--- /dev/null
+From c4ee841f602e5eef8eab673295c49c5b49d7732b Mon Sep 17 00:00:00 2001
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Date: Thu, 29 May 2014 09:32:24 -0700
+Subject: intel_pstate: add sample time scaling
+
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+
+commit c4ee841f602e5eef8eab673295c49c5b49d7732b upstream.
+
+The PID assumes that samples are of equal time, which for a deferable
+timers this is not true when the system goes idle. This causes the
+PID to take a long time to converge to the min P state and depending
+on the pattern of the idle load can make the P state appear stuck.
+
+The hold-off value of three sample times before using the scaling is
+to give a grace period for applications that have high performance
+requirements and spend a lot of time idle, The poster child for this
+behavior is the ffmpeg benchmark in the Phoronix test suite.
+
+Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/cpufreq/intel_pstate.c | 18 +++++++++++++++++-
+ 1 file changed, 17 insertions(+), 1 deletion(-)
+
+--- a/drivers/cpufreq/intel_pstate.c
++++ b/drivers/cpufreq/intel_pstate.c
+@@ -60,6 +60,7 @@ struct sample {
+ u64 aperf;
+ u64 mperf;
+ int freq;
++ ktime_t time;
+ };
+
+ struct pstate_data {
+@@ -97,6 +98,7 @@ struct cpudata {
+ struct vid_data vid;
+ struct _pid pid;
+
++ ktime_t last_sample_time;
+ u64 prev_aperf;
+ u64 prev_mperf;
+ struct sample sample;
+@@ -583,6 +585,8 @@ static inline void intel_pstate_sample(s
+ aperf = aperf >> FRAC_BITS;
+ mperf = mperf >> FRAC_BITS;
+
++ cpu->last_sample_time = cpu->sample.time;
++ cpu->sample.time = ktime_get();
+ cpu->sample.aperf = aperf;
+ cpu->sample.mperf = mperf;
+ cpu->sample.aperf -= cpu->prev_aperf;
+@@ -605,12 +609,24 @@ static inline void intel_pstate_set_samp
+
+ static inline int32_t intel_pstate_get_scaled_busy(struct cpudata *cpu)
+ {
+- int32_t core_busy, max_pstate, current_pstate;
++ int32_t core_busy, max_pstate, current_pstate, sample_ratio;
++ u32 duration_us;
++ u32 sample_time;
+
+ core_busy = cpu->sample.core_pct_busy;
+ max_pstate = int_tofp(cpu->pstate.max_pstate);
+ current_pstate = int_tofp(cpu->pstate.current_pstate);
+ core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
++
++ sample_time = (pid_params.sample_rate_ms * USEC_PER_MSEC);
++ duration_us = (u32) ktime_us_delta(cpu->sample.time,
++ cpu->last_sample_time);
++ if (duration_us > sample_time * 3) {
++ sample_ratio = div_fp(int_tofp(sample_time),
++ int_tofp(duration_us));
++ core_busy = mul_fp(core_busy, sample_ratio);
++ }
++
+ return core_busy;
+ }
+
--- /dev/null
+From f0fe3cd7e12d8290c82284b5c8aee723cbd0371a Mon Sep 17 00:00:00 2001
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Date: Thu, 29 May 2014 09:32:23 -0700
+Subject: intel_pstate: Correct rounding in busy calculation
+
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+
+commit f0fe3cd7e12d8290c82284b5c8aee723cbd0371a upstream.
+
+Changing to fixed point math throughout the busy calculation in
+commit e66c1768 (Change busy calculation to use fixed point
+math.) Introduced some inaccuracies by rounding the busy value at two
+points in the calculation. This change removes roundings and moves
+the rounding to the output of the PID where the calculations are
+complete and the value returned as an integer.
+
+Fixes: e66c17683746 (intel_pstate: Change busy calculation to use fixed point math.)
