[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <092d917b4cae2762317a8739c874855554ae913f.camel@mediatek.com>
Date: Wed, 3 Jul 2024 16:41:41 +0000
From: Jason-JH Lin (林睿祥) <Jason-JH.Lin@...iatek.com>
To: "jassisinghbrar@...il.com" <jassisinghbrar@...il.com>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-mediatek@...ts.infradead.org" <linux-mediatek@...ts.infradead.org>,
Singo Chang (張興國) <Singo.Chang@...iatek.com>,
"chunkuang.hu@...nel.org" <chunkuang.hu@...nel.org>,
Nancy Lin (林欣螢) <Nancy.Lin@...iatek.com>,
Project_Global_Chrome_Upstream_Group
<Project_Global_Chrome_Upstream_Group@...iatek.com>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>, "matthias.bgg@...il.com"
<matthias.bgg@...il.com>, "angelogioacchino.delregno@...labora.com"
<angelogioacchino.delregno@...labora.com>
Subject: Re: [PATCH 2/2] mailbox: mtk-cmdq: Move pm_runimte_get and put to
mbox_chan_ops API
On Thu, 2024-06-27 at 22:40 -0500, Jassi Brar wrote:
>
> External email : Please do not click links or open attachments until
> you have verified the sender or the content.
> On Wed, Jun 26, 2024 at 4:32 AM Jason-JH Lin (林睿祥)
> <Jason-JH.Lin@...iatek.com> wrote:
> > >
> > > The idea is that
> > > 1) If the gap between transfers on a busy channel is ~10ms or
> > > more. And ..
> > > 2) The silence on the mailbox channel is multiple of 100ms. And
> ...
> >
> > The mailbox channel would not be always busy and the gap is not
> > consistent as well.
> > For example, display driver calls mbox_send_message() every ~16ms
> while
> > UI is updating in 60fps.
> > If UI dose not update, display driver won't call
> mbox_send_message(),
> > so user may not has the gap and the silence like this.
> >
> From your logs, send_diff only increases from 16ms if the UI is not
> updating. Which makes it more prudent to release the channel.
>
> The user needs the channel for mbox_send_message, while actual power
> saving comes from cmdq_runtime_suspend/resume.
> So for your target usage pattern, trace mbox_send_message() ,
> cmdq_runtime_resume() and cmdq_runtime_suspend() and compare the
> timestamped logs with and without this patchset.
>
[snip]
I realized that we have 2 cmdq drivers:
display uses 10320000.mailbox and imgsys uses 10330000.mailbox
I added diff logs in cmdq_runtime_resume(), cmdq_runtime_suspend()
and cmdq_mbox_send_data().
Here is the log of 10320000.mailbox:
// CMDQ probe
[ 4.288408] mtk_cmdq 10320000.mailbox: IRQ=279++++++++++++++
// Show boot logo
[ 9.177741] cmdq_runtime_resume: resume_diff=9177 ms
[ 9.698045] cmdq_mbox_send_data: send_diff=14 ms
// Show UI
[ 15.620525] cmdq_mbox_send_data: send_diff=5922 ms
[ 15.652041] cmdq_mbox_send_data: send_diff=31 ms
[ 15.681308] cmdq_mbox_send_data: send_diff=29 ms
[ 16.276090] cmdq_mbox_send_data: send_diff=594 ms
[ 16.355256] cmdq_mbox_send_data: send_diff=79 ms
...
[ 35.013395] cmdq_mbox_send_data: send_diff=16 ms
[ 35.030301] cmdq_mbox_send_data: send_diff=16 ms
[ 35.047002] cmdq_mbox_send_data: send_diff=16 ms
[ 35.068360] cmdq_mbox_send_data: send_diff=21 ms
// Open camera APP
[ 37.291937] cmdq_mbox_send_data: send_diff=2223 ms
[ 37.884435] cmdq_mbox_send_data: send_diff=592 ms
[ 37.905703] cmdq_mbox_send_data: send_diff=21 ms
[ 37.928599] cmdq_mbox_send_data: send_diff=22 ms
...
