[Buildroot] package/rtl8189fs: bug in conf, and a question

[Buildroot] package/rtl8189fs: bug in conf, and a question

[Indrek Kruusa]

[-- Attachment #1.1: Type: text/plain, Size: 1063 bytes --]

Hi!
I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s RISC-V). I'm
almost there - Samuel Holland managed to mainline initial support for this
board, so vanilla linux kernel 6.3-rc2 boots fine (with proper/modified
u-boot version).
The WiFi module is based on rtl8189ftv chip (SDIO connection) and its
driver seems to be problematic.
First, buildroot's package/rtl8189fs is currently broken; it points to the
wrong commit (seems to be copy paste error, inherited from
package/rtl8189es). The result is unusable. I updated it to the correct
commit (on the correct branch) and the driver builds fine and loads fine.
There are couple of issues though:
- for some reason the debug mode in driver is set
-  driver seems to constantly restart the module, and it emits endless
debug messages

So, is there anybody for whom the rtl8189fs linux driver works fine with
kernel 6.3-rcX?

PS. I can prepare a patch to fix the package/rtl8189fs but I expect Giulio
to bump the version again for both package/rtl8189es and package/rtl8189fs
soon.

Best regards,
Indrek

[-- Attachment #1.2: Type: text/html, Size: 1420 bytes --]

[-- Attachment #2: Type: text/plain, Size: 150 bytes --]

_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

Re: [Buildroot] package/rtl8189fs: bug in conf, and a question

[Thomas Petazzoni via buildroot]

Hello Indrek,

On Sat, 18 Mar 2023 23:46:15 +0200
Indrek Kruusa <indrek.kruusa@gmail.com> wrote:

> I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s RISC-V). I'm
> almost there - Samuel Holland managed to mainline initial support for this
> board, so vanilla linux kernel 6.3-rc2 boots fine (with proper/modified
> u-boot version).
> The WiFi module is based on rtl8189ftv chip (SDIO connection) and its
> driver seems to be problematic.
> First, buildroot's package/rtl8189fs is currently broken; it points to the
> wrong commit (seems to be copy paste error, inherited from
> package/rtl8189es). The result is unusable. I updated it to the correct
> commit (on the correct branch) and the driver builds fine and loads fine.

Are you sure? It works fine here:

$ make rtl8189fs-extract
>>> rtl8189fs e58bd86c9d9408c648b1246a0dd76b16856ec172 Downloading
wget --passive-ftp -nd -t 3 -O '/home/thomas/projets/buildroot/output/build/.rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz.bpmm05/output' 'https://github.com/jwrdegoede/rtl8189ES_linux/archive/e58bd86c9d9408c648b1246a0dd76b16856ec172/rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz' 
--2023-03-18 23:20:29--  https://github.com/jwrdegoede/rtl8189ES_linux/archive/e58bd86c9d9408c648b1246a0dd76b16856ec172/rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz
Resolving github.com (github.com)... 140.82.121.4
Connecting to github.com (github.com)|140.82.121.4|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: https://codeload.github.com/jwrdegoede/rtl8189ES_linux/tar.gz/e58bd86c9d9408c648b1246a0dd76b16856ec172 [following]
--2023-03-18 23:20:29--  https://codeload.github.com/jwrdegoede/rtl8189ES_linux/tar.gz/e58bd86c9d9408c648b1246a0dd76b16856ec172
Resolving codeload.github.com (codeload.github.com)... 140.82.121.9
Connecting to codeload.github.com (codeload.github.com)|140.82.121.9|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [application/x-gzip]
Saving to: ‘/home/thomas/projets/buildroot/output/build/.rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz.bpmm05/output’

/home/thomas/projets/buildroo     [  <=>                                           ]   2,42M  9,05MB/s    in 0,3s    

2023-03-18 23:20:30 (9,05 MB/s) - ‘/home/thomas/projets/buildroot/output/build/.rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz.bpmm05/output’ saved [2539299]

rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz: OK (sha256: 96347dbfedd95a217fb89b7b295fe417b711dbad3db896ef0c2fa733c6d150fd)
>>> rtl8189fs e58bd86c9d9408c648b1246a0dd76b16856ec172 Extracting
gzip -d -c /home/thomas/dl/rtl8189fs/rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172.tar.gz | tar --strip-components=1 -C /home/thomas/projets/buildroot/output/build/rtl8189fs-e58bd86c9d9408c648b1246a0dd76b16856ec172   -xf -

Best regards,

Thomas
-- 
Thomas Petazzoni, co-owner and CEO, Bootlin
Embedded Linux and Kernel engineering and training
https://bootlin.com
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

Re: [Buildroot] package/rtl8189fs: bug in conf, and a question

[Giulio Benetti]

[-- Attachment #1.1: Type: text/plain, Size: 2183 bytes --]

