* [PATCH v13 1/5] x86: perf: Move RDPMC event flag to a common definition
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
@ 2021-12-08 20:11 ` Rob Herring
2021-12-08 20:11 ` [PATCH v13 2/5] perf: Add a counter for number of user access events in context Rob Herring
` (4 subsequent siblings)
5 siblings, 0 replies; 11+ messages in thread
From: Rob Herring @ 2021-12-08 20:11 UTC (permalink / raw)
To: Will Deacon, Mark Rutland, Peter Zijlstra
Cc: Vince Weaver, Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov, x86,
H. Peter Anvin, linux-kernel, linux-arm-kernel, linux-perf-users,
Kan Liang
In preparation to enable user counter access on arm64 and to move some
of the user access handling to perf core, create a common event flag for
user counter access and convert x86 to use it.
Since the architecture specific flags start at the LSB, starting at the
MSB for common flags.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: x86@kernel.org
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: linux-perf-users@vger.kernel.org
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Rob Herring <robh@kernel.org>
---
v11:
- Move PERF_EVENT_FLAG_USER_READ_CNT definition above struct
---
arch/x86/events/core.c | 10 +++++-----
arch/x86/events/perf_event.h | 2 +-
include/linux/perf_event.h | 9 +++++++++
3 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c
index 38b2c779146f..68dea7ce6a22 100644
--- a/arch/x86/events/core.c
+++ b/arch/x86/events/core.c
@@ -2476,7 +2476,7 @@ static int x86_pmu_event_init(struct perf_event *event)
if (READ_ONCE(x86_pmu.attr_rdpmc) &&
!(event->hw.flags & PERF_X86_EVENT_LARGE_PEBS))
- event->hw.flags |= PERF_X86_EVENT_RDPMC_ALLOWED;
+ event->hw.flags |= PERF_EVENT_FLAG_USER_READ_CNT;
return err;
}
@@ -2510,7 +2510,7 @@ void perf_clear_dirty_counters(void)
static void x86_pmu_event_mapped(struct perf_event *event, struct mm_struct *mm)
{
- if (!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+ if (!(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT))
return;
/*
@@ -2531,7 +2531,7 @@ static void x86_pmu_event_mapped(struct perf_event *event, struct mm_struct *mm)
static void x86_pmu_event_unmapped(struct perf_event *event, struct mm_struct *mm)
{
- if (!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+ if (!(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT))
return;
if (atomic_dec_and_test(&mm->context.perf_rdpmc_allowed))
@@ -2542,7 +2542,7 @@ static int x86_pmu_event_idx(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
- if (!(hwc->flags & PERF_X86_EVENT_RDPMC_ALLOWED))
+ if (!(hwc->flags & PERF_EVENT_FLAG_USER_READ_CNT))
return 0;
if (is_metric_idx(hwc->idx))
@@ -2725,7 +2725,7 @@ void arch_perf_update_userpage(struct perf_event *event,
userpg->cap_user_time = 0;
userpg->cap_user_time_zero = 0;
userpg->cap_user_rdpmc =
- !!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED);
+ !!(event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT);
userpg->pmc_width = x86_pmu.cntval_bits;
if (!using_native_sched_clock() || !sched_clock_stable())
diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h
index 5480db242083..9d376e528dfc 100644
--- a/arch/x86/events/perf_event.h
+++ b/arch/x86/events/perf_event.h
@@ -74,7 +74,7 @@ static inline bool constraint_match(struct event_constraint *c, u64 ecode)
#define PERF_X86_EVENT_PEBS_NA_HSW 0x0010 /* haswell style datala, unknown */
#define PERF_X86_EVENT_EXCL 0x0020 /* HT exclusivity on counter */
#define PERF_X86_EVENT_DYNAMIC 0x0040 /* dynamic alloc'd constraint */
-#define PERF_X86_EVENT_RDPMC_ALLOWED 0x0080 /* grant rdpmc permission */
+
#define PERF_X86_EVENT_EXCL_ACCT 0x0100 /* accounted EXCL event */
#define PERF_X86_EVENT_AUTO_RELOAD 0x0200 /* use PEBS auto-reload */
#define PERF_X86_EVENT_LARGE_PEBS 0x0400 /* use large PEBS */
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 0dcfd265beed..ba9467972c09 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -129,6 +129,15 @@ struct hw_perf_event_extra {
int idx; /* index in shared_regs->regs[] */
};
+/**
+ * hw_perf_event::flag values
+ *
+ * PERF_EVENT_FLAG_ARCH bits are reserved for architecture-specific
+ * usage.
+ */
+#define PERF_EVENT_FLAG_ARCH 0x0000ffff
+#define PERF_EVENT_FLAG_USER_READ_CNT 0x80000000
+
/**
* struct hw_perf_event - performance event hardware details:
*/
--
2.32.0
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH v13 2/5] perf: Add a counter for number of user access events in context
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
2021-12-08 20:11 ` [PATCH v13 1/5] x86: perf: Move RDPMC event flag to a common definition Rob Herring
@ 2021-12-08 20:11 ` Rob Herring
2021-12-08 20:11 ` [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch Rob Herring
` (3 subsequent siblings)
5 siblings, 0 replies; 11+ messages in thread
From: Rob Herring @ 2021-12-08 20:11 UTC (permalink / raw)
To: Will Deacon, Mark Rutland, Peter Zijlstra
Cc: Vince Weaver, Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov, x86,
H. Peter Anvin, linux-kernel, linux-arm-kernel, linux-perf-users
On arm64, user space counter access will be controlled differently
compared to x86. On x86, access in the strictest mode is enabled for all
tasks in an MM when any event is mmap'ed. For arm64, access is
explicitly requested for an event and only enabled when the event's
context is active. This avoids hooks into the arch context switch code
and gives better control of when access is enabled.
In order to configure user space access when the PMU is enabled, it is
necessary to know if any event (currently active or not) in the current
context has user space accessed enabled. Add a counter similar to other
counters in the context to avoid walking the event list every time.
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Rob Herring <robh@kernel.org>
---
v13:
- Rewrote commit message to give more background/reasoning
v10:
- Re-added.
