From: Bin Liu <b-liu@ti.com>
To: Peter Chen <hzpeterchen@gmail.com>
Cc: pawell@cadence.com, balbi@kernel.org, devicetree@vger.kernel.org,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-usb@vger.kernel.org, rogerq@ti.com,
lkml <linux-kernel@vger.kernel.org>,
adouglas@cadence.com, jbergsagel@ti.com, nsekhar@ti.com,
nm@ti.com, sureshp@cadence.com, peter.chen@nxp.com,
pjez@cadence.com, kurahul@cadence.com
Subject: [v1,2/2] usb:cdns3 Add Cadence USB3 DRD Driver
Date: Wed, 12 Dec 2018 09:53:13 -0600 [thread overview]
Message-ID: <20181212155313.GA6502@uda0271908> (raw)
Peter,
On Wed, Dec 12, 2018 at 10:04:25AM +0800, Peter Chen wrote:
[snip]
> > >I strongly advise against using dev_dbg() for debugging. Even more so
> > >inside your IRQ handler.
>
> Felipe, I use Dynamic Debug for debugging, and show debug messages with
> "dmesg" after testing/debugging. I see dwc3 using trace, any benefits
> for switching
> to trace?
The benefits I see are
- *by default*, the debug log doesn't have to go through uart console,
which is slow.
I typically 'cat trace_pipe' from a telnet which is much faster then
uart. But Dynamic Debug log by default got printed on uart, I have to
set printk level to not print them, which is an extra step.
- tracepoint uses one place to decode the message vs DD has to repeat
the similar print statement in the code; and tracepoint decodes it
offline which reduce the kernel runtime overhead too.
- with tracepoint, it is easier to turn on debug for a specific group of
messages.
- you can adjust the ftrace buffer size at runtime. can we do that for
printk? I don't remember.
Regards,
-Bin.
WARNING: multiple messages have this Message-ID (diff)
From: Bin Liu <b-liu@ti.com>
To: Peter Chen <hzpeterchen@gmail.com>
Cc: pawell@cadence.com, balbi@kernel.org, devicetree@vger.kernel.org,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
linux-usb@vger.kernel.org, rogerq@ti.com,
lkml <linux-kernel@vger.kernel.org>,
adouglas@cadence.com, jbergsagel@ti.com, nsekhar@ti.com,
nm@ti.com, sureshp@cadence.com, peter.chen@nxp.com,
pjez@cadence.com, kurahul@cadence.com
Subject: Re: [PATCH v1 2/2] usb:cdns3 Add Cadence USB3 DRD Driver
Date: Wed, 12 Dec 2018 09:53:13 -0600 [thread overview]
Message-ID: <20181212155313.GA6502@uda0271908> (raw)
In-Reply-To: <CAL411-ryK2LLZVg5_sLkCYOuhwdHLVm5XavSyC9nbEXZah_43w@mail.gmail.com>
Peter,
On Wed, Dec 12, 2018 at 10:04:25AM +0800, Peter Chen wrote:
[snip]
> > >I strongly advise against using dev_dbg() for debugging. Even more so
> > >inside your IRQ handler.
>
> Felipe, I use Dynamic Debug for debugging, and show debug messages with
> "dmesg" after testing/debugging. I see dwc3 using trace, any benefits
> for switching
> to trace?
The benefits I see are
- *by default*, the debug log doesn't have to go through uart console,
which is slow.
I typically 'cat trace_pipe' from a telnet which is much faster then
uart. But Dynamic Debug log by default got printed on uart, I have to
set printk level to not print them, which is an extra step.
- tracepoint uses one place to decode the message vs DD has to repeat
the similar print statement in the code; and tracepoint decodes it
offline which reduce the kernel runtime overhead too.
- with tracepoint, it is easier to turn on debug for a specific group of
messages.
- you can adjust the ftrace buffer size at runtime. can we do that for
printk? I don't remember.
Regards,
-Bin.
