Stefano Garzarella
2022-Jul-27 12:22 UTC
[RFC PATCH v2 3/9] vmci/vsock: use 'target' in notify_poll_in, callback
@Jorgen can you take a look at this series, especially this patch? Maybe we need to update the comments in the else branch, something like s/there is nothing/there is not enough data Thanks, Stefano On Mon, Jul 25, 2022 at 08:01:01AM +0000, Arseniy Krasnov wrote:>This callback controls setting of POLLIN,POLLRDNORM output bits of poll() >syscall,but in some cases,it is incorrectly to set it, when socket has >at least 1 bytes of available data. Use 'target' which is already exists >and equal to sk_rcvlowat in this case. > >Signed-off-by: Arseniy Krasnov <AVKrasnov at sberdevices.ru> >--- > net/vmw_vsock/vmci_transport_notify.c | 2 +- > net/vmw_vsock/vmci_transport_notify_qstate.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > >diff --git a/net/vmw_vsock/vmci_transport_notify.c b/net/vmw_vsock/vmci_transport_notify.c >index d69fc4b595ad..1684b85b0660 100644 >--- a/net/vmw_vsock/vmci_transport_notify.c >+++ b/net/vmw_vsock/vmci_transport_notify.c >@@ -340,7 +340,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, > { > struct vsock_sock *vsk = vsock_sk(sk); > >- if (vsock_stream_has_data(vsk)) { >+ if (vsock_stream_has_data(vsk) >= target) { > *data_ready_now = true; > } else { > /* We can't read right now because there is nothing in the >diff --git a/net/vmw_vsock/vmci_transport_notify_qstate.c b/net/vmw_vsock/vmci_transport_notify_qstate.c >index 0f36d7c45db3..a40407872b53 100644 >--- a/net/vmw_vsock/vmci_transport_notify_qstate.c >+++ b/net/vmw_vsock/vmci_transport_notify_qstate.c >@@ -161,7 +161,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, > { > struct vsock_sock *vsk = vsock_sk(sk); > >- if (vsock_stream_has_data(vsk)) { >+ if (vsock_stream_has_data(vsk) >= target) { > *data_ready_now = true; > } else { > /* We can't read right now because there is nothing in the >-- >2.25.1
Stefano Garzarella
2022-Jul-27 12:29 UTC
[RFC PATCH v2 3/9] vmci/vsock: use 'target' in notify_poll_in, callback
On Wed, Jul 27, 2022 at 2:22 PM Stefano Garzarella <sgarzare at redhat.com> wrote:> > @Jorgen can you take a look at this series, especially this patch?Jorgen's email bounced back, so I'm CCing VMCI maintainers. Bryan, Rajesh, Vishnu, can you take a look? Thanks, Stefano> > Maybe we need to update the comments in the else branch, something like > s/there is nothing/there is not enough data > > Thanks, > Stefano > > On Mon, Jul 25, 2022 at 08:01:01AM +0000, Arseniy Krasnov wrote: > >This callback controls setting of POLLIN,POLLRDNORM output bits of poll() > >syscall,but in some cases,it is incorrectly to set it, when socket has > >at least 1 bytes of available data. Use 'target' which is already exists > >and equal to sk_rcvlowat in this case. > > > >Signed-off-by: Arseniy Krasnov <AVKrasnov at sberdevices.ru> > >--- > > net/vmw_vsock/vmci_transport_notify.c | 2 +- > > net/vmw_vsock/vmci_transport_notify_qstate.c | 2 +- > > 2 files changed, 2 insertions(+), 2 deletions(-) > > > >diff --git a/net/vmw_vsock/vmci_transport_notify.c b/net/vmw_vsock/vmci_transport_notify.c > >index d69fc4b595ad..1684b85b0660 100644 > >--- a/net/vmw_vsock/vmci_transport_notify.c > >+++ b/net/vmw_vsock/vmci_transport_notify.c > >@@ -340,7 +340,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, > > { > > struct vsock_sock *vsk = vsock_sk(sk); > > > >- if (vsock_stream_has_data(vsk)) { > >+ if (vsock_stream_has_data(vsk) >= target) { > > *data_ready_now = true; > > } else { > > /* We can't read right now because there is nothing in the > >diff --git a/net/vmw_vsock/vmci_transport_notify_qstate.c b/net/vmw_vsock/vmci_transport_notify_qstate.c > >index 0f36d7c45db3..a40407872b53 100644 > >--- a/net/vmw_vsock/vmci_transport_notify_qstate.c > >+++ b/net/vmw_vsock/vmci_transport_notify_qstate.c > >@@ -161,7 +161,7 @@ vmci_transport_notify_pkt_poll_in(struct sock *sk, > > { > > struct vsock_sock *vsk = vsock_sk(sk); > > > >- if (vsock_stream_has_data(vsk)) { > >+ if (vsock_stream_has_data(vsk) >= target) { > > *data_ready_now = true; > > } else { > > /* We can't read right now because there is nothing in the > >-- > >2.25.1