Re: [2/3] mwifiex: support sysfs initiated device coredump
On 3/13/2018 9:19 PM, Marcel Holtmann wrote: Hi Arend, Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") it is possible to initiate a device coredump from user-space. This patch adds support for it adding the .coredump() driver callback. As there is no longer a need to initiate it through debugfs remove that code. Signed-off-by: Arend van SprielBased on the discussion I assume this is ok to take to w-d-next. If that's not the case, please let me know ASAP. It is up to the mwifiex maintainers to decide, I guess. The ABI documentation need to be revised and change the callback to void return type. I am not sure what the best approach is. 1) apply this and fix return type later, or 2) fix return type and resubmit this. What is your opinion? I guess the callback change will go through Greg's tree? Then I suspect it's easier that you submit the callback change to Greg first and wait for it to trickle down to wireless-drivers-next (after the next merge window) and then I can apply the driver patches. Otherwise there might be a conflict between my and Greg's tree. That was my assessment, but unfortunately Marcel already applied the btmrvl patch before I could reply. So how do I move from here? Option 1) revert brmrvl and fix callback return type, or 2) apply mwifiex patch and fix callback return type later for both drivers. I can take the patch back out of bluetooth-next if needed. It is your call. Thanks, Marcel Let's go for that. Please revert/remove the patch. Regards, Arend
Re: [2/3] mwifiex: support sysfs initiated device coredump
Hi Arend, > Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") > it is possible to initiate a device coredump from user-space. This > patch adds support for it adding the .coredump() driver callback. > As there is no longer a need to initiate it through debugfs remove > that code. > > Signed-off-by: Arend van SprielBased on the discussion I assume this is ok to take to w-d-next. If that's not the case, please let me know ASAP. >>> >>> It is up to the mwifiex maintainers to decide, I guess. The ABI >>> documentation need to be revised and change the callback to void >>> return type. I am not sure what the best approach is. 1) apply this >>> and fix return type later, or 2) fix return type and resubmit this. >>> What is your opinion? >> >> I guess the callback change will go through Greg's tree? Then I suspect >> it's easier that you submit the callback change to Greg first and wait >> for it to trickle down to wireless-drivers-next (after the next merge >> window) and then I can apply the driver patches. Otherwise there might >> be a conflict between my and Greg's tree. > > That was my assessment, but unfortunately Marcel already applied the btmrvl > patch before I could reply. So how do I move from here? Option 1) revert > brmrvl and fix callback return type, or 2) apply mwifiex patch and fix > callback return type later for both drivers. I can take the patch back out of bluetooth-next if needed. It is your call. Regards Marcel
Re: [2/3] mwifiex: support sysfs initiated device coredump
On 3/13/2018 2:10 PM, Kalle Valo wrote: Arend van Sprielwrites: On 3/12/2018 10:41 AM, Kalle Valo wrote: Arend Van Spriel wrote: Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") it is possible to initiate a device coredump from user-space. This patch adds support for it adding the .coredump() driver callback. As there is no longer a need to initiate it through debugfs remove that code. Signed-off-by: Arend van Spriel Based on the discussion I assume this is ok to take to w-d-next. If that's not the case, please let me know ASAP. It is up to the mwifiex maintainers to decide, I guess. The ABI documentation need to be revised and change the callback to void return type. I am not sure what the best approach is. 1) apply this and fix return type later, or 2) fix return type and resubmit this. What is your opinion? I guess the callback change will go through Greg's tree? Then I suspect it's easier that you submit the callback change to Greg first and wait for it to trickle down to wireless-drivers-next (after the next merge window) and then I can apply the driver patches. Otherwise there might be a conflict between my and Greg's tree. That was my assessment, but unfortunately Marcel already applied the btmrvl patch before I could reply. So how do I move from here? Option 1) revert brmrvl and fix callback return type, or 2) apply mwifiex patch and fix callback return type later for both drivers. Regards, Arend
Re: [2/3] mwifiex: support sysfs initiated device coredump
Arend van Sprielwrites: > On 3/12/2018 10:41 AM, Kalle Valo wrote: >> Arend Van Spriel wrote: >> >>> Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") >>> it is possible to initiate a device coredump from user-space. This >>> patch adds support for it adding the .coredump() driver callback. >>> As there is no longer a need to initiate it through debugfs remove >>> that code. >>> >>> Signed-off-by: Arend van Spriel >> >> Based on the discussion I assume this is ok to take to w-d-next. If that's >> not >> the case, please let me know ASAP. > > It is up to the mwifiex maintainers to decide, I guess. The ABI > documentation need to be revised and change the callback to void > return type. I am not sure what the best approach is. 1) apply this > and fix return type later, or 2) fix return type and resubmit this. > What is your opinion? I guess the callback change will go through Greg's tree? Then I suspect it's easier that you submit the callback change to Greg first and wait for it to trickle down to wireless-drivers-next (after the next merge window) and then I can apply the driver patches. Otherwise there might be a conflict between my and Greg's tree. -- Kalle Valo
Re: [2/3] mwifiex: support sysfs initiated device coredump
On 3/12/2018 10:41 AM, Kalle Valo wrote: Arend Van Sprielwrote: Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") it is possible to initiate a device coredump from user-space. This patch adds support for it adding the .coredump() driver callback. As there is no longer a need to initiate it through debugfs remove that code. Signed-off-by: Arend van Spriel Based on the discussion I assume this is ok to take to w-d-next. If that's not the case, please let me know ASAP. It is up to the mwifiex maintainers to decide, I guess. The ABI documentation need to be revised and change the callback to void return type. I am not sure what the best approach is. 1) apply this and fix return type later, or 2) fix return type and resubmit this. What is your opinion? Regards, Arend
Re: [2/3] mwifiex: support sysfs initiated device coredump
Arend Van Sprielwrote: > Since commit 3c47d19ff4dc ("drivers: base: add coredump driver ops") > it is possible to initiate a device coredump from user-space. This > patch adds support for it adding the .coredump() driver callback. > As there is no longer a need to initiate it through debugfs remove > that code. > > Signed-off-by: Arend van Spriel Based on the discussion I assume this is ok to take to w-d-next. If that's not the case, please let me know ASAP. -- https://patchwork.kernel.org/patch/10231933/ https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches