From 38753cbc4dca431d4354319c7481f6bd1a212baf Mon Sep 17 00:00:00 2001 From: Colton Lewis Date: Thu, 8 Aug 2024 17:42:43 +0000 Subject: [PATCH 1/4] KVM: arm64: Move data barrier to end of split walk This DSB guarantees page table updates have been made visible to the hardware table walker. Moving the DSB from stage2_split_walker() to after the walk is finished in kvm_pgtable_stage2_split() results in a roughly 70% reduction in Clear Dirty Log Time in dirty_log_perf_test (modified to use eager page splitting) when using huge pages. This gain holds steady through a range of vcpus used (tested 1-64) and memory used (tested 1-64GB). This is safe to do because nothing else is using the page tables while they are still being mapped and this is how other page table walkers already function. None of them have a data barrier in the walker itself because relative ordering of table PTEs to table contents comes from the release semantics of stage2_make_pte(). Signed-off-by: Colton Lewis Acked-by: Oliver Upton Link: https://lore.kernel.org/r/20240808174243.2836363-1-coltonlewis@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/pgtable.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 9e2bbee77491..9788af2ca8c0 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -1547,7 +1547,6 @@ static int stage2_split_walker(const struct kvm_pgtable_visit_ctx *ctx, */ new = kvm_init_table_pte(childp, mm_ops); stage2_make_pte(ctx, new); - dsb(ishst); return 0; } @@ -1559,8 +1558,11 @@ int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, .flags = KVM_PGTABLE_WALK_LEAF, .arg = mc, }; + int ret; - return kvm_pgtable_walk(pgt, addr, size, &walker); + ret = kvm_pgtable_walk(pgt, addr, size, &walker); + dsb(ishst); + return ret; } int __kvm_pgtable_stage2_init(struct kvm_pgtable *pgt, struct kvm_s2_mmu *mmu, From ae41d7dbaeb4f79134136cd65ad7015cf9ccf78a Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 26 Jul 2024 16:51:10 -0700 Subject: [PATCH 2/4] KVM: arm64: Release pfn, i.e. put page, if copying MTE tags hits ZONE_DEVICE Put the page reference acquired by gfn_to_pfn_prot() if kvm_vm_ioctl_mte_copy_tags() runs into ZONE_DEVICE memory. KVM's less- than-stellar heuristics for dealing with pfn-mapped memory means that KVM can get a page reference to ZONE_DEVICE memory. Fixes: f0376edb1ddc ("KVM: arm64: Add ioctl to fetch/store tags in a guest") Signed-off-by: Sean Christopherson Reviewed-by: Steven Price Reviewed-by: Catalin Marinas Link: https://lore.kernel.org/r/20240726235234.228822-2-seanjc@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/guest.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c index 11098eb7eb44..e1f0ff08836a 100644 --- a/arch/arm64/kvm/guest.c +++ b/arch/arm64/kvm/guest.c @@ -1059,6 +1059,7 @@ int kvm_vm_ioctl_mte_copy_tags(struct kvm *kvm, page = pfn_to_online_page(pfn); if (!page) { /* Reject ZONE_DEVICE memory */ + kvm_release_pfn_clean(pfn); ret = -EFAULT; goto out; } From e0b7de4fd18c47ebd47ec0dd1af6503d4071b943 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 26 Jul 2024 16:51:11 -0700 Subject: [PATCH 3/4] KVM: arm64: Disallow copying MTE to guest memory while KVM is dirty logging Disallow copying MTE tags to guest memory while KVM is dirty logging, as writing guest memory without marking the gfn as dirty in the memslot could result in userspace failing to migrate the updated page. Ideally (maybe?), KVM would simply mark the gfn as dirty, but there is no vCPU to work with, and presumably the only use case for copy MTE tags _to_ the guest is when restoring state on the target. Fixes: f0376edb1ddc ("KVM: arm64: Add ioctl to fetch/store tags in a guest") Signed-off-by: Sean Christopherson Reviewed-by: Steven Price Reviewed-by: Catalin Marinas Link: https://lore.kernel.org/r/20240726235234.228822-3-seanjc@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/guest.