- Maintain the count in the perf core
v9:
- Dropped
v8:
- new patch
---
include/linux/perf_event.h | 1 +
kernel/events/core.c | 4 ++++
2 files changed, 5 insertions(+)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index ba9467972c09..411e34210fbf 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -831,6 +831,7 @@ struct perf_event_context {
int nr_events;
int nr_active;
+ int nr_user;
int is_active;
int nr_stat;
int nr_freq;
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 523106a506ee..68952c134968 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1808,6 +1808,8 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx)
list_add_rcu(&event->event_entry, &ctx->event_list);
ctx->nr_events++;
+ if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT)
+ ctx->nr_user++;
if (event->attr.inherit_stat)
ctx->nr_stat++;
@@ -1999,6 +2001,8 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx)
event->attach_state &= ~PERF_ATTACH_CONTEXT;
ctx->nr_events--;
+ if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT)
+ ctx->nr_user--;
if (event->attr.inherit_stat)
ctx->nr_stat--;
--
2.32.0
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
2021-12-08 20:11 ` [PATCH v13 1/5] x86: perf: Move RDPMC event flag to a common definition Rob Herring
2021-12-08 20:11 ` [PATCH v13 2/5] perf: Add a counter for number of user access events in context Rob Herring
@ 2021-12-08 20:11 ` Rob Herring
2021-12-28 11:07 ` Geert Uytterhoeven
2021-12-08 20:11 ` [PATCH v13 4/5] arm64: perf: Enable PMU counter userspace access for perf event Rob Herring
` (2 subsequent siblings)
5 siblings, 1 reply; 11+ messages in thread
From: Rob Herring @ 2021-12-08 20:11 UTC (permalink / raw)
To: Will Deacon, Mark Rutland, Peter Zijlstra
Cc: Vince Weaver, Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov, x86,
H. Peter Anvin, linux-kernel, linux-arm-kernel, linux-perf-users
Like x86, some users may want to disable userspace PMU counter
altogether. Add a sysctl 'perf_user_access' file to control userspace
counter access. The default is '0' which is disabled. Writing '1'
enables access.
Note that x86 supports globally enabling user access by writing '2' to
/sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
userspace support to worry about, this shouldn't be necessary for Arm.
It could be added later if the need arises.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-perf-users@vger.kernel.org
Acked-by: Will Deacon <will@kernel.org>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Rob Herring <robh@kernel.org>
---
v11:
- Move custom handler to next patch
v10:
- Add documentation
- Use a custom handler (needed on the next patch)
v9:
- Use sysctl instead of sysfs attr
- Default to disabled
v8:
- New patch
---
Documentation/admin-guide/sysctl/kernel.rst | 11 +++++++++++
arch/arm64/kernel/perf_event.c | 17 +++++++++++++++++
2 files changed, 28 insertions(+)
diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
index 426162009ce9..346a0dba5703 100644
--- a/Documentation/admin-guide/sysctl/kernel.rst
+++ b/Documentation/admin-guide/sysctl/kernel.rst
@@ -905,6 +905,17 @@ enabled, otherwise writing to this file will return ``-EBUSY``.
The default value is 8.
+perf_user_access (arm64 only)
+=================================
+
+Controls user space access for reading perf event counters. When set to 1,
+user space can read performance monitor counter registers directly.
+
+The default value is 0 (access disabled).
+
+See Documentation/arm64/perf.rst for more information.
+
+
pid_max
=======
diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index b4044469527e..6ae20c4217af 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -286,6 +286,8 @@ static const struct attribute_group armv8_pmuv3_events_attr_group = {
PMU_FORMAT_ATTR(event, "config:0-15");
PMU_FORMAT_ATTR(long, "config1:0");
+static int sysctl_perf_user_access __read_mostly;
+
static inline bool armv8pmu_event_is_64bit(struct perf_event *event)
{
return event->attr.config1 & 0x1;
@@ -1104,6 +1106,19 @@ static int armv8pmu_probe_pmu(struct arm_pmu *cpu_pmu)
return probe.present ? 0 : -ENODEV;
}
+static struct ctl_table armv8_pmu_sysctl_table[] = {
+ {
+ .procname = "perf_user_access",
+ .data = &sysctl_perf_user_access,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_ONE,
+ },
+ { }
+};
+
static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
int (*map_event)(struct perf_event *event),
const struct attribute_group *events,
@@ -1136,6 +1151,8 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_CAPS] = caps ?
caps : &armv8_pmuv3_caps_attr_group;
+ register_sysctl("kernel", armv8_pmu_sysctl_table);
+
return 0;
}
--
2.32.0
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch
2021-12-08 20:11 ` [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch Rob Herring
@ 2021-12-28 11:07 ` Geert Uytterhoeven
2022-01-04 13:56 ` Will Deacon
0 siblings, 1 reply; 11+ messages in thread
From: Geert Uytterhoeven @ 2021-12-28 11:07 UTC (permalink / raw)
To: Rob Herring
Cc: Will Deacon, Mark Rutland, Peter Zijlstra, Vince Weaver,
Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov,
the arch/x86 maintainers, H. Peter Anvin,
Linux Kernel Mailing List, Linux ARM, linux-perf-users,
Linux-Renesas
Hi Rob,
On Wed, Dec 8, 2021 at 9:19 PM Rob Herring <robh@kernel.org> wrote:
> Like x86, some users may want to disable userspace PMU counter
> altogether. Add a sysctl 'perf_user_access' file to control userspace
> counter access. The default is '0' which is disabled. Writing '1'
> enables access.
>
> Note that x86 supports globally enabling user access by writing '2' to
> /sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
> userspace support to worry about, this shouldn't be necessary for Arm.
> It could be added later if the need arises.
Thanks for your patch, which is now commit e2012600810c9ded ("arm64:
perf: Add userspace counter access disable switch") in arm64/for-next/core.
This is causing two issues on Renesas Salvator-XS with R-Car H3.
