On the kernel 5.9 there's support for the RTL8821CE chipset, but my
variant seems to not be supported:
Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: rfe 2 isn't supported
Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: failed to setup
chip efuse info
Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: failed to setup
chip information
If you need any further information please reply with what steps I should take.
There's no wifi working at all here, and those are all of the messages I get.
Thanks you!
On Sun, Oct 11, 2020 at 9:13 AM David Rubio
<[email protected]> wrote:
> Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: rfe 2 isn't supported
I'm not sure anything changed since the last time this came up:
https://lore.kernel.org/linux-wireless/[email protected]/
I think this needs somebody from Realtek to look, or else ignore the
objections and merge the sub-optimal (trivial) patch. I'm not sure of
the ramifications of doing the latter.
Brian
On 10/12/20 9:11 PM, Andy Huang wrote:
> The support for RFE type 2 of 8821ce, the patches are ready, but we need to do some verifications. After the verification it will be upload.
>
> Tzu-En
>
> -----Original Message-----
> From: Brian Norris <[email protected]>
> Sent: Tuesday, October 13, 2020 2:13 AM
> To: David Rubio <[email protected]>
> Cc: linux-wireless <[email protected]>; Yan-Hsuan Chuang <[email protected]>; Kai-Heng Feng <[email protected]>
> Subject: Re: rtw88 / rtl_8821ce: rfe 2 is not supported
>
> On Sun, Oct 11, 2020 at 9:13 AM David Rubio <[email protected]> wrote:
>> Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: rfe 2 isn't
>> supported
>
> I'm not sure anything changed since the last time this came up:
>
> https://lore.kernel.org/linux-wireless/[email protected]/
>
> I think this needs somebody from Realtek to look, or else ignore the objections and merge the sub-optimal (trivial) patch. I'm not sure of the ramifications of doing the latter.
>
Once the patches are uploaded, I will merge them with rtw88 at GitHub.
Larry
The support for RFE type 2 of 8821ce, the patches are ready, but we need to do some verifications. After the verification it will be upload.
Tzu-En
-----Original Message-----
From: Brian Norris <[email protected]>
Sent: Tuesday, October 13, 2020 2:13 AM
To: David Rubio <[email protected]>
Cc: linux-wireless <[email protected]>; Yan-Hsuan Chuang <[email protected]>; Kai-Heng Feng <[email protected]>
Subject: Re: rtw88 / rtl_8821ce: rfe 2 is not supported
On Sun, Oct 11, 2020 at 9:13 AM David Rubio <[email protected]> wrote:
> Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: rfe 2 isn't
> supported
I'm not sure anything changed since the last time this came up:
https://lore.kernel.org/linux-wireless/[email protected]/
I think this needs somebody from Realtek to look, or else ignore the objections and merge the sub-optimal (trivial) patch. I'm not sure of the ramifications of doing the latter.
Brian
------Please consider the environment before printing this e-mail.
> https://lore.kernel.org/linux-wireless/[email protected]/
I tested that patch. Works fine for me for wifi, but I can't test BT
to be sure it works 100%. Most people will be fine with just wifi
though, I guess, considering the objections were mostly about BT (I
understood -from the objection- that connecting to a AP when having a
BT device paired breaks?)
Thanks for making me aware of it!
- Kodehawa
El lun., 12 oct. 2020 a las 23:37, Larry Finger
(<[email protected]>) escribió:
>
> On 10/12/20 9:11 PM, Andy Huang wrote:
> > The support for RFE type 2 of 8821ce, the patches are ready, but we need to do some verifications. After the verification it will be upload.
> >
> > Tzu-En
> >
> > -----Original Message-----
> > From: Brian Norris <[email protected]>
> > Sent: Tuesday, October 13, 2020 2:13 AM
> > To: David Rubio <[email protected]>
> > Cc: linux-wireless <[email protected]>; Yan-Hsuan Chuang <[email protected]>; Kai-Heng Feng <[email protected]>
> > Subject: Re: rtw88 / rtl_8821ce: rfe 2 is not supported
> >
> > On Sun, Oct 11, 2020 at 9:13 AM David Rubio <[email protected]> wrote:
> >> Oct 11 12:20:39 youmu kernel: rtw_8821ce 0000:02:00.0: rfe 2 isn't
> >> supported
> >
> > I'm not sure anything changed since the last time this came up:
> >
> > https://lore.kernel.org/linux-wireless/[email protected]/
> >
> > I think this needs somebody from Realtek to look, or else ignore the objections and merge the sub-optimal (trivial) patch. I'm not sure of the ramifications of doing the latter.
> >
>
> Once the patches are uploaded, I will merge them with rtw88 at GitHub.
>
> Larry
>
David Rubio <[email protected]> writes:
>> https://lore.kernel.org/linux-wireless/[email protected]/
>
> I tested that patch. Works fine for me for wifi, but I can't test BT
> to be sure it works 100%. Most people will be fine with just wifi
> though, I guess, considering the objections were mostly about BT (I
> understood -from the objection- that connecting to a AP when having a
> BT device paired breaks?)
If the patch helps people to get wifi working we should take it, BT coex
issues can be fixed in followup patches. IIRC there has been multiple
reports about this so I'm leaning towards taking the patch to v5.11.
I changed the patch to New state and my plan is to take it to
wireless-drivers-next once the tree opens:
https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
There's no possibility of including it on 5.10 changes? Seems like a
trivial patch
Or did the merge window pass already? I'm not aware of it by now.
Thanks though. Very much appreciated.
El dom., 18 oct. 2020 a las 6:11, Kalle Valo (<[email protected]>) escribió:
>
> David Rubio <[email protected]> writes:
>
> >> https://lore.kernel.org/linux-wireless/[email protected]/
> >
> > I tested that patch. Works fine for me for wifi, but I can't test BT
> > to be sure it works 100%. Most people will be fine with just wifi
> > though, I guess, considering the objections were mostly about BT (I
> > understood -from the objection- that connecting to a AP when having a
> > BT device paired breaks?)
>
> If the patch helps people to get wifi working we should take it, BT coex
> issues can be fixed in followup patches. IIRC there has been multiple
> reports about this so I'm leaning towards taking the patch to v5.11.
>
> I changed the patch to New state and my plan is to take it to
> wireless-drivers-next once the tree opens:
>
> https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
>
> --
> https://patchwork.kernel.org/project/linux-wireless/list/
>
> https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
David Rubio <[email protected]> writes:
> There's no possibility of including it on 5.10 changes? Seems like a
> trivial patch Or did the merge window pass already? I'm not aware of
> it by now.
wireless-drivers-next goes via net-next and the deadline for net-next is
before the merge window opens, so it's too late for v5.10.
--
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
On 10/18/20 4:11 AM, Kalle Valo wrote:
> David Rubio <[email protected]> writes:
>
>>> https://lore.kernel.org/linux-wireless/[email protected]/
>>
>> I tested that patch. Works fine for me for wifi, but I can't test BT
>> to be sure it works 100%. Most people will be fine with just wifi
>> though, I guess, considering the objections were mostly about BT (I
>> understood -from the objection- that connecting to a AP when having a
>> BT device paired breaks?)
>
> If the patch helps people to get wifi working we should take it, BT coex
> issues can be fixed in followup patches. IIRC there has been multiple
> reports about this so I'm leaning towards taking the patch to v5.11.
>
> I changed the patch to New state and my plan is to take it to
> wireless-drivers-next once the tree opens:
>
> https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
Kalle,
I had generated and applied that trivial patch to my GitHub repo with the rtw88
drivers a couple of months ago. Yes, it does get the user past the
initialization check; however, wifi performance is abysmal according the the
users of the repo. It seems that the antenna selection of rfe 2 models affects
wifi as well as BT. Applying this patch will get wifi running; however, the
users will need to be within 1 m of the AP for it to work! I do not have an
RTL8821CE chip, thus I have not tested myself.
Larry
Makes sense I guess. My room is 7m^2 so can't test beyond that.
Small room issues, or well, not issues, I guess.
- Kodehawa
El lun., 19 oct. 2020 a las 14:10, Larry Finger
(<[email protected]>) escribió:
>
> On 10/18/20 4:11 AM, Kalle Valo wrote:
> > David Rubio <[email protected]> writes:
> >
> >>> https://lore.kernel.org/linux-wireless/[email protected]/
> >>
> >> I tested that patch. Works fine for me for wifi, but I can't test BT
> >> to be sure it works 100%. Most people will be fine with just wifi
> >> though, I guess, considering the objections were mostly about BT (I
> >> understood -from the objection- that connecting to a AP when having a
> >> BT device paired breaks?)
> >
> > If the patch helps people to get wifi working we should take it, BT coex
> > issues can be fixed in followup patches. IIRC there has been multiple
> > reports about this so I'm leaning towards taking the patch to v5.11.
> >
> > I changed the patch to New state and my plan is to take it to
> > wireless-drivers-next once the tree opens:
> >
> > https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
>
> Kalle,
>
> I had generated and applied that trivial patch to my GitHub repo with the rtw88
> drivers a couple of months ago. Yes, it does get the user past the
> initialization check; however, wifi performance is abysmal according the the
> users of the repo. It seems that the antenna selection of rfe 2 models affects
> wifi as well as BT. Applying this patch will get wifi running; however, the
> users will need to be within 1 m of the AP for it to work! I do not have an
> RTL8821CE chip, thus I have not tested myself.
>
> Larry
>
David Rubio <[email protected]> writes,
>
> Makes sense I guess. My room is 7m^2 so can't test beyond that.
> Small room issues, or well, not issues, I guess.
>
> - Kodehawa
>
> El lun., 19 oct. 2020 a las 14:10, Larry Finger
> (<[email protected]>) escribió:
> >
> > On 10/18/20 4:11 AM, Kalle Valo wrote:
> > > David Rubio <[email protected]> writes:
> > >
> > >>>
> https://lore.kernel.org/linux-wireless/c0c336d806584361992d4b52665fbb82@r
> ealtek.com/
> > >>
> > >> I tested that patch. Works fine for me for wifi, but I can't test BT
> > >> to be sure it works 100%. Most people will be fine with just wifi
> > >> though, I guess, considering the objections were mostly about BT (I
> > >> understood -from the objection- that connecting to a AP when having a
> > >> BT device paired breaks?)
> > >
> > > If the patch helps people to get wifi working we should take it, BT coex
> > > issues can be fixed in followup patches. IIRC there has been multiple
> > > reports about this so I'm leaning towards taking the patch to v5.11.
> > >
> > > I changed the patch to New state and my plan is to take it to
> > > wireless-drivers-next once the tree opens:
> > >
> > >
> https://patchwork.kernel.org/project/linux-wireless/patch/20200805084559.30
> [email protected]/
> >
> > Kalle,
> >
> > I had generated and applied that trivial patch to my GitHub repo with the
> rtw88
> > drivers a couple of months ago. Yes, it does get the user past the
> > initialization check; however, wifi performance is abysmal according the the
> > users of the repo. It seems that the antenna selection of rfe 2 models affects
> > wifi as well as BT. Applying this patch will get wifi running; however, the
> > users will need to be within 1 m of the AP for it to work! I do not have an
> > RTL8821CE chip, thus I have not tested myself.
> >
> > Larry
> >
>
David,
What is the band for the test, 2.4GHz or 5GHz?
Without setting to a correct rf module for RFE type 2, you might encounter some
problems on 2.4GHz band.
In my test, on 2.4GHz band, with the wrong setting, it could not connect to an AP.
Tzu-En
5GHz. Haven't really tried 2.4GHz.
So basically with that patch, 5GHz works ok-ish, and 2.4GHz is broken?
- Kodehawa
El mié., 21 oct. 2020 a las 0:41, Andy Huang (<[email protected]>) escribió:
>
>
> David Rubio <[email protected]> writes,
> >
> > Makes sense I guess. My room is 7m^2 so can't test beyond that.
> > Small room issues, or well, not issues, I guess.
> >
> > - Kodehawa
> >
> > El lun., 19 oct. 2020 a las 14:10, Larry Finger
> > (<[email protected]>) escribió:
> > >
> > > On 10/18/20 4:11 AM, Kalle Valo wrote:
> > > > David Rubio <[email protected]> writes:
> > > >
> > > >>>
> > https://lore.kernel.org/linux-wireless/c0c336d806584361992d4b52665fbb82@r
> > ealtek.com/
> > > >>
> > > >> I tested that patch. Works fine for me for wifi, but I can't test BT
> > > >> to be sure it works 100%. Most people will be fine with just wifi
> > > >> though, I guess, considering the objections were mostly about BT (I
> > > >> understood -from the objection- that connecting to a AP when having a
> > > >> BT device paired breaks?)
> > > >
> > > > If the patch helps people to get wifi working we should take it, BT coex
> > > > issues can be fixed in followup patches. IIRC there has been multiple
> > > > reports about this so I'm leaning towards taking the patch to v5.11.
> > > >
> > > > I changed the patch to New state and my plan is to take it to
> > > > wireless-drivers-next once the tree opens:
> > > >
> > > >
> > https://patchwork.kernel.org/project/linux-wireless/patch/20200805084559.30
> > [email protected]/
> > >
> > > Kalle,
> > >
> > > I had generated and applied that trivial patch to my GitHub repo with the
> > rtw88
> > > drivers a couple of months ago. Yes, it does get the user past the
> > > initialization check; however, wifi performance is abysmal according the the
> > > users of the repo. It seems that the antenna selection of rfe 2 models affects
> > > wifi as well as BT. Applying this patch will get wifi running; however, the
> > > users will need to be within 1 m of the AP for it to work! I do not have an
> > > RTL8821CE chip, thus I have not tested myself.
> > >
> > > Larry
> > >
> >
> David,
>
> What is the band for the test, 2.4GHz or 5GHz?
> Without setting to a correct rf module for RFE type 2, you might encounter some
> problems on 2.4GHz band.
> In my test, on 2.4GHz band, with the wrong setting, it could not connect to an AP.
>
> Tzu-En
Larry Finger <[email protected]> writes:
> On 10/18/20 4:11 AM, Kalle Valo wrote:
>> David Rubio <[email protected]> writes:
>>
>>>>
>>> https://lore.kernel.org/linux-wireless/[email protected]/
>>>
>>> I tested that patch. Works fine for me for wifi, but I can't test BT
>>> to be sure it works 100%. Most people will be fine with just wifi
>>> though, I guess, considering the objections were mostly about BT (I
>>> understood -from the objection- that connecting to a AP when having a
>>> BT device paired breaks?)
>>
>> If the patch helps people to get wifi working we should take it, BT coex
>> issues can be fixed in followup patches. IIRC there has been multiple
>> reports about this so I'm leaning towards taking the patch to v5.11.
>>
>> I changed the patch to New state and my plan is to take it to
>> wireless-drivers-next once the tree opens:
>>
>> https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/
>
> Kalle,
>
> I had generated and applied that trivial patch to my GitHub repo with
> the rtw88 drivers a couple of months ago. Yes, it does get the user
> past the initialization check; however, wifi performance is abysmal
> according the the users of the repo. It seems that the antenna
> selection of rfe 2 models affects wifi as well as BT. Applying this
> patch will get wifi running; however, the users will need to be within
> 1 m of the AP for it to work! I do not have an RTL8821CE chip, thus I
> have not tested myself.
Ok, I will drop the patch then. But it would be really nice to enable
even some level of support for rfe 2, can't we find some solution? Like
disabling 2.4 GHz band for that rfe or something?
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
On 10/22/20 12:53 AM, Kalle Valo wrote:
> Ok, I will drop the patch then. But it would be really nice to enable
> even some level of support for rfe 2, can't we find some solution? Like
> disabling 2.4 GHz band for that rfe or something?
The E-mail from Realtek
(https://marc.info/?l=linux-wireless&m=160255510004938&w=2) says that the
patches are ready; however, they need to do some "verification". To me (without
proof), that implies a regulatory problem, and it is better that we not enable
rfe 2 without the official patches.
I hope those patches will be forthcoming very soon.
Larry
From: Larry Finger <[email protected]> On Behalf Of Larry Finger
>
> On 10/22/20 12:53 AM, Kalle Valo wrote:
> > Ok, I will drop the patch then. But it would be really nice to enable
> > even some level of support for rfe 2, can't we find some solution? Like
> > disabling 2.4 GHz band for that rfe or something?
>
> The E-mail from Realtek
> (https://marc.info/?l=linux-wireless&m=160255510004938&w=2) says that the
> patches are ready; however, they need to do some "verification". To me
> (without
> proof), that implies a regulatory problem, and it is better that we not enable
> rfe 2 without the official patches.
>
> I hope those patches will be forthcoming very soon.
>
> Larry
>
The verification is not related to regulatory. It is that our QA team tests the some
functionalities (performance, BT coex etc.) of RFE type 2. If the test run successfully,
the patch will be sent in the next 2 weeks.
Tzu-En