All of lore.kernel.org
 help / color / mirror / Atom feed
From: skannan@codeaurora.org (Saravana Kannan)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v11 8/8] perf: ARM DynamIQ Shared Unit PMU support
Date: Wed, 28 Feb 2018 14:17:33 -0800	[thread overview]
Message-ID: <5A972A7D.9020301@codeaurora.org> (raw)
In-Reply-To: <20180225143653.peb4quk3ha5h3t5x@salmiak>

On 02/25/2018 06:36 AM, Mark Rutland wrote:
> On Fri, Feb 23, 2018 at 04:53:18PM -0800, Saravana Kannan wrote:
>> On 01/02/2018 03:25 AM, Suzuki K Poulose wrote:
>>> +static void dsu_pmu_event_update(struct perf_event *event)
>>> +{
>>> +	struct hw_perf_event *hwc = &event->hw;
>>> +	u64 delta, prev_count, new_count;
>>> +
>>> +	do {
>>> +		/* We may also be called from the irq handler */
>>> +		prev_count = local64_read(&hwc->prev_count);
>>> +		new_count = dsu_pmu_read_counter(event);
>>> +	} while (local64_cmpxchg(&hwc->prev_count, prev_count, new_count) !=
>>> +			prev_count);
>>> +	delta = (new_count - prev_count) & DSU_PMU_COUNTER_MASK(hwc->idx);
>>> +	local64_add(delta, &event->count);
>>> +}
>>> +
>>> +static void dsu_pmu_read(struct perf_event *event)
>>> +{
>>> +	dsu_pmu_event_update(event);
>>> +}
>
>> I sent out a patch that'll allow PMUs to set an event flag to avoid
>> unnecessary smp calls when the event can be read from any CPU. You could
>> just always set that if you can't have multiple DSU's running the kernel (I
>> don't know if the current ARM designs support having multiple DSUs in a
>> SoC/system) or set it if associated_cpus == cpu_present_mask.
>
> As-is, that won't be safe, given the read function calls the event_update()
> function, which has side-effects on hwc->prec_count and event->count. Those
> need to be serialized somehow.

You have to grab the dsu_pmu->pmu_lock spin lock anyway because the 
system registers are shared across all CPUs. So, just expanding it a bit 
to lock the hwc->prev_count and event->count updated doesn't seem to be 
any worse. In fact, it's better than sending pointless IPIs.

The local64_read/cmpxchg/add etc makes sense when you have per-cpu 
system registers like in the case of the ARM CPU PMU registers. It 
doesn't really buy us much for registers shared across the CPUs.

Thanks,
Saravana

-- 
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

WARNING: multiple messages have this Message-ID (diff)
From: Saravana Kannan <skannan@codeaurora.org>
To: Mark Rutland <mark.rutland@arm.com>
Cc: robh@kernel.org, mathieu.poirier@linaro.org,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	peterz@infradead.org, jonathan.cameron@huawei.com,
	will.deacon@arm.com, linux-kernel@vger.kernel.org,
	marc.zyngier@arm.com, sudeep.holla@arm.com,
	frowand.list@gmail.com, leo.yan@linaro.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v11 8/8] perf: ARM DynamIQ Shared Unit PMU support
Date: Wed, 28 Feb 2018 14:17:33 -0800	[thread overview]
Message-ID: <5A972A7D.9020301@codeaurora.org> (raw)
In-Reply-To: <20180225143653.peb4quk3ha5h3t5x@salmiak>

On 02/25/2018 06:36 AM, Mark Rutland wrote:
> On Fri, Feb 23, 2018 at 04:53:18PM -0800, Saravana Kannan wrote:
>> On 01/02/2018 03:25 AM, Suzuki K Poulose wrote:
>>> +static void dsu_pmu_event_update(struct perf_event *event)
>>> +{
>>> +	struct hw_perf_event *hwc = &event->hw;
>>> +	u64 delta, prev_count, new_count;
>>> +
>>> +	do {
>>> +		/* We may also be called from the irq handler */
>>> +		prev_count = local64_read(&hwc->prev_count);
>>> +		new_count = dsu_pmu_read_counter(event);
>>> +	} while (local64_cmpxchg(&hwc->prev_count, prev_count, new_count) !=
>>> +			prev_count);
>>> +	delta = (new_count - prev_count) & DSU_PMU_COUNTER_MASK(hwc->idx);
>>> +	local64_add(delta, &event->count);
>>> +}
>>> +
>>> +static void dsu_pmu_read(struct perf_event *event)
>>> +{
>>> +	dsu_pmu_event_update(event);
>>> +}
>
>> I sent out a patch that'll allow PMUs to set an event flag to avoid
>> unnecessary smp calls when the event can be read from any CPU. You could
>> just always set that if you can't have multiple DSU's running the kernel (I
>> don't know if the current ARM designs support having multiple DSUs in a
>> SoC/system) or set it if associated_cpus == cpu_present_mask.
>
> As-is, that won't be safe, given the read function calls the event_update()
> function, which has side-effects on hwc->prec_count and event->count. Those
> need to be serialized somehow.

