diff options
Diffstat (limited to 'arch')
52 files changed, 785 insertions, 548 deletions
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 5c10af66fb5..ad6b1c084fe 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -313,7 +313,7 @@ config ARCH_ENABLE_MEMORY_HOTREMOVE config KEXEC bool "kexec system call (EXPERIMENTAL)" - depends on (PPC_PRPMC2800 || PPC_MULTIPLATFORM) && EXPERIMENTAL + depends on BOOK3S && EXPERIMENTAL help kexec is a system call that implements the ability to shutdown your current kernel, and to start another kernel. It is like a reboot diff --git a/arch/powerpc/Kconfig.debug b/arch/powerpc/Kconfig.debug index 08f7cc0a195..22091bbfdc9 100644 --- a/arch/powerpc/Kconfig.debug +++ b/arch/powerpc/Kconfig.debug @@ -129,7 +129,7 @@ config BDI_SWITCH config BOOTX_TEXT bool "Support for early boot text console (BootX or OpenFirmware only)" - depends on PPC_OF && PPC_MULTIPLATFORM + depends on PPC_OF && PPC_BOOK3S help Say Y here to see progress messages from the boot firmware in text mode. Requires either BootX or Open Firmware. diff --git a/arch/powerpc/include/asm/hw_irq.h b/arch/powerpc/include/asm/hw_irq.h index f75a5fc64d2..b7e034b0a6d 100644 --- a/arch/powerpc/include/asm/hw_irq.h +++ b/arch/powerpc/include/asm/hw_irq.h @@ -129,7 +129,7 @@ static inline int irqs_disabled_flags(unsigned long flags) * interrupt-retrigger: should we handle this via lost interrupts and IPIs * or should we not care like we do now ? --BenH. */ -struct hw_interrupt_type; +struct irq_chip; #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_HW_IRQ_H */ diff --git a/arch/powerpc/include/asm/pgtable-ppc32.h b/arch/powerpc/include/asm/pgtable-ppc32.h index 98bd7c5fcd0..67ceffc01b4 100644 --- a/arch/powerpc/include/asm/pgtable-ppc32.h +++ b/arch/powerpc/include/asm/pgtable-ppc32.h @@ -19,55 +19,6 @@ extern int icache_44x_need_flush; #endif /* __ASSEMBLY__ */ /* - * The PowerPC MMU uses a hash table containing PTEs, together with - * a set of 16 segment registers (on 32-bit implementations), to define - * the virtual to physical address mapping. - * - * We use the hash table as an extended TLB, i.e. a cache of currently - * active mappings. We maintain a two-level page table tree, much - * like that used by the i386, for the sake of the Linux memory - * management code. Low-level assembler code in hashtable.S - * (procedure hash_page) is responsible for extracting ptes from the - * tree and putting them into the hash table when necessary, and - * updating the accessed and modified bits in the page table tree. - */ - -/* - * The PowerPC MPC8xx uses a TLB with hardware assisted, software tablewalk. - * We also use the two level tables, but we can put the real bits in them - * needed for the TLB and tablewalk. These definitions require Mx_CTR.PPM = 0, - * Mx_CTR.PPCS = 0, and MD_CTR.TWAM = 1. The level 2 descriptor has - * additional page protection (when Mx_CTR.PPCS = 1) that allows TLB hit - * based upon user/super access. The TLB does not have accessed nor write - * protect. We assume that if the TLB get loaded with an entry it is - * accessed, and overload the changed bit for write protect. We use - * two bits in the software pte that are supposed to be set to zero in - * the TLB entry (24 and 25) for these indicators. Although the level 1 - * descriptor contains the guarded and writethrough/copyback bits, we can - * set these at the page level since they get copied from the Mx_TWC - * register when the TLB entry is loaded. We will use bit 27 for guard, since - * that is where it exists in the MD_TWC, and bit 26 for writethrough. - * These will get masked from the level 2 descriptor at TLB load time, and - * copied to the MD_TWC before it gets loaded. - * Large page sizes added. We currently support two sizes, 4K and 8M. - * This also allows a TLB hander optimization because we can directly - * load the PMD into MD_TWC. The 8M pages are only used for kernel - * mapping of well known areas. The PMD (PGD) entries contain control - * flags in addition to the address, so care must be taken that the - * software no longer assumes these are only pointers. - */ - -/* - * At present, all PowerPC 400-class processors share a similar TLB - * architecture. The instruction and data sides share a unified, - * 64-entry, fully-associative TLB which is maintained totally under - * software control. In addition, the instruction side has a - * hardware-managed, 4-entry, fully-associative TLB which serves as a - * first level to the shared TLB. These two TLBs are known as the UTLB - * and ITLB, respectively (see "mmu.h" for definitions). - */ - -/* * The normal case is that PTEs are 32-bits and we have a 1-page * 1024-entry pgdir pointing to 1-page 1024-entry PTE pages. -- paulus * @@ -135,261 +86,25 @@ extern int icache_44x_need_flush; */ #if defined(CONFIG_40x) - -/* There are several potential gotchas here. The 40x hardware TLBLO - field looks like this: - - 0 1 2 3 4 ... 18 19 20 21 22 23 24 25 26 27 28 29 30 31 - RPN..................... 0 0 EX WR ZSEL....... W I M G - - Where possible we make the Linux PTE bits match up with this - - - bits 20 and 21 must be cleared, because we use 4k pages (40x can - support down to 1k pages), this is done in the TLBMiss exception - handler. - - We use only zones 0 (for kernel pages) and 1 (for user pages) - of the 16 available. Bit 24-26 of the TLB are cleared in the TLB - miss handler. Bit 27 is PAGE_USER, thus selecting the correct - zone. - - PRESENT *must* be in the bottom two bits because swap cache - entries use the top 30 bits. Because 40x doesn't support SMP - anyway, M is irrelevant so we borrow it for PAGE_PRESENT. Bit 30 - is cleared in the TLB miss handler before the TLB entry is loaded. - - All other bits of the PTE are loaded into TLBLO without - modification, leaving us only the bits 20, 21, 24, 25, 26, 30 for - software PTE bits. We actually use use bits 21, 24, 25, and - 30 respectively for the software bits: ACCESSED, DIRTY, RW, and - PRESENT. -*/ - -/* Definitions for 40x embedded chips. */ -#define _PAGE_GUARDED 0x001 /* G: page is guarded from prefetch */ -#define _PAGE_FILE 0x001 /* when !present: nonlinear file mapping */ -#define _PAGE_PRESENT 0x002 /* software: PTE contains a translation */ -#define _PAGE_NO_CACHE 0x004 /* I: caching is inhibited */ -#define _PAGE_WRITETHRU 0x008 /* W: caching is write-through */ -#define _PAGE_USER 0x010 /* matches one of the zone permission bits */ -#define _PAGE_RW 0x040 /* software: Writes permitted */ -#define _PAGE_DIRTY 0x080 /* software: dirty page */ -#define _PAGE_HWWRITE 0x100 /* hardware: Dirty & RW, set in exception */ -#define _PAGE_HWEXEC 0x200 /* hardware: EX permission */ -#define _PAGE_ACCESSED 0x400 /* software: R: page referenced */ - -#define _PMD_PRESENT 0x400 /* PMD points to page of PTEs */ -#define _PMD_BAD 0x802 -#define _PMD_SIZE 0x0e0 /* size field, != 0 for large-page PMD entry */ -#define _PMD_SIZE_4M 0x0c0 -#define _PMD_SIZE_16M 0x0e0 -#define PMD_PAGE_SIZE(pmdval) (1024 << (((pmdval) & _PMD_SIZE) >> 4)) - -/* Until my rework is finished, 40x still needs atomic PTE updates */ -#define PTE_ATOMIC_UPDATES 1 - +#include <asm/pte-40x.h> #elif defined(CONFIG_44x) -/* - * Definitions for PPC440 - * - * Because of the 3 word TLB entries to support 36-bit addressing, - * the attribute are difficult to map in such a fashion that they - * are easily loaded during exception processing. I decided to - * organize the entry so the ERPN is the only portion in the - * upper word of the PTE and the attribute bits below are packed - * in as sensibly as they can be in the area below a 4KB page size - * oriented RPN. This at least makes it easy to load the RPN and - * ERPN fields in the TLB. -Matt - * - * 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 - * - * Newer 440 cores (440x6 as used on AMCC 460EX/460GT) have additional - * TLB2 storage attibute fields. Those are: - * - * TLB2: - * 0...10 11 12 13 14 15 16...31 - * no change WL1 IL1I IL1D IL2I IL2D no change - * - * 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 original PPC440 cores, - * because it doesn't support SMP. However, some later 460 variants - * have -some- form of SMP support and so I keep the bit there for - * future use - * - * With the PPC 44x Linux implementation, the 0-11th LSBs of the PTE are used - * for memory protection related functions (see PTE structure in - * include/asm-ppc/mmu.h). The _PAGE_XXX definitions in this file map to the - * above bits. Note that the bit values are CPU specific, not architecture - * specific. - * - * The kernel PTE entry holds an arch-dependent swp_entry structure under - * certain situations. In other words, in such situations some portion of - * the PTE bits are used as a swp_entry. In the PPC implementation, the - * 3-24th LSB are shared with swp_entry, however the 0-2nd three LSB still - * hold protection values. That means the three protection bits are - * reserved for both PTE and SWAP entry at the most significant 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_FILE 0x00000004 /* S: nonlinear file mapping */ -#define _PAGE_HWEXEC 0x00000004 /* H: Execute permission */ -#define _PAGE_ACCESSED 0x00000008 /* S: Page referenced */ -#define _PAGE_DIRTY 0x00000010 /* S: Page dirty */ -#define _PAGE_SPECIAL 0x00000020 /* S: Special page */ -#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_NO_CACHE 0x00000400 /* H: I bit */ -#define _PAGE_WRITETHRU 0x00000800 /* H: W bit */ - -/* TODO: Add large page lowmem mapping support */ -#define _PMD_PRESENT 0 -#define _PMD_PRESENT_MASK (PAGE_MASK) -#define _PMD_BAD (~PAGE_MASK) - -/* ERPN in a PTE never gets cleared, ignore it */ -#define _PTE_NONE_MASK 0xffffffff00000000ULL - -#define __HAVE_ARCH_PTE_SPECIAL - +#include <asm/pte-44x.h> #elif defined(CONFIG_FSL_BOOKE) -/* - MMU Assist Register 3: - - 32 33 34 35 36 ... 