This small series builds on top of the work that was started with [1].
It recently became apparent that KVM/arm64 is the last bit of the
kernel that still uses perf_num_counters().
As I went ahead to address this, it became obvious that all traces of
oprofile had been eradicated from all architectures but arm64, s390
and sh (plus a bit of cruft in the core perf code). With KVM fixed,
perf_num_counters() and perf_pmu_name() are finally gone.
Thanks,
M.
[1] https://lore.kernel.org/lkml/20210215050618.hgftdmfmslbdrg3j@vireshk-i7
Marc Zyngier (5):
KVM: arm64: Divorce the perf code from oprofile helpers
arm64: Get rid of oprofile leftovers
s390: Get rid of oprofile leftovers
sh: Get rid of oprofile leftovers
perf: Get rid of oprofile leftovers
arch/arm64/kvm/perf.c | 7 +------
arch/arm64/kvm/pmu-emul.c | 2 +-
arch/s390/kernel/perf_event.c | 21 ---------------------
arch/sh/kernel/perf_event.c | 18 ------------------
drivers/perf/arm_pmu.c | 30 ------------------------------
include/kvm/arm_pmu.h | 4 ++++
include/linux/perf_event.h | 2 --
kernel/events/core.c | 5 -----
8 files changed, 6 insertions(+), 83 deletions(-)
--
2.29.2
perf_pmu_name() and perf_num_counters() are unused. Drop them.
Signed-off-by: Marc Zyngier <[email protected]>
---
arch/s390/kernel/perf_event.c | 21 ---------------------
1 file changed, 21 deletions(-)
diff --git a/arch/s390/kernel/perf_event.c b/arch/s390/kernel/perf_event.c
index 1e75cc983546..ea7729bebaa0 100644
--- a/arch/s390/kernel/perf_event.c
+++ b/arch/s390/kernel/perf_event.c
@@ -23,27 +23,6 @@
#include <asm/sysinfo.h>
#include <asm/unwind.h>
-const char *perf_pmu_name(void)
-{
- if (cpum_cf_avail() || cpum_sf_avail())
- return "CPU-Measurement Facilities (CPU-MF)";
- return "pmu";
-}
-EXPORT_SYMBOL(perf_pmu_name);
-
-int perf_num_counters(void)
-{
- int num = 0;
-
- if (cpum_cf_avail())
- num += PERF_CPUM_CF_MAX_CTR;
- if (cpum_sf_avail())
- num += PERF_CPUM_SF_MAX_CTR;
-
- return num;
-}
-EXPORT_SYMBOL(perf_num_counters);
-
static struct kvm_s390_sie_block *sie_block(struct pt_regs *regs)
{
struct stack_frame *stack = (struct stack_frame *) regs->gprs[15];
--
2.29.2
KVM/arm64 is the sole user of perf_num_counters(), and really
could do without it. Stop using the obsolete API by relying on
the existing probing code.
Signed-off-by: Marc Zyngier <[email protected]>
---
arch/arm64/kvm/perf.c | 7 +------
arch/arm64/kvm/pmu-emul.c | 2 +-
include/kvm/arm_pmu.h | 4 ++++
3 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/arch/arm64/kvm/perf.c b/arch/arm64/kvm/perf.c
index 739164324afe..b8b398670ef2 100644
--- a/arch/arm64/kvm/perf.c
+++ b/arch/arm64/kvm/perf.c
@@ -50,12 +50,7 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = {
int kvm_perf_init(void)
{
- /*
- * Check if HW_PERF_EVENTS are supported by checking the number of
- * hardware performance counters. This could ensure the presence of
- * a physical PMU and CONFIG_PERF_EVENT is selected.
