Messages in this thread | | | Date | Thu, 04 Feb 2010 12:03:01 -0800 | From | Randy Dunlap <> | Subject | Re: mmotm 2010-02-03-20-09 uploaded (mmap.c) |
| |
On 02/04/10 10:58, Andrew Morton wrote: > On Thu, 04 Feb 2010 10:14:55 -0800 > Randy Dunlap<rdunlap@xenotime.net> wrote: > >> On 02/03/10 20:09, akpm@linux-foundation.org wrote: >>> The mm-of-the-moment snapshot 2010-02-03-20-09 has been uploaded to >>> >>> http://userweb.kernel.org/~akpm/mmotm/ >>> >>> and will soon be available at >>> >>> git://zen-kernel.org/kernel/mmotm.git >>> >>> It contains the following patches against 2.6.33-rc6: >> >> >> In some config cases, this patch seems to have a problem: >> sysctl-clean-up-vm-related-variable-declarations.patch >> >> In mm/mmap.c: >> +#ifdef HAVE_ARCH_PICK_MMAP_LAYOUT >> +/* Used by each architecture's private code and sysctl. */ >> +int sysctl_legacy_va_layout; >> +#endif >> >> mm/mmap.c:92: error: expected identifier or '(' before numeric constant >> >> I guess that sometimes sysctl_legacy_va_layout is a #define: >> from include/linux/mm.h: >> >> #ifdef CONFIG_SYSCTL >> extern int sysctl_legacy_va_layout; >> #else >> #define sysctl_legacy_va_layout 0 >> #endif >> >> and indeed the failing builds do not have CONFIG_SYSCTL enabled. > > Thanks. This, I guess: > > --- a/mm/mmap.c~sysctl-clean-up-vm-related-variable-declarations-fix > +++ a/mm/mmap.c > @@ -87,7 +87,7 @@ int sysctl_overcommit_ratio = 50; /* def > int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT; > struct percpu_counter vm_committed_as; > > -#ifdef HAVE_ARCH_PICK_MMAP_LAYOUT > +#if defined(HAVE_ARCH_PICK_MMAP_LAYOUT)&& defined(CONFIG_SYSCTL) > /* Used by each architecture's private code and sysctl. */ > int sysctl_legacy_va_layout; > #endif > _ >
Yep, that works. Thanks.
Acked-by: Randy Dunlap <rdunlap@xenotime.net>
-- ~Randy
| |