On 11/03/2017 09:01 PM, Michael Kerrisk (man-pages) wrote:
[CC += David Herrmann]
Hi Florian,
On 3 November 2017 at 20:22, Florian Weimer <fweimer@xxxxxxxxxx> wrote:
memfd_create(2) mentions a <sys/memfd.h>. Is this a typo for
<linux/memfd.h>?
Offhand, I do not know. This header file was listed in the original
draft of the page that I received from David Herrmann. (I did much
reworking of the page, but overlooked this detail.) Perhaps it was a
typo, or perhaps it was David's estimation that there would eventually
be such a glibc header.
I want to implement the system call wrapper, but we're probably going to put
it into <sys/mman.h> and have it include <linux/memfd.h> for _GNU_SOURCE.
I'm agnostic about it. Maybe David will have some thoughts.
Thanks.
My <sys/mman.h> approach is probably not such a great idea because we
can't include <linux/memfd.h> because the kernel headers may not have
it. So if we put memfd_create into <sys/mman.h>, then we'd have to
duplicate the constants. In contrast, a new header could indeed include
<linux/memfd.h>.
Florian
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
