* [PATCH v2 1/9] powerpc/pseries: Use macros for referring to the DTL enable mask
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
@ 2019-06-14 15:43 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 2/9] powerpc/pseries: Do not save the previous DTL mask value Naveen N. Rao
` (7 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:43 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
Introduce macros to encode the DTL enable mask fields and use those
instead of hardcoding numbers.
Acked-by: Nathan Lynch <nathanl@linux.ibm.com>
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/lppaca.h | 11 +++++++++++
arch/powerpc/platforms/pseries/dtl.c | 8 +-------
arch/powerpc/platforms/pseries/lpar.c | 2 +-
arch/powerpc/platforms/pseries/setup.c | 2 +-
4 files changed, 14 insertions(+), 9 deletions(-)
diff --git a/arch/powerpc/include/asm/lppaca.h b/arch/powerpc/include/asm/lppaca.h
index 806494283e2a..9837bb2b3e69 100644
--- a/arch/powerpc/include/asm/lppaca.h
+++ b/arch/powerpc/include/asm/lppaca.h
@@ -141,6 +141,17 @@ struct dtl_entry {
#define DISPATCH_LOG_BYTES 4096 /* bytes per cpu */
#define N_DISPATCH_LOG (DISPATCH_LOG_BYTES / sizeof(struct dtl_entry))
+/*
+ * Dispatch trace log event enable mask:
+ * 0x1: voluntary virtual processor waits
+ * 0x2: time-slice preempts
+ * 0x4: virtual partition memory page faults
+ */
+#define DTL_LOG_CEDE 0x1
+#define DTL_LOG_PREEMPT 0x2
+#define DTL_LOG_FAULT 0x4
+#define DTL_LOG_ALL (DTL_LOG_CEDE | DTL_LOG_PREEMPT | DTL_LOG_FAULT)
+
extern struct kmem_cache *dtl_cache;
/*
diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c
index ab5de985a787..337378f3d1fd 100644
--- a/arch/powerpc/platforms/pseries/dtl.c
+++ b/arch/powerpc/platforms/pseries/dtl.c
@@ -27,13 +27,7 @@ struct dtl {
};
static DEFINE_PER_CPU(struct dtl, cpu_dtl);
-/*
- * Dispatch trace log event mask:
- * 0x7: 0x1: voluntary virtual processor waits
- * 0x2: time-slice preempts
- * 0x4: virtual partition memory page faults
- */
-static u8 dtl_event_mask = 0x7;
+static u8 dtl_event_mask = DTL_LOG_ALL;
/*
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 73620dfb63a1..96c4677cf8ab 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -113,7 +113,7 @@ void vpa_init(int cpu)
pr_err("WARNING: DTL registration of cpu %d (hw %d) "
"failed with %ld\n", smp_processor_id(),
hwcpu, ret);
- lppaca_of(cpu).dtl_enable_mask = 2;
+ lppaca_of(cpu).dtl_enable_mask = DTL_LOG_PREEMPT;
}
}
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 8fa012a65a71..a643e45d0544 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -312,7 +312,7 @@ static int alloc_dispatch_logs(void)
pr_err("WARNING: DTL registration of cpu %d (hw %d) failed "
"with %d\n", smp_processor_id(),
hard_smp_processor_id(), ret);
- get_paca()->lppaca_ptr->dtl_enable_mask = 2;
+ get_paca()->lppaca_ptr->dtl_enable_mask = DTL_LOG_PREEMPT;
return 0;
}
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 2/9] powerpc/pseries: Do not save the previous DTL mask value
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
2019-06-14 15:43 ` [PATCH v2 1/9] powerpc/pseries: Use macros for referring to the DTL enable mask Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 3/9] powerpc/pseries: Factor out DTL buffer allocation and registration routines Naveen N. Rao
` (6 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
When CONFIG_VIRT_CPU_ACCOUNTING_NATIVE is enabled, we always initialize
DTL enable mask to DTL_LOG_PREEMPT (0x2). There are no other places
where the mask is changed. As such, when reading the DTL log buffer
through debugfs, there is no need to save and restore the previous mask
value.
We don't need to save and restore the earlier mask value if
CONFIG_VIRT_CPU_ACCOUNTING_NATIVE is not enabled. So, remove the field
from the structure as well.
Acked-by: Nathan Lynch <nathanl@linux.ibm.com>
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/platforms/pseries/dtl.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c
index 337378f3d1fd..a88176afba32 100644
--- a/arch/powerpc/platforms/pseries/dtl.c
+++ b/arch/powerpc/platforms/pseries/dtl.c
@@ -42,7 +42,6 @@ struct dtl_ring {
struct dtl_entry *write_ptr;
struct dtl_entry *buf;
struct dtl_entry *buf_end;
- u8 saved_dtl_mask;
};
static DEFINE_PER_CPU(struct dtl_ring, dtl_rings);
@@ -92,7 +91,6 @@ static int dtl_start(struct dtl *dtl)
dtlr->write_ptr = dtl->buf;
/* enable event logging */
- dtlr->saved_dtl_mask = lppaca_of(dtl->cpu).dtl_enable_mask;
lppaca_of(dtl->cpu).dtl_enable_mask |= dtl_event_mask;
dtl_consumer = consume_dtle;
@@ -110,7 +108,7 @@ static void dtl_stop(struct dtl *dtl)
dtlr->buf = NULL;
/* restore dtl_enable_mask */
- lppaca_of(dtl->cpu).dtl_enable_mask = dtlr->saved_dtl_mask;
+ lppaca_of(dtl->cpu).dtl_enable_mask = DTL_LOG_PREEMPT;
if (atomic_dec_and_test(&dtl_count))
dtl_consumer = NULL;
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 3/9] powerpc/pseries: Factor out DTL buffer allocation and registration routines
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
2019-06-14 15:43 ` [PATCH v2 1/9] powerpc/pseries: Use macros for referring to the DTL enable mask Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 2/9] powerpc/pseries: Do not save the previous DTL mask value Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 4/9] powerpc/pseries: Introduce helpers to gatekeep DTLB usage Naveen N. Rao
` (5 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
Introduce new helpers for DTL buffer allocation and registration and
have the existing code use those.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/plpar_wrappers.h | 2 +
arch/powerpc/platforms/pseries/lpar.c | 66 ++++++++++++++++-------
arch/powerpc/platforms/pseries/setup.c | 34 +-----------
3 files changed, 52 insertions(+), 50 deletions(-)
diff --git a/arch/powerpc/include/asm/plpar_wrappers.h b/arch/powerpc/include/asm/plpar_wrappers.h
index cff5a411e595..d08feb1bc2bd 100644
--- a/arch/powerpc/include/asm/plpar_wrappers.h
+++ b/arch/powerpc/include/asm/plpar_wrappers.h
@@ -88,6 +88,8 @@ static inline long register_dtl(unsigned long cpu, unsigned long vpa)
return vpa_call(H_VPA_REG_DTL, cpu, vpa);
}
+extern void register_dtl_buffer(int cpu);
+extern void alloc_dtl_buffers(void);
extern void vpa_init(int cpu);
static inline long plpar_pte_enter(unsigned long flags,
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 96c4677cf8ab..8c5377fe9985 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -52,13 +52,58 @@ EXPORT_SYMBOL(plpar_hcall);
EXPORT_SYMBOL(plpar_hcall9);
EXPORT_SYMBOL(plpar_hcall_norets);
+void alloc_dtl_buffers(void)
+{
+ int cpu;
+ struct paca_struct *pp;
+ struct dtl_entry *dtl;
+
+ for_each_possible_cpu(cpu) {
+ pp = paca_ptrs[cpu];
+ dtl = kmem_cache_alloc(dtl_cache, GFP_KERNEL);
+ if (!dtl) {
+ pr_warn("Failed to allocate dispatch trace log for cpu %d\n",
+ cpu);
+ pr_warn("Stolen time statistics will be unreliable\n");
+ break;
+ }
+
+ pp->dtl_ridx = 0;
+ pp->dispatch_log = dtl;
+ pp->dispatch_log_end = dtl + N_DISPATCH_LOG;
+ pp->dtl_curr = dtl;
+ }
+}
+
+void register_dtl_buffer(int cpu)
+{
+ long ret;
+ struct paca_struct *pp;
+ struct dtl_entry *dtl;
+ int hwcpu = get_hard_smp_processor_id(cpu);
+
+ pp = paca_ptrs[cpu];
+ dtl = pp->dispatch_log;
+ if (dtl) {
+ pp->dtl_ridx = 0;
+ pp->dtl_curr = dtl;
+ lppaca_of(cpu).dtl_idx = 0;
+
+ /* hypervisor reads buffer length from this field */
+ dtl->enqueue_to_dispatch_time = cpu_to_be32(DISPATCH_LOG_BYTES);
+ ret = register_dtl(hwcpu, __pa(dtl));
+ if (ret)
+ pr_err("WARNING: DTL registration of cpu %d (hw %d) "
+ "failed with %ld\n", cpu, hwcpu, ret);
+ lppaca_of(cpu).dtl_enable_mask = DTL_LOG_PREEMPT;
+ }
+}
+
void vpa_init(int cpu)
{
int hwcpu = get_hard_smp_processor_id(cpu);
unsigned long addr;
long ret;
- struct paca_struct *pp;
- struct dtl_entry *dtl;
/*
* The spec says it "may be problematic" if CPU x registers the VPA of
@@ -99,22 +144,7 @@ void vpa_init(int cpu)
/*
* Register dispatch trace log, if one has been allocated.
