From: Ben Greear <greearb@candelatech.com>
To: Rajkumar Manoharan <rmanohar@codeaurora.org>
Cc: linux-wireless@vger.kernel.org,
Rajkumar Manoharan <rmanohar@qti.qualcomm.com>,
ath10k@lists.infradead.org
Subject: Re: [PATCH 2/2] ath10k: remove 10.1 firmware support
Date: Fri, 3 Jun 2016 09:36:15 -0700 [thread overview]
Message-ID: <5751B1FF.2070307@candelatech.com> (raw)
In-Reply-To: <0ec837e1ff03d8a7df8bc06ebca7c635@codeaurora.org>
On 06/03/2016 09:28 AM, Rajkumar Manoharan wrote:
> On 2016-06-03 21:37, Ben Greear wrote:
>> On 06/03/2016 08:55 AM, Rajkumar Manoharan wrote:
>>> On 2016-06-03 21:18, Ben Greear wrote:
>>>> On 06/03/2016 08:33 AM, Rajkumar Manoharan wrote:
>>>>> Earlier qca9888 device was brought up using 10.1 firmware and then
>>>>> later all firmware fixes and new features are migrated to 10.2/10.2.x
>>>>> firmware branch. As all of 10.1 funtionalities are supported in 10.2
>>>>> based firmware, removing 10.1 firmware support for qca9888 device.
>>>>
>>>> Oh please do not do this. My 10.1 firmware works very nicely,
>>>> and out-performs 10.2 in my testing. Lots of people use my firmware
>>>> when they need IBSS support or other features not found in
>>>> official firmware, so it is not just me that will have
>>>> problems if you remove this support.
>>>>
>>> Aah.. I thought CT firmware is 10.2 based. Since most of firmware bug fixes and enhancements are integrated into 10.2 based firmware, we thought of get rid of
>>> 10.1 firmware to reduce code size. Moreover existing 10.1 official firmware has known issues. Is it possible to upgrade CT firmware to 10.2 WMI/HTT interfaces?
>>
>> I have a 10.2 firmware too, but it is less stable, performs worse,
>> uses more RAM on the NIC (so I can do fewer virtual stations),
>> and I am not sure I can squeeze enough RAM out of it to port some
>> of the more interesting rate-ctrl fixes from 10.1 to 10.2.
>> I have recently started backporting a lot of 10.2.4 changes
>> into 10.1, which will aid any users on pre 4.0 kernels since they cannot
>> run 10.2.4 firmware w/out backporting.
>>
>> I might could make my 10.1 work with a 10.2 driver API, but it would
>> take quite a bit of
>> effort, way more than what removing 10.1 from the driver saves in my opinion.
>> The driver already separates firmware specific logic pretty well, so I
>> don't think
>> it should be a huge maintenance effort to keep 10.1 support in the driver.
>>
>> Maybe you could delete the 10.2 (not 10.2.4) firmware support and gain
>> some space that way? I
>> doubt anyone is using that productively....
>>
> Nope. There are customers who are still using 10.2.2 firmware. My kind advice is that please try to optimize latest firmware. 10.2 is nothing but based on 10.1
> trunk so there wont be much interface difference... Will drop this series as of now...
10.2 has a significant re-work of the rate-ctrl. I think that is the main difference
that I see.
Maybe I can just copy the 10.1 rate-ctrl over 10.2 and get back the memory and performance
that I need. But, I have a bunch of other stuff to do first.
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k
next prev parent reply other threads:[~2016-06-03 16:42 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-03 15:33 [PATCH 1/2] ath10k: handle testmode events for 10.2 based firmware Rajkumar Manoharan
2016-06-03 15:33 ` [PATCH 2/2] ath10k: remove 10.1 firmware support Rajkumar Manoharan
2016-06-03 15:48 ` Ben Greear
2016-06-03 15:55 ` Rajkumar Manoharan
2016-06-03 16:07 ` Ben Greear
2016-06-03 16:28 ` Rajkumar Manoharan
2016-06-03 16:36 ` Ben Greear [this message]
2016-06-03 16:09 ` Valo, Kalle
2016-06-03 16:20 ` Ben Greear
2016-06-03 16:33 ` Valo, Kalle
2016-06-03 16:46 ` Ben Greear
2016-09-14 12:24 ` [1/2] ath10k: handle testmode events for 10.2 based firmware Kalle Valo
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=5751B1FF.2070307@candelatech.com \
--to=greearb@candelatech.com \
--cc=ath10k@lists.infradead.org \
--cc=linux-wireless@vger.kernel.org \
--cc=rmanohar@codeaurora.org \
--cc=rmanohar@qti.qualcomm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox