A long known problem of the upstream netback implementation that on the TX
path (from guest to Dom0) it copies the whole packet from guest memory into
Dom0. That simply became a bottleneck with 10Gb NICs, and generally it's a
huge perfomance penalty. The classic kernel version of netback used grant
mapping, and to get notified when the page can be unmapped, it used page
destructors. Unfortunately that destructor is not an upstreamable solution.
Ian Campbell's skb fragment destructor patch series [1] tried to solve this
problem, however it seems to be very invasive on the network stack's code,
and therefore haven't progressed very well.
This patch series use SKBTX_DEV_ZEROCOPY flags to tell the stack it needs to
know when the skb is freed up. That is the way KVM solved the same problem,
and based on my initial tests it can do the same for us. Avoiding the extra
copy boosted up TX throughput from 6.8 Gbps to 7.9 (I used a slower
Interlagos box, both Dom0 and guest on upstream kernel, on the same NUMA node,
running iperf 2.0.5, and the remote end was a bare metal box on the same 10Gb
switch)
Based on my investigations the packet get only copied if it is delivered to
Dom0 stack, which is due to this [2] patch. That's a bit unfortunate, but
luckily it doesn't cause a major regression for this usecase. In the future
we should try to eliminate that copy somehow.
There are a few spinoff tasks which will be addressed in separate patches:
- grant copy the header directly instead of map and memcpy. This should help
us avoiding TLB flushing
- use something else than ballooned pages
- fix grant map to use page->index properly
I will run some more extensive tests, but some basic XenRT tests were already
passed with good results.
I've tried to broke it down to smaller patches, with mixed results, so I
welcome suggestions on that part as well:
1: Introduce TX grant map definitions
2: Change TX path from grant copy to mapping
3: Remove old TX grant copy definitons and fix indentations
4: Change RX path for mapped SKB fragments
5: Add stat counters for zerocopy
6: Handle guests with too many frags
7: Add stat counters for frag_list skbs
8: Timeout packets in RX path
9: Aggregate TX unmap operations
v2: I've fixed some smaller things, see the individual patches. I've added a
few new stat counters, and handling the important use case when an older guest
sends lots of slots. Instead of delayed copy now we timeout packets on the RX
path, based on the assumption that otherwise packets should get stucked
anywhere else. Finally some unmap batching to avoid too much TLB flush
v3: Apart from fixing a few things mentioned in responses the important change
is the use the hypercall directly for grant [un]mapping, therefore we can
avoid m2p override.
[1] http://lwn.net/Articles/491522/
[2] https://lkml.org/lkml/2012/7/20/363
Signed-off-by: Zoltan Kiss <[email protected]>
This patch contains the new definitions necessary for grant mapping.
v2:
- move unmapping to separate thread. The NAPI instance has to be scheduled
even from thread context, which can cause huge delays
- that causes unfortunately bigger struct xenvif
- store grant handle after checking validity
v3:
- fix comment in xenvif_tx_dealloc_action()
- call unmap hypercall directly instead of gnttab_unmap_refs(), which does
unnecessary m2p_override. Also remove pages_to_[un]map members
- BUG() if grant_tx_handle corrupted
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 25 ++++++
drivers/net/xen-netback/interface.c | 1 +
drivers/net/xen-netback/netback.c | 163 +++++++++++++++++++++++++++++++++++
3 files changed, 189 insertions(+)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index d218ccd..f1071e3 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -79,6 +79,11 @@ struct pending_tx_info {
* if it is head of one or more tx
* reqs
*/
+ /* callback data for released SKBs. The callback is always
+ * xenvif_zerocopy_callback, ctx points to the next fragment, desc
+ * contains the pending_idx
+ */
+ struct ubuf_info callback_struct;
};
#define XEN_NETIF_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, PAGE_SIZE)
@@ -108,6 +113,8 @@ struct xenvif_rx_meta {
*/
#define MAX_GRANT_COPY_OPS (MAX_SKB_FRAGS * XEN_NETIF_RX_RING_SIZE)
+#define NETBACK_INVALID_HANDLE -1
+
struct xenvif {
/* Unique identifier for this interface. */
domid_t domid;
@@ -126,13 +133,23 @@ struct xenvif {
pending_ring_idx_t pending_cons;
u16 pending_ring[MAX_PENDING_REQS];
struct pending_tx_info pending_tx_info[MAX_PENDING_REQS];
+ grant_handle_t grant_tx_handle[MAX_PENDING_REQS];
/* Coalescing tx requests before copying makes number of grant
* copy ops greater or equal to number of slots required. In
* worst case a tx request consumes 2 gnttab_copy.
*/
struct gnttab_copy tx_copy_ops[2*MAX_PENDING_REQS];
+ struct gnttab_map_grant_ref tx_map_ops[MAX_PENDING_REQS];
+ struct gnttab_unmap_grant_ref tx_unmap_ops[MAX_PENDING_REQS];
+ spinlock_t dealloc_lock;
+ spinlock_t response_lock;
+ pending_ring_idx_t dealloc_prod;
+ pending_ring_idx_t dealloc_cons;
+ u16 dealloc_ring[MAX_PENDING_REQS];
+ struct task_struct *dealloc_task;
+ wait_queue_head_t dealloc_wq;
/* Use kthread for guest RX */
struct task_struct *task;
@@ -221,6 +238,8 @@ int xenvif_tx_action(struct xenvif *vif, int budget);
int xenvif_kthread(void *data);
void xenvif_kick_thread(struct xenvif *vif);
+int xenvif_dealloc_kthread(void *data);
+
/* Determine whether the needed number of slots (req) are available,
* and set req_event if not.
*/
@@ -228,6 +247,12 @@ bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed);
void xenvif_stop_queue(struct xenvif *vif);
+/* Callback from stack when TX packet can be released */
+void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success);
+
+/* Unmap a pending page, usually has to be called before xenvif_idx_release */
+void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx);
+
extern bool separate_tx_rx_irq;
#endif /* __XEN_NETBACK__COMMON_H__ */
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index 8d6def2..7170f97 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -37,6 +37,7 @@
#include <xen/events.h>
#include <asm/xen/hypercall.h>
+#include <xen/balloon.h>
#define XENVIF_QUEUE_LENGTH 32
#define XENVIF_NAPI_WEIGHT 64
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index addfe1d1..7c241f9 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -771,6 +771,19 @@ static struct page *xenvif_alloc_page(struct xenvif *vif,
return page;
}
+static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
+ struct xen_netif_tx_request *txp,
+ struct gnttab_map_grant_ref *gop)
+{
+ gnttab_set_map_op(gop, idx_to_kaddr(vif, pending_idx),
+ GNTMAP_host_map | GNTMAP_readonly,
+ txp->gref, vif->domid);
+
+ memcpy(&vif->pending_tx_info[pending_idx].req, txp,
+ sizeof(*txp));
+
+}
+
static struct gnttab_copy *xenvif_get_requests(struct xenvif *vif,
struct sk_buff *skb,
struct xen_netif_tx_request *txp,
@@ -1599,6 +1612,105 @@ static int xenvif_tx_submit(struct xenvif *vif)
return work_done;
}
+void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success)
+{
+ unsigned long flags;
+ pending_ring_idx_t index;
+ u16 pending_idx = ubuf->desc;
+ struct pending_tx_info *temp =
+ container_of(ubuf, struct pending_tx_info, callback_struct);
+ struct xenvif *vif =
+ container_of(temp - pending_idx, struct xenvif,
+ pending_tx_info[0]);
+
+ spin_lock_irqsave(&vif->dealloc_lock, flags);
+ do {
+ pending_idx = ubuf->desc;
+ ubuf = (struct ubuf_info *) ubuf->ctx;
+ index = pending_index(vif->dealloc_prod);
+ vif->dealloc_ring[index] = pending_idx;
+ /* Sync with xenvif_tx_action_dealloc:
+ * insert idx then incr producer.
+ */
+ smp_wmb();
+ vif->dealloc_prod++;
+ } while (ubuf);
+ wake_up(&vif->dealloc_wq);
+ spin_unlock_irqrestore(&vif->dealloc_lock, flags);
+}
+
+static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
+{
+ struct gnttab_unmap_grant_ref *gop;
+ pending_ring_idx_t dc, dp;
+ u16 pending_idx, pending_idx_release[MAX_PENDING_REQS];
+ unsigned int i = 0;
+
+ dc = vif->dealloc_cons;
+ gop = vif->tx_unmap_ops;
+
+ /* Free up any grants we have finished using */
+ do {
+ dp = vif->dealloc_prod;
+
+ /* Ensure we see all indices enqueued by all
+ * xenvif_zerocopy_callback().
+ */
+ smp_rmb();
+
+ while (dc != dp) {
+ pending_idx =
+ vif->dealloc_ring[pending_index(dc++)];
+
+ /* Already unmapped? */
+ if (vif->grant_tx_handle[pending_idx] ==
+ NETBACK_INVALID_HANDLE) {
+ netdev_err(vif->dev,
+ "Trying to unmap invalid handle! "
+ "pending_idx: %x\n", pending_idx);
+ continue;
+ }
+
+ pending_idx_release[gop-vif->tx_unmap_ops] =
+ pending_idx;
+ gnttab_set_unmap_op(gop,
+ idx_to_kaddr(vif, pending_idx),
+ GNTMAP_host_map,
+ vif->grant_tx_handle[pending_idx]);
+ vif->grant_tx_handle[pending_idx] =
+ NETBACK_INVALID_HANDLE;
+ ++gop;
+ }
+
+ } while (dp != vif->dealloc_prod);
+
+ vif->dealloc_cons = dc;
+
+ if (gop - vif->tx_unmap_ops > 0) {
+ int ret;
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref,
+ vif->tx_unmap_ops,
+ gop - vif->tx_unmap_ops);
+ if (ret) {
+ netdev_err(vif->dev, "Unmap fail: nr_ops %x ret %d\n",
+ gop - vif->tx_unmap_ops, ret);
+ for (i = 0; i < gop - vif->tx_unmap_ops; ++i) {
+ netdev_err(vif->dev,
+ " host_addr: %llx handle: %x status: %d\n",
+ gop[i].host_addr,
+ gop[i].handle,
+ gop[i].status);
+ }
+ BUG();
+ }
+ }
+
+ for (i = 0; i < gop - vif->tx_unmap_ops; ++i)
+ xenvif_idx_release(vif, pending_idx_release[i],
+ XEN_NETIF_RSP_OKAY);
+}
+
+
/* Called after netfront has transmitted */
int xenvif_tx_action(struct xenvif *vif, int budget)
{
@@ -1665,6 +1777,27 @@ static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
vif->mmap_pages[pending_idx] = NULL;
}
+void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
+{
+ int ret;
+ struct gnttab_unmap_grant_ref tx_unmap_op;
+
+ if (vif->grant_tx_handle[pending_idx] == NETBACK_INVALID_HANDLE) {
+ netdev_err(vif->dev,
+ "Trying to unmap invalid handle! pending_idx: %x\n",
+ pending_idx);
+ return;
+ }
+ gnttab_set_unmap_op(&tx_unmap_op,
+ idx_to_kaddr(vif, pending_idx),
+ GNTMAP_host_map,
+ vif->grant_tx_handle[pending_idx]);
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref,
+ &tx_unmap_op,
+ 1);
+ BUG_ON(ret);
+ vif->grant_tx_handle[pending_idx] = NETBACK_INVALID_HANDLE;
+}
static void make_tx_response(struct xenvif *vif,
struct xen_netif_tx_request *txp,
@@ -1726,6 +1859,14 @@ static inline int tx_work_todo(struct xenvif *vif)
return 0;
}
+static inline int tx_dealloc_work_todo(struct xenvif *vif)
+{
+ if (vif->dealloc_cons != vif->dealloc_prod)
+ return 1;
+
+ return 0;
+}
+
void xenvif_unmap_frontend_rings(struct xenvif *vif)
{
if (vif->tx.sring)
@@ -1814,6 +1955,28 @@ int xenvif_kthread(void *data)
return 0;
}
+int xenvif_dealloc_kthread(void *data)
+{
+ struct xenvif *vif = data;
+
+ while (!kthread_should_stop()) {
+ wait_event_interruptible(vif->dealloc_wq,
+ tx_dealloc_work_todo(vif) ||
+ kthread_should_stop());
+ if (kthread_should_stop())
+ break;
+
+ xenvif_tx_dealloc_action(vif);
+ cond_resched();
+ }
+
+ /* Unmap anything remaining*/
+ if (tx_dealloc_work_todo(vif))
+ xenvif_tx_dealloc_action(vif);
+
+ return 0;
+}
+
static int __init netback_init(void)
{
int rc = 0;
These became obsolate with grant mapping. I've left intentionally the
indentations in this way, to improve readability of previous patches.
v2:
- move the indentation fixup patch here
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 37 +------------------
drivers/net/xen-netback/netback.c | 72 ++++++++-----------------------------
2 files changed, 15 insertions(+), 94 deletions(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 33cb12c..f286879 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -46,39 +46,9 @@
#include <xen/xenbus.h>
typedef unsigned int pending_ring_idx_t;
-#define INVALID_PENDING_RING_IDX (~0U)
-/* For the head field in pending_tx_info: it is used to indicate
- * whether this tx info is the head of one or more coalesced requests.
- *
- * When head != INVALID_PENDING_RING_IDX, it means the start of a new
- * tx requests queue and the end of previous queue.
- *
- * An example sequence of head fields (I = INVALID_PENDING_RING_IDX):
- *
- * ...|0 I I I|5 I|9 I I I|...
- * -->|<-INUSE----------------
- *
- * After consuming the first slot(s) we have:
- *
- * ...|V V V V|5 I|9 I I I|...
- * -----FREE->|<-INUSE--------
- *
- * where V stands for "valid pending ring index". Any number other
- * than INVALID_PENDING_RING_IDX is OK. These entries are considered
- * free and can contain any number other than
- * INVALID_PENDING_RING_IDX. In practice we use 0.
- *
- * The in use non-INVALID_PENDING_RING_IDX (say 0, 5 and 9 in the
- * above example) number is the index into pending_tx_info and
- * mmap_pages arrays.
- */
struct pending_tx_info {
- struct xen_netif_tx_request req; /* coalesced tx request */
- pending_ring_idx_t head; /* head != INVALID_PENDING_RING_IDX
- * if it is head of one or more tx
- * reqs
- */
+ struct xen_netif_tx_request req; /* tx request */
/* callback data for released SKBs. The callback is always
* xenvif_zerocopy_callback, ctx points to the next fragment, desc
* contains the pending_idx
@@ -128,11 +98,6 @@ struct xenvif {
struct pending_tx_info pending_tx_info[MAX_PENDING_REQS];
grant_handle_t grant_tx_handle[MAX_PENDING_REQS];
- /* Coalescing tx requests before copying makes number of grant
- * copy ops greater or equal to number of slots required. In
- * worst case a tx request consumes 2 gnttab_copy.
