haraldwolff
/
qemu-patch-raspberry4
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

xen/9pfs: yield when there isn't enough room on the ring 

Instead of truncating replies, which is problematic, wait until the
client reads more data and frees bytes on the reply ring.

Do that by calling qemu_coroutine_yield(). The corresponding
qemu_coroutine_enter_if_inactive() is called from xen_9pfs_bh upon
receiving the next notification from the client.

We need to be careful to avoid races in case xen_9pfs_bh and the
coroutine are both active at the same time. In xen_9pfs_bh, wait until
either the critical section is over (ring->co == NULL) or until the
coroutine becomes inactive (qemu_coroutine_yield() was called) before
continuing. Then, simply wake up the coroutine if it is inactive.

Signed-off-by: Stefano Stabellini <stefano.stabellini@xilinx.com>
Reviewed-by: Christian Schoenebeck <qemu_oss@crudebyte.com>
Message-Id: <20200521192627.15259-2-sstabellini@kernel.org>
Signed-off-by: Greg Kurz <groug@kaod.org>
stable-6.0
Stefano Stabellini 2020-05-21 12:26:26 -07:00 committed by Greg Kurz
parent cf45183b71
commit a4c4d46272
1 changed files with 25 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
hw/9pfs/xen-9p-backend.c
View File

@ -37,6 +37,7 @@ typedef struct Xen9pfsRing {
struct iovec *sg;
QEMUBH *bh;
Coroutine *co;
/* local copies, so that we can read/write PDU data directly from
* the ring */
@ -198,16 +199,20 @@ static void xen_9pfs_init_in_iov_from_pdu(V9fsPDU *pdu,
g_free(ring->sg);
ring->sg = g_new0(struct iovec, 2);
xen_9pfs_in_sg(ring, ring->sg, &num, pdu->idx, size);
ring->co = qemu_coroutine_self();
/* make sure other threads see ring->co changes before continuing */
smp_wmb();
again:
xen_9pfs_in_sg(ring, ring->sg, &num, pdu->idx, size);
buf_size = iov_size(ring->sg, num);
if (buf_size < size) {
xen_pv_printf(&xen_9pfs->xendev, 0, "Xen 9pfs request type %d"
"needs %zu bytes, buffer has %zu\n", pdu->id, size,
buf_size);
xen_be_set_state(&xen_9pfs->xendev, XenbusStateClosing);
xen_9pfs_disconnect(&xen_9pfs->xendev);
qemu_coroutine_yield();
goto again;
}
ring->co = NULL;
/* make sure other threads see ring->co changes before continuing */
smp_wmb();
*piov = ring->sg;
*pniov = num;
@ -292,6 +297,20 @@ static int xen_9pfs_receive(Xen9pfsRing *ring)
static void xen_9pfs_bh(void *opaque)
{
Xen9pfsRing *ring = opaque;
bool wait;
again:
wait = ring->co != NULL && qemu_coroutine_entered(ring->co);
/* paired with the smb_wmb barriers in xen_9pfs_init_in_iov_from_pdu */
smp_rmb();
if (wait) {
cpu_relax();
goto again;
}
if (ring->co != NULL) {
qemu_coroutine_enter_if_inactive(ring->co);
}
xen_9pfs_receive(ring);
}