* [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
This patch set enables support for using the recent changes that allow for
unmapping pages without invalidating their contents via
DMA_ATTR_SKIP_CPU_SYNC. With this change DMA pages should be writable and
as a result we should be able to make use of build_skb which can be used to
drop the skb->head memory allocation, header parsing, and memcpy from the
receive path which can greatly help to improve performance.
My main concern at this point is that there might be an architecture where
I didn't get DMA_ATTR_SKIP_CPU_SYNC implemented that might still need it.
For that reason I have also added a ethtool private flag called out as
"legacy-rx". If a platform encounters an issue where the Rx can possibly
corrupt data it can be enbled by running:
ethtool --set-priv-flags DEVNAME legacy-rx on
The testing matrix for all of these patches is going to be pretty
extensive. Basically we want to test these patches on as many platforms
and architectures as possible with as many features being toggled as
possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while receiving
traffic.
Within the patches there is also some intialization changes. Specifically
I have updated the code paths to defer clearing the rings until we are
about to initialize them and give them to hardware. By doing this we are
able to avoid having to dirty memory we don't need to which should help to
improve suspend/resume times for when we start looking at possibly using
the suspend/resume approach for migration of interface in VMs.
---
Alexander Duyck (11):
ixgbe: Add function for checking to see if we can reuse page
ixgbe: Only DMA sync frame length
ixgbe: Update driver to make use of DMA attributes in Rx path
ixgbe: Update code to better handle incrementing page count
ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE
ixgbe: Use length to determine if descriptor is done
ixgbe: Break out Rx buffer page management
ixgbe: Add support for padding packet
ixgbe: Add private flag to control buffer mode
ixgbe: Add support for build_skb
ixgbe: Don't bother clearing buffer memory for descriptor rings
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 45 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 58 ++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 584 ++++++++++++++--------
3 files changed, 465 insertions(+), 222 deletions(-)
^ permalink raw reply [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan
This patch set enables support for using the recent changes that allow for
unmapping pages without invalidating their contents via
DMA_ATTR_SKIP_CPU_SYNC. With this change DMA pages should be writable and
as a result we should be able to make use of build_skb which can be used to
drop the skb->head memory allocation, header parsing, and memcpy from the
receive path which can greatly help to improve performance.
My main concern at this point is that there might be an architecture where
I didn't get DMA_ATTR_SKIP_CPU_SYNC implemented that might still need it.
For that reason I have also added a ethtool private flag called out as
"legacy-rx". If a platform encounters an issue where the Rx can possibly
corrupt data it can be enbled by running:
ethtool --set-priv-flags DEVNAME legacy-rx on
The testing matrix for all of these patches is going to be pretty
extensive. Basically we want to test these patches on as many platforms
and architectures as possible with as many features being toggled as
possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while receiving
traffic.
Within the patches there is also some intialization changes. Specifically
I have updated the code paths to defer clearing the rings until we are
about to initialize them and give them to hardware. By doing this we are
able to avoid having to dirty memory we don't need to which should help to
improve suspend/resume times for when we start looking at possibly using
the suspend/resume approach for migration of interface in VMs.
---
Alexander Duyck (11):
ixgbe: Add function for checking to see if we can reuse page
ixgbe: Only DMA sync frame length
ixgbe: Update driver to make use of DMA attributes in Rx path
ixgbe: Update code to better handle incrementing page count
ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE
ixgbe: Use length to determine if descriptor is done
ixgbe: Break out Rx buffer page management
ixgbe: Add support for padding packet
ixgbe: Add private flag to control buffer mode
ixgbe: Add support for build_skb
ixgbe: Don't bother clearing buffer memory for descriptor rings
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 45 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 58 ++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 584 ++++++++++++++--------
3 files changed, 465 insertions(+), 222 deletions(-)
--
^ permalink raw reply [flat|nested] 30+ messages in thread
* [next PATCH 01/11] ixgbe: Add function for checking to see if we can reuse page
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:06 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch consolidates the code for the ixgbe driver so that it is more
inline with what is already in igb. The general idea is to just
consolidate functions that represent logical steps in the Rx process so we
can later update them more easily.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 70 +++++++++++++++----------
1 file changed, 41 insertions(+), 29 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 3beadc8c7a0a..e80d885af4d3 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1947,6 +1947,41 @@ static inline bool ixgbe_page_is_reserved(struct page *page)
return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
}
+static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
+ struct page *page,
+ const unsigned int truesize)
+{
+#if (PAGE_SIZE >= 8192)
+ unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
+ ixgbe_rx_bufsz(rx_ring);
+#endif
+ /* avoid re-using remote pages */
+ if (unlikely(ixgbe_page_is_reserved(page)))
+ return false;
+
+#if (PAGE_SIZE < 8192)
+ /* if we are only owner of page we can reuse it */
+ if (unlikely(page_count(page) != 1))
+ return false;
+
+ /* flip page offset to other buffer */
+ rx_buffer->page_offset ^= truesize;
+#else
+ /* move offset up to the next cache line */
+ rx_buffer->page_offset += truesize;
+
+ if (rx_buffer->page_offset > last_offset)
+ return false;
+#endif
+
+ /* Even if we own the page, we are not allowed to use atomic_set()
+ * This would break get_page_unless_zero() users.
+ */
+ page_ref_inc(page);
+
+ return true;
+}
+
/**
* ixgbe_add_rx_frag - Add contents of Rx buffer to sk_buff
* @rx_ring: rx descriptor ring to transact packets on
@@ -1968,18 +2003,18 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct sk_buff *skb)
{
struct page *page = rx_buffer->page;
+ unsigned char *va = page_address(page) + rx_buffer->page_offset;
unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
#else
unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
- unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
- ixgbe_rx_bufsz(rx_ring);
#endif
- if ((size <= IXGBE_RX_HDR_SIZE) && !skb_is_nonlinear(skb)) {
- unsigned char *va = page_address(page) + rx_buffer->page_offset;
+ if (unlikely(skb_is_nonlinear(skb)))
+ goto add_tail_frag;
+ if (size <= IXGBE_RX_HDR_SIZE) {
memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
/* page is not reserved, we can reuse buffer as-is */
@@ -1991,34 +2026,11 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
return false;
}
+add_tail_frag:
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
rx_buffer->page_offset, size, truesize);
- /* avoid re-using remote pages */
- if (unlikely(ixgbe_page_is_reserved(page)))
- return false;
-
-#if (PAGE_SIZE < 8192)
- /* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != 1))
- return false;
-
- /* flip page offset to other buffer */
- rx_buffer->page_offset ^= truesize;
-#else
- /* move offset up to the next cache line */
- rx_buffer->page_offset += truesize;
-
- if (rx_buffer->page_offset > last_offset)
- return false;
-#endif
-
- /* Even if we own the page, we are not allowed to use atomic_set()
- * This would break get_page_unless_zero() users.
- */
- page_ref_inc(page);
-
- return true;
+ return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
}
static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 01/11] ixgbe: Add function for checking to see if we can reuse page
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch consolidates the code for the ixgbe driver so that it is more
inline with what is already in igb. The general idea is to just
consolidate functions that represent logical steps in the Rx process so we
can later update them more easily.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 70 +++++++++++++++----------
1 file changed, 41 insertions(+), 29 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 3beadc8c7a0a..e80d885af4d3 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1947,6 +1947,41 @@ static inline bool ixgbe_page_is_reserved(struct page *page)
return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
}
+static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
+ struct page *page,
+ const unsigned int truesize)
+{
+#if (PAGE_SIZE >= 8192)
+ unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
+ ixgbe_rx_bufsz(rx_ring);
+#endif
+ /* avoid re-using remote pages */
+ if (unlikely(ixgbe_page_is_reserved(page)))
+ return false;
+
+#if (PAGE_SIZE < 8192)
+ /* if we are only owner of page we can reuse it */
+ if (unlikely(page_count(page) != 1))
+ return false;
+
+ /* flip page offset to other buffer */
+ rx_buffer->page_offset ^= truesize;
+#else
+ /* move offset up to the next cache line */
+ rx_buffer->page_offset += truesize;
+
+ if (rx_buffer->page_offset > last_offset)
+ return false;
+#endif
+
+ /* Even if we own the page, we are not allowed to use atomic_set()
+ * This would break get_page_unless_zero() users.
+ */
+ page_ref_inc(page);
+
+ return true;
+}
+
/**
* ixgbe_add_rx_frag - Add contents of Rx buffer to sk_buff
* @rx_ring: rx descriptor ring to transact packets on
@@ -1968,18 +2003,18 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct sk_buff *skb)
{
struct page *page = rx_buffer->page;
+ unsigned char *va = page_address(page) + rx_buffer->page_offset;
unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
#else
unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
- unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
- ixgbe_rx_bufsz(rx_ring);
#endif
- if ((size <= IXGBE_RX_HDR_SIZE) && !skb_is_nonlinear(skb)) {
- unsigned char *va = page_address(page) + rx_buffer->page_offset;
+ if (unlikely(skb_is_nonlinear(skb)))
+ goto add_tail_frag;
+ if (size <= IXGBE_RX_HDR_SIZE) {
memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
/* page is not reserved, we can reuse buffer as-is */
@@ -1991,34 +2026,11 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
return false;
}
+add_tail_frag:
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
rx_buffer->page_offset, size, truesize);
- /* avoid re-using remote pages */
- if (unlikely(ixgbe_page_is_reserved(page)))
- return false;
-
-#if (PAGE_SIZE < 8192)
- /* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != 1))
- return false;
-
- /* flip page offset to other buffer */
- rx_buffer->page_offset ^= truesize;
-#else
- /* move offset up to the next cache line */
- rx_buffer->page_offset += truesize;
-
- if (rx_buffer->page_offset > last_offset)
- return false;
-#endif
-
- /* Even if we own the page, we are not allowed to use atomic_set()
- * This would break get_page_unless_zero() users.
- */
- page_ref_inc(page);
-
- return true;
+ return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
}
static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 02/11] ixgbe: Only DMA sync frame length
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:06 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
On some platforms, syncing a buffer for DMA is expensive. Rather than
sync the whole 2K receive buffer, only synchronise the length of the
frame, which will typically be the MTU, or a much smaller TCP ACK.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index e80d885af4d3..dbbf5223ace2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1858,7 +1858,7 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
dma_sync_single_range_for_cpu(rx_ring->dev,
IXGBE_CB(skb)->dma,
frag->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ skb_frag_size(frag),
DMA_FROM_DEVICE);
}
IXGBE_CB(skb)->dma = 0;
@@ -1999,12 +1999,11 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
**/
static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *rx_buffer,
- union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size,
struct sk_buff *skb)
{
struct page *page = rx_buffer->page;
unsigned char *va = page_address(page) + rx_buffer->page_offset;
- unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
#else
@@ -2036,6 +2035,7 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
union ixgbe_adv_rx_desc *rx_desc)
{
+ unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
struct ixgbe_rx_buffer *rx_buffer;
struct sk_buff *skb;
struct page *page;
@@ -2090,14 +2090,14 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
dma_sync_single_range_for_cpu(rx_ring->dev,
rx_buffer->dma,
rx_buffer->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ size,
DMA_FROM_DEVICE);
rx_buffer->skb = NULL;
}
/* pull page into skb */
- if (ixgbe_add_rx_frag(rx_ring, rx_buffer, rx_desc, skb)) {
+ if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
} else if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 02/11] ixgbe: Only DMA sync frame length
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
On some platforms, syncing a buffer for DMA is expensive. Rather than
sync the whole 2K receive buffer, only synchronise the length of the
frame, which will typically be the MTU, or a much smaller TCP ACK.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index e80d885af4d3..dbbf5223ace2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1858,7 +1858,7 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
dma_sync_single_range_for_cpu(rx_ring->dev,
IXGBE_CB(skb)->dma,
frag->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ skb_frag_size(frag),
DMA_FROM_DEVICE);
}
IXGBE_CB(skb)->dma = 0;
@@ -1999,12 +1999,11 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
**/
static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *rx_buffer,
- union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size,
struct sk_buff *skb)
{
struct page *page = rx_buffer->page;
unsigned char *va = page_address(page) + rx_buffer->page_offset;
- unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
#else
@@ -2036,6 +2035,7 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
union ixgbe_adv_rx_desc *rx_desc)
{
+ unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
struct ixgbe_rx_buffer *rx_buffer;
struct sk_buff *skb;
struct page *page;
@@ -2090,14 +2090,14 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
dma_sync_single_range_for_cpu(rx_ring->dev,
rx_buffer->dma,
rx_buffer->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ size,
DMA_FROM_DEVICE);
rx_buffer->skb = NULL;
}
/* pull page into skb */
- if (ixgbe_add_rx_frag(rx_ring, rx_buffer, rx_desc, skb)) {
+ if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
} else if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 03/11] ixgbe: Update driver to make use of DMA attributes in Rx path
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:06 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds support for DMA_ATTR_SKIP_CPU_SYNC and
DMA_ATTR_WEAK_ORDERING. By enabling both of these for the Rx path we are
able to see performance improvements on architectures that implement either
one due to the fact that page mapping and unmapping only has to sync what
is actually being used instead of the entire buffer. In addition by
enabling the weak ordering attribute enables a performance improvement for
architectures that can associate a memory ordering with a DMA buffer such
as Sparc.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 3 +
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 56 +++++++++++++++++--------
2 files changed, 40 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 9c6ccfc34177..97e74deecae2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -107,6 +107,9 @@
/* How many Rx Buffers do we bundle into one write to the hardware ? */
#define IXGBE_RX_BUFFER_WRITE 16 /* Must be power of 2 */
+#define IXGBE_RX_DMA_ATTR \
+ (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
+
enum ixgbe_tx_flags {
/* cmd_type flags */
IXGBE_TX_FLAGS_HW_VLAN = 0x01,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dbbf5223ace2..062b984ffdf4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1583,8 +1583,10 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
}
/* map page for use */
- dma = dma_map_page(rx_ring->dev, page, 0,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ dma = dma_map_page_attrs(rx_ring->dev, page, 0,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
/*
* if mapping failed free memory back to system since
@@ -1627,6 +1629,12 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
if (!ixgbe_alloc_mapped_page(rx_ring, bi))
break;
+ /* sync the buffer for use by the device */
+ dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
+ bi->page_offset,
+ ixgbe_rx_bufsz(rx_ring),
+ DMA_FROM_DEVICE);
+
/*
* Refresh the desc even if buffer_addrs didn't change
* because each write-back erases this info.