50 51 52 53 54 55 56 57 58 59 60 61 62 63 - RPN...................... 0 0 U0 U1 U2 U3 UX SX UW SW UR SR - - - PRESENT *must* be in the bottom three bits because swap cache - entries use the top 29 bits. - - - FILE *must* be in the bottom three bits because swap cache - entries use the top 29 bits. -*/ - -/* Definitions for FSL Book-E Cores */ -#define _PAGE_PRESENT 0x00001 /* S: PTE contains a translation */ -#define _PAGE_USER 0x00002 /* S: User page (maps to UR) */ -#define _PAGE_FILE 0x00002 /* S: when !present: nonlinear file mapping */ -#define _PAGE_RW 0x00004 /* S: Write permission (SW) */ -#define _PAGE_DIRTY 0x00008 /* S: Page dirty */ -#define _PAGE_HWEXEC 0x00010 /* H: SX permission */ -#define _PAGE_ACCESSED 0x00020 /* S: Page referenced */ - -#define _PAGE_ENDIAN 0x00040 /* H: E bit */ -#define _PAGE_GUARDED 0x00080 /* H: G bit */ -#define _PAGE_COHERENT 0x00100 /* H: M bit */ -#define _PAGE_NO_CACHE 0x00200 /* H: I bit */ -#define _PAGE_WRITETHRU 0x00400 /* H: W bit */ -#define _PAGE_SPECIAL 0x00800 /* S: Special page */ - -#ifdef CONFIG_PTE_64BIT -/* ERPN in a PTE never gets cleared, ignore it */ -#define _PTE_NONE_MASK 0xffffffffffff0000ULL -#endif - -#define _PMD_PRESENT 0 -#define _PMD_PRESENT_MASK (PAGE_MASK) -#define _PMD_BAD (~PAGE_MASK) - -#define __HAVE_ARCH_PTE_SPECIAL - +#include <asm/pte-fsl-booke.h> #elif defined(CONFIG_8xx) -/* Definitions for 8xx embedded chips. */ -#define _PAGE_PRESENT 0x0001 /* Page is valid */ -#define _PAGE_FILE 0x0002 /* when !present: nonlinear file mapping */ -#define _PAGE_NO_CACHE 0x0002 /* I: cache inhibit */ -#define _PAGE_SHARED 0x0004 /* No ASID (context) compare */ - -/* These five software bits must be masked out when the entry is loaded - * into the TLB. - */ -#define _PAGE_EXEC 0x0008 /* software: i-cache coherency required */ -#define _PAGE_GUARDED 0x0010 /* software: guarded access */ -#define _PAGE_DIRTY 0x0020 /* software: page changed */ -#define _PAGE_RW 0x0040 /* software: user write access allowed */ -#define _PAGE_ACCESSED 0x0080 /* software: page referenced */ - -/* Setting any bits in the nibble with the follow two controls will - * require a TLB exception handler change. It is assumed unused bits - * are always zero. - */ -#define _PAGE_HWWRITE 0x0100 /* h/w write enable: never set in Linux PTE */ -#define _PAGE_USER 0x0800 /* One of the PP bits, the other is USER&~RW */ - -#define _PMD_PRESENT 0x0001 -#define _PMD_BAD 0x0ff0 -#define _PMD_PAGE_MASK 0x000c -#define _PMD_PAGE_8M 0x000c - -#define _PTE_NONE_MASK _PAGE_ACCESSED - -/* Until my rework is finished, 8xx still needs atomic PTE updates */ -#define PTE_ATOMIC_UPDATES 1 - +#include <asm/pte-8xx.h> #else /* CONFIG_6xx */ -/* Definitions for 60x, 740/750, etc. */ -#define _PAGE_PRESENT 0x001 /* software: pte contains a translation */ -#define _PAGE_HASHPTE 0x002 /* hash_page has made an HPTE for this pte */ -#define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */ -#define _PAGE_USER 0x004 /* usermode access allowed */ -#define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */ -#define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */ -#define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */ -#define _PAGE_WRITETHRU 0x040 /* W: cache write-through */ -#define _PAGE_DIRTY 0x080 /* C: page changed */ -#define _PAGE_ACCESSED 0x100 /* R: page referenced */ -#define _PAGE_EXEC 0x200 /* software: i-cache coherency required */ -#define _PAGE_RW 0x400 /* software: user write access allowed */ -#define _PAGE_SPECIAL 0x800 /* software: Special page */ - -#ifdef CONFIG_PTE_64BIT -/* We never clear the high word of the pte */ -#define _PTE_NONE_MASK (0xffffffff00000000ULL | _PAGE_HASHPTE) -#else -#define _PTE_NONE_MASK _PAGE_HASHPTE +#include <asm/pte-hash32.h> #endif -#define _PMD_PRESENT 0 -#define _PMD_PRESENT_MASK (PAGE_MASK) -#define _PMD_BAD (~PAGE_MASK) - -/* Hash table based platforms need atomic updates of the linux PTE */ -#define PTE_ATOMIC_UPDATES 1 - +/* If _PAGE_SPECIAL is defined, then we advertise our support for it */ +#ifdef _PAGE_SPECIAL #define __HAVE_ARCH_PTE_SPECIAL - #endif /* - * Some bits are only used on some cpu families... + * Some bits are only used on some cpu families... Make sure that all + * the undefined gets defined as 0 */ #ifndef _PAGE_HASHPTE #define _PAGE_HASHPTE 0 @@ -431,9 +146,29 @@ extern int icache_44x_need_flush; #define _PAGE_HPTEFLAGS _PAGE_HASHPTE -#define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_SPECIAL) +/* Location of the PFN in the PTE. Most platforms use the same as _PAGE_SHIFT + * here (ie, naturally aligned). Platform who don't just pre-define the + * value so we don't override it here + */ +#ifndef PTE_RPN_SHIFT +#define PTE_RPN_SHIFT (PAGE_SHIFT) +#endif + +#ifdef CONFIG_PTE_64BIT +#define PTE_RPN_MAX (1ULL << (64 - PTE_RPN_SHIFT)) +#define PTE_RPN_MASK (~((1ULL<<PTE_RPN_SHIFT)-1)) +#else +#define PTE_RPN_MAX (1UL << (32 - PTE_RPN_SHIFT)) +#define PTE_RPN_MASK (~((1UL<<PTE_RPN_SHIFT)-1)) +#endif + +/* _PAGE_CHG_MASK masks of bits that are to be preserved accross + * pgprot changes + */ +#define _PAGE_CHG_MASK (PTE_RPN_MASK | _PAGE_HPTEFLAGS | _PAGE_DIRTY | \ + _PAGE_ACCESSED | _PAGE_SPECIAL) +/* Mask of bits returned by pte_pgprot() */ #define PAGE_PROT_BITS (_PAGE_GUARDED | _PAGE_COHERENT | _PAGE_NO_CACHE | \ _PAGE_WRITETHRU | _PAGE_ENDIAN | \ _PAGE_USER | _PAGE_ACCESSED | \ @@ -521,18 +256,10 @@ extern unsigned long bad_call_to_PMD_PAGE_SIZE(void); * Conversions between PTE values and page frame numbers. */ -/* in some case we want to additionaly adjust where the pfn is in the pte to - * allow room for more flags */ -#if defined(CONFIG_FSL_BOOKE) && defined(CONFIG_PTE_64BIT) -#define PFN_SHIFT_OFFSET (PAGE_SHIFT + 8) -#else -#define PFN_SHIFT_OFFSET (PAGE_SHIFT) -#endif - -#define pte_pfn(x) (pte_val(x) >> PFN_SHIFT_OFFSET) +#define pte_pfn(x) (pte_val(x) >> PTE_RPN_SHIFT) #define pte_page(x) pfn_to_page(pte_pfn(x)) -#define pfn_pte(pfn, prot) __pte(((pte_basic_t)(pfn) << PFN_SHIFT_OFFSET) |\ +#define pfn_pte(pfn, prot) __pte(((pte_basic_t)(pfn) << PTE_RPN_SHIFT) |\ pgprot_val(prot)) #define mk_pte(page, prot) pfn_pte(page_to_pfn(page), prot) #endif /* __ASSEMBLY__ */ @@ -600,11 +327,19 @@ extern void flush_hash_entry(struct mm_struct *mm, pte_t *ptep, unsigned long address); /* - * Atomic PTE updates. + * PTE updates. This function is called whenever an existing + * valid PTE is updated. This does -not- include set_pte_at() + * which nowadays only sets a new PTE. + * + * Depending on the type of MMU, we may need to use atomic updates + * and the PTE may be either 32 or 64 bit wide. In the later case, + * when using atomic updates, only the low part of the PTE is + * accessed atomically. * - * pte_update clears and sets bit atomically, and returns - * the old pte value. In the 64-bit PTE case we lock around the - * low PTE word since we expect ALL flag bits to be there + * In addition, on 44x, we also maintain a global flag indicating + * that an executable user mapping was modified, which is needed + * to properly flush the virtually tagged instruction cache of + * those implementations. */ #ifndef CONFIG_PTE_64BIT static inline unsigned long pte_update(pte_t *p, diff --git a/arch/powerpc/include/asm/pgtable-4k.h b/arch/powerpc/include/asm/pgtable-ppc64-4k.h index 1dbca4e7de6..6eefdcffa35 100644 --- a/arch/powerpc/include/asm/pgtable-4k.h +++ b/arch/powerpc/include/asm/pgtable-ppc64-4k.h @@ -1,5 +1,5 @@ -#ifndef _ASM_POWERPC_PGTABLE_4K_H -#define _ASM_POWERPC_PGTABLE_4K_H +#ifndef _ASM_POWERPC_PGTABLE_PPC64_4K_H +#define _ASM_POWERPC_PGTABLE_PPC64_4K_H /* * Entries per page directory level. The PTE level must use a 64b record * for each page table entry. The PMD and PGD level use a 32b record for @@ -40,28 +40,6 @@ #define PGDIR_SIZE (1UL << PGDIR_SHIFT) #define PGDIR_MASK (~(PGDIR_SIZE-1)) -/* PTE bits */ -#define _PAGE_HASHPTE 0x0400 /* software: pte has an associated HPTE */ -#define _PAGE_SECONDARY 0x8000 /* software: HPTE is in secondary group */ -#define _PAGE_GROUP_IX 0x7000 /* software: HPTE index within group */ -#define _PAGE_F_SECOND _PAGE_SECONDARY -#define _PAGE_F_GIX _PAGE_GROUP_IX -#define _PAGE_SPECIAL 0x10000 /* software: special page */ -#define __HAVE_ARCH_PTE_SPECIAL - -/* PTE flags to conserve for HPTE identification */ -#define _PAGE_HPTEFLAGS (_PAGE_BUSY | _PAGE_HASHPTE | \ - _PAGE_SECONDARY | _PAGE_GROUP_IX) - -/* There is no 4K PFN hack on 4K pages */ -#define _PAGE_4K_PFN 0 - -/* PAGE_MASK gives the right answer below, but only by accident */ -/* It should be preserving the high 48 bits and then specifically */ -/* preserving _PAGE_SECONDARY | _PAGE_GROUP_IX */ -#define _PAGE_CHG_MASK (PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY | \ - _PAGE_HPTEFLAGS | _PAGE_SPECIAL) - /* Bits to mask out from a PMD to get to the PTE page */ #define PMD_MASKED_BITS 0 /* Bits to mask out from a PUD to get to the PMD page */ @@ -69,30 +47,6 @@ /* Bits to mask out from a PGD to get to the PUD page */ #define PGD_MASKED_BITS 0 -/* shift to put page number into pte */ -#define PTE_RPN_SHIFT (17) - -#ifdef STRICT_MM_TYPECHECKS -#define __real_pte(e,p) ((real_pte_t){(e)}) -#define __rpte_to_pte(r) ((r).pte) -#else -#define __real_pte(e,p) (e) -#define __rpte_to_pte(r) (__pte(r)) -#endif -#define __rpte_to_hidx(r,index) (pte_val(__rpte_to_pte(r)) >> 12) - -#define pte_iterate_hashed_subpages(rpte, psize, va, index, shift) \ - do { \ - index = 0; \ - shift = mmu_psize_defs[psize].shift; \ - -#define pte_iterate_hashed_end() } while(0) - -#ifdef CONFIG_PPC_HAS_HASH_64K -#define pte_pagesize_index(mm, addr, pte) get_slice_psize(mm, addr) -#else -#define pte_pagesize_index(mm, addr, pte) MMU_PAGE_4K -#endif /* * 4-level page tables related bits @@ -112,6 +66,9 @@ #define pud_ERROR(e) \ printk("%s:%d: bad pud %08lx.