On Wed, Apr 22, 2020 at 10:17 PM Jessica Yu <jeyu@xxxxxxxxxx> wrote: > > +++ Masahiro Yamada [22/04/20 01:13 +0900]: > [snip] > >diff --git a/arch/xtensa/include/asm/module.h b/arch/xtensa/include/asm/vermagic.h > >similarity index 72% > >rename from arch/xtensa/include/asm/module.h > >rename to arch/xtensa/include/asm/vermagic.h > >index 488b40c6f9b9..6f9e359a54ac 100644 > >--- a/arch/xtensa/include/asm/module.h > >+++ b/arch/xtensa/include/asm/vermagic.h > >@@ -1,6 +1,4 @@ > > /* > >- * include/asm-xtensa/module.h > >- * > > * This file contains the module code specific to the Xtensa architecture. > > Maybe we can remove this comment too? Since it's now asm/vermagic.h and > not asm/module.h anymore. OK, I will delete it. Thanks for checking it closely. > > Thanks for the cleanup. I agree that <linux/vermagic.h> shouldn't have > any ordering dependency on <linux/module.h>. > > I just double checked to see if there were any other users of > MODULE_ARCH_VERMAGIC that needed it through module.h, and there are > none. It was literally just being defined in asm/module.h to be used > in linux/vermagic.h. So there was no reason really to confine the > MODULE_ARCH_VERMAGIC definition to asm/module.h. > > Acked-by: Jessica Yu <jeyu@xxxxxxxxxx> -- Best Regards Masahiro Yamada