*/
- pp = paca_ptrs[cpu];
- dtl = pp->dispatch_log;
- if (dtl) {
- pp->dtl_ridx = 0;
- pp->dtl_curr = dtl;
- lppaca_of(cpu).dtl_idx = 0;
-
- /* hypervisor reads buffer length from this field */
- dtl->enqueue_to_dispatch_time = cpu_to_be32(DISPATCH_LOG_BYTES);
- ret = register_dtl(hwcpu, __pa(dtl));
- if (ret)
- pr_err("WARNING: DTL registration of cpu %d (hw %d) "
- "failed with %ld\n", smp_processor_id(),
- hwcpu, ret);
- lppaca_of(cpu).dtl_enable_mask = DTL_LOG_PREEMPT;
- }
+ register_dtl_buffer(cpu);
}
#ifdef CONFIG_PPC_BOOK3S_64
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index a643e45d0544..e4d75b958593 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -273,46 +273,16 @@ struct kmem_cache *dtl_cache;
*/
static int alloc_dispatch_logs(void)
{
- int cpu, ret;
- struct paca_struct *pp;
- struct dtl_entry *dtl;
-
if (!firmware_has_feature(FW_FEATURE_SPLPAR))
return 0;
if (!dtl_cache)
return 0;
- for_each_possible_cpu(cpu) {
- pp = paca_ptrs[cpu];
- dtl = kmem_cache_alloc(dtl_cache, GFP_KERNEL);
- if (!dtl) {
- pr_warn("Failed to allocate dispatch trace log for cpu %d\n",
- cpu);
- pr_warn("Stolen time statistics will be unreliable\n");
- break;
- }
-
- pp->dtl_ridx = 0;
- pp->dispatch_log = dtl;
- pp->dispatch_log_end = dtl + N_DISPATCH_LOG;
- pp->dtl_curr = dtl;
- }
+ alloc_dtl_buffers();
/* Register the DTL for the current (boot) cpu */
- dtl = get_paca()->dispatch_log;
- get_paca()->dtl_ridx = 0;
- get_paca()->dtl_curr = dtl;
- get_paca()->lppaca_ptr->dtl_idx = 0;
-
- /* hypervisor reads buffer length from this field */
- dtl->enqueue_to_dispatch_time = cpu_to_be32(DISPATCH_LOG_BYTES);
- ret = register_dtl(hard_smp_processor_id(), __pa(dtl));
- if (ret)
- pr_err("WARNING: DTL registration of cpu %d (hw %d) failed "
- "with %d\n", smp_processor_id(),
- hard_smp_processor_id(), ret);
- get_paca()->lppaca_ptr->dtl_enable_mask = DTL_LOG_PREEMPT;
+ register_dtl_buffer(smp_processor_id());
return 0;
}
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 4/9] powerpc/pseries: Introduce helpers to gatekeep DTLB usage
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (2 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 3/9] powerpc/pseries: Factor out DTL buffer allocation and registration routines Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 5/9] powerpc/pseries: Generalize hcall_vphn() Naveen N. Rao
` (4 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
Since we would be introducing a new user of the DTL buffer in a
subsequent patch, add helpers to gatekeep use of the DTL buffer. The
current usage of the DTL buffer from debugfs is at a per-cpu level
(corresponding to the cpu debugfs file that is opened). Subsequently, we
will have users enabling/accessing DTLB for all online cpus. These
helpers allow any number of per-cpu users, or a single global user
exclusively.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/plpar_wrappers.h | 2 ++
arch/powerpc/platforms/pseries/dtl.c | 10 +++++-
arch/powerpc/platforms/pseries/lpar.c | 38 +++++++++++++++++++++++
3 files changed, 49 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/include/asm/plpar_wrappers.h b/arch/powerpc/include/asm/plpar_wrappers.h
index d08feb1bc2bd..ab7dd454b6eb 100644
--- a/arch/powerpc/include/asm/plpar_wrappers.h
+++ b/arch/powerpc/include/asm/plpar_wrappers.h
@@ -88,6 +88,8 @@ static inline long register_dtl(unsigned long cpu, unsigned long vpa)
return vpa_call(H_VPA_REG_DTL, cpu, vpa);
}
+extern bool register_dtl_buffer_access(bool global);
+extern void unregister_dtl_buffer_access(bool global);
extern void register_dtl_buffer(int cpu);
extern void alloc_dtl_buffers(void);
extern void vpa_init(int cpu);
diff --git a/arch/powerpc/platforms/pseries/dtl.c b/arch/powerpc/platforms/pseries/dtl.c
index a88176afba32..de531311cd9a 100644
--- a/arch/powerpc/platforms/pseries/dtl.c
+++ b/arch/powerpc/platforms/pseries/dtl.c
@@ -180,11 +180,15 @@ static int dtl_enable(struct dtl *dtl)
if (dtl->buf)
return -EBUSY;
+ if (register_dtl_buffer_access(false))
+ return -EBUSY;
+
n_entries = dtl_buf_entries;
buf = kmem_cache_alloc_node(dtl_cache, GFP_KERNEL, cpu_to_node(dtl->cpu));
if (!buf) {
printk(KERN_WARNING "%s: buffer alloc failed for cpu %d\n",
__func__, dtl->cpu);
+ unregister_dtl_buffer_access(false);
return -ENOMEM;
}
@@ -201,8 +205,11 @@ static int dtl_enable(struct dtl *dtl)
}
spin_unlock(&dtl->lock);
- if (rc)
+ if (rc) {
+ unregister_dtl_buffer_access(false);
kmem_cache_free(dtl_cache, buf);
+ }
+
return rc;
}
@@ -214,6 +221,7 @@ static void dtl_disable(struct dtl *dtl)
dtl->buf = NULL;
dtl->buf_entries = 0;
spin_unlock(&dtl->lock);
+ unregister_dtl_buffer_access(false);
}
/* file interface */
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 8c5377fe9985..53e005c84078 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -99,6 +99,44 @@ void register_dtl_buffer(int cpu)
}
}
+#ifdef CONFIG_PPC_SPLPAR
+static DEFINE_SPINLOCK(dtl_buffer_refctr_lock);
+static unsigned int dtl_buffer_global_refctr, dtl_buffer_percpu_refctr;
+
+bool register_dtl_buffer_access(bool global)
+{
+ int rc = 0;
+
+ spin_lock(&dtl_buffer_refctr_lock);
+
+ if ((global && (dtl_buffer_global_refctr || dtl_buffer_percpu_refctr))
+ || (!global && dtl_buffer_global_refctr)) {
+ rc = -1;
+ } else {
+ if (global)
+ dtl_buffer_global_refctr++;
+ else
+ dtl_buffer_percpu_refctr++;
+ }
+
+ spin_unlock(&dtl_buffer_refctr_lock);
+
+ return rc;
+}
+
+void unregister_dtl_buffer_access(bool global)
+{
+ spin_lock(&dtl_buffer_refctr_lock);
+
+ if (global)
+ dtl_buffer_global_refctr--;
+ else
+ dtl_buffer_percpu_refctr--;
+
+ spin_unlock(&dtl_buffer_refctr_lock);
+}
+#endif /* CONFIG_PPC_SPLPAR */
+
void vpa_init(int cpu)
{
int hwcpu = get_hard_smp_processor_id(cpu);
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 5/9] powerpc/pseries: Generalize hcall_vphn()
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (3 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 4/9] powerpc/pseries: Introduce helpers to gatekeep DTLB usage Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 6/9] powerpc/pseries: Move mm/book3s64/vphn.c under platforms/pseries/ Naveen N. Rao