You have to grab the dsu_pmu->pmu_lock spin lock anyway because the 
system registers are shared across all CPUs. So, just expanding it a bit 
to lock the hwc->prev_count and event->count updated doesn't seem to be 
any worse. In fact, it's better than sending pointless IPIs.

The local64_read/cmpxchg/add etc makes sense when you have per-cpu 
system registers like in the case of the ARM CPU PMU registers. It 
doesn't really buy us much for registers shared across the CPUs.

Thanks,
Saravana

-- 
Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

  reply	other threads:[~2018-02-28 22:17 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-02 11:25 [PATCH v11 0/8] perf: Support for ARM DynamIQ Shared Unit Suzuki K Poulose
2018-01-02 11:25 ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 1/8] perf: Export perf_event_update_userpage Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 2/8] of: Add helper for mapping device node to logical CPU number Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 3/8] coresight: of: Use of_cpu_node_to_id helper Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 4/8] irqchip: gic-v3: " Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 5/8] arm64: Use of_cpu_node_to_id helper for CPU topology parsing Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 6/8] arm_pmu: Use of_cpu_node_to_id helper Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 7/8] dt-bindings: Document devicetree binding for ARM DSU PMU Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-01-02 11:25 ` [PATCH v11 8/8] perf: ARM DynamIQ Shared Unit PMU support Suzuki K Poulose
2018-01-02 11:25   ` Suzuki K Poulose
2018-02-22  2:32   ` Saravana Kannan
2018-02-22  2:32     ` Saravana Kannan
2018-02-22 11:33     ` Mark Rutland
2018-02-22 11:33       ` Mark Rutland
2018-02-22 20:38       ` Saravana Kannan
2018-02-22 20:38         ` Saravana Kannan
2018-02-23 11:35         ` Mark Rutland
2018-02-23 11:35           ` Mark Rutland
2018-02-23 21:46           ` Saravana Kannan
2018-02-23 21:46             ` Saravana Kannan
2018-02-24  0:53   ` Saravana Kannan
2018-02-24  0:53     ` Saravana Kannan
2018-02-25 14:36     ` Mark Rutland
2018-02-25 14:36       ` Mark Rutland
2018-02-28 22:17       ` Saravana Kannan [this message]
2018-02-28 22:17         ` Saravana Kannan
2018-03-01 11:49         ` Mark Rutland
2018-03-01 11:49           ` Mark Rutland
2018-03-01 20:35           ` Saravana Kannan
2018-03-01 20:35             ` Saravana Kannan
2018-03-02 10:42             ` Mark Rutland
2018-03-02 10:42               ` Mark Rutland
2018-03-02 19:19               ` Saravana Kannan
2018-03-02 19:19                 ` Saravana Kannan
2018-03-05 10:59                 ` Mark Rutland
2018-03-05 10:59                   ` Mark Rutland
2018-03-05 22:10                   ` Saravana Kannan
2018-03-05 22:10                     ` Saravana Kannan
2018-03-07 14:59                     ` Suzuki K Poulose
2018-03-07 14:59                       ` Suzuki K Poulose
2018-03-07 21:36                       ` Saravana Kannan
2018-03-07 21:36                         ` Saravana Kannan
2018-03-19  9:50                         ` Suzuki K Poulose
2018-03-19  9:50                           ` Suzuki K Poulose
2018-03-08 11:42                     ` Mark Rutland
2018-03-08 11:42                       ` Mark Rutland
2018-03-08 23:59   ` Saravana Kannan
2018-03-08 23:59     ` Saravana Kannan
2018-03-09 10:53     ` Suzuki K Poulose
2018-03-09 10:53       ` Suzuki K Poulose
2018-03-09 13:35       ` Mark Rutland
2018-03-09 13:35         ` Mark Rutland
2018-03-09 22:49         ` Saravana Kannan
2018-03-09 22:49           ` Saravana Kannan
2018-03-10 15:45           ` Mark Rutland
2018-03-10 15:45             ` Mark Rutland

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5A972A7D.9020301@codeaurora.org \
    --to=skannan@codeaurora.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.