diff --git a/arch/powerpc/include/asm/book3s/64/mmu-hash.h b/arch/powerpc/include/asm/book3s/64/mmu-hash.h index 10a34282829e0452eac0722f30e2c06b7516ef45..c68102293a192760f40477fcd36d49e1b7f94eb0 100644 --- a/arch/powerpc/include/asm/book3s/64/mmu-hash.h +++ b/arch/powerpc/include/asm/book3s/64/mmu-hash.h @@ -39,6 +39,7 @@ /* Bits in the SLB VSID word */ #define SLB_VSID_SHIFT 12 +#define SLB_VSID_SHIFT_256M SLB_VSID_SHIFT #define SLB_VSID_SHIFT_1T 24 #define SLB_VSID_SSIZE_SHIFT 62 #define SLB_VSID_B ASM_CONST(0xc000000000000000) @@ -521,9 +522,19 @@ extern void slb_set_size(u16 size); * because of the modulo operation in vsid scramble. */ +/* + * Max Va bits we support as of now is 68 bits. We want 19 bit + * context ID. + * Restrictions: + * GPU has restrictions of not able to access beyond 128TB + * (47 bit effective address). We also cannot do more than 20bit PID. + * For p4 and p5 which can only do 65 bit VA, we restrict our CONTEXT_BITS + * to 16 bits (ie, we can only have 2^16 pids at the same time). + */ +#define VA_BITS 68 #define CONTEXT_BITS 19 -#define ESID_BITS 18 -#define ESID_BITS_1T 6 +#define ESID_BITS (VA_BITS - (SID_SHIFT + CONTEXT_BITS)) +#define ESID_BITS_1T (VA_BITS - (SID_SHIFT_1T + CONTEXT_BITS)) #define ESID_BITS_MASK ((1 << ESID_BITS) - 1) #define ESID_BITS_1T_MASK ((1 << ESID_BITS_1T) - 1) @@ -533,7 +544,7 @@ extern void slb_set_size(u16 size); * The proto-VSID space has 2^(CONTEX_BITS + ESID_BITS) - 1 segments * available for user + kernel mapping. VSID 0 is reserved as invalid, contexts * 1-4 are used for kernel mapping. Each segment contains 2^28 bytes. Each - * context maps 2^46 bytes (64TB). + * context maps 2^49 bytes (512TB). * * We also need to avoid the last segment of the last context, because that * would give a protovsid of 0x1fffffffff. That will result in a VSID 0 @@ -545,54 +556,46 @@ extern void slb_set_size(u16 size); /* Would be nice to use KERNEL_REGION_ID here */ #define KERNEL_REGION_CONTEXT_OFFSET (0xc - 1) +/* + * For platforms that support on 65bit VA we limit the context bits + */ +#define MAX_USER_CONTEXT_65BIT_VA ((ASM_CONST(1) << (65 - (SID_SHIFT + ESID_BITS))) - 2) + /* * This should be computed such that protovosid * vsid_mulitplier - * doesn't overflow 64 bits. It should also be co-prime to vsid_modulus + * doesn't overflow 64 bits. The vsid_mutliplier should also be + * co-prime to vsid_modulus. We also need to make sure that number + * of bits in multiplied result (dividend) is less than twice the number of + * protovsid bits for our modulus optmization to work. + * + * The below table shows the current values used. + * |-------+------------+----------------------+------------+-------------------| + * | | Prime Bits | proto VSID_BITS_65VA | Total Bits | 2* prot VSID_BITS | + * |-------+------------+----------------------+------------+-------------------| + * | 1T | 24 | 25 | 49 | 50 | + * |-------+------------+----------------------+------------+-------------------| + * | 256MB | 24 | 37 | 61 | 74 | + * |-------+------------+----------------------+------------+-------------------| + * + * |-------+------------+----------------------+------------+--------------------| + * | | Prime Bits | proto VSID_BITS_68VA | Total Bits | 2* proto VSID_BITS | + * |-------+------------+----------------------+------------+--------------------| + * | 1T | 24 | 28 | 52 | 56 | + * |-------+------------+----------------------+------------+--------------------| + * | 256MB | 24 | 40 | 64 | 80 | + * |-------+------------+----------------------+------------+--------------------| + * */ #define VSID_MULTIPLIER_256M ASM_CONST(12538073) /* 24-bit prime */ -#define VSID_BITS_256M (CONTEXT_BITS + ESID_BITS) -#define VSID_MODULUS_256M ((1UL<= \ - * 2^36-1, then r3+1 has the 2^36 bit set. So, if r3+1 has \ - * the bit clear, r3 already has the answer we want, if it \ - * doesn't, the answer is the low 36 bits of r3+1. So in all \ - * cases the answer is the low 36 bits of (r3 + ((r3+1) >> 36))*/\ - addi rx,rt,1; \ - srdi rx,rx,VSID_BITS_##size; /* extract 2^VSID_BITS bit */ \ - add rt,rt,rx - /* 4 bits per slice and we have one slice per 1TB */ #define SLICE_ARRAY_SIZE (H_PGTABLE_RANGE >> 41) @@ -640,7 +643,7 @@ static inline void subpage_prot_init_new_context(struct mm_struct *mm) { } #define vsid_scramble(protovsid, size) \ ((((protovsid) * VSID_MULTIPLIER_##size) % VSID_MODULUS_##size)) -#else /* 1 */ +/* simplified form avoiding mod operation */ #define vsid_scramble(protovsid, size) \ ({ \ unsigned long x; \ @@ -648,6 +651,21 @@ static inline void subpage_prot_init_new_context(struct mm_struct *mm) { } x = (x >> VSID_BITS_##size) + (x & VSID_MODULUS_##size); \ (x + ((x+1) >> VSID_BITS_##size)) & VSID_MODULUS_##size; \ }) + +#else /* 1 */ +static inline unsigned long vsid_scramble(unsigned long protovsid, + unsigned long vsid_multiplier, int vsid_bits) +{ + unsigned long vsid; + unsigned long vsid_modulus = ((1UL << vsid_bits) - 1); + /* + * We have same multipler for both 256 and 1T segements now + */ + vsid = protovsid * vsid_multiplier; + vsid = (vsid >> vsid_bits) + (vsid & vsid_modulus); + return (vsid + ((vsid + 1) >> vsid_bits)) & vsid_modulus; +} + #endif /* 1 */ /* Returns the segment size indicator for a user address */ @@ -662,17 +680,30 @@ static inline int user_segment_size(unsigned long addr) static inline unsigned long get_vsid(unsigned long context, unsigned long ea, int ssize) { + unsigned long va_bits = VA_BITS; + unsigned long vsid_bits; + unsigned long protovsid; + /* * Bad address. We return VSID 0 for that */ if ((ea & ~REGION_MASK) >= H_PGTABLE_RANGE) return 0; - if (ssize == MMU_SEGSIZE_256M) - return vsid_scramble((context << ESID_BITS) - | ((ea >> SID_SHIFT) & ESID_BITS_MASK), 256M); - return vsid_scramble((context << ESID_BITS_1T) - | ((ea >> SID_SHIFT_1T) & ESID_BITS_1T_MASK), 1T); + if (!mmu_has_feature(MMU_FTR_68_BIT_VA)) + va_bits = 65; + + if (ssize == MMU_SEGSIZE_256M) { + vsid_bits = va_bits - SID_SHIFT; + protovsid = (context << ESID_BITS) | + ((ea >> SID_SHIFT) & ESID_BITS_MASK); + return vsid_scramble(protovsid, VSID_MULTIPLIER_256M, vsid_bits); + } + /* 1T segment */ + vsid_bits = va_bits - SID_SHIFT_1T; + protovsid = (context << ESID_BITS_1T) | + ((ea >> SID_SHIFT_1T) & ESID_BITS_1T_MASK); + return vsid_scramble(protovsid, VSID_MULTIPLIER_1T, vsid_bits); } /* diff --git a/arch/powerpc/include/asm/mmu.h b/arch/powerpc/include/asm/mmu.h index 065e762fae85c7c73fa71cc4f0834827935a40eb..78260409dc9c0250f14141b0df6f8219c220f0a0 100644 --- a/arch/powerpc/include/asm/mmu.h +++ b/arch/powerpc/include/asm/mmu.h @@ -28,6 +28,10 @@ * Individual features below. */ +/* + * Support for 68 bit VA space. We added that from ISA 2.05 + */ +#define MMU_FTR_68_BIT_VA ASM_CONST(0x00002000) /* * Kernel read only support. * We added the ppp value 0b110 in ISA 2.04. @@ -109,10 +113,10 @@ #define MMU_FTRS_POWER4 MMU_FTRS_DEFAULT_HPTE_ARCH_V2 #define MMU_FTRS_PPC970 MMU_FTRS_POWER4 | MMU_FTR_TLBIE_CROP_VA #define MMU_FTRS_POWER5 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE -#define MMU_FTRS_POWER6 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE | MMU_FTR_KERNEL_RO -#define MMU_FTRS_POWER7 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE | MMU_FTR_KERNEL_RO -#define MMU_FTRS_POWER8 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE | MMU_FTR_KERNEL_RO -#define MMU_FTRS_POWER9 MMU_FTRS_POWER4 | MMU_FTR_LOCKLESS_TLBIE | MMU_FTR_KERNEL_RO +#define MMU_FTRS_POWER6 MMU_FTRS_POWER5 | MMU_FTR_KERNEL_RO | MMU_FTR_68_BIT_VA +#define MMU_FTRS_POWER7 MMU_FTRS_POWER6 +#define MMU_FTRS_POWER8 MMU_FTRS_POWER6 +#define MMU_FTRS_POWER9 MMU_FTRS_POWER6 #define MMU_FTRS_CELL MMU_FTRS_DEFAULT_HPTE_ARCH_V2 | \ MMU_FTR_CI_LARGE_PAGE #define MMU_FTRS_PA6T MMU_FTRS_DEFAULT_HPTE_ARCH_V2 | \ @@ -136,7 +140,7 @@ enum { MMU_FTR_NO_SLBIE_B | MMU_FTR_16M_PAGE | MMU_FTR_TLBIEL | MMU_FTR_LOCKLESS_TLBIE | MMU_FTR_CI_LARGE_PAGE | MMU_FTR_1T_SEGMENT | MMU_FTR_TLBIE_CROP_VA | - MMU_FTR_KERNEL_RO | + MMU_FTR_KERNEL_RO | MMU_FTR_68_BIT_VA | #ifdef CONFIG_PPC_RADIX_MMU MMU_FTR_TYPE_RADIX | #endif @@ -290,7 +294,10 @@ static inline bool early_radix_enabled(void) #define MMU_PAGE_16G 14 #define MMU_PAGE_64G 15 -/* N.B. we need to change the type of hpte_page_sizes if this gets to be > 16 */ +/* + * N.B. we need to change the type of hpte_page_sizes if this gets to be > 16 + * Also we need to change he type of mm_context.low/high_slices_psize. + */ #define MMU_PAGE_COUNT 16 #ifdef CONFIG_PPC_BOOK3S_64 diff --git a/arch/powerpc/kvm/book3s_64_mmu_host.c b/arch/powerpc/kvm/book3s_64_mmu_host.c index b35f44c98d1feaaa3bea8ed4468b58cb8a6ca0b4..74b0153780e39da60446e8bd747e50d8287f7b78 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_host.c +++ b/arch/powerpc/kvm/book3s_64_mmu_host.c @@ -229,6 +229,7 @@ void kvmppc_mmu_unmap_page(struct kvm_vcpu *vcpu, struct kvmppc_pte *pte) static struct kvmppc_sid_map *create_sid_map(struct kvm_vcpu *vcpu, u64 gvsid) { + unsigned long vsid_bits = VSID_BITS_65_256M; struct kvmppc_sid_map *map; struct kvmppc_vcpu_book3s *vcpu_book3s = to_book3s(vcpu); u16 sid_map_mask; @@ -257,7 +258,12 @@ static struct kvmppc_sid_map *create_sid_map(struct kvm_vcpu *vcpu, u64 gvsid) kvmppc_mmu_pte_flush(vcpu, 0, 0); kvmppc_mmu_flush_segments(vcpu); } - map->host_vsid = vsid_scramble(vcpu_book3s->proto_vsid_next++, 256M); + + if (mmu_has_feature(MMU_FTR_68_BIT_VA)) + vsid_bits = VSID_BITS_256M; + + map->host_vsid = vsid_scramble(vcpu_book3s->proto_vsid_next++, + VSID_MULTIPLIER_256M, vsid_bits); map->guest_vsid = gvsid; map->valid = true; diff --git a/arch/powerpc/mm/mmu_context_book3s64.c b/arch/powerpc/mm/mmu_context_book3s64.c index a10e972221c40bf801535899a294580114d18771..e5fde156e11dd8688255979b578126d2a7524955 100644 --- a/arch/powerpc/mm/mmu_context_book3s64.c +++ b/arch/powerpc/mm/mmu_context_book3s64.c @@ -59,7 +59,14 @@ static int alloc_context_id(int min_id, int max_id) int hash__alloc_context_id(void) { - return alloc_context_id(MIN_USER_CONTEXT, MAX_USER_CONTEXT); + unsigned long max; + + if (mmu_has_feature(MMU_FTR_68_BIT_VA)) + max = MAX_USER_CONTEXT; + else + max = MAX_USER_CONTEXT_65BIT_VA; + + return alloc_context_id(MIN_USER_CONTEXT, max); } EXPORT_SYMBOL_GPL(hash__alloc_context_id); diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S index ba1f8696c3388ecce9ee7b6085cc171e0d88b1d5..1c503d07e0fb5fe2819b4d3cb0c0bd4890c7456e 100644 --- a/arch/powerpc/mm/slb_low.S +++ b/arch/powerpc/mm/slb_low.S @@ -23,6 +23,48 @@ #include #include +/* + * This macro generates asm code to compute the VSID scramble + * function. Used in slb_allocate() and do_stab_bolted. The function + * computed is: (protovsid*VSID_MULTIPLIER) % VSID_MODULUS + * + * rt = register containing the proto-VSID and into which the + * VSID will be stored + * rx = scratch register (clobbered) + * rf = flags + * + * - rt and rx must be different registers + * - The answer will end up in the low VSID_BITS bits of rt. The higher + * bits may contain other garbage, so you may need to mask the + * result. + */ +#define ASM_VSID_SCRAMBLE(rt, rx, rf, size) \ + lis rx,VSID_MULTIPLIER_##size@h; \ + ori rx,rx,VSID_MULTIPLIER_##size@l; \ + mulld rt,rt,rx; /* rt = rt * MULTIPLIER */ \ +/* \ + * powermac get slb fault before feature fixup, so make 65 bit part \ + * the default part of feature fixup \ + */ \ +BEGIN_MMU_FTR_SECTION \ + srdi rx,rt,VSID_BITS_65_##size; \ + clrldi rt,rt,(64-VSID_BITS_65_##size); \ + add rt,rt,rx; \ + addi rx,rt,1; \ + srdi rx,rx,VSID_BITS_65_##size; \ + add rt,rt,rx; \ + rldimi rf,rt,SLB_VSID_SHIFT_##size,(64 - (SLB_VSID_SHIFT_##size + VSID_BITS_65_##size)); \ +MMU_FTR_SECTION_ELSE \ + srdi rx,rt,VSID_BITS_##size; \ + clrldi rt,rt,(64-VSID_BITS_##size); \ + add rt,rt,rx; /* add high and low bits */ \ + addi rx,rt,1; \ + srdi rx,rx,VSID_BITS_##size; /* extract 2^VSID_BITS bit */ \ + add rt,rt,rx; \ + rldimi rf,rt,SLB_VSID_SHIFT_##size,(64 - (SLB_VSID_SHIFT_##size + VSID_BITS_##size)); \ +ALT_MMU_FTR_SECTION_END_IFCLR(MMU_FTR_68_BIT_VA) + + /* void slb_allocate_realmode(unsigned long ea); * * Create an SLB entry for the given EA (user or kernel). @@ -179,13 +221,7 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) */ .Lslb_finish_load: rldimi r10,r9,ESID_BITS,0 - ASM_VSID_SCRAMBLE(r10,r9,256M) - /* - * bits above VSID_BITS_256M need to be ignored from r10 - * also combine VSID and flags - */ - rldimi r11,r10,SLB_VSID_SHIFT,(64 - (SLB_VSID_SHIFT + VSID_BITS_256M)) - + ASM_VSID_SCRAMBLE(r10,r9,r11,256M) /* r3 = EA, r11 = VSID data */ /* * Find a slot, round robin. Previously we tried to find a @@ -249,12 +285,12 @@ slb_compare_rr_to_size: .Lslb_finish_load_1T: srdi r10,r10,(SID_SHIFT_1T - SID_SHIFT) /* get 1T ESID */ rldimi r10,r9,ESID_BITS_1T,0 - ASM_VSID_SCRAMBLE(r10,r9,1T) + ASM_VSID_SCRAMBLE(r10,r9,r11,1T) /* * bits above VSID_BITS_1T need to be ignored from r10 * also combine VSID and flags */ - rldimi r11,r10,SLB_VSID_SHIFT_1T,(64 - (SLB_VSID_SHIFT_1T + VSID_BITS_1T)) + li r10,MMU_SEGSIZE_1T rldimi r11,r10,SLB_VSID_SSIZE_SHIFT,0 /* insert segment size */