` (3 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
H_HOME_NODE_ASSOCIATIVITY hcall can take two different flags and return
different associativity information in each case. Generalize the
existing hcall_vphn() function to take flags as an argument and to
return the result. Update the only existing user to pass the proper
arguments.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/mm/book3s64/vphn.h | 8 ++++++++
arch/powerpc/mm/numa.c | 27 +++++++++++++--------------
2 files changed, 21 insertions(+), 14 deletions(-)
diff --git a/arch/powerpc/mm/book3s64/vphn.h b/arch/powerpc/mm/book3s64/vphn.h
index f0b93c2dd578..f7ff1e0c3801 100644
--- a/arch/powerpc/mm/book3s64/vphn.h
+++ b/arch/powerpc/mm/book3s64/vphn.h
@@ -11,6 +11,14 @@
*/
#define VPHN_ASSOC_BUFSIZE (VPHN_REGISTER_COUNT*sizeof(u64)/sizeof(u16) + 1)
+/*
+ * The H_HOME_NODE_ASSOCIATIVITY hcall takes two values for flags:
+ * 1 for retrieving associativity information for a guest cpu
+ * 2 for retrieving associativity information for a host/hypervisor cpu
+ */
+#define VPHN_FLAG_VCPU 1
+#define VPHN_FLAG_PCPU 2
+
extern int vphn_unpack_associativity(const long *packed, __be32 *unpacked);
#endif
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 917904d2fe97..42ea79514420 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -1083,6 +1083,17 @@ static void reset_topology_timer(void);
static int topology_timer_secs = 1;
static int topology_inited;
+static long hcall_vphn(unsigned long cpu, u64 flags, __be32 *associativity)
+{
+ long rc;
+ long retbuf[PLPAR_HCALL9_BUFSIZE] = {0};
+
+ rc = plpar_hcall9(H_HOME_NODE_ASSOCIATIVITY, retbuf, flags, cpu);
+ vphn_unpack_associativity(retbuf, associativity);
+
+ return rc;
+}
+
/*
* Change polling interval for associativity changes.
*/
@@ -1161,25 +1172,13 @@ static int update_cpu_associativity_changes_mask(void)
* Retrieve the new associativity information for a virtual processor's
* home node.
*/
-static long hcall_vphn(unsigned long cpu, __be32 *associativity)
-{
- long rc;
- long retbuf[PLPAR_HCALL9_BUFSIZE] = {0};
- u64 flags = 1;
- int hwcpu = get_hard_smp_processor_id(cpu);
-
- rc = plpar_hcall9(H_HOME_NODE_ASSOCIATIVITY, retbuf, flags, hwcpu);
- vphn_unpack_associativity(retbuf, associativity);
-
- return rc;
-}
-
static long vphn_get_associativity(unsigned long cpu,
__be32 *associativity)
{
long rc;
- rc = hcall_vphn(cpu, associativity);
+ rc = hcall_vphn(get_hard_smp_processor_id(cpu),
+ VPHN_FLAG_VCPU, associativity);
switch (rc) {
case H_FUNCTION:
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 6/9] powerpc/pseries: Move mm/book3s64/vphn.c under platforms/pseries/
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (4 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 5/9] powerpc/pseries: Generalize hcall_vphn() Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 7/9] powerpc/pseries: Provide vcpu dispatch statistics Naveen N. Rao
` (2 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
hcall_vphn() is specific to pseries and will be used in a subsequent
patch. So, move it to a more appropriate place under
arch/powerpc/platforms/pseries. Also merge vphn.h into plpar_wrappers.h
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/plpar_wrappers.h | 19 +++++++++++++++
arch/powerpc/mm/book3s64/Makefile | 1 -
arch/powerpc/mm/book3s64/vphn.h | 24 -------------------
arch/powerpc/mm/numa.c | 13 +---------
arch/powerpc/platforms/pseries/Makefile | 1 +
.../{mm/book3s64 => platforms/pseries}/vphn.c | 16 +++++++++++--
6 files changed, 35 insertions(+), 39 deletions(-)
delete mode 100644 arch/powerpc/mm/book3s64/vphn.h
rename arch/powerpc/{mm/book3s64 => platforms/pseries}/vphn.c (83%)
diff --git a/arch/powerpc/include/asm/plpar_wrappers.h b/arch/powerpc/include/asm/plpar_wrappers.h
index ab7dd454b6eb..5f72f1fa4f81 100644
--- a/arch/powerpc/include/asm/plpar_wrappers.h
+++ b/arch/powerpc/include/asm/plpar_wrappers.h
@@ -11,6 +11,25 @@
#include <asm/paca.h>
#include <asm/page.h>
+/* The H_HOME_NODE_ASSOCIATIVITY h_call returns 6 64-bit registers. */
+#define VPHN_REGISTER_COUNT 6
+
+/*
+ * 6 64-bit registers unpacked into up to 24 be32 associativity values. To
+ * form the complete property we have to add the length in the first cell.
+ */
+#define VPHN_ASSOC_BUFSIZE (VPHN_REGISTER_COUNT*sizeof(u64)/sizeof(u16) + 1)
+
+/*
+ * The H_HOME_NODE_ASSOCIATIVITY hcall takes two values for flags:
+ * 1 for retrieving associativity information for a guest cpu
+ * 2 for retrieving associativity information for a host/hypervisor cpu
+ */
+#define VPHN_FLAG_VCPU 1
+#define VPHN_FLAG_PCPU 2
+
+long hcall_vphn(unsigned long cpu, u64 flags, __be32 *associativity);
+
static inline long poll_pending(void)
{
return plpar_hcall_norets(H_POLL_PENDING);
diff --git a/arch/powerpc/mm/book3s64/Makefile b/arch/powerpc/mm/book3s64/Makefile
index 974b4fc19f4f..fd393b8be14f 100644
--- a/arch/powerpc/mm/book3s64/Makefile
+++ b/arch/powerpc/mm/book3s64/Makefile
@@ -10,7 +10,6 @@ obj-$(CONFIG_PPC_NATIVE) += hash_native.o
obj-$(CONFIG_PPC_RADIX_MMU) += radix_pgtable.o radix_tlb.o
obj-$(CONFIG_PPC_4K_PAGES) += hash_4k.o
obj-$(CONFIG_PPC_64K_PAGES) += hash_64k.o
-obj-$(CONFIG_PPC_SPLPAR) += vphn.o
obj-$(CONFIG_HUGETLB_PAGE) += hash_hugetlbpage.o
ifdef CONFIG_HUGETLB_PAGE
obj-$(CONFIG_PPC_RADIX_MMU) += radix_hugetlbpage.o
diff --git a/arch/powerpc/mm/book3s64/vphn.h b/arch/powerpc/mm/book3s64/vphn.h
deleted file mode 100644
index f7ff1e0c3801..000000000000
--- a/arch/powerpc/mm/book3s64/vphn.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef _ARCH_POWERPC_MM_VPHN_H_
-#define _ARCH_POWERPC_MM_VPHN_H_
-
-/* The H_HOME_NODE_ASSOCIATIVITY h_call returns 6 64-bit registers. */
-#define VPHN_REGISTER_COUNT 6
-
-/*
- * 6 64-bit registers unpacked into up to 24 be32 associativity values. To
- * form the complete property we have to add the length in the first cell.
