From: Aneesh Kumar K.V Date: Tue, 1 Dec 2015 03:36:43 +0000 (+0530) Subject: powerpc/mm: Convert 4k hash insert to C X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=91f1da99792a1d133df94c4753510305353064a1;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git powerpc/mm: Convert 4k hash insert to C Acked-by: Scott Wood Signed-off-by: Aneesh Kumar K.V Signed-off-by: Michael Ellerman --- diff --git a/arch/powerpc/mm/Makefile b/arch/powerpc/mm/Makefile index 3eb73a38220d..f80ad1a76cc8 100644 --- a/arch/powerpc/mm/Makefile +++ b/arch/powerpc/mm/Makefile @@ -18,6 +18,9 @@ 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 \ mmu_context_hash$(CONFIG_WORD_SIZE).o +ifeq ($(CONFIG_PPC_STD_MMU_64),y) +obj-$(CONFIG_PPC_64K_PAGES) += hash64_64k.o +endif obj-$(CONFIG_PPC_ICSWX) += icswx.o obj-$(CONFIG_PPC_ICSWX_PID) += icswx_pid.o obj-$(CONFIG_40x) += 40x_mmu.o diff --git a/arch/powerpc/mm/hash64_64k.c b/arch/powerpc/mm/hash64_64k.c new file mode 100644 index 000000000000..9ffeae2cbb57 --- /dev/null +++ b/arch/powerpc/mm/hash64_64k.c @@ -0,0 +1,202 @@ +/* + * Copyright IBM Corporation, 2015 + * Author Aneesh Kumar K.V + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU Lesser General Public License + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it would be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * + */ + +#include +#include +#include + +int __hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid, + pte_t *ptep, unsigned long trap, unsigned long flags, + int ssize, int subpg_prot) +{ + real_pte_t rpte; + unsigned long *hidxp; + unsigned long hpte_group; + unsigned int subpg_index; + unsigned long rflags, pa, hidx; + unsigned long old_pte, new_pte, subpg_pte; + unsigned long vpn, hash, slot; + unsigned long shift = mmu_psize_defs[MMU_PAGE_4K].shift; + + /* + * atomically mark the linux large page PTE busy and dirty + */ + do { + pte_t pte = READ_ONCE(*ptep); + + old_pte = pte_val(pte); + /* If PTE busy, retry the access */ + if (unlikely(old_pte & _PAGE_BUSY)) + return 0; + /* If PTE permissions don't match, take page fault */ + if (unlikely(access & ~old_pte)) + return 1; + /* + * Try to lock the PTE, add ACCESSED and DIRTY if it was + * a write access. Since this is 4K insert of 64K page size + * also add _PAGE_COMBO + */ + new_pte = old_pte | _PAGE_BUSY | _PAGE_ACCESSED | _PAGE_COMBO; + if (access & _PAGE_RW) + new_pte |= _PAGE_DIRTY; + } while (old_pte != __cmpxchg_u64((unsigned long *)ptep, + old_pte, new_pte)); + /* + * Handle the subpage protection bits + */ + subpg_pte = new_pte & ~subpg_prot; + /* + * PP bits. _PAGE_USER is already PP bit 0x2, so we only + * need to add in 0x1 if it's a read-only user page + */ + rflags = subpg_pte & _PAGE_USER; + if ((subpg_pte & _PAGE_USER) && !((subpg_pte & _PAGE_RW) && + (subpg_pte & _PAGE_DIRTY))) + rflags |= 0x1; + /* + * _PAGE_EXEC -> HW_NO_EXEC since it's inverted + */ + rflags |= ((subpg_pte & _PAGE_EXEC) ? 0 : HPTE_R_N); + /* + * Always add C and Memory coherence bit + */ + rflags |= HPTE_R_C | HPTE_R_M; + /* + * Add in WIMG bits + */ + rflags |= (subpg_pte & (_PAGE_WRITETHRU | _PAGE_NO_CACHE | + _PAGE_COHERENT | _PAGE_GUARDED)); + + if (!cpu_has_feature(CPU_FTR_NOEXECUTE) && + !cpu_has_feature(CPU_FTR_COHERENT_ICACHE)) { + + /* + * No CPU has hugepages but lacks no execute, so we + * don't need to worry about that case + */ + rflags = hash_page_do_lazy_icache(rflags, __pte(old_pte), trap); + } + + subpg_index = (ea & (PAGE_SIZE - 1)) >> shift; + vpn = hpt_vpn(ea, vsid, ssize); + rpte = __real_pte(__pte(old_pte), ptep); + /* + *None of the sub 4k page is hashed + */ + if (!(old_pte & _PAGE_HASHPTE)) + goto htab_insert_hpte; + /* + * Check if the pte was already inserted into the hash table + * as a 64k HW page, and invalidate the 64k HPTE if so. + */ + if (!(old_pte & _PAGE_COMBO)) { + flush_hash_page(vpn, rpte, MMU_PAGE_64K, ssize, flags); + old_pte &= ~_PAGE_HPTE_SUB; + goto htab_insert_hpte; + } + /* + * Check for sub page valid and update + */ + if (__rpte_sub_valid(rpte, subpg_index)) { + int ret; + + hash = hpt_hash(vpn, shift, ssize); + hidx = __rpte_to_hidx(rpte, subpg_index); + if (hidx & _PTEIDX_SECONDARY) + hash = ~hash; + slot = (hash & htab_hash_mask) * HPTES_PER_GROUP; + slot += hidx & _PTEIDX_GROUP_IX; + + ret = ppc_md.hpte_updatepp(slot, rflags, vpn, + MMU_PAGE_4K, MMU_PAGE_4K, + ssize, flags); + /* + *if we failed because typically the HPTE wasn't really here + * we try an insertion. + */ + if (ret == -1) + goto htab_insert_hpte; + + *ptep = __pte(new_pte & ~_PAGE_BUSY); + return 0; + } + +htab_insert_hpte: + /* + * handle _PAGE_4K_PFN case + */ + if (old_pte & _PAGE_4K_PFN) { + /* + * All the sub 4k page have the same + * physical address. + */ + pa = pte_pfn(__pte(old_pte)) << HW_PAGE_SHIFT; + } else { + pa = pte_pfn(__pte(old_pte)) << PAGE_SHIFT; + pa += (subpg_index << shift); + } + hash = hpt_hash(vpn, shift, ssize); +repeat: + hpte_group = ((hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL; + + /* Insert into the hash table, primary slot */ + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, 0, + MMU_PAGE_4K, MMU_PAGE_4K, ssize); + /* + * Primary is full, try the secondary + */ + if (unlikely(slot == -1)) { + hpte_group = ((~hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL; + slot = ppc_md.hpte_insert(hpte_group, vpn, pa, + rflags, HPTE_V_SECONDARY, + MMU_PAGE_4K, MMU_PAGE_4K, ssize); + if (slot == -1) { + if (mftb() & 0x1) + hpte_group = ((hash & htab_hash_mask) * + HPTES_PER_GROUP) & ~0x7UL; + ppc_md.hpte_remove(hpte_group); + /* + * FIXME!! Should be try the group from which we removed ? + */ + goto repeat; + } + } + /* + * Hypervisor failure. Restore old pmd and return -1 + * similar to __hash_page_* + */ + if (unlikely(slot == -2)) { + *ptep = __pte(old_pte); + hash_failure_debug(ea, access, vsid, trap, ssize, + MMU_PAGE_4K, MMU_PAGE_4K, old_pte); + return -1; + } + /* + * Insert slot number & secondary bit in PTE second half, + * clear _PAGE_BUSY and set appropriate HPTE slot bit + * Since we have _PAGE_BUSY set on ptep, we can be sure + * nobody is undating hidx. + */ + hidxp = (unsigned long *)(ptep + PTRS_PER_PTE); + /* __real_pte use pte_val() any idea why ? FIXME!! */ + rpte.hidx &= ~(0xfUL << (subpg_index << 2)); + *hidxp = rpte.hidx | (slot << (subpg_index << 2)); + new_pte |= (_PAGE_HPTE_SUB0 >> subpg_index); + /* + * check __real_pte for details on matching smp_rmb() + */ + smp_wmb(); + *ptep = __pte(new_pte & ~_PAGE_BUSY); + return 0; +} diff --git a/arch/powerpc/mm/hash_low_64.S b/arch/powerpc/mm/hash_low_64.S index 3b49e3295901..6b4d4c1d0628 100644 --- a/arch/powerpc/mm/hash_low_64.S +++ b/arch/powerpc/mm/hash_low_64.S @@ -328,381 +328,8 @@ htab_pte_insert_failure: li r3,-1 b htab_bail - #else /* CONFIG_PPC_64K_PAGES */ - -/***************************************************************************** - * * - * 64K SW & 4K or 64K HW in a 4K segment pages implementation * - * * - *****************************************************************************/ - -/* _hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid, - * pte_t *ptep, unsigned long trap, unsigned local flags, - * int ssize, int subpg_prot) - */ - -/* - * For now, we do NOT implement Admixed pages - */ -_GLOBAL(__hash_page_4K) - mflr r0 - std r0,16(r1) - stdu r1,-STACKFRAMESIZE(r1) - /* Save all params that we need after a function call */ - std r6,STK_PARAM(R6)(r1) - std r8,STK_PARAM(R8)(r1) - std r9,STK_PARAM(R9)(r1) - - /* Save non-volatile registers. - * r31 will hold "old PTE" - * r30 is "new PTE" - * r29 is vpn - * r28 is a hash value - * r27 is hashtab mask (maybe dynamic patched instead ?) - * r26 is the hidx mask - * r25 is the index in combo page - */ - std r25,STK_REG(R25)(r1) - std r26,STK_REG(R26)(r1) - std r27,STK_REG(R27)(r1) - std r28,STK_REG(R28)(r1) - std r29,STK_REG(R29)(r1) - std r30,STK_REG(R30)(r1) - std r31,STK_REG(R31)(r1) - - /* Step 1: - * - * Check permissions, atomically mark the linux PTE busy - * and hashed. - */ -1: - ldarx r31,0,r6 - /* Check access rights (access & ~(pte_val(*ptep))) */ - andc. r0,r4,r31 - bne- htab_wrong_access - /* Check if PTE is busy */ - andi. r0,r31,_PAGE_BUSY - /* If so, just bail out and refault if needed. Someone else - * is changing this PTE anyway and might hash it. - */ - bne- htab_bail_ok - /* Prepare new PTE value (turn access RW into DIRTY, then - * add BUSY and ACCESSED) - */ - rlwinm r30,r4,32-9+7,31-7,31-7 /* _PAGE_RW -> _PAGE_DIRTY */ - or r30,r30,r31 - ori r30,r30,_PAGE_BUSY | _PAGE_ACCESSED - oris r30,r30,_PAGE_COMBO@h - /* Write the linux PTE atomically (setting busy) */ - stdcx. r30,0,r6 - bne- 1b - isync - - /* Step 2: - * - * Insert/Update the HPTE in the hash table. At this point, - * r4 (access) is re-useable, we use it for the new HPTE flags - */ - - /* Load the hidx index */ - rldicl r25,r3,64-12,60 - -BEGIN_FTR_SECTION - cmpdi r9,0 /* check segment size */ - bne 3f -END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT) - /* Calc vpn and put it in r29 */ - sldi r29,r5,SID_SHIFT - VPN_SHIFT - /* - * clrldi r3,r3,64 - SID_SHIFT --> ea & 0xfffffff - * srdi r28,r3,VPN_SHIFT - */ - rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT - VPN_SHIFT) - or r29,r28,r29 - /* - * Calculate hash value for primary slot and store it in r28 - * r3 = va, r5 = vsid - * r0 = (va >> 12) & ((1ul << (28 - 12)) -1) - */ - rldicl r0,r3,64-12,48 - xor r28,r5,r0 /* hash */ - b 4f - -3: /* Calc vpn and put it in r29 */ - sldi r29,r5,SID_SHIFT_1T - VPN_SHIFT - /* - * clrldi r3,r3,64 - SID_SHIFT_1T --> ea & 0xffffffffff - * srdi r28,r3,VPN_SHIFT - */ - rldicl r28,r3,64 - VPN_SHIFT,64 - (SID_SHIFT_1T - VPN_SHIFT) - or r29,r28,r29 - - /* - * Calculate hash value for primary slot and - * store it in r28 for 1T segment - * r3 = va, r5 = vsid - */ - sldi r28,r5,25 /* vsid << 25 */ - /* r0 = (va >> 12) & ((1ul << (40 - 12)) -1) */ - rldicl r0,r3,64-12,36 - xor r28,r28,r5 /* vsid ^ ( vsid << 25) */ - xor r28,r28,r0 /* hash */ - - /* Convert linux PTE bits into HW equivalents */ -4: -#ifdef CONFIG_PPC_SUBPAGE_PROT - andc r10,r30,r10 - andi. r3,r10,0x1fe /* Get basic set of flags */ - rlwinm r0,r10,32-9+1,30,30 /* _PAGE_RW -> _PAGE_USER (r0) */ -#else - andi. r3,r30,0x1fe /* Get basic set of flags */ - rlwinm r0,r30,32-9+1,30,30 /* _PAGE_RW -> _PAGE_USER (r0) */ -#endif - xori r3,r3,HPTE_R_N /* _PAGE_EXEC -> NOEXEC */ - rlwinm r4,r30,32-7+1,30,30 /* _PAGE_DIRTY -> _PAGE_USER (r4) */ - and r0,r0,r4 /* _PAGE_RW & _PAGE_DIRTY ->r0 bit 30*/ - andc r0,r3,r0 /* r0 = pte & ~r0 */ - rlwimi r3,r0,32-1,31,31 /* Insert result into PP lsb */ - /* - * Always add "C" bit for perf. Memory coherence is always enabled - */ - ori r3,r3,HPTE_R_C | HPTE_R_M - - /* We eventually do the icache sync here (maybe inline that - * code rather than call a C function...) - */ -BEGIN_FTR_SECTION - mr r4,r30 - mr r5,r7 - bl hash_page_do_lazy_icache -END_FTR_SECTION(CPU_FTR_NOEXECUTE|CPU_FTR_COHERENT_ICACHE, CPU_FTR_NOEXECUTE) - - /* At this point, r3 contains new PP bits, save them in - * place of "access" in the param area (sic) - */ - std r3,STK_PARAM(R4)(r1) - - /* Get htab_hash_mask */ - ld r4,htab_hash_mask@got(2) - ld r27,0(r4) /* htab_hash_mask -> r27 */ - - /* Check if we may already be in the hashtable, in this case, we - * go to out-of-line code to try to modify the HPTE. We look for - * the bit at (1 >> (index + 32)) - */ - rldicl. r0,r31,64-12,48 - li r26,0 /* Default hidx */ - beq htab_insert_pte - - /* - * Check if the pte was already inserted into the hash table - * as a 64k HW page, and invalidate the 64k HPTE if so. - */ - andis. r0,r31,_PAGE_COMBO@h - beq htab_inval_old_hpte - - ld r6,STK_PARAM(R6)(r1) - ori r26,r6,PTE_PAGE_HIDX_OFFSET /* Load the hidx mask. */ - ld r26,0(r26) - addi r5,r25,36 /* Check actual HPTE_SUB bit, this */ - rldcr. r0,r31,r5,0 /* must match pgtable.h definition */ - bne htab_modify_pte - -htab_insert_pte: - /* real page number in r5, PTE RPN