Re: [maemo-leste] [PATCH] nokia_rx51: disable obsolete VIDEO config

On Sat, Oct 24, 2020 at 12:46:45PM +0200, Merlijn Wajer wrote:
Hi,
On 24/10/2020 12:06, Pali Rohár wrote:
On Wednesday 21 October 2020 08:47:02 Tom Rini wrote:
On Wed, Oct 21, 2020 at 11:08:37AM +0200, Pali Rohár wrote:
On Tuesday 20 October 2020 10:17:07 Tom Rini wrote:
On Tue, Oct 20, 2020 at 11:30:51AM +0200, Pali Rohár wrote:
<snip> >>> Yes, there is mmc issue, I have found which commit caused it and >>> because nobody was able to debug / understand what is happening I have >>> sent revert patch, which fixes that issue, until somebody solve issue in >>> original patch. >>> >>> But I do not see any progress on either applying revert patch or fixing >>> original patch. >>> >>> And I do not like to have custom patch in my local branch (and >>> periodically rebasing it) as it just another layer of complication. >>> >>> Could we move somehow in this issue? >> >> No, it's probably stuck until you can fix the problem or otherwise add a >> "quirk" for your platform. What does Linux do in this case? > > Well, I provided all requested information, just I'm lacking any new > response from developers / maintainers of particular code. > > I have already identified problematic place / commit and I have already > sent revert patch which fixing this issue. > > Tom, what else do you need? > > I see the main issue now that nobody reviewed / merged patch and nobody > responded to emails which I have sent more then 2 months ago.
I would like to add my voice for reverting the patches that break the support, and work from that point to improve support and eventually migrate to DT.
Maemo Leste has a lot of Nokia N900 users, and this is blocking us from moving to mainline u-boot. Is it possible to get the opinion of the original patch authors on this?
I am set up with a serial module on the Nokia N900, so I can help provide specific debug info.
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.

On Saturday 24 October 2020 08:19:31 Tom Rini wrote:
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
Well, I do not see any warnings at build, I'm building U-Boot by command:
$ make nokia_rx51_defconfig && make SOURCE_DATE_EPOCH=0 ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- u-boot.bin
Maybe I'm doing something wrong that I do not see it? Or is there any other way to "list all issues for N900" which needs to be addressed?
Anyway, I'm looking at DM and as the first step I will do conversion of code to CONFIG_DM_I2C. It is OK for you? Maybe it brings some light to non-working second i2c bus.
I hope that TI (or other) developers look at that mmc issue as this is blocking all N900 users.

On Mon, Oct 26, 2020 at 12:04:54AM +0100, Pali Rohár wrote:
On Saturday 24 October 2020 08:19:31 Tom Rini wrote:
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
Well, I do not see any warnings at build, I'm building U-Boot by command:
$ make nokia_rx51_defconfig && make SOURCE_DATE_EPOCH=0 ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- u-boot.bin
Note that ARCH=arm is meaningless in U-Boot.
Maybe I'm doing something wrong that I do not see it? Or is there any other way to "list all issues for N900" which needs to be addressed?
Yes, the "all" target.
Anyway, I'm looking at DM and as the first step I will do conversion of code to CONFIG_DM_I2C. It is OK for you? Maybe it brings some light to non-working second i2c bus.
I hope that TI (or other) developers look at that mmc issue as this is blocking all N900 users.
Maybe converting to DM_MMC will also clear up or shed more light on the problem you have there as well.

On Sunday 25 October 2020 19:55:28 Tom Rini wrote:
On Mon, Oct 26, 2020 at 12:04:54AM +0100, Pali Rohár wrote:
On Saturday 24 October 2020 08:19:31 Tom Rini wrote:
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
Well, I do not see any warnings at build, I'm building U-Boot by command:
$ make nokia_rx51_defconfig && make SOURCE_DATE_EPOCH=0 ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- u-boot.bin
Note that ARCH=arm is meaningless in U-Boot.
Maybe I'm doing something wrong that I do not see it? Or is there any other way to "list all issues for N900" which needs to be addressed?
Yes, the "all" target.
Ok, now I see them! I used in all my scripts only u-boot.bin target (as I needed only this one binary), so basically all warnings were perfectly hidden also in past and I did not know about it...
Maybe there are more people who do not know about these warnings? What about printing them as part of u-boot.bin target?
Anyway, I'm looking at DM and as the first step I will do conversion of code to CONFIG_DM_I2C. It is OK for you? Maybe it brings some light to non-working second i2c bus.
I hope that TI (or other) developers look at that mmc issue as this is blocking all N900 users.
Maybe converting to DM_MMC will also clear up or shed more light on the problem you have there as well.
-- Tom

On Mon, Oct 26, 2020 at 01:19:43AM +0100, Pali Rohár wrote:
On Sunday 25 October 2020 19:55:28 Tom Rini wrote:
On Mon, Oct 26, 2020 at 12:04:54AM +0100, Pali Rohár wrote:
On Saturday 24 October 2020 08:19:31 Tom Rini wrote:
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
Well, I do not see any warnings at build, I'm building U-Boot by command:
$ make nokia_rx51_defconfig && make SOURCE_DATE_EPOCH=0 ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- u-boot.bin
Note that ARCH=arm is meaningless in U-Boot.
Maybe I'm doing something wrong that I do not see it? Or is there any other way to "list all issues for N900" which needs to be addressed?
Yes, the "all" target.
Ok, now I see them! I used in all my scripts only u-boot.bin target (as I needed only this one binary), so basically all warnings were perfectly hidden also in past and I did not know about it...
Maybe there are more people who do not know about these warnings? What about printing them as part of u-boot.bin target?
I was wondering about that too, but I'm not sure.