\n", __FILE__, __LINE__, pud_val(e)) +/* + * On all 4K setups, remap_4k_pfn() equates to remap_pfn_range() */ #define remap_4k_pfn(vma, addr, pfn, prot) \ remap_pfn_range((vma), (addr), (pfn), PAGE_SIZE, (prot)) -#endif /* _ASM_POWERPC_PGTABLE_4K_H */ + +#endif /* _ASM_POWERPC_PGTABLE_PPC64_4K_H */ diff --git a/arch/powerpc/include/asm/pgtable-ppc64-64k.h b/arch/powerpc/include/asm/pgtable-ppc64-64k.h new file mode 100644 index 00000000000..6cc085b945a --- /dev/null +++ b/arch/powerpc/include/asm/pgtable-ppc64-64k.h @@ -0,0 +1,42 @@ +#ifndef _ASM_POWERPC_PGTABLE_PPC64_64K_H +#define _ASM_POWERPC_PGTABLE_PPC64_64K_H + +#include <asm-generic/pgtable-nopud.h> + + +#define PTE_INDEX_SIZE 12 +#define PMD_INDEX_SIZE 12 +#define PUD_INDEX_SIZE 0 +#define PGD_INDEX_SIZE 4 + +#ifndef __ASSEMBLY__ + +#define PTE_TABLE_SIZE (sizeof(real_pte_t) << PTE_INDEX_SIZE) +#define PMD_TABLE_SIZE (sizeof(pmd_t) << PMD_INDEX_SIZE) +#define PGD_TABLE_SIZE (sizeof(pgd_t) << PGD_INDEX_SIZE) + +#define PTRS_PER_PTE (1 << PTE_INDEX_SIZE) +#define PTRS_PER_PMD (1 << PMD_INDEX_SIZE) +#define PTRS_PER_PGD (1 << PGD_INDEX_SIZE) + +/* With 4k base page size, hugepage PTEs go at the PMD level */ +#define MIN_HUGEPTE_SHIFT PAGE_SHIFT + +/* PMD_SHIFT determines what a second-level page table entry can map */ +#define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE) +#define PMD_SIZE (1UL << PMD_SHIFT) +#define PMD_MASK (~(PMD_SIZE-1)) + +/* PGDIR_SHIFT determines what a third-level page table entry can map */ +#define PGDIR_SHIFT (PMD_SHIFT + PMD_INDEX_SIZE) +#define PGDIR_SIZE (1UL << PGDIR_SHIFT) +#define PGDIR_MASK (~(PGDIR_SIZE-1)) + +#endif /* __ASSEMBLY__ */ + +/* Bits to mask out from a PMD to get to the PTE page */ +#define PMD_MASKED_BITS 0x1ff +/* Bits to mask out from a PGD/PUD to get to the PMD page */ +#define PUD_MASKED_BITS 0x1ff + +#endif /* _ASM_POWERPC_PGTABLE_PPC64_64K_H */ diff --git a/arch/powerpc/include/asm/pgtable-ppc64.h b/arch/powerpc/include/asm/pgtable-ppc64.h index c627877fcf1..542073836b2 100644 --- a/arch/powerpc/include/asm/pgtable-ppc64.h +++ b/arch/powerpc/include/asm/pgtable-ppc64.h @@ -11,9 +11,9 @@ #endif /* __ASSEMBLY__ */ #ifdef CONFIG_PPC_64K_PAGES -#include <asm/pgtable-64k.h> +#include <asm/pgtable-ppc64-64k.h> #else -#include <asm/pgtable-4k.h> +#include <asm/pgtable-ppc64-4k.h> #endif #define FIRST_USER_ADDRESS 0 @@ -25,6 +25,8 @@ PUD_INDEX_SIZE + PGD_INDEX_SIZE + PAGE_SHIFT) #define PGTABLE_RANGE (ASM_CONST(1) << PGTABLE_EADDR_SIZE) + +/* Some sanity checking */ #if TASK_SIZE_USER64 > PGTABLE_RANGE #error TASK_SIZE_USER64 exceeds pagetable range #endif @@ -33,7 +35,6 @@ #error TASK_SIZE_USER64 exceeds user VSID range #endif - /* * Define the address range of the vmalloc VM area. */ @@ -76,29 +77,26 @@ /* - * Common bits in a linux-style PTE. These match the bits in the - * (hardware-defined) PowerPC PTE as closely as possible. Additional - * bits may be defined in pgtable-*.h + * Include the PTE bits definitions */ -#define _PAGE_PRESENT 0x0001 /* software: pte contains a translation */ -#define _PAGE_USER 0x0002 /* matches one of the PP bits */ -#define _PAGE_FILE 0x0002 /* (!present only) software: pte holds file offset */ -#define _PAGE_EXEC 0x0004 /* No execute on POWER4 and newer (we invert) */ -#define _PAGE_GUARDED 0x0008 -#define _PAGE_COHERENT 0x0010 /* M: enforce memory coherence (SMP systems) */ -#define _PAGE_NO_CACHE 0x0020 /* I: cache inhibit */ -#define _PAGE_WRITETHRU 0x0040 /* W: cache write-through */ -#define _PAGE_DIRTY 0x0080 /* C: page changed */ -#define _PAGE_ACCESSED 0x0100 /* R: page referenced */ -#define _PAGE_RW 0x0200 /* software: user write access allowed */ -#define _PAGE_BUSY 0x0800 /* software: PTE & hash are busy */ - -/* Strong Access Ordering */ -#define _PAGE_SAO (_PAGE_WRITETHRU | _PAGE_NO_CACHE | _PAGE_COHERENT) - -#define _PAGE_BASE (_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_COHERENT) - -#define _PAGE_WRENABLE (_PAGE_RW | _PAGE_DIRTY) +#include <asm/pte-hash64.h> + +/* To make some generic powerpc code happy */ +#ifndef _PAGE_HWEXEC +#define _PAGE_HWEXEC 0 +#endif + +/* Some other useful definitions */ +#define PTE_RPN_MAX (1UL << (64 - PTE_RPN_SHIFT)) +#define PTE_RPN_MASK (~((1UL<<PTE_RPN_SHIFT)-1)) + +/* _PAGE_CHG_MASK masks of bits that are to be preserved accross + * pgprot changes + */ +#define _PAGE_CHG_MASK (PTE_RPN_MASK | _PAGE_HPTEFLAGS | _PAGE_DIRTY | \ + _PAGE_ACCESSED | _PAGE_SPECIAL) + + /* __pgprot defined in arch/powerpc/include/asm/page.h */ #define PAGE_NONE __pgprot(_PAGE_PRESENT | _PAGE_ACCESSED) @@ -117,16 +115,9 @@ #define PAGE_AGP __pgprot(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_NO_CACHE) #define HAVE_PAGE_AGP -#define PAGE_PROT_BITS (_PAGE_GUARDED | _PAGE_COHERENT | \ - _PAGE_NO_CACHE | _PAGE_WRITETHRU | \ - _PAGE_4K_PFN | _PAGE_RW | _PAGE_USER | \ - _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_EXEC) -/* PTEIDX nibble */ -#define _PTEIDX_SECONDARY 0x8 -#define _PTEIDX_GROUP_IX 0x7 +/* We always have _PAGE_SPECIAL on 64 bit */ +#define __HAVE_ARCH_PTE_SPECIAL -/* To make some generic powerpc code happy */ -#define _PAGE_HWEXEC 0 /* * POWER4 and newer have per page execute protection, older chips can only @@ -163,6 +154,38 @@ #ifndef __ASSEMBLY__ /* + * This is the default implementation of various PTE accessors, it's + * used in all cases except Book3S with 64K pages where we have a + * concept of sub-pages + */ +#ifndef __real_pte + +#ifdef STRICT_MM_TYPECHECKS +#define __real_pte(e,p) ((real_pte_t){(e)}) +#define __rpte_to_pte(r) ((r).pte) +#else +#define __real_pte(e,p) (e) +#define __rpte_to_pte(r) (__pte(r)) +#endif +#define __rpte_to_hidx(r,index) (pte_val(__rpte_to_pte(r)) >> 12) + +#define pte_iterate_hashed_subpages(rpte, psize, va, index, shift) \ + do { \ + index = 0; \ + shift = mmu_psize_defs[psize].shift; \ + +#define pte_iterate_hashed_end() } while(0) + +#ifdef CONFIG_PPC_HAS_HASH_64K +#define pte_pagesize_index(mm, addr, pte) get_slice_psize(mm, addr) +#else +#define pte_pagesize_index(mm, addr, pte) MMU_PAGE_4K +#endif + +#endif /* __real_pte */ + + +/* * Conversion functions: convert a page and protection to a page entry, * and a page entry and page directory to the page they refer to. * diff --git a/arch/powerpc/include/asm/pte-40x.h b/arch/powerpc/include/asm/pte-40x.h new file mode 100644 index 00000000000..07630faae02 --- /dev/null +++ b/arch/powerpc/include/asm/pte-40x.h @@ -0,0 +1,64 @@ +#ifndef _ASM_POWERPC_PTE_40x_H +#define _ASM_POWERPC_PTE_40x_H +#ifdef __KERNEL__ + +/* + * At present, all PowerPC 400-class processors share a similar TLB + * architecture. The instruction and data sides share a unified, + * 64-entry, fully-associative TLB which is maintained totally under + * software control. In addition, the instruction side has a + * hardware-managed, 4-entry, fully-associative TLB which serves as a + * first level to the shared TLB. These two TLBs are known as the UTLB + * and ITLB, respectively (see "mmu.h" for definitions). + * + * There are several potential gotchas here. The 40x hardware TLBLO + * field looks like this: + * + * 0 1 2 3 4 ... 18 19 20 21 22 23 24 25 26 27 28 29 30 31 + * RPN..................... 0 0 EX WR ZSEL....... W I M G + * + * Where possible we make the Linux PTE bits match up with this + * + * - bits 20 and 21 must be cleared, because we use 4k pages (40x can + * support down to 1k pages), this is done in the TLBMiss exception + * handler. + * - We use only zones 0 (for kernel pages) and 1 (for user pages) + * of the 16 available. Bit 24-26 of the TLB are cleared in the TLB + * miss handler. Bit 27 is PAGE_USER, thus selecting the correct + * zone. + * - PRESENT *must* be in the bottom two bits because swap cache + * entries use the top 30 bits. Because 40x doesn't support SMP + * anyway, M is irrelevant so we borrow it for PAGE_PRESENT. Bit 30 + * is cleared in the TLB miss handler before the TLB entry is loaded. + * - All other bits of the PTE are loaded into TLBLO without + * modification, leaving us only the bits 20, 21, 24, 25, 26, 30 for + * software PTE bits. We actually use use bits 21, 24, 25, and + * 30 respectively for the software bits: ACCESSED, DIRTY, RW, and + * PRESENT. + */ + +#define _PAGE_GUARDED 0x001 /* G: page is guarded from prefetch */ +#define _PAGE_FILE 0x001 /* when !present: nonlinear file mapping */ +#define _PAGE_PRESENT 0x002 /* software: PTE contains a translation */ +#define _PAGE_NO_CACHE 0x004 /* I: caching is inhibited */ +#define _PAGE_WRITETHRU 0x008 /* W: caching is write-through */ +#define _PAGE_USER 0x010 /* matches one of the zone