Skip to contentSkip to conversation

powerpc/hv-24x7: Expose chip/sockets info to add json file metric support for the hv_24x7 socket/chip level events

PATCH Applied Kajol Jain 2 participants 7 messages May 25, 2020 – Jun 9, 2020 PowerPC

lore
Share & tools
lore

Kajol Jain

author
#0 lore raw

v10 cover powerpc/hv-24x7: Expose chip/sockets info to add json file metric support for the hv_24x7 socket/chip level events

Patchset fixes the inconsistent results we are getting when we run multiple 24x7 events.

"hv_24x7" pmu interface events needs system dependent parameter like socket/chip/core. For example, hv_24x7 chip level events needs specific chip-id to which the data is requested should be added as part of pmu events.

So to enable JSON file support to "hv_24x7" interface, patchset expose total number of sockets and chips per-socket details in sysfs files (sockets, chips) under "/sys/devices/hv_24x7/interface/".

To get number of sockets, chips per sockets and cores per chip patchset adds a rtas call with token "PROCESSOR_MODULE_INFO" to get these details. Patchset also handles partition migration case to re-init these system depended parameters by adding proper calls in post_mobility_fixup()(mobility.c).

Changelog:
v9 -> v10
- Change function name from 'read_sys_info_pseries()' to 'read_24x7_sys_info()'.
- Remove rtas_local_data_buf use, as we can directly use rtas_data_buf itself as we just need to do few loads and it also remove overload of memcpy in the lock
Suggested by: Michael Ellerman.
- Move part of updating 24x7 system info variables inside 'rtas_data_buf_lock' and make sure we release spinlock in each case.
- Added Tested-by: Madhavan Srinivasan tag for patch 1

v8 -> v9
- As we not need 8k data buffer we also not require new lock and buffer size. So, removed that part.
- Since we are using rtas_data_buf itself for rtas call, Switch to kmalloc for data buffer so that we can free that memory.
- Made more changes as suggested by Nathan Lynch.

v7 -> v8
- Add support for exposing cores per details as well. Suggested by: Madhavan Srinivasan.
- Remove config check for 'CONFIG_PPC_RTAS' in previous implementation and address other comments by Michael Ellerman.

v6 -> v7
- Split patchset into two patch series, one with kernel changes and another with perf tool side changes. This pachset contain all kernel side changes.

Kajol Jain (5):
powerpc/perf/hv-24x7: Fix inconsistent output values incase multiple
hv-24x7 events run
powerpc/hv-24x7: Add rtas call in hv-24x7 driver to get processor
details
powerpc/hv-24x7: Add sysfs files inside hv-24x7 device to show
processor details
Documentation/ABI: Add ABI documentation for chips and sockets powerpc/pseries: Update hv-24x7 information after migration

 .../sysfs-bus-event_source-devices-hv_24x7    | 21 ++++
 arch/powerpc/include/asm/rtas.h               |  6 ++
 arch/powerpc/perf/hv-24x7.c                   | 98 +++++++++++++++++--
 arch/powerpc/platforms/pseries/mobility.c     |  3 +
 4 files changed, 118 insertions(+), 10 deletions(-)
Signature · 1 line
2.18.2
author
#1 lore raw

PATCH v10 1/5 powerpc/perf/hv-24x7: Fix inconsistent output values incase multiple hv-24x7 events run

Commit 2b206ee6b0df ("powerpc/perf/hv-24x7: Display change in counter values")' added to print _change_ in the counter value rather then raw value for 24x7 counters. Incase of transactions, the event count is set to 0 at the beginning of the transaction. It also sets the event's prev_count to the raw value at the time of initialization. Because of setting event count to 0, we are seeing some weird behaviour, whenever we run multiple 24x7 events at a time.

For example:

command#: ./perf stat -e "{hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/,

			   hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/}"
	  		   -C 0 -I 1000 sleep 100

     1.000121704                120 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     1.000121704                  5 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     2.000357733                  8 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     2.000357733                 10 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     3.000495215 18,446,744,073,709,551,616 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     3.000495215 18,446,744,073,709,551,616 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     4.000641884                 56 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     4.000641884 18,446,744,073,709,551,616 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     5.000791887 18,446,744,073,709,551,616 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/

Getting these large values in case we do -I.

As we are setting event_count to 0, for interval case, overall event_count is not coming in incremental order. As we may can get new delta lesser then previous count. Because of which when we print intervals, we are getting negative value which create these large values.

This patch removes part where we set event_count to 0 in function 'h_24x7_event_read'. There won't be much impact as we do set event->hw.prev_count to the raw value at the time of initialization to print change value.