- */
- struct gnttab_copy tx_copy_ops[2*MAX_PENDING_REQS];
struct gnttab_map_grant_ref tx_map_ops[MAX_PENDING_REQS];
struct gnttab_unmap_grant_ref tx_unmap_ops[MAX_PENDING_REQS];
/* passed to gnttab_[un]map_refs with pages under (un)mapping */
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 20352be..88a0fad 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -71,16 +71,6 @@ module_param(fatal_skb_slots, uint, 0444);
*/
#define XEN_NETBK_LEGACY_SLOTS_MAX XEN_NETIF_NR_SLOTS_MIN
-/*
- * If head != INVALID_PENDING_RING_IDX, it means this tx request is head of
- * one or more merged tx requests, otherwise it is the continuation of
- * previous tx request.
- */
-static inline int pending_tx_is_head(struct xenvif *vif, RING_IDX idx)
-{
- return vif->pending_tx_info[idx].head != INVALID_PENDING_RING_IDX;
-}
-
static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
u8 status);
@@ -762,19 +752,6 @@ static int xenvif_count_requests(struct xenvif *vif,
return slots;
}
-static struct page *xenvif_alloc_page(struct xenvif *vif,
- u16 pending_idx)
-{
- struct page *page;
-
- page = alloc_page(GFP_ATOMIC|__GFP_COLD);
- if (!page)
- return NULL;
- vif->mmap_pages[pending_idx] = page;
-
- return page;
-}
-
static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
struct xen_netif_tx_request *txp,
struct gnttab_map_grant_ref *gop)
@@ -797,13 +774,9 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
struct skb_shared_info *shinfo = skb_shinfo(skb);
skb_frag_t *frags = shinfo->frags;
u16 pending_idx = *((u16 *)skb->data);
- u16 head_idx = 0;
- int slot, start;
- struct page *page;
- pending_ring_idx_t index, start_idx = 0;
- uint16_t dst_offset;
+ int start;
+ pending_ring_idx_t index;
unsigned int nr_slots;
- struct pending_tx_info *first = NULL;
/* At this point shinfo->nr_frags is in fact the number of
* slots, which can be as large as XEN_NETBK_LEGACY_SLOTS_MAX.
@@ -815,8 +788,8 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
for (shinfo->nr_frags = start; shinfo->nr_frags < nr_slots;
shinfo->nr_frags++, txp++, gop++) {
- index = pending_index(vif->pending_cons++);
- pending_idx = vif->pending_ring[index];
+ index = pending_index(vif->pending_cons++);
+ pending_idx = vif->pending_ring[index];
xenvif_tx_create_gop(vif, pending_idx, txp, gop);
frag_set_pending_idx(&frags[shinfo->nr_frags], pending_idx);
}
@@ -824,18 +797,6 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
BUG_ON(shinfo->nr_frags > MAX_SKB_FRAGS);
return gop;
-err:
- /* Unwind, freeing all pages and sending error responses. */
- while (shinfo->nr_frags-- > start) {
- xenvif_idx_release(vif,
- frag_get_pending_idx(&frags[shinfo->nr_frags]),
- XEN_NETIF_RSP_ERROR);
- }
- /* The head too, if necessary. */
- if (start)
- xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
-
- return NULL;
}
static int xenvif_tx_check_gop(struct xenvif *vif,
@@ -848,7 +809,6 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
struct pending_tx_info *tx_info;
int nr_frags = shinfo->nr_frags;
int i, err, start;
- u16 peek; /* peek into next tx request */
/* Check status of header. */
err = gop->status;
@@ -870,14 +830,12 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
for (i = start; i < nr_frags; i++) {
int j, newerr;
- pending_ring_idx_t head;
pending_idx = frag_get_pending_idx(&shinfo->frags[i]);
tx_info = &vif->pending_tx_info[pending_idx];
- head = tx_info->head;
/* Check error status: if okay then remember grant handle. */
- newerr = (++gop)->status;
+ newerr = (++gop)->status;
if (likely(!newerr)) {
if (vif->grant_tx_handle[pending_idx] !=
@@ -1343,7 +1301,6 @@ static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
(skb_queue_len(&vif->tx_queue) < budget)) {
struct xen_netif_tx_request txreq;
struct xen_netif_tx_request txfrags[XEN_NETBK_LEGACY_SLOTS_MAX];
- struct page *page;
struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX-1];
u16 pending_idx;
RING_IDX idx;
@@ -1705,18 +1662,17 @@ static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
{
struct pending_tx_info *pending_tx_info;
pending_ring_idx_t index;
- u16 peek; /* peek into next tx request */
unsigned long flags;
- pending_tx_info = &vif->pending_tx_info[pending_idx];
- spin_lock_irqsave(&vif->response_lock, flags);
- make_tx_response(vif, &pending_tx_info->req, status);
- index = pending_index(vif->pending_prod);
- vif->pending_ring[index] = pending_idx;
- /* TX shouldn't use the index before we give it back here */
- mb();
- vif->pending_prod++;
- spin_unlock_irqrestore(&vif->response_lock, flags);
+ pending_tx_info = &vif->pending_tx_info[pending_idx];
+ spin_lock_irqsave(&vif->response_lock, flags);
+ make_tx_response(vif, &pending_tx_info->req, status);
+ index = pending_index(vif->pending_prod);
+ vif->pending_ring[index] = pending_idx;
+ /* TX shouldn't use the index before we give it back here */
+ mb();
+ vif->pending_prod++;
+ spin_unlock_irqrestore(&vif->response_lock, flags);
}
void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
These counters help determine how often the guest sends a packet with more
than MAX_SKB_FRAGS frags.
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 1 +
drivers/net/xen-netback/interface.c | 7 +++++++
drivers/net/xen-netback/netback.c | 1 +
3 files changed, 9 insertions(+)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index e3c28ff..c037efb 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -158,6 +158,7 @@ struct xenvif {
unsigned long tx_zerocopy_sent;
unsigned long tx_zerocopy_success;
unsigned long tx_zerocopy_fail;
+ unsigned long tx_frag_overflow;
/* Miscellaneous private stuff. */
struct net_device *dev;
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index ac27af3..b7daf8d 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -254,6 +254,13 @@ static const struct xenvif_stat {
"tx_zerocopy_fail",
offsetof(struct xenvif, tx_zerocopy_fail)
},
+ /* Number of packets exceeding MAX_SKB_FRAG slots. You should use
+ * a guest with the same MAX_SKB_FRAG
+ */
+ {
+ "tx_frag_overflow",
+ offsetof(struct xenvif, tx_frag_overflow)
+ },
};
static int xenvif_get_sset_count(struct net_device *dev, int string_set)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 9841429..4305965 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -1656,6 +1656,7 @@ static int xenvif_tx_submit(struct xenvif *vif, int budget)
skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
vif->tx_zerocopy_sent += 2;
+ vif->tx_frag_overflow++;
nskb = skb;
skb = skb_copy_expand(skb, 0, 0, GFP_ATOMIC | __GFP_NOWARN);
Xen network protocol had implicit dependency on MAX_SKB_FRAGS. Netback has to
handle guests sending up to XEN_NETBK_LEGACY_SLOTS_MAX slots. To achieve that:
- create a new skb
- map the leftover slots to its frags (no linear buffer here!)
- chain it to the previous through skb_shinfo(skb)->frag_list
- map them
- copy the whole stuff into a brand new skb and send it to the stack
- unmap the 2 old skb's pages
v3:
- adding extra check for frag number
- consolidate alloc_skb's into xenvif_alloc_skb()
- BUG_ON(frag_overflow > MAX_SKB_FRAGS)
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/netback.c | 115 +++++++++++++++++++++++++++++++++----
1 file changed, 105 insertions(+), 10 deletions(-)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index ea1e27d..3796cb3 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -800,6 +800,19 @@ static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
}
+static inline struct sk_buff *xenvif_alloc_skb(unsigned int size)
+{
+ struct sk_buff *skb = alloc_skb(size + NET_SKB_PAD + NET_IP_ALIGN,
+ GFP_ATOMIC | __GFP_NOWARN);
+ if (unlikely(skb == NULL))
+ return NULL;
+
+ /* Packets passed to netif_rx() must have some headroom. */
+ skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
+
+ return skb;
+}
+
static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
struct sk_buff *skb,
struct xen_netif_tx_request *txp,
@@ -810,11 +823,16 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
u16 pending_idx = *((u16 *)skb->data);
int start;
pending_ring_idx_t index;
- unsigned int nr_slots;
+ unsigned int nr_slots, frag_overflow = 0;
/* At this point shinfo->nr_frags is in fact the number of
* slots, which can be as large as XEN_NETBK_LEGACY_SLOTS_MAX.
*/
+ if (shinfo->nr_frags > MAX_SKB_FRAGS) {
+ frag_overflow = shinfo->nr_frags - MAX_SKB_FRAGS;
+ BUG_ON(frag_overflow > MAX_SKB_FRAGS);
+ shinfo->nr_frags = MAX_SKB_FRAGS;
+ }
nr_slots = shinfo->nr_frags;
/* Skip first skb fragment if it is on same page as header fragment. */
@@ -830,6 +848,29 @@ static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
BUG_ON(shinfo->nr_frags > MAX_SKB_FRAGS);
+ if (frag_overflow) {
+ struct sk_buff *nskb = xenvif_alloc_skb(0);
+ if (unlikely(nskb == NULL)) {
+ netdev_err(vif->dev,
+ "Can't allocate the frag_list skb.\n");
+ return NULL;
+ }
+
+ shinfo = skb_shinfo(nskb);
+ frags = shinfo->frags;
+
+ for (shinfo->nr_frags = 0; shinfo->nr_frags < frag_overflow;
+ shinfo->nr_frags++, txp++, gop++) {
+ index = pending_index(vif->pending_cons++);
+ pending_idx = vif->pending_ring[index];
+ xenvif_tx_create_gop(vif, pending_idx, txp, gop);
+ frag_set_pending_idx(&frags[shinfo->nr_frags],
+ pending_idx);
+ }
+
+ skb_shinfo(skb)->frag_list = nskb;
+ }
+
return gop;
}
@@ -843,6 +884,7 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
struct pending_tx_info *tx_info;
int nr_frags = shinfo->nr_frags;
int i, err, start;
+ struct sk_buff *first_skb = NULL;
/* Check status of header. */
err = gop->status;
@@ -862,6 +904,7 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
/* Skip first skb fragment if it is on same page as header fragment. */
start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx);
+check_frags:
for (i = start; i < nr_frags; i++) {
int j, newerr;
@@ -896,11 +939,20 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
/* Not the first error? Preceding frags already invalidated. */
if (err)
continue;
-
/* First error: invalidate header and preceding fragments. */
- pending_idx = *((u16 *)skb->data);
- xenvif_idx_unmap(vif, pending_idx);
- xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_OKAY);
+ if (!first_skb) {
+ pending_idx = *((u16 *)skb->data);
+ xenvif_idx_unmap(vif, pending_idx);
+ xenvif_idx_release(vif,
+ pending_idx,
+ XEN_NETIF_RSP_OKAY);
+ } else {
+ pending_idx = *((u16 *)first_skb->data);
+ xenvif_idx_unmap(vif, pending_idx);
+ xenvif_idx_release(vif,
+ pending_idx,
+ XEN_NETIF_RSP_OKAY);
+ }
for (j = start; j < i; j++) {
pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
xenvif_idx_unmap(vif, pending_idx);
@@ -912,6 +964,32 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
err = newerr;
}
+ if (shinfo->frag_list) {
+ first_skb = skb;
+ skb = shinfo->frag_list;
+ shinfo = skb_shinfo(skb);
+ nr_frags = shinfo->nr_frags;
+ start = 0;
+
+ goto check_frags;
+ }
+
+ /* There was a mapping error in the frag_list skb. We have to unmap
+ * the first skb's frags
+ */
+ if (first_skb && err) {
+ int j;
+ shinfo = skb_shinfo(first_skb);
+ pending_idx = *((u16 *)first_skb->data);
+ start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx);
+ for (j = start; j < shinfo->nr_frags; j++) {
+ pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
+ xenvif_idx_unmap(vif, pending_idx);
+ xenvif_idx_release(vif, pending_idx,
+ XEN_NETIF_RSP_OKAY);
+ }
+ }
+
*gopp = gop + 1;
return err;
}
@@ -1403,8 +1481,7 @@ static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
ret < XEN_NETBK_LEGACY_SLOTS_MAX) ?
PKT_PROT_LEN : txreq.size;
- skb = alloc_skb(data_len + NET_SKB_PAD + NET_IP_ALIGN,
- GFP_ATOMIC | __GFP_NOWARN);
+ skb = xenvif_alloc_skb(data_len);
if (unlikely(skb == NULL)) {
netdev_dbg(vif->dev,
"Can't allocate a skb in start_xmit.\n");
@@ -1412,9 +1489,6 @@ static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
break;
}
- /* Packets passed to netif_rx() must have some headroom. */
- skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN);
-
if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) {
struct xen_netif_extra_info *gso;
gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
@@ -1476,6 +1550,7 @@ static int xenvif_tx_submit(struct xenvif *vif)
struct xen_netif_tx_request *txp;
u16 pending_idx;
unsigned data_len;
+ struct sk_buff *nskb = NULL;
pending_idx = *((u16 *)skb->data);
txp = &vif->pending_tx_info[pending_idx].req;
@@ -1518,6 +1593,23 @@ static int xenvif_tx_submit(struct xenvif *vif)
pending_idx :
INVALID_PENDING_IDX);
+ if (skb_shinfo(skb)->frag_list) {
+ nskb = skb_shinfo(skb)->frag_list;
+ xenvif_fill_frags(vif, nskb, INVALID_PENDING_IDX);
+ skb->len += nskb->len;
+ skb->data_len += nskb->len;
+ skb->truesize += nskb->truesize;
+ skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ vif->tx_zerocopy_sent += 2;
+ nskb = skb;
+
+ skb = skb_copy_expand(skb,
+ 0,
+ 0,
+ GFP_ATOMIC | __GFP_NOWARN);
+ skb_shinfo(skb)->destructor_arg = NULL;
+ }
if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
int target = min_t(int, skb->len, PKT_PROT_LEN);
__pskb_pull_tail(skb, target - skb_headlen(skb));
@@ -1568,6 +1660,9 @@ static int xenvif_tx_submit(struct xenvif *vif)
}
netif_receive_skb(skb);
+
+ if (nskb)
+ kfree_skb(nskb);
}
return work_done;
A malicious or buggy guest can leave its queue filled indefinitely, in which
case qdisc start to queue packets for that VIF. If those packets came from an
another guest, it can block its slots and prevent shutdown. To avoid that, we
make sure the queue is drained in every 10 seconds.
v3:
- remove stale debug log
- tie unmap timeout in xenvif_free to this timeout
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 5 +++++
drivers/net/xen-netback/interface.c | 22 ++++++++++++++++++++--
drivers/net/xen-netback/netback.c | 9 +++++++++
3 files changed, 34 insertions(+), 2 deletions(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index dda3fd5..063fcda 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -130,6 +130,8 @@ struct xenvif {
*/
bool rx_event;
+ struct timer_list wake_queue;
+
/* This array is allocated seperately as it is large */
struct gnttab_copy *grant_copy_op;
@@ -224,4 +226,7 @@ void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx);
extern bool separate_tx_rx_irq;
+extern unsigned int rx_drain_timeout_msecs;
+extern unsigned int rx_drain_timeout_jiffies;
+
#endif /* __XEN_NETBACK__COMMON_H__ */
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index 95fcd63..ce032f9 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -114,6 +114,16 @@ static irqreturn_t xenvif_interrupt(int irq, void *dev_id)
return IRQ_HANDLED;
}
+static void xenvif_wake_queue(unsigned long data)
+{
+ struct xenvif *vif = (struct xenvif *)data;
+
+ if (netif_queue_stopped(vif->dev)) {
+ netdev_err(vif->dev, "draining TX queue\n");
+ netif_wake_queue(vif->dev);
+ }
+}
+
static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct xenvif *vif = netdev_priv(dev);
@@ -143,8 +153,13 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
* then turn off the queue to give the ring a chance to
* drain.