value + index */ - andis. r0,r31,_PAGE_4K_PFN@h - srdi r5,r31,PTE_RPN_SHIFT - bne- htab_special_pfn - sldi r5,r5,PAGE_FACTOR - add r5,r5,r25 -htab_special_pfn: - sldi r5,r5,HW_PAGE_SHIFT - - /* Calculate primary group hash */ - and r0,r28,r27 - rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - - /* Call ppc_md.hpte_insert */ - ld r6,STK_PARAM(R4)(r1) /* Retrieve new pp bits */ - mr r4,r29 /* Retrieve vpn */ - li r7,0 /* !bolted, !secondary */ - li r8,MMU_PAGE_4K /* page size */ - li r9,MMU_PAGE_4K /* actual page size */ - ld r10,STK_PARAM(R9)(r1) /* segment size */ -.globl htab_call_hpte_insert1 -htab_call_hpte_insert1: - bl . /* patched by htab_finish_init() */ - cmpdi 0,r3,0 - bge htab_pte_insert_ok /* Insertion successful */ - cmpdi 0,r3,-2 /* Critical failure */ - beq- htab_pte_insert_failure - - /* Now try secondary slot */ - - /* real page number in r5, PTE RPN value + index */ - andis. r0,r31,_PAGE_4K_PFN@h - srdi r5,r31,PTE_RPN_SHIFT - bne- 3f - sldi r5,r5,PAGE_FACTOR - add r5,r5,r25 -3: sldi r5,r5,HW_PAGE_SHIFT - - /* Calculate secondary group hash */ - andc r0,r27,r28 - rldicr r3,r0,3,63-3 /* r0 = (~hash & mask) << 3 */ - - /* Call ppc_md.hpte_insert */ - ld r6,STK_PARAM(R4)(r1) /* Retrieve new pp bits */ - mr r4,r29 /* Retrieve vpn */ - li r7,HPTE_V_SECONDARY /* !bolted, secondary */ - li r8,MMU_PAGE_4K /* page size */ - li r9,MMU_PAGE_4K /* actual page size */ - ld r10,STK_PARAM(R9)(r1) /* segment size */ -.globl htab_call_hpte_insert2 -htab_call_hpte_insert2: - bl . /* patched by htab_finish_init() */ - cmpdi 0,r3,0 - bge+ htab_pte_insert_ok /* Insertion successful */ - cmpdi 0,r3,-2 /* Critical failure */ - beq- htab_pte_insert_failure - - /* Both are full, we need to evict something */ - mftb r0 - /* Pick a random group based on TB */ - andi. r0,r0,1 - mr r5,r28 - bne 2f - not r5,r5 -2: and r0,r5,r27 - rldicr r3,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - /* Call ppc_md.hpte_remove */ -.globl htab_call_hpte_remove -htab_call_hpte_remove: - bl . /* patched by htab_finish_init() */ - - /* Try all again */ - b htab_insert_pte - - /* - * Call out to C code to invalidate an 64k HW HPTE that is - * useless now that the segment has been switched to 4k pages. - */ -htab_inval_old_hpte: - mr r3,r29 /* vpn */ - mr r4,r31 /* PTE.pte */ - li r5,0 /* PTE.hidx */ - li r6,MMU_PAGE_64K /* psize */ - ld r7,STK_PARAM(R9)(r1) /* ssize */ - ld r8,STK_PARAM(R8)(r1) /* flags */ - bl flush_hash_page - /* Clear out _PAGE_HPTE_SUB bits in the new linux PTE */ - lis r0,_PAGE_HPTE_SUB@h - ori r0,r0,_PAGE_HPTE_SUB@l - andc r30,r30,r0 - b htab_insert_pte - -htab_bail_ok: - li r3,0 - b htab_bail - -htab_pte_insert_ok: - /* Insert slot number & secondary bit in PTE second half, - * clear _PAGE_BUSY and set approriate HPTE slot bit - */ - ld r6,STK_PARAM(R6)(r1) - li r0,_PAGE_BUSY - andc r30,r30,r0 - /* HPTE SUB bit */ - li r0,1 - subfic r5,r25,27 /* Must match bit position in */ - sld r0,r0,r5 /* pgtable.h */ - or r30,r30,r0 - /* hindx */ - sldi r5,r25,2 - sld r3,r3,r5 - li r4,0xf - sld r4,r4,r5 - andc r26,r26,r4 - or r26,r26,r3 - ori r5,r6,PTE_PAGE_HIDX_OFFSET - std r26,0(r5) - lwsync - std r30,0(r6) - li r3, 0 -htab_bail: - ld r25,STK_REG(R25)(r1) - ld r26,STK_REG(R26)(r1) - ld r27,STK_REG(R27)(r1) - ld r28,STK_REG(R28)(r1) - ld r29,STK_REG(R29)(r1) - ld r30,STK_REG(R30)(r1) - ld r31,STK_REG(R31)(r1) - addi r1,r1,STACKFRAMESIZE - ld r0,16(r1) - mtlr r0 - blr - -htab_modify_pte: - /* Keep PP bits in r4 and slot idx from the PTE around in r3 */ - mr r4,r3 - sldi r5,r25,2 - srd r3,r26,r5 - - /* Secondary group ? if yes, get a inverted hash value */ - mr r5,r28 - andi. r0,r3,0x8 /* page secondary ? */ - beq 1f - not r5,r5 -1: andi. r3,r3,0x7 /* extract idx alone */ - - /* Calculate proper slot value for ppc_md.hpte_updatepp */ - and r0,r5,r27 - rldicr r0,r0,3,63-3 /* r0 = (hash & mask) << 3 */ - add r3,r0,r3 /* add slot idx */ - - /* Call ppc_md.hpte_updatepp */ - mr r5,r29 /* vpn */ - li r6,MMU_PAGE_4K /* base page size */ - li r7,MMU_PAGE_4K /* actual page size */ - ld r8,STK_PARAM(R9)(r1) /* segment size */ - ld r9,STK_PARAM(R8)(r1) /* get "flags" param */ -.globl htab_call_hpte_updatepp -htab_call_hpte_updatepp: - bl . /* patched by htab_finish_init() */ - - /* if we failed because typically the HPTE wasn't really here - * we try an insertion. - */ - cmpdi 0,r3,-1 - beq- htab_insert_pte - - /* Clear the BUSY bit and Write out the PTE */ - li r0,_PAGE_BUSY - andc r30,r30,r0 - ld r6,STK_PARAM(R6)(r1) - std r30,0(r6) - li r3,0 - b htab_bail - -htab_wrong_access: - /* Bail out clearing reservation */ - stdcx. r31,0,r6 - li r3,1 - b htab_bail - -htab_pte_insert_failure: - /* Bail out restoring old PTE */ - ld r6,STK_PARAM(R6)(r1) - std r31,0(r6) - li r3,-1 - b htab_bail - -#endif /* CONFIG_PPC_64K_PAGES */ - -#ifdef CONFIG_PPC_64K_PAGES - /***************************************************************************** * * * 64K SW & 64K HW in a 64K segment pages implementation * @@ -994,10 +621,3 @@ ht64_pte_insert_failure: #endif /* CONFIG_PPC_64K_PAGES */ - - -/***************************************************************************** - * * - * Huge pages implementation is in hugetlbpage.c * - * * - *****************************************************************************/ diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c index 7d4f254a2671..995809911f17 100644 --- a/arch/powerpc/mm/hash_utils_64.c +++ b/arch/powerpc/mm/hash_utils_64.c @@ -653,7 +653,7 @@ static void __init htab_finish_init(void) patch_branch(ht64_call_hpte_updatepp, ppc_function_entry(ppc_md.hpte_updatepp), BRANCH_SET_LINK); -#endif /* CONFIG_PPC_64K_PAGES */ +#else /* !CONFIG_PPC_64K_PAGES */ patch_branch(htab_call_hpte_insert1, ppc_function_entry(ppc_md.hpte_insert), @@ -667,6 +667,8 @@ static void __init htab_finish_init(void) patch_branch(htab_call_hpte_updatepp, ppc_function_entry(ppc_md.hpte_updatepp), BRANCH_SET_LINK); +#endif + } static void __init htab_initialize(void)