diff options
| author | Chris Metcalf <cmetcalf@tilera.com> | 2012-03-29 19:58:43 +0200 |
|---|---|---|
| committer | Chris Metcalf <cmetcalf@tilera.com> | 2012-05-25 18:48:24 +0200 |
| commit | d5d14ed6f2db7287a5088e1350cf422bf72140b3 (patch) | |
| tree | 19f0bc20bb6f1995a1e4f75dc58e388c047f7d23 /arch/tile/mm/init.c | |
| parent | arch/tile: optimize get_user/put_user and friends (diff) | |
| download | linux-d5d14ed6f2db7287a5088e1350cf422bf72140b3.tar.xz linux-d5d14ed6f2db7287a5088e1350cf422bf72140b3.zip | |
arch/tile: Allow tilegx to build with either 16K or 64K page size
This change introduces new flags for the hv_install_context()
API that passes a page table pointer to the hypervisor. Clients
can explicitly request 4K, 16K, or 64K small pages when they
install a new context. In practice, the page size is fixed at
kernel compile time and the same size is always requested every
time a new page table is installed.
The <hv/hypervisor.h> header changes so that it provides more abstract
macros for managing "page" things like PFNs and page tables. For
example there is now a HV_DEFAULT_PAGE_SIZE_SMALL instead of the old
HV_PAGE_SIZE_SMALL. The various PFN routines have been eliminated and
only PA- or PTFN-based ones remain (since PTFNs are always expressed
in fixed 2KB "page" size). The page-table management macros are
renamed with a leading underscore and take page-size arguments with
the presumption that clients will use those macros in some single
place to provide the "real" macros they will use themselves.
I happened to notice the old hv_set_caching() API was totally broken
(it assumed 4KB pages) so I changed it so it would nominally work
correctly with other page sizes.
Tag modules with the page size so you can't load a module built with
a conflicting page size. (And add a test for SMP while we're at it.)
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
Diffstat (limited to 'arch/tile/mm/init.c')
| -rw-r--r-- | arch/tile/mm/init.c | 11 |
1 files changed, 3 insertions, 8 deletions
diff --git a/arch/tile/mm/init.c b/arch/tile/mm/init.c index 1e4633520b35..c04fbfd93fc5 100644 --- a/arch/tile/mm/init.c +++ b/arch/tile/mm/init.c @@ -82,7 +82,7 @@ static int num_l2_ptes[MAX_NUMNODES]; static void init_prealloc_ptes(int node, int pages) { - BUG_ON(pages & (HV_L2_ENTRIES-1)); + BUG_ON(pages & (PTRS_PER_PTE - 1)); if (pages) { num_l2_ptes[node] = pages; l2_ptes[node] = __alloc_bootmem(pages * sizeof(pte_t), @@ -131,14 +131,9 @@ static void __init assign_pte(pmd_t *pmd, pte_t *page_table) #ifdef __tilegx__ -#if HV_L1_SIZE != HV_L2_SIZE -# error Rework assumption that L1 and L2 page tables are same size. -#endif - -/* Since pmd_t arrays and pte_t arrays are the same size, just use casts. */ static inline pmd_t *alloc_pmd(void) { - return (pmd_t *)alloc_pte(); + return __alloc_bootmem(L1_KERNEL_PGTABLE_SIZE, HV_PAGE_TABLE_ALIGN, 0); } static inline void assign_pmd(pud_t *pud, pmd_t *pmd) @@ -811,7 +806,7 @@ void __init paging_init(void) * changing init_mm once we get up and running, and there's no * need for e.g. vmalloc_sync_all(). */ - BUILD_BUG_ON(pgd_index(VMALLOC_START) != pgd_index(VMALLOC_END)); + BUILD_BUG_ON(pgd_index(VMALLOC_START) != pgd_index(VMALLOC_END - 1)); pud = pud_offset(pgd_base + pgd_index(VMALLOC_START), VMALLOC_START); assign_pmd(pud, alloc_pmd()); #endif |