One during kernel boot:
hw perfevents: enabled with armv8_cortex_a53 PMU driver, 7
counters available
+sysctl duplicate entry: /kernel//perf_user_access
+CPU: 0 PID: 1 Comm: swapper/0 Not tainted
5.16.0-rc3-arm64-renesas-00003-ge2012600810c #1420
+Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
+Call trace:
+ dump_backtrace+0x0/0x190
+ show_stack+0x14/0x20
+ dump_stack_lvl+0x88/0xb0
+ dump_stack+0x14/0x2c
+ __register_sysctl_table+0x384/0x818
+ register_sysctl+0x20/0x28
+ armv8_pmu_init.constprop.0+0x118/0x150
+ armv8_a57_pmu_init+0x1c/0x28
+ arm_pmu_device_probe+0x1b4/0x558
+ armv8_pmu_device_probe+0x18/0x20
+ platform_probe+0x64/0xd0
+ really_probe+0xb4/0x2f8
+ __driver_probe_device+0x74/0xd8
+ driver_probe_device+0x3c/0xe0
+ __driver_attach+0x80/0x110
+ bus_for_each_dev+0x6c/0xc0
+ driver_attach+0x20/0x28
+ bus_add_driver+0x138/0x1e0
+ driver_register+0x60/0x110
+ __platform_driver_register+0x24/0x30
+ armv8_pmu_driver_init+0x18/0x20
+ do_one_initcall+0x15c/0x31c
+ kernel_init_freeable+0x2f0/0x354
+ kernel_init+0x20/0x120
+ ret_from_fork+0x10/0x20
hw perfevents: enabled with armv8_cortex_a57 PMU driver, 7
counters available
Presumably the same entry is added twice, once for the A53 PMU,
and a second time for the A57 PMU?
A second during systemd startup:
systemd-journald[326]: Failed to open runtime journal: No such
file or directory
systemd-journald[345]: File
/run/log/journal/09223238c0464b38ad4fc1d505d98e17/system.journal
corrupted or uncleanly shut down, renaming and replacing.
followed by lots of
systemd[<n>]: <foo>.service: Failed to connect stdout to the
journal socket, ignoring: Connection refused
failures.
> --- a/arch/arm64/kernel/perf_event.c
> +++ b/arch/arm64/kernel/perf_event.c
> @@ -286,6 +286,8 @@ static const struct attribute_group armv8_pmuv3_events_attr_group = {
> PMU_FORMAT_ATTR(event, "config:0-15");
> PMU_FORMAT_ATTR(long, "config1:0");
>
> +static int sysctl_perf_user_access __read_mostly;
> +
> static inline bool armv8pmu_event_is_64bit(struct perf_event *event)
> {
> return event->attr.config1 & 0x1;
> @@ -1104,6 +1106,19 @@ static int armv8pmu_probe_pmu(struct arm_pmu *cpu_pmu)
> return probe.present ? 0 : -ENODEV;
> }
>
> +static struct ctl_table armv8_pmu_sysctl_table[] = {
> + {
> + .procname = "perf_user_access",
> + .data = &sysctl_perf_user_access,
> + .maxlen = sizeof(unsigned int),
> + .mode = 0644,
> + .proc_handler = proc_dointvec_minmax,
> + .extra1 = SYSCTL_ZERO,
> + .extra2 = SYSCTL_ONE,
> + },
> + { }
> +};
> +
> static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
> int (*map_event)(struct perf_event *event),
> const struct attribute_group *events,
> @@ -1136,6 +1151,8 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
> cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_CAPS] = caps ?
> caps : &armv8_pmuv3_caps_attr_group;
>
> + register_sysctl("kernel", armv8_pmu_sysctl_table);
> +
> return 0;
> }
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch
2021-12-28 11:07 ` Geert Uytterhoeven
@ 2022-01-04 13:56 ` Will Deacon
2022-01-05 11:25 ` Will Deacon
2022-01-07 10:22 ` Geert Uytterhoeven
0 siblings, 2 replies; 11+ messages in thread
From: Will Deacon @ 2022-01-04 13:56 UTC (permalink / raw)
To: Geert Uytterhoeven
Cc: Rob Herring, Mark Rutland, Peter Zijlstra, Vince Weaver,
Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov,
the arch/x86 maintainers, H. Peter Anvin,
Linux Kernel Mailing List, Linux ARM, linux-perf-users,
Linux-Renesas
Hi Geert,
On Tue, Dec 28, 2021 at 12:07:02PM +0100, Geert Uytterhoeven wrote:
> On Wed, Dec 8, 2021 at 9:19 PM Rob Herring <robh@kernel.org> wrote:
> > Like x86, some users may want to disable userspace PMU counter
> > altogether. Add a sysctl 'perf_user_access' file to control userspace
> > counter access. The default is '0' which is disabled. Writing '1'
> > enables access.
> >
> > Note that x86 supports globally enabling user access by writing '2' to
> > /sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
> > userspace support to worry about, this shouldn't be necessary for Arm.
> > It could be added later if the need arises.
>
> Thanks for your patch, which is now commit e2012600810c9ded ("arm64:
> perf: Add userspace counter access disable switch") in arm64/for-next/core.
>
> This is causing two issues on Renesas Salvator-XS with R-Car H3.
> One during kernel boot:
>
> hw perfevents: enabled with armv8_cortex_a53 PMU driver, 7
> counters available
> +sysctl duplicate entry: /kernel//perf_user_access
> +CPU: 0 PID: 1 Comm: swapper/0 Not tainted
> 5.16.0-rc3-arm64-renesas-00003-ge2012600810c #1420
> +Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
> +Call trace:
> + dump_backtrace+0x0/0x190
> + show_stack+0x14/0x20
> + dump_stack_lvl+0x88/0xb0
> + dump_stack+0x14/0x2c
> + __register_sysctl_table+0x384/0x818
> + register_sysctl+0x20/0x28
> + armv8_pmu_init.constprop.0+0x118/0x150
> + armv8_a57_pmu_init+0x1c/0x28
> + arm_pmu_device_probe+0x1b4/0x558
> + armv8_pmu_device_probe+0x18/0x20
> + platform_probe+0x64/0xd0
> + really_probe+0xb4/0x2f8
> + __driver_probe_device+0x74/0xd8
> + driver_probe_device+0x3c/0xe0
> + __driver_attach+0x80/0x110
> + bus_for_each_dev+0x6c/0xc0
> + driver_attach+0x20/0x28
> + bus_add_driver+0x138/0x1e0
> + driver_register+0x60/0x110
> + __platform_driver_register+0x24/0x30
> + armv8_pmu_driver_init+0x18/0x20
> + do_one_initcall+0x15c/0x31c
> + kernel_init_freeable+0x2f0/0x354
> + kernel_init+0x20/0x120
> + ret_from_fork+0x10/0x20
> hw perfevents: enabled with armv8_cortex_a57 PMU driver, 7
> counters available
>
> Presumably the same entry is added twice, once for the A53 PMU,
> and a second time for the A57 PMU?
