[PATCH] ppc32: fix ppc440 pagetable attributes
This patch fixes a bug in the PPC440 pagetable attributes that breaks swap support. It also adds some notes on the PPC440 attribute fields. Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com> for CELF Signed-off-by: Matt Porter <mporter@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This commit is contained in:
Родитель
4aad724d3e
Коммит
534afb90a9
|
@ -202,18 +202,64 @@ extern unsigned long ioremap_bot, ioremap_base;
|
|||
*
|
||||
* Note that these bits preclude future use of a page size
|
||||
* less than 4KB.
|
||||
*
|
||||
*
|
||||
* PPC 440 core has following TLB attribute fields;
|
||||
*
|
||||
* TLB1:
|
||||
* 0 1 2 3 4 ... 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
|
||||
* RPN................................. - - - - - - ERPN.......
|
||||
*
|
||||
* TLB2:
|
||||
* 0 1 2 3 4 ... 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
|
||||
* - - - - - - U0 U1 U2 U3 W I M G E - UX UW UR SX SW SR
|
||||
*
|
||||
* There are some constrains and options, to decide mapping software bits
|
||||
* into TLB entry.
|
||||
*
|
||||
* - PRESENT *must* be in the bottom three bits because swap cache
|
||||
* entries use the top 29 bits for TLB2.
|
||||
*
|
||||
* - FILE *must* be in the bottom three bits because swap cache
|
||||
* entries use the top 29 bits for TLB2.
|
||||
*
|
||||
* - CACHE COHERENT bit (M) has no effect on PPC440 core, because it
|
||||
* doesn't support SMP. So we can use this as software bit, like
|
||||
* DIRTY.
|
||||
*
|
||||
* PPC Book-E Linux implementation uses PPC HW PTE bit field definition,
|
||||
* even it doesn't have HW PTE. 0-11th LSB of PTE stand for memory
|
||||
* protection-related function. (See PTE structure in include/asm-ppc/mmu.h)
|
||||
* Definition of _PAGE_XXX in "include/asm-ppc/pagetable.h" stands for
|
||||
* above bits. Note that those bits values are CPU dependent, not
|
||||
* architecture.
|
||||
*
|
||||
* Kernel PTE entry holds arch-dependent swp_entry structure under certain
|
||||
* situation. In other words, in such situation, some portion of PTE bits
|
||||
* are used as swp_entry. In PPC implementation, 3-24th LSB are shared with
|
||||
* swp_entry, however 0-2nd three LSB still hold protection values.
|
||||
* That means three protection bits are reserved for both PTE and SWAP
|
||||
* entry at the most three LSBs.
|
||||
*
|
||||
* There are three protection bits available for SWAP entry;
|
||||
* _PAGE_PRESENT
|
||||
* _PAGE_FILE
|
||||
* _PAGE_HASHPTE (if HW has)
|
||||
*
|
||||
* So those three bits have to be inside of 0-2nd LSB of PTE.
|
||||
*
|
||||
*/
|
||||
|
||||
#define _PAGE_PRESENT 0x00000001 /* S: PTE valid */
|
||||
#define _PAGE_RW 0x00000002 /* S: Write permission */
|
||||
#define _PAGE_DIRTY 0x00000004 /* S: Page dirty */
|
||||
#define _PAGE_FILE 0x00000004 /* S: nonlinear file mapping */
|
||||
#define _PAGE_ACCESSED 0x00000008 /* S: Page referenced */
|
||||
#define _PAGE_HWWRITE 0x00000010 /* H: Dirty & RW */
|
||||
#define _PAGE_HWEXEC 0x00000020 /* H: Execute permission */
|
||||
#define _PAGE_USER 0x00000040 /* S: User page */
|
||||
#define _PAGE_ENDIAN 0x00000080 /* H: E bit */
|
||||
#define _PAGE_GUARDED 0x00000100 /* H: G bit */
|
||||
#define _PAGE_COHERENT 0x00000200 /* H: M bit */
|
||||
#define _PAGE_FILE 0x00000400 /* S: nonlinear file mapping */
|
||||
#define _PAGE_DIRTY 0x00000200 /* S: Page dirty */
|
||||
#define _PAGE_NO_CACHE 0x00000400 /* H: I bit */
|
||||
#define _PAGE_WRITETHRU 0x00000800 /* H: W bit */
|
||||
|
||||
|
|
Загрузка…
Ссылка в новой задаче