Re: [PATCH v2] rust: error: add missing error codes

[Greg KH <gregkh@linuxfoundation.org>]

On Tue, May 09, 2023 at 08:07:00AM +0000, Alice Ryhl wrote:
> On Mon, 8 May 2023 12:47:01 +0100
> Gary Guo <gary@garyguo.net> wrote:
> > On Thu,  4 May 2023 06:48:54 +0000
> > Alice Ryhl <aliceryhl@google.com> wrote:
> > > This adds the error codes from `include/linux/errno.h` to the list of
> > > Rust error constants. These errors were not included originally, because
> > > they are not supposed to be visible from userspace. However, they are
> > > still a perfectly valid error to use when writing a kernel driver. For
> > > example, you might want to return ERESTARTSYS if you receive a signal
> > > during a call to `schedule`.
> > 
> > `include/linux/errno.h` also includes all of `asm/errno.h`,
> > which defines EDEADLK - EHWPOISON, which is not included in this patch.
> > I feel like these error codes should be added first?
> 
> It seems like there are a lot of asm/errno.h files:
> 
> $ find . -name errno.h
> ./arch/powerpc/include/uapi/asm/errno.h
> ./arch/mips/include/asm/errno.h
> ./arch/mips/include/uapi/asm/errno.h
> ./arch/alpha/include/uapi/asm/errno.h
> ./arch/parisc/include/uapi/asm/errno.h
> ./arch/sparc/include/uapi/asm/errno.h
> ./arch/x86/include/generated/uapi/asm/errno.h
> ./tools/arch/powerpc/include/uapi/asm/errno.h
> ./tools/arch/mips/include/asm/errno.h
> ./tools/arch/mips/include/uapi/asm/errno.h
> ./tools/arch/alpha/include/uapi/asm/errno.h
> ./tools/arch/parisc/include/uapi/asm/errno.h
> ./tools/arch/sparc/include/uapi/asm/errno.h
> ./tools/arch/x86/include/uapi/asm/errno.h
> ./tools/include/nolibc/errno.h
> ./tools/include/uapi/asm/errno.h
> ./tools/include/uapi/asm-generic/errno.h

You can ignore the tool ones.

> ./include/uapi/asm-generic/errno.h
> ./include/uapi/linux/errno.h
> ./include/linux/errno.h
> 
> How should I proceed with this? You mention EDEADLK - EHWPOISON, but its
> not clear to me which asm/errno.h file I should base this on.

It depends on which arch you are building for.  That's why we have
per-platform errno.h files, the values are different for different ones.
So you need to handle them all properly somehow.  How is rust going to
handle per-arch stuff like this?

thanks,

greg k-h