The function xxx_find_related_event() scan all working events to find
related events. During this process, we also can find the idle counters.
If not found related events, return the first idle counter to simplify
the code.
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-8-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
If we use two events with the same filter and related event type
(see the following example), the driver check whether they are related
events and are in the same group, otherwise the function
hisi_pcie_pmu_find_related_event() return -EINVAL, then the 2nd event
cannot count but the 1st event is running, although the PCIe PMU has
other idle counters.
In this case, The perf event scheduler will make the two events to
multiplex a counter, if the user use the formula
(1st event_value / 2nd event_value) to calculate the bandwidth, he/she
won't get the correct value, because they are not counting at the
same period.
This patch tries to fix this by making the related events to use
different idle counters if they are not in the same event group.
And finally, I'm going to say. The related events are best used in the
same group [1]. There are two ways to know if they are related events.
a) By event name, such as the latency events "xxx_latency, xxx_cnt" or
bandwidth events "xxx_flux, xxx_time".
b) By event type, such as "event=0xXXXX, event=0x1XXXX".
Use group to count the related events:
[1] -e "{pmu_name/xxx_latency,port=1/,pmu_name/xxx_cnt,port=1/}"
example:
1st event: hisi_pcie0_core1/event=0x804,port=1
2nd event: hisi_pcie0_core1/event=0x10804,port=1
test cmd:
perf stat -e hisi_pcie0_core1/event=0x804,port=1/ \
-e hisi_pcie0_core1/event=0x10804,port=1/
before patch:
25,281 hisi_pcie0_core1/event=0x804,port=1/ (49.91%)
470,598 hisi_pcie0_core1/event=0x10804,port=1/ (50.09%)
after patch:
24,147 hisi_pcie0_core1/event=0x804,port=1/
474,558 hisi_pcie0_core1/event=0x10804,port=1/
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huwei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-7-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The PMU can monitor traffic of certain target Root Port or downstream
target Endpoint. User can specify the target filter by the "port" or
"bdf" option respectively. The PMU can only monitor the Root Port or
Endpoint on the same PCIe core so the value of "port" or "bdf" should
be valid and will be checked by the driver.
Currently at least and only one of "port" and "bdf" option must be set.
If "port" filter is not set or is set explicitly to zero (default),
driver will regard the user specifies a "bdf" option since "port" option
is a bitmask of the target Root Ports and zero is not a valid
value.
If user not explicitly set "port" or "bdf" filter, the driver uses "bdf"
default value (zero) to set target filter, but driver will skip the
check of bdf=0, although it's a valid value (meaning 0000:000:00.0).
Then the user just gets zero.
Therefore, we need to check if both "port" and "bdf" are invalid, then
return failure and report warning.
Testing:
before the patch:
0 hisi_pcie0_core1/rx_mrd_flux/
0 hisi_pcie0_core1/rx_mrd_flux,port=0/
24,124 hisi_pcie0_core1/rx_mrd_flux,port=1/
0 hisi_pcie0_core1/rx_mrd_flux,bdf=0/
0 hisi_pcie0_core1/rx_mrd_flux,port=0x800/
<not supported> hisi_pcie0_core1/rx_mrd_flux,bdf=1/
24,132 hisi_pcie0_core1/rx_mrd_flux,bdf=0x1700/
<not supported> hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x0/
<not supported> hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x1/
24,138 hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x1700/
24,126 hisi_pcie0_core1/rx_mrd_flux,port=0x1,bdf=0x0/
after the patch:
<not supported> hisi_pcie0_core1/rx_mrd_flux/
<not supported> hisi_pcie0_core1/rx_mrd_flux,port=0/
24,153 hisi_pcie0_core1/rx_mrd_flux,port=1/
0 hisi_pcie0_core1/rx_mrd_flux,port=0x800/
<not supported> hisi_pcie0_core1/rx_mrd_flux,bdf=0/
<not supported> hisi_pcie0_core1/rx_mrd_flux,bdf=1/
24,117 hisi_pcie0_core1/rx_mrd_flux,bdf=0x1700/
<not supported> hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x0/
<not supported> hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x1/
24,120 hisi_pcie0_core1/rx_mrd_flux,port=0x0,bdf=0x1700/
24,123 hisi_pcie0_core1/rx_mrd_flux,port=0x1,bdf=0x0/
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-6-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
A typical PCIe transaction is consisted of various TLP packets in both
direction. For counting bandwidth only memory read events are exported
currently. Add memory write and completion counting events of both
direction to complete the bandwidth counting.
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-5-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The metric counting shows incorrect results if the events in the
metric group using the same event but different filter options.
This is because we only judge the event code to decide whether
the event in the metric group should share the same hardware
counter, but ignore the settings of the filter.
For example, on a platform of 2 ports 0x1 and 0x2 but only port
0x1 has a downstream PCIe NVME device. The metric counting
shows both ports have the same counts because we misassign these
two events to one same hardware counter:
[root@localhost perf-iostat]# ./perf stat -e '{hisi_pcie0_core1/event=0x0104,port=0x2/,hisi_pcie0_core1/event=0x0104,port=0x1/}'
Performance counter stats for 'system wide':
7907484924 hisi_pcie0_core1/event=0x0104,port=0x2/
7907484924 hisi_pcie0_core1/event=0x0104,port=0x1/
10.153863691 seconds time elapsed
Fix this by using the whole config rather than the event only
to judge whether two events are the same and should share the
same hardware counter. With this patch, the metric counting in
the above case tends to be corrected:
[root@localhost perf-iostat]# ./perf stat -e '{hisi_pcie0_core1/event=0x0104,port=0x2/,hisi_pcie0_core1/event=0x0104,port=0x1/}'
Performance counter stats for 'system wide':
0 hisi_pcie0_core1/event=0x0104,port=0x2/
8123122077 hisi_pcie0_core1/event=0x0104,port=0x1/
10.152875631 seconds time elapsed
Fixes: 8404b0fbc7 ("drivers/perf: hisi: Add driver for HiSilicon PCIe PMU")
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-4-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Factor out retrieving of the register value for the
corresponding event from hisi_pcie_config_event_ctrl() into a
new function hisi_pcie_pmu_get_event_ctrl_val() allowing future
reuse.
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-3-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
hisi_pcie_pmu_{config,clear}_filter() are config/clear HISI_PCIE_EVENT_CTRL
register which contains not only the filter but also the event code. The
function names are bit misleading. Rename it to
hisi_pcie_pmu_{config,clear}_event_ctrl() to reflects their functions
more accurately.
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20240223103359.18669-2-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
HiSilicon UC PMU v2 suffers the erratum 162700402 that the PMU counter
cannot be set due to the lack of clock under power saving mode. This will
lead to error or inaccurate counts. The clock can be enabled by the PMU
global enabling control.
This patch tries to fix this by set the UC PMU enable before set event
period to turn on the clock, and then restore the UC PMU configuration.
The counter register can hold its value without a clock.
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20240227125231.53127-1-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
Trivially convert these drivers from always returning zero in the remove
callback to the void returning variant.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/33a8be0641b9447469fb7f6af0a10fb65efa97a3.1702648125.git.u.kleine-koenig@pengutronix.de
Signed-off-by: Will Deacon <will@kernel.org>
Some event id of HiSilicon uncore UC PMU driver is incorrect, fix them.
Fixes: 312eca95e2 ("drivers/perf: hisi: Add support for HiSilicon UC PMU driver")
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20231204110425.20354-1-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
When we fail to register the uncore pmu, the pmu context may not been
allocated. The error handing will call cpuhp_state_remove_instance()
to call uncore pmu offline callback, which migrate the pmu context.
Since that's liable to lead to some kind of use-after-free.