+Reported-by: Doug Smythies <dsmythies@telus.net>
+Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/cpufreq/intel_pstate.c | 12 +++++++-----
+ 1 file changed, 7 insertions(+), 5 deletions(-)
+
+--- a/drivers/cpufreq/intel_pstate.c
++++ b/drivers/cpufreq/intel_pstate.c
+@@ -40,10 +40,10 @@
+ #define BYT_TURBO_VIDS 0x66d
+
+
+-#define FRAC_BITS 6
++#define FRAC_BITS 8
+ #define int_tofp(X) ((int64_t)(X) << FRAC_BITS)
+ #define fp_toint(X) ((X) >> FRAC_BITS)
+-#define FP_ROUNDUP(X) ((X) += 1 << FRAC_BITS)
++
+
+ static inline int32_t mul_fp(int32_t x, int32_t y)
+ {
+@@ -198,7 +198,10 @@ static signed int pid_calc(struct _pid *
+ pid->last_err = fp_error;
+
+ result = pterm + mul_fp(pid->integral, pid->i_gain) + dterm;
+-
++ if (result >= 0)
++ result = result + (1 << (FRAC_BITS-1));
++ else
++ result = result - (1 << (FRAC_BITS-1));
+ return (signed int)fp_toint(result);
+ }
+
+@@ -563,7 +566,6 @@ static inline void intel_pstate_calc_bus
+ core_pct = div_fp(int_tofp((sample->aperf)),
+ int_tofp((sample->mperf)));
+ core_pct = mul_fp(core_pct, int_tofp(100));
+- FP_ROUNDUP(core_pct);
+
+ sample->freq = fp_toint(
+ mul_fp(int_tofp(cpu->pstate.max_pstate * 1000), core_pct));
+@@ -609,7 +611,7 @@ static inline int32_t intel_pstate_get_s
+ max_pstate = int_tofp(cpu->pstate.max_pstate);
+ current_pstate = int_tofp(cpu->pstate.current_pstate);
+ core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
+- return FP_ROUNDUP(core_busy);
++ return core_busy;
+ }
+
+ static inline void intel_pstate_adjust_busy_pstate(struct cpudata *cpu)
--- /dev/null
+From bf8102228a8bf053051f311e5486042fe0542894 Mon Sep 17 00:00:00 2001
+From: Doug Smythies <dsmythies@telus.net>
+Date: Fri, 30 May 2014 10:10:57 -0700
+Subject: intel_pstate: Improve initial busy calculation
+
+From: Doug Smythies <dsmythies@telus.net>
+
+commit bf8102228a8bf053051f311e5486042fe0542894 upstream.
+
+This change makes the busy calculation using 64 bit math which prevents
+overflow for large values of aperf/mperf.
+
+Signed-off-by: Doug Smythies <dsmythies@telus.net>
+Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/cpufreq/intel_pstate.c | 13 ++++++++-----
+ 1 file changed, 8 insertions(+), 5 deletions(-)
+
+--- a/drivers/cpufreq/intel_pstate.c
++++ b/drivers/cpufreq/intel_pstate.c
+@@ -563,16 +563,19 @@ static void intel_pstate_get_cpu_pstates
+ static inline void intel_pstate_calc_busy(struct cpudata *cpu,
+ struct sample *sample)
+ {
+- int32_t core_pct;
++ int64_t core_pct;
++ int32_t rem;
+
+- core_pct = div_fp(int_tofp((sample->aperf)),
+- int_tofp((sample->mperf)));
+- core_pct = mul_fp(core_pct, int_tofp(100));
++ core_pct = int_tofp(sample->aperf) * int_tofp(100);
++ core_pct = div_u64_rem(core_pct, int_tofp(sample->mperf), &rem);
++
++ if ((rem << 1) >= int_tofp(sample->mperf))
++ core_pct += 1;
+
+ sample->freq = fp_toint(
+ mul_fp(int_tofp(cpu->pstate.max_pstate * 1000), core_pct));
+
+- sample->core_pct_busy = core_pct;
++ sample->core_pct_busy = (int32_t)core_pct;
+ }
+
+ static inline void intel_pstate_sample(struct cpudata *cpu)
--- /dev/null
+From adacdf3f2b8e65aa441613cf61c4f598e9042690 Mon Sep 17 00:00:00 2001
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Date: Thu, 29 May 2014 09:32:22 -0700
+Subject: intel_pstate: Remove C0 tracking
+
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+
+commit adacdf3f2b8e65aa441613cf61c4f598e9042690 upstream.
+
+Commit fcb6a15c (intel_pstate: Take core C0 time into account for core
+busy calculation) introduced a regression referenced below. The issue
+with "lockup" after suspend that this commit was addressing is now dealt
+with in the suspend path.
