|
From: Garrett C. <yan...@gm...> - 2010-02-22 18:17:53
|
On Mon, Feb 22, 2010 at 8:49 AM, Rishikesh K Rajak <ris...@li...> wrote: > Hi Iranna, > > > Thanks for reporting this build failure and helping, as it will > really help to make ltp more stable. > > Here is the quick patch i am attaching which solves the problem. > > Signed-off By: Rishikesh K Rajak<ris...@li...> > Tested By: Rishikesh K Rajak<ris...@li...> > --- > diff --git a/testcases/kernel/syscalls/mbind/mbind01.c > b/testcases/kernel/syscalls/mbind/mbind01.c > index 5d7eb69..847186a 100644 > --- a/testcases/kernel/syscalls/mbind/mbind01.c > +++ b/testcases/kernel/syscalls/mbind/mbind01.c > @@ -74,6 +74,7 @@ int TST_TOTAL = 2; /* total number of tests > in this file. */ > #include <unistd.h> > #include "include_j_h.h" > #include "numa_helpers.h" > +#include <linux/mempolicy.h> > > #if ! defined(LIBNUMA_API_VERSION) || LIBNUMA_API_VERSION < 2 No. This patch shouldn't be committed because of namespace collisions between numa and mempolicy (thank the developers for just copy-pasting constants, macros, etc instead of properly requiring numa on the target system in order for mempolicy to work...). Thanks, -Garrett |