With this patch
In power9 platform

command#: ./perf stat -e "{hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/,

		           hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/}"
			   -C 0 -I 1000 sleep 100

     1.000117685                 93 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     1.000117685                  1 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     2.000349331                 98 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     2.000349331                  2 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     3.000495900                131 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     3.000495900                  4 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     4.000645920                204 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
     4.000645920                 61 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=1/
     4.284169997                 22 hv_24x7/PM_MCS01_128B_RD_DISP_PORT01,chip=0/
---
 arch/powerpc/perf/hv-24x7.c | 10 ----------
 1 file changed, 10 deletions(-)
arch/powerpc/perf/hv-24x7.c +0 -10
index 573e0b309c0c..48e8f4b17b91 100644
@@ -1400,16 +1400,6 @@ static void h_24x7_event_read(struct perf_event *event)
 			h24x7hw = &get_cpu_var(hv_24x7_hw);
 			h24x7hw->events[i] = event;
 			put_cpu_var(h24x7hw);
-			/*
-			 * Clear the event count so we can compute the _change_
-			 * in the 24x7 raw counter value at the end of the txn.
-			 *
-			 * Note that we could alternatively read the 24x7 value
-			 * now and save its value in event->hw.prev_count. But
-			 * that would require issuing a hcall, which would then
-			 * defeat the purpose of using the txn interface.
-			 */
-			local64_set(&event->count, 0);
 		}
 
 		put_cpu_var(hv_24x7_reqb);
Signature · 1 line
2.18.2
author
#2 lore raw

PATCH v10 2/5 powerpc/hv-24x7: Add rtas call in hv-24x7 driver to get processor details

In reply to Kajol Jain

For hv_24x7 socket/chip level events, specific chip-id to which the data requested should be added as part of pmu events. But number of chips/socket in the system details are not exposed.

Patch implements read_24x7_sys_info() to get system parameter values like number of sockets, cores per chip and chips per socket. Rtas_call with token "PROCESSOR_MODULE_INFO" is used to get these values.

Sub-sequent patch exports these values via sysfs.

Patch also make these parameters default to 1

---
 arch/powerpc/include/asm/rtas.h |  6 ++++
 arch/powerpc/perf/hv-24x7.c     | 64 +++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+)
arch/powerpc/include/asm/rtas.h +6 -0
index 3c1887351c71..7ef81b3e33f6 100644
@@ -483,5 +483,11 @@ static inline void rtas_initialize(void) { };
 
 extern int call_rtas(const char *, int, int, unsigned long *, ...);
 
+#ifdef CONFIG_HV_PERF_CTRS
+void read_24x7_sys_info(void);
+#else
+static inline void read_24x7_sys_info(void) { }
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* _POWERPC_RTAS_H */
arch/powerpc/perf/hv-24x7.c +64 -0

82 lines not on this page · raw message

Signature · 1 line
2.18.2
author
#3 lore raw

PATCH v10 3/5 powerpc/hv-24x7: Add sysfs files inside hv-24x7 device to show processor details

In reply to Kajol Jain

To expose the system dependent parameter like total number of sockets and numbers of chips per socket, patch adds two sysfs files. "sockets" and "chips" are added to /sys/devices/hv_24x7/interface/ of the "hv_24x7" pmu.

---
 arch/powerpc/perf/hv-24x7.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
arch/powerpc/perf/hv-24x7.c +24 -0
index 5cd954af5445..995dd170ca4b 100644
@@ -448,6 +448,24 @@ static ssize_t device_show_string(struct device *dev,
 	return sprintf(buf, "%s\n", (char *)d->var);
 }
 
+static ssize_t sockets_show(struct device *dev,
+			    struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%d\n", phys_sockets);
+}
+
+static ssize_t chipspersocket_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%d\n", phys_chipspersocket);
+}
+
+static ssize_t coresperchip_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
+{
+	return sprintf(buf, "%d\n", phys_coresperchip);
+}
+
 static struct attribute *device_str_attr_create_(char *name, char *str)
 {
 	struct dev_ext_attribute *attr = kzalloc(sizeof(*attr), GFP_KERNEL);
@@ -1094,6 +1112,9 @@ PAGE_0_ATTR(catalog_len, "%lld\n",
 		(unsigned long long)be32_to_cpu(page_0->length) * 4096);
 static BIN_ATTR_RO(catalog, 0/* real length varies */);
 static DEVICE_ATTR_RO(domains);
+static DEVICE_ATTR_RO(sockets);
+static DEVICE_ATTR_RO(chipspersocket);
+static DEVICE_ATTR_RO(coresperchip);
 
 static struct bin_attribute *if_bin_attrs[] = {
 	&bin_attr_catalog,
@@ -1104,6 +1125,9 @@ static struct attribute *if_attrs[] = {
 	&dev_attr_catalog_len.attr,
 	&dev_attr_catalog_version.attr,
 	&dev_attr_domains.attr,
+	&dev_attr_sockets.attr,
+	&dev_attr_chipspersocket.attr,
+	&dev_attr_coresperchip.attr,
 	NULL,
 };
 
Signature · 1 line
2.18.2
author
#4 lore raw

PATCH v10 4/5 Documentation/ABI: Add ABI documentation for chips and sockets

In reply to Kajol Jain

Add documentation for the following sysfs files:
/sys/devices/hv_24x7/interface/chipspersocket,
/sys/devices/hv_24x7/interface/sockets,
/sys/devices/hv_24x7/interface/coresperchip

---
 .../sysfs-bus-event_source-devices-hv_24x7    | 21 +++++++++++++++++++
 1 file changed, 21 insertions(+)
Documentation/ABI/testing/sysfs-bus-event_source-devices-hv_24x7 +21 -0
index ec27c6c9e737..e8698afcd952 100644
@@ -22,6 +22,27 @@ Description:
 		Exposes the "version" field of the 24x7 catalog. This is also
 		extractable from the provided binary "catalog" sysfs entry.
 
+What:		/sys/devices/hv_24x7/interface/sockets
+Date:		May 2020
+Contact:	Linux on PowerPC Developer List <[email protected]>
+Description:	read only
+		This sysfs interface exposes the number of sockets present in the
+		system.
+
+What:		/sys/devices/hv_24x7/interface/chipspersocket
+Date:		May 2020
+Contact:	Linux on PowerPC Developer List <[email protected]>
+Description:	read only
+		This sysfs interface exposes the number of chips per socket
+		present in the system.
+
+What:		/sys/devices/hv_24x7/interface/coresperchip
+Date:		May 2020
+Contact:	Linux on PowerPC Developer List <[email protected]>
+Description:	read only
+		This sysfs interface exposes the number of cores per chip
+		present in the system.
+
 What:		/sys/bus/event_source/devices/hv_24x7/event_descs/<event-name>
 Date:		February 2014
 Contact:	Linux on PowerPC Developer List <[email protected]>
Signature · 1 line
2.18.2
author
#5 lore raw

PATCH v10 5/5 powerpc/pseries: Update hv-24x7 information after migration

In reply to Kajol Jain

Function 'read_sys_info_pseries()' is added to get system parameter values like number of sockets and chips per socket.
and it gets these details via rtas_call with token "PROCESSOR_MODULE_INFO".

Incase lpar migrate from one system to another, system parameter details like chips per sockets or number of sockets might change. So, it needs to be re-initialized otherwise, these values corresponds to previous system values.
This patch adds a call to 'read_sys_info_pseries()' from 'post-mobility_fixup()' to re-init the physsockets and physchips values

---
 arch/powerpc/platforms/pseries/mobility.c | 3 +++
 1 file changed, 3 insertions(+)
arch/powerpc/platforms/pseries/mobility.c +3 -0
index b571285f6c14..10d982997736 100644
@@ -371,6 +371,9 @@ void post_mobility_fixup(void)
 	/* Possibly switch to a new RFI flush type */
 	pseries_setup_rfi_flush();
 
+	/* Reinitialise system information for hv-24x7 */
+	read_24x7_sys_info();
+
 	return;
 }
 
Signature · 1 line
2.18.2
#6 lore raw
In reply to Kajol Jain
Quoted 9 lines · Kajol Jain

Patchset fixes the inconsistent results we are getting when we run multiple 24x7 events.

"hv_24x7" pmu interface events needs system dependent parameter like socket/chip/core. For example, hv_24x7 chip level events needs specific chip-id to which the data is requested should be added as part of pmu events.

[...]

Applied to powerpc/next.

[1/5] powerpc/perf/hv-24x7: Fix inconsistent output values incase multiple hv-24x7 events run

      https://git.kernel.org/powerpc/c/b4ac18eead28611ff470d0f47a35c4e0ac080d9c
[2/5] powerpc/hv-24x7: Add rtas call in hv-24x7 driver to get processor details
      https://git.kernel.org/powerpc/c/8ba21426738207711347335b2cf3e99c690fc777
[3/5] powerpc/hv-24x7: Add sysfs files inside hv-24x7 device to show processor details
      https://git.kernel.org/powerpc/c/60beb65da1efd4cc23d05141181c39b98487950f
[4/5] Documentation/ABI: Add ABI documentation for chips and sockets
      https://git.kernel.org/powerpc/c/15cd1d35ba4a59832df693858ef046457107bd8d
[5/5] powerpc/pseries: Update hv-24x7 information after migration
      https://git.kernel.org/powerpc/c/373b373053384f12951ae9f916043d955501d482

cheers