Re: [PATCH 26/32] perf/x86/intel/cqm: integrate CQM cgroups with scheduler

From: Vikas Shivappa
Date: Fri Apr 29 2016 - 17:01:47 EST




On Fri, 29 Apr 2016, David Carrillo-Cisneros wrote:

Not sure I see the problem you point here. In step 3, PQR_ASSOC is
updated with RMID1, __pqr_update is the one called using the scheduler
hook, right after perf sched_in .

On Fri, Apr 29, 2016 at 1:25 PM, Vikas Shivappa
<vikas.shivappa@xxxxxxxxxxxxxxx> wrote:


On Thu, 28 Apr 2016, David Carrillo-Cisneros wrote:

Allow monitored cgroups to update the PQR MSR during task switch even
without an associated perf_event.

The package RMID for the current monr associated with a monitored
cgroup is written to hw during task switch (after perf_events is run)
if perf_event did not write a RMID for an event.

perf_event and any other caller of pqr_cache_update_rmid can update the
CPU's RMID using one of two modes:
- PQR_RMID_MODE_NOEVENT: A RMID that do not correspond to an event.
e.g. the RMID of the root pmonr when no event is scheduled.
- PQR_RMID_MODE_EVENT: A RMID used by an event. Set during pmu::add
unset on pmu::del. This mode prevents from using a non-event
cgroup RMID.

This patch also introduces caching of writes to PQR MSR within the per-pcu
pqr state variable. This interface to update RMIDs and CLOSIDs will be
also utilized in upcoming versions of Intel's MBM and CAT drivers.

Reviewed-by: Stephane Eranian <eranian@xxxxxxxxxx>
Signed-off-by: David Carrillo-Cisneros <davidcc@xxxxxxxxxx>
---
arch/x86/events/intel/cqm.c | 65
+++++++++++++++++++++++++++++----------
arch/x86/events/intel/cqm.h | 2 --
arch/x86/include/asm/pqr_common.h | 53 +++++++++++++++++++++++++++----
arch/x86/kernel/cpu/pqr_common.c | 46 +++++++++++++++++++++++----
4 files changed, 135 insertions(+), 31 deletions(-)

diff --git a/arch/x86/events/intel/cqm.c b/arch/x86/events/intel/cqm.c
index daf9fdf..4ece0a4 100644
--- a/arch/x86/events/intel/cqm.c
+++ b/arch/x86/events/intel/cqm.c
@@ -198,19 +198,6 @@ static inline int cqm_prmid_update(struct prmid
*prmid)
return __cqm_prmid_update(prmid, __rmid_min_update_time);
}

-/*
- * Updates caller cpu's cache.
- */
-static inline void __update_pqr_prmid(struct prmid *prmid)
-{
- struct intel_pqr_state *state = this_cpu_ptr(&pqr_state);
-
- if (state->rmid == prmid->rmid)
- return;
- state->rmid = prmid->rmid;
- wrmsr(MSR_IA32_PQR_ASSOC, prmid->rmid, state->closid);
-}
-
static inline bool __valid_pkg_id(u16 pkg_id)
{
return pkg_id < PQR_MAX_NR_PKGS;
@@ -2531,12 +2518,11 @@ static inline bool cqm_group_leader(struct
perf_event *event)
static inline void __intel_cqm_event_start(
struct perf_event *event, union prmid_summary summary)
{
- u16 pkg_id = topology_physical_package_id(smp_processor_id());
if (!(event->hw.state & PERF_HES_STOPPED))
return;
-
event->hw.state &= ~PERF_HES_STOPPED;
- __update_pqr_prmid(__prmid_from_rmid(pkg_id, summary.sched_rmid));
+
+ pqr_cache_update_rmid(summary.sched_rmid, PQR_RMID_MODE_EVENT);
}

static void intel_cqm_event_start(struct perf_event *event, int mode)
@@ -2566,7 +2552,7 @@ static void intel_cqm_event_stop(struct perf_event
*event, int mode)
/* Occupancy of CQM events is obtained at read. No need to read
* when event is stopped since read on inactive cpus succeed.
*/
- __update_pqr_prmid(__prmid_from_rmid(pkg_id, summary.sched_rmid));
+ pqr_cache_update_rmid(summary.sched_rmid, PQR_RMID_MODE_NOEVENT);
}

static int intel_cqm_event_add(struct perf_event *event, int mode)
@@ -2977,6 +2963,8 @@ static void intel_cqm_cpu_starting(unsigned int cpu)

state->rmid = 0;
state->closid = 0;
+ state->next_rmid = 0;
+ state->next_closid = 0;

/* XXX: lock */
/* XXX: Make sure this case is handled when hotplug happens. */
@@ -3152,6 +3140,12 @@ static int __init intel_cqm_init(void)
pr_info("Intel CQM monitoring enabled with at least %u rmids per
package.\n",
min_max_rmid + 1);

+ /* Make sure pqr_common_enable_key is enabled after
+ * cqm_initialized_key.
+ */
+ barrier();
+
+ static_branch_enable(&pqr_common_enable_key);
return ret;

error_init_mutex:
@@ -3163,4 +3157,41 @@ error:
return ret;
}

+/* Schedule task without a CQM perf_event. */
+inline void __intel_cqm_no_event_sched_in(void)
+{
+#ifdef CONFIG_CGROUP_PERF
+ struct monr *monr;
+ struct pmonr *pmonr;
+ union prmid_summary summary;
+ u16 pkg_id = topology_physical_package_id(smp_processor_id());
+ struct pmonr *root_pmonr = monr_hrchy_root->pmonrs[pkg_id];
+
+ /* Assume CQM enabled is likely given that PQR is enabled. */
+ if (!static_branch_likely(&cqm_initialized_key))
+ return;
+
+ /* Safe to call from_task since we are in scheduler lock. */
+ monr = monr_from_perf_cgroup(perf_cgroup_from_task(current,
NULL));
+ pmonr = monr->pmonrs[pkg_id];
+
+ /* Utilize most up to date pmonr summary. */
+ monr_hrchy_get_next_prmid_summary(pmonr);
+ summary.value = atomic64_read(&pmonr->prmid_summary_atomic);
+
+ if (!prmid_summary__is_mon_active(summary))
+ goto no_rmid;
+
+ if (WARN_ON_ONCE(!__valid_rmid(pkg_id, summary.sched_rmid)))
+ goto no_rmid;
+
+ pqr_cache_update_rmid(summary.sched_rmid, PQR_RMID_MODE_NOEVENT);
+ return;
+
+no_rmid:
+ summary.value = atomic64_read(&root_pmonr->prmid_summary_atomic);
+ pqr_cache_update_rmid(summary.sched_rmid, PQR_RMID_MODE_NOEVENT);
+#endif
+}
+
device_initcall(intel_cqm_init);
diff --git a/arch/x86/events/intel/cqm.h b/arch/x86/events/intel/cqm.h
index 0f3da94..e1f8bd0 100644
--- a/arch/x86/events/intel/cqm.h
+++ b/arch/x86/events/intel/cqm.h
@@ -82,8 +82,6 @@ union prmid_summary {
};
};

-# define INVALID_RMID (-1)
-
/* A pmonr in (U)state has no sched_rmid, read_rmid can be 0 or
INVALID_RMID
* depending on whether monitoring is active or not.
*/
diff --git a/arch/x86/include/asm/pqr_common.h
b/arch/x86/include/asm/pqr_common.h
index f770637..abbb235 100644
--- a/arch/x86/include/asm/pqr_common.h
+++ b/arch/x86/include/asm/pqr_common.h
@@ -3,31 +3,72 @@

#if defined(CONFIG_INTEL_RDT)

+#include <linux/jump_label.h>
#include <linux/types.h>
#include <asm/percpu.h>
+#include <asm/msr.h>

#define MSR_IA32_PQR_ASSOC 0x0c8f
+#define INVALID_RMID (-1)
+#define INVALID_CLOSID (-1)
+
+
+extern struct static_key_false pqr_common_enable_key;
+
+enum intel_pqr_rmid_mode {
+ /* RMID has no perf_event associated. */
+ PQR_RMID_MODE_NOEVENT = 0,
+ /* RMID has a perf_event associated. */
+ PQR_RMID_MODE_EVENT
+};

