diff options
author | Martin Schwidefsky | 2017-04-20 14:43:51 +0200 |
---|---|---|
committer | Martin Schwidefsky | 2017-04-25 07:47:32 +0200 |
commit | ee71d16d22bb268c1f6a64ef6d3654ace5f1e8c7 (patch) | |
tree | 8e8ecc1da0ba8947a9b024a435262460d67caeff /arch/s390/mm/gup.c | |
parent | 5037c22c53603acd4e8c465d750fdd604c600bd3 (diff) |
s390/mm: make TASK_SIZE independent from the number of page table levels
The TASK_SIZE for a process should be maximum possible size of the address
space, 2GB for a 31-bit process and 8PB for a 64-bit process. The number
of page table levels required for a given memory layout is a consequence
of the mapped memory areas and their location.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Diffstat (limited to 'arch/s390/mm/gup.c')
-rw-r--r-- | arch/s390/mm/gup.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/arch/s390/mm/gup.c b/arch/s390/mm/gup.c index 18d4107e10ee..b7b779c40a5b 100644 --- a/arch/s390/mm/gup.c +++ b/arch/s390/mm/gup.c @@ -211,7 +211,7 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write, addr = start; len = (unsigned long) nr_pages << PAGE_SHIFT; end = start + len; - if ((end <= start) || (end > TASK_SIZE)) + if ((end <= start) || (end > mm->context.asce_limit)) return 0; /* * local_irq_save() doesn't prevent pagetable teardown, but does |