Hi Indkrek,

> Il giorno 18 mar 2023, alle ore 22:46, Indrek Kruusa <indrek.kruusa@gmail.com> ha scritto:
> 
> 
> Hi!
> I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s RISC-V). I'm almost there - Samuel Holland managed to mainline initial support for this board, so vanilla linux kernel 6.3-rc2 boots fine (with proper/modified u-boot version).
> The WiFi module is based on rtl8189ftv chip (SDIO connection) and its driver seems to be problematic.
> First, buildroot's package/rtl8189fs is currently broken; it points to the wrong commit (seems to be copy paste error, inherited from package/rtl8189es). The result is unusable.

as Thomas already pointed the commit works ok for me too.

> I updated it to the correct commit (on the correct branch) and the driver builds fine and loads fine.

Which branch and commit have you used? Take care that the commit actually used has not the support for Linux 6.3 that has been added later.

> There are couple of issues though:
> - for some reason the debug mode in driver is set
> -  driver seems to constantly restart the module, and it emits endless debug messages

This could be bound to the branch/commit of the driver, if it includes Linux 6.3 support or not.
But it should, otherwise it should not build.
Anyway I don’t know that much about other branches.
Maybe you mean that master branch is for rtl8189es and rtl8189fs branch is for rtl8189fs?

Also, how do you determine that rtl8189ftv is compatible with rtl8189fs?

> 
> So, is there anybody for whom the rtl8189fs linux driver works fine with kernel 6.3-rcX?
> 
> PS. I can prepare a patch to fix the package/rtl8189fs but I expect Giulio to bump the version again for both package/rtl8189es and package/rtl8189fs soon.

Sure I can do that, there is already a pending patch for rtl8189es but it’s already superseded at this point. 
By Monday I’ll send the 2 patches or feel free to do it instead.

Best regards
Giulio

> 
> Best regards,
> Indrek
> 
> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot

[-- Attachment #1.2: Type: text/html, Size: 3560 bytes --]

[-- Attachment #2: Type: text/plain, Size: 150 bytes --]

_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

Re: [Buildroot] package/rtl8189fs: bug in conf, and a question

[Indrek Kruusa]

[-- Attachment #1.1: Type: text/plain, Size: 3161 bytes --]

Kontakt Giulio Benetti (<giulio.benetti@benettiengineering.com>) kirjutas
kuupäeval P, 19. märts 2023 kell 01:01:

> Hi Indkrek,
>
> Il giorno 18 mar 2023, alle ore 22:46, Indrek Kruusa <
> indrek.kruusa@gmail.com> ha scritto:
>
> 
> Hi!
> I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s RISC-V). I'm
> almost there - Samuel Holland managed to mainline initial support for this
> board, so vanilla linux kernel 6.3-rc2 boots fine (with proper/modified
> u-boot version).
> The WiFi module is based on rtl8189ftv chip (SDIO connection) and its
> driver seems to be problematic.
> First, buildroot's package/rtl8189fs is currently broken; it points to the
> wrong commit (seems to be copy paste error, inherited from
> package/rtl8189es). The result is unusable.
>
>
> as Thomas already pointed the commit works ok for me too.
>
> I updated it to the correct commit (on the correct branch) and the driver
> builds fine and loads fine.
>
>
> Which branch and commit have you used? Take care that the commit actually
> used has not the support for Linux 6.3 that has been added later.
>
> There are couple of issues though:
> - for some reason the debug mode in driver is set
>
> -  driver seems to constantly restart the module, and it emits endless
> debug messages
>
>
> This could be bound to the branch/commit of the driver, if it includes
> Linux 6.3 support or not.
> But it should, otherwise it should not build.
> Anyway I don’t know that much about other branches.
> Maybe you mean that master branch is for rtl8189es and rtl8189fs branch is
> for rtl8189fs?
>
> Also, how do you determine that rtl8189ftv is compatible with rtl8189fs?
>
>
> So, is there anybody for whom the rtl8189fs linux driver works fine with
> kernel 6.3-rcX?
>
>
> PS. I can prepare a patch to fix the package/rtl8189fs but I expect Giulio
> to bump the version again for both package/rtl8189es and
> package/rtl8189fs soon.
>
>
> Sure I can do that, there is already a pending patch for rtl8189es but
> it’s already superseded at this point.
> By Monday I’ll send the 2 patches or feel free to do it instead.
>
>

Hi!

Yes, for the rtl8189fs kernel module the code must be checked out from the
rtl8189fs branch. Currently the commit hash for the package/rtl8189es and
package/rtl8189fs is the same and it's from rtl8189ES_linux/master. It
builds but it produces the wrong thing in the case of the rtl8189fs. I'm
fine if it gets fixed next week ;)
About rtl8189ftv and compatibility. The difference between rtl8188ftv and
rtl8189ftv chip seems to be the connection interface type (usb and sdio
accordingly). It looks to me that the rtl8189fs driver still relies on the
8188f specific code but it adds sdio capability (roughly speaking). Also,
https://linux-sunxi.org/Wifi#RTL8189FTV says that the rtl8189fs branch is
the only working solution currently (or has been at least).
There can be several reasons why it doesn't work for me. That's why I was
asking first about the others' experience on actual hardware, maybe with
the different (mature) SOC/platform.

Thanks,
Indrek

[-- Attachment #1.2: Type: text/html, Size: 4511 bytes --]

[-- Attachment #2: Type: text/plain, Size: 150 bytes --]

_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

Re: [Buildroot] package/rtl8189fs: bug in conf, and a question

[Indrek Kruusa]

[-- Attachment #1.1: Type: text/plain, Size: 3637 bytes --]

Kontakt Indrek Kruusa (<indrek.kruusa@gmail.com>) kirjutas kuupäeval P, 19.
märts 2023 kell 09:43:

> Kontakt Giulio Benetti (<giulio.benetti@benettiengineering.com>) kirjutas
> kuupäeval P, 19. märts 2023 kell 01:01:
>
>> Hi Indkrek,
>>
>> Il giorno 18 mar 2023, alle ore 22:46, Indrek Kruusa <
>> indrek.kruusa@gmail.com> ha scritto:
>>
>> 
>> Hi!
>> I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s RISC-V). I'm
>> almost there - Samuel Holland managed to mainline initial support for this
>> board, so vanilla linux kernel 6.3-rc2 boots fine (with proper/modified
>> u-boot version).
>> The WiFi module is based on rtl8189ftv chip (SDIO connection) and its
>> driver seems to be problematic.
>> First, buildroot's package/rtl8189fs is currently broken; it points to
>> the wrong commit (seems to be copy paste error, inherited from
>> package/rtl8189es). The result is unusable.
>>
>>
>> as Thomas already pointed the commit works ok for me too.
>>
>> I updated it to the correct commit (on the correct branch) and the driver
>> builds fine and loads fine.
>>
>>
>> Which branch and commit have you used? Take care that the commit actually
>> used has not the support for Linux 6.3 that has been added later.
>>
>> There are couple of issues though:
>> - for some reason the debug mode in driver is set
>>
>> -  driver seems to constantly restart the module, and it emits endless
>> debug messages
>>
>>
>> This could be bound to the branch/commit of the driver, if it includes
>> Linux 6.3 support or not.
>> But it should, otherwise it should not build.
>> Anyway I don’t know that much about other branches.
>> Maybe you mean that master branch is for rtl8189es and rtl8189fs branch
>> is for rtl8189fs?
>>
>> Also, how do you determine that rtl8189ftv is compatible with rtl8189fs?
>>
>>
>> So, is there anybody for whom the rtl8189fs linux driver works fine with
>> kernel 6.3-rcX?
>>
>>
>> PS. I can prepare a patch to fix the package/rtl8189fs but I expect Giulio
>> to bump the version again for both package/rtl8189es and
>> package/rtl8189fs soon.
>>
>>
>> Sure I can do that, there is already a pending patch for rtl8189es but
>> it’s already superseded at this point.
>> By Monday I’ll send the 2 patches or feel free to do it instead.
>>
>>
>
> Hi!
>
> Yes, for the rtl8189fs kernel module the code must be checked out from the
> rtl8189fs branch. Currently the commit hash for the package/rtl8189es and
> package/rtl8189fs is the same and it's from rtl8189ES_linux/master. It
> builds but it produces the wrong thing in the case of the rtl8189fs. I'm
> fine if it gets fixed next week ;)
> About rtl8189ftv and compatibility. The difference between rtl8188ftv and
> rtl8189ftv chip seems to be the connection interface type (usb and sdio
> accordingly). It looks to me that the rtl8189fs driver still relies on the
> 8188f specific code but it adds sdio capability (roughly speaking). Also,
> https://linux-sunxi.org/Wifi#RTL8189FTV says that the rtl8189fs branch is
> the only working solution currently (or has been at least).
> There can be several reasons why it doesn't work for me. That's why I was
> asking first about the others' experience on actual hardware, maybe with
> the different (mature) SOC/platform.
>
>
Oh, apparently the MangoPi MQ requires the external wifi antenna (which is
bundled with the board) to be connected ;) Wifi works now for me too (with
the fixed rtl8189fs commit). It seems that previously it got no signal and
was rapidly jumping in and out from power saving.

