Re: [PATCH v3 1/4] rtl8723ae: Take the FW LPS mode handling out

[Bernd Edlinger]

On 1/8/19 6:27 PM, Kalle Valo wrote:
> Bernd Edlinger <bernd.edlinger@xxxxxxxxxx> writes:
> 
>> This appears to trigger a firmware bug and causes severe
>> problems with rtl8723ae PCI devices.
>>
>> When the power save mode is activated for longer periods
>> of time the firmware stops to receive any packets.
>>
>> This problem was exposed by commit 873ffe154ae0 ("rtlwifi:
>> Fix logic error in enter/exit power-save mode").
>>
>> Previously the power save mode was only active rarely and
>> only for a short time so that the problem was not noticeable.
>>
>> Signed-off-by: Bernd Edlinger <bernd.edlinger@xxxxxxxxxx>
>> ---
>> v2: Adjust the defaults of swlps and fwlps module
>> parameters to match the firmware capabilities instead of removing
>> the whole code, so it can be easily re-activated once a firmware
>> update is available.
>>
>> v3: Make the title fit in one line.
> 
> hotmail.de made me suspicious and indeed my Spidey sense was right:
> 
> Applying: rtl8723ae: Take the FW LPS mode handling out
> Using index info to reconstruct a base tree...
> .git/rebase-apply/patch:22: trailing whitespace.
>         .swctrl_lps = true,
> .git/rebase-apply/patch:23: trailing whitespace.
>         .fwctrl_lps = false,
> .git/rebase-apply/patch:33: trailing whitespace.
> MODULE_PARM_DESC(swlps, "Set to 1 to use SW control power save (default 1)\n");
> .git/rebase-apply/patch:34: trailing whitespace.
> MODULE_PARM_DESC(fwlps, "Set to 1 to use FW control power save (default 0)\n");
> error: patch failed: drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c:266
> error: drivers/net/wireless/realtek/rtlwifi/rtl8723ae/sw.c: patch does not apply
> error: Did you hand edit your patch?
> It does not apply to blobs recorded in its index.
> Patch failed at 0001 rtl8723ae: Take the FW LPS mode handling out
> 
> Did you use Outlook to send this patch or what? Anyway, it's strongly
> suggested to use git to submit the patches (and not use Outlook servers
> either as they are notorious in breaking our patches):
> 

No, I did use firefox with the settings which I found at
https://www.kernel.org/doc/html/latest/process/email-clients.html#email-clients

I think all that is wrong with the message body is it is base64 encoded, and all the
lines end with CRLF, and it seems the CR is the trailing whitespace.

I think the Character Encoding/UTF-8 does not work right with
Exchange Servers.

What Character Encoding would git send-email use?  UTF-8?

So I will try to send the patches again, with 7-bit encoded message bodies.
Actually it will be quoted-printable, but I do not use any umlauts.


So for the records, I did open about:config in the firefox preferences,
and change:

mailnews.send_default_charset from UTF-8 to ISO-8859-1

that I have already done before, but it was not sufficient:

mailnews.send_plaintext_flowed from true to false
mailnews.wraplength from 72 to 0


Thanks
Bernd.