From: Fuad Tabba <tabba@google.com>
To: kvm@vger.kernel.org, linux-arm-msm@vger.kernel.org, linux-mm@kvack.org
Cc: pbonzini@redhat.com, chenhuacai@kernel.org, mpe@ellerman.id.au,
anup@brainfault.org, paul.walmsley@sifive.com,
palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com,
viro@zeniv.linux.org.uk, brauner@kernel.org,
willy@infradead.org, akpm@linux-foundation.org,
xiaoyao.li@intel.com, yilun.xu@intel.com,
chao.p.peng@linux.intel.com, jarkko@kernel.org,
amoorthy@google.com, dmatlack@google.com,
isaku.yamahata@intel.com, mic@digikod.net, vbabka@suse.cz,
vannapurve@google.com, ackerleytng@google.com,
mail@maciej.szmigiero.name, david@redhat.com,
michael.roth@amd.com, wei.w.wang@intel.com,
liam.merwick@oracle.com, isaku.yamahata@gmail.com,
kirill.shutemov@linux.intel.com, suzuki.poulose@arm.com,
steven.price@arm.com, quic_eberman@quicinc.com,
quic_mnalajal@quicinc.com, quic_tsoni@quicinc.com,
quic_svaddagi@quicinc.com, quic_cvanscha@quicinc.com,
quic_pderrin@quicinc.com, quic_pheragu@quicinc.com,
catalin.marinas@arm.com, james.morse@arm.com,
yuzenghui@huawei.com, oliver.upton@linux.dev, maz@kernel.org,
will@kernel.org, qperret@google.com, keirf@google.com,
roypat@amazon.co.uk, shuah@kernel.org, hch@infradead.org,
jgg@nvidia.com, rientjes@google.com, jhubbard@nvidia.com,
fvdl@google.com, hughd@google.com, jthoughton@google.com,
peterx@redhat.com, pankaj.gupta@amd.com, tabba@google.com
Subject: [PATCH v7 6/7] KVM: guest_memfd: Handle invalidation of shared memory
Date: Fri, 28 Mar 2025 15:31:32 +0000 [thread overview]
Message-ID: <20250328153133.3504118-7-tabba@google.com> (raw)
In-Reply-To: <20250328153133.3504118-1-tabba@google.com>
When guest_memfd backed memory is invalidated, e.g., on punching holes,
releasing, ensure that the sharing states are updated and that any
folios in a transient state are restored to an appropriate state.
Signed-off-by: Fuad Tabba <tabba@google.com>
---
virt/kvm/guest_memfd.c | 57 ++++++++++++++++++++++++++++++++++++++++++
1 file changed, 57 insertions(+)
diff --git a/virt/kvm/guest_memfd.c b/virt/kvm/guest_memfd.c
index ce19bd6c2031..eec9d5e09f09 100644
--- a/virt/kvm/guest_memfd.c
+++ b/virt/kvm/guest_memfd.c
@@ -118,6 +118,16 @@ static struct folio *kvm_gmem_get_folio(struct inode *inode, pgoff_t index)
return filemap_grab_folio(inode->i_mapping, index);
}
+#ifdef CONFIG_KVM_GMEM_SHARED_MEM
+static void kvm_gmem_offset_range_invalidate_shared(struct inode *inode,
+ pgoff_t start, pgoff_t end);
+#else
+static inline void kvm_gmem_offset_range_invalidate_shared(struct inode *inode,
+ pgoff_t start, pgoff_t end)
+{
+}
+#endif
+
static void kvm_gmem_invalidate_begin(struct kvm_gmem *gmem, pgoff_t start,
pgoff_t end)
{
@@ -127,6 +137,7 @@ static void kvm_gmem_invalidate_begin(struct kvm_gmem *gmem, pgoff_t start,
unsigned long index;
xa_for_each_range(&gmem->bindings, index, slot, start, end - 1) {
+ struct file *file = READ_ONCE(slot->gmem.file);
pgoff_t pgoff = slot->gmem.pgoff;
struct kvm_gfn_range gfn_range = {
@@ -146,6 +157,16 @@ static void kvm_gmem_invalidate_begin(struct kvm_gmem *gmem, pgoff_t start,
}
flush |= kvm_mmu_unmap_gfn_range(kvm, &gfn_range);
+
+ /*
+ * If this gets called after kvm_gmem_unbind() it means that all
+ * in-flight operations are gone, and the file has been closed.
+ */
+ if (file) {
+ kvm_gmem_offset_range_invalidate_shared(file_inode(file),
+ gfn_range.start,
+ gfn_range.end);
+ }
}
if (flush)
@@ -512,6 +533,42 @@ static int kvm_gmem_offset_clear_shared(struct inode *inode, pgoff_t index)
return r;
}
+/*
+ * Callback when invalidating memory that is potentially shared.
+ *
+ * Must be called with the offsets_lock write lock held.
+ */
+static void kvm_gmem_offset_range_invalidate_shared(struct inode *inode,
+ pgoff_t start, pgoff_t end)
+{
+ struct xarray *shared_offsets = &kvm_gmem_private(inode)->shared_offsets;
+ rwlock_t *offsets_lock = &kvm_gmem_private(inode)->offsets_lock;
+ pgoff_t i;
+
+ lockdep_assert_held_write(offsets_lock);
+
+ for (i = start; i < end; i++) {
+ /*
+ * If the folio is NONE_SHARED, it indicates that it's
+ * transitioning to private (GUEST_SHARED). Transition it to
+ * shared (ALL_SHARED) and remove the callback.