permission bits */ +#define _PAGE_RW 0x040 /* software: Writes permitted */ +#define _PAGE_DIRTY 0x080 /* software: dirty page */ +#define _PAGE_HWWRITE 0x100 /* hardware: Dirty & RW, set in exception */ +#define _PAGE_HWEXEC 0x200 /* hardware: EX permission */ +#define _PAGE_ACCESSED 0x400 /* software: R: page referenced */ + +#define _PMD_PRESENT 0x400 /* PMD points to page of PTEs */ +#define _PMD_BAD 0x802 +#define _PMD_SIZE 0x0e0 /* size field, != 0 for large-page PMD entry */ +#define _PMD_SIZE_4M 0x0c0 +#define _PMD_SIZE_16M 0x0e0 + +#define PMD_PAGE_SIZE(pmdval) (1024 << (((pmdval) & _PMD_SIZE) >> 4)) + +/* Until my rework is finished, 40x still needs atomic PTE updates */ +#define PTE_ATOMIC_UPDATES 1 + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_40x_H */ diff --git a/arch/powerpc/include/asm/pte-44x.h b/arch/powerpc/include/asm/pte-44x.h new file mode 100644 index 00000000000..37e98bcf83e --- /dev/null +++ b/arch/powerpc/include/asm/pte-44x.h @@ -0,0 +1,102 @@ +#ifndef _ASM_POWERPC_PTE_44x_H +#define _ASM_POWERPC_PTE_44x_H +#ifdef __KERNEL__ + +/* + * Definitions for PPC440 + * + * Because of the 3 word TLB entries to support 36-bit addressing, + * the attribute are difficult to map in such a fashion that they + * are easily loaded during exception processing. I decided to + * organize the entry so the ERPN is the only portion in the + * upper word of the PTE and the attribute bits below are packed + * in as sensibly as they can be in the area below a 4KB page size + * oriented RPN. This at least makes it easy to load the RPN and + * ERPN fields in the TLB. -Matt + * + * This isn't entirely true anymore, at least some bits are now + * easier to move into the TLB from the PTE. -BenH. + * + * 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 + * + * Newer 440 cores (440x6 as used on AMCC 460EX/460GT) have additional + * TLB2 storage attibute fields. Those are: + * + * TLB2: + * 0...10 11 12 13 14 15 16...31 + * no change WL1 IL1I IL1D IL2I IL2D no change + * + * 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 original PPC440 cores, + * because it doesn't support SMP. However, some later 460 variants + * have -some- form of SMP support and so I keep the bit there for + * future use + * + * With the PPC 44x Linux implementation, the 0-11th LSBs of the PTE are used + * for memory protection related functions (see PTE structure in + * include/asm-ppc/mmu.h). The _PAGE_XXX definitions in this file map to the + * above bits. Note that the bit values are CPU specific, not architecture + * specific. + * + * The kernel PTE entry holds an arch-dependent swp_entry structure under + * certain situations. In other words, in such situations some portion of + * the PTE bits are used as a swp_entry. In the PPC implementation, the + * 3-24th LSB are shared with swp_entry, however the 0-2nd three LSB still + * hold protection values. That means the three protection bits are + * reserved for both PTE and SWAP entry at the most significant 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_FILE 0x00000004 /* S: nonlinear file mapping */ +#define _PAGE_HWEXEC 0x00000004 /* H: Execute permission */ +#define _PAGE_ACCESSED 0x00000008 /* S: Page referenced */ +#define _PAGE_DIRTY 0x00000010 /* S: Page dirty */ +#define _PAGE_SPECIAL 0x00000020 /* S: Special page */ +#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_NO_CACHE 0x00000400 /* H: I bit */ +#define _PAGE_WRITETHRU 0x00000800 /* H: W bit */ + +/* TODO: Add large page lowmem mapping support */ +#define _PMD_PRESENT 0 +#define _PMD_PRESENT_MASK (PAGE_MASK) +#define _PMD_BAD (~PAGE_MASK) + +/* ERPN in a PTE never gets cleared, ignore it */ +#define _PTE_NONE_MASK 0xffffffff00000000ULL + + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_44x_H */ diff --git a/arch/powerpc/include/asm/pte-8xx.h b/arch/powerpc/include/asm/pte-8xx.h new file mode 100644 index 00000000000..b07acfd330b --- /dev/null +++ b/arch/powerpc/include/asm/pte-8xx.h @@ -0,0 +1,64 @@ +#ifndef _ASM_POWERPC_PTE_8xx_H +#define _ASM_POWERPC_PTE_8xx_H +#ifdef __KERNEL__ + +/* + * The PowerPC MPC8xx uses a TLB with hardware assisted, software tablewalk. + * We also use the two level tables, but we can put the real bits in them + * needed for the TLB and tablewalk. These definitions require Mx_CTR.PPM = 0, + * Mx_CTR.PPCS = 0, and MD_CTR.TWAM = 1. The level 2 descriptor has + * additional page protection (when Mx_CTR.PPCS = 1) that allows TLB hit + * based upon user/super access. The TLB does not have accessed nor write + * protect. We assume that if the TLB get loaded with an entry it is + * accessed, and overload the changed bit for write protect. We use + * two bits in the software pte that are supposed to be set to zero in + * the TLB entry (24 and 25) for these indicators. Although the level 1 + * descriptor contains the guarded and writethrough/copyback bits, we can + * set these at the page level since they get copied from the Mx_TWC + * register when the TLB entry is loaded. We will use bit 27 for guard, since + * that is where it exists in the MD_TWC, and bit 26 for writethrough. + * These will get masked from the level 2 descriptor at TLB load time, and + * copied to the MD_TWC before it gets loaded. + * Large page sizes added. We currently support two sizes, 4K and 8M. + * This also allows a TLB hander optimization because we can directly + * load the PMD into MD_TWC. The 8M pages are only used for kernel + * mapping of well known areas. The PMD (PGD) entries contain control + * flags in addition to the address, so care must be taken that the + * software no longer assumes these are only pointers. + */ + +/* Definitions for 8xx embedded chips. */ +#define _PAGE_PRESENT 0x0001 /* Page is valid */ +#define _PAGE_FILE 0x0002 /* when !present: nonlinear file mapping */ +#define _PAGE_NO_CACHE 0x0002 /* I: cache inhibit */ +#define _PAGE_SHARED 0x0004 /* No ASID (context) compare */ + +/* These five software bits must be masked out when the entry is loaded + * into the TLB. + */ +#define _PAGE_EXEC 0x0008 /* software: i-cache coherency required */ +#define _PAGE_GUARDED 0x0010 /* software: guarded access */ +#define _PAGE_DIRTY 0x0020 /* software: page changed */ +#define _PAGE_RW 0x0040 /* software: user write access allowed */ +#define _PAGE_ACCESSED 0x0080 /* software: page referenced */ + +/* Setting any bits in the nibble with the follow two controls will + * require a TLB exception handler change. It is assumed unused bits + * are always zero. + */ +#define _PAGE_HWWRITE 0x0100 /* h/w write enable: never set in Linux PTE */ +#define _PAGE_USER 0x0800 /* One of the PP bits, the other is USER&~RW */ + +#define _PMD_PRESENT 0x0001 +#define _PMD_BAD 0x0ff0 +#define _PMD_PAGE_MASK 0x000c +#define _PMD_PAGE_8M 0x000c + +#define _PTE_NONE_MASK _PAGE_ACCESSED + +/* Until my rework is finished, 8xx still needs atomic PTE updates */ +#define PTE_ATOMIC_UPDATES 1 + + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_8xx_H */ diff --git a/arch/powerpc/include/asm/pte-fsl-booke.h b/arch/powerpc/include/asm/pte-fsl-booke.h new file mode 100644 index 00000000000..10820f58acf --- /dev/null +++ b/arch/powerpc/include/asm/pte-fsl-booke.h @@ -0,0 +1,48 @@ +#ifndef _ASM_POWERPC_PTE_FSL_BOOKE_H +#define _ASM_POWERPC_PTE_FSL_BOOKE_H +#ifdef __KERNEL__ + +/* PTE bit definitions for Freescale BookE SW loaded TLB MMU based + * processors + * + MMU Assist Register 3: + + 32 33 34 35 36 ... 50 51 52 53 54 55 56 57 58 59 60 61 62 63 + RPN...................... 0 0 U0 U1 U2 U3 UX SX UW SW UR SR + + - PRESENT *must* be in the bottom three bits because swap cache + entries use the top 29 bits. + + - FILE *must* be in the bottom three bits because swap cache + entries use the top 29 bits. +*/ + +/* Definitions for FSL Book-E Cores */ +#define _PAGE_PRESENT 0x00001 /* S: PTE contains a translation */ +#define _PAGE_USER 0x00002 /* S: User page (maps to UR) */ +#define _PAGE_FILE 0x00002 /* S: when !present: nonlinear file mapping */ +#define _PAGE_RW 0x00004 /* S: Write permission (SW) */ +#define _PAGE_DIRTY 0x00008 /* S: Page dirty */ +#define _PAGE_HWEXEC 0x00010 /* H: SX permission */ +#define _PAGE_ACCESSED 0x00020 /* S: Page referenced */ + +#define _PAGE_ENDIAN 0x00040 /* H: E bit */ +#define _PAGE_GUARDED 0x00080 /* H: G bit */ +#define _PAGE_COHERENT 0x00100 /* H: M bit */ +#define _PAGE_NO_CACHE 0x00200 /* H: I bit */ +#define _PAGE_WRITETHRU 0x00400 /* H: W bit */ +#define _PAGE_SPECIAL 0x00800 /* S: Special page */ + +#ifdef CONFIG_PTE_64BIT +/* ERPN in a PTE never gets cleared, ignore it */ +#define _PTE_NONE_MASK 0xffffffffffff0000ULL +/* We extend the size of the PTE flags area when using 64-bit PTEs */ +#define PTE_RPN_SHIFT (PAGE_SHIFT + 8) +#endif + +#define _PMD_PRESENT 0 +#define _PMD_PRESENT_MASK (PAGE_MASK) +#define _PMD_BAD (~PAGE_MASK) + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_FSL_BOOKE_H */ diff --git a/arch/powerpc/include/asm/pte-hash32.h b/arch/powerpc/include/asm/pte-hash32.h new file mode 100644 index 00000000000..6afe22b02f2 --- /dev/null +++ b/arch/powerpc/include/asm/pte-hash32.h @@ -0,0 +1,49 @@ +#ifndef _ASM_POWERPC_PTE_HASH32_H +#define _ASM_POWERPC_PTE_HASH32_H +#ifdef __KERNEL__ + +/* + * The "classic" 32-bit implementation of the PowerPC MMU uses a hash + * table containing PTEs, together with a set of 16 segment registers, + * to define the virtual to physical address mapping. + * + * We use the hash table as an extended TLB, i.e. a cache of currently + * active mappings. We maintain a two-level page table tree, much + * like that used by the i386, for the sake of the Linux memory + * management code. Low-level assembler code in hash_low_32.S + * (procedure hash_page) is responsible for extracting ptes from the + * tree and putting them into the hash table when necessary, and + * updating the accessed and modified bits in the page table tree. + */ + +#define _PAGE_PRESENT 0x001 /* software: pte contains a translation */ +#define _PAGE_HASHPTE 0x002 /* hash_page has made an HPTE for this pte */ +#define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */ +#define _PAGE_USER 0x004 /* usermode access allowed */ +#define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */ +#define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */ +#define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */ +#define _PAGE_WRITETHRU 0x040 /* W: cache write-through */ +#define _PAGE_DIRTY 0x080 /* C: page changed */ +#define _PAGE_ACCESSED 0x100 /* R: page referenced */ +#define _PAGE_EXEC 0x200 /* software: i-cache coherency required */ +#define _PAGE_RW 0x400 /* software: user write access allowed */ +#define _PAGE_SPECIAL 0x800 /* software: Special page */ + +#ifdef CONFIG_PTE_64BIT +/* We never clear the high word of the pte */ +#define _PTE_NONE_MASK (0xffffffff00000000ULL | _PAGE_HASHPTE) +#else +#define _PTE_NONE_MASK _PAGE_HASHPTE +#endif + +#define _PMD_PRESENT 0 +#define _PMD_PRESENT_MASK (PAGE_MASK) +#define _PMD_BAD (~PAGE_MASK) + +/* Hash table based platforms need atomic updates of the linux PTE */ +#define PTE_ATOMIC_UPDATES 1 + + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_HASH32_H */ diff --git a/arch/powerpc/include/asm/pte-hash64-4k.h b/arch/powerpc/include/asm/pte-hash64-4k.h new file mode 100644 index 00000000000..29fdc158fe3 --- /dev/null +++ b/arch/powerpc/include/asm/pte-hash64-4k.h @@ -0,0 +1,20 @@ +/* To be include by pgtable-hash64.h only */ + +/* PTE bits */ +#define _PAGE_HASHPTE 0x0400 /* software: pte has an associated HPTE */ +#define _PAGE_SECONDARY 0x8000 /* software: HPTE is in secondary group */ +#define _PAGE_GROUP_IX 0x7000 /* software: HPTE index within group */ +#define _PAGE_F_SECOND _PAGE_SECONDARY +#define _PAGE_F_GIX _PAGE_GROUP_IX +#define _PAGE_SPECIAL 0x10000 /* software: special page */ + +/* There is no 4K PFN hack on 4K pages */ +#define _PAGE_4K_PFN 0 + +/* PTE flags to conserve for HPTE identification */ +#define _PAGE_HPTEFLAGS (_PAGE_BUSY | _PAGE_HASHPTE | \ + _PAGE_SECONDARY | _PAGE_GROUP_IX) + +/* shift to put page number into pte */ +#define PTE_RPN_SHIFT (17) + diff --git a/arch/powerpc/include/asm/pgtable-64k.h b/arch/powerpc/include/asm/pte-hash64-64k.h index 7389003349a..e05d26fa372 100644 --- a/arch/powerpc/include/asm/pgtable-64k.h +++ b/arch/powerpc/include/asm/pte-hash64-64k.h @@ -1,76 +1,6 @@ -#ifndef _ASM_POWERPC_PGTABLE_64K_H -#define _ASM_POWERPC_PGTABLE_64K_H - -#include <asm-generic/pgtable-nopud.h> - - -#define PTE_INDEX_SIZE 12 -#define PMD_INDEX_SIZE 12 -#define PUD_INDEX_SIZE 0 -#define PGD_INDEX_SIZE 4 - -#ifndef __ASSEMBLY__ -#define PTE_TABLE_SIZE (sizeof(real_pte_t) << PTE_INDEX_SIZE) -#define PMD_TABLE_SIZE (sizeof(pmd_t) << PMD_INDEX_SIZE) -#define PGD_TABLE_SIZE (sizeof(pgd_t) << PGD_INDEX_SIZE) - -#define PTRS_PER_PTE (1 << PTE_INDEX_SIZE) -#define PTRS_PER_PMD (1 << PMD_INDEX_SIZE) -#define PTRS_PER_PGD (1 << PGD_INDEX_SIZE) - -#ifdef CONFIG_PPC_SUBPAGE_PROT -/* - * For the sub-page protection option, we extend the PGD with one of - * these. Basically we have a 3-level tree, with the top level being - * the protptrs array. To optimize speed and memory consumption when - * only addresses < 4GB are being protected, pointers to the first - * four pages of sub-page protection words are stored in the low_prot - * array. - * Each page of sub-page protection words protects 1GB (4 bytes - * protects 64k). For the 3-level tree, each page of pointers then - * protects 8TB. - */ -struct subpage_prot_table { - unsigned long maxaddr; /* only addresses < this are protected */ - unsigned int **protptrs[2]; - unsigned int *low_prot[4]; -}; - -#undef PGD_TABLE_SIZE -#define PGD_TABLE_SIZE ((sizeof(pgd_t) << PGD_INDEX_SIZE) + \ - sizeof(struct subpage_prot_table)) - -#define SBP_L1_BITS (PAGE_SHIFT - 2) -#define SBP_L2_BITS (PAGE_SHIFT - 3) -#define SBP_L1_COUNT (1 << SBP_L1_BITS) -#define SBP_L2_COUNT (1 << SBP_L2_BITS) -#define SBP_L2_SHIFT (PAGE_SHIFT + SBP_L1_BITS) -#define SBP_L3_SHIFT (SBP_L2_SHIFT + SBP_L2_BITS) - -extern void subpage_prot_free(pgd_t *pgd); - -static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) -{ - return (struct subpage_prot_table *)(pgd + PTRS_PER_PGD); -} -#endif /* CONFIG_PPC_SUBPAGE_PROT */ -#endif /* __ASSEMBLY__ */ - -/* With 4k base page size, hugepage PTEs go at the PMD level */ -#define MIN_HUGEPTE_SHIFT PAGE_SHIFT - -/* PMD_SHIFT determines what a second-level page table entry can map */ -#define PMD_SHIFT (PAGE_SHIFT + PTE_INDEX_SIZE) -#define PMD_SIZE (1UL << PMD_SHIFT) -#define PMD_MASK (~(PMD_SIZE-1)) - -/* PGDIR_SHIFT determines what a third-level page table entry can map */ -#define PGDIR_SHIFT (PMD_SHIFT + PMD_INDEX_SIZE) -#define PGDIR_SIZE (1UL << PGDIR_SHIFT) -#define PGDIR_MASK (~(PGDIR_SIZE-1)) +/* To be include by pgtable-hash64.h only */ /* Additional PTE bits (don't change without checking asm in hash_low.S) */ -#define __HAVE_ARCH_PTE_SPECIAL #define _PAGE_SPECIAL 0x00000400 /* software: special page */ #define _PAGE_HPTE_SUB 0x0ffff000 /* combo only: sub pages HPTE bits */ #define _PAGE_HPTE_SUB0 0x08000000 /* combo only: first sub page */ @@ -107,21 +37,15 @@ static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) * of addressable physical space, or 46 bits for the special 4k PFNs. */ #define PTE_RPN_SHIFT (30) -#define PTE_RPN_MAX (1UL << (64 - PTE_RPN_SHIFT)) -#define PTE_RPN_MASK (~((1UL<<PTE_RPN_SHIFT)-1)) - -/* _PAGE_CHG_MASK masks of bits that are to be preserved accross - * pgprot changes - */ -#define _PAGE_CHG_MASK (PTE_RPN_MASK | _PAGE_HPTEFLAGS | _PAGE_DIRTY | \ - _PAGE_ACCESSED | _PAGE_SPECIAL) -/* Bits to mask out from a PMD to get to the PTE page */ -#define PMD_MASKED_BITS 0x1ff -/* Bits to mask out from a PGD/PUD to get to the PMD page */ -#define PUD_MASKED_BITS 0x1ff +#ifndef __ASSEMBLY__ -/* Manipulate "rpte" values */ +/* + * With 64K pages on hash table, we have a special PTE format that + * uses a second "half" of the page table to encode sub-page information + * in order to deal with 64K made of 4K HW pages. Thus we override the + * generic accessors and iterators here + */ #define __real_pte(e,p) ((real_pte_t) { \ (e), pte_val(*((p) + PTRS_PER_PTE)) }) #define __rpte_to_hidx(r,index) ((pte_val((r).pte) & _PAGE_COMBO) ? \ @@ -130,7 +54,6 @@ static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) #define __rpte_sub_valid(rpte, index) \ (pte_val(rpte.pte) & (_PAGE_HPTE_SUB0 >> (index))) - /* Trick: we set __end to va + 64k, which happens works for * a 16M page as well as we want only one iteration */ @@ -152,4 +75,41 @@ static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) remap_pfn_range((vma), (addr), (pfn), PAGE_SIZE, \ __pgprot(pgprot_val((prot)) | _PAGE_4K_PFN)) -#endif /* _ASM_POWERPC_PGTABLE_64K_H */ + +#ifdef CONFIG_PPC_SUBPAGE_PROT +/* + * For the sub-page protection option, we extend the PGD with one of + * these. Basically we have a 3-level tree, with the top level being + * the protptrs array. To optimize speed and memory consumption when + * only addresses < 4GB are being protected, pointers to the first + * four pages of sub-page protection words are stored in the low_prot + * array. + * Each page of sub-page protection words protects 1GB (4 bytes + * protects 64k). For the 3-level tree, each page of pointers then + * protects 8TB. + */ +struct subpage_prot_table { + unsigned long maxaddr; /* only addresses < this are protected */ + unsigned int **protptrs[2]; + unsigned int *low_prot[4]; +}; + +#undef PGD_TABLE_SIZE +#define PGD_TABLE_SIZE ((sizeof(pgd_t) << PGD_INDEX_SIZE) + \ + sizeof(struct subpage_prot_table)) + +#define SBP_L1_BITS (PAGE_SHIFT - 2) +#define SBP_L2_BITS (PAGE_SHIFT - 3) +#define SBP_L1_COUNT (1 << SBP_L1_BITS) +#define SBP_L2_COUNT (1 << SBP_L2_BITS) +#define SBP_L2_SHIFT (PAGE_SHIFT + SBP_L1_BITS) +#define SBP_L3_SHIFT (SBP_L2_SHIFT + SBP_L2_BITS) + +extern void subpage_prot_free(pgd_t *pgd); + +static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd) +{ + return (struct subpage_prot_table *)(pgd + PTRS_PER_PGD); +} +#endif /* CONFIG_PPC_SUBPAGE_PROT */ +#endif /* __ASSEMBLY__ */ diff --git a/arch/powerpc/include/asm/pte-hash64.h b/arch/powerpc/include/asm/pte-hash64.h new file mode 100644 index 00000000000..62766636cc1 --- /dev/null +++ b/arch/powerpc/include/asm/pte-hash64.h @@ -0,0 +1,47 @@ +#ifndef _ASM_POWERPC_PTE_HASH64_H +#define _ASM_POWERPC_PTE_HASH64_H +#ifdef __KERNEL__ + +/* + * Common bits between 4K and 64K pages in a linux-style PTE. + * These match the bits in the (hardware-defined) PowerPC PTE as closely + * as possible. Additional bits may be defined in pgtable-hash64-*.h + */ +#define _PAGE_PRESENT 0x0001 /* software: pte contains a translation */ +#define _PAGE_USER 0x0002 /* matches one of the PP bits */ +#define _PAGE_FILE 0x0002 /* (!present only) software: pte holds file offset */ +#define _PAGE_EXEC 0x0004 /* No execute on POWER4 and newer (we invert) */ +#define _PAGE_GUARDED 0x0008 +#define _PAGE_COHERENT 0x0010 /* M: enforce memory coherence (SMP systems) */ +#define _PAGE_NO_CACHE 0x0020 /* I: cache inhibit */ +#define _PAGE_WRITETHRU 0x0040 /* W: cache write-through */ +#define _PAGE_DIRTY 0x0080 /* C: page changed */ +#define _PAGE_ACCESSED 0x0100 /* R: page referenced */ +#define _PAGE_RW 0x0200 /* software: user write access allowed */ +#define _PAGE_BUSY 0x0800 /* software: PTE & hash are busy */ + +/* Strong Access Ordering */ +#define _PAGE_SAO (_PAGE_WRITETHRU | _PAGE_NO_CACHE | _PAGE_COHERENT) + +#define _PAGE_BASE (_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_COHERENT) + +#define _PAGE_WRENABLE (_PAGE_RW | _PAGE_DIRTY) + +/* PTEIDX nibble */ +#define _PTEIDX_SECONDARY 0x8 +#define _PTEIDX_GROUP_IX 0x7 + +#define PAGE_PROT_BITS (_PAGE_GUARDED | _PAGE_COHERENT | \ + _PAGE_NO_CACHE | _PAGE_WRITETHRU | \ + _PAGE_4K_PFN | _PAGE_RW | _PAGE_USER | \ + _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_EXEC) + + +#ifdef CONFIG_PPC_64K_PAGES +#include <asm/pte-hash64-64k.h> +#else +#include <asm/pte-hash64-4k.h> +#endif + +#endif /* __KERNEL__ */ +#endif /* _ASM_POWERPC_PTE_HASH64_H */ diff --git a/arch/powerpc/include/asm/udbg.h b/arch/powerpc/include/asm/udbg.h index 6418ceea44b..cd21e5e6b04 100644 --- a/arch/powerpc/include/asm/udbg.h +++ b/arch/powerpc/include/asm/udbg.h @@ -15,6 +15,7 @@ #include <linux/init.h> extern void (*udbg_putc)(char c); +extern void (*udbg_flush)(void); extern int (*udbg_getc)(void); extern int (*udbg_getc_poll)(void); diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile index dfec3d2790b..71901fbda4a 100644 --- a/arch/powerpc/kernel/Makefile +++ b/arch/powerpc/kernel/Makefile @@ -75,7 +75,7 @@ obj-y += time.o prom.o traps.o setup-common.o \ obj-$(CONFIG_PPC32) += entry_32.o setup_32.o obj-$(CONFIG_PPC64) += dma-iommu.o iommu.o obj-$(CONFIG_KGDB) += kgdb.o -obj-$(CONFIG_PPC_MULTIPLATFORM) += prom_init.o +obj-$(CONFIG_PPC_OF_BOOT_TRAMPOLINE) += prom_init.o obj-$(CONFIG_MODULES) += ppc_ksyms.o obj-$(CONFIG_BOOTX_TEXT) += btext.o obj-$(CONFIG_SMP) += smp.o diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c index 638838691b2..ccea2431ddf 100644 --- a/arch/powerpc/kernel/cputable.c +++ b/arch/powerpc/kernel/cputable.c @@ -731,6 +731,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_setup = __setup_cpu_750, .machine_check = machine_check_generic, .platform = "ppc750", + .oprofile_cpu_type = "ppc/750", + .oprofile_type = PPC_OPROFILE_G4, }, { /* 750FX rev 2.0 must disable HID0[DPM] */ .pvr_mask = 0xffffffff, @@ -746,6 +748,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_setup = __setup_cpu_750, .machine_check = machine_check_generic, .platform = "ppc750", + .oprofile_cpu_type = "ppc/750", + .oprofile_type = PPC_OPROFILE_G4, }, { /* 750FX (All revs except 2.0) */ .pvr_mask = 0xffff0000, @@ -761,6 +765,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_setup = __setup_cpu_750fx, .machine_check = machine_check_generic, .platform = "ppc750", + .oprofile_cpu_type = "ppc/750", + .oprofile_type = PPC_OPROFILE_G4, }, { /* 750GX */ .pvr_mask = 0xffff0000, @@ -776,6 +782,8 @@ static struct cpu_spec __initdata cpu_specs[] = { .cpu_setup = __setup_cpu_750fx, .machine_check = machine_check_generic, .platform = "ppc750", + .oprofile_cpu_type = "ppc/750", + .oprofile_type = PPC_OPROFILE_G4, }, { /* 740/750 (L2CR bit need fixup for 740) */ .pvr_mask = 0xffff0000, diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S index a1c4cfd25de..f8c2e6b6f45 100644 --- a/arch/powerpc/kernel/head_32.S +++ b/arch/powerpc/kernel/head_32.S @@ -108,18 +108,21 @@ __start: * because OF may have I/O devices mapped into that area * (particularly on CHRP). */ -#ifdef CONFIG_PPC_MULTIPLATFORM cmpwi 0,r5,0 beq 1f +#ifdef CONFIG_PPC_OF_BOOT_TRAMPOLINE /* find out where we are now */ bcl 20,31,$+4 0: mflr r8 /* r8 = runtime addr here */ addis r8,r8,(_stext - 0b)@ha addi r8,r8,(_stext - 0b)@l /* current runtime base addr */ bl prom_init +#endif /* CONFIG_PPC_OF_BOOT_TRAMPOLINE */ + + /* We never return. We also hit that trap if trying to boot + * from OF while CONFIG_PPC_OF_BOOT_TRAMPOLINE isn't selected */ trap -#endif /* * Check for BootX signature when supporting PowerMac and branch to diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index ebaedafc8e6..50ef505b8fb 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -1360,6 +1360,7 @@ _GLOBAL(__start_initialization_multiplatform) b .__after_prom_start _INIT_STATIC(__boot_from_prom) +#ifdef CONFIG_PPC_OF_BOOT_TRAMPOLINE /* Save parameters */ mr r31,r3 mr r30,r4 @@ -1390,7 +1391,10 @@ _INIT_STATIC(__boot_from_prom) /* Do all of the interaction with OF client interface */ mr r8,r26 bl .prom_init - /* We never return */ +#endif /* #CONFIG_PPC_OF_BOOT_TRAMPOLINE */ + + /* We never return. We also hit that trap if trying to boot + * from OF while CONFIG_PPC_OF_BOOT_TRAMPOLINE isn't selected */ trap _STATIC(__after_prom_start) diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c index 23b8b5e36f9..48ea2008b20 100644 --- a/arch/powerpc/kernel/irq.c +++ b/arch/powerpc/kernel/irq.c @@ -171,7 +171,7 @@ int show_interrupts(struct seq_file *p, void *v) { int i = *(loff_t *)v, j; struct irqaction *action; - irq_desc_t *desc; + struct irq_desc *desc; unsigned long flags; if (i == 0) { @@ -1038,7 +1038,7 @@ arch_initcall(irq_late_init); static int virq_debug_show(struct seq_file *m, void *private) { unsigned long flags; - irq_desc_t *desc; + struct irq_desc *desc; const char *p; char none[] = "none"; int i; diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index 2ad17315fc8..2603f20984c 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -1482,7 +1482,7 @@ void __init pcibios_resource_survey(void) * we proceed to assigning things that were left unassigned */ if (!(ppc_pci_flags & PPC_PCI_PROBE_ONLY)) { - pr_debug("PCI: Assigning unassigned resouces...\n"); + pr_debug("PCI: Assigning unassigned resources...\n"); pci_assign_unassigned_resources(); } diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 7f1b33d5e30..4d5ebb46b2c 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -2283,6 +2283,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, */ prom_init_stdout(); + prom_printf("Preparing to boot %s", PTRRELOC((char *)linux_banner)); + /* * Get default machine type. At this point, we do not differentiate * between pSeries SMP and pSeries LPAR diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh index ea3a2ec03ff..1ac136b128f 100644 --- a/arch/powerpc/kernel/prom_init_check.sh +++ b/arch/powerpc/kernel/prom_init_check.sh @@ -20,7 +20,7 @@ WHITELIST="add_reloc_offset __bss_start __bss_stop copy_and_flush _end enter_prom memcpy memset reloc_offset __secondary_hold __secondary_hold_acknowledge __secondary_hold_spinloop __start strcmp strcpy strlcpy strlen strncmp strstr logo_linux_clut224 -reloc_got2 kernstart_addr memstart_addr" +reloc_got2 kernstart_addr memstart_addr linux_banner" NM="$1" OBJ="$2" diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c index 7d6c9bb8c77..fc9af47e212 100644 --- a/arch/powerpc/kernel/udbg.c +++ b/arch/powerpc/kernel/udbg.c @@ -18,6 +18,7 @@ #include <asm/udbg.h> void (*udbg_putc)(char c); +void (*udbg_flush)(void); int (*udbg_getc)(void); int (*udbg_getc_poll)(void); @@ -76,6 +77,9 @@ void udbg_puts(const char *s) while ((c = *s++) != '\0') udbg_putc(c); } + + if (udbg_flush) + udbg_flush(); } #if 0 else { @@ -98,6 +102,9 @@ int udbg_write(const char *s, int n) } } + if (udbg_flush) + udbg_flush(); + return n - remain; } diff --git a/arch/powerpc/kernel/udbg_16550.c b/arch/powerpc/kernel/udbg_16550.c index 7b7da8cfd5e..0362a891e54 100644 --- a/arch/powerpc/kernel/udbg_16550.c +++ b/arch/powerpc/kernel/udbg_16550.c @@ -48,14 +48,21 @@ struct NS16550 { static struct NS16550 __iomem *udbg_comport; -static void udbg_550_putc(char c) +static void udbg_550_flush(void) { if (udbg_comport) { while ((in_8(&udbg_comport->lsr) & LSR_THRE) == 0) /* wait for idle */; - out_8(&udbg_comport->thr, c); + } +} + +static void udbg_550_putc(char c) +{ + if (udbg_comport) { if (c == '\n') udbg_550_putc('\r'); + udbg_550_flush(); + out_8(&udbg_comport->thr, c); } } @@ -108,6 +115,7 @@ void udbg_init_uart(void __iomem *comport, unsigned int speed, /* Clear & enable FIFOs */ out_8(&udbg_comport->fcr ,0x07); udbg_putc = udbg_550_putc; + udbg_flush = udbg_550_flush; udbg_getc = udbg_550_getc; udbg_getc_poll = udbg_550_getc_poll; } @@ -149,14 +157,21 @@ unsigned int udbg_probe_uart_speed(void __iomem *comport, unsigned int clock) } #ifdef CONFIG_PPC_MAPLE -void udbg_maple_real_putc(char c) +void udbg_maple_real_flush(void) { if (udbg_comport) { while ((real_readb(&udbg_comport->lsr) & LSR_THRE) == 0) /* wait for idle */; - real_writeb(c, &udbg_comport->thr); eieio(); + } +} + +void udbg_maple_real_putc(char c) +{ + if (udbg_comport) { if (c == '\n') udbg_maple_real_putc('\r'); + udbg_maple_real_flush(); + real_writeb(c, &udbg_comport->thr); eieio(); } } @@ -165,20 +180,28 @@ void __init udbg_init_maple_realmode(void) udbg_comport = (struct NS16550 __iomem *)0xf40003f8; udbg_putc = udbg_maple_real_putc; + udbg_flush = udbg_maple_real_flush; udbg_getc = NULL; udbg_getc_poll = NULL; } #endif /* CONFIG_PPC_MAPLE */ #ifdef CONFIG_PPC_PASEMI -void udbg_pas_real_putc(char c) +void udbg_pas_real_flush(void) { if (udbg_comport) { while ((real_205_readb(&udbg_comport->lsr) & LSR_THRE) == 