search for: rsp_con

Hi, We are running into a situation where rsp_prod index in the shared ring is not getting updated for the netfront tx ring by the netback. We see that rsp_cons is the same value as rsp_prod, with req_prod 236 slots away(tx ring is full). From looking at the netfront driver code, it looks as if xennet_tx_buf_gc processing only happens if rsp_prod is more than rsp_cons. Our understanding is that netfront sets rsp_cons to tell the netback to start proce...

Hi, We are running into a situation where rsp_prod index in the shared ring is not getting updated for the netfront tx ring by the netback. We see that rsp_cons is the same value as rsp_prod, with req_prod 236 slots away(tx ring is full). From looking at the netfront driver code, it looks as if xennet_tx_buf_gc processing only happens if rsp_prod is more than rsp_cons. Our understanding is that netfront sets rsp_cons to tell the netback to start proce...

...and it looks like the skb is an id. Investigating further, the condition occurs if the gnttab_query_foreign_access returns non zero on a second or latter iteration through the for loop. If it return non zero, the the code takes the ''goto out'' which by passes fixing up np->tx.rsp_cons. Then the next time in network_tx_buf_gc we reuse np->tx.rsp_cons which is at the location of a previously completed skb and the skb gets an id and not a skb. Looking at the unstable tree, the goto has been removed and replaced with a break. However, it looks like if gnttab_query_foreign_ac...

upstream commit 7e96287ddc4f42081e18248b6167041c0908004c Author: Vivek Goyal <vgoyal@in.ibm.com> [PATCH] kdump: introduce "reset_devices" command line option Resetting the devices during driver initialization can be a costly operation in terms of time (especially scsi devices). This option can be used by drivers to know that user forcibly wants the devices to

...nd reading function to operate it ,but it came to oops:     struct xenstore_domain_interface {     char req[XENSTORE_RING_SIZE]; /* Requests to xenstore daemon. */     char rsp[XENSTORE_RING_SIZE]; /* Replies and async watch events. */     XENSTORE_RING_IDX req_cons, req_prod;     XENSTORE_RING_IDX rsp_cons, rsp_prod;     };     struct xenstore_domain_interface *xenstore;     xen_hvm_param_t param;     param.domid = DOMID_SELF;     param.index = HVM_PARAM_STORE_PFN;     if ( hypercall_hvm_op(HVMOP_get_param, &param) )         BUG();     xenstore = (void *) (unsigned long) (param.value << PA...

...xennet_get_rx_skb(np, cons + slots); + ref = xennet_get_rx_ref(np, cons + slots); + slots++; } - if (unlikely(frags > max)) { + if (unlikely(slots > max)) { if (net_ratelimit()) dev_warn(dev, "Too many frags\n"); err = -E2BIG; } if (unlikely(err)) - np->rx.rsp_cons = cons + frags; + np->rx.rsp_cons = cons + slots; return err; } -- 1.7.10.4

...g.len), + *volatile *events = NULL; + char *data = (char*)event + sizeof(*event); + struct watch *watch; memcpy_from_ring(xenstore_buf->rsp, - payload, - MASK_XENSTORE_IDX(xenstore_buf->rsp_cons), - msg.len + sizeof(msg)); + data, + MASK_XENSTORE_IDX(xenstore_buf->rsp_cons + sizeof(msg)), + msg.len); - path = payload + sizeof(msg); - token = path + strlen(path) + 1; + event->path = data; +...

The following two patches allow the parallel use of memsharing, xenpaging and xen-access by using an independent ring buffer for each feature. Please review. v2: - update mem_event_check_ring arguments, check domain rather than domain_id - check ring_full first because its value was just evaluated - check if ring buffer is initialized before calling mem_access_domctl/mem_paging_domctl

...*data, unsigned len) avail = len; /* Must read data /after/ reading the producer index. */ - rmb(); + virt_rmb(); memcpy(data, src, avail); data += avail; len -= avail; /* Other side must not see free space until we've copied out */ - mb(); + virt_mb(); intf->rsp_cons += avail; pr_debug("Finished read of %i bytes (%i to go)\n", avail, len); -- MST

...\ + virt_mb(); \ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ } while (0) @@ -279,7 +279,7 @@ struct __name##_back_ring { \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ if (_work_to_do) break; \ (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \ - mb(); \ + virt_mb(); \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ } while (0) -- MST

...data, unsigned len) avail = len; /* Must read data /after/ reading the producer index. */ - rmb(); + __smp_rmb(); memcpy(data, src, avail); data += avail; len -= avail; /* Other side must not see free space until we've copied out */ - mb(); + __smp_mb(); intf->rsp_cons += avail; pr_debug("Finished read of %i bytes (%i to go)\n", avail, len); -- MST

...\ + __smp_mb(); \ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ } while (0) @@ -279,7 +279,7 @@ struct __name##_back_ring { \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ if (_work_to_do) break; \ (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \ - mb(); \ + __smp_mb(); \ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ } while (0) -- MST

...a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 3c3dd40..7416f41 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -239,11 +239,14 @@ static void xennet_alloc_rx_buffers(struct net_device *dev) */ batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { - skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD, + skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) goto no_skb; + /* Align ip header to...

...a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 44aed80..2724688 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -239,11 +239,14 @@ static void xennet_alloc_rx_buffers(struct net_device *dev) */ batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { - skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD, + skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) goto no_skb; + /* Align ip header to...

...a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 3c3dd40..7416f41 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -239,11 +239,14 @@ static void xennet_alloc_rx_buffers(struct net_device *dev) */ batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { - skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD, + skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) goto no_skb; + /* Align ip header to...

...a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 3c3dd40..7416f41 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -239,11 +239,14 @@ static void xennet_alloc_rx_buffers(struct net_device *dev) */ batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { - skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD, + skb = __netdev_alloc_skb(dev, RX_COPY_THRESHOLD + NET_IP_ALIGN, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) goto no_skb; + /* Align ip header to...

...gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly); @@ -762,23 +762,22 @@ static RING_IDX xennet_fill_frags(struct netfront_info *np, struct skb_shared_info *shinfo = skb_shinfo(skb); int nr_frags = shinfo->nr_frags; RING_IDX cons = np->rx.rsp_cons; - skb_frag_t *frag = shinfo->frags + nr_frags; struct sk_buff *nskb; while ((nskb = __skb_dequeue(list))) { struct xen_netif_rx_response *rx = RING_GET_RESPONSE(&np->rx, ++cons); + skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0]; - frag->page = skb_shinfo(nskb...

Hi Andi, This series of patches implements the Xen paravirt-ops interface. It applies to 2.6.21-rc7 + your patches + the last batch of pv_ops patches I posted. This patch generally restricts itself to Xen-specific parts of the tree, though it does make a few small changes elsewhere. These patches include: - some helper routines for allocating address space and walking pagetables - Xen

Hi Andi, This series of patches implements the Xen paravirt-ops interface. It applies to 2.6.21-rc7 + your patches + the last batch of pv_ops patches I posted. This patch generally restricts itself to Xen-specific parts of the tree, though it does make a few small changes elsewhere. These patches include: - some helper routines for allocating address space and walking pagetables - Xen

Hi Andi, This series of patches implements the Xen paravirt-ops interface. It applies to 2.6.21 + 2.6.21-070428-1 + the patches I've posted over the last couple of days. Changes since the last posting: - update per review comments - comment each file - comment barrier use - use common cpu sibling setup - fixed a few minor bugs, specifically a traceirq mismatch warning - some