+ */
+ if (xa_to_value(xa_load(shared_offsets, i)) == KVM_GMEM_NONE_SHARED) {
+ struct folio *folio = filemap_lock_folio(inode->i_mapping, i);
+
+ if (!WARN_ON_ONCE(IS_ERR(folio))) {
+ if (folio_test_guestmem(folio))
+ kvm_gmem_restore_pending_folio(folio, inode);
+
+ folio_unlock(folio);
+ folio_put(folio);
+ }
+ }
+
+ xa_erase(shared_offsets, i);
+ }
+}
+
/*
* Marks the range [start, end) as not shared with the host. If the host doesn't
* have any references to a particular folio, then that folio is marked as
--
2.49.0.472.ge94155a9ec-goog
next prev parent reply other threads:[~2025-03-28 15:31 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-28 15:31 [PATCH v7 0/7] KVM: Restricted mapping of guest_memfd at the host and arm64 support Fuad Tabba
2025-03-28 15:31 ` [PATCH v7 1/7] KVM: guest_memfd: Make guest mem use guest mem inodes instead of anonymous inodes Fuad Tabba
2025-04-08 11:53 ` Shivank Garg
2025-03-28 15:31 ` [PATCH v7 2/7] KVM: guest_memfd: Introduce kvm_gmem_get_pfn_locked(), which retains the folio lock Fuad Tabba
2025-03-28 15:31 ` [PATCH v7 3/7] KVM: guest_memfd: Track folio sharing within a struct kvm_gmem_private Fuad Tabba
2025-04-02 22:25 ` Ackerley Tng
2025-04-02 23:56 ` Sean Christopherson
2025-04-03 8:58 ` Fuad Tabba
2025-04-03 14:51 ` Sean Christopherson
2025-04-04 6:43 ` Fuad Tabba
2025-03-28 15:31 ` [PATCH v7 4/7] KVM: guest_memfd: Folio sharing states and functions that manage their transition Fuad Tabba
2025-04-02 23:48 ` Ackerley Tng
2025-04-03 8:58 ` Fuad Tabba
2025-04-03 0:19 ` Sean Christopherson
2025-04-03 9:11 ` Fuad Tabba
2025-04-03 14:52 ` Sean Christopherson
2025-04-04 6:44 ` Fuad Tabba
2025-03-28 15:31 ` [PATCH v7 5/7] KVM: guest_memfd: Restore folio state after final folio_put() Fuad Tabba
2025-04-05 16:23 ` Matthew Wilcox
2025-03-28 15:31 ` Fuad Tabba [this message]
2025-03-28 15:31 ` [PATCH v7 7/7] KVM: guest_memfd: Add a guest_memfd() flag to initialize it as shared Fuad Tabba
2025-04-02 22:47 ` Ackerley Tng
2025-04-04 7:39 ` Fuad Tabba
2025-04-02 23:03 ` Ackerley Tng
2025-04-03 11:32 ` Fuad Tabba
2025-04-04 18:05 ` [PATCH v7 0/7] KVM: Restricted mapping of guest_memfd at the host and arm64 support Liam R. Howlett
2025-04-04 20:10 ` David Hildenbrand
2025-04-04 21:48 ` Sean Christopherson
2025-04-05 2:45 ` Liam R. Howlett
2025-04-07 6:28 ` Fuad Tabba
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250328153133.3504118-7-tabba@google.com \
--to=tabba@google.com \
--cc=ackerleytng@google.com \
--cc=akpm@linux-foundation.org \
--cc=amoorthy@google.com \
--cc=anup@brainfault.org \
--cc=aou@eecs.berkeley.edu \
--cc=brauner@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=chao.p.peng@linux.intel.com \
--cc=chenhuacai@kernel.org \
--cc=david@redhat.com \
--cc=dmatlack@google.com \
--cc=fvdl@google.com \
--cc=hch@infradead.org \
--cc=hughd@google.com \
--cc=isaku.yamahata@gmail.com \
--cc=isaku.yamahata@intel.com \
--cc=james.morse@arm.com \
--cc=jarkko@kernel.org \
--cc=jgg@nvidia.com \
--cc=jhubbard@nvidia.com \
--cc=jthoughton@google.com \
--cc=keirf@google.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=liam.merwick@oracle.com \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mail@maciej.szmigiero.name \
--cc=maz@kernel.org \
--cc=mic@digikod.net \
--cc=michael.roth@amd.com \
--cc=mpe@ellerman.id.au \
--cc=oliver.upton@linux.dev \
--cc=palmer@dabbelt.com \
--cc=pankaj.gupta@amd.com \
--cc=paul.walmsley@sifive.com \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=qperret@google.com \
--cc=quic_cvanscha@quicinc.com \
--cc=quic_eberman@quicinc.com \
--cc=quic_mnalajal@quicinc.com \
--cc=quic_pderrin@quicinc.com \
--cc=quic_pheragu@quicinc.com \
--cc=quic_svaddagi@quicinc.com \
--cc=quic_tsoni@quicinc.com \
--cc=rientjes@google.com \
--cc=roypat@amazon.co.uk \
--cc=seanjc@google.com \
--cc=shuah@kernel.org \
--cc=steven.price@arm.com \
--cc=suzuki.poulose@arm.com \
--cc=vannapurve@google.com \
--cc=vbabka@suse.cz \
--cc=viro@zeniv.linux.org.uk \
--cc=wei.w.wang@intel.com \
--cc=will@kernel.org \
--cc=willy@infradead.org \
--cc=xiaoyao.li@intel.com \
--cc=yilun.xu@intel.com \
--cc=yuzenghui@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox