_vma+0xaa/0xf0 [i915] i915_vma_bind+0xaf/0x1e0 [i915] i915_gem_execbuffer_relocate_entry+0x513/0x6f0 [i915] i915_gem_execbuffer_relocate_vma.isra.34+0x188/0x250 [i915] ? trace_hardirqs_on+0xd/0x10 ? i915_gem_execbuffer_reserve_vma.isra.31+0x152/0x1f0 [i915] ? i915_gem_execbuffer_reserve.isra.32+0x372/0x3a0 [i915] i915_gem_do_execbuffer.isra.38+0xa70/0x1a40 [i915] ? __might_fault+0x4e/0xb0 i915_gem_execbuffer2+0xc5/0x260 [i915] ? __might_fault+0x4e/0xb0 drm_ioctl+0x206/0x450 [drm] ? i915_gem_execbuffer+0x340/0x340 [i915] ? __fget+0x5/0x200 do_vfs_ioctl+0x91/0x6f0 ? __fget+0x111/0x200 ? __fget+0x5/0x200 SyS_ioctl+0x79/0x90 entry_SYSCALL_64_fastpath+0x23/0xc6 even though the code triggering it is correct. Unfortunately, the might_sleep_if() assertions in question are too coarse-grained to cover such cases correctly, so make them a bit less sensitive in order to avoid the false-positives. Reported-and-tested-by: Sedat Dilek Signed-off-by: Rafael J. Wysocki --- drivers/base/power/runtime.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 872eac4cb1df..a14fac6a01d3 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -966,13 +966,13 @@ int __pm_runtime_idle(struct device *dev, int rpmflags) unsigned long flags; int retval; - might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); - if (rpmflags & RPM_GET_PUT) { if (!atomic_dec_and_test(&dev->power.usage_count)) return 0; } + might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); + spin_lock_irqsave(&dev->power.lock, flags); retval = rpm_idle(dev, rpmflags); spin_unlock_irqrestore(&dev->power.lock, flags); @@ -998,13 +998,13 @@ int __pm_runtime_suspend(struct device *dev, int rpmflags) unsigned long flags; int retval; - might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); - if (rpmflags & RPM_GET_PUT) { if (!atomic_dec_and_test(&dev->power.usage_count)) return 0; } + might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); + spin_lock_irqsave(&dev->power.lock, flags); retval = rpm_suspend(dev, rpmflags); spin_unlock_irqrestore(&dev->power.lock, flags); @@ -1029,7 +1029,8 @@ int __pm_runtime_resume(struct device *dev, int rpmflags) unsigned long flags; int retval; - might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe); + might_sleep_if(!(rpmflags & RPM_ASYNC) && !dev->power.irq_safe && + dev->power.runtime_status != RPM_ACTIVE); if (rpmflags & RPM_GET_PUT) atomic_inc(&dev->power.usage_count); -- cgit v1.2.3-59-g8ed1b