On Mon, 26 Oct 2020 00:04:54 +0100 Pali Rohár pali@kernel.org wrote: ...
Maybe I'm doing something wrong that I do not see it? Or is there any other way to "list all issues for N900" which needs to be addressed?
$ ./tools/buildman/buildman nokia_rx51 Building current source for 1 boards (1 thread, 12 jobs per thread) arm: w+ nokia_rx51 +===================== WARNING ====================== +This board does not use CONFIG_DM. CONFIG_DM will be +compulsory starting with the v2020.01 release. +Failure to update may result in board removal. +See doc/driver-model/migration.rst for more info. +==================================================== +This board does not use CONFIG_DM_MMC. Please update +the board to use CONFIG_DM_MMC before the v2019.04 release. +Failure to update by the deadline may result in board removal. +This board does not use CONFIG_DM_USB. Please update +the board to use CONFIG_DM_USB before the v2019.07 release. +This board does not use CONFIG_DM_VIDEO Please update +the board to use CONFIG_DM_VIDEO before the v2019.07 release. +This board does not use CONFIG_WDT (DM watchdog support). +Please update the board to use CONFIG_WDT before the +v2019.10 release. 0 1 0 /1 nokia_rx51 Completed: 1 total built, duration 0:00:04, rate 0.25
-- Anatolij

On 24/10/20 5:49 pm, Tom Rini wrote:
On Sat, Oct 24, 2020 at 12:46:45PM +0200, Merlijn Wajer wrote:
Hi,
On 24/10/2020 12:06, Pali Rohár wrote:
On Wednesday 21 October 2020 08:47:02 Tom Rini wrote:
On Wed, Oct 21, 2020 at 11:08:37AM +0200, Pali Rohár wrote:
On Tuesday 20 October 2020 10:17:07 Tom Rini wrote:
On Tue, Oct 20, 2020 at 11:30:51AM +0200, Pali Rohár wrote:
<snip> >>> Yes, there is mmc issue, I have found which commit caused it and >>> because nobody was able to debug / understand what is happening I have >>> sent revert patch, which fixes that issue, until somebody solve issue in >>> original patch. >>> >>> But I do not see any progress on either applying revert patch or fixing >>> original patch. >>> >>> And I do not like to have custom patch in my local branch (and >>> periodically rebasing it) as it just another layer of complication. >>> >>> Could we move somehow in this issue? >> >> No, it's probably stuck until you can fix the problem or otherwise add a >> "quirk" for your platform. What does Linux do in this case? > > Well, I provided all requested information, just I'm lacking any new > response from developers / maintainers of particular code. > > I have already identified problematic place / commit and I have already > sent revert patch which fixing this issue. > > Tom, what else do you need? > > I see the main issue now that nobody reviewed / merged patch and nobody > responded to emails which I have sent more then 2 months ago.
I would like to add my voice for reverting the patches that break the support, and work from that point to improve support and eventually migrate to DT.
Maemo Leste has a lot of Nokia N900 users, and this is blocking us from moving to mainline u-boot. Is it possible to get the opinion of the original patch authors on this?
I am set up with a serial module on the Nokia N900, so I can help provide specific debug info.
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
IIUC, the breakage is happening in non-DM code. I don't think non-DM code is maintained anymore for mmc. As Tom suggested we should start using DM for Nokia RX51
Thanks and regards, Lokesh

On Thursday 29 October 2020 14:06:30 Lokesh Vutla wrote:
On 24/10/20 5:49 pm, Tom Rini wrote:
On Sat, Oct 24, 2020 at 12:46:45PM +0200, Merlijn Wajer wrote:
Hi,
On 24/10/2020 12:06, Pali Rohár wrote:
On Wednesday 21 October 2020 08:47:02 Tom Rini wrote:
On Wed, Oct 21, 2020 at 11:08:37AM +0200, Pali Rohár wrote:
On Tuesday 20 October 2020 10:17:07 Tom Rini wrote: > On Tue, Oct 20, 2020 at 11:30:51AM +0200, Pali Rohár wrote:
<snip> >>> Yes, there is mmc issue, I have found which commit caused it and >>> because nobody was able to debug / understand what is happening I have >>> sent revert patch, which fixes that issue, until somebody solve issue in >>> original patch. >>> >>> But I do not see any progress on either applying revert patch or fixing >>> original patch. >>> >>> And I do not like to have custom patch in my local branch (and >>> periodically rebasing it) as it just another layer of complication. >>> >>> Could we move somehow in this issue? >> >> No, it's probably stuck until you can fix the problem or otherwise add a >> "quirk" for your platform. What does Linux do in this case? > > Well, I provided all requested information, just I'm lacking any new > response from developers / maintainers of particular code. > > I have already identified problematic place / commit and I have already > sent revert patch which fixing this issue. > > Tom, what else do you need? > > I see the main issue now that nobody reviewed / merged patch and nobody > responded to emails which I have sent more then 2 months ago.
I would like to add my voice for reverting the patches that break the support, and work from that point to improve support and eventually migrate to DT.
Maemo Leste has a lot of Nokia N900 users, and this is blocking us from moving to mainline u-boot. Is it possible to get the opinion of the original patch authors on this?
I am set up with a serial module on the Nokia N900, so I can help provide specific debug info.
Let me add in the TI folks that might be able to look in to any of this today. But the next problem is that the rx51 needs to be converted to use DM in U-Boot. It's throwing up a ton of "you need to convert to X by <some release in 2019> or this board may be removed" warnings at build. I really really want to see some of this start to be addressed.
IIUC, the breakage is happening in non-DM code.
No, this code is not non-DM specific. It is used also for DM builds.
participants (4)
-
Anatolij Gustschin
-
Lokesh Vutla
-
Pali Rohár
-
Tom Rini