Use cpuhp_state_remove_instance_nocalls() instead of
cpuhp_state_remove_instance() so that the notifiers don't execute after
the PMU device has been failed to register.
Fixes: a0ab25cd82 ("drivers/perf: hisi: Add support for HiSilicon PA PMU driver")
FIxes: 3bf30882c3 ("drivers/perf: hisi: Add support for HiSilicon SLLC PMU driver")
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Link: https://lore.kernel.org/r/20231024113630.13472-1-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Check whether the event type matches the PMU type firstly in
pmu::event_init() before touching the event. Otherwise we'll
change the events of others and lead to incorrect results.
Since in perf_init_event() we may call every pmu's event_init()
in a certain case, we should not modify the event if it's not
ours.
Fixes: 8404b0fbc7 ("drivers/perf: hisi: Add driver for HiSilicon PCIe PMU")
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20231024092954.42297-2-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
When tearing down a 'hisi_hns3' PMU, we mistakenly run the CPU hotplug
callbacks after the device has been unregistered, leading to fireworks
when we try to execute empty function callbacks within the driver:
| Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
| CPU: 0 PID: 15 Comm: cpuhp/0 Tainted: G W O 5.12.0-rc4+ #1
| Hardware name: , BIOS KpxxxFPGA 1P B600 V143 04/22/2021
| pstate: 80400009 (Nzcv daif +PAN -UAO -TCO BTYPE=--)
| pc : perf_pmu_migrate_context+0x98/0x38c
| lr : perf_pmu_migrate_context+0x94/0x38c
|
| Call trace:
| perf_pmu_migrate_context+0x98/0x38c
| hisi_hns3_pmu_offline_cpu+0x104/0x12c [hisi_hns3_pmu]
Use cpuhp_state_remove_instance_nocalls() instead of
cpuhp_state_remove_instance() so that the notifiers don't execute after
the PMU device has been unregistered.
Fixes: 66637ab137 ("drivers/perf: hisi: add driver for HNS3 PMU")
Signed-off-by: Hao Chen <chenhao418@huawei.com>
Signed-off-by: Jijie Shao <shaojijie@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20231019091352.998964-1-shaojijie@huawei.com
[will: Rewrote commit message]
Signed-off-by: Will Deacon <will@kernel.org>
The PCIe PMUs locate on different NUMA node but currently we don't
consider it and likely stack all the sessions on the same CPU:
[root@localhost tmp]# cat /sys/devices/hisi_pcie*/cpumask
0
0
0
0
0
0
This can be optimize a bit to use a local CPU for the PMU.
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20230815131010.2147-1-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon Hip09 platform, there are 4 UC (unified cache) modules
on each chip CCL (CPU Cluster). UC is a cache that provides
coherence between NUMA and UMA domains. It is located between L2
and Memory System. Many PMU events are supported. Let's support
the UC PMU driver using the HiSilicon uncore PMU framework.
* rd_req_en : rd_req_en is the abbreviation of read request tracetag
enable and allows user to count only read operations. Details are listed
in the hisi-pmu document at Documentation/admin-guide/perf/hisi-pmu.rst
* srcid_en & srcid: Allows users to filter statistical information based
on specific CPU/ICL by srcid.
srcid_en depends on rd_req_en being enabled.
* uring_channel: Allows users to filter statistical information based on
the specified tx request uring channel.
uring_channel only supported events: [0x47 ~ 0x59].
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20230615125926.29832-3-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Compared to the original PA device, H60PA offers higher bandwidth.
The H60PA is a new device and we use HID to differentiate them.
The events supported by PAv3 and PAv2 are different. The PAv3 PMU
removed some events which are supported by PAv2 PMU. The older PA
PMU driver will probe v3 as v2. Therefore PA events displayed by
"perf list" cannot work properly. We add the HISI0275 HID for PAv3
PMU to distinguish different.
For each H60PA PMU, except for the overflow interrupt register, other
functions of the H60PA PMU are the same as the original PA PMU module.