- */
-#define VPHN_ASSOC_BUFSIZE (VPHN_REGISTER_COUNT*sizeof(u64)/sizeof(u16) + 1)
-
-/*
- * The H_HOME_NODE_ASSOCIATIVITY hcall takes two values for flags:
- * 1 for retrieving associativity information for a guest cpu
- * 2 for retrieving associativity information for a host/hypervisor cpu
- */
-#define VPHN_FLAG_VCPU 1
-#define VPHN_FLAG_PCPU 2
-
-extern int vphn_unpack_associativity(const long *packed, __be32 *unpacked);
-
-#endif
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 42ea79514420..5027c9cfe56a 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -1064,7 +1064,7 @@ u64 memory_hotplug_max(void)
/* Virtual Processor Home Node (VPHN) support */
#ifdef CONFIG_PPC_SPLPAR
-#include "book3s64/vphn.h"
+#include <asm/plpar_wrappers.h>
struct topology_update_data {
struct topology_update_data *next;
@@ -1083,17 +1083,6 @@ static void reset_topology_timer(void);
static int topology_timer_secs = 1;
static int topology_inited;
-static long hcall_vphn(unsigned long cpu, u64 flags, __be32 *associativity)
-{
- long rc;
- long retbuf[PLPAR_HCALL9_BUFSIZE] = {0};
-
- rc = plpar_hcall9(H_HOME_NODE_ASSOCIATIVITY, retbuf, flags, cpu);
- vphn_unpack_associativity(retbuf, associativity);
-
- return rc;
-}
-
/*
* Change polling interval for associativity changes.
*/
diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
index a43ec843c8e2..ab3d59aeacca 100644
--- a/arch/powerpc/platforms/pseries/Makefile
+++ b/arch/powerpc/platforms/pseries/Makefile
@@ -25,6 +25,7 @@ obj-$(CONFIG_LPARCFG) += lparcfg.o
obj-$(CONFIG_IBMVIO) += vio.o
obj-$(CONFIG_IBMEBUS) += ibmebus.o
obj-$(CONFIG_PAPR_SCM) += papr_scm.o
+obj-$(CONFIG_PPC_SPLPAR) += vphn.o
ifdef CONFIG_PPC_PSERIES
obj-$(CONFIG_SUSPEND) += suspend.o
diff --git a/arch/powerpc/mm/book3s64/vphn.c b/arch/powerpc/platforms/pseries/vphn.c
similarity index 83%
rename from arch/powerpc/mm/book3s64/vphn.c
rename to arch/powerpc/platforms/pseries/vphn.c
index 0ee7734afb50..6bbed333e4f2 100644
--- a/arch/powerpc/mm/book3s64/vphn.c
+++ b/arch/powerpc/platforms/pseries/vphn.c
@@ -1,6 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
+#include <asm/plpar_wrappers.h>
#include <asm/byteorder.h>
-#include "vphn.h"
+#include <asm/hvcall.h>
/*
* The associativity domain numbers are returned from the hypervisor as a
@@ -22,7 +23,7 @@
*
* Convert to the sequence they would appear in the ibm,associativity property.
*/
-int vphn_unpack_associativity(const long *packed, __be32 *unpacked)
+static int vphn_unpack_associativity(const long *packed, __be32 *unpacked)
{
__be64 be_packed[VPHN_REGISTER_COUNT];
int i, nr_assoc_doms = 0;
@@ -71,3 +72,14 @@ int vphn_unpack_associativity(const long *packed, __be32 *unpacked)
return nr_assoc_doms;
}
+
+long hcall_vphn(unsigned long cpu, u64 flags, __be32 *associativity)
+{
+ long rc;
+ long retbuf[PLPAR_HCALL9_BUFSIZE] = {0};
+
+ rc = plpar_hcall9(H_HOME_NODE_ASSOCIATIVITY, retbuf, flags, cpu);
+ vphn_unpack_associativity(retbuf, associativity);
+
+ return rc;
+}
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 7/9] powerpc/pseries: Provide vcpu dispatch statistics
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (5 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 6/9] powerpc/pseries: Move mm/book3s64/vphn.c under platforms/pseries/ Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 8/9] powerpc/pseries: Protect against hogging the cpu while setting up the stats Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 9/9] powerpc/pseries: Add documentation for vcpudispatch_stats Naveen N. Rao
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
For Shared Processor LPARs, the POWER Hypervisor maintains a relatively
static mapping of the LPAR processors (vcpus) to physical processor
chips (representing the "home" node) and tries to always dispatch vcpus
on their associated physical processor chip. However, under certain
scenarios, vcpus may be dispatched on a different processor chip (away
from its home node). The actual physical processor number on which a
certain vcpu is dispatched is available to the guest in the
'processor_id' field of each DTL entry.
The guest can discover the home node of each vcpu through the
H_HOME_NODE_ASSOCIATIVITY(flags=1) hcall. The guest can also discover
the associativity of physical processors, as represented in the DTL
entry, through the H_HOME_NODE_ASSOCIATIVITY(flags=2) hcall.
These can then be compared to determine if the vcpu was dispatched on
its home node or not. If the vcpu was not dispatched on the home node,
it is possible to determine if the vcpu was dispatched in a different
chip, socket or drawer.
Introduce a procfs file /proc/powerpc/vcpudispatch_stats that can be
used to obtain these statistics. Writing '1' to this file enables
collecting the statistics, while writing '0' disables the statistics.