Indrek

[-- Attachment #1.2: Type: text/html, Size: 5286 bytes --]

[-- Attachment #2: Type: text/plain, Size: 150 bytes --]

_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot

Re: [Buildroot] package/rtl8189fs: bug in conf, and a question

[Giulio Benetti]

Hi Indrek,

On 19/03/23 11:07, Indrek Kruusa wrote:
> 
> 
> Kontakt Indrek Kruusa (<indrek.kruusa@gmail.com 
> <mailto:indrek.kruusa@gmail.com>>) kirjutas kuupäeval P, 19. märts 2023 
> kell 09:43:
> 
>     Kontakt Giulio Benetti (<giulio.benetti@benettiengineering.com
>     <mailto:giulio.benetti@benettiengineering.com>>) kirjutas kuupäeval
>     P, 19. märts 2023 kell 01:01:
> 
>         Hi Indkrek,
> 
>>         Il giorno 18 mar 2023, alle ore 22:46, Indrek Kruusa
>>         <indrek.kruusa@gmail.com <mailto:indrek.kruusa@gmail.com>> ha
>>         scritto:
>>
>>         
>>         Hi!
>>         I'm trying to prepare a patch for MangoPi MQ (Allwinner D1s
>>         RISC-V). I'm almost there - Samuel Holland managed to mainline
>>         initial support for this board, so vanilla linux
>>         kernel 6.3-rc2 boots fine (with proper/modified u-boot version).
>>         The WiFi module is based on rtl8189ftv chip (SDIO connection)
>>         and its driver seems to be problematic.
>>         First, buildroot's package/rtl8189fs is currently broken; it
>>         points to the wrong commit (seems to be copy paste error,
>>         inherited from package/rtl8189es). The result is unusable.
> 
>         as Thomas already pointed the commit works ok for me too.
> 
>>         I updated it to the correct commit (on the correct branch) and
>>         the driver builds fine and loads fine.
> 
>         Which branch and commit have you used? Take care that the commit
>         actually used has not the support for Linux 6.3 that has been
>         added later.
> 
>>         There are couple of issues though:
>>         - for some reason the debug mode in driver is set
>>         -  driver seems to constantly restart the module, and it emits
>>         endless debug messages
> 
>         This could be bound to the branch/commit of the driver, if it
>         includes Linux 6.3 support or not.
>         But it should, otherwise it should not build.
>         Anyway I don’t know that much about other branches.
>         Maybe you mean that master branch is for rtl8189es and rtl8189fs
>         branch is for rtl8189fs?
> 
>         Also, how do you determine that rtl8189ftv is compatible with
>         rtl8189fs?
> 
>>
>>         So, is there anybody for whom the rtl8189fs linux driver works
>>         fine with kernel 6.3-rcX?
>>
>>         PS. I can prepare a patch to fix the package/rtl8189fs but I
>>         expect Giulio to bump the version again for both
>>         package/rtl8189es and package/rtl8189fs soon.
> 
>         Sure I can do that, there is already a pending patch for
>         rtl8189es but it’s already superseded at this point.
>         By Monday I’ll send the 2 patches or feel free to do it instead.
> 
> 
> 
>     Hi!
> 
>     Yes, for the rtl8189fs kernel module the code must be checked out
>     from the rtl8189fs branch. Currently the commit hash for the
>     package/rtl8189es and package/rtl8189fs is the same and it's from
>     rtl8189ES_linux/master. It builds but it produces the wrong thing in
>     the case of the rtl8189fs. I'm fine if it gets fixed next week ;)

Fine I've checked deeper today and found the same as you've found.

>     About rtl8189ftv and compatibility. The difference between
>     rtl8188ftv and rtl8189ftv chip seems to be the connection interface
>     type (usb and sdio accordingly). It looks to me that the rtl8189fs
>     driver still relies on the 8188f specific code but it adds sdio
>     capability (roughly speaking). Also,
>     https://linux-sunxi.org/Wifi#RTL8189FTV
>     <https://linux-sunxi.org/Wifi#RTL8189FTV> says that the rtl8189fs
>     branch is the only working solution currently (or has been at least).

Ok

>     There can be several reasons why it doesn't work for me. That's why
>     I was asking first about the others' experience on actual hardware,
>     maybe with the different (mature) SOC/platform.
> 
> 
> Oh, apparently the MangoPi MQ requires the external wifi antenna (which 
> is bundled with the board) to be connected ;) Wifi works now for me too 
> (with the fixed rtl8189fs commit). It seems that previously it got no 
> signal and was rapidly jumping in and out from power saving.

Perfect, so I'm creating the 2 patches, 1 for rtl8189eu keeping master
branch but bumping to last commit and 1 for rtl8189fs that bumps to
latest commit but of rtl8189fs branch.

Thank you for pointing rtl8189fs branch and testing!

Best regards
-- 
Giulio Benetti
Benetti Engineering sas

> 
> Indrek
> 
> 
> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot

_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot