diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2012-07-24 23:01:20 +0400 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2012-07-24 23:01:20 +0400 |
commit | 5fecc9d8f59e765c2a48379dd7c6f5cf88c7d75a (patch) | |
tree | d1fc25d9650d3ac24591bba6f5e2e7a1afc54796 /arch/powerpc/kvm/book3s_64_mmu_hv.c | |
parent | 3c4cfadef6a1665d9cd02a543782d03d3e6740c6 (diff) | |
parent | 1a577b72475d161b6677c05abe57301362023bb2 (diff) | |
download | linux-5fecc9d8f59e765c2a48379dd7c6f5cf88c7d75a.tar.xz |
Merge tag 'kvm-3.6-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM updates from Avi Kivity:
"Highlights include
- full big real mode emulation on pre-Westmere Intel hosts (can be
disabled with emulate_invalid_guest_state=0)
- relatively small ppc and s390 updates
- PCID/INVPCID support in guests
- EOI avoidance; 3.6 guests should perform better on 3.6 hosts on
interrupt intensive workloads)
- Lockless write faults during live migration
- EPT accessed/dirty bits support for new Intel processors"
Fix up conflicts in:
- Documentation/virtual/kvm/api.txt:
Stupid subchapter numbering, added next to each other.
- arch/powerpc/kvm/booke_interrupts.S:
PPC asm changes clashing with the KVM fixes
- arch/s390/include/asm/sigp.h, arch/s390/kvm/sigp.c:
Duplicated commits through the kvm tree and the s390 tree, with
subsequent edits in the KVM tree.
* tag 'kvm-3.6-1' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (93 commits)
KVM: fix race with level interrupts
x86, hyper: fix build with !CONFIG_KVM_GUEST
Revert "apic: fix kvm build on UP without IOAPIC"
KVM guest: switch to apic_set_eoi_write, apic_write
apic: add apic_set_eoi_write for PV use
KVM: VMX: Implement PCID/INVPCID for guests with EPT
KVM: Add x86_hyper_kvm to complete detect_hypervisor_platform check
KVM: PPC: Critical interrupt emulation support
KVM: PPC: e500mc: Fix tlbilx emulation for 64-bit guests
KVM: PPC64: booke: Set interrupt computation mode for 64-bit host
KVM: PPC: bookehv: Add ESR flag to Data Storage Interrupt
KVM: PPC: bookehv64: Add support for std/ld emulation.
booke: Added crit/mc exception handler for e500v2
booke/bookehv: Add host crit-watchdog exception support
KVM: MMU: document mmu-lock and fast page fault
KVM: MMU: fix kvm_mmu_pagetable_walk tracepoint
KVM: MMU: trace fast page fault
KVM: MMU: fast path of handling guest page fault
KVM: MMU: introduce SPTE_MMU_WRITEABLE bit
KVM: MMU: fold tlb flush judgement into mmu_spte_update
...
Diffstat (limited to 'arch/powerpc/kvm/book3s_64_mmu_hv.c')
-rw-r--r-- | arch/powerpc/kvm/book3s_64_mmu_hv.c | 123 |
1 files changed, 95 insertions, 28 deletions
diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index 80a577517584..d03eb6f7b058 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -37,56 +37,121 @@ /* POWER7 has 10-bit LPIDs, PPC970 has 6-bit LPIDs */ #define MAX_LPID_970 63 -long kvmppc_alloc_hpt(struct kvm *kvm) +/* Power architecture requires HPT is at least 256kB */ +#define PPC_MIN_HPT_ORDER 18 + +long kvmppc_alloc_hpt(struct kvm *kvm, u32 *htab_orderp) { unsigned long hpt; - long lpid; struct revmap_entry *rev; struct kvmppc_linear_info *li; + long order = kvm_hpt_order; - /* Allocate guest's hashed page table */ - li = kvm_alloc_hpt(); - if (li) { - /* using preallocated memory */ - hpt = (ulong)li->base_virt; - kvm->arch.hpt_li = li; - } else { - /* using dynamic memory */ + if (htab_orderp) { + order = *htab_orderp; + if (order < PPC_MIN_HPT_ORDER) + order = PPC_MIN_HPT_ORDER; + } + + /* + * If the user wants a different size from default, + * try first to allocate it from the kernel page allocator. + */ + hpt = 0; + if (order != kvm_hpt_order) { hpt = __get_free_pages(GFP_KERNEL|__GFP_ZERO|__GFP_REPEAT| - __GFP_NOWARN, HPT_ORDER - PAGE_SHIFT); + __GFP_NOWARN, order - PAGE_SHIFT); + if (!hpt) + --order; } + /* Next try to allocate from the preallocated pool */ if (!hpt) { - pr_err("kvm_alloc_hpt: Couldn't alloc HPT\n"); - return -ENOMEM; + li = kvm_alloc_hpt(); + if (li) { + hpt = (ulong)li->base_virt; + kvm->arch.hpt_li = li; + order = kvm_hpt_order; + } } + + /* Lastly try successively smaller sizes from the page allocator */ + while (!hpt && order > PPC_MIN_HPT_ORDER) { + hpt = __get_free_pages(GFP_KERNEL|__GFP_ZERO|__GFP_REPEAT| + __GFP_NOWARN, order - PAGE_SHIFT); + if (!hpt) + --order; + } + + if (!hpt) + return -ENOMEM; + kvm->arch.hpt_virt = hpt; + kvm->arch.hpt_order = order; + /* HPTEs are 2**4 bytes long */ + kvm->arch.hpt_npte = 1ul << (order - 4); + /* 128 (2**7) bytes in each HPTEG */ + kvm->arch.hpt_mask = (1ul << (order - 7)) - 1; /* Allocate reverse map array */ - rev = vmalloc(sizeof(struct revmap_entry) * HPT_NPTE); + rev = vmalloc(sizeof(struct revmap_entry) * kvm->arch.hpt_npte); if (!rev) { pr_err("kvmppc_alloc_hpt: Couldn't alloc reverse map array\n"); goto out_freehpt; } kvm->arch.revmap = rev; + kvm->arch.sdr1 = __pa(hpt) | (order - 18); - lpid = kvmppc_alloc_lpid(); - if (lpid < 0) - goto out_freeboth; + pr_info("KVM guest htab at %lx (order %ld), LPID %x\n", + hpt, order, kvm->arch.lpid); - kvm->arch.sdr1 = __pa(hpt) | (HPT_ORDER - 18); - kvm->arch.lpid = lpid; - - pr_info("KVM guest htab at %lx, LPID %lx\n", hpt, lpid); + if (htab_orderp) + *htab_orderp = order; return 0; - out_freeboth: - vfree(rev); out_freehpt: - free_pages(hpt, HPT_ORDER - PAGE_SHIFT); + if (kvm->arch.hpt_li) + kvm_release_hpt(kvm->arch.hpt_li); + else + free_pages(hpt, order - PAGE_SHIFT); return -ENOMEM; } +long kvmppc_alloc_reset_hpt(struct kvm *kvm, u32 *htab_orderp) +{ + long err = -EBUSY; + long order; + + mutex_lock(&kvm->lock); + if (kvm->arch.rma_setup_done) { + kvm->arch.rma_setup_done = 0; + /* order rma_setup_done vs. vcpus_running */ + smp_mb(); + if (atomic_read(&kvm->arch.vcpus_running)) { + kvm->arch.rma_setup_done = 1; + goto out; + } + } + if (kvm->arch.hpt_virt) { + order = kvm->arch.hpt_order; + /* Set the entire HPT to 0, i.e. invalid HPTEs */ + memset((void *)kvm->arch.hpt_virt, 0, 1ul << order); + /* + * Set the whole last_vcpu array to an invalid vcpu number. + * This ensures that each vcpu will flush its TLB on next entry. + */ + memset(kvm->arch.last_vcpu, 0xff, sizeof(kvm->arch.last_vcpu)); + *htab_orderp = order; + err = 0; + } else { + err = kvmppc_alloc_hpt(kvm, htab_orderp); + order = *htab_orderp; + } + out: + mutex_unlock(&kvm->lock); + return err; +} + void kvmppc_free_hpt(struct kvm *kvm) { kvmppc_free_lpid(kvm->arch.lpid); @@ -94,7 +159,8 @@ void kvmppc_free_hpt(struct kvm *kvm) if (kvm->arch.hpt_li) kvm_release_hpt(kvm->arch.hpt_li); else - free_pages(kvm->arch.hpt_virt, HPT_ORDER - PAGE_SHIFT); + free_pages(kvm->arch.hpt_virt, + kvm->arch.hpt_order - PAGE_SHIFT); } /* Bits in first HPTE dword for pagesize 4k, 64k or 16M */ @@ -119,6 +185,7 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot, unsigned long psize; unsigned long hp0, hp1; long ret; + struct kvm *kvm = vcpu->kvm; psize = 1ul << porder; npages = memslot->npages >> (porder - PAGE_SHIFT); @@ -127,8 +194,8 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot, if (npages > 1ul << (40 - porder)) npages = 1ul << (40 - porder); /* Can't use more than 1 HPTE per HPTEG */ - if (npages > HPT_NPTEG) - npages = HPT_NPTEG; + if (npages > kvm->arch.hpt_mask + 1) + npages = kvm->arch.hpt_mask + 1; hp0 = HPTE_V_1TB_SEG | (VRMA_VSID << (40 - 16)) | HPTE_V_BOLTED | hpte0_pgsize_encoding(psize); @@ -138,7 +205,7 @@ void kvmppc_map_vrma(struct kvm_vcpu *vcpu, struct kvm_memory_slot *memslot, for (i = 0; i < npages; ++i) { addr = i << porder; /* can't use hpt_hash since va > 64 bits */ - hash = (i ^ (VRMA_VSID ^ (VRMA_VSID << 25))) & HPT_HASH_MASK; + hash = (i ^ (VRMA_VSID ^ (VRMA_VSID << 25))) & kvm->arch.hpt_mask; /* * We assume that the hash table is empty and no * vcpus are using it at this stage. Since we create |