The statistics themselves are available by reading the procfs file. By
default, the DTLB log for each vcpu is processed 50 times a second so as
not to miss any entries. This processing frequency can be changed
through /proc/powerpc/vcpudispatch_stats_freq.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/topology.h | 6 +
arch/powerpc/mm/numa.c | 16 +
arch/powerpc/platforms/pseries/lpar.c | 536 +++++++++++++++++++++++++-
3 files changed, 556 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h
index f85e2b01c3df..2f7e1ea5089e 100644
--- a/arch/powerpc/include/asm/topology.h
+++ b/arch/powerpc/include/asm/topology.h
@@ -35,6 +35,7 @@ static inline int pcibus_to_node(struct pci_bus *bus)
cpu_all_mask : \
cpumask_of_node(pcibus_to_node(bus)))
+extern int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc);
extern int __node_distance(int, int);
#define node_distance(a, b) __node_distance(a, b)
@@ -84,6 +85,11 @@ static inline int numa_update_cpu_topology(bool cpus_locked)
static inline void update_numa_cpu_lookup_table(unsigned int cpu, int node) {}
+static inline int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)
+{
+ return 0;
+}
+
#endif /* CONFIG_NUMA */
#if defined(CONFIG_NUMA) && defined(CONFIG_PPC_SPLPAR)
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 5027c9cfe56a..3afa884feead 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -163,6 +163,22 @@ static void unmap_cpu_from_node(unsigned long cpu)
}
#endif /* CONFIG_HOTPLUG_CPU || CONFIG_PPC_SPLPAR */
+int cpu_distance(__be32 *cpu1_assoc, __be32 *cpu2_assoc)
+{
+ int dist = 0;
+
+ int i, index;
+
+ for (i = 0; i < distance_ref_points_depth; i++) {
+ index = be32_to_cpu(distance_ref_points[i]);
+ if (cpu1_assoc[index] == cpu2_assoc[index])
+ break;
+ dist++;
+ }
+
+ return dist;
+}
+
/* must hold reference to node during call */
static const __be32 *of_get_associativity(struct device_node *dev)
{
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 53e005c84078..200a96c8626a 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -17,6 +17,10 @@
#include <linux/jump_label.h>
#include <linux/delay.h>
#include <linux/stop_machine.h>
+#include <linux/spinlock.h>
+#include <linux/cpuhotplug.h>
+#include <linux/workqueue.h>
+#include <linux/proc_fs.h>
#include <asm/processor.h>
#include <asm/mmu.h>
#include <asm/page.h>
@@ -52,6 +56,12 @@ EXPORT_SYMBOL(plpar_hcall);
EXPORT_SYMBOL(plpar_hcall9);
EXPORT_SYMBOL(plpar_hcall_norets);
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+static u8 dtl_mask = DTL_LOG_PREEMPT;
+#else
+static u8 dtl_mask;
+#endif
+
void alloc_dtl_buffers(void)
{
int cpu;
@@ -60,11 +70,15 @@ void alloc_dtl_buffers(void)
for_each_possible_cpu(cpu) {
pp = paca_ptrs[cpu];
+ if (pp->dispatch_log)
+ continue;
dtl = kmem_cache_alloc(dtl_cache, GFP_KERNEL);
if (!dtl) {
pr_warn("Failed to allocate dispatch trace log for cpu %d\n",
cpu);
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
pr_warn("Stolen time statistics will be unreliable\n");
+#endif
break;
}
@@ -84,7 +98,7 @@ void register_dtl_buffer(int cpu)
pp = paca_ptrs[cpu];
dtl = pp->dispatch_log;
- if (dtl) {
+ if (dtl && dtl_mask) {
pp->dtl_ridx = 0;
pp->dtl_curr = dtl;
lppaca_of(cpu).dtl_idx = 0;
@@ -95,13 +109,72 @@ void register_dtl_buffer(int cpu)
if (ret)
pr_err("WARNING: DTL registration of cpu %d (hw %d) "
"failed with %ld\n", cpu, hwcpu, ret);
- lppaca_of(cpu).dtl_enable_mask = DTL_LOG_PREEMPT;
+ lppaca_of(cpu).dtl_enable_mask = dtl_mask;
}
}
+
#ifdef CONFIG_PPC_SPLPAR
+struct dtl_worker {
+ struct delayed_work work;
+ int cpu;
+};
+
+struct vcpu_dispatch_data {
+ int last_disp_cpu;
+
+ int total_disp;
+
+ int same_cpu_disp;
+ int same_chip_disp;
+ int diff_chip_disp;
+ int far_chip_disp;
+
+ int numa_home_disp;
+ int numa_remote_disp;
+ int numa_far_disp;
+};
+
+/*
+ * This represents the number of cpus in the hypervisor. Since there is no
+ * architected way to discover the number of processors in the host, we
+ * provision for dealing with NR_CPUS. This is currently 2048 by default, and
+ * is sufficient for our purposes. This will need to be tweaked if
+ * CONFIG_NR_CPUS is changed.
+ */
+#define NR_CPUS_H NR_CPUS
+
+static DEFINE_PER_CPU(struct vcpu_dispatch_data, vcpu_disp_data);
+static DEFINE_PER_CPU(u64, dtl_entry_ridx);
+static DEFINE_PER_CPU(struct dtl_worker, dtl_workers);
+static enum cpuhp_state dtl_worker_state;
+static DEFINE_MUTEX(dtl_worker_mutex);
+static unsigned int dtl_worker_refctr;
static DEFINE_SPINLOCK(dtl_buffer_refctr_lock);
static unsigned int dtl_buffer_global_refctr, dtl_buffer_percpu_refctr;
+static int vcpudispatch_stats_on __read_mostly;
+static int vcpudispatch_stats_freq = 50;
+static __be32 *vcpu_associativity, *pcpu_associativity;
+
+
+static void free_dtl_buffers(void)
+{
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+ int cpu;
+ struct paca_struct *pp;
+
+ for_each_possible_cpu(cpu) {
+ pp = paca_ptrs[cpu];
+ if (!pp->dispatch_log)
+ continue;
+ kmem_cache_free(dtl_cache, pp->dispatch_log);
+ pp->dtl_ridx = 0;
+ pp->dispatch_log = 0;
+ pp->dispatch_log_end = 0;
+ pp->dtl_curr = 0;
+ }
+#endif
+}
bool register_dtl_buffer_access(bool global)
{
@@ -135,6 +208,465 @@ void unregister_dtl_buffer_access(bool global)
spin_unlock(&dtl_buffer_refctr_lock);
}
+
+static int init_cpu_associativity(void)
+{
+ vcpu_associativity = kcalloc(num_possible_cpus() / threads_per_core,
+ VPHN_ASSOC_BUFSIZE * sizeof(__be32), GFP_KERNEL);
+ pcpu_associativity = kcalloc(NR_CPUS_H / threads_per_core,
+ VPHN_ASSOC_BUFSIZE * sizeof(__be32), GFP_KERNEL);
+
+ if (!vcpu_associativity || !pcpu_associativity) {
+ pr_err("error allocating memory for associativity information\n");
+ return -ENOMEM;
+ }
+
+ return 0;
+}
+
+static void destroy_cpu_associativity(void)
+{
+ kfree(vcpu_associativity);
+ kfree(pcpu_associativity);
+ vcpu_associativity = pcpu_associativity = 0;
+}
+
+static __be32 *__get_cpu_associativity(int cpu, __be32 *cpu_assoc, int flag)
+{
+ __be32 *assoc;
+ int rc = 0;
+
+ assoc = &cpu_assoc[(int)(cpu / threads_per_core) * VPHN_ASSOC_BUFSIZE];
+ if (!assoc[0]) {
+ rc = hcall_vphn(cpu, flag, &assoc[0]);
+ if (rc)
+ return NULL;
+ }
+
+ return assoc;
+}
+
+static __be32 *get_pcpu_associativity(int cpu)
+{
+ return __get_cpu_associativity(cpu, pcpu_associativity, VPHN_FLAG_PCPU);
+}
+
+static __be32 *get_vcpu_associativity(int cpu)
+{
+ return __get_cpu_associativity(cpu, vcpu_associativity, VPHN_FLAG_VCPU);
+}
+
+static int cpu_relative_dispatch_distance(int last_disp_cpu, int cur_disp_cpu)
+{
+ __be32 *last_disp_cpu_assoc, *cur_disp_cpu_assoc;
+
+ if (last_disp_cpu >= NR_CPUS_H || cur_disp_cpu >= NR_CPUS_H)
+ return -EINVAL;
+
+ last_disp_cpu_assoc = get_pcpu_associativity(last_disp_cpu);
+ cur_disp_cpu_assoc = get_pcpu_associativity(cur_disp_cpu);
+
+ if (!last_disp_cpu_assoc || !cur_disp_cpu_assoc)
+ return -EIO;
+
+ return cpu_distance(last_disp_cpu_assoc, cur_disp_cpu_assoc);
+}
+
+static int cpu_home_node_dispatch_distance(int disp_cpu)
+{
+ __be32 *disp_cpu_assoc, *vcpu_assoc;
+ int vcpu_id = smp_processor_id();
+
+ if (disp_cpu >= NR_CPUS_H) {
+ pr_debug_ratelimited("vcpu dispatch cpu %d > %d\n",
+ disp_cpu, NR_CPUS_H);
+ return -EINVAL;
+ }
+
+ disp_cpu_assoc = get_pcpu_associativity(disp_cpu);
+ vcpu_assoc = get_vcpu_associativity(vcpu_id);
+
+ if (!disp_cpu_assoc || !vcpu_assoc)
+ return -EIO;
+
+ return cpu_distance(disp_cpu_assoc, vcpu_assoc);
+}
+
+static void update_vcpu_disp_stat(int disp_cpu)
+{
+ struct vcpu_dispatch_data *disp;
+ int distance;
+
+ disp = this_cpu_ptr(&vcpu_disp_data);
+ if (disp->last_disp_cpu == -1) {
+ disp->last_disp_cpu = disp_cpu;
+ return;
+ }
+
+ disp->total_disp++;
+
+ if (disp->last_disp_cpu == disp_cpu ||
+ (cpu_first_thread_sibling(disp->last_disp_cpu) ==
+ cpu_first_thread_sibling(disp_cpu)))
+ disp->same_cpu_disp++;
+ else {
+ distance = cpu_relative_dispatch_distance(disp->last_disp_cpu,
+ disp_cpu);
+ if (distance < 0)
+ pr_debug_ratelimited("vcpudispatch_stats: cpu %d: error determining associativity\n",
+ smp_processor_id());
+ else {
+ switch (distance) {
+ case 0:
+ disp->same_chip_disp++;
+ break;
+ case 1:
+ disp->diff_chip_disp++;
+ break;
+ case 2:
+ disp->far_chip_disp++;
+ break;
+ default:
+ pr_debug_ratelimited("vcpudispatch_stats: cpu %d (%d -> %d): unexpected relative dispatch distance %d\n",
+ smp_processor_id(),
+ disp->last_disp_cpu,
+ disp_cpu,
+ distance);
+ }
+ }
+ }
+
+ distance = cpu_home_node_dispatch_distance(disp_cpu);
+ if (distance < 0)
+ pr_debug_ratelimited("vcpudispatch_stats: cpu %d: error determining associativity\n",
+ smp_processor_id());
+ else {
+ switch (distance) {
+ case 0:
+ disp->numa_home_disp++;
+ break;
+ case 1:
+ disp->numa_remote_disp++;
+ break;
+ case 2:
+ disp->numa_far_disp++;
+ break;
+ default:
+ pr_debug_ratelimited("vcpudispatch_stats: cpu %d on %d: unexpected numa dispatch distance %d\n",
+ smp_processor_id(),
+ disp_cpu,
+ distance);
+ }
+ }
+
+ disp->last_disp_cpu = disp_cpu;
+}
+
+static void process_dtl_buffer(struct work_struct *work)
+{
+ struct dtl_entry dtle;
+ u64 i = __this_cpu_read(dtl_entry_ridx);
+ struct dtl_entry *dtl = local_paca->dispatch_log + (i % N_DISPATCH_LOG);
+ struct dtl_entry *dtl_end = local_paca->dispatch_log_end;
+ struct lppaca *vpa = local_paca->lppaca_ptr;
+ struct dtl_worker *d = container_of(work, struct dtl_worker, work.work);
+
+ if (!local_paca->dispatch_log)
+ return;
+
+ /* if we have been migrated away, we cancel ourself */
+ if (d->cpu != smp_processor_id()) {
+ pr_debug("vcpudispatch_stats: cpu %d worker migrated -- canceling worker\n",
+ smp_processor_id());
+ return;
+ }
+
+ if (i == be64_to_cpu(vpa->dtl_idx))
+ goto out;
+
+ while (i < be64_to_cpu(vpa->dtl_idx)) {
+ dtle = *dtl;
+ barrier();
+ if (i + N_DISPATCH_LOG < be64_to_cpu(vpa->dtl_idx)) {
+ /* buffer has overflowed */
+ pr_debug_ratelimited("vcpudispatch_stats: cpu %d lost %lld DTL samples\n",
+ d->cpu,
+ be64_to_cpu(vpa->dtl_idx) - N_DISPATCH_LOG - i);
+ i = be64_to_cpu(vpa->dtl_idx) - N_DISPATCH_LOG;
+ dtl = local_paca->dispatch_log + (i % N_DISPATCH_LOG);
+ continue;
+ }
+ update_vcpu_disp_stat(be16_to_cpu(dtle.processor_id));
+ ++i;
+ ++dtl;
+ if (dtl == dtl_end)
+ dtl = local_paca->dispatch_log;
+ }
+
+ __this_cpu_write(dtl_entry_ridx, i);
+
+out:
+ schedule_delayed_work_on(d->cpu, to_delayed_work(work),
+ HZ / vcpudispatch_stats_freq);
+}
+
+static int dtl_worker_online(unsigned int cpu)
+{
+ struct dtl_worker *d = &per_cpu(dtl_workers, cpu);
+
+ memset(d, 0, sizeof(*d));
+ INIT_DELAYED_WORK(&d->work, process_dtl_buffer);
+ d->cpu = cpu;
+
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+ per_cpu(dtl_entry_ridx, cpu) = 0;
+ register_dtl_buffer(cpu);
+#else
+ per_cpu(dtl_entry_ridx, cpu) = be64_to_cpu(lppaca_of(cpu).dtl_idx);
+#endif
+
+ schedule_delayed_work_on(cpu, &d->work, HZ / vcpudispatch_stats_freq);
+ return 0;
+}
+
+static int dtl_worker_offline(unsigned int cpu)
+{
+ struct dtl_worker *d = &per_cpu(dtl_workers, cpu);
+
+ cancel_delayed_work_sync(&d->work);
+
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+ unregister_dtl(get_hard_smp_processor_id(cpu));
+#endif
+
+ return 0;
+}
+
+static void set_global_dtl_mask(u8 mask)
+{
+ int cpu;
+
+ dtl_mask = mask;
+ for_each_present_cpu(cpu)
+ lppaca_of(cpu).dtl_enable_mask = dtl_mask;
+}
+
+static void reset_global_dtl_mask(void)
+{
+ int cpu;
+
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
+ dtl_mask = DTL_LOG_PREEMPT;
+#else
+ dtl_mask = 0;
+#endif
+ for_each_present_cpu(cpu)
+ lppaca_of(cpu).dtl_enable_mask = dtl_mask;
+}
+
+static int dtl_worker_enable(void)
+{
+ int rc = 0, state;
+
+ mutex_lock(&dtl_worker_mutex);
+
+ if (dtl_worker_refctr) {
+ dtl_worker_refctr++;
+ goto out;
+ }
+
+ if (register_dtl_buffer_access(1)) {
+ rc = -EBUSY;
+ goto out;
+ }
+
+ set_global_dtl_mask(DTL_LOG_ALL);
+
+ /* Setup dtl buffers and register those */
+ alloc_dtl_buffers();
+
+ state = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "powerpc/dtl:online",
+ dtl_worker_online, dtl_worker_offline);
+ if (state < 0) {
+ pr_err("vcpudispatch_stats: unable to setup workqueue for DTL processing\n");
+ free_dtl_buffers();
+ reset_global_dtl_mask();
+ unregister_dtl_buffer_access(1);
+ rc = -EINVAL;
+ goto out;
+ }
+ dtl_worker_state = state;
+ dtl_worker_refctr++;
+
+out:
+ mutex_unlock(&dtl_worker_mutex);
+ return rc;
+}
+
+static void dtl_worker_disable(void)
+{
+ mutex_lock(&dtl_worker_mutex);
+ dtl_worker_refctr--;
+ if (!dtl_worker_refctr) {
+ cpuhp_remove_state(dtl_worker_state);
+ dtl_worker_state = 0;
+ free_dtl_buffers();
+ reset_global_dtl_mask();
+ unregister_dtl_buffer_access(1);
+ }
+ mutex_unlock(&dtl_worker_mutex);
+}
+
+static ssize_t vcpudispatch_stats_write(struct file *file, const char __user *p,
+ size_t count, loff_t *ppos)
+{
+ struct vcpu_dispatch_data *disp;
+ int rc, cmd, cpu;
+ char buf[16];
+
+ if (count > 15)