- */
- if (IS_ENABLED(CONFIG_ARM_PMU) && perf_num_counters() > 0)
+ if (kvm_pmu_probe_pmuver() != 0xf)
static_branch_enable(&kvm_arm_pmu_available);
return perf_register_guest_info_callbacks(&kvm_guest_cbs);
diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c
index e32c6e139a09..fd167d4f4215 100644
--- a/arch/arm64/kvm/pmu-emul.c
+++ b/arch/arm64/kvm/pmu-emul.c
@@ -739,7 +739,7 @@ void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u64 data,
kvm_pmu_create_perf_event(vcpu, select_idx);
}
-static int kvm_pmu_probe_pmuver(void)
+int kvm_pmu_probe_pmuver(void)
{
struct perf_event_attr attr = { };
struct perf_event *event;
diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
index 6fd3cda608e4..864b9997efb2 100644
--- a/include/kvm/arm_pmu.h
+++ b/include/kvm/arm_pmu.h
@@ -61,6 +61,7 @@ int kvm_arm_pmu_v3_get_attr(struct kvm_vcpu *vcpu,
int kvm_arm_pmu_v3_has_attr(struct kvm_vcpu *vcpu,
struct kvm_device_attr *attr);
int kvm_arm_pmu_v3_enable(struct kvm_vcpu *vcpu);
+int kvm_pmu_probe_pmuver(void);
#else
struct kvm_pmu {
};
@@ -116,6 +117,9 @@ static inline u64 kvm_pmu_get_pmceid(struct kvm_vcpu *vcpu, bool pmceid1)
{
return 0;
}
+
+static inline int kvm_pmu_probe_pmuver(void) { return 0xf; }
+
#endif
#endif
--
2.29.2
perf_pmu_name() and perf_num_counters() are now unused. Drop them.
Signed-off-by: Marc Zyngier <[email protected]>
---
drivers/perf/arm_pmu.c | 30 ------------------------------
1 file changed, 30 deletions(-)
diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
index 2d10d84fb79c..d4f7f1f9cc77 100644
--- a/drivers/perf/arm_pmu.c
+++ b/drivers/perf/arm_pmu.c
@@ -581,33 +581,6 @@ static const struct attribute_group armpmu_common_attr_group = {
.attrs = armpmu_common_attrs,
};
-/* Set at runtime when we know what CPU type we are. */
-static struct arm_pmu *__oprofile_cpu_pmu;
-
-/*
- * Despite the names, these two functions are CPU-specific and are used
- * by the OProfile/perf code.
- */
-const char *perf_pmu_name(void)
-{
- if (!__oprofile_cpu_pmu)
- return NULL;
-
- return __oprofile_cpu_pmu->name;
-}
-EXPORT_SYMBOL_GPL(perf_pmu_name);
-
-int perf_num_counters(void)
-{
- int max_events = 0;
-
- if (__oprofile_cpu_pmu != NULL)
- max_events = __oprofile_cpu_pmu->num_events;
-
- return max_events;
-}
-EXPORT_SYMBOL_GPL(perf_num_counters);
-
static int armpmu_count_irq_users(const int irq)
{
int cpu, count = 0;
@@ -979,9 +952,6 @@ int armpmu_register(struct arm_pmu *pmu)
if (ret)
goto out_destroy;
- if (!__oprofile_cpu_pmu)
- __oprofile_cpu_pmu = pmu;
-
pr_info("enabled with %s PMU driver, %d counters available%s\n",
pmu->name, pmu->num_events,
has_nmi ? ", using NMIs" : "");
--
2.29.2
perf_pmu_name() and perf_num_counters() are unused. Drop them.
Signed-off-by: Marc Zyngier <[email protected]>
---
arch/sh/kernel/perf_event.c | 18 ------------------
1 file changed, 18 deletions(-)
diff --git a/arch/sh/kernel/perf_event.c b/arch/sh/kernel/perf_event.c
index 445e3ece4c23..1d2507f22437 100644
--- a/arch/sh/kernel/perf_event.c
+++ b/arch/sh/kernel/perf_event.c
@@ -57,24 +57,6 @@ static inline int sh_pmu_initialized(void)
return !!sh_pmu;
}
-const char *perf_pmu_name(void)
-{
- if (!sh_pmu)
- return NULL;
-
- return sh_pmu->name;
-}
-EXPORT_SYMBOL_GPL(perf_pmu_name);
-
-int perf_num_counters(void)
-{
- if (!sh_pmu)
- return 0;
-
- return sh_pmu->num_events;
-}
-EXPORT_SYMBOL_GPL(perf_num_counters);
-
/*
* Release the PMU if this is the last perf_event.