+
+Fixes: fcb6a15c2e7e (intel_pstate: Take core C0 time into account for core busy calculation)
+Link: https://bugzilla.kernel.org/show_bug.cgi?id=66581
+Link: https://bugzilla.kernel.org/show_bug.cgi?id=75121
+Reported-by: Doug Smythies <dsmythies@telus.net>
+Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/cpufreq/intel_pstate.c | 13 +------------
+ 1 file changed, 1 insertion(+), 12 deletions(-)
+
+--- a/drivers/cpufreq/intel_pstate.c
++++ b/drivers/cpufreq/intel_pstate.c
+@@ -59,7 +59,6 @@ struct sample {
+ int32_t core_pct_busy;
+ u64 aperf;
+ u64 mperf;
+- unsigned long long tsc;
+ int freq;
+ };
+
+@@ -100,7 +99,6 @@ struct cpudata {
+
+ u64 prev_aperf;
+ u64 prev_mperf;
+- unsigned long long prev_tsc;
+ struct sample sample;
+ };
+
+@@ -561,46 +559,37 @@ static inline void intel_pstate_calc_bus
+ struct sample *sample)
+ {
+ int32_t core_pct;
+- int32_t c0_pct;
+
+ core_pct = div_fp(int_tofp((sample->aperf)),
+ int_tofp((sample->mperf)));
+ core_pct = mul_fp(core_pct, int_tofp(100));
+ FP_ROUNDUP(core_pct);
+
+- c0_pct = div_fp(int_tofp(sample->mperf), int_tofp(sample->tsc));
+-
+ sample->freq = fp_toint(
+ mul_fp(int_tofp(cpu->pstate.max_pstate * 1000), core_pct));
+
+- sample->core_pct_busy = mul_fp(core_pct, c0_pct);
++ sample->core_pct_busy = core_pct;
+ }
+
+ static inline void intel_pstate_sample(struct cpudata *cpu)
+ {
+ u64 aperf, mperf;
+- unsigned long long tsc;
+
+ rdmsrl(MSR_IA32_APERF, aperf);
+ rdmsrl(MSR_IA32_MPERF, mperf);
+- tsc = native_read_tsc();
+
+ aperf = aperf >> FRAC_BITS;
+ mperf = mperf >> FRAC_BITS;
+- tsc = tsc >> FRAC_BITS;
+
+ cpu->sample.aperf = aperf;
+ cpu->sample.mperf = mperf;
+- cpu->sample.tsc = tsc;
+ cpu->sample.aperf -= cpu->prev_aperf;
+ cpu->sample.mperf -= cpu->prev_mperf;
+- cpu->sample.tsc -= cpu->prev_tsc;
+
+ intel_pstate_calc_busy(cpu, &cpu->sample);
+
+ cpu->prev_aperf = aperf;
+ cpu->prev_mperf = mperf;
+- cpu->prev_tsc = tsc;
+ }
+
+ static inline void intel_pstate_set_sample_time(struct cpudata *cpu)
--- /dev/null
+From d37e2b764499e092ebc493d6f980827feb952e23 Mon Sep 17 00:00:00 2001
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Date: Wed, 12 Feb 2014 10:01:04 -0800
+Subject: intel_pstate: remove unneeded sample buffers
+
+From: Dirk Brandewie <dirk.j.brandewie@intel.com>
+
+commit d37e2b764499e092ebc493d6f980827feb952e23 upstream.
+
+Remove unneeded sample buffers, intel_pstate operates on the most
+recent sample only. This save some memory and make the code more
+readable.
+
+Signed-off-by: Dirk Brandewie <dirk.j.brandewie@intel.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/cpufreq/intel_pstate.c | 24 +++++++++++-------------
+ 1 file changed, 11 insertions(+), 13 deletions(-)
+
+--- a/drivers/cpufreq/intel_pstate.c
++++ b/drivers/cpufreq/intel_pstate.c
+@@ -101,8 +101,7 @@ struct cpudata {
+ u64 prev_aperf;
+ u64 prev_mperf;
+ unsigned long long prev_tsc;
+- int sample_ptr;
+- struct sample samples[SAMPLE_COUNT];
++ struct sample sample;
+ };
+
+ static struct cpudata **all_cpu_data;
+@@ -590,15 +589,14 @@ static inline void intel_pstate_sample(s
+ mperf = mperf >> FRAC_BITS;
+ tsc = tsc >> FRAC_BITS;
+
+- cpu->sample_ptr = (cpu->sample_ptr + 1) % SAMPLE_COUNT;
+- cpu->samples[cpu->sample_ptr].aperf = aperf;
+- cpu->samples[cpu->sample_ptr].mperf = mperf;
+- cpu->samples[cpu->sample_ptr].tsc = tsc;
+- cpu->samples[cpu->sample_ptr].aperf -= cpu->prev_aperf;
+- cpu->samples[cpu->sample_ptr].mperf -= cpu->prev_mperf;
+- cpu->samples[cpu->sample_ptr].tsc -= cpu->prev_tsc;
++ cpu->sample.aperf = aperf;
++ cpu->sample.mperf = mperf;
++ cpu->sample.tsc = tsc;
++ cpu->sample.aperf -= cpu->prev_aperf;
++ cpu->sample.mperf -= cpu->prev_mperf;
++ cpu->sample.tsc -= cpu->prev_tsc;
+
+- intel_pstate_calc_busy(cpu, &cpu->samples[cpu->sample_ptr]);
++ intel_pstate_calc_busy(cpu, &cpu->sample);
+
+ cpu->prev_aperf = aperf;
+ cpu->prev_mperf = mperf;
+@@ -618,7 +616,7 @@ static inline int32_t intel_pstate_get_s
