Re: [LTP] The Best Practice to Skip Syscall Tests

[Mike Frysinger <vapier@gentoo.org>]

[-- Attachment #1.1: Type: Text/Plain, Size: 2586 bytes --]

On Saturday 22 August 2009 22:30:37 CAI Qian wrote:
> From: Mike Frysinger <vapier@gentoo.org>
> > On Friday 21 August 2009 07:57:39 CAI Qian wrote:
> >> Test cases unshare failed to compile on systems without such a
> >> syscall. I am unsure what the best way to handle it. The old ways are,
> >>
> >> * add a kernel version check in the unshare tests code.
> >>
> >> or
> >>
> >> * add a m4 rule.
> >> * modify the unshare tests code to check if the macro, and add stub
> >>   functions if the macro is false.
> >>
> >> The first approach is unable to handle non-vanilla kernels. The second
> >> approach needs to add stub functions which are unclean. What I would
> >> like to do is,
> >>
> >> * add a m4 rule.
> >> * modify unshare Makefile like this,
> >>
> >>   ifdef HAVE_UNSHARE  <-- if the syscall is present, compile the tests.
> >
> > the configure test should always be there if the code needs new
> > headers (like sys/signalfd.h).  if it's only a new syscall, then the
> > configure test should only be for the new symbol (assuming one is
> > provided).  in all cases, the tests should be compiled.  your proposal
> > doesnt handle the case where the C library is in place, but the kernel
> > is older so returns ENOSYS.
>
> OK, how about those two approaches?
>
> * add a stub C program to check if kernel returns ENOSYS. If so, returns
>   1. Otherwise, 0.
> * add a m4 rule to check the headers or/and functions.
> * if ./configure found all headers and functions, and execute the stub
>   program to return 0, it will set HAVE_UNSHARE.
> * modify unshare Makefile to complie it only if HAVE_UNSHARE set.
>
> or
>
> * add a m4 rule to check the headers or/and functions.
> * if ./configure found all headers and functions, , it will set
>   HAVE_UNSHARE.
> * modify the test code to deal with three situations.
>   * HAVE_UNSHARE = 0
>   * HAVE_UNSHARE = 1 + ENOSYS
>   * HAVE_UNSHARE = 1 + no ENOSYS

in addition to the reasoning Garrett provided, this makes for a fairly 
complicated setup for no real gain.  the solution i mentioned (and is being 
used) covers all known build/runtime issues.

if you need a new C library function, check for it.  if you need a new C 
library header, check for it.  if a test may be run on an older system and 
thus the call fails with ENOSYS, the test must check for it and error out with 
TCONF (and a helper message).  whether someone chooses to also add a runtime 
kernel version check (as can be seen in a few tests) is up to whoever feels 
like doing the work.
-mike

[-- Attachment #1.2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[-- Attachment #2: Type: text/plain, Size: 355 bytes --]

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july

[-- Attachment #3: Type: text/plain, Size: 155 bytes --]

_______________________________________________
Ltp-list mailing list
Ltp-list@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ltp-list