2.6.12-rc1, ./drivers/base/platform.c

The latest prepatch, 2.6.12-rc1, introduced the following change.

drivers/base/platform.c 2005-03-17 17:35:04 -08:00 +1 -1

@@ -131,7 +131,7 @@
         pdev->dev.bus = &platform_bus_type;
 
         if (pdev->id != -1)
-                snprintf(pdev->dev.bus_id, BUS_ID_SIZE, "%s%u",
 pdev->name, pdev->id);
+                snprintf(pdev->dev.bus_id, BUS_ID_SIZE, "%s.%u",
 pdev->name, pdev->id);
         else

                 strlcpy(pdev->dev.bus_id, pdev->name, BUS_ID_SIZE);

It causes problem, as most platform files, for example, arch/ppc/platforms/katana.c, still use the old name without ".". I do not understand why bus_id "mpsc.0" is better than "mpsc0". Please explain what is the benefit of introducing such a change, before I can submit a patch for all those platform files to work with this change.
Please CC me, as I am currently not in the list.

It causes problem, as most platform files, for example, arch/ppc/platforms/katana.c, still use the old name without ".". I do

Quoted 2 lines · All Linux

It causes problem, as most platform files, for example, arch/ppc/platforms/katana.c, still use the old name without ".". I do

Mark Greer recently produced a patch for the katana board among other PPC platforms to fix this breakage. I'll look for the announcement mail but I recall seeing it a day or two ago.

Thanks
Brian