Re: linux-next: manual merge of the scsi-mkp tree with the origin tree

[Uwe Kleine-König]

Hello Mark,

On Mon, Jul 13, 2026 at 04:13:24PM +0100, Mark Brown wrote:
> Today's linux-next merge of the scsi-mkp tree got a conflict in:
> 
>   include/linux/mod_devicetable.h
> 
> between commit:
> 
>   ad428f5811bd7 ("mod_devicetable.h: Split into per subsystem headers")
> 
> from the origin tree and commit:
> 
>   e73ba3d6ed03b ("scsi: zorro: Simplify storing pointers in device id struct")
> 
> from the scsi-mkp tree.
> 
> I fixed it up (see below) 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.
> 
> diff --cc include/linux/mod_devicetable.h
> index a397213bedace,2673a1bd82c45..0000000000000
> --- a/include/linux/mod_devicetable.h
> +++ b/include/linux/mod_devicetable.h
> diff --git a/include/linux/device-id/zorro.h b/include/linux/device-id/zorro.h
> index 5fdac81689839..26827b9b90b67 100644
> --- a/include/linux/device-id/zorro.h
> +++ b/include/linux/device-id/zorro.h
> @@ -13,7 +13,11 @@ typedef unsigned long kernel_ulong_t;
>  
>  struct zorro_device_id {
>  	__u32 id;			/* Device ID or ZORRO_WILDCARD */
> -	kernel_ulong_t driver_data;	/* Data private to the driver */
> +	union {
> +		/* Data private to the driver */
> +		kernel_ulong_t driver_data;
> +		const void *driver_data_ptr;
> +	};
>  };

While this looks right, I think it's wrong in git:

$ git show next-20260715~32
commit b36e50419ebc00bef45b574c6ce3e7edbbda5eb9
Merge: 20361c12febf cf1af0ccca54
Author: Mark Brown <broonie@xxxxxxxxxx>
Date:   Wed Jul 15 14:23:48 2026 +0100

    Merge branch 'for-next' of https://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi.git

diff --cc include/linux/device-id/zorro.h
index 5fdac8168983,000000000000..26827b9b90b6
mode 100644,000000..100644
--- a/include/linux/device-id/zorro.h
+++ b/include/linux/device-id/zorro.h
@@@ -1,19 -1,0 +1,23 @@@
 +/* SPDX-License-Identifier: GPL-2.0 */
 +#ifndef LINUX_DEVICE_ID_ZORRO_H
 +#define LINUX_DEVICE_ID_ZORRO_H
 +
 +#ifdef __KERNEL__
 +#include <linux/types.h>
 +typedef unsigned long kernel_ulong_t;
 +#endif
 +
 +#define ZORRO_WILDCARD                        (0xffffffff)    /* not official */
 +
 +#define ZORRO_DEVICE_MODALIAS_FMT     "zorro:i%08X"
 +
 +struct zorro_device_id {
 +      __u32 id;                       /* Device ID or ZORRO_WILDCARD */
-       kernel_ulong_t driver_data;     /* Data private to the driver */
++      union {
++              /* Data private to the driver */
++              kernel_ulong_t driver_data;
++              const void *driver_data_ptr;
++      };
 +};
 +
 +#endif /* ifndef LINUX_DEVICE_ID_ZORRO_H */

So the fixup is applied to the merge of the scsi tree and not the
scsi-mkp tree. The latter wasn't pulled into next-20260715 at all. Given
that e73ba3d6ed03b isn't in next, the fixup shouldn't be there either.

Looking at merge.log I see merging scsi-mkp is tried, the fixup is
applied, yielding 0fd66dadace89 and then `git reset --hard HEAD^`
follows throwing away that commit to replace it by
next-20260714/scsi-mkp which is already included and thus the
patches/device-id-zorro fixup is applied to the wrong commit.

Having said that e73ba3d6ed03b is still in
https://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi.git for-next
without the patch preparing drivers/ata/pata_buddha.c for that change
and thus resulting in a build failure on m68k.

Best regards
Uwe

Attachment: signature.asc
Description: PGP signature