@@ -1849,8 +1857,10 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
{
/* if the page was released unmap it, else just sync our portion */
if (unlikely(IXGBE_CB(skb)->page_released)) {
- dma_unmap_page(rx_ring->dev, IXGBE_CB(skb)->dma,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(rx_ring->dev, IXGBE_CB(skb)->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
IXGBE_CB(skb)->page_released = false;
} else {
struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
@@ -1934,12 +1944,6 @@ static void ixgbe_reuse_rx_page(struct ixgbe_ring *rx_ring,
/* transfer page from old buffer to new buffer */
*new_buff = *old_buff;
-
- /* sync the buffer for use by the device */
- dma_sync_single_range_for_device(rx_ring->dev, new_buff->dma,
- new_buff->page_offset,
- ixgbe_rx_bufsz(rx_ring),
- DMA_FROM_DEVICE);
}
static inline bool ixgbe_page_is_reserved(struct page *page)
@@ -2105,9 +2109,10 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
IXGBE_CB(skb)->page_released = true;
} else {
/* we are not reusing the buffer so unmap it */
- dma_unmap_page(rx_ring->dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring),
- DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
}
/* clear contents of buffer_info */
@@ -4941,10 +4946,11 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
if (rx_buffer->skb) {
struct sk_buff *skb = rx_buffer->skb;
if (IXGBE_CB(skb)->page_released)
- dma_unmap_page(dev,
- IXGBE_CB(skb)->dma,
- ixgbe_rx_bufsz(rx_ring),
- DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(dev,
+ IXGBE_CB(skb)->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
dev_kfree_skb(skb);
rx_buffer->skb = NULL;
}
@@ -4952,8 +4958,20 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
if (!rx_buffer->page)
continue;
- dma_unmap_page(dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ /* Invalidate cache lines that may have been written to by
+ * device so that we avoid corrupting memory.
+ */
+ dma_sync_single_range_for_cpu(rx_ring->dev,
+ rx_buffer->dma,
+ rx_buffer->page_offset,
+ ixgbe_rx_bufsz(rx_ring),
+ DMA_FROM_DEVICE);
+
+ /* free resources associated with mapping */
+ dma_unmap_page_attrs(dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
__free_pages(rx_buffer->page, ixgbe_rx_pg_order(rx_ring));
rx_buffer->page = NULL;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 03/11] ixgbe: Update driver to make use of DMA attributes in Rx path
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds support for DMA_ATTR_SKIP_CPU_SYNC and
DMA_ATTR_WEAK_ORDERING. By enabling both of these for the Rx path we are
able to see performance improvements on architectures that implement either
one due to the fact that page mapping and unmapping only has to sync what
is actually being used instead of the entire buffer. In addition by
enabling the weak ordering attribute enables a performance improvement for
architectures that can associate a memory ordering with a DMA buffer such
as Sparc.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 3 +
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 56 +++++++++++++++++--------
2 files changed, 40 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 9c6ccfc34177..97e74deecae2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -107,6 +107,9 @@
/* How many Rx Buffers do we bundle into one write to the hardware ? */
#define IXGBE_RX_BUFFER_WRITE 16 /* Must be power of 2 */
+#define IXGBE_RX_DMA_ATTR \
+ (DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
+
enum ixgbe_tx_flags {
/* cmd_type flags */
IXGBE_TX_FLAGS_HW_VLAN = 0x01,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dbbf5223ace2..062b984ffdf4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1583,8 +1583,10 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
}
/* map page for use */
- dma = dma_map_page(rx_ring->dev, page, 0,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ dma = dma_map_page_attrs(rx_ring->dev, page, 0,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
/*
* if mapping failed free memory back to system since
@@ -1627,6 +1629,12 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
if (!ixgbe_alloc_mapped_page(rx_ring, bi))
break;
+ /* sync the buffer for use by the device */
+ dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
+ bi->page_offset,
+ ixgbe_rx_bufsz(rx_ring),
+ DMA_FROM_DEVICE);
+
/*
* Refresh the desc even if buffer_addrs didn't change
* because each write-back erases this info.
@@ -1849,8 +1857,10 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
{
/* if the page was released unmap it, else just sync our portion */
if (unlikely(IXGBE_CB(skb)->page_released)) {
- dma_unmap_page(rx_ring->dev, IXGBE_CB(skb)->dma,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(rx_ring->dev, IXGBE_CB(skb)->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
IXGBE_CB(skb)->page_released = false;
} else {
struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
@@ -1934,12 +1944,6 @@ static void ixgbe_reuse_rx_page(struct ixgbe_ring *rx_ring,
/* transfer page from old buffer to new buffer */
*new_buff = *old_buff;
-
- /* sync the buffer for use by the device */
- dma_sync_single_range_for_device(rx_ring->dev, new_buff->dma,
- new_buff->page_offset,
- ixgbe_rx_bufsz(rx_ring),
- DMA_FROM_DEVICE);
}
static inline bool ixgbe_page_is_reserved(struct page *page)
@@ -2105,9 +2109,10 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
IXGBE_CB(skb)->page_released = true;
} else {
/* we are not reusing the buffer so unmap it */
- dma_unmap_page(rx_ring->dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring),
- DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
}
/* clear contents of buffer_info */
@@ -4941,10 +4946,11 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
if (rx_buffer->skb) {
struct sk_buff *skb = rx_buffer->skb;
if (IXGBE_CB(skb)->page_released)
- dma_unmap_page(dev,
- IXGBE_CB(skb)->dma,
- ixgbe_rx_bufsz(rx_ring),
- DMA_FROM_DEVICE);
+ dma_unmap_page_attrs(dev,
+ IXGBE_CB(skb)->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
dev_kfree_skb(skb);
rx_buffer->skb = NULL;
}
@@ -4952,8 +4958,20 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
if (!rx_buffer->page)
continue;
- dma_unmap_page(dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring), DMA_FROM_DEVICE);
+ /* Invalidate cache lines that may have been written to by
+ * device so that we avoid corrupting memory.
+ */
+ dma_sync_single_range_for_cpu(rx_ring->dev,
+ rx_buffer->dma,
+ rx_buffer->page_offset,
+ ixgbe_rx_bufsz(rx_ring),
+ DMA_FROM_DEVICE);
+
+ /* free resources associated with mapping */
+ dma_unmap_page_attrs(dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
__free_pages(rx_buffer->page, ixgbe_rx_pg_order(rx_ring));
rx_buffer->page = NULL;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 04/11] ixgbe: Update code to better handle incrementing page count
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:06 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
Batch the page count updates instead of doing them one at a time. By doing
this we can improve the overall performance as the atomic increment
operations can be expensive due to the fact that on x86 they are locked
operations which can cause stalls. By doing bulk updates we can
consolidate the stall which should help to improve the overall receive
performance.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 7 ++++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 39 ++++++++++++++++---------
2 files changed, 31 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 97e74deecae2..717c65b0deb2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -198,7 +198,12 @@ struct ixgbe_rx_buffer {
struct sk_buff *skb;
dma_addr_t dma;
struct page *page;
- unsigned int page_offset;
+#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
+ __u32 page_offset;
+#else
+ __u16 page_offset;
+#endif
+ __u16 pagecnt_bias;
};
struct ixgbe_queue_stats {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 062b984ffdf4..519b6a2b65c1 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1602,6 +1602,7 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
bi->dma = dma;
bi->page = page;
bi->page_offset = 0;
+ bi->pagecnt_bias = 1;
return true;
}
@@ -1959,13 +1960,15 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
ixgbe_rx_bufsz(rx_ring);
#endif
+ unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
+
/* avoid re-using remote pages */
if (unlikely(ixgbe_page_is_reserved(page)))
return false;
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != 1))
+ if (unlikely(page_count(page) != pagecnt_bias))
return false;
/* flip page offset to other buffer */
@@ -1978,10 +1981,14 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
return false;
#endif
- /* Even if we own the page, we are not allowed to use atomic_set()
- * This would break get_page_unless_zero() users.
+ /* If we have drained the page fragment pool we need to update
+ * the pagecnt_bias and page count so that we fully restock the
+ * number of references the driver holds.
*/
- page_ref_inc(page);
+ if (unlikely(pagecnt_bias == 1)) {
+ page_ref_add(page, USHRT_MAX);
+ rx_buffer->pagecnt_bias = USHRT_MAX;
+ }
return true;
}
@@ -2025,7 +2032,6 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
return true;
/* this page cannot be reused so discard it */
- __free_pages(page, ixgbe_rx_pg_order(rx_ring));
return false;
}
@@ -2104,15 +2110,19 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
- } else if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
- /* the page has been released from the ring */
- IXGBE_CB(skb)->page_released = true;
} else {
- /* we are not reusing the buffer so unmap it */
- dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring),
- DMA_FROM_DEVICE,
- IXGBE_RX_DMA_ATTR);
+ if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
+ /* the page has been released from the ring */
+ IXGBE_CB(skb)->page_released = true;
+ } else {
+ /* we are not reusing the buffer so unmap it */
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
+ }
+ __page_frag_cache_drain(page,
+ rx_buffer->pagecnt_bias);
}
/* clear contents of buffer_info */
@@ -4972,7 +4982,8 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
- __free_pages(rx_buffer->page, ixgbe_rx_pg_order(rx_ring));
+ __page_frag_cache_drain(rx_buffer->page,
+ rx_buffer->pagecnt_bias);
rx_buffer->page = NULL;
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 04/11] ixgbe: Update code to better handle incrementing page count
@ 2017-01-06 16:06 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:06 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
Batch the page count updates instead of doing them one at a time. By doing
this we can improve the overall performance as the atomic increment
operations can be expensive due to the fact that on x86 they are locked
operations which can cause stalls. By doing bulk updates we can
consolidate the stall which should help to improve the overall receive
performance.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 7 ++++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 39 ++++++++++++++++---------
2 files changed, 31 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 97e74deecae2..717c65b0deb2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -198,7 +198,12 @@ struct ixgbe_rx_buffer {
struct sk_buff *skb;
dma_addr_t dma;
struct page *page;
- unsigned int page_offset;
+#if (BITS_PER_LONG > 32) || (PAGE_SIZE >= 65536)
+ __u32 page_offset;
+#else
+ __u16 page_offset;
+#endif
+ __u16 pagecnt_bias;
};
struct ixgbe_queue_stats {
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 062b984ffdf4..519b6a2b65c1 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1602,6 +1602,7 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
bi->dma = dma;
bi->page = page;
bi->page_offset = 0;
+ bi->pagecnt_bias = 1;
return true;
}
@@ -1959,13 +1960,15 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
ixgbe_rx_bufsz(rx_ring);
#endif
+ unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
+
/* avoid re-using remote pages */
if (unlikely(ixgbe_page_is_reserved(page)))
return false;
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != 1))
+ if (unlikely(page_count(page) != pagecnt_bias))
return false;
/* flip page offset to other buffer */
@@ -1978,10 +1981,14 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
return false;
#endif
- /* Even if we own the page, we are not allowed to use atomic_set()
- * This would break get_page_unless_zero() users.
+ /* If we have drained the page fragment pool we need to update
+ * the pagecnt_bias and page count so that we fully restock the
+ * number of references the driver holds.
*/
- page_ref_inc(page);
+ if (unlikely(pagecnt_bias == 1)) {
+ page_ref_add(page, USHRT_MAX);
+ rx_buffer->pagecnt_bias = USHRT_MAX;
+ }
return true;
}
@@ -2025,7 +2032,6 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
return true;
/* this page cannot be reused so discard it */
- __free_pages(page, ixgbe_rx_pg_order(rx_ring));
return false;
}
@@ -2104,15 +2110,19 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
- } else if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
- /* the page has been released from the ring */
- IXGBE_CB(skb)->page_released = true;
} else {
- /* we are not reusing the buffer so unmap it */
- dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
- ixgbe_rx_pg_size(rx_ring),
- DMA_FROM_DEVICE,
- IXGBE_RX_DMA_ATTR);
+ if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
+ /* the page has been released from the ring */
+ IXGBE_CB(skb)->page_released = true;
+ } else {
+ /* we are not reusing the buffer so unmap it */
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
+ ixgbe_rx_pg_size(rx_ring),
+ DMA_FROM_DEVICE,
+ IXGBE_RX_DMA_ATTR);
+ }
+ __page_frag_cache_drain(page,
+ rx_buffer->pagecnt_bias);
}
/* clear contents of buffer_info */
@@ -4972,7 +4982,8 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
- __free_pages(rx_buffer->page, ixgbe_rx_pg_order(rx_ring));
+ __page_frag_cache_drain(rx_buffer->page,
+ rx_buffer->pagecnt_bias);
rx_buffer->page = NULL;
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 05/11] ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
In order to support build_skb with jumbo frames it will be necessary to use
3K buffers for the Rx path with 8K pages backing them. This is needed on
architectures that implement 4K pages because we can't support 2K buffers
plus padding in a 4K page.
In the case of systems that support page sizes larger than 4K the 3K
attribute will only be applied to FCoE as we can fall back to using just 2K
buffers and adding the padding.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 20 +++++++++-----------
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 19 +++++++++++++------
2 files changed, 22 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 717c65b0deb2..80328e657d6a 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -234,13 +234,14 @@ struct ixgbe_rx_queue_stats {
#define IXGBE_TS_HDR_LEN 8
enum ixgbe_ring_state_t {
+ __IXGBE_RX_3K_BUFFER,
+ __IXGBE_RX_RSC_ENABLED,
+ __IXGBE_RX_CSUM_UDP_ZERO_ERR,
+ __IXGBE_RX_FCOE,
__IXGBE_TX_FDIR_INIT_DONE,
__IXGBE_TX_XPS_INIT_DONE,
__IXGBE_TX_DETECT_HANG,
__IXGBE_HANG_CHECK_ARMED,
- __IXGBE_RX_RSC_ENABLED,
- __IXGBE_RX_CSUM_UDP_ZERO_ERR,
- __IXGBE_RX_FCOE,
};
struct ixgbe_fwd_adapter {
@@ -352,19 +353,16 @@ struct ixgbe_ring_feature {
*/
static inline unsigned int ixgbe_rx_bufsz(struct ixgbe_ring *ring)
{
-#ifdef IXGBE_FCOE
- if (test_bit(__IXGBE_RX_FCOE, &ring->state))
- return (PAGE_SIZE < 8192) ? IXGBE_RXBUFFER_4K :
- IXGBE_RXBUFFER_3K;
-#endif
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ return IXGBE_RXBUFFER_3K;
return IXGBE_RXBUFFER_2K;
}
static inline unsigned int ixgbe_rx_pg_order(struct ixgbe_ring *ring)
{
-#ifdef IXGBE_FCOE
- if (test_bit(__IXGBE_RX_FCOE, &ring->state))
- return (PAGE_SIZE < 8192) ? 1 : 0;
+#if (PAGE_SIZE < 8192)
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ return 1;
#endif
return 0;
}
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 519b6a2b65c1..1b6f1f584044 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1617,6 +1617,7 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
union ixgbe_adv_rx_desc *rx_desc;
struct ixgbe_rx_buffer *bi;
u16 i = rx_ring->next_to_use;
+ u16 bufsz;
/* nothing to do */
if (!cleaned_count)
@@ -1626,14 +1627,15 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
bi = &rx_ring->rx_buffer_info[i];
i -= rx_ring->count;
+ bufsz = ixgbe_rx_bufsz(rx_ring);
+
do {
if (!ixgbe_alloc_mapped_page(rx_ring, bi))
break;
/* sync the buffer for use by the device */
dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
- bi->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ bi->page_offset, bufsz,
DMA_FROM_DEVICE);
/*
@@ -2016,9 +2018,9 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct page *page = rx_buffer->page;
unsigned char *va = page_address(page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
- unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+ unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
if (unlikely(skb_is_nonlinear(skb)))
@@ -3917,10 +3919,15 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
*/
for (i = 0; i < adapter->num_rx_queues; i++) {
rx_ring = adapter->rx_ring[i];
+
+ clear_ring_rsc_enabled(rx_ring);
+ clear_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
set_ring_rsc_enabled(rx_ring);
- else
- clear_ring_rsc_enabled(rx_ring);
+
+ if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
}
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 05/11] ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
In order to support build_skb with jumbo frames it will be necessary to use
3K buffers for the Rx path with 8K pages backing them. This is needed on
architectures that implement 4K pages because we can't support 2K buffers
plus padding in a 4K page.