WARNING: multiple messages have this Message-ID (diff)
From: Bin Liu <b-liu@ti.com>
To: Peter Chen <hzpeterchen@gmail.com>
Cc: <pawell@cadence.com>, <balbi@kernel.org>,
<devicetree@vger.kernel.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
<linux-usb@vger.kernel.org>, <rogerq@ti.com>,
lkml <linux-kernel@vger.kernel.org>, <adouglas@cadence.com>,
<jbergsagel@ti.com>, <nsekhar@ti.com>, <nm@ti.com>,
<sureshp@cadence.com>, <peter.chen@nxp.com>, <pjez@cadence.com>,
<kurahul@cadence.com>
Subject: Re: [PATCH v1 2/2] usb:cdns3 Add Cadence USB3 DRD Driver
Date: Wed, 12 Dec 2018 09:53:13 -0600 [thread overview]
Message-ID: <20181212155313.GA6502@uda0271908> (raw)
In-Reply-To: <CAL411-ryK2LLZVg5_sLkCYOuhwdHLVm5XavSyC9nbEXZah_43w@mail.gmail.com>
Peter,
On Wed, Dec 12, 2018 at 10:04:25AM +0800, Peter Chen wrote:
[snip]
> > >I strongly advise against using dev_dbg() for debugging. Even more so
> > >inside your IRQ handler.
>
> Felipe, I use Dynamic Debug for debugging, and show debug messages with
> "dmesg" after testing/debugging. I see dwc3 using trace, any benefits
> for switching
> to trace?
The benefits I see are
- *by default*, the debug log doesn't have to go through uart console,
which is slow.
I typically 'cat trace_pipe' from a telnet which is much faster then
uart. But Dynamic Debug log by default got printed on uart, I have to
set printk level to not print them, which is an extra step.
- tracepoint uses one place to decode the message vs DD has to repeat
the similar print statement in the code; and tracepoint decodes it
offline which reduce the kernel runtime overhead too.
- with tracepoint, it is easier to turn on debug for a specific group of
messages.
- you can adjust the ftrace buffer size at runtime. can we do that for
printk? I don't remember.
Regards,
-Bin.
next reply other threads:[~2018-12-12 15:53 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-12 15:53 Bin Liu [this message]
2018-12-12 15:53 ` [PATCH v1 2/2] usb:cdns3 Add Cadence USB3 DRD Driver Bin Liu
2018-12-12 15:53 ` Bin Liu
-- strict thread matches above, loose matches on Subject: below --
2018-12-31 5:35 [v1,1/2] dt-bindings: add binding for USBSS-DRD controller Pawel Laszczak
2018-12-31 5:35 ` [PATCH v1 1/2] " Pawel Laszczak
2018-12-27 21:01 [v1,1/2] " Rob Herring
2018-12-27 21:01 ` [PATCH v1 1/2] " Rob Herring
2018-12-22 22:24 [v1,1/2] " Pawel Laszczak
2018-12-22 22:24 ` [PATCH v1 1/2] " Pawel Laszczak
2018-12-20 20:01 [v1,1/2] " Rob Herring
2018-12-20 20:01 ` [PATCH v1 1/2] " Rob Herring
2018-12-17 11:56 [v1,2/2] usb:cdns3 Add Cadence USB3 DRD Driver Felipe Balbi
2018-12-17 11:56 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-17 11:56 ` Felipe Balbi
2018-12-17 11:51 [v1,2/2] " Pawel Laszczak
2018-12-17 11:51 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-17 11:34 [v1,2/2] " Felipe Balbi
2018-12-17 11:34 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-17 11:34 ` Felipe Balbi
2018-12-17 11:25 [v1,2/2] " Pawel Laszczak
2018-12-17 11:25 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-17 5:46 [v1,2/2] " Pawel Laszczak
2018-12-17 5:46 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-16 13:01 [v1,2/2] " Pawel Laszczak
2018-12-16 13:01 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-14 22:56 [v1,2/2] " kbuild test robot
2018-12-14 22:56 ` [PATCH v1 2/2] " kbuild test robot
2018-12-14 22:56 ` kbuild test robot
2018-12-14 3:46 [v1,2/2] " Kishon Vijay Abraham I
2018-12-14 3:46 ` [PATCH v1 2/2] " Kishon Vijay Abraham I
2018-12-13 9:35 [v1,2/2] " Peter Chen
2018-12-13 9:35 ` [PATCH v1 2/2] " Peter Chen
2018-12-13 9:25 [v1,1/2] dt-bindings: add binding for USBSS-DRD controller Pawel Laszczak
2018-12-13 9:25 ` [PATCH v1 1/2] " Pawel Laszczak
2018-12-13 9:20 [v1,1/2] " Peter Chen
2018-12-13 9:20 ` [PATCH v1 1/2] " Peter Chen
2018-12-13 1:21 [v1,2/2] usb:cdns3 Add Cadence USB3 DRD Driver Peter Chen
2018-12-13 1:21 ` [PATCH v1 2/2] " Peter Chen
2018-12-12 9:24 [v1,2/2] " Peter Chen
2018-12-12 9:24 ` [PATCH v1 2/2] " Peter Chen
2018-12-12 8:34 [v1,2/2] " Felipe Balbi
2018-12-12 8:34 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-12 8:34 ` Felipe Balbi
2018-12-12 7:38 [v1,2/2] " Peter Chen
2018-12-12 7:38 ` [PATCH v1 2/2] " Peter Chen
2018-12-12 6:55 [v1,2/2] " Felipe Balbi
2018-12-12 6:55 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-12 6:52 [v1,2/2] " Felipe Balbi
2018-12-12 6:52 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-12 6:52 ` Felipe Balbi
2018-12-12 2:04 [v1,2/2] " Peter Chen
2018-12-12 2:04 ` [PATCH v1 2/2] " Peter Chen
2018-12-11 19:04 [v1,2/2] " Pawel Laszczak
2018-12-11 19:04 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-11 19:04 ` Pawel Laszczak
2018-12-11 12:15 [v1,2/2] " Felipe Balbi
2018-12-11 12:15 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-11 12:15 ` Felipe Balbi
2018-12-11 12:14 [v1,2/2] " Felipe Balbi
2018-12-11 12:14 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-11 12:14 ` Felipe Balbi
2018-12-11 11:46 [v1,2/2] " Felipe Balbi
2018-12-11 11:46 ` [PATCH v1 2/2] " Felipe Balbi
2018-12-11 10:16 [v1,1/2] dt-bindings: add binding for USBSS-DRD controller Roger Quadros
2018-12-11 10:16 ` [PATCH v1 1/2] " Roger Quadros
2018-12-11 10:16 ` Roger Quadros
2018-12-11 10:01 [v1,2/2] usb:cdns3 Add Cadence USB3 DRD Driver Pawel Laszczak
2018-12-11 10:01 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-11 9:39 [v1,2/2] " Roger Quadros
2018-12-11 9:39 ` [PATCH v1 2/2] " Roger Quadros
2018-12-11 9:39 ` Roger Quadros
2018-12-10 12:39 [v1,2/2] " Pawel Laszczak
2018-12-10 12:39 ` [PATCH v1 2/2] " Pawel Laszczak
2018-12-10 12:39 ` Pawel Laszczak
2018-12-10 12:39 [v1,1/2] dt-bindings: add binding for USBSS-DRD controller Pawel Laszczak
2018-12-10 12:39 ` [PATCH v1 1/2] " Pawel Laszczak
2018-12-10 12:39 ` Pawel Laszczak
2018-12-10 12:39 [PATCH v1 0/2] Introduced new Cadence USBSS DRD Driver Pawel Laszczak
2018-12-10 12:39 ` Pawel Laszczak
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=20181212155313.GA6502@uda0271908 \
--to=b-liu@ti.com \
--cc=adouglas@cadence.com \
--cc=balbi@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=gregkh@linuxfoundation.org \
--cc=hzpeterchen@gmail.com \
--cc=jbergsagel@ti.com \
--cc=kurahul@cadence.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=nm@ti.com \
--cc=nsekhar@ti.com \
--cc=pawell@cadence.com \
--cc=peter.chen@nxp.com \
--cc=pjez@cadence.com \
--cc=rogerq@ti.com \
--cc=sureshp@cadence.com \
/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.