[ 41.396577] cmdq_mbox_send_data: send_diff=16 ms
[ 41.413107] cmdq_mbox_send_data: send_diff=16 ms
[ 42.440862] cmdq_mbox_send_data: send_diff=1027 ms
[ 42.647045] cmdq_mbox_send_data: send_diff=206 ms
// Close camera APP
[ 89.940631] cmdq_mbox_send_data: send_diff=47293 ms
[ 106.704928] cmdq_mbox_send_data: send_diff=16764 ms
...
[ 113.130841] cmdq_mbox_send_data: send_diff=211 ms
// push power key to disable panel
[ 113.281999] cmdq_runtime_suspend: R/S_diff=104104 ms
// push power key to enable panel
[ 131.287190] cmdq_runtime_resume: resume_diff=122109 ms
Here is the log of 10330000.mailbox:
// CMDQ probe
[ 4.290404] mtk_cmdq 10330000.mailbox: IRQ=280++++++++++++++
// Open camera APP
[ 36.174867] BUG: sleeping function called from invalid context at
drivers/base/power/runtime.c:1164
[ 36.183942] in_atomic(): 1, irqs_disabled(): 128, non_block: 0, pid:
95, name: kworker/u17:0
[ 36.192383] preempt_count: 1, expected: 0
[ 36.196395] RCU nest depth: 0, expected: 0
[ 36.200489] INFO: lockdep is turned off.
[ 36.204405] irq event stamp: 16
[ 36.238636] CPU: 2 PID: 95 Comm: kworker/u17:0 Not tainted
[ 36.251508] Hardware name: Google Ciri sku0/unprovisioned board (DT)
[ 36.257864] Workqueue: imgsys_runner imgsys_runner_func
[ 36.263107] Call trace:
[ 36.265555] dump_backtrace+0x100/0x120
[ 36.269400] show_stack+0x20/0x2c
[ 36.272722] dump_stack_lvl+0x84/0xb4
[ 36.276390] dump_stack+0x18/0x44
[ 36.279706] __might_resched+0x204/0x214
[ 36.283631] __might_sleep+0x50/0x80
[ 36.287209] __pm_runtime_resume+0x48/0xbc
[ 36.291316] cmdq_mbox_send_data+0x84/0x43c
[ 36.295510] msg_submit+0x84/0x104
[ 36.298919] mbox_send_message+0xb8/0x124
[ 36.302934] imgsys_cmdq_sendtask+0x6b0/0x990
[ 36.307300] imgsys_runner_func+0x44/0x78
[ 36.311318] process_one_work+0x274/0x574
[ 36.315336] worker_thread+0x240/0x3f0
[ 36.319093] kthread+0xfc/0x11c
[ 36.322242] ret_from_fork+0x10/0x20
[ 36.325989] cmdq_runtime_resume: resume_diff=36325 ms
[ 36.342235] cmdq_mbox_send_data: send_diff=167 ms
[ 36.366714] cmdq_mbox_send_data: send_diff=24 ms
[ 36.382882] cmdq_mbox_send_data: send_diff=16 ms
[ 36.399512] cmdq_mbox_send_data: send_diff=16 ms
...
[ 42.817244] cmdq_mbox_send_data: send_diff=16 ms
[ 42.844891] cmdq_mbox_send_data: send_diff=27 ms
[ 42.856340] cmdq_mbox_send_data: send_diff=11 ms
[ 42.856364] cmdq_mbox_send_data: send_diff=0 ms
[ 42.856384] cmdq_mbox_send_data: send_diff=0 ms
[ 42.856410] cmdq_mbox_send_data: send_diff=0 ms
[ 42.856431] cmdq_mbox_send_data: send_diff=0 ms
[ 42.856458] cmdq_mbox_send_data: send_diff=0 ms
[ 42.856482] cmdq_mbox_send_data: send_diff=0 ms
[ 42.985695] cmdq_mbox_send_data: send_diff=129 ms
[ 42.992017] cmdq_runtime_suspend: R/S_diff=6666 ms
Regards,
Jason-JH.Lin
Powered by blists - more mailing lists