In the case of systems that support page sizes larger than 4K the 3K
attribute will only be applied to FCoE as we can fall back to using just 2K
buffers and adding the padding.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 20 +++++++++-----------
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 19 +++++++++++++------
2 files changed, 22 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 717c65b0deb2..80328e657d6a 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -234,13 +234,14 @@ struct ixgbe_rx_queue_stats {
#define IXGBE_TS_HDR_LEN 8
enum ixgbe_ring_state_t {
+ __IXGBE_RX_3K_BUFFER,
+ __IXGBE_RX_RSC_ENABLED,
+ __IXGBE_RX_CSUM_UDP_ZERO_ERR,
+ __IXGBE_RX_FCOE,
__IXGBE_TX_FDIR_INIT_DONE,
__IXGBE_TX_XPS_INIT_DONE,
__IXGBE_TX_DETECT_HANG,
__IXGBE_HANG_CHECK_ARMED,
- __IXGBE_RX_RSC_ENABLED,
- __IXGBE_RX_CSUM_UDP_ZERO_ERR,
- __IXGBE_RX_FCOE,
};
struct ixgbe_fwd_adapter {
@@ -352,19 +353,16 @@ struct ixgbe_ring_feature {
*/
static inline unsigned int ixgbe_rx_bufsz(struct ixgbe_ring *ring)
{
-#ifdef IXGBE_FCOE
- if (test_bit(__IXGBE_RX_FCOE, &ring->state))
- return (PAGE_SIZE < 8192) ? IXGBE_RXBUFFER_4K :
- IXGBE_RXBUFFER_3K;
-#endif
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ return IXGBE_RXBUFFER_3K;
return IXGBE_RXBUFFER_2K;
}
static inline unsigned int ixgbe_rx_pg_order(struct ixgbe_ring *ring)
{
-#ifdef IXGBE_FCOE
- if (test_bit(__IXGBE_RX_FCOE, &ring->state))
- return (PAGE_SIZE < 8192) ? 1 : 0;
+#if (PAGE_SIZE < 8192)
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ return 1;
#endif
return 0;
}
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 519b6a2b65c1..1b6f1f584044 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1617,6 +1617,7 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
union ixgbe_adv_rx_desc *rx_desc;
struct ixgbe_rx_buffer *bi;
u16 i = rx_ring->next_to_use;
+ u16 bufsz;
/* nothing to do */
if (!cleaned_count)
@@ -1626,14 +1627,15 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
bi = &rx_ring->rx_buffer_info[i];
i -= rx_ring->count;
+ bufsz = ixgbe_rx_bufsz(rx_ring);
+
do {
if (!ixgbe_alloc_mapped_page(rx_ring, bi))
break;
/* sync the buffer for use by the device */
dma_sync_single_range_for_device(rx_ring->dev, bi->dma,
- bi->page_offset,
- ixgbe_rx_bufsz(rx_ring),
+ bi->page_offset, bufsz,
DMA_FROM_DEVICE);
/*
@@ -2016,9 +2018,9 @@ static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct page *page = rx_buffer->page;
unsigned char *va = page_address(page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
- unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+ unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
if (unlikely(skb_is_nonlinear(skb)))
@@ -3917,10 +3919,15 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
*/
for (i = 0; i < adapter->num_rx_queues; i++) {
rx_ring = adapter->rx_ring[i];
+
+ clear_ring_rsc_enabled(rx_ring);
+ clear_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
set_ring_rsc_enabled(rx_ring);
- else
- clear_ring_rsc_enabled(rx_ring);
+
+ if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
}
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 06/11] ixgbe: Use length to determine if descriptor is done
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This change makes it so that we use the length of the packet instead of the
DD status bit to determine if a new descriptor is ready to be processed.
The obvious advantage is that it cuts down on reads as we don't really even
need the DD bit if going from a 0 to a non-zero value on size is enough to
inform us that the packet has been completed.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 1b6f1f584044..6ce596816954 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1653,8 +1653,8 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
i -= rx_ring->count;
}
- /* clear the status bits for the next_to_use descriptor */
- rx_desc->wb.upper.status_error = 0;
+ /* clear the length for the next_to_use descriptor */
+ rx_desc->wb.upper.length = 0;
cleaned_count--;
} while (cleaned_count);
@@ -2170,7 +2170,7 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
rx_desc = IXGBE_RX_DESC(rx_ring, rx_ring->next_to_clean);
- if (!rx_desc->wb.upper.status_error)
+ if (!rx_desc->wb.upper.length)
break;
/* This memory barrier is needed to keep us from reading
@@ -3749,6 +3749,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
struct ixgbe_ring *ring)
{
struct ixgbe_hw *hw = &adapter->hw;
+ union ixgbe_adv_rx_desc *rx_desc;
u64 rdba = ring->dma;
u32 rxdctl;
u8 reg_idx = ring->reg_idx;
@@ -3783,6 +3784,10 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
rxdctl |= 0x080420;
}
+ /* initialize Rx descriptor 0 */
+ rx_desc = IXGBE_RX_DESC(ring, 0);
+ rx_desc->wb.upper.length = 0;
+
/* enable receive descriptor ring */
rxdctl |= IXGBE_RXDCTL_ENABLE;
IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx), rxdctl);
@@ -4998,9 +5003,6 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
size = sizeof(struct ixgbe_rx_buffer) * rx_ring->count;
memset(rx_ring->rx_buffer_info, 0, size);
- /* Zero out the descriptor ring */
- memset(rx_ring->desc, 0, rx_ring->size);
-
rx_ring->next_to_alloc = 0;
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 06/11] ixgbe: Use length to determine if descriptor is done
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This change makes it so that we use the length of the packet instead of the
DD status bit to determine if a new descriptor is ready to be processed.
The obvious advantage is that it cuts down on reads as we don't really even
need the DD bit if going from a 0 to a non-zero value on size is enough to
inform us that the packet has been completed.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 1b6f1f584044..6ce596816954 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1653,8 +1653,8 @@ void ixgbe_alloc_rx_buffers(struct ixgbe_ring *rx_ring, u16 cleaned_count)
i -= rx_ring->count;
}
- /* clear the status bits for the next_to_use descriptor */
- rx_desc->wb.upper.status_error = 0;
+ /* clear the length for the next_to_use descriptor */
+ rx_desc->wb.upper.length = 0;
cleaned_count--;
} while (cleaned_count);
@@ -2170,7 +2170,7 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
rx_desc = IXGBE_RX_DESC(rx_ring, rx_ring->next_to_clean);
- if (!rx_desc->wb.upper.status_error)
+ if (!rx_desc->wb.upper.length)
break;
/* This memory barrier is needed to keep us from reading
@@ -3749,6 +3749,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
struct ixgbe_ring *ring)
{
struct ixgbe_hw *hw = &adapter->hw;
+ union ixgbe_adv_rx_desc *rx_desc;
u64 rdba = ring->dma;
u32 rxdctl;
u8 reg_idx = ring->reg_idx;
@@ -3783,6 +3784,10 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
rxdctl |= 0x080420;
}
+ /* initialize Rx descriptor 0 */
+ rx_desc = IXGBE_RX_DESC(ring, 0);
+ rx_desc->wb.upper.length = 0;
+
/* enable receive descriptor ring */
rxdctl |= IXGBE_RXDCTL_ENABLE;
IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx), rxdctl);
@@ -4998,9 +5003,6 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
size = sizeof(struct ixgbe_rx_buffer) * rx_ring->count;
memset(rx_ring->rx_buffer_info, 0, size);
- /* Zero out the descriptor ring */
- memset(rx_ring->desc, 0, rx_ring->size);
-
rx_ring->next_to_alloc = 0;
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 07/11] ixgbe: Break out Rx buffer page management
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
We are going to be expanding the number of Rx paths in the driver. Instead
of duplicating all that code I am pulling it apart into separate functions
so that we don't have so much code duplication.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 230 +++++++++++++------------
1 file changed, 122 insertions(+), 108 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 6ce596816954..79495cc990c2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1864,7 +1864,6 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
- IXGBE_CB(skb)->page_released = false;
} else {
struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
@@ -1874,7 +1873,6 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
skb_frag_size(frag),
DMA_FROM_DEVICE);
}
- IXGBE_CB(skb)->dma = 0;
}
/**
@@ -1945,8 +1943,14 @@ static void ixgbe_reuse_rx_page(struct ixgbe_ring *rx_ring,
nta++;
rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
- /* transfer page from old buffer to new buffer */
- *new_buff = *old_buff;
+ /* Transfer page from old buffer to new buffer.
+ * Move each member individually to avoid possible store
+ * forwarding stalls and unnecessary copy of skb.
+ */
+ new_buff->dma = old_buff->dma;
+ new_buff->page = old_buff->page;
+ new_buff->page_offset = old_buff->page_offset;
+ new_buff->pagecnt_bias = old_buff->pagecnt_bias;
}
static inline bool ixgbe_page_is_reserved(struct page *page)
@@ -1954,15 +1958,10 @@ static inline bool ixgbe_page_is_reserved(struct page *page)
return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
}
-static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
- struct page *page,
- const unsigned int truesize)
+static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer)
{
-#if (PAGE_SIZE >= 8192)
- unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
- ixgbe_rx_bufsz(rx_ring);
-#endif
- unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
+ unsigned int pagecnt_bias = rx_buffer->pagecnt_bias;
+ struct page *page = rx_buffer->page;
/* avoid re-using remote pages */
if (unlikely(ixgbe_page_is_reserved(page)))
@@ -1970,16 +1969,17 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != pagecnt_bias))
+ if (unlikely((page_ref_count(page) - pagecnt_bias) > 1))
return false;
-
- /* flip page offset to other buffer */
- rx_buffer->page_offset ^= truesize;
#else
- /* move offset up to the next cache line */
- rx_buffer->page_offset += truesize;
-
- if (rx_buffer->page_offset > last_offset)
+ /* The last offset is a bit aggressive in that we assume the
+ * worst case of FCoE being enabled and using a 3K buffer.
+ * However this should have minimal impact as the 1K extra is
+ * still less than one buffer in size.
+ */
+#define IXGBE_LAST_OFFSET \
+ (SKB_WITH_OVERHEAD(PAGE_SIZE) - IXGBE_RXBUFFER_3K)
+ if (rx_buffer->page_offset > IXGBE_LAST_OFFSET)
return false;
#endif
@@ -1987,7 +1987,7 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
* the pagecnt_bias and page count so that we fully restock the
* number of references the driver holds.
*/
- if (unlikely(pagecnt_bias == 1)) {
+ if (unlikely(!pagecnt_bias)) {
page_ref_add(page, USHRT_MAX);
rx_buffer->pagecnt_bias = USHRT_MAX;
}
@@ -2010,106 +2010,65 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
* The function will then update the page offset if necessary and return
* true if the buffer can be reused by the adapter.
**/
-static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
+static void ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *rx_buffer,
- unsigned int size,
- struct sk_buff *skb)
+ struct sk_buff *skb,
+ unsigned int size)
{
- struct page *page = rx_buffer->page;
- unsigned char *va = page_address(page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
-
- if (unlikely(skb_is_nonlinear(skb)))
- goto add_tail_frag;
-
- if (size <= IXGBE_RX_HDR_SIZE) {
- memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
-
- /* page is not reserved, we can reuse buffer as-is */
- if (likely(!ixgbe_page_is_reserved(page)))
- return true;
-
- /* this page cannot be reused so discard it */
- return false;
- }
-
-add_tail_frag:
- skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
+ skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
rx_buffer->page_offset, size, truesize);
-
- return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
}
-static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
- union ixgbe_adv_rx_desc *rx_desc)
+static struct ixgbe_rx_buffer *ixgbe_get_rx_buffer(struct ixgbe_ring *rx_ring,
+ union ixgbe_adv_rx_desc *rx_desc,
+ struct sk_buff **skb,
+ const unsigned int size)
{
- unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
struct ixgbe_rx_buffer *rx_buffer;
- struct sk_buff *skb;
- struct page *page;
rx_buffer = &rx_ring->rx_buffer_info[rx_ring->next_to_clean];
- page = rx_buffer->page;
- prefetchw(page);
-
- skb = rx_buffer->skb;
-
- if (likely(!skb)) {
- void *page_addr = page_address(page) +
- rx_buffer->page_offset;
+ prefetchw(rx_buffer->page);
+ *skb = rx_buffer->skb;
- /* prefetch first cache line of first page */
- prefetch(page_addr);
-#if L1_CACHE_BYTES < 128
- prefetch(page_addr + L1_CACHE_BYTES);
-#endif
-
- /* allocate a skb to store the frags */
- skb = napi_alloc_skb(&rx_ring->q_vector->napi,
- IXGBE_RX_HDR_SIZE);
- if (unlikely(!skb)) {
- rx_ring->rx_stats.alloc_rx_buff_failed++;
- return NULL;
- }
-
- /*
- * we will be copying header into skb->data in
- * pskb_may_pull so it is in our interest to prefetch
- * it now to avoid a possible cache miss
- */
- prefetchw(skb->data);
-
- /*
- * Delay unmapping of the first packet. It carries the
- * header information, HW may still access the header
- * after the writeback. Only unmap it when EOP is
- * reached
- */
- if (likely(ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)))
- goto dma_sync;
-
- IXGBE_CB(skb)->dma = rx_buffer->dma;
+ /* Delay unmapping of the first packet. It carries the header
+ * information, HW may still access the header after the writeback.
+ * Only unmap it when EOP is reached
+ */
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)) {
+ if (!*skb)
+ goto skip_sync;
} else {
- if (ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
- ixgbe_dma_sync_frag(rx_ring, skb);
+ if (*skb)
+ ixgbe_dma_sync_frag(rx_ring, *skb);
+ }
-dma_sync:
- /* we are reusing so sync this buffer for CPU use */
- dma_sync_single_range_for_cpu(rx_ring->dev,
- rx_buffer->dma,
- rx_buffer->page_offset,
- size,
- DMA_FROM_DEVICE);
+ /* we are reusing so sync this buffer for CPU use */
+ dma_sync_single_range_for_cpu(rx_ring->dev,
+ rx_buffer->dma,
+ rx_buffer->page_offset,
+ size,
+ DMA_FROM_DEVICE);
+skip_sync:
+ rx_buffer->pagecnt_bias--;
- rx_buffer->skb = NULL;
- }
+ return rx_buffer;
+}
- /* pull page into skb */
- if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
+static void ixgbe_put_rx_buffer(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ struct sk_buff *skb)
+{
+ if (ixgbe_can_reuse_rx_page(rx_buffer)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
} else {
@@ -2123,12 +2082,55 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
}
- __page_frag_cache_drain(page,
+ __page_frag_cache_drain(rx_buffer->page,
rx_buffer->pagecnt_bias);
}
- /* clear contents of buffer_info */
+ /* clear contents of rx_buffer */
rx_buffer->page = NULL;
+ rx_buffer->skb = NULL;
+}
+
+static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size)
+{
+ void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+#if (PAGE_SIZE < 8192)
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
+#else
+ unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+#endif
+ struct sk_buff *skb;
+
+ /* prefetch first cache line of first page */
+ prefetch(va);
+#if L1_CACHE_BYTES < 128
+ prefetch(va + L1_CACHE_BYTES);
+#endif
+
+ /* allocate a skb to store the frags */
+ skb = napi_alloc_skb(&rx_ring->q_vector->napi, IXGBE_RX_HDR_SIZE);
+ if (unlikely(!skb))
+ return NULL;
+
+ if (size > IXGBE_RX_HDR_SIZE) {
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
+ IXGBE_CB(skb)->dma = rx_buffer->dma;
+
+ skb_add_rx_frag(skb, 0, rx_buffer->page,
+ rx_buffer->page_offset,
+ size, truesize);
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
+ } else {
+ memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
+ rx_buffer->pagecnt_bias++;
+ }
return skb;
}
@@ -2160,7 +2162,9 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
while (likely(total_rx_packets < budget)) {
union ixgbe_adv_rx_desc *rx_desc;
+ struct ixgbe_rx_buffer *rx_buffer;
struct sk_buff *skb;
+ unsigned int size;
/* return some buffers to hardware, one at a time is too slow */
if (cleaned_count >= IXGBE_RX_BUFFER_WRITE) {
@@ -2169,8 +2173,8 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
}
rx_desc = IXGBE_RX_DESC(rx_ring, rx_ring->next_to_clean);
-
- if (!rx_desc->wb.upper.length)
+ size = le16_to_cpu(rx_desc->wb.upper.length);
+ if (!size)
break;
/* This memory barrier is needed to keep us from reading
@@ -2179,13 +2183,23 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
*/
dma_rmb();
+ rx_buffer = ixgbe_get_rx_buffer(rx_ring, rx_desc, &skb, size);
+
/* retrieve a buffer from the ring */
- skb = ixgbe_fetch_rx_buffer(rx_ring, rx_desc);
+ if (skb)
+ ixgbe_add_rx_frag(rx_ring, rx_buffer, skb, size);
+ else
+ skb = ixgbe_construct_skb(rx_ring, rx_buffer,
+ rx_desc, size);
/* exit if we failed to retrieve a buffer */
- if (!skb)
+ if (!skb) {
+ rx_ring->rx_stats.alloc_rx_buff_failed++;
+ rx_buffer->pagecnt_bias++;
break;
+ }
+ ixgbe_put_rx_buffer(rx_ring, rx_buffer, skb);
cleaned_count++;
/* place incomplete frames back on ring for completion */
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 07/11] ixgbe: Break out Rx buffer page management
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
We are going to be expanding the number of Rx paths in the driver. Instead
of duplicating all that code I am pulling it apart into separate functions
so that we don't have so much code duplication.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 230 +++++++++++++------------
1 file changed, 122 insertions(+), 108 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 6ce596816954..79495cc990c2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1864,7 +1864,6 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
- IXGBE_CB(skb)->page_released = false;
} else {
struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[0];
@@ -1874,7 +1873,6 @@ static void ixgbe_dma_sync_frag(struct ixgbe_ring *rx_ring,
skb_frag_size(frag),
DMA_FROM_DEVICE);
}
- IXGBE_CB(skb)->dma = 0;
}
/**
@@ -1945,8 +1943,14 @@ static void ixgbe_reuse_rx_page(struct ixgbe_ring *rx_ring,
nta++;
rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0;
- /* transfer page from old buffer to new buffer */
- *new_buff = *old_buff;
+ /* Transfer page from old buffer to new buffer.
+ * Move each member individually to avoid possible store
+ * forwarding stalls and unnecessary copy of skb.
+ */
+ new_buff->dma = old_buff->dma;
+ new_buff->page = old_buff->page;
+ new_buff->page_offset = old_buff->page_offset;
+ new_buff->pagecnt_bias = old_buff->pagecnt_bias;
}
static inline bool ixgbe_page_is_reserved(struct page *page)
@@ -1954,15 +1958,10 @@ static inline bool ixgbe_page_is_reserved(struct page *page)
return (page_to_nid(page) != numa_mem_id()) || page_is_pfmemalloc(page);
}
-static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
- struct page *page,
- const unsigned int truesize)
+static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer)
{
-#if (PAGE_SIZE >= 8192)
- unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
- ixgbe_rx_bufsz(rx_ring);
-#endif
- unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
+ unsigned int pagecnt_bias = rx_buffer->pagecnt_bias;
+ struct page *page = rx_buffer->page;
/* avoid re-using remote pages */
if (unlikely(ixgbe_page_is_reserved(page)))
@@ -1970,16 +1969,17 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
#if (PAGE_SIZE < 8192)
/* if we are only owner of page we can reuse it */
- if (unlikely(page_count(page) != pagecnt_bias))
+ if (unlikely((page_ref_count(page) - pagecnt_bias) > 1))
return false;
-
- /* flip page offset to other buffer */
- rx_buffer->page_offset ^= truesize;
#else
- /* move offset up to the next cache line */
- rx_buffer->page_offset += truesize;
-
- if (rx_buffer->page_offset > last_offset)
+ /* The last offset is a bit aggressive in that we assume the
+ * worst case of FCoE being enabled and using a 3K buffer.
+ * However this should have minimal impact as the 1K extra is
+ * still less than one buffer in size.
+ */
+#define IXGBE_LAST_OFFSET \
+ (SKB_WITH_OVERHEAD(PAGE_SIZE) - IXGBE_RXBUFFER_3K)
+ if (rx_buffer->page_offset > IXGBE_LAST_OFFSET)
return false;
#endif
@@ -1987,7 +1987,7 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
* the pagecnt_bias and page count so that we fully restock the
* number of references the driver holds.
*/
- if (unlikely(pagecnt_bias == 1)) {
+ if (unlikely(!pagecnt_bias)) {
page_ref_add(page, USHRT_MAX);
rx_buffer->pagecnt_bias = USHRT_MAX;
}
@@ -2010,106 +2010,65 @@ static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
* The function will then update the page offset if necessary and return
* true if the buffer can be reused by the adapter.
**/
-static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
+static void ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *rx_buffer,
- unsigned int size,
- struct sk_buff *skb)
+ struct sk_buff *skb,
+ unsigned int size)
{
- struct page *page = rx_buffer->page;
- unsigned char *va = page_address(page) + rx_buffer->page_offset;
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
-
- if (unlikely(skb_is_nonlinear(skb)))
- goto add_tail_frag;
-
- if (size <= IXGBE_RX_HDR_SIZE) {
- memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
-
- /* page is not reserved, we can reuse buffer as-is */
- if (likely(!ixgbe_page_is_reserved(page)))
- return true;
-
- /* this page cannot be reused so discard it */
- return false;
- }
-
-add_tail_frag:
- skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
+ skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
rx_buffer->page_offset, size, truesize);
-
- return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
}
-static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
- union ixgbe_adv_rx_desc *rx_desc)
+static struct ixgbe_rx_buffer *ixgbe_get_rx_buffer(struct ixgbe_ring *rx_ring,
+ union ixgbe_adv_rx_desc *rx_desc,
+ struct sk_buff **skb,
+ const unsigned int size)
{
- unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
struct ixgbe_rx_buffer *rx_buffer;
- struct sk_buff *skb;
- struct page *page;
rx_buffer = &rx_ring->rx_buffer_info[rx_ring->next_to_clean];
- page = rx_buffer->page;
- prefetchw(page);
-
- skb = rx_buffer->skb;
-
- if (likely(!skb)) {
- void *page_addr = page_address(page) +
- rx_buffer->page_offset;
+ prefetchw(rx_buffer->page);
+ *skb = rx_buffer->skb;
- /* prefetch first cache line of first page */
- prefetch(page_addr);
-#if L1_CACHE_BYTES < 128
- prefetch(page_addr + L1_CACHE_BYTES);
-#endif
-
- /* allocate a skb to store the frags */
- skb = napi_alloc_skb(&rx_ring->q_vector->napi,
- IXGBE_RX_HDR_SIZE);
- if (unlikely(!skb)) {
- rx_ring->rx_stats.alloc_rx_buff_failed++;
- return NULL;
- }
-
- /*
- * we will be copying header into skb->data in
- * pskb_may_pull so it is in our interest to prefetch
- * it now to avoid a possible cache miss
- */
- prefetchw(skb->data);
-
- /*
- * Delay unmapping of the first packet. It carries the
- * header information, HW may still access the header
- * after the writeback. Only unmap it when EOP is
- * reached
- */
- if (likely(ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)))
- goto dma_sync;
-
- IXGBE_CB(skb)->dma = rx_buffer->dma;
+ /* Delay unmapping of the first packet. It carries the header
+ * information, HW may still access the header after the writeback.
+ * Only unmap it when EOP is reached
+ */
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)) {
+ if (!*skb)
+ goto skip_sync;
} else {
- if (ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
- ixgbe_dma_sync_frag(rx_ring, skb);
+ if (*skb)
+ ixgbe_dma_sync_frag(rx_ring, *skb);
+ }
-dma_sync:
- /* we are reusing so sync this buffer for CPU use */
- dma_sync_single_range_for_cpu(rx_ring->dev,
- rx_buffer->dma,
- rx_buffer->page_offset,
- size,
- DMA_FROM_DEVICE);
+ /* we are reusing so sync this buffer for CPU use */
+ dma_sync_single_range_for_cpu(rx_ring->dev,
+ rx_buffer->dma,
+ rx_buffer->page_offset,
+ size,
+ DMA_FROM_DEVICE);
+skip_sync:
+ rx_buffer->pagecnt_bias--;
- rx_buffer->skb = NULL;
- }
+ return rx_buffer;
+}
- /* pull page into skb */
- if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
+static void ixgbe_put_rx_buffer(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ struct sk_buff *skb)
+{
+ if (ixgbe_can_reuse_rx_page(rx_buffer)) {
/* hand second half of page back to the ring */
ixgbe_reuse_rx_page(rx_ring, rx_buffer);
} else {
@@ -2123,12 +2082,55 @@ static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
}
- __page_frag_cache_drain(page,
+ __page_frag_cache_drain(rx_buffer->page,
rx_buffer->pagecnt_bias);
}
- /* clear contents of buffer_info */
+ /* clear contents of rx_buffer */
rx_buffer->page = NULL;
+ rx_buffer->skb = NULL;
+}
+
+static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size)
+{
+ void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+#if (PAGE_SIZE < 8192)
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
+#else
+ unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+#endif
+ struct sk_buff *skb;
+
+ /* prefetch first cache line of first page */
+ prefetch(va);
+#if L1_CACHE_BYTES < 128
+ prefetch(va + L1_CACHE_BYTES);
+#endif
+
+ /* allocate a skb to store the frags */
+ skb = napi_alloc_skb(&rx_ring->q_vector->napi, IXGBE_RX_HDR_SIZE);
+ if (unlikely(!skb))
+ return NULL;
+
+ if (size > IXGBE_RX_HDR_SIZE) {
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
+ IXGBE_CB(skb)->dma = rx_buffer->dma;
+
+ skb_add_rx_frag(skb, 0, rx_buffer->page,
+ rx_buffer->page_offset,
+ size, truesize);
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
+ } else {
+ memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
+ rx_buffer->pagecnt_bias++;
+ }
return skb;
}
@@ -2160,7 +2162,9 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
while (likely(total_rx_packets < budget)) {
union ixgbe_adv_rx_desc *rx_desc;
+ struct ixgbe_rx_buffer *rx_buffer;
struct sk_buff *skb;
+ unsigned int size;
/* return some buffers to hardware, one@a time is too slow */
if (cleaned_count >= IXGBE_RX_BUFFER_WRITE) {
@@ -2169,8 +2173,8 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
}
rx_desc = IXGBE_RX_DESC(rx_ring, rx_ring->next_to_clean);
-
- if (!rx_desc->wb.upper.length)
+ size = le16_to_cpu(rx_desc->wb.upper.length);
+ if (!size)
break;
/* This memory barrier is needed to keep us from reading
@@ -2179,13 +2183,23 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
*/
dma_rmb();
+ rx_buffer = ixgbe_get_rx_buffer(rx_ring, rx_desc, &skb, size);
+
/* retrieve a buffer from the ring */
- skb = ixgbe_fetch_rx_buffer(rx_ring, rx_desc);
+ if (skb)
+ ixgbe_add_rx_frag(rx_ring, rx_buffer, skb, size);
+ else
+ skb = ixgbe_construct_skb(rx_ring, rx_buffer,
+ rx_desc, size);
/* exit if we failed to retrieve a buffer */
- if (!skb)
+ if (!skb) {
+ rx_ring->rx_stats.alloc_rx_buff_failed++;
+ rx_buffer->pagecnt_bias++;
break;
+ }
+ ixgbe_put_rx_buffer(rx_ring, rx_buffer, skb);
cleaned_count++;
/* place incomplete frames back on ring for completion */
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 08/11] ixgbe: Add support for padding packet
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds support for providing a buffer with headroom and tailroom
to allow for shared info, NET_SKB_PAD, and NET_IP_ALIGN. With this
combined with the DMA changes we can start using build_skb to build frames
around an incoming Rx buffer instead of having to memcpy the headers.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 17 ++++++++++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 43 +++++++++++++++++++++++--
2 files changed, 56 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 80328e657d6a..3537d07b4807 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -94,6 +94,14 @@
#define IXGBE_RXBUFFER_4K 4096
#define IXGBE_MAX_RXBUFFER 16384 /* largest size for a single descriptor */
+#define IXGBE_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN)
+#if (PAGE_SIZE < 8192)
+#define IXGBE_MAX_FRAME_BUILD_SKB \
+ (SKB_WITH_OVERHEAD(IXGBE_RXBUFFER_2K) - IXGBE_SKB_PAD)
+#else
+#define IGB_MAX_FRAME_BUILD_SKB IXGBE_RXBUFFER_2K
+#endif
+
/*
* NOTE: netdev_alloc_skb reserves up to 64 bytes, NET_IP_ALIGN means we
* reserve 64 more, and skb_shared_info adds an additional 320 bytes more,
@@ -235,6 +243,7 @@ struct ixgbe_rx_queue_stats {
enum ixgbe_ring_state_t {
__IXGBE_RX_3K_BUFFER,
+ __IXGBE_RX_BUILD_SKB_ENABLED,
__IXGBE_RX_RSC_ENABLED,
__IXGBE_RX_CSUM_UDP_ZERO_ERR,
__IXGBE_RX_FCOE,
@@ -244,6 +253,9 @@ enum ixgbe_ring_state_t {
__IXGBE_HANG_CHECK_ARMED,
};
+#define ring_uses_build_skb(ring) \
+ test_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &(ring)->state)
+
struct ixgbe_fwd_adapter {
unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
struct net_device *netdev;
@@ -355,6 +367,10 @@ static inline unsigned int ixgbe_rx_bufsz(struct ixgbe_ring *ring)
{
if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
return IXGBE_RXBUFFER_3K;
+#if (PAGE_SIZE < 8192)
+ if (ring_uses_build_skb(ring))
+ return IXGBE_MAX_FRAME_BUILD_SKB;
+#endif
return IXGBE_RXBUFFER_2K;
}
@@ -670,6 +686,7 @@ struct ixgbe_adapter {
#define IXGBE_FLAG2_VLAN_PROMISC BIT(13)
#define IXGBE_FLAG2_EEE_CAPABLE BIT(14)
#define IXGBE_FLAG2_EEE_ENABLED BIT(15)
+#define IXGBE_FLAG2_RX_LEGACY BIT(16)
/* Tx fast path data */
int num_tx_queues;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 79495cc990c2..8529eafb9717 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1565,6 +1565,11 @@ static inline void ixgbe_rx_checksum(struct ixgbe_ring *ring,
}
}
+static inline unsigned int ixgbe_rx_offset(struct ixgbe_ring *rx_ring)
+{
+ return ring_uses_build_skb(rx_ring) ? IXGBE_SKB_PAD : 0;
+}
+
static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *bi)
{
@@ -1601,7 +1606,7 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
bi->dma = dma;
bi->page = page;
- bi->page_offset = 0;
+ bi->page_offset = ixgbe_rx_offset(rx_ring);
bi->pagecnt_bias = 1;
return true;
@@ -2018,7 +2023,9 @@ static void ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = SKB_DATA_ALIGN(size);
+ unsigned int truesize = ring_uses_build_skb(rx_ring) ?
+ SKB_DATA_ALIGN(IXGBE_SKB_PAD + size) :
+ SKB_DATA_ALIGN(size);
#endif
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
rx_buffer->page_offset, size, truesize);
@@ -2100,7 +2107,7 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+ unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
struct sk_buff *skb;
@@ -3462,7 +3469,10 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter,
srrctl = IXGBE_RX_HDR_SIZE << IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT;
/* configure the packet buffer length */
- srrctl |= ixgbe_rx_bufsz(rx_ring) >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state))
+ srrctl |= IXGBE_RXBUFFER_3K >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
+ else
+ srrctl |= IXGBE_RXBUFFER_2K >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
/* configure descriptor type */
srrctl |= IXGBE_SRRCTL_DESCTYPE_ADV_ONEBUF;
@@ -3796,6 +3806,17 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
*/
rxdctl &= ~0x3FFFFF;
rxdctl |= 0x080420;
+#if (PAGE_SIZE < 8192)
+ } else {
+ rxdctl &= ~(IXGBE_RXDCTL_RLPMLMASK |
+ IXGBE_RXDCTL_RLPML_EN);
+
+ /* Limit the maximum frame size so we don't overrun the skb */
+ if (ring_uses_build_skb(ring) &&
+ !test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ rxdctl |= IXGBE_MAX_FRAME_BUILD_SKB |
+ IXGBE_RXDCTL_RLPML_EN;
+#endif
}
/* initialize Rx descriptor 0 */
@@ -3941,12 +3962,26 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
clear_ring_rsc_enabled(rx_ring);
clear_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+ clear_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
set_ring_rsc_enabled(rx_ring);
if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
+ if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
+ continue;
+
+ set_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
+
+#if (PAGE_SIZE < 8192)
+ if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
+ if (max_frame > (ETH_FRAME_LEN + ETH_FCS_LEN))
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+#endif
}
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 08/11] ixgbe: Add support for padding packet
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds support for providing a buffer with headroom and tailroom
to allow for shared info, NET_SKB_PAD, and NET_IP_ALIGN. With this
combined with the DMA changes we can start using build_skb to build frames
around an incoming Rx buffer instead of having to memcpy the headers.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 17 ++++++++++
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 43 +++++++++++++++++++++++--
2 files changed, 56 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
index 80328e657d6a..3537d07b4807 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h
@@ -94,6 +94,14 @@
#define IXGBE_RXBUFFER_4K 4096
#define IXGBE_MAX_RXBUFFER 16384 /* largest size for a single descriptor */
+#define IXGBE_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN)
+#if (PAGE_SIZE < 8192)
+#define IXGBE_MAX_FRAME_BUILD_SKB \
+ (SKB_WITH_OVERHEAD(IXGBE_RXBUFFER_2K) - IXGBE_SKB_PAD)
+#else
+#define IGB_MAX_FRAME_BUILD_SKB IXGBE_RXBUFFER_2K
+#endif
+
/*
* NOTE: netdev_alloc_skb reserves up to 64 bytes, NET_IP_ALIGN means we
* reserve 64 more, and skb_shared_info adds an additional 320 bytes more,
@@ -235,6 +243,7 @@ struct ixgbe_rx_queue_stats {
enum ixgbe_ring_state_t {
__IXGBE_RX_3K_BUFFER,
+ __IXGBE_RX_BUILD_SKB_ENABLED,
__IXGBE_RX_RSC_ENABLED,
__IXGBE_RX_CSUM_UDP_ZERO_ERR,
__IXGBE_RX_FCOE,
@@ -244,6 +253,9 @@ enum ixgbe_ring_state_t {
__IXGBE_HANG_CHECK_ARMED,
};
+#define ring_uses_build_skb(ring) \
+ test_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &(ring)->state)
+
struct ixgbe_fwd_adapter {
unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
struct net_device *netdev;
@@ -355,6 +367,10 @@ static inline unsigned int ixgbe_rx_bufsz(struct ixgbe_ring *ring)
{
if (test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
return IXGBE_RXBUFFER_3K;
+#if (PAGE_SIZE < 8192)
+ if (ring_uses_build_skb(ring))
+ return IXGBE_MAX_FRAME_BUILD_SKB;
+#endif
return IXGBE_RXBUFFER_2K;
}
@@ -670,6 +686,7 @@ struct ixgbe_adapter {
#define IXGBE_FLAG2_VLAN_PROMISC BIT(13)
#define IXGBE_FLAG2_EEE_CAPABLE BIT(14)
#define IXGBE_FLAG2_EEE_ENABLED BIT(15)
+#define IXGBE_FLAG2_RX_LEGACY BIT(16)
/* Tx fast path data */
int num_tx_queues;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 79495cc990c2..8529eafb9717 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1565,6 +1565,11 @@ static inline void ixgbe_rx_checksum(struct ixgbe_ring *ring,
}
}
+static inline unsigned int ixgbe_rx_offset(struct ixgbe_ring *rx_ring)
+{
+ return ring_uses_build_skb(rx_ring) ? IXGBE_SKB_PAD : 0;
+}
+
static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
struct ixgbe_rx_buffer *bi)
{
@@ -1601,7 +1606,7 @@ static bool ixgbe_alloc_mapped_page(struct ixgbe_ring *rx_ring,
bi->dma = dma;
bi->page = page;
- bi->page_offset = 0;
+ bi->page_offset = ixgbe_rx_offset(rx_ring);
bi->pagecnt_bias = 1;
return true;
@@ -2018,7 +2023,9 @@ static void ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = SKB_DATA_ALIGN(size);
+ unsigned int truesize = ring_uses_build_skb(rx_ring) ?
+ SKB_DATA_ALIGN(IXGBE_SKB_PAD + size) :
+ SKB_DATA_ALIGN(size);
#endif
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
rx_buffer->page_offset, size, truesize);
@@ -2100,7 +2107,7 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
#if (PAGE_SIZE < 8192)
unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
#else
- unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
+ unsigned int truesize = SKB_DATA_ALIGN(size);
#endif
struct sk_buff *skb;
@@ -3462,7 +3469,10 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter,
srrctl = IXGBE_RX_HDR_SIZE << IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT;
/* configure the packet buffer length */
- srrctl |= ixgbe_rx_bufsz(rx_ring) >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
+ if (test_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state))
+ srrctl |= IXGBE_RXBUFFER_3K >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
+ else
+ srrctl |= IXGBE_RXBUFFER_2K >> IXGBE_SRRCTL_BSIZEPKT_SHIFT;
/* configure descriptor type */
srrctl |= IXGBE_SRRCTL_DESCTYPE_ADV_ONEBUF;
@@ -3796,6 +3806,17 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
*/
rxdctl &= ~0x3FFFFF;
rxdctl |= 0x080420;
+#if (PAGE_SIZE < 8192)
+ } else {
+ rxdctl &= ~(IXGBE_RXDCTL_RLPMLMASK |
+ IXGBE_RXDCTL_RLPML_EN);
+
+ /* Limit the maximum frame size so we don't overrun the skb */
+ if (ring_uses_build_skb(ring) &&
+ !test_bit(__IXGBE_RX_3K_BUFFER, &ring->state))
+ rxdctl |= IXGBE_MAX_FRAME_BUILD_SKB |
+ IXGBE_RXDCTL_RLPML_EN;
+#endif
}
/* initialize Rx descriptor 0 */
@@ -3941,12 +3962,26 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
clear_ring_rsc_enabled(rx_ring);
clear_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+ clear_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
set_ring_rsc_enabled(rx_ring);
if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
+ if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
+ continue;
+
+ set_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
+
+#if (PAGE_SIZE < 8192)
+ if (adapter->flags2 & IXGBE_FLAG2_RSC_ENABLED)
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+
+ if (max_frame > (ETH_FRAME_LEN + ETH_FCS_LEN))
+ set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+#endif
}
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 09/11] ixgbe: Add private flag to control buffer mode
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
Since there are potential drawbacks to the new Rx allocation approach I
thought it best to add a "chicken bit" so that we can turn the feature off
if in the event that a problem is found.
It also provides a means of validating the legacy Rx path in the event that
we are forced to fall back. At some point in the future when we are
convinced we don't need it anymore we might be able to drop the legacy-rx
flag.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 47 ++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index aa80ece9ea96..6466d828bc6f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -151,6 +151,13 @@ struct ixgbe_stats {
};
#define IXGBE_TEST_LEN sizeof(ixgbe_gstrings_test) / ETH_GSTRING_LEN
+static const char ixgbe_priv_flags_strings[][ETH_GSTRING_LEN] = {
+#define IXGBE_PRIV_FLAGS_LEGACY_RX BIT(0)
+ "legacy-rx",
+};
+
+#define IXGBE_PRIV_FLAGS_STR_LEN ARRAY_SIZE(ixgbe_priv_flags_strings)
+
/* currently supported speeds for 10G */
#define ADVRTSD_MSK_10G (SUPPORTED_10000baseT_Full | \
SUPPORTED_10000baseKX4_Full | \
@@ -1002,6 +1009,8 @@ static void ixgbe_get_drvinfo(struct net_device *netdev,
strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
sizeof(drvinfo->bus_info));
+
+ drvinfo->n_priv_flags = IXGBE_PRIV_FLAGS_STR_LEN;
}
static void ixgbe_get_ringparam(struct net_device *netdev,
@@ -1141,6 +1150,8 @@ static int ixgbe_get_sset_count(struct net_device *netdev, int sset)
return IXGBE_TEST_LEN;
case ETH_SS_STATS:
return IXGBE_STATS_LEN;
+ case ETH_SS_PRIV_FLAGS:
+ return IXGBE_PRIV_FLAGS_STR_LEN;
default:
return -EOPNOTSUPP;
}
@@ -1305,6 +1316,9 @@ static void ixgbe_get_strings(struct net_device *netdev, u32 stringset,
}
/* BUG_ON(p - data != IXGBE_STATS_LEN * ETH_GSTRING_LEN); */
break;
+ case ETH_SS_PRIV_FLAGS:
+ memcpy(data, ixgbe_priv_flags_strings,
+ IXGBE_PRIV_FLAGS_STR_LEN * ETH_GSTRING_LEN);
}
}
@@ -3386,6 +3400,37 @@ static int ixgbe_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
return 0;
}
+static u32 ixgbe_get_priv_flags(struct net_device *netdev)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(netdev);
+ u32 priv_flags = 0;
+
+ if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
+ priv_flags |= IXGBE_PRIV_FLAGS_LEGACY_RX;
+
+ return priv_flags;
+}
+
+static int ixgbe_set_priv_flags(struct net_device *netdev, u32 priv_flags)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(netdev);
+ unsigned int flags2 = adapter->flags2;
+
+ flags2 &= ~IXGBE_FLAG2_RX_LEGACY;
+ if (priv_flags & IXGBE_PRIV_FLAGS_LEGACY_RX)
+ flags2 |= IXGBE_FLAG2_RX_LEGACY;
+
+ if (flags2 != adapter->flags2) {
+ adapter->flags2 = flags2;
+
+ /* reset interface to repopulate queues */
+ if (netif_running(netdev))
+ ixgbe_reinit_locked(adapter);
+ }
+
+ return 0;
+}
+
static const struct ethtool_ops ixgbe_ethtool_ops = {
.get_settings = ixgbe_get_settings,
.set_settings = ixgbe_set_settings,
@@ -3422,6 +3467,8 @@ static int ixgbe_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
.set_eee = ixgbe_set_eee,
.get_channels = ixgbe_get_channels,
.set_channels = ixgbe_set_channels,
+ .get_priv_flags = ixgbe_get_priv_flags,
+ .set_priv_flags = ixgbe_set_priv_flags,
.get_ts_info = ixgbe_get_ts_info,
.get_module_info = ixgbe_get_module_info,
.get_module_eeprom = ixgbe_get_module_eeprom,
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 09/11] ixgbe: Add private flag to control buffer mode
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
Since there are potential drawbacks to the new Rx allocation approach I
thought it best to add a "chicken bit" so that we can turn the feature off
if in the event that a problem is found.