Looks like it, and perhaps that's also what is confusing systemd?
Rob -- how come you didn't see this during your testing?
Anywho, please can you try the untested diff below?
Thanks,
Will
--->8
diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index 81cc9f0e718a..639f632aaa66 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -1214,6 +1214,14 @@ static struct ctl_table armv8_pmu_sysctl_table[] = {
{ }
};
+static void armv8_pmu_register_sysctl_table(void)
+{
+ static u32 tbl_registered = 0;
+
+ if (!cmpxchg_relaxed(&tbl_registered, 0, 1))
+ register_sysctl("kernel", armv8_pmu_sysctl_table);
+}
+
static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
int (*map_event)(struct perf_event *event),
const struct attribute_group *events,
@@ -1248,8 +1256,7 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_CAPS] = caps ?
caps : &armv8_pmuv3_caps_attr_group;
- register_sysctl("kernel", armv8_pmu_sysctl_table);
-
+ armv8_pmu_register_sysctl_table();
return 0;
}
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch
2022-01-04 13:56 ` Will Deacon
@ 2022-01-05 11:25 ` Will Deacon
2022-01-07 10:22 ` Geert Uytterhoeven
1 sibling, 0 replies; 11+ messages in thread
From: Will Deacon @ 2022-01-05 11:25 UTC (permalink / raw)
To: Geert Uytterhoeven
Cc: Rob Herring, Mark Rutland, Peter Zijlstra, Vince Weaver,
Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov,
the arch/x86 maintainers, H. Peter Anvin,
Linux Kernel Mailing List, Linux ARM, linux-perf-users,
Linux-Renesas
On Tue, Jan 04, 2022 at 01:56:59PM +0000, Will Deacon wrote:
> On Tue, Dec 28, 2021 at 12:07:02PM +0100, Geert Uytterhoeven wrote:
> > On Wed, Dec 8, 2021 at 9:19 PM Rob Herring <robh@kernel.org> wrote:
> > > Like x86, some users may want to disable userspace PMU counter
> > > altogether. Add a sysctl 'perf_user_access' file to control userspace
> > > counter access. The default is '0' which is disabled. Writing '1'
> > > enables access.
> > >
> > > Note that x86 supports globally enabling user access by writing '2' to
> > > /sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
> > > userspace support to worry about, this shouldn't be necessary for Arm.
> > > It could be added later if the need arises.
> >
> > Thanks for your patch, which is now commit e2012600810c9ded ("arm64:
> > perf: Add userspace counter access disable switch") in arm64/for-next/core.
> >
> > This is causing two issues on Renesas Salvator-XS with R-Car H3.
> > One during kernel boot:
> >
> > hw perfevents: enabled with armv8_cortex_a53 PMU driver, 7
> > counters available
> > +sysctl duplicate entry: /kernel//perf_user_access
> > +CPU: 0 PID: 1 Comm: swapper/0 Not tainted
> > 5.16.0-rc3-arm64-renesas-00003-ge2012600810c #1420
> > +Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
> > +Call trace:
> > + dump_backtrace+0x0/0x190
> > + show_stack+0x14/0x20
> > + dump_stack_lvl+0x88/0xb0
> > + dump_stack+0x14/0x2c
> > + __register_sysctl_table+0x384/0x818
> > + register_sysctl+0x20/0x28
> > + armv8_pmu_init.constprop.0+0x118/0x150
> > + armv8_a57_pmu_init+0x1c/0x28
> > + arm_pmu_device_probe+0x1b4/0x558
> > + armv8_pmu_device_probe+0x18/0x20
> > + platform_probe+0x64/0xd0
> > + really_probe+0xb4/0x2f8
> > + __driver_probe_device+0x74/0xd8
> > + driver_probe_device+0x3c/0xe0
> > + __driver_attach+0x80/0x110
> > + bus_for_each_dev+0x6c/0xc0
> > + driver_attach+0x20/0x28
> > + bus_add_driver+0x138/0x1e0
> > + driver_register+0x60/0x110
> > + __platform_driver_register+0x24/0x30
> > + armv8_pmu_driver_init+0x18/0x20
> > + do_one_initcall+0x15c/0x31c
> > + kernel_init_freeable+0x2f0/0x354
> > + kernel_init+0x20/0x120
> > + ret_from_fork+0x10/0x20
> > hw perfevents: enabled with armv8_cortex_a57 PMU driver, 7
> > counters available
> >
> > Presumably the same entry is added twice, once for the A53 PMU,
> > and a second time for the A57 PMU?
>
> Looks like it, and perhaps that's also what is confusing systemd?
> Rob -- how come you didn't see this during your testing?
>
> Anywho, please can you try the untested diff below?
I just remembered I have a big/little SoC on my desk after borrowing a
NanoPi (RK3399) from Marc Z, so I took this diff for a spin there and
both the kernel and systemd seem happy.
Will
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch
2022-01-04 13:56 ` Will Deacon
2022-01-05 11:25 ` Will Deacon
@ 2022-01-07 10:22 ` Geert Uytterhoeven
1 sibling, 0 replies; 11+ messages in thread
From: Geert Uytterhoeven @ 2022-01-07 10:22 UTC (permalink / raw)
To: Will Deacon
Cc: Rob Herring, Mark Rutland, Peter Zijlstra, Vince Weaver,
Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov,
the arch/x86 maintainers, H. Peter Anvin,
Linux Kernel Mailing List, Linux ARM, linux-perf-users,
Linux-Renesas
Hi Will,
On Tue, Jan 4, 2022 at 2:57 PM Will Deacon <will@kernel.org> wrote:
> On Tue, Dec 28, 2021 at 12:07:02PM +0100, Geert Uytterhoeven wrote:
> > On Wed, Dec 8, 2021 at 9:19 PM Rob Herring <robh@kernel.org> wrote:
> > > Like x86, some users may want to disable userspace PMU counter
> > > altogether. Add a sysctl 'perf_user_access' file to control userspace
> > > counter access. The default is '0' which is disabled. Writing '1'
> > > enables access.
> > >
> > > Note that x86 supports globally enabling user access by writing '2' to
> > > /sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
> > > userspace support to worry about, this shouldn't be necessary for Arm.
> > > It could be added later if the need arises.