*/
- if (!xenvif_rx_ring_slots_available(vif, min_slots_needed))
+ if (!xenvif_rx_ring_slots_available(vif, min_slots_needed)) {
+ vif->wake_queue.function = xenvif_wake_queue;
+ vif->wake_queue.data = (unsigned long)vif;
xenvif_stop_queue(vif);
+ mod_timer(&vif->wake_queue,
+ jiffies + rx_drain_timeout_jiffies);
+ }
skb_queue_tail(&vif->rx_queue, skb);
xenvif_kick_thread(vif);
@@ -353,6 +368,8 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
/* Initialize 'expires' now: it's used to track the credit window. */
vif->credit_timeout.expires = jiffies;
+ init_timer(&vif->wake_queue);
+
dev->netdev_ops = &xenvif_netdev_ops;
dev->hw_features = NETIF_F_SG |
NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM |
@@ -528,6 +545,7 @@ void xenvif_disconnect(struct xenvif *vif)
xenvif_carrier_off(vif);
if (vif->task) {
+ del_timer_sync(&vif->wake_queue);
kthread_stop(vif->task);
vif->task = NULL;
}
@@ -558,7 +576,7 @@ void xenvif_free(struct xenvif *vif)
if (vif->grant_tx_handle[i] != NETBACK_INVALID_HANDLE) {
unmap_timeout++;
schedule_timeout(msecs_to_jiffies(1000));
- if (unmap_timeout > 9 &&
+ if (unmap_timeout > (rx_drain_timeout_msecs/1000) &&
net_ratelimit())
netdev_err(vif->dev,
"Page still granted! Index: %x\n", i);
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index f815395..6bc5413 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -62,6 +62,13 @@ module_param(separate_tx_rx_irq, bool, 0644);
static unsigned int fatal_skb_slots = FATAL_SKB_SLOTS_DEFAULT;
module_param(fatal_skb_slots, uint, 0444);
+/* When guest ring is filled up, qdisc queues the packets for us, but we have
+ * to timeout them, otherwise other guests' packets can get stucked there
+ */
+unsigned int rx_drain_timeout_msecs = 10000;
+module_param(rx_drain_timeout_msecs, uint, 0444);
+unsigned int rx_drain_timeout_jiffies;
+
/*
* To avoid confusion, we define XEN_NETBK_LEGACY_SLOTS_MAX indicating
* the maximum slots a valid packet can use. Now this value is defined
@@ -2032,6 +2039,8 @@ static int __init netback_init(void)
if (rc)
goto failed_init;
+ rx_drain_timeout_jiffies = msecs_to_jiffies(rx_drain_timeout_msecs);
+
return 0;
failed_init:
Sorry, the version number in the subject should be v3
Zoli
Unmapping causes TLB flushing, therefore we should make it in the largest
possible batches. However we shouldn't starve the guest for too long. So if
the guest has space for at least two big packets and we don't have at least a
quarter ring to unmap, delay it for at most 1 milisec.
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 2 ++
drivers/net/xen-netback/interface.c | 2 ++
drivers/net/xen-netback/netback.c | 31 ++++++++++++++++++++++++++++++-
3 files changed, 34 insertions(+), 1 deletion(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 063fcda..55d1f14 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -115,6 +115,8 @@ struct xenvif {
u16 dealloc_ring[MAX_PENDING_REQS];
struct task_struct *dealloc_task;
wait_queue_head_t dealloc_wq;
+ struct timer_list dealloc_delay;
+ bool dealloc_delay_timed_out;
/* Use kthread for guest RX */
struct task_struct *task;
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index ce032f9..0287d62 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -406,6 +406,7 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
.desc = i };
vif->grant_tx_handle[i] = NETBACK_INVALID_HANDLE;
}
+ init_timer(&vif->dealloc_delay);
/*
* Initialise a dummy MAC address. We choose the numerically
@@ -551,6 +552,7 @@ void xenvif_disconnect(struct xenvif *vif)
}
if (vif->dealloc_task) {
+ del_timer_sync(&vif->dealloc_delay);
kthread_stop(vif->dealloc_task);
vif->dealloc_task = NULL;
}
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 6bc5413..27cc36c 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -134,6 +134,11 @@ static inline pending_ring_idx_t nr_pending_reqs(struct xenvif *vif)
vif->pending_prod + vif->pending_cons;
}
+static inline pending_ring_idx_t nr_free_slots(struct xen_netif_tx_back_ring *ring)
+{
+ return ring->nr_ents - (ring->sring->req_prod - ring->rsp_prod_pvt);
+}
+
bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed)
{
RING_IDX prod, cons;
@@ -1904,10 +1909,34 @@ static inline int tx_work_todo(struct xenvif *vif)
return 0;
}
+static void xenvif_dealloc_delay(unsigned long data)
+{
+ struct xenvif *vif = (struct xenvif *)data;
+
+ vif->dealloc_delay_timed_out = true;
+ wake_up(&vif->dealloc_wq);
+}
+
static inline int tx_dealloc_work_todo(struct xenvif *vif)
{
- if (vif->dealloc_cons != vif->dealloc_prod)
+ if (vif->dealloc_cons != vif->dealloc_prod) {
+ if ((nr_free_slots(&vif->tx) > 2 * XEN_NETBK_LEGACY_SLOTS_MAX) &&
+ (vif->dealloc_prod - vif->dealloc_cons < MAX_PENDING_REQS / 4) &&
+ !vif->dealloc_delay_timed_out) {
+ if (!timer_pending(&vif->dealloc_delay)) {
+ vif->dealloc_delay.function =
+ xenvif_dealloc_delay;
+ vif->dealloc_delay.data = (unsigned long)vif;
+ mod_timer(&vif->dealloc_delay,
+ jiffies + msecs_to_jiffies(1));
+
+ }
+ return 0;
+ }
+ del_timer_sync(&vif->dealloc_delay);
+ vif->dealloc_delay_timed_out = false;
return 1;
+ }
return 0;
}
These counters help determine how often the buffers had to be copied. Also
they help find out if packets are leaked, as if "sent != success + fail",
there are probably packets never freed up properly.
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/common.h | 3 +++
drivers/net/xen-netback/interface.c | 15 +++++++++++++++
drivers/net/xen-netback/netback.c | 9 ++++++++-
3 files changed, 26 insertions(+), 1 deletion(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 419e63c..e3c28ff 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -155,6 +155,9 @@ struct xenvif {
/* Statistics */
unsigned long rx_gso_checksum_fixup;
+ unsigned long tx_zerocopy_sent;
+ unsigned long tx_zerocopy_success;
+ unsigned long tx_zerocopy_fail;
/* Miscellaneous private stuff. */
struct net_device *dev;
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index af5216f..75fe683 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -239,6 +239,21 @@ static const struct xenvif_stat {
"rx_gso_checksum_fixup",
offsetof(struct xenvif, rx_gso_checksum_fixup)
},
+ /* If (sent != success + fail), there are probably packets never
+ * freed up properly!
+ */
+ {
+ "tx_zerocopy_sent",
+ offsetof(struct xenvif, tx_zerocopy_sent),
+ },
+ {
+ "tx_zerocopy_success",
+ offsetof(struct xenvif, tx_zerocopy_success),
+ },
+ {
+ "tx_zerocopy_fail",
+ offsetof(struct xenvif, tx_zerocopy_fail)
+ },
};
static int xenvif_get_sset_count(struct net_device *dev, int string_set)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index a1b03e4..e2dd565 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -1611,8 +1611,10 @@ static int xenvif_tx_submit(struct xenvif *vif, int budget)
* skb_copy_ubufs while we are still in control of the skb. E.g.
* the __pskb_pull_tail earlier can do such thing.
*/
- if (skb_shinfo(skb)->destructor_arg)
+ if (skb_shinfo(skb)->destructor_arg) {
skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ vif->tx_zerocopy_sent++;
+ }
netif_receive_skb(skb);
}
@@ -1645,6 +1647,11 @@ void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success)
napi_schedule(&vif->napi);
} while (ubuf);
spin_unlock_irqrestore(&vif->dealloc_lock, flags);
+
+ if (likely(zerocopy_success))
+ vif->tx_zerocopy_success++;
+ else
+ vif->tx_zerocopy_fail++;
}
static inline void xenvif_tx_action_dealloc(struct xenvif *vif)
RX path need to know if the SKB fragments are stored on pages from another
domain.
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/netback.c | 46 +++++++++++++++++++++++++++++++++----
1 file changed, 41 insertions(+), 5 deletions(-)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 10d0cf0..e070475 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -322,7 +322,9 @@ static struct xenvif_rx_meta *get_next_rx_buffer(struct xenvif *vif,
static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
struct netrx_pending_operations *npo,
struct page *page, unsigned long size,
- unsigned long offset, int *head)
+ unsigned long offset, int *head,
+ struct xenvif *foreign_vif,
+ grant_ref_t foreign_gref)
{
struct gnttab_copy *copy_gop;
struct xenvif_rx_meta *meta;
@@ -364,8 +366,15 @@ static void xenvif_gop_frag_copy(struct xenvif *vif, struct sk_buff *skb,
copy_gop->flags = GNTCOPY_dest_gref;
copy_gop->len = bytes;
- copy_gop->source.domid = DOMID_SELF;
- copy_gop->source.u.gmfn = virt_to_mfn(page_address(page));
+ if (foreign_vif) {
+ copy_gop->source.domid = foreign_vif->domid;
+ copy_gop->source.u.ref = foreign_gref;
+ copy_gop->flags |= GNTCOPY_source_gref;
+ } else {
+ copy_gop->source.domid = DOMID_SELF;
+ copy_gop->source.u.gmfn =
+ virt_to_mfn(page_address(page));
+ }
copy_gop->source.offset = offset;
copy_gop->dest.domid = vif->domid;
@@ -426,6 +435,9 @@ static int xenvif_gop_skb(struct sk_buff *skb,
int old_meta_prod;
int gso_type;
int gso_size;
+ struct ubuf_info *ubuf = skb_shinfo(skb)->destructor_arg;
+ grant_ref_t foreign_grefs[MAX_SKB_FRAGS];
+ struct xenvif *foreign_vif = NULL;
old_meta_prod = npo->meta_prod;
@@ -466,6 +478,26 @@ static int xenvif_gop_skb(struct sk_buff *skb,
npo->copy_off = 0;
npo->copy_gref = req->gref;
+ if ((skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) &&
+ (ubuf->callback == &xenvif_zerocopy_callback)) {
+ u16 pending_idx = ubuf->desc;
+ int i = 0;
+ struct pending_tx_info *temp =
+ container_of(ubuf,
+ struct pending_tx_info,
+ callback_struct);
+ foreign_vif =
+ container_of(temp - pending_idx,
+ struct xenvif,
+ pending_tx_info[0]);
+ do {
+ pending_idx = ubuf->desc;
+ foreign_grefs[i++] =
+ foreign_vif->pending_tx_info[pending_idx].req.gref;
+ ubuf = (struct ubuf_info *) ubuf->ctx;
+ } while (ubuf);
+ }
+
data = skb->data;
while (data < skb_tail_pointer(skb)) {
unsigned int offset = offset_in_page(data);
@@ -475,7 +507,9 @@ static int xenvif_gop_skb(struct sk_buff *skb,
len = skb_tail_pointer(skb) - data;
xenvif_gop_frag_copy(vif, skb, npo,
- virt_to_page(data), len, offset, &head);
+ virt_to_page(data), len, offset, &head,
+ NULL,
+ 0);
data += len;
}
@@ -484,7 +518,9 @@ static int xenvif_gop_skb(struct sk_buff *skb,
skb_frag_page(&skb_shinfo(skb)->frags[i]),
skb_frag_size(&skb_shinfo(skb)->frags[i]),
skb_shinfo(skb)->frags[i].page_offset,
- &head);
+ &head,
+ foreign_vif,
+ foreign_grefs[i]);
}
return npo->meta_prod - old_meta_prod;
This patch changes the grant copy on the TX patch to grant mapping
v2:
- delete branch for handling fragmented packets fit PKT_PROT_LEN sized first
request
- mark the effect of using ballooned pages in a comment
- place setting of skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY right
before netif_receive_skb, and mark the importance of it
- grab dealloc_lock before __napi_complete to avoid contention with the
callback's napi_schedule
- handle fragmented packets where first request < PKT_PROT_LEN
- fix up error path when checksum_setup failed
- check before teardown for pending grants, and start complain if they are
there after 10 second
v3:
- delete a surplus checking from tx_action
- remove stray line
- squash xenvif_idx_unmap changes into the first patch
- init spinlocks
- call map hypercall directly instead of gnttab_map_refs()
- fix unmapping timeout in xenvif_free()
Signed-off-by: Zoltan Kiss <[email protected]>
---
drivers/net/xen-netback/interface.c | 57 +++++++-
drivers/net/xen-netback/netback.c | 251 ++++++++++++++---------------------
2 files changed, 153 insertions(+), 155 deletions(-)
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index 7170f97..3b2b249 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -122,7 +122,9 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
BUG_ON(skb->dev != dev);
/* Drop the packet if vif is not ready */
- if (vif->task == NULL || !xenvif_schedulable(vif))
+ if (vif->task == NULL ||
+ vif->dealloc_task == NULL ||
+ !xenvif_schedulable(vif))
goto drop;
/* At best we'll need one slot for the header and one for each
@@ -345,8 +347,26 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
vif->pending_prod = MAX_PENDING_REQS;
for (i = 0; i < MAX_PENDING_REQS; i++)
vif->pending_ring[i] = i;
- for (i = 0; i < MAX_PENDING_REQS; i++)
- vif->mmap_pages[i] = NULL;
+ spin_lock_init(&vif->dealloc_lock);
+ spin_lock_init(&vif->response_lock);
+ /* If ballooning is disabled, this will consume real memory, so you
+ * better enable it. The long term solution would be to use just a
+ * bunch of valid page descriptors, without dependency on ballooning
+ */
+ err = alloc_xenballooned_pages(MAX_PENDING_REQS,
+ vif->mmap_pages,
+ false);
+ if (err) {
+ netdev_err(dev, "Could not reserve mmap_pages\n");
+ return NULL;
+ }
+ for (i = 0; i < MAX_PENDING_REQS; i++) {
+ vif->pending_tx_info[i].callback_struct = (struct ubuf_info)
+ { .callback = xenvif_zerocopy_callback,
+ .ctx = NULL,
+ .desc = i };
+ vif->grant_tx_handle[i] = NETBACK_INVALID_HANDLE;
+ }
/*
* Initialise a dummy MAC address. We choose the numerically
@@ -390,6 +410,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
goto err;
init_waitqueue_head(&vif->wq);
+ init_waitqueue_head(&vif->dealloc_wq);
if (tx_evtchn == rx_evtchn) {
/* feature-split-event-channels == 0 */
@@ -431,6 +452,14 @@ int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
goto err_rx_unbind;
}
+ vif->dealloc_task = kthread_create(xenvif_dealloc_kthread,
+ (void *)vif, "%s-dealloc", vif->dev->name);
+ if (IS_ERR(vif->dealloc_task)) {
+ pr_warn("Could not allocate kthread for %s\n", vif->dev->name);
+ err = PTR_ERR(vif->dealloc_task);
+ goto err_rx_unbind;
+ }
+
vif->task = task;
rtnl_lock();
@@ -443,6 +472,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
rtnl_unlock();
wake_up_process(vif->task);
+ wake_up_process(vif->dealloc_task);
return 0;
@@ -480,6 +510,11 @@ void xenvif_disconnect(struct xenvif *vif)
vif->task = NULL;
}
+ if (vif->dealloc_task) {
+ kthread_stop(vif->dealloc_task);
+ vif->dealloc_task = NULL;
+ }
+
if (vif->tx_irq) {
if (vif->tx_irq == vif->rx_irq)
unbind_from_irqhandler(vif->tx_irq, vif);
@@ -495,6 +530,22 @@ void xenvif_disconnect(struct xenvif *vif)
void xenvif_free(struct xenvif *vif)
{
+ int i, unmap_timeout = 0;
+
+ for (i = 0; i < MAX_PENDING_REQS; ++i) {
+ if (vif->grant_tx_handle[i] != NETBACK_INVALID_HANDLE) {
+ unmap_timeout++;
+ schedule_timeout(msecs_to_jiffies(1000));
+ if (unmap_timeout > 9 &&
+ net_ratelimit())
+ netdev_err(vif->dev,
+ "Page still granted! Index: %x\n", i);
+ i = -1;
+ }
+ }
+
+ free_xenballooned_pages(MAX_PENDING_REQS, vif->mmap_pages);
+
netif_napi_del(&vif->napi);
unregister_netdev(vif->dev);
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 7c241f9..53d7e78 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -644,9 +644,12 @@ static void xenvif_tx_err(struct xenvif *vif,
struct xen_netif_tx_request *txp, RING_IDX end)
{
RING_IDX cons = vif->tx.req_cons;
+ unsigned long flags;
do {
+ spin_lock_irqsave(&vif->response_lock, flags);
make_tx_response(vif, txp, XEN_NETIF_RSP_ERROR);
+ spin_unlock_irqrestore(&vif->response_lock, flags);
if (cons == end)
break;
txp = RING_GET_REQUEST(&vif->tx, cons++);
@@ -784,10 +787,10 @@ static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
}
-static struct gnttab_copy *xenvif_get_requests(struct xenvif *vif,
+static struct gnttab_map_grant_ref *xenvif_get_requests(struct xenvif *vif,
struct sk_buff *skb,
struct xen_netif_tx_request *txp,
- struct gnttab_copy *gop)
+ struct gnttab_map_grant_ref *gop)
{
struct skb_shared_info *shinfo = skb_shinfo(skb);
skb_frag_t *frags = shinfo->frags;
@@ -808,83 +811,12 @@ static struct gnttab_copy *xenvif_get_requests(struct xenvif *vif,
/* Skip first skb fragment if it is on same page as header fragment. */
start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx);
- /* Coalesce tx requests, at this point the packet passed in
- * should be <= 64K. Any packets larger than 64K have been
- * handled in xenvif_count_requests().
- */
- for (shinfo->nr_frags = slot = start; slot < nr_slots;
- shinfo->nr_frags++) {
- struct pending_tx_info *pending_tx_info =
- vif->pending_tx_info;
-
- page = alloc_page(GFP_ATOMIC|__GFP_COLD);
- if (!page)
- goto err;
-
- dst_offset = 0;
- first = NULL;
- while (dst_offset < PAGE_SIZE && slot < nr_slots) {
- gop->flags = GNTCOPY_source_gref;
-
- gop->source.u.ref = txp->gref;
- gop->source.domid = vif->domid;
- gop->source.offset = txp->offset;
-
- gop->dest.domid = DOMID_SELF;
-
- gop->dest.offset = dst_offset;
- gop->dest.u.gmfn = virt_to_mfn(page_address(page));
-
- if (dst_offset + txp->size > PAGE_SIZE) {
- /* This page can only merge a portion
- * of tx request. Do not increment any
- * pointer / counter here. The txp
- * will be dealt with in future
- * rounds, eventually hitting the
- * `else` branch.
- */
- gop->len = PAGE_SIZE - dst_offset;
- txp->offset += gop->len;
- txp->size -= gop->len;
- dst_offset += gop->len; /* quit loop */
- } else {
- /* This tx request can be merged in the page */
- gop->len = txp->size;
- dst_offset += gop->len;
-
+ for (shinfo->nr_frags = start; shinfo->nr_frags < nr_slots;
+ shinfo->nr_frags++, txp++, gop++) {
index = pending_index(vif->pending_cons++);
-
pending_idx = vif->pending_ring[index];
-
- memcpy(&pending_tx_info[pending_idx].req, txp,
- sizeof(*txp));
-
- /* Poison these fields, corresponding
- * fields for head tx req will be set
- * to correct values after the loop.
- */
- vif->mmap_pages[pending_idx] = (void *)(~0UL);
- pending_tx_info[pending_idx].head =
- INVALID_PENDING_RING_IDX;
-
- if (!first) {
- first = &pending_tx_info[pending_idx];
- start_idx = index;
- head_idx = pending_idx;
- }
-
- txp++;
- slot++;
- }
-
- gop++;
- }
-
- first->req.offset = 0;
- first->req.size = dst_offset;
- first->head = start_idx;
- vif->mmap_pages[head_idx] = page;
- frag_set_pending_idx(&frags[shinfo->nr_frags], head_idx);
+ xenvif_tx_create_gop(vif, pending_idx, txp, gop);
+ frag_set_pending_idx(&frags[shinfo->nr_frags], pending_idx);
}
BUG_ON(shinfo->nr_frags > MAX_SKB_FRAGS);
@@ -906,9 +838,9 @@ err:
static int xenvif_tx_check_gop(struct xenvif *vif,
struct sk_buff *skb,
- struct gnttab_copy **gopp)
+ struct gnttab_map_grant_ref **gopp)
{
- struct gnttab_copy *gop = *gopp;
+ struct gnttab_map_grant_ref *gop = *gopp;
u16 pending_idx = *((u16 *)skb->data);
struct skb_shared_info *shinfo = skb_shinfo(skb);
struct pending_tx_info *tx_info;
@@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
err = gop->status;
if (unlikely(err))
xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
+ else {
+ if (vif->grant_tx_handle[pending_idx] !=
+ NETBACK_INVALID_HANDLE) {
+ netdev_err(vif->dev,
+ "Stale mapped handle! pending_idx %x handle %x\n",
+ pending_idx, vif->grant_tx_handle[pending_idx]);
+ BUG();
+ }
+ set_phys_to_machine(idx_to_pfn(vif, pending_idx),
+ FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
+ vif->grant_tx_handle[pending_idx] = gop->handle;
+ }
/* Skip first skb fragment if it is on same page as header fragment. */
start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx);
@@ -933,18 +877,26 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
head = tx_info->head;
/* Check error status: if okay then remember grant handle. */
- do {
newerr = (++gop)->status;
- if (newerr)
- break;
- peek = vif->pending_ring[pending_index(++head)];
- } while (!pending_tx_is_head(vif, peek));
if (likely(!newerr)) {
+ if (vif->grant_tx_handle[pending_idx] !=
+ NETBACK_INVALID_HANDLE) {
+ netdev_err(vif->dev,
+ "Stale mapped handle! pending_idx %x handle %x\n",
+ pending_idx,
+ vif->grant_tx_handle[pending_idx]);
+ xenvif_fatal_tx_err(vif);
+ }
+ set_phys_to_machine(idx_to_pfn(vif, pending_idx),
+ FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
+ vif->grant_tx_handle[pending_idx] = gop->handle;
/* Had a previous error? Invalidate this fragment. */
- if (unlikely(err))
+ if (unlikely(err)) {
+ xenvif_idx_unmap(vif, pending_idx);
xenvif_idx_release(vif, pending_idx,
XEN_NETIF_RSP_OKAY);
+ }
continue;
}
@@ -957,9 +909,11 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
/* First error: invalidate header and preceding fragments. */
pending_idx = *((u16 *)skb->data);
+ xenvif_idx_unmap(vif, pending_idx);
xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_OKAY);
for (j = start; j < i; j++) {
pending_idx = frag_get_pending_idx(&shinfo->frags[j]);
+ xenvif_idx_unmap(vif, pending_idx);
xenvif_idx_release(vif, pending_idx,
XEN_NETIF_RSP_OKAY);
}
@@ -972,7 +926,8 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
return err;
}
-static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
+static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb,
+ u16 prev_pending_idx)
{
struct skb_shared_info *shinfo = skb_shinfo(skb);
int nr_frags = shinfo->nr_frags;
@@ -986,6 +941,17 @@ static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
pending_idx = frag_get_pending_idx(frag);
+ /* If this is not the first frag, chain it to the previous*/
+ if (unlikely(prev_pending_idx == INVALID_PENDING_IDX))
+ skb_shinfo(skb)->destructor_arg =
+ &vif->pending_tx_info[pending_idx].callback_struct;
+ else if (likely(pending_idx != prev_pending_idx))
+ vif->pending_tx_info[prev_pending_idx].callback_struct.ctx =
+ &(vif->pending_tx_info[pending_idx].callback_struct);
+
+ vif->pending_tx_info[pending_idx].callback_struct.ctx = NULL;
+ prev_pending_idx = pending_idx;
+
txp = &vif->pending_tx_info[pending_idx].req;
page = virt_to_page(idx_to_kaddr(vif, pending_idx));
__skb_fill_page_desc(skb, i, page, txp->offset, txp->size);
@@ -993,10 +959,15 @@ static void xenvif_fill_frags(struct xenvif *vif, struct sk_buff *skb)
skb->data_len += txp->size;
skb->truesize += txp->size;
- /* Take an extra reference to offset xenvif_idx_release */
+ /* Take an extra reference to offset network stack's put_page */
get_page(vif->mmap_pages[pending_idx]);
- xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_OKAY);
}
+ /* FIXME: __skb_fill_page_desc set this to true because page->pfmemalloc
+ * overlaps with "index", and "mapping" is not set. I think mapping
+ * should be set. If delivered to local stack, it would drop this
+ * skb in sk_filter unless the socket has the right to use it.
+ */
+ skb->pfmemalloc = false;
}
static int xenvif_get_extras(struct xenvif *vif,
@@ -1358,7 +1329,7 @@ static bool tx_credit_exceeded(struct xenvif *vif, unsigned size)
static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
{
- struct gnttab_copy *gop = vif->tx_copy_ops, *request_gop;
+ struct gnttab_map_grant_ref *gop = vif->tx_map_ops, *request_gop;
struct sk_buff *skb;
int ret;
@@ -1466,30 +1437,10 @@ static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
}
}
- /* XXX could copy straight to head */
- page = xenvif_alloc_page(vif, pending_idx);
- if (!page) {
- kfree_skb(skb);
- xenvif_tx_err(vif, &txreq, idx);
- break;
- }
-
- gop->source.u.ref = txreq.gref;
- gop->source.domid = vif->domid;
- gop->source.offset = txreq.offset;
-
- gop->dest.u.gmfn = virt_to_mfn(page_address(page));
- gop->dest.domid = DOMID_SELF;
- gop->dest.offset = txreq.offset;
-
- gop->len = txreq.size;
- gop->flags = GNTCOPY_source_gref;
+ xenvif_tx_create_gop(vif, pending_idx, &txreq, gop);
gop++;
- memcpy(&vif->pending_tx_info[pending_idx].req,
- &txreq, sizeof(txreq));
- vif->pending_tx_info[pending_idx].head = index;
*((u16 *)skb->data) = pending_idx;
__skb_put(skb, data_len);
@@ -1518,17 +1469,17 @@ static unsigned xenvif_tx_build_gops(struct xenvif *vif, int budget)
vif->tx.req_cons = idx;
- if ((gop-vif->tx_copy_ops) >= ARRAY_SIZE(vif->tx_copy_ops))
+ if ((gop-vif->tx_map_ops) >= ARRAY_SIZE(vif->tx_map_ops))
break;
}
- return gop - vif->tx_copy_ops;
+ return gop - vif->tx_map_ops;
}
static int xenvif_tx_submit(struct xenvif *vif)
{
- struct gnttab_copy *gop = vif->tx_copy_ops;
+ struct gnttab_map_grant_ref *gop = vif->tx_map_ops;
struct sk_buff *skb;
int work_done = 0;
@@ -1552,12 +1503,17 @@ static int xenvif_tx_submit(struct xenvif *vif)
memcpy(skb->data,
(void *)(idx_to_kaddr(vif, pending_idx)|txp->offset),
data_len);
+ vif->pending_tx_info[pending_idx].callback_struct.ctx = NULL;
if (data_len < txp->size) {
/* Append the packet payload as a fragment. */
txp->offset += data_len;
txp->size -= data_len;
+ skb_shinfo(skb)->destructor_arg =
+ &vif->pending_tx_info[pending_idx].callback_struct;
} else {
/* Schedule a response immediately. */
+ skb_shinfo(skb)->destructor_arg = NULL;
+ xenvif_idx_unmap(vif, pending_idx);
xenvif_idx_release(vif, pending_idx,
XEN_NETIF_RSP_OKAY);
}
@@ -1567,7 +1523,11 @@ static int xenvif_tx_submit(struct xenvif *vif)
else if (txp->flags & XEN_NETTXF_data_validated)
skb->ip_summed = CHECKSUM_UNNECESSARY;
- xenvif_fill_frags(vif, skb);
+ xenvif_fill_frags(vif,
+ skb,
+ skb_shinfo(skb)->destructor_arg ?
+ pending_idx :
+ INVALID_PENDING_IDX);
if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
int target = min_t(int, skb->len, PKT_PROT_LEN);
@@ -1581,6 +1541,8 @@ static int xenvif_tx_submit(struct xenvif *vif)
if (checksum_setup(vif, skb)) {
netdev_dbg(vif->dev,
"Can't setup checksum in net_tx_action\n");
+ if (skb_shinfo(skb)->destructor_arg)
+ skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
kfree_skb(skb);
continue;
}
@@ -1606,6 +1568,14 @@ static int xenvif_tx_submit(struct xenvif *vif)
work_done++;
+ /* Set this flag right before netif_receive_skb, otherwise
+ * someone might think this packet already left netback, and
+ * do a skb_copy_ubufs while we are still in control of the
+ * skb. E.g. the __pskb_pull_tail earlier can do such thing.
+ */
+ if (skb_shinfo(skb)->destructor_arg)
+ skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+
netif_receive_skb(skb);
}
@@ -1715,7 +1685,7 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
int xenvif_tx_action(struct xenvif *vif, int budget)
{
unsigned nr_gops;
- int work_done;
+ int work_done, ret;
if (unlikely(!tx_work_todo(vif)))
return 0;
@@ -1725,7 +1695,10 @@ int xenvif_tx_action(struct xenvif *vif, int budget)
if (nr_gops == 0)
return 0;
- gnttab_batch_copy(vif->tx_copy_ops, nr_gops);
+ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
+ vif->tx_map_ops,
+ nr_gops);
+ BUG_ON(ret);
work_done = xenvif_tx_submit(vif);
@@ -1736,45 +1709,19 @@ static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
u8 status)
{
struct pending_tx_info *pending_tx_info;
- pending_ring_idx_t head;
+ pending_ring_idx_t index;
u16 peek; /* peek into next tx request */
+ unsigned long flags;
- BUG_ON(vif->mmap_pages[pending_idx] == (void *)(~0UL));
-
- /* Already complete? */
- if (vif->mmap_pages[pending_idx] == NULL)
- return;
-
- pending_tx_info = &vif->pending_tx_info[pending_idx];
-
- head = pending_tx_info->head;
-
- BUG_ON(!pending_tx_is_head(vif, head));
- BUG_ON(vif->pending_ring[pending_index(head)] != pending_idx);
-
- do {
- pending_ring_idx_t index;
- pending_ring_idx_t idx = pending_index(head);
- u16 info_idx = vif->pending_ring[idx];
-
- pending_tx_info = &vif->pending_tx_info[info_idx];
+ pending_tx_info = &vif->pending_tx_info[pending_idx];
+ spin_lock_irqsave(&vif->response_lock, flags);
make_tx_response(vif, &pending_tx_info->req, status);
-
- /* Setting any number other than
- * INVALID_PENDING_RING_IDX indicates this slot is
- * starting a new packet / ending a previous packet.
- */
- pending_tx_info->head = 0;
-
- index = pending_index(vif->pending_prod++);
- vif->pending_ring[index] = vif->pending_ring[info_idx];
-
- peek = vif->pending_ring[pending_index(++head)];
-
- } while (!pending_tx_is_head(vif, peek));
-
- put_page(vif->mmap_pages[pending_idx]);
- vif->mmap_pages[pending_idx] = NULL;
+ index = pending_index(vif->pending_prod);
+ vif->pending_ring[index] = pending_idx;
+ /* TX shouldn't use the index before we give it back here */
+ mb();
+ vif->pending_prod++;
+ spin_unlock_irqrestore(&vif->response_lock, flags);
}
void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
From: Zoltan Kiss <[email protected]>
Date: Wed, 8 Jan 2014 00:10:10 +0000
> This patch contains the new definitions necessary for grant mapping.
>
> v2:
> - move unmapping to separate thread. The NAPI instance has to be scheduled
> even from thread context, which can cause huge delays
> - that causes unfortunately bigger struct xenvif
> - store grant handle after checking validity
>
> v3:
> - fix comment in xenvif_tx_dealloc_action()
> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
> unnecessary m2p_override. Also remove pages_to_[un]map members
> - BUG() if grant_tx_handle corrupted
>
> Signed-off-by: Zoltan Kiss <[email protected]>
>
> ---
> drivers/net/xen-netback/common.h | 25 ++++++
> drivers/net/xen-netback/interface.c | 1 +
> drivers/net/xen-netback/netback.c | 163 +++++++++++++++++++++++++++++++++++
> 3 files changed, 189 insertions(+)
>
> diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
> index d218ccd..f1071e3 100644
> --- a/drivers/net/xen-netback/common.h
> +++ b/drivers/net/xen-netback/common.h
> @@ -79,6 +79,11 @@ struct pending_tx_info {
> * if it is head of one or more tx
> * reqs
> */
> + /* callback data for released SKBs. The callback is always
> + * xenvif_zerocopy_callback, ctx points to the next fragment, desc
> + * contains the pending_idx
> + */
> + struct ubuf_info callback_struct;
> };
>
> #define XEN_NETIF_TX_RING_SIZE __CONST_RING_SIZE(xen_netif_tx, PAGE_SIZE)
> @@ -108,6 +113,8 @@ struct xenvif_rx_meta {
> */
> #define MAX_GRANT_COPY_OPS (MAX_SKB_FRAGS * XEN_NETIF_RX_RING_SIZE)
>
> +#define NETBACK_INVALID_HANDLE -1
> +
> struct xenvif {
> /* Unique identifier for this interface. */
> domid_t domid;
> @@ -126,13 +133,23 @@ struct xenvif {
> pending_ring_idx_t pending_cons;
> u16 pending_ring[MAX_PENDING_REQS];
> struct pending_tx_info pending_tx_info[MAX_PENDING_REQS];
> + grant_handle_t grant_tx_handle[MAX_PENDING_REQS];
>
> /* Coalescing tx requests before copying makes number of grant
> * copy ops greater or equal to number of slots required. In
> * worst case a tx request consumes 2 gnttab_copy.
> */
> struct gnttab_copy tx_copy_ops[2*MAX_PENDING_REQS];
> + struct gnttab_map_grant_ref tx_map_ops[MAX_PENDING_REQS];
> + struct gnttab_unmap_grant_ref tx_unmap_ops[MAX_PENDING_REQS];
>
> + spinlock_t dealloc_lock;
> + spinlock_t response_lock;
> + pending_ring_idx_t dealloc_prod;
> + pending_ring_idx_t dealloc_cons;
> + u16 dealloc_ring[MAX_PENDING_REQS];
> + struct task_struct *dealloc_task;
> + wait_queue_head_t dealloc_wq;
>
> /* Use kthread for guest RX */
> struct task_struct *task;
> @@ -221,6 +238,8 @@ int xenvif_tx_action(struct xenvif *vif, int budget);
> int xenvif_kthread(void *data);
> void xenvif_kick_thread(struct xenvif *vif);
>
> +int xenvif_dealloc_kthread(void *data);
> +
> /* Determine whether the needed number of slots (req) are available,
> * and set req_event if not.
> */
> @@ -228,6 +247,12 @@ bool xenvif_rx_ring_slots_available(struct xenvif *vif, int needed);
>
> void xenvif_stop_queue(struct xenvif *vif);
>
> +/* Callback from stack when TX packet can be released */
> +void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success);
> +
> +/* Unmap a pending page, usually has to be called before xenvif_idx_release */
> +void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx);
> +
> extern bool separate_tx_rx_irq;
>
> #endif /* __XEN_NETBACK__COMMON_H__ */
> diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
> index 8d6def2..7170f97 100644
> --- a/drivers/net/xen-netback/interface.c
> +++ b/drivers/net/xen-netback/interface.c
> @@ -37,6 +37,7 @@
>
> #include <xen/events.h>
> #include <asm/xen/hypercall.h>
> +#include <xen/balloon.h>
>
> #define XENVIF_QUEUE_LENGTH 32
> #define XENVIF_NAPI_WEIGHT 64
> diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
> index addfe1d1..7c241f9 100644
> --- a/drivers/net/xen-netback/netback.c
> +++ b/drivers/net/xen-netback/netback.c
> @@ -771,6 +771,19 @@ static struct page *xenvif_alloc_page(struct xenvif *vif,
> return page;
> }
>
> +static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
> + struct xen_netif_tx_request *txp,
> + struct gnttab_map_grant_ref *gop)
> +{
> + gnttab_set_map_op(gop, idx_to_kaddr(vif, pending_idx),
> + GNTMAP_host_map | GNTMAP_readonly,
> + txp->gref, vif->domid);
> +
> + memcpy(&vif->pending_tx_info[pending_idx].req, txp,
> + sizeof(*txp));
> +
> +}
> +
> static struct gnttab_copy *xenvif_get_requests(struct xenvif *vif,
> struct sk_buff *skb,
> struct xen_netif_tx_request *txp,
> @@ -1599,6 +1612,105 @@ static int xenvif_tx_submit(struct xenvif *vif)
> return work_done;
> }
>
> +void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success)
> +{
> + unsigned long flags;
> + pending_ring_idx_t index;
> + u16 pending_idx = ubuf->desc;
> + struct pending_tx_info *temp =
> + container_of(ubuf, struct pending_tx_info, callback_struct);
> + struct xenvif *vif =
> + container_of(temp - pending_idx, struct xenvif,
> + pending_tx_info[0]);
> +
> + spin_lock_irqsave(&vif->dealloc_lock, flags);
> + do {
> + pending_idx = ubuf->desc;
> + ubuf = (struct ubuf_info *) ubuf->ctx;
> + index = pending_index(vif->dealloc_prod);
> + vif->dealloc_ring[index] = pending_idx;
> + /* Sync with xenvif_tx_action_dealloc:
> + * insert idx then incr producer.
> + */
> + smp_wmb();
> + vif->dealloc_prod++;
> + } while (ubuf);
> + wake_up(&vif->dealloc_wq);
> + spin_unlock_irqrestore(&vif->dealloc_lock, flags);
> +}
> +
> +static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
> +{
> + struct gnttab_unmap_grant_ref *gop;
> + pending_ring_idx_t dc, dp;
> + u16 pending_idx, pending_idx_release[MAX_PENDING_REQS];
> + unsigned int i = 0;
> +
> + dc = vif->dealloc_cons;
> + gop = vif->tx_unmap_ops;
> +
> + /* Free up any grants we have finished using */
> + do {
> + dp = vif->dealloc_prod;
> +
> + /* Ensure we see all indices enqueued by all
> + * xenvif_zerocopy_callback().
> + */
> + smp_rmb();
> +
> + while (dc != dp) {
> + pending_idx =
> + vif->dealloc_ring[pending_index(dc++)];
> +
> + /* Already unmapped? */
> + if (vif->grant_tx_handle[pending_idx] ==
> + NETBACK_INVALID_HANDLE) {
> + netdev_err(vif->dev,
> + "Trying to unmap invalid handle! "
> + "pending_idx: %x\n", pending_idx);
> + continue;
> + }
> +
> + pending_idx_release[gop-vif->tx_unmap_ops] =
> + pending_idx;
> + gnttab_set_unmap_op(gop,
> + idx_to_kaddr(vif, pending_idx),
> + GNTMAP_host_map,
> + vif->grant_tx_handle[pending_idx]);
> + vif->grant_tx_handle[pending_idx] =
> + NETBACK_INVALID_HANDLE;
> + ++gop;
> + }
> +
> + } while (dp != vif->dealloc_prod);
> +
> + vif->dealloc_cons = dc;
> +
> + if (gop - vif->tx_unmap_ops > 0) {
> + int ret;
> + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref,
> + vif->tx_unmap_ops,
> + gop - vif->tx_unmap_ops);
> + if (ret) {
> + netdev_err(vif->dev, "Unmap fail: nr_ops %x ret %d\n",
> + gop - vif->tx_unmap_ops, ret);
> + for (i = 0; i < gop - vif->tx_unmap_ops; ++i) {
> + netdev_err(vif->dev,
> + " host_addr: %llx handle: %x status: %d\n",
> + gop[i].host_addr,
> + gop[i].handle,
> + gop[i].status);
> + }
> + BUG();
> + }
> + }
> +
> + for (i = 0; i < gop - vif->tx_unmap_ops; ++i)
> + xenvif_idx_release(vif, pending_idx_release[i],
> + XEN_NETIF_RSP_OKAY);
> +}
> +
> +
> /* Called after netfront has transmitted */
> int xenvif_tx_action(struct xenvif *vif, int budget)
> {
> @@ -1665,6 +1777,27 @@ static void xenvif_idx_release(struct xenvif *vif, u16 pending_idx,
> vif->mmap_pages[pending_idx] = NULL;
> }
>
> +void xenvif_idx_unmap(struct xenvif *vif, u16 pending_idx)
> +{
> + int ret;
> + struct gnttab_unmap_grant_ref tx_unmap_op;
> +
> + if (vif->grant_tx_handle[pending_idx] == NETBACK_INVALID_HANDLE) {
> + netdev_err(vif->dev,
> + "Trying to unmap invalid handle! pending_idx: %x\n",
> + pending_idx);
> + return;
> + }
> + gnttab_set_unmap_op(&tx_unmap_op,
> + idx_to_kaddr(vif, pending_idx),
> + GNTMAP_host_map,
> + vif->grant_tx_handle[pending_idx]);
> + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref,
> + &tx_unmap_op,
> + 1);
> + BUG_ON(ret);
> + vif->grant_tx_handle[pending_idx] = NETBACK_INVALID_HANDLE;
> +}
>
> static void make_tx_response(struct xenvif *vif,
> struct xen_netif_tx_request *txp,
> @@ -1726,6 +1859,14 @@ static inline int tx_work_todo(struct xenvif *vif)
> return 0;
> }
>
> +static inline int tx_dealloc_work_todo(struct xenvif *vif)
Make this return bool.
> + return 1;
return true;
> + return 0;
return false;
> + wait_event_interruptible(vif->dealloc_wq,
> + tx_dealloc_work_todo(vif) ||
> + kthread_should_stop());
Inconsistent indentation. You should make the arguments line up at
exactly the first column after the openning parenthesis of the function
call.
On Wed, 2014-01-08 at 00:10 +0000, Zoltan Kiss wrote:
>
> + if (skb_shinfo(skb)->frag_list) {
> + nskb = skb_shinfo(skb)->frag_list;
> + xenvif_fill_frags(vif, nskb, INVALID_PENDING_IDX);
> + skb->len += nskb->len;
> + skb->data_len += nskb->len;
> + skb->truesize += nskb->truesize;
> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> + skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> + vif->tx_zerocopy_sent += 2;
> + nskb = skb;
> +
> + skb = skb_copy_expand(skb,
> + 0,
> + 0,
> + GFP_ATOMIC | __GFP_NOWARN);
skb can be NULL here
> + skb_shinfo(skb)->destructor_arg = NULL;
> + }
> if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
> int target = min_t(int, skb->len, PKT_PROT_LEN);
> __pskb_pull_tail(skb, target - skb_headlen(skb));
> @@ -1568,6 +1660,9 @@ static int xenvif_tx_submit(struct xenvif *vif)
> }
>
On 08/01/14 02:12, Eric Dumazet wrote:
> On Wed, 2014-01-08 at 00:10 +0000, Zoltan Kiss wrote:
>
>>
>> + if (skb_shinfo(skb)->frag_list) {
>> + nskb = skb_shinfo(skb)->frag_list;
>> + xenvif_fill_frags(vif, nskb, INVALID_PENDING_IDX);
>> + skb->len += nskb->len;
>> + skb->data_len += nskb->len;
>> + skb->truesize += nskb->truesize;
>> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
>> + skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
>> + vif->tx_zerocopy_sent += 2;
>> + nskb = skb;
>> +
>> + skb = skb_copy_expand(skb,
>> + 0,
>> + 0,
>> + GFP_ATOMIC | __GFP_NOWARN);
>
> skb can be NULL here
Thanks, fixed that.
Zoli
On Wed, 2014-01-08 at 13:49 +0000, Zoltan Kiss wrote:
> On 08/01/14 02:12, Eric Dumazet wrote:
> > On Wed, 2014-01-08 at 00:10 +0000, Zoltan Kiss wrote:
> >
> >>
> >> + if (skb_shinfo(skb)->frag_list) {
> >> + nskb = skb_shinfo(skb)->frag_list;
> >> + xenvif_fill_frags(vif, nskb, INVALID_PENDING_IDX);
> >> + skb->len += nskb->len;
> >> + skb->data_len += nskb->len;
> >> + skb->truesize += nskb->truesize;
> >> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> >> + skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> >> + vif->tx_zerocopy_sent += 2;
> >> + nskb = skb;
> >> +
> >> + skb = skb_copy_expand(skb,
> >> + 0,
> >> + 0,
> >> + GFP_ATOMIC | __GFP_NOWARN);
> >
> > skb can be NULL here
>
> Thanks, fixed that.
BTW, I am not sure why you copy the skb.
Is it to get rid of frag_list, and why ?
On 08/01/14 01:29, David Miller wrote:
>> +static inline int tx_dealloc_work_todo(struct xenvif *vif)
>
> Make this return bool.
Done, also in the last patch.
>> + wait_event_interruptible(vif->dealloc_wq,
>> + tx_dealloc_work_todo(vif) ||
>> + kthread_should_stop());
>
> Inconsistent indentation. You should make the arguments line up at
> exactly the first column after the openning parenthesis of the function
> call.
Done, thanks.
Zoli
On 08/01/14 13:54, Eric Dumazet wrote:
> On Wed, 2014-01-08 at 13:49 +0000, Zoltan Kiss wrote:
>> On 08/01/14 02:12, Eric Dumazet wrote:
>>> On Wed, 2014-01-08 at 00:10 +0000, Zoltan Kiss wrote:
>>>
>>>>
>>>> + if (skb_shinfo(skb)->frag_list) {
>>>> + nskb = skb_shinfo(skb)->frag_list;
>>>> + xenvif_fill_frags(vif, nskb, INVALID_PENDING_IDX);
>>>> + skb->len += nskb->len;
>>>> + skb->data_len += nskb->len;
>>>> + skb->truesize += nskb->truesize;
>>>> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
>>>> + skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
>>>> + vif->tx_zerocopy_sent += 2;
>>>> + nskb = skb;
>>>> +
>>>> + skb = skb_copy_expand(skb,
>>>> + 0,
>>>> + 0,
>>>> + GFP_ATOMIC | __GFP_NOWARN);
>>>
>>> skb can be NULL here
>>
>> Thanks, fixed that.
>
> BTW, I am not sure why you copy the skb.
>
> Is it to get rid of frag_list, and why ?
Yes, it is to get rid of the frag_list, just to be on the safe side. I'm
not sure if it is normal to send a big skb with MAX_SKB_FRAGS frags plus
an empty skb on the frag_list with one frag, so I just consolidate them
here. This scenario shouldn't happen very often anyway, even guests
which can send more than MAX_SKB_FRAGS slots tends to do it rarely.
Zoli
You once mentioned that you have a trick to avoid touching TLB, is it in
this series?
(Haven't really looked at this series as I'm in today. Will have a
closer look tonight. I'm just curious now.)
Wei.
On 08/01/14 14:43, Wei Liu wrote:
> You once mentioned that you have a trick to avoid touching TLB, is it in
> this series?
>
> (Haven't really looked at this series as I'm in today. Will have a
> closer look tonight. I'm just curious now.)
>
> Wei.
>
No, I'm currently working on that, it will be a separate series, as it
also needs some Xen modifications which haven't reached upstream yet AFAIK.
Zoli
I just realized when answering Ma's mail that this doesn't cause the
desired effect after Paul's flow control improvement: starting the queue
doesn't drop the packets which cannot fit the ring. Which in fact might
be not good. We are adding the skb to vif->rx_queue even when
xenvif_rx_ring_slots_available(vif, min_slots_needed) said there is no
space for that. Or am I missing something? Paul?
Zoli
On 08/01/14 00:10, Zoltan Kiss wrote:
> A malicious or buggy guest can leave its queue filled indefinitely, in which
> case qdisc start to queue packets for that VIF. If those packets came from an
> another guest, it can block its slots and prevent shutdown. To avoid that, we
> make sure the queue is drained in every 10 seconds.
...
> diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
> index 95fcd63..ce032f9 100644
> --- a/drivers/net/xen-netback/interface.c
> +++ b/drivers/net/xen-netback/interface.c
> @@ -114,6 +114,16 @@ static irqreturn_t xenvif_interrupt(int irq, void *dev_id)
> return IRQ_HANDLED;
> }
>
> +static void xenvif_wake_queue(unsigned long data)
> +{
> + struct xenvif *vif = (struct xenvif *)data;
> +
> + if (netif_queue_stopped(vif->dev)) {
> + netdev_err(vif->dev, "draining TX queue\n");
> + netif_wake_queue(vif->dev);
> + }
> +}
> +
> static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
> {
> struct xenvif *vif = netdev_priv(dev);
> @@ -143,8 +153,13 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
> * then turn off the queue to give the ring a chance to
> * drain.
> */
> - if (!xenvif_rx_ring_slots_available(vif, min_slots_needed))
> + if (!xenvif_rx_ring_slots_available(vif, min_slots_needed)) {
> + vif->wake_queue.function = xenvif_wake_queue;
> + vif->wake_queue.data = (unsigned long)vif;
> xenvif_stop_queue(vif);
> + mod_timer(&vif->wake_queue,
> + jiffies + rx_drain_timeout_jiffies);
> + }
>
> skb_queue_tail(&vif->rx_queue, skb);
> xenvif_kick_thread(vif);
> -----Original Message-----
> From: Zoltan Kiss
> Sent: 08 January 2014 21:34
> To: Ian Campbell; Wei Liu; [email protected];
> [email protected]; [email protected]; Jonathan Davies
> Cc: Zoltan Kiss; Paul Durrant
> Subject: Re: [PATCH net-next v3 8/9] xen-netback: Timeout packets in RX
> path
>
> I just realized when answering Ma's mail that this doesn't cause the
> desired effect after Paul's flow control improvement: starting the queue
> doesn't drop the packets which cannot fit the ring. Which in fact might
> be not good.
No, that would not be good.
> We are adding the skb to vif->rx_queue even when
> xenvif_rx_ring_slots_available(vif, min_slots_needed) said there is no
> space for that. Or am I missing something? Paul?
>
That's correct. Part of the flow control improvement was to get rid of needless packet drops. For your purposes, you basically need to avoid using the queuing discipline and take packets into netback's vif->rx_queue regardless of the state of the shared ring so that you can drop them if they get beyond a certain age. So, perhaps you should never stop the netif queue, place an upper limit on vif->rx_queue (either packet or byte count) and drop when that is exceeded (i.e. mimicking pfifo or bfifo internally).
Paul
> Zoli
>
> On 08/01/14 00:10, Zoltan Kiss wrote:
> > A malicious or buggy guest can leave its queue filled indefinitely, in which
> > case qdisc start to queue packets for that VIF. If those packets came from
> an
> > another guest, it can block its slots and prevent shutdown. To avoid that,
> we
> > make sure the queue is drained in every 10 seconds.
> ...
> > diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-
> netback/interface.c
> > index 95fcd63..ce032f9 100644
> > --- a/drivers/net/xen-netback/interface.c
> > +++ b/drivers/net/xen-netback/interface.c
> > @@ -114,6 +114,16 @@ static irqreturn_t xenvif_interrupt(int irq, void
> *dev_id)
> > return IRQ_HANDLED;
> > }
> >
> > +static void xenvif_wake_queue(unsigned long data)
> > +{
> > + struct xenvif *vif = (struct xenvif *)data;
> > +
> > + if (netif_queue_stopped(vif->dev)) {
> > + netdev_err(vif->dev, "draining TX queue\n");
> > + netif_wake_queue(vif->dev);
> > + }
> > +}
> > +
> > static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
> > {
> > struct xenvif *vif = netdev_priv(dev);
> > @@ -143,8 +153,13 @@ static int xenvif_start_xmit(struct sk_buff *skb,
> struct net_device *dev)
> > * then turn off the queue to give the ring a chance to
> > * drain.
> > */
> > - if (!xenvif_rx_ring_slots_available(vif, min_slots_needed))
> > + if (!xenvif_rx_ring_slots_available(vif, min_slots_needed)) {
> > + vif->wake_queue.function = xenvif_wake_queue;
> > + vif->wake_queue.data = (unsigned long)vif;
> > xenvif_stop_queue(vif);
> > + mod_timer(&vif->wake_queue,
> > + jiffies + rx_drain_timeout_jiffies);
> > + }
> >
> > skb_queue_tail(&vif->rx_queue, skb);
> > xenvif_kick_thread(vif);
On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
> This patch contains the new definitions necessary for grant mapping.
>
> v2:
> - move unmapping to separate thread. The NAPI instance has to be scheduled
> even from thread context, which can cause huge delays
> - that causes unfortunately bigger struct xenvif
> - store grant handle after checking validity
>
> v3:
> - fix comment in xenvif_tx_dealloc_action()
> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
> unnecessary m2p_override. Also remove pages_to_[un]map members
Is it worthy to have another function call
gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
parameter to control wether we need to touch m2p_override? I *think* it
will benefit block driver as well?
(CC Roger and David for input)
> - BUG() if grant_tx_handle corrupted
>
> Signed-off-by: Zoltan Kiss <[email protected]>
>
> ---
[...]
>
> #define XENVIF_QUEUE_LENGTH 32
> #define XENVIF_NAPI_WEIGHT 64
> diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
> index addfe1d1..7c241f9 100644
> --- a/drivers/net/xen-netback/netback.c
> +++ b/drivers/net/xen-netback/netback.c
> @@ -771,6 +771,19 @@ static struct page *xenvif_alloc_page(struct xenvif *vif,
> return page;
> }
>
> +static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
> + struct xen_netif_tx_request *txp,
> + struct gnttab_map_grant_ref *gop)
Indentation.
> +{
> + gnttab_set_map_op(gop, idx_to_kaddr(vif, pending_idx),
> + GNTMAP_host_map | GNTMAP_readonly,
> + txp->gref, vif->domid);
> +
> + memcpy(&vif->pending_tx_info[pending_idx].req, txp,
> + sizeof(*txp));
> +
> +}
> +
[...]
> +void xenvif_zerocopy_callback(struct ubuf_info *ubuf, bool zerocopy_success)
> +{
> + unsigned long flags;
> + pending_ring_idx_t index;
> + u16 pending_idx = ubuf->desc;
> + struct pending_tx_info *temp =
> + container_of(ubuf, struct pending_tx_info, callback_struct);
> + struct xenvif *vif =
> + container_of(temp - pending_idx, struct xenvif,
> + pending_tx_info[0]);
Indentation.
> +
> + spin_lock_irqsave(&vif->dealloc_lock, flags);
> + do {
> + pending_idx = ubuf->desc;
> + ubuf = (struct ubuf_info *) ubuf->ctx;
> + index = pending_index(vif->dealloc_prod);
> + vif->dealloc_ring[index] = pending_idx;
> + /* Sync with xenvif_tx_action_dealloc:
xenvif_tx_dealloc_action I suppose.
> + * insert idx then incr producer.
> + */
> + smp_wmb();
> + vif->dealloc_prod++;
> + } while (ubuf);
> + wake_up(&vif->dealloc_wq);
> + spin_unlock_irqrestore(&vif->dealloc_lock, flags);
> +}
> +
> +static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
> +{
> + struct gnttab_unmap_grant_ref *gop;
> + pending_ring_idx_t dc, dp;
> + u16 pending_idx, pending_idx_release[MAX_PENDING_REQS];
> + unsigned int i = 0;
> +
> + dc = vif->dealloc_cons;
> + gop = vif->tx_unmap_ops;
> +
> + /* Free up any grants we have finished using */
> + do {
> + dp = vif->dealloc_prod;
> +
> + /* Ensure we see all indices enqueued by all
> + * xenvif_zerocopy_callback().
> + */
> + smp_rmb();
> +
> + while (dc != dp) {
> + pending_idx =
> + vif->dealloc_ring[pending_index(dc++)];
> +
> + /* Already unmapped? */
> + if (vif->grant_tx_handle[pending_idx] ==
> + NETBACK_INVALID_HANDLE) {
> + netdev_err(vif->dev,
> + "Trying to unmap invalid handle! "
> + "pending_idx: %x\n", pending_idx);
> + continue;
You seemed to miss the BUG_ON we discussed?
See thread starting <[email protected]>.
Wei.
> + }
> +
> + pending_idx_release[gop-vif->tx_unmap_ops] =
> + pending_idx;
> + gnttab_set_unmap_op(gop,
> + idx_to_kaddr(vif, pending_idx),
> + GNTMAP_host_map,
> + vif->grant_tx_handle[pending_idx]);
> + vif->grant_tx_handle[pending_idx] =
> + NETBACK_INVALID_HANDLE;
> + ++gop;
> + }
> +
On Wed, Jan 08, 2014 at 12:10:11AM +0000, Zoltan Kiss wrote:
> This patch changes the grant copy on the TX patch to grant mapping
>
> v2:
> - delete branch for handling fragmented packets fit PKT_PROT_LEN sized first
> request
> - mark the effect of using ballooned pages in a comment
> - place setting of skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY right
> before netif_receive_skb, and mark the importance of it
> - grab dealloc_lock before __napi_complete to avoid contention with the
> callback's napi_schedule
> - handle fragmented packets where first request < PKT_PROT_LEN
> - fix up error path when checksum_setup failed
> - check before teardown for pending grants, and start complain if they are
> there after 10 second
>
> v3:
> - delete a surplus checking from tx_action
> - remove stray line
> - squash xenvif_idx_unmap changes into the first patch
> - init spinlocks
> - call map hypercall directly instead of gnttab_map_refs()
I suppose this is to avoid touching m2p override as well, just as
previous patch uses unmap hypercall directly.
> - fix unmapping timeout in xenvif_free()
>
> Signed-off-by: Zoltan Kiss <[email protected]>
> ---
> drivers/net/xen-netback/interface.c | 57 +++++++-
> drivers/net/xen-netback/netback.c | 251 ++++++++++++++---------------------
> 2 files changed, 153 insertions(+), 155 deletions(-)
>
> diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
> index 7170f97..3b2b249 100644
> --- a/drivers/net/xen-netback/interface.c
> +++ b/drivers/net/xen-netback/interface.c
> @@ -122,7 +122,9 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
> BUG_ON(skb->dev != dev);
>
> /* Drop the packet if vif is not ready */
> - if (vif->task == NULL || !xenvif_schedulable(vif))
> + if (vif->task == NULL ||
> + vif->dealloc_task == NULL ||
> + !xenvif_schedulable(vif))
Indentation.
> goto drop;
>
> /* At best we'll need one slot for the header and one for each
> @@ -345,8 +347,26 @@ struct xenvif *xenvif_alloc(struct device *parent, domid_t domid,
> vif->pending_prod = MAX_PENDING_REQS;
> for (i = 0; i < MAX_PENDING_REQS; i++)
> vif->pending_ring[i] = i;
> - for (i = 0; i < MAX_PENDING_REQS; i++)
> - vif->mmap_pages[i] = NULL;
> + spin_lock_init(&vif->dealloc_lock);
> + spin_lock_init(&vif->response_lock);
> + /* If ballooning is disabled, this will consume real memory, so you
> + * better enable it. The long term solution would be to use just a
> + * bunch of valid page descriptors, without dependency on ballooning
> + */
> + err = alloc_xenballooned_pages(MAX_PENDING_REQS,
> + vif->mmap_pages,
> + false);
Ditto.
> + if (err) {
> + netdev_err(dev, "Could not reserve mmap_pages\n");
> + return NULL;
> + }
> + for (i = 0; i < MAX_PENDING_REQS; i++) {
> + vif->pending_tx_info[i].callback_struct = (struct ubuf_info)
> + { .callback = xenvif_zerocopy_callback,
> + .ctx = NULL,
> + .desc = i };
> + vif->grant_tx_handle[i] = NETBACK_INVALID_HANDLE;
> + }
>
> /*
> * Initialise a dummy MAC address. We choose the numerically
> @@ -390,6 +410,7 @@ int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
> goto err;
>
> init_waitqueue_head(&vif->wq);
> + init_waitqueue_head(&vif->dealloc_wq);
>
> if (tx_evtchn == rx_evtchn) {
> /* feature-split-event-channels == 0 */
> @@ -431,6 +452,14 @@ int xenvif_connect(struct xenvif *vif, unsigned long tx_ring_ref,
> goto err_rx_unbind;
> }
>
> + vif->dealloc_task = kthread_create(xenvif_dealloc_kthread,
> + (void *)vif, "%s-dealloc", vif->dev->name);
Ditto.
> + if (IS_ERR(vif->dealloc_task)) {
> + pr_warn("Could not allocate kthread for %s\n", vif->dev->name);
> + err = PTR_ERR(vif->dealloc_task);
> + goto err_rx_unbind;
> + }
> +
> vif->task = task;
>
> rtnl_lock();
[...]
>
> static int xenvif_tx_check_gop(struct xenvif *vif,
> struct sk_buff *skb,
> - struct gnttab_copy **gopp)
> + struct gnttab_map_grant_ref **gopp)
> {
> - struct gnttab_copy *gop = *gopp;
> + struct gnttab_map_grant_ref *gop = *gopp;
> u16 pending_idx = *((u16 *)skb->data);
> struct skb_shared_info *shinfo = skb_shinfo(skb);
> struct pending_tx_info *tx_info;
> @@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
> err = gop->status;
> if (unlikely(err))
> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
> + else {
> + if (vif->grant_tx_handle[pending_idx] !=
> + NETBACK_INVALID_HANDLE) {
> + netdev_err(vif->dev,
> + "Stale mapped handle! pending_idx %x handle %x\n",
> + pending_idx, vif->grant_tx_handle[pending_idx]);
> + BUG();
> + }
> + set_phys_to_machine(idx_to_pfn(vif, pending_idx),
> + FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
What happens when you don't have this?
> + vif->grant_tx_handle[pending_idx] = gop->handle;
> + }
>
> /* Skip first skb fragment if it is on same page as header fragment. */
> start = (frag_get_pending_idx(&shinfo->frags[0]) == pending_idx);
> @@ -933,18 +877,26 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
> head = tx_info->head;
>
[...]
> }
> @@ -1567,7 +1523,11 @@ static int xenvif_tx_submit(struct xenvif *vif)
> else if (txp->flags & XEN_NETTXF_data_validated)
> skb->ip_summed = CHECKSUM_UNNECESSARY;
>
> - xenvif_fill_frags(vif, skb);
> + xenvif_fill_frags(vif,
> + skb,
> + skb_shinfo(skb)->destructor_arg ?
> + pending_idx :
> + INVALID_PENDING_IDX);
>
Indentation.
> if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
> int target = min_t(int, skb->len, PKT_PROT_LEN);
> @@ -1581,6 +1541,8 @@ static int xenvif_tx_submit(struct xenvif *vif)
> if (checksum_setup(vif, skb)) {
> netdev_dbg(vif->dev,
> "Can't setup checksum in net_tx_action\n");
> + if (skb_shinfo(skb)->destructor_arg)
> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
Do you still care setting the flag even if this skb is not going to be
delivered? If so can you state clearly the reason just like the
following hunk?
> kfree_skb(skb);
> continue;
> }
> @@ -1606,6 +1568,14 @@ static int xenvif_tx_submit(struct xenvif *vif)
>
> work_done++;
>
> + /* Set this flag right before netif_receive_skb, otherwise
> + * someone might think this packet already left netback, and
> + * do a skb_copy_ubufs while we are still in control of the
> + * skb. E.g. the __pskb_pull_tail earlier can do such thing.
> + */
> + if (skb_shinfo(skb)->destructor_arg)
> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> +
> netif_receive_skb(skb);
> }
>
> @@ -1715,7 +1685,7 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
> int xenvif_tx_action(struct xenvif *vif, int budget)
> {
> unsigned nr_gops;
> - int work_done;
> + int work_done, ret;
>
> if (unlikely(!tx_work_todo(vif)))
> return 0;
> @@ -1725,7 +1695,10 @@ int xenvif_tx_action(struct xenvif *vif, int budget)
> if (nr_gops == 0)
> return 0;
>
> - gnttab_batch_copy(vif->tx_copy_ops, nr_gops);
> + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
> + vif->tx_map_ops,
> + nr_gops);
Why do you need to replace gnttab_batch_copy with hypercall? In the
ideal situation gnttab_batch_copy should behave the same as directly
hypercall but it also handles GNTST_eagain for you.
Wei.
On 09/01/14 15:30, Wei Liu wrote:
> On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
>> This patch contains the new definitions necessary for grant mapping.
>>
>> v2:
>> - move unmapping to separate thread. The NAPI instance has to be scheduled
>> even from thread context, which can cause huge delays
>> - that causes unfortunately bigger struct xenvif
>> - store grant handle after checking validity
>>
>> v3:
>> - fix comment in xenvif_tx_dealloc_action()
>> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
>> unnecessary m2p_override. Also remove pages_to_[un]map members
>
> Is it worthy to have another function call
> gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
> parameter to control wether we need to touch m2p_override? I *think* it
> will benefit block driver as well?
add_m2p_override and remove_m2p_override calls should be moved into the
gntdev device as that should be the only user.
David
On 09/01/14 16:30, Wei Liu wrote:
> On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
>> This patch contains the new definitions necessary for grant mapping.
>>
>> v2:
>> - move unmapping to separate thread. The NAPI instance has to be scheduled
>> even from thread context, which can cause huge delays
>> - that causes unfortunately bigger struct xenvif
>> - store grant handle after checking validity
>>
>> v3:
>> - fix comment in xenvif_tx_dealloc_action()
>> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
>> unnecessary m2p_override. Also remove pages_to_[un]map members
>
> Is it worthy to have another function call
> gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
> parameter to control wether we need to touch m2p_override? I *think* it
> will benefit block driver as well?
Anthony Liguori posted a patch to perform something similar in blkback,
but I think the patch never made it upstream:
https://lkml.org/lkml/2013/11/12/749
Probably a good time to revisit it so this mechanism can be used by both
blkback and netback?
Roger.
On Thu, 9 Jan 2014, David Vrabel wrote:
> On 09/01/14 15:30, Wei Liu wrote:
> > On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
> >> This patch contains the new definitions necessary for grant mapping.
> >>
> >> v2:
> >> - move unmapping to separate thread. The NAPI instance has to be scheduled
> >> even from thread context, which can cause huge delays
> >> - that causes unfortunately bigger struct xenvif
> >> - store grant handle after checking validity
> >>
> >> v3:
> >> - fix comment in xenvif_tx_dealloc_action()
> >> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
> >> unnecessary m2p_override. Also remove pages_to_[un]map members
> >
> > Is it worthy to have another function call
> > gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
> > parameter to control wether we need to touch m2p_override? I *think* it
> > will benefit block driver as well?
>
> add_m2p_override and remove_m2p_override calls should be moved into the
> gntdev device as that should be the only user.
First of all the gntdev device is common code, while the m2p_override is
an x86 concept.
Then I would like to point out that there are no guarantees that a
network driver, or any other kernel subsystems, don't come to rely on
mfn_to_pfn translations for any reasons at any time.
It just happens that today the only known user is gupf, but tomorrow,
who knows?
If we move the m2p_override calls to the gntdev device somehow (avoif
ifdefs please), we should be very well aware of the risks involved.
Of course my practical self realizes that we don't want a performance
regression and this is the quickest way to fix it, so I am not
completely oppose to it.
On 09/01/14 17:28, Stefano Stabellini wrote:
> On Thu, 9 Jan 2014, David Vrabel wrote:
>> On 09/01/14 15:30, Wei Liu wrote:
>>> On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
>>>> This patch contains the new definitions necessary for grant mapping.
>>>>
>>>> v2:
>>>> - move unmapping to separate thread. The NAPI instance has to be scheduled
>>>> even from thread context, which can cause huge delays
>>>> - that causes unfortunately bigger struct xenvif
>>>> - store grant handle after checking validity
>>>>
>>>> v3:
>>>> - fix comment in xenvif_tx_dealloc_action()
>>>> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
>>>> unnecessary m2p_override. Also remove pages_to_[un]map members
>>>
>>> Is it worthy to have another function call
>>> gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
>>> parameter to control wether we need to touch m2p_override? I *think* it
>>> will benefit block driver as well?
>>
>> add_m2p_override and remove_m2p_override calls should be moved into the
>> gntdev device as that should be the only user.
>
> First of all the gntdev device is common code, while the m2p_override is
> an x86 concept.
m2p_add_override() and m2p_remove_override() are already called from
common code and ARM already provides inline stubs.
The m2p override mechanism is also broken by design (local PFN to
foreign MFN may be many-to-one, but the m2p override only works if local
PFN to foreign MFN is one-to-one). So I want the m2p override to be only
used where it is /currently/ necessary. I think there should be no new
users of it nor should it be considered a fix for any other use case.
David
On Thu, 9 Jan 2014, David Vrabel wrote:
> On 09/01/14 17:28, Stefano Stabellini wrote:
> > On Thu, 9 Jan 2014, David Vrabel wrote:
> >> On 09/01/14 15:30, Wei Liu wrote:
> >>> On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
> >>>> This patch contains the new definitions necessary for grant mapping.
> >>>>
> >>>> v2:
> >>>> - move unmapping to separate thread. The NAPI instance has to be scheduled
> >>>> even from thread context, which can cause huge delays
> >>>> - that causes unfortunately bigger struct xenvif
> >>>> - store grant handle after checking validity
> >>>>
> >>>> v3:
> >>>> - fix comment in xenvif_tx_dealloc_action()
> >>>> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
> >>>> unnecessary m2p_override. Also remove pages_to_[un]map members
> >>>
> >>> Is it worthy to have another function call
> >>> gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
> >>> parameter to control wether we need to touch m2p_override? I *think* it
> >>> will benefit block driver as well?
> >>
> >> add_m2p_override and remove_m2p_override calls should be moved into the
> >> gntdev device as that should be the only user.
> >
> > First of all the gntdev device is common code, while the m2p_override is
> > an x86 concept.
>
> m2p_add_override() and m2p_remove_override() are already called from
> common code and ARM already provides inline stubs.
This is the right time to fix it, then :)
Maybe we should add the m2p_add_override call to the x86 implementation
of set_phys_to_machine, or maybe we need a new generic
set_machine_to_phys call.
> The m2p override mechanism is also broken by design (local PFN to
> foreign MFN may be many-to-one, but the m2p override only works if local
> PFN to foreign MFN is one-to-one). So I want the m2p override to be only
> used where it is /currently/ necessary. I think there should be no new
> users of it nor should it be considered a fix for any other use case.
I agree, but I think that we have different views on the use case.
To me the m2p_override use case is "everywhere an mfn_to_pfn translation
is required", that unfortunately is potentially everywhere at this time.
I would love to restrict it further but at the very least we would need
something written down under Documentation. Otherwise when the next
Linux hacker comes along with a performance optimization for her new
network driver that breaks Xen because Xen is incapable of doing mfn to
pfn translations, the maintainers might (rightfully) decide that it is
simply our problem.
On 09/01/14 18:09, Stefano Stabellini wrote:
>
> I agree, but I think that we have different views on the use case.
> To me the m2p_override use case is "everywhere an mfn_to_pfn translation
> is required", that unfortunately is potentially everywhere at this time.
mfn_to_pfn() cannot be made to work correctly with foreign MFNs. It's a
fundamentally unsolvable problem.
IMO, the only sensible use of the m2p_override is to cause mfn_to_pfn()
to BUG() if a foreign MFN is used.
David
On 09/01/14 15:30, Wei Liu wrote:
> On Wed, Jan 08, 2014 at 12:10:10AM +0000, Zoltan Kiss wrote:
>> v3:
>> - fix comment in xenvif_tx_dealloc_action()
>> - call unmap hypercall directly instead of gnttab_unmap_refs(), which does
>> unnecessary m2p_override. Also remove pages_to_[un]map members
>
> Is it worthy to have another function call
> gnttab_unmap_refs_no_m2p_override in Xen core driver, or just add a
> parameter to control wether we need to touch m2p_override? I *think* it
> will benefit block driver as well?
>
> (CC Roger and David for input)
Yep, it worth, but let's make it a different patch
>> --- a/drivers/net/xen-netback/netback.c
>> +++ b/drivers/net/xen-netback/netback.c
>> @@ -771,6 +771,19 @@ static struct page *xenvif_alloc_page(struct xenvif *vif,
>> return page;
>> }
>>
>> +static inline void xenvif_tx_create_gop(struct xenvif *vif, u16 pending_idx,
>> + struct xen_netif_tx_request *txp,
>> + struct gnttab_map_grant_ref *gop)
>
> Indentation.
I fixed it and the later ones up, hopefully I haven't missed anything.
>
>> +
>> + spin_lock_irqsave(&vif->dealloc_lock, flags);
>> + do {
>> + pending_idx = ubuf->desc;
>> + ubuf = (struct ubuf_info *) ubuf->ctx;
>> + index = pending_index(vif->dealloc_prod);
>> + vif->dealloc_ring[index] = pending_idx;
>> + /* Sync with xenvif_tx_action_dealloc:
>
> xenvif_tx_dealloc_action I suppose.
Yes.
>> + /* Already unmapped? */
>> + if (vif->grant_tx_handle[pending_idx] ==
>> + NETBACK_INVALID_HANDLE) {
>> + netdev_err(vif->dev,
>> + "Trying to unmap invalid handle! "
>> + "pending_idx: %x\n", pending_idx);
>> + continue;
>
> You seemed to miss the BUG_ON we discussed?
>
> See thread starting <[email protected]>.
Indeed, despite I wrote it in the version history :)
On 09/01/14 15:30, Wei Liu wrote:
> On Wed, Jan 08, 2014 at 12:10:11AM +0000, Zoltan Kiss wrote:
>> v3:
>> - delete a surplus checking from tx_action
>> - remove stray line
>> - squash xenvif_idx_unmap changes into the first patch
>> - init spinlocks
>> - call map hypercall directly instead of gnttab_map_refs()
>
> I suppose this is to avoid touching m2p override as well, just as
> previous patch uses unmap hypercall directly.
Yes.
>> --- a/drivers/net/xen-netback/interface.c
>> +++ b/drivers/net/xen-netback/interface.c
>> @@ -122,7 +122,9 @@ static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev)
>> BUG_ON(skb->dev != dev);
>>
>> /* Drop the packet if vif is not ready */
>> - if (vif->task == NULL || !xenvif_schedulable(vif))
>> + if (vif->task == NULL ||
>> + vif->dealloc_task == NULL ||
>> + !xenvif_schedulable(vif))
>
> Indentation.
Fixed, and the later ones as well
>> @@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
>> err = gop->status;
>> if (unlikely(err))
>> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
>> + else {
>> + if (vif->grant_tx_handle[pending_idx] !=
>> + NETBACK_INVALID_HANDLE) {
>> + netdev_err(vif->dev,
>> + "Stale mapped handle! pending_idx %x handle %x\n",
>> + pending_idx, vif->grant_tx_handle[pending_idx]);
>> + BUG();
>> + }
>> + set_phys_to_machine(idx_to_pfn(vif, pending_idx),
>> + FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
>
> What happens when you don't have this?
Your frags will be filled with garbage. I don't understand exactly what
this function does, someone might want to enlighten us? I've took it's
usage from classic kernel.
Also, it might be worthwhile to check the return value and BUG if it's
false, but I don't know what exactly that return value means.
>
>> if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
>> int target = min_t(int, skb->len, PKT_PROT_LEN);
>> @@ -1581,6 +1541,8 @@ static int xenvif_tx_submit(struct xenvif *vif)
>> if (checksum_setup(vif, skb)) {
>> netdev_dbg(vif->dev,
>> "Can't setup checksum in net_tx_action\n");
>> + if (skb_shinfo(skb)->destructor_arg)
>> + skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
>
> Do you still care setting the flag even if this skb is not going to be
> delivered? If so can you state clearly the reason just like the
> following hunk?
Of course, otherwise the pages wouldn't be sent back to the guest. I've
added a comment.
>> @@ -1715,7 +1685,7 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
>> int xenvif_tx_action(struct xenvif *vif, int budget)
>> {
>> unsigned nr_gops;
>> - int work_done;
>> + int work_done, ret;
>>
>> if (unlikely(!tx_work_todo(vif)))
>> return 0;
>> @@ -1725,7 +1695,10 @@ int xenvif_tx_action(struct xenvif *vif, int budget)
>> if (nr_gops == 0)
>> return 0;
>>
>> - gnttab_batch_copy(vif->tx_copy_ops, nr_gops);
>> + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
>> + vif->tx_map_ops,
>> + nr_gops);
>
> Why do you need to replace gnttab_batch_copy with hypercall? In the
> ideal situation gnttab_batch_copy should behave the same as directly
> hypercall but it also handles GNTST_eagain for you.
I don't need gnttab_batch_copy at all, I'm using the grant mapping
hypercall here.
Regards,
Zoli
On Fri, Jan 10, 2014 at 11:35:08AM +0000, Zoltan Kiss wrote:
[...]
>
> >>@@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
> >> err = gop->status;
> >> if (unlikely(err))
> >> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
> >>+ else {
> >>+ if (vif->grant_tx_handle[pending_idx] !=
> >>+ NETBACK_INVALID_HANDLE) {
> >>+ netdev_err(vif->dev,
> >>+ "Stale mapped handle! pending_idx %x handle %x\n",
> >>+ pending_idx, vif->grant_tx_handle[pending_idx]);
> >>+ BUG();
> >>+ }
> >>+ set_phys_to_machine(idx_to_pfn(vif, pending_idx),
> >>+ FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
> >
> >What happens when you don't have this?
> Your frags will be filled with garbage. I don't understand exactly
> what this function does, someone might want to enlighten us? I've
> took it's usage from classic kernel.
> Also, it might be worthwhile to check the return value and BUG if
> it's false, but I don't know what exactly that return value means.
>
This is actually part of gnttab_map_refs. As you're using hypercall
directly this becomes very fragile.
So the right thing to do is to fix gnttab_map_refs.
> >
> >> if (skb_is_nonlinear(skb) && skb_headlen(skb) < PKT_PROT_LEN) {
> >> int target = min_t(int, skb->len, PKT_PROT_LEN);
> >>@@ -1581,6 +1541,8 @@ static int xenvif_tx_submit(struct xenvif *vif)
> >> if (checksum_setup(vif, skb)) {
> >> netdev_dbg(vif->dev,
> >> "Can't setup checksum in net_tx_action\n");
> >>+ if (skb_shinfo(skb)->destructor_arg)
> >>+ skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
> >
> >Do you still care setting the flag even if this skb is not going to be
> >delivered? If so can you state clearly the reason just like the
> >following hunk?
> Of course, otherwise the pages wouldn't be sent back to the guest.
> I've added a comment.
>
OK, Thanks! That means whenever SKB leaves netback we need to add this
flag.
> >>@@ -1715,7 +1685,7 @@ static inline void xenvif_tx_dealloc_action(struct xenvif *vif)
> >> int xenvif_tx_action(struct xenvif *vif, int budget)
> >> {
> >> unsigned nr_gops;
> >>- int work_done;
> >>+ int work_done, ret;
> >>
> >> if (unlikely(!tx_work_todo(vif)))
> >> return 0;
> >>@@ -1725,7 +1695,10 @@ int xenvif_tx_action(struct xenvif *vif, int budget)
> >> if (nr_gops == 0)
> >> return 0;
> >>
> >>- gnttab_batch_copy(vif->tx_copy_ops, nr_gops);
> >>+ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
> >>+ vif->tx_map_ops,
> >>+ nr_gops);
> >
> >Why do you need to replace gnttab_batch_copy with hypercall? In the
> >ideal situation gnttab_batch_copy should behave the same as directly
> >hypercall but it also handles GNTST_eagain for you.
>
> I don't need gnttab_batch_copy at all, I'm using the grant mapping
> hypercall here.
>
Oops, my bad! Ignore that one.
Wei.
> Regards,
>
> Zoli
On Fri, Jan 10, 2014 at 11:45:34AM +0000, Wei Liu wrote:
> On Fri, Jan 10, 2014 at 11:35:08AM +0000, Zoltan Kiss wrote:
> [...]
> >
> > >>@@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
> > >> err = gop->status;
> > >> if (unlikely(err))
> > >> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
> > >>+ else {
> > >>+ if (vif->grant_tx_handle[pending_idx] !=
> > >>+ NETBACK_INVALID_HANDLE) {
> > >>+ netdev_err(vif->dev,
> > >>+ "Stale mapped handle! pending_idx %x handle %x\n",
> > >>+ pending_idx, vif->grant_tx_handle[pending_idx]);
> > >>+ BUG();
> > >>+ }
> > >>+ set_phys_to_machine(idx_to_pfn(vif, pending_idx),
> > >>+ FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
> > >
> > >What happens when you don't have this?
> > Your frags will be filled with garbage. I don't understand exactly
> > what this function does, someone might want to enlighten us? I've
> > took it's usage from classic kernel.
> > Also, it might be worthwhile to check the return value and BUG if
> > it's false, but I don't know what exactly that return value means.
> >
>
> This is actually part of gnttab_map_refs. As you're using hypercall
> directly this becomes very fragile.
>
To make it clear, set_phys_to_machine is done within m2p_add_override.
Wei.
On 10/01/14 11:45, Wei Liu wrote:
> On Fri, Jan 10, 2014 at 11:35:08AM +0000, Zoltan Kiss wrote:
> [...]
>>
>>>> @@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif *vif,
>>>> err = gop->status;
>>>> if (unlikely(err))
>>>> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
>>>> + else {
>>>> + if (vif->grant_tx_handle[pending_idx] !=
>>>> + NETBACK_INVALID_HANDLE) {
>>>> + netdev_err(vif->dev,
>>>> + "Stale mapped handle! pending_idx %x handle %x\n",
>>>> + pending_idx, vif->grant_tx_handle[pending_idx]);
>>>> + BUG();
>>>> + }
>>>> + set_phys_to_machine(idx_to_pfn(vif, pending_idx),
>>>> + FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
>>>
>>> What happens when you don't have this?
>> Your frags will be filled with garbage. I don't understand exactly
>> what this function does, someone might want to enlighten us? I've
>> took it's usage from classic kernel.
>> Also, it might be worthwhile to check the return value and BUG if
>> it's false, but I don't know what exactly that return value means.
>>
>
> This is actually part of gnttab_map_refs. As you're using hypercall
> directly this becomes very fragile.
>
> So the right thing to do is to fix gnttab_map_refs.
I agree, as I mentioned in other email in this thread, I think that
should be the topic of an another patchseries. In the meantime, I will
use gnttab_batch_map instead of the direct hypercall, it handles the
GNTST_eagain scenario, and I will use set_phys_to_machine the same way
as m2p_override does:
if (unlikely(!set_phys_to_machine(idx_to_pfn(vif, pending_idx),
FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT))
BUG();
Zoli
On 10/01/14 15:24, Zoltan Kiss wrote:
> On 10/01/14 11:45, Wei Liu wrote:
>> On Fri, Jan 10, 2014 at 11:35:08AM +0000, Zoltan Kiss wrote:
>> [...]
>>>
>>>>> @@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif
>>>>> *vif,
>>>>> err = gop->status;
>>>>> if (unlikely(err))
>>>>> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
>>>>> + else {
>>>>> + if (vif->grant_tx_handle[pending_idx] !=
>>>>> + NETBACK_INVALID_HANDLE) {
>>>>> + netdev_err(vif->dev,
>>>>> + "Stale mapped handle! pending_idx %x handle %x\n",
>>>>> + pending_idx, vif->grant_tx_handle[pending_idx]);
>>>>> + BUG();
>>>>> + }
>>>>> + set_phys_to_machine(idx_to_pfn(vif, pending_idx),
>>>>> + FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
>>>>
>>>> What happens when you don't have this?
>>> Your frags will be filled with garbage. I don't understand exactly
>>> what this function does, someone might want to enlighten us? I've
>>> took it's usage from classic kernel.
>>> Also, it might be worthwhile to check the return value and BUG if
>>> it's false, but I don't know what exactly that return value means.
>>>
>>
>> This is actually part of gnttab_map_refs. As you're using hypercall
>> directly this becomes very fragile.
>>
>> So the right thing to do is to fix gnttab_map_refs.
> I agree, as I mentioned in other email in this thread, I think that
> should be the topic of an another patchseries. In the meantime, I will
> use gnttab_batch_map instead of the direct hypercall, it handles the
> GNTST_eagain scenario, and I will use set_phys_to_machine the same way
> as m2p_override does:
If the grant table code doesn't provide the API calls you need you can
either:
a) add the new API as a prerequisite patch.
b) use the existing API calls and live with the performance problem,
until you can refactor the API later on.
Adding a netback-specific hack isn't a valid option.
David
On Fri, Jan 10, 2014 at 04:02:20PM +0000, David Vrabel wrote:
> On 10/01/14 15:24, Zoltan Kiss wrote:
> > On 10/01/14 11:45, Wei Liu wrote:
> >> On Fri, Jan 10, 2014 at 11:35:08AM +0000, Zoltan Kiss wrote:
> >> [...]
> >>>
> >>>>> @@ -920,6 +852,18 @@ static int xenvif_tx_check_gop(struct xenvif
> >>>>> *vif,
> >>>>> err = gop->status;
> >>>>> if (unlikely(err))
> >>>>> xenvif_idx_release(vif, pending_idx, XEN_NETIF_RSP_ERROR);
> >>>>> + else {
> >>>>> + if (vif->grant_tx_handle[pending_idx] !=
> >>>>> + NETBACK_INVALID_HANDLE) {
> >>>>> + netdev_err(vif->dev,
> >>>>> + "Stale mapped handle! pending_idx %x handle %x\n",
> >>>>> + pending_idx, vif->grant_tx_handle[pending_idx]);
> >>>>> + BUG();
> >>>>> + }
> >>>>> + set_phys_to_machine(idx_to_pfn(vif, pending_idx),
> >>>>> + FOREIGN_FRAME(gop->dev_bus_addr >> PAGE_SHIFT));
> >>>>
> >>>> What happens when you don't have this?
> >>> Your frags will be filled with garbage. I don't understand exactly
> >>> what this function does, someone might want to enlighten us? I've
> >>> took it's usage from classic kernel.
> >>> Also, it might be worthwhile to check the return value and BUG if
> >>> it's false, but I don't know what exactly that return value means.
> >>>
> >>
> >> This is actually part of gnttab_map_refs. As you're using hypercall
> >> directly this becomes very fragile.
> >>
> >> So the right thing to do is to fix gnttab_map_refs.
> > I agree, as I mentioned in other email in this thread, I think that
> > should be the topic of an another patchseries. In the meantime, I will
> > use gnttab_batch_map instead of the direct hypercall, it handles the
> > GNTST_eagain scenario, and I will use set_phys_to_machine the same way
> > as m2p_override does:
>
> If the grant table code doesn't provide the API calls you need you can
> either:
>
> a) add the new API as a prerequisite patch.
> b) use the existing API calls and live with the performance problem,
> until you can refactor the API later on.
>
> Adding a netback-specific hack isn't a valid option.
>
Agreed.
Wei.
> David
On 10/01/14 16:02, David Vrabel wrote:
> On 10/01/14 15:24, Zoltan Kiss wrote:
> If the grant table code doesn't provide the API calls you need you can
> either:
>
> a) add the new API as a prerequisite patch.
> b) use the existing API calls and live with the performance problem,
> until you can refactor the API later on.
>
> Adding a netback-specific hack isn't a valid option.
>
> David
Ok, I've sent in the patch which does a)
Zoli
On 09/01/14 09:20, Paul Durrant wrote:
>> We are adding the skb to vif->rx_queue even when
>> xenvif_rx_ring_slots_available(vif, min_slots_needed) said there is no
>> space for that. Or am I missing something? Paul?
>>
> That's correct. Part of the flow control improvement was to get rid of needless packet drops. For your purposes, you basically need to avoid using the queuing discipline and take packets into netback's vif->rx_queue regardless of the state of the shared ring so that you can drop them if they get beyond a certain age. So, perhaps you should never stop the netif queue, place an upper limit on vif->rx_queue (either packet or byte count) and drop when that is exceeded (i.e. mimicking pfifo or bfifo internally).
>
How about this:
- when the timer fires first we wake up the thread an tell it to drop
all the packets in rx_queue
- start_xmit then can drain the qdisc queue into the device queue
- additionally, the RX thread should stop that timer when it was able to
do some work
Regards,
Zoli
> -----Original Message-----
> From: Zoltan Kiss
> Sent: 13 January 2014 00:20
> To: Paul Durrant; Ian Campbell; Wei Liu; [email protected];
> [email protected]; [email protected]; Jonathan Davies
> Subject: Re: [PATCH net-next v3 8/9] xen-netback: Timeout packets in RX
> path
>
> On 09/01/14 09:20, Paul Durrant wrote:
> >> We are adding the skb to vif->rx_queue even when
> >> xenvif_rx_ring_slots_available(vif, min_slots_needed) said there is no
> >> space for that. Or am I missing something? Paul?
> >>
> > That's correct. Part of the flow control improvement was to get rid of
> needless packet drops. For your purposes, you basically need to avoid using
> the queuing discipline and take packets into netback's vif->rx_queue
> regardless of the state of the shared ring so that you can drop them if they
> get beyond a certain age. So, perhaps you should never stop the netif
> queue, place an upper limit on vif->rx_queue (either packet or byte count)
> and drop when that is exceeded (i.e. mimicking pfifo or bfifo internally).
> >
> How about this:
> - when the timer fires first we wake up the thread an tell it to drop
> all the packets in rx_queue
> - start_xmit then can drain the qdisc queue into the device queue
> - additionally, the RX thread should stop that timer when it was able to
> do some work
>
Yes, you could do it that way.
Paul