It also provides a means of validating the legacy Rx path in the event that
we are forced to fall back. At some point in the future when we are
convinced we don't need it anymore we might be able to drop the legacy-rx
flag.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 47 ++++++++++++++++++++++
1 file changed, 47 insertions(+)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index aa80ece9ea96..6466d828bc6f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -151,6 +151,13 @@ struct ixgbe_stats {
};
#define IXGBE_TEST_LEN sizeof(ixgbe_gstrings_test) / ETH_GSTRING_LEN
+static const char ixgbe_priv_flags_strings[][ETH_GSTRING_LEN] = {
+#define IXGBE_PRIV_FLAGS_LEGACY_RX BIT(0)
+ "legacy-rx",
+};
+
+#define IXGBE_PRIV_FLAGS_STR_LEN ARRAY_SIZE(ixgbe_priv_flags_strings)
+
/* currently supported speeds for 10G */
#define ADVRTSD_MSK_10G (SUPPORTED_10000baseT_Full | \
SUPPORTED_10000baseKX4_Full | \
@@ -1002,6 +1009,8 @@ static void ixgbe_get_drvinfo(struct net_device *netdev,
strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
sizeof(drvinfo->bus_info));
+
+ drvinfo->n_priv_flags = IXGBE_PRIV_FLAGS_STR_LEN;
}
static void ixgbe_get_ringparam(struct net_device *netdev,
@@ -1141,6 +1150,8 @@ static int ixgbe_get_sset_count(struct net_device *netdev, int sset)
return IXGBE_TEST_LEN;
case ETH_SS_STATS:
return IXGBE_STATS_LEN;
+ case ETH_SS_PRIV_FLAGS:
+ return IXGBE_PRIV_FLAGS_STR_LEN;
default:
return -EOPNOTSUPP;
}
@@ -1305,6 +1316,9 @@ static void ixgbe_get_strings(struct net_device *netdev, u32 stringset,
}
/* BUG_ON(p - data != IXGBE_STATS_LEN * ETH_GSTRING_LEN); */
break;
+ case ETH_SS_PRIV_FLAGS:
+ memcpy(data, ixgbe_priv_flags_strings,
+ IXGBE_PRIV_FLAGS_STR_LEN * ETH_GSTRING_LEN);
}
}
@@ -3386,6 +3400,37 @@ static int ixgbe_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
return 0;
}
+static u32 ixgbe_get_priv_flags(struct net_device *netdev)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(netdev);
+ u32 priv_flags = 0;
+
+ if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
+ priv_flags |= IXGBE_PRIV_FLAGS_LEGACY_RX;
+
+ return priv_flags;
+}
+
+static int ixgbe_set_priv_flags(struct net_device *netdev, u32 priv_flags)
+{
+ struct ixgbe_adapter *adapter = netdev_priv(netdev);
+ unsigned int flags2 = adapter->flags2;
+
+ flags2 &= ~IXGBE_FLAG2_RX_LEGACY;
+ if (priv_flags & IXGBE_PRIV_FLAGS_LEGACY_RX)
+ flags2 |= IXGBE_FLAG2_RX_LEGACY;
+
+ if (flags2 != adapter->flags2) {
+ adapter->flags2 = flags2;
+
+ /* reset interface to repopulate queues */
+ if (netif_running(netdev))
+ ixgbe_reinit_locked(adapter);
+ }
+
+ return 0;
+}
+
static const struct ethtool_ops ixgbe_ethtool_ops = {
.get_settings = ixgbe_get_settings,
.set_settings = ixgbe_set_settings,
@@ -3422,6 +3467,8 @@ static int ixgbe_set_eee(struct net_device *netdev, struct ethtool_eee *edata)
.set_eee = ixgbe_set_eee,
.get_channels = ixgbe_get_channels,
.set_channels = ixgbe_set_channels,
+ .get_priv_flags = ixgbe_get_priv_flags,
+ .set_priv_flags = ixgbe_set_priv_flags,
.get_ts_info = ixgbe_get_ts_info,
.get_module_info = ixgbe_get_module_info,
.get_module_eeprom = ixgbe_get_module_eeprom,
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 10/11] ixgbe: Add support for build_skb
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds build_skb support to the Rx path. There are several
advantages to this change.
1. It avoids the memcpy and skb->head allocation for small packets which
improves performance by about 5% in my tests.
2. It avoids the memcpy, skb->head allocation, and eth_get_headlen
for larger packets improving performance by about 10% in my tests.
3. For VXLAN packets it allows the full header to be in skb->data which
improves the performance by as much as 30% in some of my tests.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 49 ++++++++++++++++++++++++-
1 file changed, 48 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 8529eafb9717..dd04b76ae0bd 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1913,7 +1913,7 @@ static bool ixgbe_cleanup_headers(struct ixgbe_ring *rx_ring,
}
/* place header in linear portion of buffer */
- if (skb_is_nonlinear(skb))
+ if (!skb_headlen(skb))
ixgbe_pull_tail(rx_ring, skb);
#ifdef IXGBE_FCOE
@@ -2142,6 +2142,49 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
return skb;
}
+static struct sk_buff *ixgbe_build_skb(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size)
+{
+ void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+#if (PAGE_SIZE < 8192)
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
+#else
+ unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
+ SKB_DATA_ALIGN(IXGBE_SKB_PAD + size);
+#endif
+ struct sk_buff *skb;
+
+ /* prefetch first cache line of first page */
+ prefetch(va);
+#if L1_CACHE_BYTES < 128
+ prefetch(va + L1_CACHE_BYTES);
+#endif
+
+ /* build an skb to around the page buffer */
+ skb = build_skb(va - IXGBE_SKB_PAD, truesize);
+ if (unlikely(!skb))
+ return NULL;
+
+ /* update pointers within the skb to store the data */
+ skb_reserve(skb, IXGBE_SKB_PAD);
+ __skb_put(skb, size);
+
+ /* record DMA address if this is the start of a chain of buffers */
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
+ IXGBE_CB(skb)->dma = rx_buffer->dma;
+
+ /* update buffer offset */
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
+
+ return skb;
+}
+
/**
* ixgbe_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf
* @q_vector: structure containing interrupt and ring information
@@ -2195,6 +2238,9 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
/* retrieve a buffer from the ring */
if (skb)
ixgbe_add_rx_frag(rx_ring, rx_buffer, skb, size);
+ else if (ring_uses_build_skb(rx_ring))
+ skb = ixgbe_build_skb(rx_ring, rx_buffer,
+ rx_desc, size);
else
skb = ixgbe_construct_skb(rx_ring, rx_buffer,
rx_desc, size);
@@ -3970,6 +4016,7 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+ clear_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
continue;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 10/11] ixgbe: Add support for build_skb
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch adds build_skb support to the Rx path. There are several
advantages to this change.
1. It avoids the memcpy and skb->head allocation for small packets which
improves performance by about 5% in my tests.
2. It avoids the memcpy, skb->head allocation, and eth_get_headlen
for larger packets improving performance by about 10% in my tests.
3. For VXLAN packets it allows the full header to be in skb->data which
improves the performance by as much as 30% in some of my tests.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 49 ++++++++++++++++++++++++-
1 file changed, 48 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 8529eafb9717..dd04b76ae0bd 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -1913,7 +1913,7 @@ static bool ixgbe_cleanup_headers(struct ixgbe_ring *rx_ring,
}
/* place header in linear portion of buffer */
- if (skb_is_nonlinear(skb))
+ if (!skb_headlen(skb))
ixgbe_pull_tail(rx_ring, skb);
#ifdef IXGBE_FCOE
@@ -2142,6 +2142,49 @@ static struct sk_buff *ixgbe_construct_skb(struct ixgbe_ring *rx_ring,
return skb;
}
+static struct sk_buff *ixgbe_build_skb(struct ixgbe_ring *rx_ring,
+ struct ixgbe_rx_buffer *rx_buffer,
+ union ixgbe_adv_rx_desc *rx_desc,
+ unsigned int size)
+{
+ void *va = page_address(rx_buffer->page) + rx_buffer->page_offset;
+#if (PAGE_SIZE < 8192)
+ unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2;
+#else
+ unsigned int truesize = SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) +
+ SKB_DATA_ALIGN(IXGBE_SKB_PAD + size);
+#endif
+ struct sk_buff *skb;
+
+ /* prefetch first cache line of first page */
+ prefetch(va);
+#if L1_CACHE_BYTES < 128
+ prefetch(va + L1_CACHE_BYTES);
+#endif
+
+ /* build an skb to around the page buffer */
+ skb = build_skb(va - IXGBE_SKB_PAD, truesize);
+ if (unlikely(!skb))
+ return NULL;
+
+ /* update pointers within the skb to store the data */
+ skb_reserve(skb, IXGBE_SKB_PAD);
+ __skb_put(skb, size);
+
+ /* record DMA address if this is the start of a chain of buffers */
+ if (!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
+ IXGBE_CB(skb)->dma = rx_buffer->dma;
+
+ /* update buffer offset */
+#if (PAGE_SIZE < 8192)
+ rx_buffer->page_offset ^= truesize;
+#else
+ rx_buffer->page_offset += truesize;
+#endif
+
+ return skb;
+}
+
/**
* ixgbe_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf
* @q_vector: structure containing interrupt and ring information
@@ -2195,6 +2238,9 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
/* retrieve a buffer from the ring */
if (skb)
ixgbe_add_rx_frag(rx_ring, rx_buffer, skb, size);
+ else if (ring_uses_build_skb(rx_ring))
+ skb = ixgbe_build_skb(rx_ring, rx_buffer,
+ rx_desc, size);
else
skb = ixgbe_construct_skb(rx_ring, rx_buffer,
rx_desc, size);
@@ -3970,6 +4016,7 @@ static void ixgbe_set_rx_buffer_len(struct ixgbe_adapter *adapter)
if (test_bit(__IXGBE_RX_FCOE, &rx_ring->state))
set_bit(__IXGBE_RX_3K_BUFFER, &rx_ring->state);
+ clear_bit(__IXGBE_RX_BUILD_SKB_ENABLED, &rx_ring->state);
if (adapter->flags2 & IXGBE_FLAG2_RX_LEGACY)
continue;
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [next PATCH 11/11] ixgbe: Don't bother clearing buffer memory for descriptor rings
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 16:07 ` Alexander Duyck
-1 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan, jeffrey.t.kirsher; +Cc: netdev
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch makes it so that we don't need to bother with clearing the
memory out for the descriptor rings. The general idea is to only free
buffers associated with buffers in use which are located between the
next_to_clean and next_to_use or next_to_alloc values. Everything outside
of those regions can be safely ignored since they should have no buffers
associated with them.
The advantage to doing things this way is that is should speed up bring-up
and tear-down of the rings. Specifically we can avoid the 512 or more
cycles required to memset the rings in tear-down. In the bring-up phase we
then clear the memory as a part of initialization. The general idea is
that the clearing in initialization can act as a prefetch of sorts for the
buffer info structures so they are in the local CPU when we go to populate
them. This should help to improve overall time needed to perform a
suspend/resume.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 11 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 158 ++++++++++++----------
2 files changed, 98 insertions(+), 71 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6466d828bc6f..8c6f40d4db6f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -1945,7 +1945,16 @@ static u16 ixgbe_clean_test_rings(struct ixgbe_ring *rx_ring,
/* unmap buffer on Tx side */
tx_buffer = &tx_ring->tx_buffer_info[tx_ntc];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer);
+
+ /* Free all the Tx ring sk_buffs */
+ dev_kfree_skb_any(tx_buffer->skb);
+
+ /* unmap skb header data */
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
/* increment Rx/Tx next to clean counters */
rx_ntc++;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dd04b76ae0bd..6ec0ebf7f174 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -958,28 +958,6 @@ static inline void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter,
}
}
-void ixgbe_unmap_and_free_tx_resource(struct ixgbe_ring *ring,
- struct ixgbe_tx_buffer *tx_buffer)
-{
- if (tx_buffer->skb) {
- dev_kfree_skb_any(tx_buffer->skb);
- if (dma_unmap_len(tx_buffer, len))
- dma_unmap_single(ring->dev,
- dma_unmap_addr(tx_buffer, dma),
- dma_unmap_len(tx_buffer, len),
- DMA_TO_DEVICE);
- } else if (dma_unmap_len(tx_buffer, len)) {
- dma_unmap_page(ring->dev,
- dma_unmap_addr(tx_buffer, dma),
- dma_unmap_len(tx_buffer, len),
- DMA_TO_DEVICE);
- }
- tx_buffer->next_to_watch = NULL;
- tx_buffer->skb = NULL;
- dma_unmap_len_set(tx_buffer, len, 0);
- /* tx_buffer must be completely set up in the transmit path */
-}
-
static void ixgbe_update_xoff_rx_lfc(struct ixgbe_adapter *adapter)
{
struct ixgbe_hw *hw = &adapter->hw;
@@ -1211,7 +1189,6 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
DMA_TO_DEVICE);
/* clear tx_buffer data */
- tx_buffer->skb = NULL;
dma_unmap_len_set(tx_buffer, len, 0);
/* unmap remaining buffers */
@@ -3345,6 +3322,10 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
clear_bit(__IXGBE_HANG_CHECK_ARMED, &ring->state);
+ /* reinitialize tx_buffer_info */
+ memset(ring->tx_buffer_info, 0,
+ sizeof(struct ixgbe_tx_buffer) * ring->count);
+
/* enable queue */
IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx), txdctl);
@@ -3865,6 +3846,10 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
#endif
}
+ /* initialize rx_buffer_info */
+ memset(ring->rx_buffer_info, 0,
+ sizeof(struct ixgbe_rx_buffer) * ring->count);
+
/* initialize Rx descriptor 0 */
rx_desc = IXGBE_RX_DESC(ring, 0);
rx_desc->wb.upper.length = 0;
@@ -5049,33 +5034,22 @@ static void ixgbe_fwd_psrtype(struct ixgbe_fwd_adapter *vadapter)
**/
static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
{
- struct device *dev = rx_ring->dev;
- unsigned long size;
- u16 i;
-
- /* ring already cleared, nothing to do */
- if (!rx_ring->rx_buffer_info)
- return;
+ u16 i = rx_ring->next_to_clean;
+ struct ixgbe_rx_buffer *rx_buffer = &rx_ring->rx_buffer_info[i];
/* Free all the Rx ring sk_buffs */
- for (i = 0; i < rx_ring->count; i++) {
- struct ixgbe_rx_buffer *rx_buffer = &rx_ring->rx_buffer_info[i];
-
+ while (i != rx_ring->next_to_alloc) {
if (rx_buffer->skb) {
struct sk_buff *skb = rx_buffer->skb;
if (IXGBE_CB(skb)->page_released)
- dma_unmap_page_attrs(dev,
+ dma_unmap_page_attrs(rx_ring->dev,
IXGBE_CB(skb)->dma,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
dev_kfree_skb(skb);
- rx_buffer->skb = NULL;
}
- if (!rx_buffer->page)
- continue;
-
/* Invalidate cache lines that may have been written to by
* device so that we avoid corrupting memory.