/**
* struct intel_pqr_state - State cache for the PQR MSR
- * @rmid: The cached Resource Monitoring ID
- * @closid: The cached Class Of Service ID
+ * @rmid: Last rmid written to hw.
+ * @next_rmid: Next rmid to write to hw.
+ * @next_rmid_mode: Next rmid's mode.
+ * @closid: The current Class Of Service ID
+ * @next_closid: The Class Of Service ID to use.
*
* The upper 32 bits of MSR_IA32_PQR_ASSOC contain closid and the
* lower 10 bits rmid. The update to MSR_IA32_PQR_ASSOC always
* contains both parts, so we need to cache them.
*
- * The cache also helps to avoid pointless updates if the value does
- * not change.
+ * The cache also helps to avoid pointless updates if the value does not
+ * change. It also keeps track of the type of RMID set (event vs no
event)
+ * used to determine when a cgroup RMID is required.
*/
struct intel_pqr_state {
- u32 rmid;
- u32 closid;
+ u32 rmid;
+ u32 next_rmid;
+ enum intel_pqr_rmid_mode next_rmid_mode;
+ u32 closid;
+ u32 next_closid;
};

DECLARE_PER_CPU(struct intel_pqr_state, pqr_state);

#define PQR_MAX_NR_PKGS 8

+void __pqr_update(void);
+
+inline void __intel_cqm_no_event_sched_in(void);
+
+inline void pqr_cache_update_rmid(u32 rmid, enum intel_pqr_rmid_mode
mode);
+
+inline void pqr_cache_update_closid(u32 closid);
+
+static inline void pqr_update(void)
+{
+ if (static_branch_unlikely(&pqr_common_enable_key))
+ __pqr_update();
+}
+
+#else
+
+static inline void pqr_update(void)
+{
+}
+
#endif
#endif
diff --git a/arch/x86/kernel/cpu/pqr_common.c
b/arch/x86/kernel/cpu/pqr_common.c
index 9eff5d9..d91c127 100644
--- a/arch/x86/kernel/cpu/pqr_common.c
+++ b/arch/x86/kernel/cpu/pqr_common.c
@@ -1,9 +1,43 @@
#include <asm/pqr_common.h>

-/*
- * The cached intel_pqr_state is strictly per CPU and can never be
- * updated from a remote CPU. Both functions which modify the state
- * (intel_cqm_event_start and intel_cqm_event_stop) are called with
- * interrupts disabled, which is sufficient for the protection.
- */
DEFINE_PER_CPU(struct intel_pqr_state, pqr_state);
+
+DEFINE_STATIC_KEY_FALSE(pqr_common_enable_key);
+
+inline void pqr_cache_update_rmid(u32 rmid, enum intel_pqr_rmid_mode
mode)
+{
+ struct intel_pqr_state *state = this_cpu_ptr(&pqr_state);
+
+ state->next_rmid_mode = mode;
+ state->next_rmid = rmid;
+}
+
+inline void pqr_cache_update_closid(u32 closid)
+{
+ struct intel_pqr_state *state = this_cpu_ptr(&pqr_state);
+
+ state->next_closid = closid;
+}
+
+/* Update hw's RMID using cgroup's if perf_event did not.
+ * Sync pqr cache with MSR.
+ */
+inline void __pqr_update(void)
+{
+ struct intel_pqr_state *state = this_cpu_ptr(&pqr_state);
+
+ /* If perf_event has set a next_rmid that is used, do not try
+ * to obtain another one from current task.
+ */
+ if (state->next_rmid_mode == PQR_RMID_MODE_NOEVENT)
+ __intel_cqm_no_event_sched_in();


if perf_cgroup is not defined then the state is not updated here so the
state might have state->rmid might have IDs ?

1.event1 for PID1 has RMID1,
2.perf sched_in - state->next_rmid = rmid1
3.pqr_update - state->rmid = rmid1
4.sched_out - write PQR_NOEVENT -
5.next switch_to - state->rmid not reset nothing changes (when no
perf_cgroup) ?

Between 4 and 5 say event1 is dead. Basically on the next context switch(#5) if perf sched_in wasnt called the PQR still has RMID1.

When you have perf_cgroup you call the __intel_cqm_no_event.. then it sees if there is continuous monitoring or you set the next_rmid to 0- but all the code there is inside #ifdef PERF_CGROUP, so without cgroup its never reset to zero ?




+
+ /* __intel_cqm_no_event_sched_in might have changed next_rmid. */
+ if (state->rmid == state->next_rmid &&
+ state->closid == state->next_closid)
+ return;
+
+ state->rmid = state->next_rmid;
+ state->closid = state->next_closid;
+ wrmsr(MSR_IA32_PQR_ASSOC, state->rmid, state->closid);
+}
--
2.8.0.rc3.226.g39d4020