0) /* wait for idle */; - real_205_writeb(c, &udbg_comport->thr); eieio(); + } +} + +void udbg_pas_real_putc(char c) +{ + if (udbg_comport) { if (c == '\n') udbg_pas_real_putc('\r'); + udbg_pas_real_flush(); + real_205_writeb(c, &udbg_comport->thr); eieio(); } } @@ -187,6 +210,7 @@ void udbg_init_pas_realmode(void) udbg_comport = (struct NS16550 __iomem *)0xfcff03f8UL; udbg_putc = udbg_pas_real_putc; + udbg_flush = udbg_pas_real_flush; udbg_getc = NULL; udbg_getc_poll = NULL; } @@ -195,14 +219,21 @@ void udbg_init_pas_realmode(void) #ifdef CONFIG_PPC_EARLY_DEBUG_44x #include <platforms/44x/44x.h> -static void udbg_44x_as1_putc(char c) +static int udbg_44x_as1_flush(void) { if (udbg_comport) { while ((as1_readb(&udbg_comport->lsr) & LSR_THRE) == 0) /* wait for idle */; - as1_writeb(c, &udbg_comport->thr); eieio(); + } +} + +static void udbg_44x_as1_putc(char c) +{ + if (udbg_comport) { if (c == '\n') udbg_44x_as1_putc('\r'); + udbg_44x_as1_flush(); + as1_writeb(c, &udbg_comport->thr); eieio(); } } @@ -222,19 +253,27 @@ void __init udbg_init_44x_as1(void) (struct NS16550 __iomem *)PPC44x_EARLY_DEBUG_VIRTADDR; udbg_putc = udbg_44x_as1_putc; + udbg_flush = udbg_44x_as1_flush; udbg_getc = udbg_44x_as1_getc; } #endif /* CONFIG_PPC_EARLY_DEBUG_44x */ #ifdef CONFIG_PPC_EARLY_DEBUG_40x -static void udbg_40x_real_putc(char c) +static void udbg_40x_real_flush(void) { if (udbg_comport) { while ((real_readb(&udbg_comport->lsr) & LSR_THRE) == 0) /* wait for idle */; - real_writeb(c, &udbg_comport->thr); eieio(); + } +} + +static void udbg_40x_real_putc(char c) +{ + if (udbg_comport) { if (c == '\n') udbg_40x_real_putc('\r'); + udbg_40x_real_flush(); + real_writeb(c, &udbg_comport->thr); eieio(); } } @@ -254,6 +293,7 @@ void __init udbg_init_40x_realmode(void) CONFIG_PPC_EARLY_DEBUG_40x_PHYSADDR; udbg_putc = udbg_40x_real_putc; + udbg_flush = udbg_40x_real_flush; udbg_getc = udbg_40x_real_getc; udbg_getc_poll = NULL; } diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile index 953cc4a1cde..6d2838fc879 100644 --- a/arch/powerpc/mm/Makefile +++ b/arch/powerpc/mm/Makefile @@ -6,7 +6,7 @@ ifeq ($(CONFIG_PPC64),y) EXTRA_CFLAGS += -mno-minimal-toc endif -obj-y := fault.o mem.o pgtable.o \ +obj-y := fault.o mem.o pgtable.o gup.o \ init_$(CONFIG_WORD_SIZE).o \ pgtable_$(CONFIG_WORD_SIZE).o obj-$(CONFIG_PPC_MMU_NOHASH) += mmu_context_nohash.o tlb_nohash.o \ @@ -14,7 +14,7 @@ obj-$(CONFIG_PPC_MMU_NOHASH) += mmu_context_nohash.o tlb_nohash.o \ hash-$(CONFIG_PPC_NATIVE) := hash_native_64.o obj-$(CONFIG_PPC64) += hash_utils_64.o \ slb_low.o slb.o stab.o \ - gup.o mmap.o $(hash-y) + mmap.o $(hash-y) obj-$(CONFIG_PPC_STD_MMU_32) += ppc_mmu_32.o obj-$(CONFIG_PPC_STD_MMU) += hash_low_$(CONFIG_WORD_SIZE).o \ tlb_hash$(CONFIG_WORD_SIZE).o \ diff --git a/arch/powerpc/mm/gup.c b/arch/powerpc/mm/gup.c index 28a114db3ba..bc400c78c97 100644 --- a/arch/powerpc/mm/gup.c +++ b/arch/powerpc/mm/gup.c @@ -14,6 +14,8 @@ #include <linux/rwsem.h> #include <asm/pgtable.h> +#ifdef __HAVE_ARCH_PTE_SPECIAL + /* * The performance critical leaf functions are made noinline otherwise gcc * inlines everything into a single function which results in too much @@ -151,8 +153,11 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, unsigned long addr, len, end; unsigned long next; pgd_t *pgdp; - int psize, nr = 0; + int nr = 0; +#ifdef CONFIG_PPC64 unsigned int shift; + int psize; +#endif pr_debug("%s(%lx,%x,%s)\n", __func__, start, nr_pages, write ? "write" : "read"); @@ -205,8 +210,13 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, */ local_irq_disable(); +#ifdef CONFIG_PPC64 + /* Those bits are related to hugetlbfs implementation and only exist + * on 64-bit for now + */ psize = get_slice_psize(mm, addr); shift = mmu_psize_defs[psize].shift; +#endif /* CONFIG_PPC64 */ #ifdef CONFIG_HUGETLB_PAGE if (unlikely(mmu_huge_psizes[psize])) { @@ -236,7 +246,9 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, do { pgd_t pgd = *pgdp; +#ifdef CONFIG_PPC64 VM_BUG_ON(shift != mmu_psize_defs[get_slice_psize(mm, addr)].shift); +#endif pr_debug(" %016lx: normal pgd %p\n", addr, (void *)pgd_val(pgd)); next = pgd_addr_end(addr, end); @@ -279,3 +291,5 @@ slow_irqon: return ret; } } + +#endif /* __HAVE_ARCH_PTE_SPECIAL */ diff --git a/arch/powerpc/platforms/512x/Kconfig b/arch/powerpc/platforms/512x/Kconfig index 326852c78b8..4dac9b0525a 100644 --- a/arch/powerpc/platforms/512x/Kconfig +++ b/arch/powerpc/platforms/512x/Kconfig @@ -12,7 +12,7 @@ config PPC_MPC5121 config MPC5121_ADS bool "Freescale MPC5121E ADS" - depends on PPC_MULTIPLATFORM && PPC32 + depends on 6xx select DEFAULT_UIMAGE select PPC_MPC5121 select MPC5121_ADS_CPLD @@ -21,7 +21,7 @@ config MPC5121_ADS config MPC5121_GENERIC bool "Generic support for simple MPC5121 based boards" - depends on PPC_MULTIPLATFORM && PPC32 + depends on 6xx select DEFAULT_UIMAGE select PPC_MPC5121 help diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig index 0465e5b36e6..e0b9454ae69 100644 --- a/arch/powerpc/platforms/52xx/Kconfig +++ b/arch/powerpc/platforms/52xx/Kconfig @@ -1,6 +1,6 @@ config PPC_MPC52xx bool "52xx-based boards" - depends on PPC_MULTIPLATFORM && PPC32 + depends on 6xx select PPC_CLOCK select PPC_PCI_CHOICE diff --git a/arch/powerpc/platforms/82xx/Kconfig b/arch/powerpc/platforms/82xx/Kconfig index 30f008b2f92..7c7df400382 100644 --- a/arch/powerpc/platforms/82xx/Kconfig +++ b/arch/powerpc/platforms/82xx/Kconfig @@ -1,6 +1,6 @@ menuconfig PPC_82xx bool "82xx-based boards (PQ II)" - depends on 6xx && PPC_MULTIPLATFORM + depends on 6xx if PPC_82xx diff --git a/arch/powerpc/platforms/83xx/Kconfig b/arch/powerpc/platforms/83xx/Kconfig index 83c664afc89..437d29a59d7 100644 --- a/arch/powerpc/platforms/83xx/Kconfig +++ b/arch/powerpc/platforms/83xx/Kconfig @@ -1,6 +1,6 @@ menuconfig PPC_83xx bool "83xx-based boards" - depends on 6xx && PPC_MULTIPLATFORM + depends on 6xx select PPC_UDBG_16550 select PPC_PCI_CHOICE select FSL_PCI if PCI diff --git a/arch/powerpc/platforms/86xx/Kconfig b/arch/powerpc/platforms/86xx/Kconfig index 611d0d19eb5..fdaf4ddaa95 100644 --- a/arch/powerpc/platforms/86xx/Kconfig +++ b/arch/powerpc/platforms/86xx/Kconfig @@ -1,7 +1,7 @@ config PPC_86xx menuconfig PPC_86xx bool "86xx-based boards" - depends on 6xx && PPC_MULTIPLATFORM + depends on 6xx select FSL_SOC select ALTIVEC help diff --git a/arch/powerpc/platforms/Kconfig b/arch/powerpc/platforms/Kconfig index b4ab3728653..68b9b8fd9f8 100644 --- a/arch/powerpc/platforms/Kconfig +++ b/arch/powerpc/platforms/Kconfig @@ -1,14 +1,5 @@ menu "Platform support" -config PPC_MULTIPLATFORM - bool - depends on PPC64 || 6xx - default y - -config CLASSIC32 - def_bool y - depends on 6xx && PPC_MULTIPLATFORM - source "arch/powerpc/platforms/pseries/Kconfig" source "arch/powerpc/platforms/iseries/Kconfig" source "arch/powerpc/platforms/chrp/Kconfig" @@ -32,12 +23,24 @@ source "arch/powerpc/platforms/amigaone/Kconfig" config PPC_NATIVE bool - depends on PPC_MULTIPLATFORM + depends on 6xx || PPC64 help Support for running natively on the hardware, i.e. without a hypervisor. This option is not user-selectable but should be selected by all platforms that need it. +config PPC_OF_BOOT_TRAMPOLINE + bool "Support booting from Open Firmware or yaboot" + depends on 6xx || PPC64 + default y + help + Support from booting from Open Firmware or yaboot using an + Open Firmware client interface. This enables the kernel to + communicate with open firmware to retrieve system informations + such as the device tree. + + In case of doubt, say Y + config UDBG_RTAS_CONSOLE bool "RTAS based debug console" depends on PPC_RTAS @@ -71,7 +74,7 @@ config PPC_I8259 config U3_DART bool - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 default n config PPC_RTAS @@ -188,7 +191,7 @@ config PPC601_SYNC_FIX config TAU bool "On-chip CPU temperature sensor support" - depends on CLASSIC32 + depends on 6xx help G3 and G4 processors have an on-chip temperature sensor called the 'Thermal Assist Unit (TAU)', which, in theory, can measure the on-die diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype index 9428c0e11b2..9da795e4933 100644 --- a/arch/powerpc/platforms/Kconfig.cputype +++ b/arch/powerpc/platforms/Kconfig.cputype @@ -57,9 +57,17 @@ config E200 endchoice +# Until we have a choice of exclusive CPU types on 64-bit, we always +# use PPC_BOOK3S. On 32-bit, this is equivalent to 6xx which is +# "classic" MMU + +config PPC_BOOK3S + def_bool y + depends on PPC64 || 6xx + config POWER4_ONLY bool "Optimize for POWER4" - depends on PPC64 + depends on PPC64 && PPC_BOOK3S default n ---help--- Cause the compiler to optimize for POWER4/POWER5/PPC970 processors. @@ -68,16 +76,16 @@ config POWER4_ONLY config POWER3 bool - depends on PPC64 + depends on PPC64 && PPC_BOOK3S default y if !POWER4_ONLY config POWER4 - depends on PPC64 + depends on PPC64 && PPC_BOOK3S def_bool y config TUNE_CELL bool "Optimize for Cell Broadband Engine" - depends on PPC64 + depends on PPC64 && PPC_BOOK3S help Cause the compiler to optimize for the PPE of the Cell Broadband Engine. This will make the code run considerably faster on Cell @@ -147,7 +155,7 @@ config PHYS_64BIT config ALTIVEC bool "AltiVec Support" - depends on CLASSIC32 || POWER4 + depends on 6xx || POWER4 ---help--- This option enables kernel support for the Altivec extensions to the PowerPC processor. The kernel currently supports saving and restoring diff --git