From 1189be6508d45183013ddb82b18f4934193de274 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Thu, 11 Oct 2007 20:37:10 +1000 Subject: [POWERPC] Use 1TB segments This makes the kernel use 1TB segments for all kernel mappings and for user addresses of 1TB and above, on machines which support them (currently POWER5+, POWER6 and PA6T). We detect that the machine supports 1TB segments by looking at the ibm,processor-segment-sizes property in the device tree. We don't currently use 1TB segments for user addresses < 1T, since that would effectively prevent 32-bit processes from using huge pages unless we also had a way to revert to using 256MB segments. That would be possible but would involve extra complications (such as keeping track of which segment size was used when HPTEs were inserted) and is not addressed here. Parts of this patch were originally written by Ben Herrenschmidt. Signed-off-by: Paul Mackerras --- arch/powerpc/mm/hash_native_64.c | 92 ++++++++++++++++++++-------------------- 1 file changed, 46 insertions(+), 46 deletions(-) (limited to 'arch/powerpc/mm/hash_native_64.c') diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c index 6ba9b47e55a..34e5c0b219b 100644 --- a/arch/powerpc/mm/hash_native_64.c +++ b/arch/powerpc/mm/hash_native_64.c @@ -38,7 +38,7 @@ static DEFINE_SPINLOCK(native_tlbie_lock); -static inline void __tlbie(unsigned long va, unsigned int psize) +static inline void __tlbie(unsigned long va, int psize, int ssize) { unsigned int penc; @@ -48,18 +48,20 @@ static inline void __tlbie(unsigned long va, unsigned int psize) switch (psize) { case MMU_PAGE_4K: va &= ~0xffful; + va |= ssize << 8; asm volatile("tlbie %0,0" : : "r" (va) : "memory"); break; default: penc = mmu_psize_defs[psize].penc; va &= ~((1ul << mmu_psize_defs[psize].shift) - 1); va |= penc << 12; + va |= ssize << 8; asm volatile("tlbie %0,1" : : "r" (va) : "memory"); break; } } -static inline void __tlbiel(unsigned long va, unsigned int psize) +static inline void __tlbiel(unsigned long va, int psize, int ssize) { unsigned int penc; @@ -69,6 +71,7 @@ static inline void __tlbiel(unsigned long va, unsigned int psize) switch (psize) { case MMU_PAGE_4K: va &= ~0xffful; + va |= ssize << 8; asm volatile(".long 0x7c000224 | (%0 << 11) | (0 << 21)" : : "r"(va) : "memory"); break; @@ -76,6 +79,7 @@ static inline void __tlbiel(unsigned long va, unsigned int psize) penc = mmu_psize_defs[psize].penc; va &= ~((1ul << mmu_psize_defs[psize].shift) - 1); va |= penc << 12; + va |= ssize << 8; asm volatile(".long 0x7c000224 | (%0 << 11) | (1 << 21)" : : "r"(va) : "memory"); break; @@ -83,7 +87,7 @@ static inline void __tlbiel(unsigned long va, unsigned int psize) } -static inline void tlbie(unsigned long va, int psize, int local) +static inline void tlbie(unsigned long va, int psize, int ssize, int local) { unsigned int use_local = local && cpu_has_feature(CPU_FTR_TLBIEL); int lock_tlbie = !cpu_has_feature(CPU_FTR_LOCKLESS_TLBIE); @@ -94,10 +98,10 @@ static inline void tlbie(unsigned long va, int psize, int local) spin_lock(&native_tlbie_lock); asm volatile("ptesync": : :"memory"); if (use_local) { - __tlbiel(va, psize); + __tlbiel(va, psize, ssize); asm volatile("ptesync": : :"memory"); } else { - __tlbie(va, psize); + __tlbie(va, psize, ssize); asm volatile("eieio; tlbsync; ptesync": : :"memory"); } if (lock_tlbie && !use_local) @@ -126,7 +130,7 @@ static inline void native_unlock_hpte(struct hash_pte *hptep) static long native_hpte_insert(unsigned long hpte_group, unsigned long va, unsigned long pa, unsigned long rflags, - unsigned long vflags, int psize) + unsigned long vflags, int psize, int ssize) { struct hash_pte *hptep = htab_address + hpte_group; unsigned long hpte_v, hpte_r; @@ -153,7 +157,7 @@ static long native_hpte_insert(unsigned long hpte_group, unsigned long va, if (i == HPTES_PER_GROUP) return -1; - hpte_v = hpte_encode_v(va, psize) | vflags | HPTE_V_VALID; + hpte_v = hpte_encode_v(va, psize, ssize) | vflags | HPTE_V_VALID; hpte_r = hpte_encode_r(pa, psize) | rflags; if (!(vflags & HPTE_V_BOLTED)) { @@ -215,13 +219,14 @@ static long native_hpte_remove(unsigned long hpte_group) } static long native_hpte_updatepp(unsigned long slot, unsigned long newpp, - unsigned long va, int psize, int local) + unsigned long va, int psize, int ssize, + int local) { struct hash_pte *hptep = htab_address + slot; unsigned long hpte_v, want_v; int ret = 0; - want_v = hpte_encode_v(va, psize); + want_v = hpte_encode_v(va, psize, ssize); DBG_LOW(" update(va=%016lx, avpnv=%016lx, hash=%016lx, newpp=%x)", va, want_v & HPTE_V_AVPN, slot, newpp); @@ -243,39 +248,32 @@ static long native_hpte_updatepp(unsigned long slot, unsigned long newpp, native_unlock_hpte(hptep); /* Ensure it is out of the tlb too. */ - tlbie(va, psize, local); + tlbie(va, psize, ssize, local); return ret; } -static long native_hpte_find(unsigned long va, int psize) +static long native_hpte_find(unsigned long va, int psize, int ssize) { struct hash_pte *hptep; unsigned long hash; - unsigned long i, j; + unsigned long i; long slot; unsigned long want_v, hpte_v; - hash = hpt_hash(va, mmu_psize_defs[psize].shift); - want_v = hpte_encode_v(va, psize); + hash = hpt_hash(va, mmu_psize_defs[psize].shift, ssize); + want_v = hpte_encode_v(va, psize, ssize); - for (j = 0; j < 2; j++) { - slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; - for (i = 0; i < HPTES_PER_GROUP; i++) { - hptep = htab_address + slot; - hpte_v = hptep->v; + /* Bolted mappings are only ever in the primary group */ + slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; + for (i = 0; i < HPTES_PER_GROUP; i++) { + hptep = htab_address + slot; + hpte_v = hptep->v; - if (HPTE_V_COMPARE(hpte_v, want_v) - && (hpte_v & HPTE_V_VALID) - && ( !!(hpte_v & HPTE_V_SECONDARY) == j)) { - /* HPTE matches */ - if (j) - slot = -slot; - return slot; - } - ++slot; - } - hash = ~hash; + if (HPTE_V_COMPARE(hpte_v, want_v) && (hpte_v & HPTE_V_VALID)) + /* HPTE matches */ + return slot; + ++slot; } return -1; @@ -289,16 +287,16 @@ static long native_hpte_find(unsigned long va, int psize) * No need to lock here because we should be the only user. */ static void native_hpte_updateboltedpp(unsigned long newpp, unsigned long ea, - int psize) + int psize, int ssize) { unsigned long vsid, va; long slot; struct hash_pte *hptep; - vsid = get_kernel_vsid(ea); - va = (vsid << 28) | (ea & 0x0fffffff); + vsid = get_kernel_vsid(ea, ssize); + va = hpt_va(ea, vsid, ssize); - slot = native_hpte_find(va, psize); + slot = native_hpte_find(va, psize, ssize); if (slot == -1) panic("could not find page to bolt\n"); hptep = htab_address + slot; @@ -308,11 +306,11 @@ static void native_hpte_updateboltedpp(unsigned long newpp, unsigned long ea, (newpp & (HPTE_R_PP | HPTE_R_N)); /* Ensure it is out of the tlb too. */ - tlbie(va, psize, 0); + tlbie(va, psize, ssize, 0); } static void native_hpte_invalidate(unsigned long slot, unsigned long va, - int psize, int local) + int psize, int ssize, int local) { struct hash_pte *hptep = htab_address + slot; unsigned long hpte_v; @@ -323,7 +321,7 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long va, DBG_LOW(" invalidate(va=%016lx, hash: %x)\n", va, slot); - want_v = hpte_encode_v(va, psize); + want_v = hpte_encode_v(va, psize, ssize); native_lock_hpte(hptep); hpte_v = hptep->v; @@ -335,7 +333,7 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long va, hptep->v = 0; /* Invalidate the TLB */ - tlbie(va, psize, local); + tlbie(va, psize, ssize, local); local_irq_restore(flags); } @@ -345,7 +343,7 @@ static void native_hpte_invalidate(unsigned long slot, unsigned long va, #define LP_MASK(i) ((0xFF >> (i)) << LP_SHIFT) static void hpte_decode(struct hash_pte *hpte, unsigned long slot, - int *psize, unsigned long *va) + int *psize, int *ssize, unsigned long *va) { unsigned long hpte_r = hpte->r; unsigned long hpte_v = hpte->v; @@ -401,6 +399,7 @@ static void hpte_decode(struct hash_pte *hpte, unsigned long slot, *va = avpn; *psize = size; + *ssize = hpte_v >> HPTE_V_SSIZE_SHIFT; } /* @@ -417,7 +416,7 @@ static void native_hpte_clear(void) struct hash_pte *hptep = htab_address; unsigned long hpte_v, va; unsigned long pteg_count; - int psize; + int psize, ssize; pteg_count = htab_hash_mask + 1; @@ -443,9 +442,9 @@ static void native_hpte_clear(void) * already hold the native_tlbie_lock. */ if (hpte_v & HPTE_V_VALID) { - hpte_decode(hptep, slot, &psize, &va); + hpte_decode(hptep, slot, &psize, &ssize, &va); hptep->v = 0; - __tlbie(va, psize); + __tlbie(va, psize, ssize); } } @@ -468,6 +467,7 @@ static void native_flush_hash_range(unsigned long number, int local) real_pte_t pte; struct ppc64_tlb_batch *batch = &__get_cpu_var(ppc64_tlb_batch); unsigned long psize = batch->psize; + int ssize = batch->ssize; int i; local_irq_save(flags); @@ -477,14 +477,14 @@ static void native_flush_hash_range(unsigned long number, int local) pte = batch->pte[i]; pte_iterate_hashed_subpages(pte, psize, va, index, shift) { - hash = hpt_hash(va, shift); + hash = hpt_hash(va, shift, ssize); hidx = __rpte_to_hidx(pte, index); if (hidx & _PTEIDX_SECONDARY) hash = ~hash; slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; slot += hidx & _PTEIDX_GROUP_IX; hptep = htab_address + slot; - want_v = hpte_encode_v(va, psize); + want_v = hpte_encode_v(va, psize, ssize); native_lock_hpte(hptep); hpte_v = hptep->v; if (!HPTE_V_COMPARE(hpte_v, want_v) || @@ -504,7 +504,7 @@ static void native_flush_hash_range(unsigned long number, int local) pte_iterate_hashed_subpages(pte, psize, va, index, shift) { - __tlbiel(va, psize); + __tlbiel(va, psize, ssize); } pte_iterate_hashed_end(); } asm volatile("ptesync":::"memory"); @@ -521,7 +521,7 @@ static void native_flush_hash_range(unsigned long number, int local) pte_iterate_hashed_subpages(pte, psize, va, index, shift) { - __tlbie(va, psize); + __tlbie(va, psize, ssize); } pte_iterate_hashed_end(); } asm volatile("eieio; tlbsync; ptesync":::"memory"); -- cgit v1.2.3