*/
@@ -5086,19 +5060,21 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
DMA_FROM_DEVICE);
/* free resources associated with mapping */
- dma_unmap_page_attrs(dev, rx_buffer->dma,
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
__page_frag_cache_drain(rx_buffer->page,
rx_buffer->pagecnt_bias);
- rx_buffer->page = NULL;
+ i++;
+ rx_buffer++;
+ if (i == rx_ring->count) {
+ i = 0;
+ rx_buffer = rx_ring->rx_buffer_info;
+ }
}
- size = sizeof(struct ixgbe_rx_buffer) * rx_ring->count;
- memset(rx_ring->rx_buffer_info, 0, size);
-
rx_ring->next_to_alloc = 0;
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
@@ -5567,28 +5543,57 @@ void ixgbe_reset(struct ixgbe_adapter *adapter)
**/
static void ixgbe_clean_tx_ring(struct ixgbe_ring *tx_ring)
{
- struct ixgbe_tx_buffer *tx_buffer_info;
- unsigned long size;
- u16 i;
+ u16 i = tx_ring->next_to_clean;
+ struct ixgbe_tx_buffer *tx_buffer = &tx_ring->tx_buffer_info[i];
- /* ring already cleared, nothing to do */
- if (!tx_ring->tx_buffer_info)
- return;
+ while (i != tx_ring->next_to_use) {
+ union ixgbe_adv_tx_desc *eop_desc, *tx_desc;
- /* Free all the Tx ring sk_buffs */
- for (i = 0; i < tx_ring->count; i++) {
- tx_buffer_info = &tx_ring->tx_buffer_info[i];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer_info);
- }
+ /* Free all the Tx ring sk_buffs */
+ dev_kfree_skb_any(tx_buffer->skb);
- netdev_tx_reset_queue(txring_txq(tx_ring));
+ /* unmap skb header data */
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
- size = sizeof(struct ixgbe_tx_buffer) * tx_ring->count;
- memset(tx_ring->tx_buffer_info, 0, size);
+ /* check for eop_desc to determine the end of the packet */
+ eop_desc = tx_buffer->next_to_watch;
+ tx_desc = IXGBE_TX_DESC(tx_ring, i);
- /* Zero out the descriptor ring */
- memset(tx_ring->desc, 0, tx_ring->size);
+ /* unmap remaining buffers */
+ while (tx_desc != eop_desc) {
+ tx_buffer++;
+ tx_desc++;
+ i++;
+ if (unlikely(i == tx_ring->count)) {
+ i = 0;
+ tx_buffer = tx_ring->tx_buffer_info;
+ tx_desc = IXGBE_TX_DESC(tx_ring, 0);
+ }
+
+ /* unmap any remaining paged data */
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_page(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ }
+ /* move us one more past the eop_desc for start of next pkt */
+ tx_buffer++;
+ i++;
+ if (unlikely(i == tx_ring->count)) {
+ i = 0;
+ tx_buffer = tx_ring->tx_buffer_info;
+ }
+ }
+
+ /* reset BQL for queue */
+ netdev_tx_reset_queue(txring_txq(tx_ring));
+
+ /* reset next_to_use and next_to_clean */
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
}
@@ -6034,9 +6039,9 @@ int ixgbe_setup_tx_resources(struct ixgbe_ring *tx_ring)
if (tx_ring->q_vector)
ring_node = tx_ring->q_vector->numa_node;
- tx_ring->tx_buffer_info = vzalloc_node(size, ring_node);
+ tx_ring->tx_buffer_info = vmalloc_node(size, ring_node);
if (!tx_ring->tx_buffer_info)
- tx_ring->tx_buffer_info = vzalloc(size);
+ tx_ring->tx_buffer_info = vmalloc(size);
if (!tx_ring->tx_buffer_info)
goto err;
@@ -6116,9 +6121,9 @@ int ixgbe_setup_rx_resources(struct ixgbe_ring *rx_ring)
if (rx_ring->q_vector)
ring_node = rx_ring->q_vector->numa_node;
- rx_ring->rx_buffer_info = vzalloc_node(size, ring_node);
+ rx_ring->rx_buffer_info = vmalloc_node(size, ring_node);
if (!rx_ring->rx_buffer_info)
- rx_ring->rx_buffer_info = vzalloc(size);
+ rx_ring->rx_buffer_info = vmalloc(size);
if (!rx_ring->rx_buffer_info)
goto err;
@@ -7836,16 +7841,29 @@ static void ixgbe_tx_map(struct ixgbe_ring *tx_ring,
dev_err(tx_ring->dev, "TX DMA map failed\n");
/* clear dma mappings for failed tx_buffer_info map */
- for (;;) {
+ while (tx_buffer != first) {
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_page(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
+
+ if (i--)
+ i += tx_ring->count;
tx_buffer = &tx_ring->tx_buffer_info[i];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer);
- if (tx_buffer == first)
- break;
- if (i == 0)
- i = tx_ring->count;
- i--;
}
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
+
+ dev_kfree_skb_any(first->skb);
+ first->skb = NULL;
+
tx_ring->next_to_use = i;
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 11/11] ixgbe: Don't bother clearing buffer memory for descriptor rings
@ 2017-01-06 16:07 ` Alexander Duyck
0 siblings, 0 replies; 30+ messages in thread
From: Alexander Duyck @ 2017-01-06 16:07 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.h.duyck@intel.com>
This patch makes it so that we don't need to bother with clearing the
memory out for the descriptor rings. The general idea is to only free
buffers associated with buffers in use which are located between the
next_to_clean and next_to_use or next_to_alloc values. Everything outside
of those regions can be safely ignored since they should have no buffers
associated with them.
The advantage to doing things this way is that is should speed up bring-up
and tear-down of the rings. Specifically we can avoid the 512 or more
cycles required to memset the rings in tear-down. In the bring-up phase we
then clear the memory as a part of initialization. The general idea is
that the clearing in initialization can act as a prefetch of sorts for the
buffer info structures so they are in the local CPU when we go to populate
them. This should help to improve overall time needed to perform a
suspend/resume.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 11 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 158 ++++++++++++----------
2 files changed, 98 insertions(+), 71 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6466d828bc6f..8c6f40d4db6f 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -1945,7 +1945,16 @@ static u16 ixgbe_clean_test_rings(struct ixgbe_ring *rx_ring,
/* unmap buffer on Tx side */
tx_buffer = &tx_ring->tx_buffer_info[tx_ntc];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer);
+
+ /* Free all the Tx ring sk_buffs */
+ dev_kfree_skb_any(tx_buffer->skb);
+
+ /* unmap skb header data */
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
/* increment Rx/Tx next to clean counters */
rx_ntc++;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index dd04b76ae0bd..6ec0ebf7f174 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -958,28 +958,6 @@ static inline void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter,
}
}
-void ixgbe_unmap_and_free_tx_resource(struct ixgbe_ring *ring,
- struct ixgbe_tx_buffer *tx_buffer)
-{
- if (tx_buffer->skb) {
- dev_kfree_skb_any(tx_buffer->skb);
- if (dma_unmap_len(tx_buffer, len))
- dma_unmap_single(ring->dev,
- dma_unmap_addr(tx_buffer, dma),
- dma_unmap_len(tx_buffer, len),
- DMA_TO_DEVICE);
- } else if (dma_unmap_len(tx_buffer, len)) {
- dma_unmap_page(ring->dev,
- dma_unmap_addr(tx_buffer, dma),
- dma_unmap_len(tx_buffer, len),
- DMA_TO_DEVICE);
- }
- tx_buffer->next_to_watch = NULL;
- tx_buffer->skb = NULL;
- dma_unmap_len_set(tx_buffer, len, 0);
- /* tx_buffer must be completely set up in the transmit path */
-}
-
static void ixgbe_update_xoff_rx_lfc(struct ixgbe_adapter *adapter)
{
struct ixgbe_hw *hw = &adapter->hw;
@@ -1211,7 +1189,6 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
DMA_TO_DEVICE);
/* clear tx_buffer data */
- tx_buffer->skb = NULL;
dma_unmap_len_set(tx_buffer, len, 0);
/* unmap remaining buffers */
@@ -3345,6 +3322,10 @@ void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
clear_bit(__IXGBE_HANG_CHECK_ARMED, &ring->state);
+ /* reinitialize tx_buffer_info */
+ memset(ring->tx_buffer_info, 0,
+ sizeof(struct ixgbe_tx_buffer) * ring->count);
+
/* enable queue */
IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx), txdctl);
@@ -3865,6 +3846,10 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
#endif
}
+ /* initialize rx_buffer_info */
+ memset(ring->rx_buffer_info, 0,
+ sizeof(struct ixgbe_rx_buffer) * ring->count);
+
/* initialize Rx descriptor 0 */
rx_desc = IXGBE_RX_DESC(ring, 0);
rx_desc->wb.upper.length = 0;
@@ -5049,33 +5034,22 @@ static void ixgbe_fwd_psrtype(struct ixgbe_fwd_adapter *vadapter)
**/
static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
{
- struct device *dev = rx_ring->dev;
- unsigned long size;
- u16 i;
-
- /* ring already cleared, nothing to do */
- if (!rx_ring->rx_buffer_info)
- return;
+ u16 i = rx_ring->next_to_clean;
+ struct ixgbe_rx_buffer *rx_buffer = &rx_ring->rx_buffer_info[i];
/* Free all the Rx ring sk_buffs */
- for (i = 0; i < rx_ring->count; i++) {
- struct ixgbe_rx_buffer *rx_buffer = &rx_ring->rx_buffer_info[i];
-
+ while (i != rx_ring->next_to_alloc) {
if (rx_buffer->skb) {
struct sk_buff *skb = rx_buffer->skb;
if (IXGBE_CB(skb)->page_released)
- dma_unmap_page_attrs(dev,
+ dma_unmap_page_attrs(rx_ring->dev,
IXGBE_CB(skb)->dma,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
dev_kfree_skb(skb);
- rx_buffer->skb = NULL;
}
- if (!rx_buffer->page)
- continue;
-
/* Invalidate cache lines that may have been written to by
* device so that we avoid corrupting memory.
*/
@@ -5086,19 +5060,21 @@ static void ixgbe_clean_rx_ring(struct ixgbe_ring *rx_ring)
DMA_FROM_DEVICE);
/* free resources associated with mapping */
- dma_unmap_page_attrs(dev, rx_buffer->dma,
+ dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
ixgbe_rx_pg_size(rx_ring),
DMA_FROM_DEVICE,
IXGBE_RX_DMA_ATTR);
__page_frag_cache_drain(rx_buffer->page,
rx_buffer->pagecnt_bias);
- rx_buffer->page = NULL;
+ i++;
+ rx_buffer++;
+ if (i == rx_ring->count) {
+ i = 0;
+ rx_buffer = rx_ring->rx_buffer_info;
+ }
}
- size = sizeof(struct ixgbe_rx_buffer) * rx_ring->count;
- memset(rx_ring->rx_buffer_info, 0, size);
-
rx_ring->next_to_alloc = 0;
rx_ring->next_to_clean = 0;
rx_ring->next_to_use = 0;
@@ -5567,28 +5543,57 @@ void ixgbe_reset(struct ixgbe_adapter *adapter)
**/
static void ixgbe_clean_tx_ring(struct ixgbe_ring *tx_ring)
{
- struct ixgbe_tx_buffer *tx_buffer_info;
- unsigned long size;
- u16 i;
+ u16 i = tx_ring->next_to_clean;
+ struct ixgbe_tx_buffer *tx_buffer = &tx_ring->tx_buffer_info[i];
- /* ring already cleared, nothing to do */
- if (!tx_ring->tx_buffer_info)
- return;
+ while (i != tx_ring->next_to_use) {
+ union ixgbe_adv_tx_desc *eop_desc, *tx_desc;
- /* Free all the Tx ring sk_buffs */
- for (i = 0; i < tx_ring->count; i++) {
- tx_buffer_info = &tx_ring->tx_buffer_info[i];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer_info);
- }
+ /* Free all the Tx ring sk_buffs */
+ dev_kfree_skb_any(tx_buffer->skb);
- netdev_tx_reset_queue(txring_txq(tx_ring));
+ /* unmap skb header data */
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
- size = sizeof(struct ixgbe_tx_buffer) * tx_ring->count;
- memset(tx_ring->tx_buffer_info, 0, size);
+ /* check for eop_desc to determine the end of the packet */
+ eop_desc = tx_buffer->next_to_watch;
+ tx_desc = IXGBE_TX_DESC(tx_ring, i);
- /* Zero out the descriptor ring */
- memset(tx_ring->desc, 0, tx_ring->size);
+ /* unmap remaining buffers */
+ while (tx_desc != eop_desc) {
+ tx_buffer++;
+ tx_desc++;
+ i++;
+ if (unlikely(i == tx_ring->count)) {
+ i = 0;
+ tx_buffer = tx_ring->tx_buffer_info;
+ tx_desc = IXGBE_TX_DESC(tx_ring, 0);
+ }
+
+ /* unmap any remaining paged data */
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_page(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ }
+ /* move us one more past the eop_desc for start of next pkt */
+ tx_buffer++;
+ i++;
+ if (unlikely(i == tx_ring->count)) {
+ i = 0;
+ tx_buffer = tx_ring->tx_buffer_info;
+ }
+ }
+
+ /* reset BQL for queue */
+ netdev_tx_reset_queue(txring_txq(tx_ring));
+
+ /* reset next_to_use and next_to_clean */
tx_ring->next_to_use = 0;
tx_ring->next_to_clean = 0;
}
@@ -6034,9 +6039,9 @@ int ixgbe_setup_tx_resources(struct ixgbe_ring *tx_ring)
if (tx_ring->q_vector)
ring_node = tx_ring->q_vector->numa_node;
- tx_ring->tx_buffer_info = vzalloc_node(size, ring_node);
+ tx_ring->tx_buffer_info = vmalloc_node(size, ring_node);
if (!tx_ring->tx_buffer_info)
- tx_ring->tx_buffer_info = vzalloc(size);
+ tx_ring->tx_buffer_info = vmalloc(size);
if (!tx_ring->tx_buffer_info)
goto err;
@@ -6116,9 +6121,9 @@ int ixgbe_setup_rx_resources(struct ixgbe_ring *rx_ring)
if (rx_ring->q_vector)
ring_node = rx_ring->q_vector->numa_node;
- rx_ring->rx_buffer_info = vzalloc_node(size, ring_node);
+ rx_ring->rx_buffer_info = vmalloc_node(size, ring_node);
if (!rx_ring->rx_buffer_info)
- rx_ring->rx_buffer_info = vzalloc(size);
+ rx_ring->rx_buffer_info = vmalloc(size);
if (!rx_ring->rx_buffer_info)
goto err;
@@ -7836,16 +7841,29 @@ static void ixgbe_tx_map(struct ixgbe_ring *tx_ring,
dev_err(tx_ring->dev, "TX DMA map failed\n");
/* clear dma mappings for failed tx_buffer_info map */
- for (;;) {
+ while (tx_buffer != first) {
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_page(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
+
+ if (i--)
+ i += tx_ring->count;
tx_buffer = &tx_ring->tx_buffer_info[i];
- ixgbe_unmap_and_free_tx_resource(tx_ring, tx_buffer);
- if (tx_buffer == first)
- break;
- if (i == 0)
- i = tx_ring->count;
- i--;
}
+ if (dma_unmap_len(tx_buffer, len))
+ dma_unmap_single(tx_ring->dev,
+ dma_unmap_addr(tx_buffer, dma),
+ dma_unmap_len(tx_buffer, len),
+ DMA_TO_DEVICE);
+ dma_unmap_len_set(tx_buffer, len, 0);
+
+ dev_kfree_skb_any(first->skb);
+ first->skb = NULL;
+
tx_ring->next_to_use = i;
}
^ permalink raw reply related [flat|nested] 30+ messages in thread
* Re: [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-06 21:41 ` David Miller
-1 siblings, 0 replies; 30+ messages in thread
From: David Miller @ 2017-01-06 21:41 UTC (permalink / raw)
To: alexander.duyck; +Cc: intel-wired-lan, jeffrey.t.kirsher, netdev
From: Alexander Duyck <alexander.duyck@gmail.com>
Date: Fri, 06 Jan 2017 08:06:16 -0800
> The testing matrix for all of these patches is going to be pretty
> extensive. Basically we want to test these patches on as many platforms
> and architectures as possible with as many features being toggled as
> possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while receiving
> traffic.
