On Mon, 2013-01-28 at 19:25 +0800, Lingzhu Xiang wrote:
> On 01/27/2013 06:47 AM, Jeremy Kerr wrote:
> > + if (!(rc < 0 && errno == EIO)) {
> > + fprintf(stderr, "Reading a new var should return -EIO\n");
>
> What does the -EIO imply for reading a known empty file?
>
> The file is empty. There is nothing to be read. No IO should actually
> happens.
> The variable doesn't exist in nvram. What is the error? This won't make
> sense
> for userspace.
I was the one who originally suggested using EIO for reading a variable
that hasn't hit the backing store, mainly because it simplified the
EFI_NOT_FOUND logic, and I felt that reading from a non-existent
variable should be an error condition. But I can see the case for
returning zero instead.
> Empty variable never exists in nvram per spec. Userspace doesn't need an
> extra
> EIO to figure out this known fact. In the meantime, user would wonder if
> something else has gone wrong? Returning zero for reading an empty file can
> reserve EIO for something more informational.
Perhaps it's not unreasonable to expect users to understand that
zero-length files don't have a corresponding EFI variable, and that the
file won't persist across a reboot.
Jeremy, thoughts?
> Maybe at this time we just leave it as it is for other reasons. But at least
> I don't think it's a good idea to mandate this behavior.
I do think it's important that we hash out these semantics as soon as
possible, before userspace starts relying on them.
--
Matt Fleming, Intel Open Source Technology Center
--
To unsubscribe from this list: send the line "unsubscribe linux-efi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
