diff options
| author | Chris Metcalf <cmetcalf@tilera.com> | 2011-02-28 16:37:34 -0500 |
|---|---|---|
| committer | Chris Metcalf <cmetcalf@tilera.com> | 2011-03-10 13:17:53 -0500 |
| commit | 76c567fbba50c3da2f4d40e2e551bab26cfd4381 (patch) | |
| tree | 6e3c92a266d0ec255e1930adf5ba5268cd71dee9 /arch/tile/lib | |
| parent | 09c17eab075ceeafb53935d858c575b6776394d1 (diff) | |
arch/tile: support 4KB page size as well as 64KB
The Tilera architecture traditionally supports 64KB page sizes
to improve TLB utilization and improve performance when the
hardware is being used primarily to run a single application.
For more generic server scenarios, it can be beneficial to run
with 4KB page sizes, so this commit allows that to be specified
(by modifying the arch/tile/include/hv/pagesize.h header).
As part of this change, we also re-worked the PTE management
slightly so that PTE writes all go through a __set_pte() function
where we can do some additional validation. The set_pte_order()
function was eliminated since the "order" argument wasn't being used.
One bug uncovered was in the PCI DMA code, which wasn't properly
flushing the specified range. This was benign with 64KB pages,
but with 4KB pages we were getting some larger flushes wrong.
The per-cpu memory reservation code also needed updating to
conform with the newer percpu stuff; before it always chose 64KB,
and that was always correct, but with 4KB granularity we now have
to pay closer attention and reserve the amount of memory that will
be requested when the percpu code starts allocating.
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Diffstat (limited to 'arch/tile/lib')
| -rw-r--r-- | arch/tile/lib/memcpy_tile64.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/arch/tile/lib/memcpy_tile64.c b/arch/tile/lib/memcpy_tile64.c index f7d4a6ad61e..b2fe15e0107 100644 --- a/arch/tile/lib/memcpy_tile64.c +++ b/arch/tile/lib/memcpy_tile64.c @@ -96,7 +96,7 @@ static void memcpy_multicache(void *dest, const void *source, newsrc = __fix_to_virt(idx) + ((unsigned long)source & (PAGE_SIZE-1)); pmdp = pmd_offset(pud_offset(pgd_offset_k(newsrc), newsrc), newsrc); ptep = pte_offset_kernel(pmdp, newsrc); - *ptep = src_pte; /* set_pte() would be confused by this */ + __set_pte(ptep, src_pte); /* set_pte() would be confused by this */ local_flush_tlb_page(NULL, newsrc, PAGE_SIZE); /* Actually move the data. */ @@ -109,7 +109,7 @@ static void memcpy_multicache(void *dest, const void *source, */ src_pte = hv_pte_set_mode(src_pte, HV_PTE_MODE_CACHE_NO_L3); src_pte = hv_pte_set_writable(src_pte); /* need write access for inv */ - *ptep = src_pte; /* set_pte() would be confused by this */ + __set_pte(ptep, src_pte); /* set_pte() would be confused by this */ local_flush_tlb_page(NULL, newsrc, PAGE_SIZE); /* |