*/
--
2.29.2
perf_pmu_name() and perf_num_counters() are unused. Drop them.
Signed-off-by: Marc Zyngier <[email protected]>
---
include/linux/perf_event.h | 2 --
kernel/events/core.c | 5 -----
2 files changed, 7 deletions(-)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 3f7f89ea5e51..51154ed9a206 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -951,8 +951,6 @@ extern void perf_event_itrace_started(struct perf_event *event);
extern int perf_pmu_register(struct pmu *pmu, const char *name, int type);
extern void perf_pmu_unregister(struct pmu *pmu);
-extern int perf_num_counters(void);
-extern const char *perf_pmu_name(void);
extern void __perf_event_task_sched_in(struct task_struct *prev,
struct task_struct *task);
extern void __perf_event_task_sched_out(struct task_struct *prev,
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 03db40f6cba9..88cb0ba5690b 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -580,11 +580,6 @@ static u64 perf_event_time(struct perf_event *event);
void __weak perf_event_print_debug(void) { }
-extern __weak const char *perf_pmu_name(void)
-{
- return "pmu";
-}
-
static inline u64 perf_clock(void)
{
return local_clock();
--
2.29.2
On Wed, Apr 14, 2021 at 3:53 PM Marc Zyngier <[email protected]> wrote:
> perf_pmu_name() and perf_num_counters() are unused. Drop them.
>
> Signed-off-by: Marc Zyngier <[email protected]>
Reviewed-by: Geert Uytterhoeven <[email protected]>
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
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
Hi Marc,
On 2021/4/14 21:44, Marc Zyngier wrote:
> KVM/arm64 is the sole user of perf_num_counters(), and really
> could do without it. Stop using the obsolete API by relying on
> the existing probing code.
>
> Signed-off-by: Marc Zyngier <[email protected]>
> ---
> arch/arm64/kvm/perf.c | 7 +------
> arch/arm64/kvm/pmu-emul.c | 2 +-
> include/kvm/arm_pmu.h | 4 ++++
> 3 files changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/arch/arm64/kvm/perf.c b/arch/arm64/kvm/perf.c
> index 739164324afe..b8b398670ef2 100644
> --- a/arch/arm64/kvm/perf.c
> +++ b/arch/arm64/kvm/perf.c
> @@ -50,12 +50,7 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = {
>
> int kvm_perf_init(void)
> {
> - /*
> - * Check if HW_PERF_EVENTS are supported by checking the number of
> - * hardware performance counters. This could ensure the presence of
> - * a physical PMU and CONFIG_PERF_EVENT is selected.
> - */
> - if (IS_ENABLED(CONFIG_ARM_PMU) && perf_num_counters() > 0)
> + if (kvm_pmu_probe_pmuver() != 0xf)
The probe() function may be called many times (kvm_arm_pmu_v3_set_attr also calls it).
I don't know whether the first calling is enough. If so, can we use a static variable
in it, so the following calling can return the result right away?
Thanks,
Keqian
On Wed, Apr 14, 2021 at 02:44:07PM +0100, Marc Zyngier wrote:
> perf_pmu_name() and perf_num_counters() are unused. Drop them.
>
> Signed-off-by: Marc Zyngier <[email protected]>
> ---
> arch/s390/kernel/perf_event.c | 21 ---------------------
> 1 file changed, 21 deletions(-)
Acked-by: Heiko Carstens <[email protected]>
...or do you want me to pick this up and route via the s390 tree(?).
