Hi Ackerley,
On Wed, 2 Apr 2025 at 23:47, Ackerley Tng <ackerleytng@xxxxxxxxxx> wrote:
>
> Fuad Tabba <tabba@xxxxxxxxxx> writes:
>
> > Not all use cases require guest_memfd() to be shared with the host when
> > first created. Add a new flag, GUEST_MEMFD_FLAG_INIT_SHARED, which when
> > set on KVM_CREATE_GUEST_MEMFD initializes the memory as shared with the
> > host, and therefore mappable by it. Otherwise, memory is private until
> > explicitly shared by the guest with the host.
> >
> > Signed-off-by: Fuad Tabba <tabba@xxxxxxxxxx>
> > ---
> > Documentation/virt/kvm/api.rst | 4 ++++
> > include/uapi/linux/kvm.h | 1 +
> > tools/testing/selftests/kvm/guest_memfd_test.c | 7 +++++--
> > virt/kvm/guest_memfd.c | 12 ++++++++++++
> > 4 files changed, 22 insertions(+), 2 deletions(-)
> >
> > diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst
> > index 2b52eb77e29c..a5496d7d323b 100644
> > --- a/Documentation/virt/kvm/api.rst
> > +++ b/Documentation/virt/kvm/api.rst
> > @@ -6386,6 +6386,10 @@ most one mapping per page, i.e. binding multiple memory regions to a single
> > guest_memfd range is not allowed (any number of memory regions can be bound to
> > a single guest_memfd file, but the bound ranges must not overlap).
> >
> > +If the capability KVM_CAP_GMEM_SHARED_MEM is supported, then the flags field
> > +supports GUEST_MEMFD_FLAG_INIT_SHARED, which initializes the memory as shared
> > +with the host, and thereby, mappable by it.
> > +
> > See KVM_SET_USER_MEMORY_REGION2 for additional details.
> >
> > 4.143 KVM_PRE_FAULT_MEMORY
> > diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
> > index 117937a895da..22d7e33bf09c 100644
> > --- a/include/uapi/linux/kvm.h
> > +++ b/include/uapi/linux/kvm.h
> > @@ -1566,6 +1566,7 @@ struct kvm_memory_attributes {
> > #define KVM_MEMORY_ATTRIBUTE_PRIVATE (1ULL << 3)
> >
> > #define KVM_CREATE_GUEST_MEMFD _IOWR(KVMIO, 0xd4, struct kvm_create_guest_memfd)
> > +#define GUEST_MEMFD_FLAG_INIT_SHARED (1UL << 0)
> >
> > struct kvm_create_guest_memfd {
> > __u64 size;
> > diff --git a/tools/testing/selftests/kvm/guest_memfd_test.c b/tools/testing/selftests/kvm/guest_memfd_test.c
> > index 38c501e49e0e..4a7fcd6aa372 100644
> > --- a/tools/testing/selftests/kvm/guest_memfd_test.c
> > +++ b/tools/testing/selftests/kvm/guest_memfd_test.c
> > @@ -159,7 +159,7 @@ static void test_invalid_punch_hole(int fd, size_t page_size, size_t total_size)
> > static void test_create_guest_memfd_invalid(struct kvm_vm *vm)
> > {
> > size_t page_size = getpagesize();
> > - uint64_t flag;
> > + uint64_t flag = BIT(0);
> > size_t size;
> > int fd;
> >
> > @@ -170,7 +170,10 @@ static void test_create_guest_memfd_invalid(struct kvm_vm *vm)
> > size);
> > }
> >
> > - for (flag = BIT(0); flag; flag <<= 1) {
> > + if (kvm_has_cap(KVM_CAP_GMEM_SHARED_MEM))
> > + flag = GUEST_MEMFD_FLAG_INIT_SHARED << 1;
> > +
> > + for (; flag; flag <<= 1) {
>
> This would end up shifting the GUEST_MEMFD_FLAG_INIT_SHARED flag for
> each loop.
Yes. That's my intention. This tests whether the flags are invalid.
Without GUEST_MEMFD_FLAG_INIT_SHARED, no flag is valid, so it starts
with bit zero and goes through all the flags.
If KVM_CAP_GMEM_SHARED_MEM is available, then we want to start from
bit 1 (i.e., skip bit 0, which is GUEST_MEMFD_FLAG_INIT_SHARED).
Cheers,
/fuad
> > fd = __vm_create_guest_memfd(vm, page_size, flag);
> > TEST_ASSERT(fd == -1 && errno == EINVAL,
> > "guest_memfd() with flag '0x%lx' should fail with EINVAL",
> > diff --git a/virt/kvm/guest_memfd.c b/virt/kvm/guest_memfd.c
> > index eec9d5e09f09..32e149478b04 100644
> > --- a/virt/kvm/guest_memfd.c
> > +++ b/virt/kvm/guest_memfd.c
> > @@ -1069,6 +1069,15 @@ static int __kvm_gmem_create(struct kvm *kvm, loff_t size, u64 flags)
> > goto err_gmem;
> > }
> >
> > + if (IS_ENABLED(CONFIG_KVM_GMEM_SHARED_MEM) &&
> > + (flags & GUEST_MEMFD_FLAG_INIT_SHARED)) {
> > + err = kvm_gmem_offset_range_set_shared(file_inode(file), 0, size >> PAGE_SHIFT);
> > + if (err) {
> > + fput(file);
> > + goto err_gmem;
> > + }
> > + }
> > +
> > kvm_get_kvm(kvm);
> > gmem->kvm = kvm;
> > xa_init(&gmem->bindings);
> > @@ -1090,6 +1099,9 @@ int kvm_gmem_create(struct kvm *kvm, struct kvm_create_guest_memfd *args)
> > u64 flags = args->flags;
> > u64 valid_flags = 0;
> >
> > + if (IS_ENABLED(CONFIG_KVM_GMEM_SHARED_MEM))
> > + valid_flags |= GUEST_MEMFD_FLAG_INIT_SHARED;
> > +
> > if (flags & ~valid_flags)
> > return -EINVAL;
