On Wed, Dec 16, 2015 at 12:42:39AM +0300, Yury Norov wrote: > statfs64, fstat64 and mmap_pgoff has wrappers that needed both by aarch32 and > ilp32 to workaround some issues. Here we create common file to share aarch32 > workarounds to with ilp32 code. [...] > --- /dev/null > +++ b/arch/arm64/kernel/entry32-common.S > @@ -0,0 +1,37 @@ > +#include <linux/linkage.h> > +#include <linux/const.h> > + > +#include <asm/assembler.h> > +#include <asm/asm-offsets.h> > +#include <asm/errno.h> > +#include <asm/page.h> > + > +ENTRY(compat_sys_statfs64_wrapper) > + mov w3, #84 > + cmp w1, #88 > + csel w1, w3, w1, eq > + b compat_sys_statfs64 > +ENDPROC(compat_sys_statfs64_wrapper) > + > +ENTRY(compat_sys_fstatfs64_wrapper) > + mov w3, #84 > + cmp w1, #88 > + csel w1, w3, w1, eq > + b compat_sys_fstatfs64 > +ENDPROC(compat_sys_fstatfs64_wrapper)
I'm not convinced we need these wrappers for ILP32. They've been introduced on arch/arm many years ago by commit Fixes: 713c481519f1 ([ARM] 3108/2: old ABI compat: statfs64 and fstatfs64) to deal with user space passing a size of 88 (the EABI size of struct compat_statfs64 without the packing and alignment attribute). Since that commit, the sizeof(struct compat_statfs64) is 84 already. This should be the case with the new ILP32 exported headers (no backwards compatibility), so user space should never pass 88 as size. Therefore we could call compat_sys_(f)statfs64 directly without wrappers. -- Catalin -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
