Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v8 2/2] drivers/perf: hisi: Add driver for HiSilicon PCIe
 PMU
To:     Will Deacon <will@kernel.org>, Linuxarm <linuxarm@huawei.com>
CC:     <mark.rutland@arm.com>, <bhelgaas@google.com>,
        <linux-pci@vger.kernel.org>,
        <linux-arm-kernel@lists.infradead.org>,
        <linux-kernel@vger.kernel.org>, <zhangshaokun@hisilicon.com>
References: <20210728080932.72515-1-liuqi115@huawei.com>
 <20210728080932.72515-3-liuqi115@huawei.com>
 <20210802100343.GA27282@willie-the-truck>
From:   "liuqi (BA)" <liuqi115@huawei.com>
Message-ID: <a56266c4-c434-f078-6027-f30c021bd593@huawei.com>
Date:   Wed, 4 Aug 2021 15:29:54 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:68.0) Gecko/20100101
 Thunderbird/68.6.0
MIME-Version: 1.0
In-Reply-To: <20210802100343.GA27282@willie-the-truck>
Content-Type: text/plain; charset="utf-8"; format=flowed
Content-Language: en-GB
Content-Transfer-Encoding: 7bit
Precedence: bulk


Hi Will,
> Hmm, I was hoping that you would expose all the events as proper perf_events
> and get rid of the subevents entirely.
> 
> Then userspace could do things like:
> 
>    // Count number of RX memory reads
>    $ perf stat -e hisi_pcie0_0/rx_memory_read/
> 
>    // Count delay cycles
>    $ perf stat -e hisi_pcie0_0/latency/
> 
>    // Count both of the above (events must be in the same group)
>    $ perf stat -g -e hisi_pcie0_0/latency/ -e hisi_pcie0_0/rx_memory_read/
> 
> Note that in all three of these cases the hardware will be programmed in
> the same way and both HISI_PCIE_CNT and HISI_PCIE_EXT_CNT are allocated!
> 
> So for example, doing this (i.e. without the '-g'):
> 
>    $ perf stat -e hisi_pcie0_0/latency/ -e hisi_pcie0_0/rx_memory_read/
> 
> would fail because the first event would allocate both of the counters.

I'm confused with this situation when getting rid of subevent:

$ perf stat -e hisi_pcie0_0/latency/ -e hisi_pcie0_0/rx_memory_read/

In this case, driver checks the relationship of "latency" and 
"rx_memory_read" in pmu->add() function and return a -EINVAL, but this 
seems lead to time division multiplexing.

	if (event->pmu->add(event, PERF_EF_START)) {
		perf_event_set_state(event, PERF_EVENT_STATE_INACTIVE);
		event->oncpu = -1;
		ret = -EAGAIN;
		goto out;
	}
	...
out:
	perf_pmu_enable(event->pmu);

This result doesn't meet our expection, do I miss something here?

How about add an array to record events and check the relationship in 
event_init() function? It seems that perf stat could only failed when 
driver return invalid value in pmu->event_init() function.

Thanks,
Qi
> 
> All you need to do is check the counter scheduling constraints when
> accepting an event group in the driver. No need for subevents at all.
> 
> Does that make sense?
> 
> Will
> .
>