a/arch/powerpc/platforms/amigaone/Kconfig b/arch/powerpc/platforms/amigaone/Kconfig index 9276a96cede..02247671771 100644 --- a/arch/powerpc/platforms/amigaone/Kconfig +++ b/arch/powerpc/platforms/amigaone/Kconfig @@ -1,6 +1,6 @@ config AMIGAONE bool "Eyetech AmigaOne/MAI Teron" - depends on PPC32 && BROKEN_ON_SMP && PPC_MULTIPLATFORM + depends on 6xx && BROKEN_ON_SMP select PPC_I8259 select PPC_INDIRECT_PCI select PPC_UDBG_16550 diff --git a/arch/powerpc/platforms/cell/Kconfig b/arch/powerpc/platforms/cell/Kconfig index 037f59a4bfe..40e24c39ad0 100644 --- a/arch/powerpc/platforms/cell/Kconfig +++ b/arch/powerpc/platforms/cell/Kconfig @@ -23,7 +23,7 @@ config PPC_CELL_NATIVE config PPC_IBM_CELL_BLADE bool "IBM Cell Blade" - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S select PPC_CELL_NATIVE select MMIO_NVRAM select PPC_UDBG_16550 @@ -31,7 +31,7 @@ config PPC_IBM_CELL_BLADE config PPC_CELLEB bool "Toshiba's Cell Reference Set 'Celleb' Architecture" - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S select PPC_CELL_NATIVE select HAS_TXX9_SERIAL select PPC_UDBG_BEAT @@ -40,9 +40,14 @@ config PPC_CELLEB config PPC_CELL_QPACE bool "IBM Cell - QPACE" - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S select PPC_CELL_COMMON +config AXON_MSI + bool + depends on PPC_IBM_CELL_BLADE && PCI_MSI + default y + menu "Cell Broadband Engine options" depends on PPC_CELL diff --git a/arch/powerpc/platforms/cell/Makefile b/arch/powerpc/platforms/cell/Makefile index 43eccb27030..83fafe92264 100644 --- a/arch/powerpc/platforms/cell/Makefile +++ b/arch/powerpc/platforms/cell/Makefile @@ -28,7 +28,7 @@ obj-$(CONFIG_SPU_BASE) += spu_callbacks.o spu_base.o \ $(spu-manage-y) \ spufs/ -obj-$(CONFIG_PCI_MSI) += axon_msi.o +obj-$(CONFIG_AXON_MSI) += axon_msi.o # qpace setup obj-$(CONFIG_PPC_CELL_QPACE) += qpace_setup.o diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c index 9e4f2739341..d6a519e6e1c 100644 --- a/arch/powerpc/platforms/cell/spufs/file.c +++ b/arch/powerpc/platforms/cell/spufs/file.c @@ -568,16 +568,17 @@ spufs_regs_write(struct file *file, const char __user *buffer, struct spu_lscsa *lscsa = ctx->csa.lscsa; int ret; - size = min_t(ssize_t, sizeof lscsa->gprs - *pos, size); - if (size <= 0) + if (*pos >= sizeof(lscsa->gprs)) return -EFBIG; + + size = min_t(ssize_t, sizeof(lscsa->gprs) - *pos, size); *pos += size; ret = spu_acquire_saved(ctx); if (ret) return ret; - ret = copy_from_user(lscsa->gprs + *pos - size, + ret = copy_from_user((char *)lscsa->gprs + *pos - size, buffer, size) ? -EFAULT : size; spu_release_saved(ctx); @@ -623,10 +624,11 @@ spufs_fpcr_write(struct file *file, const char __user * buffer, struct spu_lscsa *lscsa = ctx->csa.lscsa; int ret; - size = min_t(ssize_t, sizeof(lscsa->fpcr) - *pos, size); - if (size <= 0) + if (*pos >= sizeof(lscsa->fpcr)) return -EFBIG; + size = min_t(ssize_t, sizeof(lscsa->fpcr) - *pos, size); + ret = spu_acquire_saved(ctx); if (ret) return ret; diff --git a/arch/powerpc/platforms/chrp/Kconfig b/arch/powerpc/platforms/chrp/Kconfig index 22b4b4e3b6f..37d438bd5b7 100644 --- a/arch/powerpc/platforms/chrp/Kconfig +++ b/arch/powerpc/platforms/chrp/Kconfig @@ -1,6 +1,6 @@ config PPC_CHRP bool "Common Hardware Reference Platform (CHRP) based machines" - depends on PPC_MULTIPLATFORM && PPC32 + depends on 6xx select MPIC select PPC_I8259 select PPC_INDIRECT_PCI diff --git a/arch/powerpc/platforms/embedded6xx/Kconfig b/arch/powerpc/platforms/embedded6xx/Kconfig index 4f9f8184d16..291ac9d8cbe 100644 --- a/arch/powerpc/platforms/embedded6xx/Kconfig +++ b/arch/powerpc/platforms/embedded6xx/Kconfig @@ -1,6 +1,6 @@ config EMBEDDED6xx bool "Embedded 6xx/7xx/7xxx-based boards" - depends on PPC32 && BROKEN_ON_SMP && PPC_MULTIPLATFORM + depends on 6xx && BROKEN_ON_SMP config LINKSTATION bool "Linkstation / Kurobox(HG) from Buffalo" diff --git a/arch/powerpc/platforms/iseries/Kconfig b/arch/powerpc/platforms/iseries/Kconfig index 7ddd0a2c802..647e8778743 100644 --- a/arch/powerpc/platforms/iseries/Kconfig +++ b/arch/powerpc/platforms/iseries/Kconfig @@ -1,6 +1,6 @@ config PPC_ISERIES bool "IBM Legacy iSeries" - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S select PPC_INDIRECT_IO select PPC_PCI_CHOICE if EMBEDDED diff --git a/arch/powerpc/platforms/iseries/irq.c b/arch/powerpc/platforms/iseries/irq.c index 701d9297c20..94f44475883 100644 --- a/arch/powerpc/platforms/iseries/irq.c +++ b/arch/powerpc/platforms/iseries/irq.c @@ -214,7 +214,7 @@ void __init iSeries_activate_IRQs() unsigned long flags; for_each_irq (irq) { - irq_desc_t *desc = get_irq_desc(irq); + struct irq_desc *desc = get_irq_desc(irq); if (desc && desc->chip && desc->chip->startup) { spin_lock_irqsave(&desc->lock, flags); diff --git a/arch/powerpc/platforms/maple/Kconfig b/arch/powerpc/platforms/maple/Kconfig index a6467a5591f..1ea621a94c3 100644 --- a/arch/powerpc/platforms/maple/Kconfig +++ b/arch/powerpc/platforms/maple/Kconfig @@ -1,5 +1,5 @@ config PPC_MAPLE - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S bool "Maple 970FX Evaluation Board" select PCI select MPIC diff --git a/arch/powerpc/platforms/pasemi/Kconfig b/arch/powerpc/platforms/pasemi/Kconfig index 348e0619e3e..a2aeb327d18 100644 --- a/arch/powerpc/platforms/pasemi/Kconfig +++ b/arch/powerpc/platforms/pasemi/Kconfig @@ -1,5 +1,5 @@ config PPC_PASEMI - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S bool "PA Semi SoC-based platforms" default n select MPIC diff --git a/arch/powerpc/platforms/powermac/Kconfig b/arch/powerpc/platforms/powermac/Kconfig index 055990ca8ce..1e1a0873e1d 100644 --- a/arch/powerpc/platforms/powermac/Kconfig +++ b/arch/powerpc/platforms/powermac/Kconfig @@ -1,6 +1,6 @@ config PPC_PMAC bool "Apple PowerMac based machines" - depends on PPC_MULTIPLATFORM + depends on PPC_BOOK3S select MPIC select PCI select PPC_INDIRECT_PCI if PPC32 diff --git a/arch/powerpc/platforms/powermac/pic.h b/arch/powerpc/platforms/powermac/pic.h index c44c89f5e53..d622a8345aa 100644 --- a/arch/powerpc/platforms/powermac/pic.h +++ b/arch/powerpc/platforms/powermac/pic.h @@ -3,7 +3,7 @@ #include <linux/irq.h> -extern struct hw_interrupt_type pmac_pic; +extern struct irq_chip pmac_pic; extern void pmac_pic_init(void); extern int pmac_get_irq(void); diff --git a/arch/powerpc/platforms/prep/Kconfig b/arch/powerpc/platforms/prep/Kconfig index 29d411279b0..bf8330ef2e7 100644 --- a/arch/powerpc/platforms/prep/Kconfig +++ b/arch/powerpc/platforms/prep/Kconfig @@ -1,6 +1,6 @@ config PPC_PREP bool "PowerPC Reference Platform (PReP) based machines" - depends on PPC_MULTIPLATFORM && PPC32 && BROKEN + depends on 6xx && BROKEN select MPIC select PPC_I8259 select PPC_INDIRECT_PCI diff --git a/arch/powerpc/platforms/ps3/Kconfig b/arch/powerpc/platforms/ps3/Kconfig index 920cf7a454b..204ae5c402f 100644 --- a/arch/powerpc/platforms/ps3/Kconfig +++ b/arch/powerpc/platforms/ps3/Kconfig @@ -1,6 +1,6 @@ config PPC_PS3 bool "Sony PS3" - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S select PPC_CELL select USB_ARCH_HAS_OHCI select USB_OHCI_LITTLE_ENDIAN diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig index ddc2a307cd5..c0e6ec240f4 100644 --- a/arch/powerpc/platforms/pseries/Kconfig +++ b/arch/powerpc/platforms/pseries/Kconfig @@ -1,5 +1,5 @@ config PPC_PSERIES - depends on PPC_MULTIPLATFORM && PPC64 + depends on PPC64 && PPC_BOOK3S bool "IBM pSeries & new (POWER5-based) iSeries" select MPIC select PPC_I8259 @@ -25,6 +25,11 @@ config EEH depends on PPC_PSERIES && PCI default y if !EMBEDDED +config PSERIES_MSI + bool + depends on PCI_MSI && EEH + default y + config SCANLOG tristate "Scanlog dump interface" depends on RTAS_PROC && PPC_PSERIES diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile index dfe574af2dc..0ce691df595 100644 --- a/arch/powerpc/platforms/pseries/Makefile +++ b/arch/powerpc/platforms/pseries/Makefile @@ -15,7 +15,7 @@ obj-$(CONFIG_SCANLOG) += scanlog.o obj-$(CONFIG_EEH) += eeh.o eeh_cache.o eeh_driver.o eeh_event.o eeh_sysfs.o obj-$(CONFIG_KEXEC) += kexec.o obj-$(CONFIG_PCI) += pci.o pci_dlpar.o -obj-$(CONFIG_PCI_MSI) += msi.o +obj-$(CONFIG_PSERIES_MSI) += msi.o obj-$(CONFIG_HOTPLUG_CPU) += hotplug-cpu.o obj-$(CONFIG_MEMORY_HOTPLUG) += hotplug-memory.o diff --git a/arch/powerpc/platforms/pseries/msi.c b/arch/powerpc/platforms/pseries/msi.c index 3e0d6ef3eca..bf2e1ac4130 100644 --- a/arch/powerpc/platforms/pseries/msi.c +++ b/arch/powerpc/platforms/pseries/msi.c @@ -356,6 +356,27 @@ static int rtas_msi_check_device(struct pci_dev *pdev, int nvec, int type) return 0; } +static int check_msix_entries(struct pci_dev *pdev) +{ + struct msi_desc *entry; + int expected; + + /* There's no way for us to express to firmware that we want + * a discontiguous, or non-zero based, range of MSI-X entries. + * So we must reject such requests. */ + + expected = 0; + list_for_each_entry(entry, &pdev->msi_list, list) { + if (entry->msi_attrib.entry_nr != expected) { + pr_debug("rtas_msi: bad MSI-X entries.\n"); + return -EINVAL; + } + expected++; + } + + return 0; +} + static int rtas_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) { struct pci_dn *pdn; @@ -367,6 +388,9 @@ static int rtas_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type) if (!pdn) return -ENODEV; + if (type == PCI_CAP_ID_MSIX && check_msix_entries(pdev)) + return -EINVAL; + /* * Try the new more explicit firmware interface, if that fails fall * back to the old interface. The old interface is known to never |