> >
> > Thanks for your patch, which is now commit e2012600810c9ded ("arm64:
> > perf: Add userspace counter access disable switch") in arm64/for-next/core.
> >
> > This is causing two issues on Renesas Salvator-XS with R-Car H3.
> > One during kernel boot:
> >
> > hw perfevents: enabled with armv8_cortex_a53 PMU driver, 7
> > counters available
> > +sysctl duplicate entry: /kernel//perf_user_access
> > +CPU: 0 PID: 1 Comm: swapper/0 Not tainted
> > 5.16.0-rc3-arm64-renesas-00003-ge2012600810c #1420
> > +Hardware name: Renesas Salvator-X 2nd version board based on r8a77951 (DT)
> > +Call trace:
> > + dump_backtrace+0x0/0x190
> > + show_stack+0x14/0x20
> > + dump_stack_lvl+0x88/0xb0
> > + dump_stack+0x14/0x2c
> > + __register_sysctl_table+0x384/0x818
> > + register_sysctl+0x20/0x28
> > + armv8_pmu_init.constprop.0+0x118/0x150
> > + armv8_a57_pmu_init+0x1c/0x28
> > + arm_pmu_device_probe+0x1b4/0x558
> > + armv8_pmu_device_probe+0x18/0x20
> > + platform_probe+0x64/0xd0
> > + really_probe+0xb4/0x2f8
> > + __driver_probe_device+0x74/0xd8
> > + driver_probe_device+0x3c/0xe0
> > + __driver_attach+0x80/0x110
> > + bus_for_each_dev+0x6c/0xc0
> > + driver_attach+0x20/0x28
> > + bus_add_driver+0x138/0x1e0
> > + driver_register+0x60/0x110
> > + __platform_driver_register+0x24/0x30
> > + armv8_pmu_driver_init+0x18/0x20
> > + do_one_initcall+0x15c/0x31c
> > + kernel_init_freeable+0x2f0/0x354
> > + kernel_init+0x20/0x120
> > + ret_from_fork+0x10/0x20
> > hw perfevents: enabled with armv8_cortex_a57 PMU driver, 7
> > counters available
> >
> > Presumably the same entry is added twice, once for the A53 PMU,
> > and a second time for the A57 PMU?
>
> Looks like it, and perhaps that's also what is confusing systemd?
> Rob -- how come you didn't see this during your testing?
>
> Anywho, please can you try the untested diff below?
Thank you, this fixes the issue.
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
> --- a/arch/arm64/kernel/perf_event.c
> +++ b/arch/arm64/kernel/perf_event.c
> @@ -1214,6 +1214,14 @@ static struct ctl_table armv8_pmu_sysctl_table[] = {
> { }
> };
>
> +static void armv8_pmu_register_sysctl_table(void)
> +{
> + static u32 tbl_registered = 0;
> +
> + if (!cmpxchg_relaxed(&tbl_registered, 0, 1))
> + register_sysctl("kernel", armv8_pmu_sysctl_table);
> +}
> +
> static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
> int (*map_event)(struct perf_event *event),
> const struct attribute_group *events,
> @@ -1248,8 +1256,7 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
> cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_CAPS] = caps ?
> caps : &armv8_pmuv3_caps_attr_group;
>
> - register_sysctl("kernel", armv8_pmu_sysctl_table);
> -
> + armv8_pmu_register_sysctl_table();
> return 0;
> }
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply [flat|nested] 11+ messages in thread
* [PATCH v13 4/5] arm64: perf: Enable PMU counter userspace access for perf event
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
` (2 preceding siblings ...)
2021-12-08 20:11 ` [PATCH v13 3/5] arm64: perf: Add userspace counter access disable switch Rob Herring
@ 2021-12-08 20:11 ` Rob Herring
2021-12-08 20:11 ` [PATCH v13 5/5] Documentation: arm64: Document PMU counters access from userspace Rob Herring
2021-12-14 14:04 ` [PATCH v13 0/5] arm64 userspace counter support Will Deacon
5 siblings, 0 replies; 11+ messages in thread
From: Rob Herring @ 2021-12-08 20:11 UTC (permalink / raw)
To: Will Deacon, Mark Rutland, Peter Zijlstra
Cc: Vince Weaver, Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov, x86,
H. Peter Anvin, linux-kernel, linux-arm-kernel, linux-perf-users
Arm PMUs can support direct userspace access of counters which allows for
low overhead (i.e. no syscall) self-monitoring of tasks. The same feature
exists on x86 called 'rdpmc'. Unlike x86, userspace access will only be
enabled for thread bound events. This could be extended if needed, but
simplifies the implementation and reduces the chances for any
information leaks (which the x86 implementation suffers from).
PMU EL0 access will be enabled when an event with userspace access is
part of the thread's context. This includes when the event is not
scheduled on the PMU. There's some additional overhead clearing
dirty counters when access is enabled in order to prevent leaking
disabled counter data from other tasks.
Unlike x86, enabling of userspace access must be requested with a new
attr bit: config1:1. If the user requests userspace access with 64-bit
counters, then the event open will fail if the h/w doesn't support
64-bit counters. Chaining is not supported with userspace access. The
modes for config1 are as follows:
config1 = 0 : user access disabled and always 32-bit
config1 = 1 : user access disabled and always 64-bit (using chaining if needed)
config1 = 2 : user access enabled and always 32-bit
config1 = 3 : user access enabled and always 64-bit
Based on work by Raphael Gault <raphael.gault@arm.com>, but has been
completely re-written.
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-perf-users@vger.kernel.org
Signed-off-by: Rob Herring <robh@kernel.org>
---
v12:
- Clear pmuserenr_el0 when enabling user access
- Return -EOPNOTSUPP instead of -EINVAL if h/w doesn't have 64-bit
counters
v11:
- Add and use armv8pmu_event_has_user_read() helper
- s/armv8pmu_access_event_idx/armv8pmu_user_event_idx/
- Return error for user access when not a task bound event or no
64-bit counters when requested.
- Move custom sysctl handler function from prior patch to here
v10:
- Don't control enabling user access based on mmap(). Changing the
event_(un)mapped to run on the event's cpu doesn't work for x86.
Triggering on mmap() doesn't limit access in any way and complicates
the implementation.
- Drop dirty counter tracking and just clear all unused counters.
- Make the sysctl immediately disable access via IPI.
