Hi!
Add a document describing the process of adding a new system call,
including the need for a flags argument for future compatibility, and
covering 32-bit/64-bit concerns (albeit in an x86-centric way).
Signed-off-by: David Drysdale <redacted>
Reviewed-by: Michael Kerrisk <redacted>
Reviewed-by: Eric B Munson <redacted>
Reviewed-by: Kees Cook <redacted>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
---
Documentation/adding-syscalls.txt | 531 ++++++++++++++++++++++++++++++++++++++
1 file changed, 531 insertions(+)
create mode 100644 Documentation/adding-syscalls.txt
We usually align documentation to less than 80 columns, afaict.
quoted hunk ↗ jump to hunk
+call. To make sure that userspace programs can safely use flags between kernel
+versions, check whether the flags value holds any unknown flags, and reject the
+sycall (with EINVAL) if it does:
syscall?
+New system call proposals, like any change to the kernel's API, should always
+be cc'ed to linux-api@vger.kernel.org
. at and of sentence?
quoted hunk ↗ jump to hunk
+System Calls Returning Elsewhere
+--------------------------------
+
+For most system calls, once the system call is complete the user program
+continues exactly where it left off -- at the next instruction, with the same
+stack and registers as before the system call, and with the same virtual
+memory space.
Umm. Normally we place return value in register. And I'm not sure.. do
syscalls preserve registers that are normally caller-clobbered in the ABI?
Thanks,
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html