+ return -EINVAL;
+
+ if (copy_from_user(buf, p, count))
+ return -EFAULT;
+
+ buf[count] = 0;
+ rc = kstrtoint(buf, 0, &cmd);
+ if (rc || cmd < 0 || cmd > 1) {
+ pr_err("vcpudispatch_stats: please use 0 to disable or 1 to enable dispatch statistics\n");
+ return rc ? rc : -EINVAL;
+ }
+
+ if ((cmd == 0 && !vcpudispatch_stats_on) ||
+ (cmd == 1 && vcpudispatch_stats_on))
+ return count;
+
+ if (cmd) {
+ rc = init_cpu_associativity();
+ if (rc)
+ return rc;
+
+ for_each_possible_cpu(cpu) {
+ disp = per_cpu_ptr(&vcpu_disp_data, cpu);
+ memset(disp, 0, sizeof(*disp));
+ disp->last_disp_cpu = -1;
+ }
+
+ rc = dtl_worker_enable();
+ if (rc) {
+ destroy_cpu_associativity();
+ return rc;
+ }
+ } else {
+ dtl_worker_disable();
+ destroy_cpu_associativity();
+ }
+
+ vcpudispatch_stats_on = cmd;
+
+ return count;
+}
+
+static int vcpudispatch_stats_display(struct seq_file *p, void *v)
+{
+ int cpu;
+ struct vcpu_dispatch_data *disp;
+
+ if (!vcpudispatch_stats_on) {
+ seq_puts(p, "off\n");
+ return 0;
+ }
+
+ for_each_online_cpu(cpu) {
+ disp = per_cpu_ptr(&vcpu_disp_data, cpu);
+ seq_printf(p, "cpu%d", cpu);
+ seq_put_decimal_ull(p, " ", disp->total_disp);
+ seq_put_decimal_ull(p, " ", disp->same_cpu_disp);
+ seq_put_decimal_ull(p, " ", disp->same_chip_disp);
+ seq_put_decimal_ull(p, " ", disp->diff_chip_disp);
+ seq_put_decimal_ull(p, " ", disp->far_chip_disp);
+ seq_put_decimal_ull(p, " ", disp->numa_home_disp);
+ seq_put_decimal_ull(p, " ", disp->numa_remote_disp);
+ seq_put_decimal_ull(p, " ", disp->numa_far_disp);
+ seq_puts(p, "\n");
+ }
+
+ return 0;
+}
+
+static int vcpudispatch_stats_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, vcpudispatch_stats_display, NULL);
+}
+
+static const struct file_operations vcpudispatch_stats_proc_ops = {
+ .open = vcpudispatch_stats_open,
+ .read = seq_read,
+ .write = vcpudispatch_stats_write,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static ssize_t vcpudispatch_stats_freq_write(struct file *file,
+ const char __user *p, size_t count, loff_t *ppos)
+{
+ int rc, freq;
+ char buf[16];
+
+ if (count > 15)
+ return -EINVAL;
+
+ if (copy_from_user(buf, p, count))
+ return -EFAULT;
+
+ buf[count] = 0;
+ rc = kstrtoint(buf, 0, &freq);
+ if (rc || freq < 1 || freq > HZ) {
+ pr_err("vcpudispatch_stats_freq: please specify a frequency between 1 and %d\n",
+ HZ);
+ return rc ? rc : -EINVAL;
+ }
+
+ vcpudispatch_stats_freq = freq;
+
+ return count;
+}
+
+static int vcpudispatch_stats_freq_display(struct seq_file *p, void *v)
+{
+ seq_printf(p, "%d\n", vcpudispatch_stats_freq);
+ return 0;
+}
+
+static int vcpudispatch_stats_freq_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, vcpudispatch_stats_freq_display, NULL);
+}
+
+static const struct file_operations vcpudispatch_stats_freq_proc_ops = {
+ .open = vcpudispatch_stats_freq_open,
+ .read = seq_read,
+ .write = vcpudispatch_stats_freq_write,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+static int __init vcpudispatch_stats_procfs_init(void)
+{
+ if (!lppaca_shared_proc(get_lppaca()))
+ return 0;
+
+ if (!proc_create("powerpc/vcpudispatch_stats", 0600, NULL,
+ &vcpudispatch_stats_proc_ops))
+ pr_err("vcpudispatch_stats: error creating procfs file\n");
+ else if (!proc_create("powerpc/vcpudispatch_stats_freq", 0600, NULL,
+ &vcpudispatch_stats_freq_proc_ops))
+ pr_err("vcpudispatch_stats_freq: error creating procfs file\n");
+
+ return 0;
+}
+
+machine_device_initcall(pseries, vcpudispatch_stats_procfs_init);
#endif /* CONFIG_PPC_SPLPAR */
void vpa_init(int cpu)
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 8/9] powerpc/pseries: Protect against hogging the cpu while setting up the stats
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (6 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 7/9] powerpc/pseries: Provide vcpu dispatch statistics Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
2019-06-14 15:44 ` [PATCH v2 9/9] powerpc/pseries: Add documentation for vcpudispatch_stats Naveen N. Rao
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
When enabling or disabling the vcpu dispatch statistics, we do a lot of
work including allocating/deallocating memory across all possible cpus
for the DTL buffer. In order to guard against hogging the cpu for too
long, track the time we're taking and yield the processor if necessary.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/plpar_wrappers.h | 2 +-
arch/powerpc/platforms/pseries/lpar.c | 29 ++++++++++++++++-------
arch/powerpc/platforms/pseries/setup.c | 2 +-
3 files changed, 22 insertions(+), 11 deletions(-)
diff --git a/arch/powerpc/include/asm/plpar_wrappers.h b/arch/powerpc/include/asm/plpar_wrappers.h
index 5f72f1fa4f81..5e5faabd7ba2 100644
--- a/arch/powerpc/include/asm/plpar_wrappers.h
+++ b/arch/powerpc/include/asm/plpar_wrappers.h
@@ -110,7 +110,7 @@ static inline long register_dtl(unsigned long cpu, unsigned long vpa)
extern bool register_dtl_buffer_access(bool global);
extern void unregister_dtl_buffer_access(bool global);
extern void register_dtl_buffer(int cpu);
-extern void alloc_dtl_buffers(void);
+extern void alloc_dtl_buffers(unsigned long *time_limit);
extern void vpa_init(int cpu);
static inline long plpar_pte_enter(unsigned long flags,
diff --git a/arch/powerpc/platforms/pseries/lpar.c b/arch/powerpc/platforms/pseries/lpar.c
index 200a96c8626a..318b08ada086 100644
--- a/arch/powerpc/platforms/pseries/lpar.c
+++ b/arch/powerpc/platforms/pseries/lpar.c
@@ -62,7 +62,7 @@ static u8 dtl_mask = DTL_LOG_PREEMPT;
static u8 dtl_mask;
#endif
-void alloc_dtl_buffers(void)
+void alloc_dtl_buffers(unsigned long *time_limit)
{
int cpu;
struct paca_struct *pp;
@@ -86,6 +86,11 @@ void alloc_dtl_buffers(void)
pp->dispatch_log = dtl;
pp->dispatch_log_end = dtl + N_DISPATCH_LOG;
pp->dtl_curr = dtl;
+
+ if (time_limit && time_after(jiffies, *time_limit)) {
+ cond_resched();
+ *time_limit = jiffies + HZ;
+ }
}
}
@@ -157,7 +162,7 @@ static int vcpudispatch_stats_freq = 50;
static __be32 *vcpu_associativity, *pcpu_associativity;
-static void free_dtl_buffers(void)
+static void free_dtl_buffers(unsigned long *time_limit)
{
#ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
int cpu;
@@ -172,6 +177,11 @@ static void free_dtl_buffers(void)
pp->dispatch_log = 0;
pp->dispatch_log_end = 0;
pp->dtl_curr = 0;
+
+ if (time_limit && time_after(jiffies, *time_limit)) {
+ cond_resched();
+ *time_limit = jiffies + HZ;
+ }
}
#endif
}
@@ -464,7 +474,7 @@ static void reset_global_dtl_mask(void)
lppaca_of(cpu).dtl_enable_mask = dtl_mask;
}
-static int dtl_worker_enable(void)
+static int dtl_worker_enable(unsigned long *time_limit)
{
int rc = 0, state;
@@ -483,13 +493,13 @@ static int dtl_worker_enable(void)
set_global_dtl_mask(DTL_LOG_ALL);
/* Setup dtl buffers and register those */
- alloc_dtl_buffers();
+ alloc_dtl_buffers(time_limit);
state = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "powerpc/dtl:online",
dtl_worker_online, dtl_worker_offline);
if (state < 0) {
pr_err("vcpudispatch_stats: unable to setup workqueue for DTL processing\n");
- free_dtl_buffers();
+ free_dtl_buffers(time_limit);
reset_global_dtl_mask();
unregister_dtl_buffer_access(1);
rc = -EINVAL;
@@ -503,14 +513,14 @@ static int dtl_worker_enable(void)
return rc;
}
-static void dtl_worker_disable(void)
+static void dtl_worker_disable(unsigned long *time_limit)
{
mutex_lock(&dtl_worker_mutex);
dtl_worker_refctr--;
if (!dtl_worker_refctr) {
cpuhp_remove_state(dtl_worker_state);
dtl_worker_state = 0;
- free_dtl_buffers();
+ free_dtl_buffers(time_limit);
reset_global_dtl_mask();
unregister_dtl_buffer_access(1);
}
@@ -520,6 +530,7 @@ static void dtl_worker_disable(void)
static ssize_t vcpudispatch_stats_write(struct file *file, const char __user *p,
size_t count, loff_t *ppos)
{
+ unsigned long time_limit = jiffies + HZ;
struct vcpu_dispatch_data *disp;
int rc, cmd, cpu;
char buf[16];
@@ -552,13 +563,13 @@ static ssize_t vcpudispatch_stats_write(struct file *file, const char __user *p,
disp->last_disp_cpu = -1;
}
- rc = dtl_worker_enable();
+ rc = dtl_worker_enable(&time_limit);
if (rc) {
destroy_cpu_associativity();
return rc;
}
} else {
- dtl_worker_disable();
+ dtl_worker_disable(&time_limit);
destroy_cpu_associativity();
}
diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index e4d75b958593..76fd6de90581 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -279,7 +279,7 @@ static int alloc_dispatch_logs(void)
if (!dtl_cache)
return 0;
- alloc_dtl_buffers();
+ alloc_dtl_buffers(0);
/* Register the DTL for the current (boot) cpu */
register_dtl_buffer(smp_processor_id());
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v2 9/9] powerpc/pseries: Add documentation for vcpudispatch_stats
2019-06-14 15:43 [PATCH v2 0/9] Provide vcpu dispatch statistics Naveen N. Rao
` (7 preceding siblings ...)
2019-06-14 15:44 ` [PATCH v2 8/9] powerpc/pseries: Protect against hogging the cpu while setting up the stats Naveen N. Rao
@ 2019-06-14 15:44 ` Naveen N. Rao
8 siblings, 0 replies; 10+ messages in thread
From: Naveen N. Rao @ 2019-06-14 15:44 UTC (permalink / raw)
To: Michael Ellerman
Cc: Nathan Lynch, Aneesh Kumar K.V, Mingming Cao, linuxppc-dev
Add a document describing the fields provided by
/proc/powerpc/vcpudispatch_stats.
Signed-off-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
---
Documentation/powerpc/vcpudispatch_stats.txt | 68 ++++++++++++++++++++
1 file changed, 68 insertions(+)
create mode 100644 Documentation/powerpc/vcpudispatch_stats.txt
diff --git a/Documentation/powerpc/vcpudispatch_stats.txt b/Documentation/powerpc/vcpudispatch_stats.txt
new file mode 100644
index 000000000000..e21476bfd78c
--- /dev/null
+++ b/Documentation/powerpc/vcpudispatch_stats.txt
@@ -0,0 +1,68 @@
+VCPU Dispatch Statistics:
+=========================
+
+For Shared Processor LPARs, the POWER Hypervisor maintains a relatively
+static mapping of the LPAR processors (vcpus) to physical processor
+chips (representing the "home" node) and tries to always dispatch vcpus
+on their associated physical processor chip. However, under certain
+scenarios, vcpus may be dispatched on a different processor chip (away
+from its home node).
+
+/proc/powerpc/vcpudispatch_stats can be used to obtain statistics
+related to the vcpu dispatch behavior. Writing '1' to this file enables
+collecting the statistics, while writing '0' disables the statistics.
+By default, the DTLB log for each vcpu is processed 50 times a second so
+as not to miss any entries. This processing frequency can be changed
+through /proc/powerpc/vcpudispatch_stats_freq.
+
+The statistics themselves are available by reading the procfs file
+/proc/powerpc/vcpudispatch_stats. Each line in the output corresponds to
+a vcpu as represented by the first field, followed by 8 numbers.
+
+The first number corresponds to:
+1. total vcpu dispatches since the beginning of statistics collection
+
+The next 4 numbers represent vcpu dispatch dispersions:
+2. number of times this vcpu was dispatched on the same processor as last
+ time
+3. number of times this vcpu was dispatched on a different processor core
+ as last time, but within the same chip
+4. number of times this vcpu was dispatched on a different chip
+5. number of times this vcpu was dispatches on a different socket/drawer
+(next numa boundary)
+
+The final 3 numbers represent statistics in relation to the home node of
+the vcpu:
+6. number of times this vcpu was dispatched in its home node (chip)
+7. number of times this vcpu was dispatched in a different node
+8. number of times this vcpu was dispatched in a node further away (numa
+distance)
+
+An example output:
+ $ sudo cat /proc/powerpc/vcpudispatch_stats
+ cpu0 6839 4126 2683 30 0 6821 18 0
+ cpu1 2515 1274 1229 12 0 2509 6 0
+ cpu2 2317 1198 1109 10 0 2312 5 0
+ cpu3 2259 1165 1088 6 0 2256 3 0
+ cpu4 2205 1143 1056 6 0 2202 3 0
+ cpu5 2165 1121 1038 6 0 2162 3 0
+ cpu6 2183 1127 1050 6 0 2180 3 0
+ cpu7 2193 1133 1052 8 0 2187 6 0
+ cpu8 2165 1115 1032 18 0 2156 9 0
+ cpu9 2301 1252 1033 16 0 2293 8 0
+ cpu10 2197 1138 1041 18 0 2187 10 0
+ cpu11 2273 1185 1062 26 0 2260 13 0
+ cpu12 2186 1125 1043 18 0 2177 9 0
+ cpu13 2161 1115 1030 16 0 2153 8 0
+ cpu14 2206 1153 1033 20 0 2196 10 0
+ cpu15 2163 1115 1032 16 0 2155 8 0
+
+In the output above, for vcpu0, there have been 6839 dispatches since
+statistics were enabled. 4126 of those dispatches were on the same
+physical cpu as the last time. 2683 were on a different core, but within
+the same chip, while 30 dispatches were on a different chip compared to
+its last dispatch.
+
+Also, out of the total of 6839 dispatches, we see that there have been
+6821 dispatches on the vcpu's home node, while 18 dispatches were
+outside its home node, on a neighbouring chip.
--
2.21.0
^ permalink raw reply related [flat|nested] 10+ messages in thread