It has 8-programable counters and each counter is free-running.
Interrupt is supported to handle counter (64-bits) overflow.
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20230615125926.29832-2-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The driver needs to migrate the perf context if the current using CPU going
to teardown. By the time calling the cpuhp::teardown() callback the
cpu_online_mask() hasn't updated yet and still includes the CPU going to
teardown. In current driver's implementation we may migrate the context
to the teardown CPU and leads to the below calltrace:
...
[ 368.104662][ T932] task:cpuhp/0 state:D stack: 0 pid: 15 ppid: 2 flags:0x00000008
[ 368.113699][ T932] Call trace:
[ 368.116834][ T932] __switch_to+0x7c/0xbc
[ 368.120924][ T932] __schedule+0x338/0x6f0
[ 368.125098][ T932] schedule+0x50/0xe0
[ 368.128926][ T932] schedule_preempt_disabled+0x18/0x24
[ 368.134229][ T932] __mutex_lock.constprop.0+0x1d4/0x5dc
[ 368.139617][ T932] __mutex_lock_slowpath+0x1c/0x30
[ 368.144573][ T932] mutex_lock+0x50/0x60
[ 368.148579][ T932] perf_pmu_migrate_context+0x84/0x2b0
[ 368.153884][ T932] hisi_pcie_pmu_offline_cpu+0x90/0xe0 [hisi_pcie_pmu]
[ 368.160579][ T932] cpuhp_invoke_callback+0x2a0/0x650
[ 368.165707][ T932] cpuhp_thread_fun+0xe4/0x190
[ 368.170316][ T932] smpboot_thread_fn+0x15c/0x1a0
[ 368.175099][ T932] kthread+0x108/0x13c
[ 368.179012][ T932] ret_from_fork+0x10/0x18
...
Use function cpumask_any_but() to find one correct active cpu to fixes
this issue.
Fixes: 8404b0fbc7 ("drivers/perf: hisi: Add driver for HiSilicon PCIe PMU")
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20230608114326.27649-1-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
When allocations fails that can be NULL now.
If the name provided is NULL, then the initialization process of the PMU
type and dev will be skipped in function perf_pmu_register().
Consequently, the PMU will not be able to register into the kernel.
Moreover, in the case of unregister the PMU, the function device_del()
will need to handle NULL pointers, which potentially can cause issues.
So move this allocation above the cpuhp_state_add_instance() and directly
return if it does fail.
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20230403081423.62460-3-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
"pmu->name" is initialized by perf_pmu_register() function, so remove
the redundant initialized in hisi_pmu_init().
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20230403081423.62460-2-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Missed initialization the variable of pmu::capabilities when extract
the initialization code of hisi_pmu->pmu into a function.
HISI UNCORE PMU drivers counters that not support context exclusion.
So we have to advertise the PERF_PMU_CAP_NO_EXCLUDE capability.
This ensures that perf will prevent us from handling events where
any exclusion flags are set.
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Link: https://lore.kernel.org/r/20230119100307.3660-2-hejunhao3@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The PMU support to filter the TLP when counting the bandwidth with below
options:
- only count the TLP headers
- only count the TLP payloads
- count both TLP headers and payloads
In the current driver it's default to count the TLP payloads only, which
will have an implicity side effects that on the traffic only have header
only TLPs, we'll get no data.
Make this user configuration through "len_mode" parameter and make it
default to count both TLP headers and payloads when user not specified.
Also update the documentation for it.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20221117084136.53572-5-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Some event id of hisi-pcie-pmu is incorrect, fix them.
Fixes: 8404b0fbc7 ("drivers/perf: hisi: Add driver for HiSilicon PCIe PMU")
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20221117084136.53572-2-yangyicong@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
HNS3(HiSilicon Network System 3) PMU is RCiEP device in HiSilicon SoC NIC,
supports collection of performance statistics such as bandwidth, latency,
packet rate and interrupt rate.
NIC of each SICL has one PMU device for it. Driver registers each PMU
device to perf, and exports information of supported events, filter mode of
each event, bdf range, hardware clock frequency, identifier and so on via
sysfs.
Each PMU device has its own registers of control, counters and interrupt,
and it supports 8 hardware events, each hardward event has its own
registers for configuration, counters and interrupt.
Filter options contains:
config - select event
port - select physical port of nic
tc - select tc(must be used with port)
func - select PF/VF
queue - select queue of PF/VF(must be used with func)
intr - select interrupt number(must be used with func)
global - select all functions of IO DIE
Signed-off-by: Guangbin Huang <huangguangbin2@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Reviewed-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20220628063419.38514-3-huangguangbin2@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon Hip09 platform, there is a CPA (Coherency Protocol Agent) on
each SICL (Super IO Cluster) which implements packet format translation,
route parsing and traffic statistics.
CPA PMU has 8 PMU counters and interrupt is supported to handle counter
overflow. Let's support its driver under the framework of HiSilicon PMU
driver.
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Reviewed-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20220415102352.6665-3-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
If a PMU is in a SICL (Super IO cluster), it is not appropriate to
associate this PMU with a CPU die. So we associate it with all CPUs
online, rather than CPUs in the nearest SCCL.
As the firmware of Hip09 platform hasn't been published yet, change
of PMU driver will not influence backwards compatibility between
driver and firmware.
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Link: https://lore.kernel.org/r/20220415102352.6665-2-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
In some places, drivers/perf code calls bitmap_weight() to check if any
bit of a given bitmap is set. It's better to use bitmap_empty() in that
case because bitmap_empty() stops traversing the bitmap as soon as it
finds first set bit, while bitmap_weight() counts all bits unconditionally.
Signed-off-by: Yury Norov <yury.norov@gmail.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20220210224933.379149-13-yury.norov@gmail.com
Signed-off-by: Will Deacon <will@kernel.org>
PCIe PMU Root Complex Integrated End Point(RCiEP) device is supported
to sample bandwidth, latency, buffer occupation etc.
Each PMU RCiEP device monitors multiple Root Ports, and each RCiEP is
registered as a PMU in /sys/bus/event_source/devices, so users can
select target PMU, and use filter to do further sets.
Filtering options contains:
event - select the event.
port - select target Root Ports. Information of Root Ports are
shown under sysfs.
bdf - select requester_id of target EP device.
trig_len - set trigger condition for starting event statistics.
trig_mode - set trigger mode. 0 means starting to statistic when bigger
than trigger condition, and 1 means smaller.
thr_len - set threshold for statistics.
thr_mode - set threshold mode. 0 means count when bigger than threshold,
and 1 means smaller.
Acked-by: Krzysztof Wilczyński <kw@linux.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20211202080633.2919-3-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The PA PMU counter offset was correct in [1] and the driver has
already been verified. We want to keep the register offset using
lower case character in later version that is consistent with
the existed driver. Since there was no functional change, we
didn't do more test. However there is typo when modified the PA
PMU counter offset by mistake, so fix this bad mistake.
[1] https://www.spinics.net/lists/arm-kernel/msg865263.html
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20210928123022.23467-1-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
These are only put in an array of pointers to const attribute_group
structs. Make them const like the other static attribute_group structs
to allow the compiler to put them in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210605221514.73449-1-rikard.falkeborn@gmail.com
Signed-off-by: Will Deacon <will@kernel.org>
'Data source' is a new function for HHA PMU and config / clear
interface was wrong by mistake. 'HHA_DATSRC_CTRL' register is
mainly used for data source configuration, if we enable bit0
as driver, it will go on count the event and we didn't check
it carefully. So fix the issue and do as the initial purpose.
Fixes: 932f6a99f9 ("drivers/perf: hisi: Add new functions for HHA PMU")
Reported-by: kernel test robot <lkp@intel.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1622709291-37996-1-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
These drivers use irq_set_affinity_hint() to set the affinity for the PMU
interrupts, which relies on the undocumented side effect that this function
actually sets the affinity under the hood.
Setting an hint is clearly not a guarantee and for these PMU interrupts an
affinity hint, which is supposed to guide userspace for setting affinity,
is beyond pointless, because the affinity of these interrupts cannot be
modified from user space.
Aside of that the error checks are bogus because the only error which is
returned from irq_set_affinity_hint() is when there is no irq descriptor
for the interrupt number, but not when the affinity set fails. That's on
purpose because the hint can point to an offline CPU.
Replace the mindless abuse with irq_set_affinity().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Acked-by: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/20210518093118.813375875@linutronix.de
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon Hip09 platform, there is a PA (Protocol Adapter) module on
each chip SICL (Super I/O Cluster) which incorporates three Hydra interface
and facilitates the cache coherency between the dies on the chip. While PA
uncore PMU model is the same as other Hip09 PMU modules and many PMU events
are supported. Let's support the PMU driver using the HiSilicon uncore PMU
framework.
PA PMU supports the following filter functions:
* tracetag_en: allows user to count events according to tt_req or
tt_core set in L3C PMU. It's the same as other PMUs.
* srcid_cmd & srcid_msk: allows user to filter statistics that come from
specific CCL/ICL by configuration source ID.
* tgtid_cmd & tgtid_msk: it is the similar function to srcid_cmd &
srcid_msk. Both are used to check where the data comes from or go to.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-9-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
HiSilicon's Hip09 is comprised by multi-dies that can be connected by SLLC
module (Skyros Link Layer Controller), its has separate PMU registers which
the driver can program it freely and interrupt is supported to handle
counter overflow. Let's support its driver under the framework of HiSilicon
uncore PMU driver.
SLLC PMU supports the following filter functions:
* tracetag_en: allows user to count data according to tt_req or
tt_core set in L3C PMU.
* srcid_cmd & srcid_msk: allows user to filter statistics that come from
specific CCL/ICL by configuration source ID.
* tgtid_hi & tgtid_lo: it also supports event statistics that these
operations will go to the CCL/ICL by configuration target ID or
target ID range. It's the same as source ID with 11-bit width in
the SoC. More introduction is added in documentation:
Documentation/admin-guide/perf/hisi-pmu.rst
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-8-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
DDRC PMU's events are useful for performance profiling, but the events
are limited and counter is fixed. On HiSilicon Hip09 platform, PMU
counters are the programmable and more events are supported. Let's
add the DDRC PMU v2 driver.
Bandwidth events are exposed directly in driver and some more events
will listed in JSON file later.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-7-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon Hip09 platform, some new functions are also supported on
HHA PMU.
* tracetag_en: it is the abbreviation of tracetag enable and allows user
to count events according to tt_req or tt_core set in L3C PMU.
* datasrc_skt: it is the abbreviation of data source from another
socket and it is used in the multi-chips. It's the same as L3C PMU.
* srcid_cmd & srcid_msk: pair of the fields are used to filter
statistics that come from the specific CCL/ICL by the configuration.
These are the abbreviation of source ID command and mask. The source
ID is 11-bit and detailed descriptions are documented in
Documentation/admin-guide/perf/hisi-pmu.rst.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-6-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon Hip09 platform, some new functions are enhanced on L3C PMU:
* tt_req: it is the abbreviation of tracetag request and allows user to
count only read/write/atomic operations. tt_req is 3-bit and details are
listed in the hisi-pmu document.
$# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_req=0x4/ sleep 5
* tt_core: it is the abbreviation of tracetag core and allows user to
filter by core/thread within the cluster, it is a 8-bit bitmap that each
bit represents the corresponding core/thread in this L3C.
$# perf stat -a -e hisi_sccl3_l3c0/config=0x02,tt_core=0xf/ sleep 5
* datasrc_cfg: it is the abbreviation of data source configuration and
allows user to check where the data comes from, such as: from local DDR,
cross-die DDR or cross-socket DDR. Its is 5-bit and represents different
data source in the SoC.
$# perf stat -a -e hisi_sccl3_l3c0/dat_access,datasrc_cfg=0xe/ sleep 5
* datasrc_skt: it is the abbreviation of data source from another socket
and is used in the multi-chips, if user wants to check the cross-socket
datat source, it shall be added in perf command. Only one bit is used to
control this.
$# perf stat -a -e hisi_sccl3_l3c0/dat_access,datasrc_cfg=0x10,datasrc_skt=1/ sleep 5
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-5-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
For HiSilicon uncore PMU, more versions are supported and some variables
shall be added suffix to distinguish the version which are prepared for
the new drivers.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-4-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
On HiSilicon uncore PMU drivers, interrupt handling function and interrupt
registration function are very similar in differents PMU modules. Let's
refactor the frame.
Two new callbacks are added for the HW accessors:
* hisi_uncore_ops::get_int_status returns a bitmap of events which
have overflowed and raised an interrupt
* hisi_uncore_ops::clear_int_status clears the overflow status for a
specific event
These callback functions are used by a common IRQ handler,
hisi_uncore_pmu_isr().
One more function hisi_uncore_pmu_init_irq() is added to replace each
PMU initialization IRQ interface and simplify the code.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-3-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
The sanity check for counter index has been done in the function
hisi_uncore_pmu_get_event_idx, so remove the redundant interface
hisi_uncore_pmu_counter_valid() and sanity check.
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: John Garry <john.garry@huawei.com>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Co-developed-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1615186237-22263-2-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
sprintf does not know the PAGE_SIZE maximum of the temporary buffer
used for sysfs content and it's possible to overrun the buffer length.
Use sysfs_emit() function to ensures that no overrun is done.
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Link: https://lore.kernel.org/r/1616148273-16374-4-git-send-email-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
Fix the following coccicheck warning:
./drivers/perf/hisilicon/hisi_uncore_pmu.c:128:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/fsl_imx8_ddr_perf.c:173:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm_spe_pmu.c:129:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm_smmu_pmu.c:563:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm_dsu_pmu.c:149:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm_dsu_pmu.c:139:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cmn.c:563:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cmn.c:351:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-ccn.c:224:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cci.c:708:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cci.c:699:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cci.c:528:8-16: WARNING: use scnprintf or sprintf.
./drivers/perf/arm-cci.c:309:8-16: WARNING: use scnprintf or sprintf.
Signed-off-by: Zihao Tang <tangzihao1@hisilicon.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Link: https://lore.kernel.org/r/1616148273-16374-2-git-send-email-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
The only usage is to put their addresses in an array of pointers to
const struct attribute group. Make them const to allow the compiler
to put them in read-only memory.
Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210117212847.21319-4-rikard.falkeborn@gmail.com
Signed-off-by: Will Deacon <will@kernel.org>
To allow userspace to identify the specific implementation of the device,
add an "identifier" sysfs file.
Encoding is as follows (same for all uncore drivers):
hi1620: 0x0
hi1630: 0x30
Signed-off-by: John Garry <john.garry@huawei.com>
Link: https://lore.kernel.org/r/1602149181-237415-2-git-send-email-john.garry@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
MODULE_*** is used in HiSilicon uncore PMU drivers and is provided by
linux/module.h, but the header file is not directly included. Add the
missing include.
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Link: https://lore.kernel.org/r/1599186097-18599-1-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Will Deacon <will@kernel.org>
Forcefully unbinding PMU drivers during perf sampling will lead to
a kernel panic, because the perf upper-layer framework call a NULL
pointer in this situation.
To solve this issue, "suppress_bind_attrs" should be set to true, so
that bind/unbind can be disabled via sysfs and prevent unbinding PMU
drivers during perf sampling.
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Link: https://lore.kernel.org/r/1594975763-32966-1-git-send-email-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