- Merge armv8pmu_event_is_chained() and armv8pmu_event_can_chain()
v9:
- Enabling/disabling of user access is now controlled in .start() and
mmap hooks which are now called on CPUs that the event is on.
Depends on rework of perf core and x86 RDPMC code posted here:
https://lore.kernel.org/lkml/20210728230230.1911468-1-robh@kernel.org/
v8:
- Rework user access tracking and enabling to be done on task
context changes using sched_task() hook. This avoids the need for any
IPIs, mm_switch hooks or undef instr handler.
- Only support user access when explicitly requested on open and
only for a thread bound events. This avoids some of the information
leaks x86 has and simplifies the implementation.
v7:
- Clear disabled counters when user access is enabled for a task to
avoid leaking other tasks counter data.
- Rework context switch handling utilizing sched_task callback
- Add armv8pmu_event_can_chain() helper
- Rework config1 flags handling structure
- Use ARMV8_IDX_CYCLE_COUNTER_USER define for remapped user cycle
counter index
v6:
- Add new attr.config1 rdpmc bit for userspace to hint it wants
userspace access when also requesting 64-bit counters.
v5:
- Only set cap_user_rdpmc if event is on current cpu
- Limit enabling/disabling access to CPUs associated with the PMU
(supported_cpus) and with the mm_struct matching current->active_mm.
v2:
- Move mapped/unmapped into arm64 code. Fixes arm32.
- Rebase on cap_user_time_short changes
Changes from Raphael's v4:
- Drop homogeneous check
- Disable access for chained counters
- Set pmc_width in user page
---
arch/arm64/kernel/perf_event.c | 119 +++++++++++++++++++++++++++++++--
1 file changed, 112 insertions(+), 7 deletions(-)
diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index 6ae20c4217af..028d9d3aadab 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -285,6 +285,7 @@ static const struct attribute_group armv8_pmuv3_events_attr_group = {
PMU_FORMAT_ATTR(event, "config:0-15");
PMU_FORMAT_ATTR(long, "config1:0");
+PMU_FORMAT_ATTR(rdpmc, "config1:1");
static int sysctl_perf_user_access __read_mostly;
@@ -293,9 +294,15 @@ static inline bool armv8pmu_event_is_64bit(struct perf_event *event)
return event->attr.config1 & 0x1;
}
+static inline bool armv8pmu_event_want_user_access(struct perf_event *event)
+{
+ return event->attr.config1 & 0x2;
+}
+
static struct attribute *armv8_pmuv3_format_attrs[] = {
&format_attr_event.attr,
&format_attr_long.attr,
+ &format_attr_rdpmc.attr,
NULL,
};
@@ -364,7 +371,7 @@ static const struct attribute_group armv8_pmuv3_caps_attr_group = {
*/
#define ARMV8_IDX_CYCLE_COUNTER 0
#define ARMV8_IDX_COUNTER0 1
-
+#define ARMV8_IDX_CYCLE_COUNTER_USER 32
/*
* We unconditionally enable ARMv8.5-PMU long event counter support
@@ -376,18 +383,22 @@ static bool armv8pmu_has_long_event(struct arm_pmu *cpu_pmu)
return (cpu_pmu->pmuver >= ID_AA64DFR0_PMUVER_8_5);
}
+static inline bool armv8pmu_event_has_user_read(struct perf_event *event)
+{
+ return event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT;
+}
+
/*
* We must chain two programmable counters for 64 bit events,
* except when we have allocated the 64bit cycle counter (for CPU
- * cycles event). This must be called only when the event has
- * a counter allocated.
+ * cycles event) or when user space counter access is enabled.
*/
static inline bool armv8pmu_event_is_chained(struct perf_event *event)
{
int idx = event->hw.idx;
struct arm_pmu *cpu_pmu = to_arm_pmu(event->pmu);
- return !WARN_ON(idx < 0) &&
+ return !armv8pmu_event_has_user_read(event) &&
armv8pmu_event_is_64bit(event) &&
!armv8pmu_has_long_event(cpu_pmu) &&
(idx != ARMV8_IDX_CYCLE_COUNTER);
@@ -720,6 +731,28 @@ static inline u32 armv8pmu_getreset_flags(void)
return value;
}
+static void armv8pmu_disable_user_access(void)
+{
+ write_sysreg(0, pmuserenr_el0);
+}
+
+static void armv8pmu_enable_user_access(struct arm_pmu *cpu_pmu)
+{
+ int i;
+ struct pmu_hw_events *cpuc = this_cpu_ptr(cpu_pmu->hw_events);
+
+ /* Clear any unused counters to avoid leaking their contents */
+ for_each_clear_bit(i, cpuc->used_mask, cpu_pmu->num_events) {
+ if (i == ARMV8_IDX_CYCLE_COUNTER)
+ write_sysreg(0, pmccntr_el0);
+ else
+ armv8pmu_write_evcntr(i, 0);
+ }
+
+ write_sysreg(0, pmuserenr_el0);
+ write_sysreg(ARMV8_PMU_USERENR_ER | ARMV8_PMU_USERENR_CR, pmuserenr_el0);
+}
+
static void armv8pmu_enable_event(struct perf_event *event)
{
/*
@@ -763,6 +796,14 @@ static void armv8pmu_disable_event(struct perf_event *event)
static void armv8pmu_start(struct arm_pmu *cpu_pmu)
{
+ struct perf_event_context *task_ctx =
+ this_cpu_ptr(cpu_pmu->pmu.pmu_cpu_context)->task_ctx;
+
+ if (sysctl_perf_user_access && task_ctx && task_ctx->nr_user)
+ armv8pmu_enable_user_access(cpu_pmu);
+ else
+ armv8pmu_disable_user_access();
+
/* Enable all counters */
armv8pmu_pmcr_write(armv8pmu_pmcr_read() | ARMV8_PMU_PMCR_E);
}
@@ -880,13 +921,16 @@ static int armv8pmu_get_event_idx(struct pmu_hw_events *cpuc,
if (evtype == ARMV8_PMUV3_PERFCTR_CPU_CYCLES) {
if (!test_and_set_bit(ARMV8_IDX_CYCLE_COUNTER, cpuc->used_mask))
return ARMV8_IDX_CYCLE_COUNTER;
+ else if (armv8pmu_event_is_64bit(event) &&
+ armv8pmu_event_want_user_access(event) &&
+ !armv8pmu_has_long_event(cpu_pmu))
+ return -EAGAIN;
}
/*
* Otherwise use events counters
*/
- if (armv8pmu_event_is_64bit(event) &&
- !armv8pmu_has_long_event(cpu_pmu))
+ if (armv8pmu_event_is_chained(event))
return armv8pmu_get_chain_idx(cpuc, cpu_pmu);
else
return armv8pmu_get_single_idx(cpuc, cpu_pmu);
@@ -902,6 +946,22 @@ static void armv8pmu_clear_event_idx(struct pmu_hw_events *cpuc,
clear_bit(idx - 1, cpuc->used_mask);
}
+static int armv8pmu_user_event_idx(struct perf_event *event)
+{
+ if (!sysctl_perf_user_access || !armv8pmu_event_has_user_read(event))
+ return 0;
+
+ /*
+ * We remap the cycle counter index to 32 to
+ * match the offset applied to the rest of
+ * the counter indices.
+ */
+ if (event->hw.idx == ARMV8_IDX_CYCLE_COUNTER)
+ return ARMV8_IDX_CYCLE_COUNTER_USER;
+
+ return event->hw.idx;
+}
+
/*
* Add an event filter to a given event.
*/
@@ -998,6 +1058,25 @@ static int __armv8_pmuv3_map_event(struct perf_event *event,
if (armv8pmu_event_is_64bit(event))
event->hw.flags |= ARMPMU_EVT_64BIT;
+ /*
+ * User events must be allocated into a single counter, and so
+ * must not be chained.
+ *
+ * Most 64-bit events require long counter support, but 64-bit
+ * CPU_CYCLES events can be placed into the dedicated cycle
+ * counter when this is free.
+ */
+ if (armv8pmu_event_want_user_access(event)) {
+ if (!(event->attach_state & PERF_ATTACH_TASK))
+ return -EINVAL;
+ if (armv8pmu_event_is_64bit(event) &&
+ (hw_event_id != ARMV8_PMUV3_PERFCTR_CPU_CYCLES) &&
+ !armv8pmu_has_long_event(armpmu))
+ return -EOPNOTSUPP;
+
+ event->hw.flags |= PERF_EVENT_FLAG_USER_READ_CNT;
+ }
+
/* Only expose micro/arch events supported by this PMU */
if ((hw_event_id > 0) && (hw_event_id < ARMV8_PMUV3_MAX_COMMON_EVENTS)
&& test_bit(hw_event_id, armpmu->pmceid_bitmap)) {
@@ -1106,13 +1185,29 @@ static int armv8pmu_probe_pmu(struct arm_pmu *cpu_pmu)
return probe.present ? 0 : -ENODEV;
}
+static void armv8pmu_disable_user_access_ipi(void *unused)
+{
+ armv8pmu_disable_user_access();
+}
+
+int armv8pmu_proc_user_access_handler(struct ctl_table *table, int write,
+ void *buffer, size_t *lenp, loff_t *ppos)
+{
+ int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
+ if (ret || !write || sysctl_perf_user_access)
+ return ret;
+
+ on_each_cpu(armv8pmu_disable_user_access_ipi, NULL, 1);
+ return 0;
+}
+
static struct ctl_table armv8_pmu_sysctl_table[] = {
{
.procname = "perf_user_access",
.data = &sysctl_perf_user_access,
.maxlen = sizeof(unsigned int),
.mode = 0644,
- .proc_handler = proc_dointvec_minmax,
+ .proc_handler = armv8pmu_proc_user_access_handler,
.extra1 = SYSCTL_ZERO,
.extra2 = SYSCTL_ONE,
},
@@ -1142,6 +1237,8 @@ static int armv8_pmu_init(struct arm_pmu *cpu_pmu, char *name,
cpu_pmu->set_event_filter = armv8pmu_set_event_filter;
cpu_pmu->filter_match = armv8pmu_filter_match;
+ cpu_pmu->pmu.event_idx = armv8pmu_user_event_idx;
+
cpu_pmu->name = name;
cpu_pmu->map_event = map_event;
cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_EVENTS] = events ?
@@ -1318,6 +1415,14 @@ void arch_perf_update_userpage(struct perf_event *event,
userpg->cap_user_time = 0;
userpg->cap_user_time_zero = 0;
userpg->cap_user_time_short = 0;
+ userpg->cap_user_rdpmc = armv8pmu_event_has_user_read(event);
+
+ if (userpg->cap_user_rdpmc) {
+ if (event->hw.flags & ARMPMU_EVT_64BIT)
+ userpg->pmc_width = 64;
+ else
+ userpg->pmc_width = 32;
+ }
do {
rd = sched_clock_read_begin(&seq);
--
2.32.0
^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH v13 5/5] Documentation: arm64: Document PMU counters access from userspace
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
` (3 preceding siblings ...)
2021-12-08 20:11 ` [PATCH v13 4/5] arm64: perf: Enable PMU counter userspace access for perf event Rob Herring
@ 2021-12-08 20:11 ` Rob Herring
2021-12-14 14:04 ` [PATCH v13 0/5] arm64 userspace counter support Will Deacon
5 siblings, 0 replies; 11+ messages in thread
From: Rob Herring @ 2021-12-08 20:11 UTC (permalink / raw)
To: Will Deacon, Mark Rutland, Peter Zijlstra
Cc: Vince Weaver, Jonathan Corbet, Catalin Marinas, Ingo Molnar,
Arnaldo Carvalho de Melo, Alexander Shishkin, Jiri Olsa,
Namhyung Kim, Thomas Gleixner, Borislav Petkov, x86,
H. Peter Anvin, linux-kernel, linux-arm-kernel, linux-perf-users,
Raphael Gault
From: Raphael Gault <raphael.gault@arm.com>
Add documentation to describe the access to the pmu hardware counters from
userspace.