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c index e1f0ff08836a..962f985977c2 100644 --- a/arch/arm64/kvm/guest.c +++ b/arch/arm64/kvm/guest.c @@ -1045,6 +1045,11 @@ int kvm_vm_ioctl_mte_copy_tags(struct kvm *kvm, mutex_lock(&kvm->slots_lock); + if (write && atomic_read(&kvm->nr_memslots_dirty_logging)) { + ret = -EBUSY; + goto out; + } + while (length > 0) { kvm_pfn_t pfn = gfn_to_pfn_prot(kvm, gfn, write, NULL); void *maddr; From f26a525b77e040d584e967369af1e018d2d59112 Mon Sep 17 00:00:00 2001 From: Snehal Koukuntla Date: Mon, 9 Sep 2024 18:01:54 +0000 Subject: [PATCH 4/4] KVM: arm64: Add memory length checks and remove inline in do_ffa_mem_xfer When we share memory through FF-A and the description of the buffers exceeds the size of the mapped buffer, the fragmentation API is used. The fragmentation API allows specifying chunks of descriptors in subsequent FF-A fragment calls and no upper limit has been established for this. The entire memory region transferred is identified by a handle which can be used to reclaim the transferred memory. To be able to reclaim the memory, the description of the buffers has to fit in the ffa_desc_buf. Add a bounds check on the FF-A sharing path to prevent the memory reclaim from failing. Also do_ffa_mem_xfer() does not need __always_inline, except for the BUILD_BUG_ON() aspect, which gets moved to a macro. [maz: fixed the BUILD_BUG_ON() breakage with LLVM, thanks to Wei-Lin Chang for the timely report] Fixes: 634d90cf0ac65 ("KVM: arm64: Handle FFA_MEM_LEND calls from the host") Cc: stable@vger.kernel.org Reviewed-by: Sebastian Ene Signed-off-by: Snehal Koukuntla Reviewed-by: Oliver Upton Link: https://lore.kernel.org/r/20240909180154.3267939-1-snehalreddy@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/ffa.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/arch/arm64/kvm/hyp/nvhe/ffa.c b/arch/arm64/kvm/hyp/nvhe/ffa.c index e715c157c2c4..e433dfab882a 100644 --- a/arch/arm64/kvm/hyp/nvhe/ffa.c +++ b/arch/arm64/kvm/hyp/nvhe/ffa.c @@ -426,9 +426,9 @@ static void do_ffa_mem_frag_tx(struct arm_smccc_res *res, return; } -static __always_inline void do_ffa_mem_xfer(const u64 func_id, - struct arm_smccc_res *res, - struct kvm_cpu_context *ctxt) +static void __do_ffa_mem_xfer(const u64 func_id, + struct arm_smccc_res *res, + struct kvm_cpu_context *ctxt) { DECLARE_REG(u32, len, ctxt, 1); DECLARE_REG(u32, fraglen, ctxt, 2); @@ -440,9 +440,6 @@ static __always_inline void do_ffa_mem_xfer(const u64 func_id, u32 offset, nr_ranges; int ret = 0; - BUILD_BUG_ON(func_id != FFA_FN64_MEM_SHARE && - func_id != FFA_FN64_MEM_LEND); - if (addr_mbz || npages_mbz || fraglen > len || fraglen > KVM_FFA_MBOX_NR_PAGES * PAGE_SIZE) { ret = FFA_RET_INVALID_PARAMETERS; @@ -461,6 +458,11 @@ static __always_inline void do_ffa_mem_xfer(const u64 func_id, goto out_unlock; } + if (len > ffa_desc_buf.len) { + ret = FFA_RET_NO_MEMORY; + goto out_unlock; + } + buf = hyp_buffers.tx; memcpy(buf, host_buffers.tx, fraglen); @@ -512,6 +514,13 @@ static __always_inline void do_ffa_mem_xfer(const u64 func_id, goto out_unlock; } +#define do_ffa_mem_xfer(fid, res, ctxt) \ + do { \ + BUILD_BUG_ON((fid) != FFA_FN64_MEM_SHARE && \ + (fid) != FFA_FN64_MEM_LEND); \ + __do_ffa_mem_xfer((fid), (res), (ctxt)); \ + } while (0); + static void do_ffa_mem_reclaim(struct arm_smccc_res *res, struct kvm_cpu_context *ctxt) {