+ {
+ int32_t core_busy, max_pstate, current_pstate;
+
+- core_busy = cpu->samples[cpu->sample_ptr].core_pct_busy;
++ core_busy = cpu->sample.core_pct_busy;
+ max_pstate = int_tofp(cpu->pstate.max_pstate);
+ current_pstate = int_tofp(cpu->pstate.current_pstate);
+ core_busy = mul_fp(core_busy, div_fp(max_pstate, current_pstate));
+@@ -652,7 +650,7 @@ static void intel_pstate_timer_func(unsi
+
+ intel_pstate_sample(cpu);
+
+- sample = &cpu->samples[cpu->sample_ptr];
++ sample = &cpu->sample;
+
+ intel_pstate_adjust_busy_pstate(cpu);
+
+@@ -727,7 +725,7 @@ static unsigned int intel_pstate_get(uns
+ cpu = all_cpu_data[cpu_num];
+ if (!cpu)
+ return 0;
+- sample = &cpu->samples[cpu->sample_ptr];
++ sample = &cpu->sample;
+ return sample->freq;
+ }
+
--- /dev/null
+From 0c36b390a546055b6815d4b93a2c9fed4d980ffb Mon Sep 17 00:00:00 2001
+From: Sebastian Ott <sebott@linux.vnet.ibm.com>
+Date: Wed, 4 Jun 2014 15:58:24 +0200
+Subject: percpu-refcount: fix usage of this_cpu_ops
+
+From: Sebastian Ott <sebott@linux.vnet.ibm.com>
+
+commit 0c36b390a546055b6815d4b93a2c9fed4d980ffb upstream.
+
+The percpu-refcount infrastructure uses the underscore variants of
+this_cpu_ops in order to modify percpu reference counters.
+(e.g. __this_cpu_inc()).
+
+However the underscore variants do not atomically update the percpu
+variable, instead they may be implemented using read-modify-write
+semantics (more than one instruction). Therefore it is only safe to
+use the underscore variant if the context is always the same (process,
+softirq, or hardirq). Otherwise it is possible to lose updates.
+
+This problem is something that Sebastian has seen within the aio
+subsystem which uses percpu refcounters both in process and softirq
+context leading to reference counts that never dropped to zeroes; even
+though the number of "get" and "put" calls matched.
+
+Fix this by using the non-underscore this_cpu_ops variant which
+provides correct per cpu atomic semantics and fixes the corrupted
+reference counts.
+
+Cc: Kent Overstreet <kmo@daterainc.com>
+Reported-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
+Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
+Signed-off-by: Tejun Heo <tj@kernel.org>
+References: http://lkml.kernel.org/g/alpine.LFD.2.11.1406041540520.21183@denkbrett
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ include/linux/percpu-refcount.h | 6 +++---
+ 1 file changed, 3 insertions(+), 3 deletions(-)
+
+--- a/include/linux/percpu-refcount.h
++++ b/include/linux/percpu-refcount.h
+@@ -110,7 +110,7 @@ static inline void percpu_ref_get(struct
+ pcpu_count = ACCESS_ONCE(ref->pcpu_count);
+
+ if (likely(REF_STATUS(pcpu_count) == PCPU_REF_PTR))
+- __this_cpu_inc(*pcpu_count);
++ this_cpu_inc(*pcpu_count);
+ else
+ atomic_inc(&ref->count);
+
+@@ -139,7 +139,7 @@ static inline bool percpu_ref_tryget(str
+ pcpu_count = ACCESS_ONCE(ref->pcpu_count);
+
+ if (likely(REF_STATUS(pcpu_count) == PCPU_REF_PTR)) {
+- __this_cpu_inc(*pcpu_count);
++ this_cpu_inc(*pcpu_count);
+ ret = true;
+ }
+
+@@ -164,7 +164,7 @@ static inline void percpu_ref_put(struct
+ pcpu_count = ACCESS_ONCE(ref->pcpu_count);
+
+ if (likely(REF_STATUS(pcpu_count) == PCPU_REF_PTR))
+- __this_cpu_dec(*pcpu_count);
++ this_cpu_dec(*pcpu_count);
+ else if (unlikely(atomic_dec_and_test(&ref->count)))
+ ref->release(ref);
+
usb-io_ti-fix-firmware-download-on-big-endian-machines-part-2.patch
usb-pci-quirks-prevent-sony-vaio-t-series-from-switching-usb-ports.patch
usb-avoid-runtime-suspend-loops-for-hcds-that-can-t-handle-suspend-resume.patch
+percpu-refcount-fix-usage-of-this_cpu_ops.patch
+intel_pstate-remove-unneeded-sample-buffers.patch
+intel_pstate-remove-c0-tracking.patch
+intel_pstate-correct-rounding-in-busy-calculation.patch
+intel_pstate-add-sample-time-scaling.patch
+intel_pstate-improve-initial-busy-calculation.patch
projects / thirdparty / kernel / stable-queue.git / commitdiff
