Skip to contentSkip to conversation

linux-next: manual merge of the staging tree with the net-next tree

DISCUSSION Mark Brown 4 participants 7 messages Aug 2, 2021 – Aug 9, 2021 Linux-next tree

lore
Share & tools
lore

Mark Brown

author maintainer
#0 lore raw

Hi all,

Today's linux-next merge of the staging tree got conflicts in:

drivers/staging/r8188eu/include/rtw_android.h
drivers/staging/r8188eu/os_dep/rtw_android.c

between commit:

89939e890605 ("staging: rtlwifi: use siocdevprivate")

from the net-next tree and commit:

2b42bd58b321 ("staging: r8188eu: introduce new os_dep dir for RTL8188eu driver")

from the staging tree.

I fixed it up using the net-next version and can carry the fix as necessary. This is now fixed as far as linux-next is concerned, but any non trivial conflicts should be mentioned to your upstream maintainer when your tree is submitted for merging. You may also want to consider cooperating with the maintainer of the conflicting tree to minimise any particularly complex conflicts.

#1 lore raw
Quoted 16 lines · Mark Brown

Hi all,

Today's linux-next merge of the staging tree got conflicts in:

drivers/staging/r8188eu/include/rtw_android.h
drivers/staging/r8188eu/os_dep/rtw_android.c

between commit:

89939e890605 ("staging: rtlwifi: use siocdevprivate")

from the net-next tree and commit:

2b42bd58b321 ("staging: r8188eu: introduce new os_dep dir for RTL8188eu driver")

Quoted 8 lines

from the staging tree.

I fixed it up using the net-next version and can carry the fix as necessary. This is now fixed as far as linux-next is concerned, but any non trivial conflicts should be mentioned to your upstream maintainer when your tree is submitted for merging. You may also want to consider cooperating with the maintainer of the conflicting tree to minimise any particularly complex conflicts.

Dear Mark,

Thank you for your e-mail - not sure what is going on here as the r8188eu files are totally new - the net-next commit is touching files from a different directory (rtl8188eu) so perhaps it is getting confused and considering it a rename or similar? Regardless, all commits for r8188eu subdirectory should come from staging, as it is a different driver.

Regards,
Phil

#2 lore raw
Quoted 8 lines · Phillip Potter

On Tue, 3 Aug 2021 at 00:33, Mark Brown <[email protected]> wrote:

Thank you for your e-mail - not sure what is going on here as the r8188eu files are totally new - the net-next commit is touching files from a different directory (rtl8188eu) so perhaps it is getting confused and considering it a rename or similar? Regardless, all commits for r8188eu subdirectory should come from staging, as it is a different driver.

In net-next, the .ndo_do_ioctl() callback behaves differently, as it no longer gets called from user space. If you have any SIOCDEVPRIVATE ioctls in the driver, those need to be
moved over to the new .ndo_siocdevprivate callback, while any nonstandard ioctl commands outside of that range can just be removed.

Arnd

#3 lore raw

Hi all,

Quoted 11 lines · On Tue, 3 Aug 2021 11:23:29 +0200 Arnd Bergmann <[email protected]> wrote:
Quoted 8 lines · Phillip Potter

On Tue, 3 Aug 2021 at 00:33, Mark Brown <[email protected]> wrote:

Thank you for your e-mail - not sure what is going on here as the r8188eu files are totally new - the net-next commit is touching files from a different directory (rtl8188eu) so perhaps it is getting confused and considering it a rename or similar? Regardless, all commits for r8188eu subdirectory should come from staging, as it is a different driver.

Quoted 6 lines

In net-next, the .ndo_do_ioctl() callback behaves differently, as it no longer gets called from user space. If you have any SIOCDEVPRIVATE ioctls in the driver, those need to be
moved over to the new .ndo_siocdevprivate callback, while any nonstandard ioctl commands outside of that range can just be removed.

So, can that be done in the staging tree as it is, or does Greg need to merge (part of) the net-next tree? Or will someone just supply me with a reasonable merge resolution patch? Or does Greg reset the staging tree to 5c872e1d2595 and try again next time? Or similar for Dave (revert Arnd's work)?

Currently it is a mess in linux-next and probably will not work even though it does build.

Signature · 2 lines
Cheers,
Stephen Rothwell
#4 lore raw
Quoted 28 lines · Stephen Rothwell

Hi all,

Quoted 11 lines · On Tue, 3 Aug 2021 11:23:29 +0200 Arnd Bergmann <[email protected]> wrote:
Quoted 8 lines · Phillip Potter

On Tue, 3 Aug 2021 at 00:33, Mark Brown <[email protected]> wrote:

Thank you for your e-mail - not sure what is going on here as the r8188eu files are totally new - the net-next commit is touching files from a different directory (rtl8188eu) so perhaps it is getting confused and considering it a rename or similar? Regardless, all commits for r8188eu subdirectory should come from staging, as it is a different driver.

Quoted 6 lines

In net-next, the .ndo_do_ioctl() callback behaves differently, as it no longer gets called from user space. If you have any SIOCDEVPRIVATE ioctls in the driver, those need to be
moved over to the new .ndo_siocdevprivate callback, while any nonstandard ioctl commands outside of that range can just be removed.

So, can that be done in the staging tree as it is, or does Greg need to merge (part of) the net-next tree? Or will someone just supply me with a reasonable merge resolution patch? Or does Greg reset the staging tree to 5c872e1d2595 and try again next time? Or similar for Dave (revert Arnd's work)?

Quoted 6 lines

Currently it is a mess in linux-next and probably will not work even though it does build.

--
Cheers,
Stephen Rothwell

Dear Stephen,

Happy to try and provide a merge resolution patch if that is easier? Will be this evening UK time though.

Regards,
Phil

#5 lore raw
Quoted 9 lines · Phillip Potter
Quoted 8 lines · Stephen Rothwell

On Tue, 3 Aug 2021 11:23:29 +0200 Arnd Bergmann <[email protected]> wrote:

So, can that be done in the staging tree as it is, or does Greg need to merge (part of) the net-next tree? Or will someone just supply me with a reasonable merge resolution patch? Or does Greg reset the staging tree to 5c872e1d2595 and try again next time? Or similar for Dave (revert Arnd's work)?

Quoted 6 lines
Quoted 3 lines

Currently it is a mess in linux-next and probably will not work even though it does build.

Happy to try and provide a merge resolution patch if that is easier? Will be this evening UK time though.

Most likely there is no user space that actually wants this function at all, so I think the easiest way would be to remove all of rtw_ioctl in the staging tree version of this driver.

Arnd

#6 lore raw
Quoted 15 lines · Arnd Bergmann
Quoted 13 lines · Phillip Potter
Quoted 11 lines · Stephen Rothwell

On Tue, 3 Aug 2021 11:23:29 +0200 Arnd Bergmann <[email protected]> wrote:

So, can that be done in the staging tree as it is, or does Greg need to merge (part of) the net-next tree? Or will someone just supply me with a reasonable merge resolution patch? Or does Greg reset the staging tree to 5c872e1d2595 and try again next time? Or similar for Dave (revert Arnd's work)?

Currently it is a mess in linux-next and probably will not work even though it does build.

Quoted 8 lines
Quoted 2 lines

Happy to try and provide a merge resolution patch if that is easier? Will be this evening UK time though.

Most likely there is no user space that actually wants this function at all, so I think the easiest way would be to remove all of rtw_ioctl in the staging tree version of this driver.

Arnd

Sounds good. I will put something together to take out rtw_ioctl in that case, and submit through Greg/linux-staging. Many thanks.

Regards,
Phil