diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2019-03-07 19:25:37 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2019-03-07 19:25:37 -0800 |
| commit | b5dd0c658c31b469ccff1b637e5124851e7a4a1c (patch) | |
| tree | dc0b31a5aa62bb4e1fa653a4f176c2faae51f9e0 /arch/powerpc/kernel/setup_64.c | |
| parent | 610cd4eadec4f97acd25d3108b0e50d1362b3319 (diff) | |
| parent | fe0436e10c8845aed24cad3a1c719efcd6e583eb (diff) | |
Merge branch 'akpm' (patches from Andrew)
Merge more updates from Andrew Morton:
- some of the rest of MM
- various misc things
- dynamic-debug updates
- checkpatch
- some epoll speedups
- autofs
- rapidio
- lib/, lib/lzo/ updates
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (83 commits)
samples/mic/mpssd/mpssd.h: remove duplicate header
kernel/fork.c: remove duplicated include
include/linux/relay.h: fix percpu annotation in struct rchan
arch/nios2/mm/fault.c: remove duplicate include
unicore32: stop printing the virtual memory layout
MAINTAINERS: fix GTA02 entry and mark as orphan
mm: create the new vm_fault_t type
arm, s390, unicore32: remove oneliner wrappers for memblock_alloc()
arch: simplify several early memory allocations
openrisc: simplify pte_alloc_one_kernel()
sh: prefer memblock APIs returning virtual address
microblaze: prefer memblock API returning virtual address
powerpc: prefer memblock APIs returning virtual address
lib/lzo: separate lzo-rle from lzo
lib/lzo: implement run-length encoding
lib/lzo: fast 8-byte copy on arm64
lib/lzo: 64-bit CTZ on arm64
lib/lzo: tidy-up ifdefs
ipc/sem.c: replace kvmalloc/memset with kvzalloc and use struct_size
ipc: annotate implicit fall through
...
Diffstat (limited to 'arch/powerpc/kernel/setup_64.c')
| -rw-r--r-- | arch/powerpc/kernel/setup_64.c | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index daa361fc6a24..ff0aac42bb33 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -902,8 +902,9 @@ static void __ref init_fallback_flush(void) * hardware prefetch runoff. We don't have a recipe for load patterns to * reliably avoid the prefetcher. */ - l1d_flush_fallback_area = __va(memblock_alloc_base(l1d_size * 2, l1d_size, limit)); - memset(l1d_flush_fallback_area, 0, l1d_size * 2); + l1d_flush_fallback_area = memblock_alloc_try_nid(l1d_size * 2, + l1d_size, MEMBLOCK_LOW_LIMIT, + limit, NUMA_NO_NODE); for_each_possible_cpu(cpu) { struct paca_struct *paca = paca_ptrs[cpu]; |