On Thu, 15 Apr 2021 11:38:52 +0100,
Heiko Carstens <[email protected]> wrote:
>
> On Wed, Apr 14, 2021 at 02:44:07PM +0100, Marc Zyngier wrote:
> > perf_pmu_name() and perf_num_counters() are unused. Drop them.
> >
> > Signed-off-by: Marc Zyngier <[email protected]>
> > ---
> > arch/s390/kernel/perf_event.c | 21 ---------------------
> > 1 file changed, 21 deletions(-)
>
> Acked-by: Heiko Carstens <[email protected]>
>
> ...or do you want me to pick this up and route via the s390 tree(?).
Either way work for me, but I just want to make sure the last patch
doesn't get applied before the previous ones.
Thanks,
M.
--
Without deviation from the norm, progress is not possible.
Hi Marc,
On 2021/4/15 18:42, Marc Zyngier wrote:
> On Thu, 15 Apr 2021 07:59:26 +0100,
> Keqian Zhu <[email protected]> wrote:
>>
>> Hi Marc,
>>
>> On 2021/4/14 21:44, Marc Zyngier wrote:
>>> KVM/arm64 is the sole user of perf_num_counters(), and really
>>> could do without it. Stop using the obsolete API by relying on
>>> the existing probing code.
>>>
>>> Signed-off-by: Marc Zyngier <[email protected]>
>>> ---
>>> arch/arm64/kvm/perf.c | 7 +------
>>> arch/arm64/kvm/pmu-emul.c | 2 +-
>>> include/kvm/arm_pmu.h | 4 ++++
>>> 3 files changed, 6 insertions(+), 7 deletions(-)
>>>
>>> diff --git a/arch/arm64/kvm/perf.c b/arch/arm64/kvm/perf.c
>>> index 739164324afe..b8b398670ef2 100644
>>> --- a/arch/arm64/kvm/perf.c
>>> +++ b/arch/arm64/kvm/perf.c
>>> @@ -50,12 +50,7 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = {
>>>
>>> int kvm_perf_init(void)
>>> {
>>> - /*
>>> - * Check if HW_PERF_EVENTS are supported by checking the number of
>>> - * hardware performance counters. This could ensure the presence of
>>> - * a physical PMU and CONFIG_PERF_EVENT is selected.
>>> - */
>>> - if (IS_ENABLED(CONFIG_ARM_PMU) && perf_num_counters() > 0)
>>> + if (kvm_pmu_probe_pmuver() != 0xf)
>> The probe() function may be called many times
>> (kvm_arm_pmu_v3_set_attr also calls it). I don't know whether the
>> first calling is enough. If so, can we use a static variable in it,
>> so the following calling can return the result right away?
>
> No, because that wouldn't help with crappy big-little implementations
> that could have PMUs with different versions. We want to find the
> version at the point where the virtual PMU is created, which is why we
> call the probe function once per vcpu.
I see.
But AFAICS the pmuver is placed in kvm->arch, and the probe function is called
once per VM. Maybe I miss something.
>
> This of course is broken in other ways (BL+KVM is a total disaster
> when it comes to PMU), but making this static would just make it
> worse.
OK.
Thanks,
Keqian
On Thu, Apr 15, 2021 at 11:47:26AM +0100, Marc Zyngier wrote:
> On Thu, 15 Apr 2021 11:38:52 +0100,
> Heiko Carstens <[email protected]> wrote:
> >
> > On Wed, Apr 14, 2021 at 02:44:07PM +0100, Marc Zyngier wrote:
> > > perf_pmu_name() and perf_num_counters() are unused. Drop them.
> > >
> > > Signed-off-by: Marc Zyngier <[email protected]>
> > > ---
> > > arch/s390/kernel/perf_event.c | 21 ---------------------
> > > 1 file changed, 21 deletions(-)
> >
> > Acked-by: Heiko Carstens <[email protected]>
> >
> > ...or do you want me to pick this up and route via the s390 tree(?).
>
> Either way work for me, but I just want to make sure the last patch
> doesn't get applied before the previous ones.
Ok, I applied this one to the s390 tree. Thanks!
On Wed, Apr 14, 2021 at 02:44:05PM +0100, Marc Zyngier wrote:
> KVM/arm64 is the sole user of perf_num_counters(), and really
> could do without it. Stop using the obsolete API by relying on
> the existing probing code.
>
> Signed-off-by: Marc Zyngier <[email protected]>
> ---
> arch/arm64/kvm/perf.c | 7 +------
> arch/arm64/kvm/pmu-emul.c | 2 +-
> include/kvm/arm_pmu.h | 4 ++++
> 3 files changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/arch/arm64/kvm/perf.c b/arch/arm64/kvm/perf.c
> index 739164324afe..b8b398670ef2 100644
> --- a/arch/arm64/kvm/perf.c
> +++ b/arch/arm64/kvm/perf.c
> @@ -50,12 +50,7 @@ static struct perf_guest_info_callbacks kvm_guest_cbs = {
>
> int kvm_perf_init(void)
> {
> - /*
> - * Check if HW_PERF_EVENTS are supported by checking the number of
> - * hardware performance counters. This could ensure the presence of
> - * a physical PMU and CONFIG_PERF_EVENT is selected.
> - */
> - if (IS_ENABLED(CONFIG_ARM_PMU) && perf_num_counters() > 0)
> + if (kvm_pmu_probe_pmuver() != 0xf)
Took me a while to figure out that this returns 0xf if the hardware has a
PMUVer of 0x0, so it's all good:
Acked-by: Will Deacon <[email protected]>
Will
On Wed, Apr 14, 2021 at 02:44:06PM +0100, Marc Zyngier wrote:
> perf_pmu_name() and perf_num_counters() are now unused. Drop them.
>
> Signed-off-by: Marc Zyngier <[email protected]>
> ---
> drivers/perf/arm_pmu.c | 30 ------------------------------
> 1 file changed, 30 deletions(-)
Nice! This was some of the first code I ever wrote in the kernel but I can't
say I miss it:
Acked-by: Will Deacon <[email protected]>
Will
On Wed, Apr 14, 2021 at 02:44:09PM +0100, Marc Zyngier wrote:
> perf_pmu_name() and perf_num_counters() are unused. Drop them.
>
> Signed-off-by: Marc Zyngier <[email protected]>
> ---
> include/linux/perf_event.h | 2 --
> kernel/events/core.c | 5 -----
> 2 files changed, 7 deletions(-)
Acked-by: Will Deacon <[email protected]>
Will
On Wed, 14 Apr 2021 14:44:04 +0100, Marc Zyngier wrote:
> This small series builds on top of the work that was started with [1].
>
> It recently became apparent that KVM/arm64 is the last bit of the
> kernel that still uses perf_num_counters().
>
> As I went ahead to address this, it became obvious that all traces of
> oprofile had been eradicated from all architectures but arm64, s390
> and sh (plus a bit of cruft in the core perf code). With KVM fixed,
> perf_num_counters() and perf_pmu_name() are finally gone.
>
> [...]
Applied to kvm-arm64/kill_oprofile_dependency, thanks!
[1/5] KVM: arm64: Divorce the perf code from oprofile helpers
commit: 5421db1be3b11c5e469cce3760d5c8a013a90f2c
[2/5] arm64: Get rid of oprofile leftovers
commit: e9c74a686a45e54b2e1c4586b14c84f3ee2f2014
[3/5] s390: Get rid of oprofile leftovers
commit: 8c3f7913a106aa8b94d331cb59709c84a9a1d55b
[4/5] sh: Get rid of oprofile leftovers
commit: ac21ecf5ad32b89909bee2b50161ce93d6462b7d
[5/5] perf: Get rid of oprofile leftovers
commit: 7f318847a0f37b96d8927e8d30ae7b8f149b11f1
Cheers,
M.
--
Without deviation from the norm, progress is not possible.