Re: [PATCH] media: pci: Use PCI_DEVICE macros and drop zeros from pci_device_id arrays
From: Uwe Kleine-König (The Capable Hub)
Date: Thu May 07 2026 - 11:18:35 EST
Hello,
On Wed, May 06, 2026 at 07:44:12PM +0200, Uwe Kleine-König (The Capable Hub) wrote:
> Some pci_device_id arrays already used the PCI_DEVICE macros, others
> used named initializers or list initializers. Unify all to use the
> PCI_DEVICE macros and (if applicable) a named initializer for
> .driver_data. Also drop all zeros that the compiler can care for and
> unify indention and spacing.
>
> There are no changes introduced by this commit to the compiled arrays for
> ARCH=x86 and ARCH=arm64 builds.
>
> Signed-off-by: Uwe Kleine-König (The Capable Hub) <u.kleine-koenig@xxxxxxxxxxxx>
> ---
> Hello,
>
> this is a preparing change for making struct pci_device_id::driver_data
> an anonymous union (similar to
> https://lore.kernel.org/all/cover.1776579304.git.u.kleine-koenig@xxxxxxxxxxxx/).
> This requires named initializers for .driver_data. Not all changes here
> are required for that, but I expanded the quest to unify a bit more than
> necessary.
>
> I put all changes in a single patch, if you prefer it being split I can do that.
I just noticed I failed to put the following hunk into the patch:
diff --git a/include/media/drv-intf/saa7146.h b/include/media/drv-intf/saa7146.h
index 71ce63c99cb4..b8f2fe5680f0 100644
--- a/include/media/drv-intf/saa7146.h
+++ b/include/media/drv-intf/saa7146.h
@@ -77,13 +77,10 @@ struct saa7146_pci_extension_data {
void *ext_priv; /* most likely a name string */
};
-#define MAKE_EXTENSION_PCI(x_var, x_vendor, x_device) \
- { \
- .vendor = PCI_VENDOR_ID_PHILIPS, \
- .device = PCI_DEVICE_ID_PHILIPS_SAA7146, \
- .subvendor = x_vendor, \
- .subdevice = x_device, \
- .driver_data = (unsigned long)& x_var, \
+#define MAKE_EXTENSION_PCI(x_var, x_vendor, x_device) \
+ { \
+ PCI_VDEVICE_SUB(PHILIPS, PCI_DEVICE_ID_PHILIPS_SAA7146, (x_vendor), (x_device)),\
+ .driver_data = (unsigned long)&(x_var), \
}
struct saa7146_extension
but that's not urgent as the current definition of MAKE_EXTENSION_PCI()
is good enough to not block the next step of my quest. I'll address that
in a v2 (if it comes to that) or later together with the next step
making the driver_data line:
.driver_data_ptr = &(x_var),
(or if you apply this version, feel free to squash the diff into the
commit).
Best regards
Uwe
Attachment:
signature.asc
Description: PGP signature