Signed-off-by: Raphael Gault <raphael.gault@arm.com>
Signed-off-by: Rob Herring <robh@kernel.org>
---
v12:
- Add note about accessing PMU registers and PMSELR_EL0 being zeroed
v11:
- Rework chained event section to reflect 64-bit counter request may
fail and 32-bit counters may be 64-bit with upper bits UNKNOWN.
v10:
- Add details on perf_user_access sysctl
v9:
- No change
v8:
- Reword that config1:1 must always be set to request user access
v7:
- Merge into existing arm64 perf.rst
v6:
- Update the chained event section with attr.config1 details
v2:
- Update links to test examples
Changes from Raphael's v4:
- Convert to rSt
- Update chained event status
- Add section for heterogeneous systems
---
Documentation/arm64/perf.rst | 78 +++++++++++++++++++++++++++++++++++-
1 file changed, 77 insertions(+), 1 deletion(-)
diff --git a/Documentation/arm64/perf.rst b/Documentation/arm64/perf.rst
index b567f177d385..1f87b57c2332 100644
--- a/Documentation/arm64/perf.rst
+++ b/Documentation/arm64/perf.rst
@@ -2,7 +2,10 @@
.. _perf_index:
-=====================
+====
+Perf
+====
+
Perf Event Attributes
=====================
@@ -88,3 +91,76 @@ exclude_host. However when using !exclude_hv there is a small blackout
window at the guest entry/exit where host events are not captured.
On VHE systems there are no blackout windows.
+
+Perf Userspace PMU Hardware Counter Access
+==========================================
+
+Overview
+--------
+The perf userspace tool relies on the PMU to monitor events. It offers an
+abstraction layer over the hardware counters since the underlying
+implementation is cpu-dependent.
+Arm64 allows userspace tools to have access to the registers storing the
+hardware counters' values directly.
+
+This targets specifically self-monitoring tasks in order to reduce the overhead
+by directly accessing the registers without having to go through the kernel.
+
+How-to
+------
+The focus is set on the armv8 PMUv3 which makes sure that the access to the pmu
+registers is enabled and that the userspace has access to the relevant
+information in order to use them.
+
+In order to have access to the hardware counters, the global sysctl
+kernel/perf_user_access must first be enabled:
+
+.. code-block:: sh
+
+ echo 1 > /proc/sys/kernel/perf_user_access
+
+It is necessary to open the event using the perf tool interface with config1:1
+attr bit set: the sys_perf_event_open syscall returns a fd which can
+subsequently be used with the mmap syscall in order to retrieve a page of memory
+containing information about the event. The PMU driver uses this page to expose
+to the user the hardware counter's index and other necessary data. Using this
+index enables the user to access the PMU registers using the `mrs` instruction.
+Access to the PMU registers is only valid while the sequence lock is unchanged.
+In particular, the PMSELR_EL0 register is zeroed each time the sequence lock is
+changed.
+
+The userspace access is supported in libperf using the perf_evsel__mmap()
+and perf_evsel__read() functions. See `tools/lib/perf/tests/test-evsel.c`_ for
+an example.
+
+About heterogeneous systems
+---------------------------
+On heterogeneous systems such as big.LITTLE, userspace PMU counter access can
+only be enabled when the tasks are pinned to a homogeneous subset of cores and
+the corresponding PMU instance is opened by specifying the 'type' attribute.
+The use of generic event types is not supported in this case.
+
+Have a look at `tools/perf/arch/arm64/tests/user-events.c`_ for an example. It
+can be run using the perf tool to check that the access to the registers works
+correctly from userspace:
+
+.. code-block:: sh
+
+ perf test -v user
+
+About chained events and counter sizes
+--------------------------------------
+The user can request either a 32-bit (config1:0 == 0) or 64-bit (config1:0 == 1)
+counter along with userspace access. The sys_perf_event_open syscall will fail
+if a 64-bit counter is requested and the hardware doesn't support 64-bit
+counters. Chained events are not supported in conjunction with userspace counter
+access. If a 32-bit counter is requested on hardware with 64-bit counters, then
+userspace must treat the upper 32-bits read from the counter as UNKNOWN. The
+'pmc_width' field in the user page will indicate the valid width of the counter
+and should be used to mask the upper bits as needed.
+
+.. Links
+.. _tools/perf/arch/arm64/tests/user-events.c:
+ https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/perf/arch/arm64/tests/user-events.c
+.. _tools/lib/perf/tests/test-evsel.c:
+ https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/lib/perf/tests/test-evsel.c
--
2.32.0
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH v13 0/5] arm64 userspace counter support
2021-12-08 20:11 [PATCH v13 0/5] arm64 userspace counter support Rob Herring
` (4 preceding siblings ...)
2021-12-08 20:11 ` [PATCH v13 5/5] Documentation: arm64: Document PMU counters access from userspace Rob Herring
@ 2021-12-14 14:04 ` Will Deacon
5 siblings, 0 replies; 11+ messages in thread
From: Will Deacon @ 2021-12-14 14:04 UTC (permalink / raw)
To: Mark Rutland, Rob Herring, Peter Zijlstra
Cc: catalin.marinas, kernel-team, Will Deacon, Borislav Petkov,
Vince Weaver, Thomas Gleixner, Ingo Molnar, linux-kernel,
Arnaldo Carvalho de Melo, Jonathan Corbet, x86, H. Peter Anvin,
linux-arm-kernel, Namhyung Kim, Jiri Olsa, Alexander Shishkin,
linux-perf-users
On Wed, 8 Dec 2021 14:11:19 -0600, Rob Herring wrote:
> Another version of arm64 userspace counter access support. No changes from
> the last version except rebasing and rewriting a commit message.
>
> The arm64 support departs from the x86 implementation by requiring the user
> to explicitly request user access (via attr.config1) and only enables access
> for task bound events. Since usage is explicitly requested, access is
> enabled at perf_event_open() rather than on mmap() as that greatly
> simplifies the implementation. Rather than trying to lock down the access
> as the x86 implementation has been doing, we can start with only a limited
> use case enabled and later expand it if needed.
>
> [...]
Applied to will (for-next/perf-user-counter-access), thanks!
[1/5] x86: perf: Move RDPMC event flag to a common definition
https://git.kernel.org/arm64/c/369461ce8fb6
[2/5] perf: Add a counter for number of user access events in context
https://git.kernel.org/arm64/c/82ff0c022d19
[3/5] arm64: perf: Add userspace counter access disable switch
https://git.kernel.org/arm64/c/e2012600810c
[4/5] arm64: perf: Enable PMU counter userspace access for perf event
https://git.kernel.org/arm64/c/83a7a4d643d3
[5/5] Documentation: arm64: Document PMU counters access from userspace
https://git.kernel.org/arm64/c/aa1005d15d2a
Cheers,
--
Will
https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev
^ permalink raw reply [flat|nested] 11+ messages in thread