Overall looks very nice to me.
I am assuming that I will get a formal submission once the necessary
amount of testing is performed.
^ permalink raw reply [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
@ 2017-01-06 21:41 ` David Miller
0 siblings, 0 replies; 30+ messages in thread
From: David Miller @ 2017-01-06 21:41 UTC (permalink / raw)
To: intel-wired-lan
From: Alexander Duyck <alexander.duyck@gmail.com>
Date: Fri, 06 Jan 2017 08:06:16 -0800
> The testing matrix for all of these patches is going to be pretty
> extensive. Basically we want to test these patches on as many platforms
> and architectures as possible with as many features being toggled as
> possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while receiving
> traffic.
Overall looks very nice to me.
I am assuming that I will get a formal submission once the necessary
amount of testing is performed.
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
2017-01-06 21:41 ` [Intel-wired-lan] " David Miller
@ 2017-01-07 1:12 ` Jeff Kirsher
-1 siblings, 0 replies; 30+ messages in thread
From: Jeff Kirsher @ 2017-01-07 1:12 UTC (permalink / raw)
To: David Miller, alexander.duyck; +Cc: intel-wired-lan, netdev
[-- Attachment #1: Type: text/plain, Size: 669 bytes --]
On Fri, 2017-01-06 at 16:41 -0500, David Miller wrote:
> From: Alexander Duyck <alexander.duyck@gmail.com>
> Date: Fri, 06 Jan 2017 08:06:16 -0800
>
> > The testing matrix for all of these patches is going to be pretty
> > extensive. Basically we want to test these patches on as many
> > platforms
> > and architectures as possible with as many features being toggled as
> > possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while
> > receiving
> > traffic.
>
> Overall looks very nice to me.
>
> I am assuming that I will get a formal submission once the necessary
> amount of testing is performed.
Yes, your assumption is correct. :-)
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb
@ 2017-01-07 1:12 ` Jeff Kirsher
0 siblings, 0 replies; 30+ messages in thread
From: Jeff Kirsher @ 2017-01-07 1:12 UTC (permalink / raw)
To: intel-wired-lan
On Fri, 2017-01-06 at 16:41 -0500, David Miller wrote:
> From: Alexander Duyck <alexander.duyck@gmail.com>
> Date: Fri, 06 Jan 2017 08:06:16 -0800
>
> > The testing matrix for all of these patches is going to be pretty
> > extensive.? Basically we want to test these patches on as many
> > platforms
> > and architectures as possible with as many features being toggled as
> > possible including RSC, FCoE, SR-IOV, and Jumbo Frames all while
> > receiving
> > traffic.
>
> Overall looks very nice to me.
>
> I am assuming that I will get a formal submission once the necessary
> amount of testing is performed.
Yes, your assumption is correct. :-)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <http://lists.osuosl.org/pipermail/intel-wired-lan/attachments/20170106/1c31d7a5/attachment.asc>
^ permalink raw reply [flat|nested] 30+ messages in thread
* Re: [next PATCH 04/11] ixgbe: Update code to better handle incrementing page count
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
@ 2017-01-07 13:06 ` kbuild test robot
-1 siblings, 0 replies; 30+ messages in thread
From: kbuild test robot @ 2017-01-07 13:06 UTC (permalink / raw)
To: Alexander Duyck; +Cc: kbuild-all, intel-wired-lan, jeffrey.t.kirsher, netdev
[-- Attachment #1: Type: text/plain, Size: 8788 bytes --]
Hi Alexander,
[auto build test ERROR on jkirsher-next-queue/dev-queue]
[also build test ERROR on v4.10-rc2 next-20170106]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Alexander-Duyck/ixgbe-Add-support-for-writable-pages-and-build_skb/20170107-193738
base: https://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue.git dev-queue
config: alpha-allyesconfig (attached as .config)
compiler: alpha-linux-gnu-gcc (Debian 6.1.1-9) 6.1.1 20160705
reproduce:
wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
make.cross ARCH=alpha
All errors (new ones prefixed by >>):
In file included from drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:60:0:
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c: In function 'ixgbe_can_reuse_rx_page':
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:1960:46: error: 'rx_ring' undeclared (first use in this function)
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
^
drivers/net/ethernet/intel/ixgbe/ixgbe.h:371:65: note: in definition of macro 'ixgbe_rx_pg_size'
#define ixgbe_rx_pg_size(_ring) (PAGE_SIZE << ixgbe_rx_pg_order(_ring))
^~~~~
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:1960:46: note: each undeclared identifier is reported only once for each function it appears in
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
^
drivers/net/ethernet/intel/ixgbe/ixgbe.h:371:65: note: in definition of macro 'ixgbe_rx_pg_size'
#define ixgbe_rx_pg_size(_ring) (PAGE_SIZE << ixgbe_rx_pg_order(_ring))
^~~~~
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c: In function 'ixgbe_fetch_rx_buffer':
>> drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:2124:3: error: implicit declaration of function '__page_frag_cache_drain' [-Werror=implicit-function-declaration]
__page_frag_cache_drain(page,
^~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/__page_frag_cache_drain +2124 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
1954
1955 static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
1956 struct page *page,
1957 const unsigned int truesize)
1958 {
1959 #if (PAGE_SIZE >= 8192)
> 1960 unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
1961 ixgbe_rx_bufsz(rx_ring);
1962 #endif
1963 unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
1964
1965 /* avoid re-using remote pages */
1966 if (unlikely(ixgbe_page_is_reserved(page)))
1967 return false;
1968
1969 #if (PAGE_SIZE < 8192)
1970 /* if we are only owner of page we can reuse it */
1971 if (unlikely(page_count(page) != pagecnt_bias))
1972 return false;
1973
1974 /* flip page offset to other buffer */
1975 rx_buffer->page_offset ^= truesize;
1976 #else
1977 /* move offset up to the next cache line */
1978 rx_buffer->page_offset += truesize;
1979
1980 if (rx_buffer->page_offset > last_offset)
1981 return false;
1982 #endif
1983
1984 /* If we have drained the page fragment pool we need to update
1985 * the pagecnt_bias and page count so that we fully restock the
1986 * number of references the driver holds.
1987 */
1988 if (unlikely(pagecnt_bias == 1)) {
1989 page_ref_add(page, USHRT_MAX);
1990 rx_buffer->pagecnt_bias = USHRT_MAX;
1991 }
1992
1993 return true;
1994 }
1995
1996 /**
1997 * ixgbe_add_rx_frag - Add contents of Rx buffer to sk_buff
1998 * @rx_ring: rx descriptor ring to transact packets on
1999 * @rx_buffer: buffer containing page to add
2000 * @rx_desc: descriptor containing length of buffer written by hardware
2001 * @skb: sk_buff to place the data into
2002 *
2003 * This function will add the data contained in rx_buffer->page to the skb.
2004 * This is done either through a direct copy if the data in the buffer is
2005 * less than the skb header size, otherwise it will just attach the page as
2006 * a frag to the skb.
2007 *
2008 * The function will then update the page offset if necessary and return
2009 * true if the buffer can be reused by the adapter.
2010 **/
2011 static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
2012 struct ixgbe_rx_buffer *rx_buffer,
2013 unsigned int size,
2014 struct sk_buff *skb)
2015 {
2016 struct page *page = rx_buffer->page;
2017 unsigned char *va = page_address(page) + rx_buffer->page_offset;
2018 #if (PAGE_SIZE < 8192)
2019 unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
2020 #else
2021 unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
2022 #endif
2023
2024 if (unlikely(skb_is_nonlinear(skb)))
2025 goto add_tail_frag;
2026
2027 if (size <= IXGBE_RX_HDR_SIZE) {
2028 memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
2029
2030 /* page is not reserved, we can reuse buffer as-is */
2031 if (likely(!ixgbe_page_is_reserved(page)))
2032 return true;
2033
2034 /* this page cannot be reused so discard it */
2035 return false;
2036 }
2037
2038 add_tail_frag:
2039 skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
2040 rx_buffer->page_offset, size, truesize);
2041
2042 return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
2043 }
2044
2045 static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
2046 union ixgbe_adv_rx_desc *rx_desc)
2047 {
2048 unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
2049 struct ixgbe_rx_buffer *rx_buffer;
2050 struct sk_buff *skb;
2051 struct page *page;
2052
2053 rx_buffer = &rx_ring->rx_buffer_info[rx_ring->next_to_clean];
2054 page = rx_buffer->page;
2055 prefetchw(page);
2056
2057 skb = rx_buffer->skb;
2058
2059 if (likely(!skb)) {
2060 void *page_addr = page_address(page) +
2061 rx_buffer->page_offset;
2062
2063 /* prefetch first cache line of first page */
2064 prefetch(page_addr);
2065 #if L1_CACHE_BYTES < 128
2066 prefetch(page_addr + L1_CACHE_BYTES);
2067 #endif
2068
2069 /* allocate a skb to store the frags */
2070 skb = napi_alloc_skb(&rx_ring->q_vector->napi,
2071 IXGBE_RX_HDR_SIZE);
2072 if (unlikely(!skb)) {
2073 rx_ring->rx_stats.alloc_rx_buff_failed++;
2074 return NULL;
2075 }
2076
2077 /*
2078 * we will be copying header into skb->data in
2079 * pskb_may_pull so it is in our interest to prefetch
2080 * it now to avoid a possible cache miss
2081 */
2082 prefetchw(skb->data);
2083
2084 /*
2085 * Delay unmapping of the first packet. It carries the
2086 * header information, HW may still access the header
2087 * after the writeback. Only unmap it when EOP is
2088 * reached
2089 */
2090 if (likely(ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)))
2091 goto dma_sync;
2092
2093 IXGBE_CB(skb)->dma = rx_buffer->dma;
2094 } else {
2095 if (ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
2096 ixgbe_dma_sync_frag(rx_ring, skb);
2097
2098 dma_sync:
2099 /* we are reusing so sync this buffer for CPU use */
2100 dma_sync_single_range_for_cpu(rx_ring->dev,
2101 rx_buffer->dma,
2102 rx_buffer->page_offset,
2103 size,
2104 DMA_FROM_DEVICE);
2105
2106 rx_buffer->skb = NULL;
2107 }
2108
2109 /* pull page into skb */
2110 if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
2111 /* hand second half of page back to the ring */
2112 ixgbe_reuse_rx_page(rx_ring, rx_buffer);
2113 } else {
2114 if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
2115 /* the page has been released from the ring */
2116 IXGBE_CB(skb)->page_released = true;
2117 } else {
2118 /* we are not reusing the buffer so unmap it */
2119 dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
2120 ixgbe_rx_pg_size(rx_ring),
2121 DMA_FROM_DEVICE,
2122 IXGBE_RX_DMA_ATTR);
2123 }
> 2124 __page_frag_cache_drain(page,
2125 rx_buffer->pagecnt_bias);
2126 }
2127
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 48633 bytes --]
^ permalink raw reply [flat|nested] 30+ messages in thread
* [Intel-wired-lan] [next PATCH 04/11] ixgbe: Update code to better handle incrementing page count
@ 2017-01-07 13:06 ` kbuild test robot
0 siblings, 0 replies; 30+ messages in thread
From: kbuild test robot @ 2017-01-07 13:06 UTC (permalink / raw)
To: intel-wired-lan
Hi Alexander,
[auto build test ERROR on jkirsher-next-queue/dev-queue]
[also build test ERROR on v4.10-rc2 next-20170106]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Alexander-Duyck/ixgbe-Add-support-for-writable-pages-and-build_skb/20170107-193738
base: https://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue.git dev-queue
config: alpha-allyesconfig (attached as .config)
compiler: alpha-linux-gnu-gcc (Debian 6.1.1-9) 6.1.1 20160705
reproduce:
wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
make.cross ARCH=alpha
All errors (new ones prefixed by >>):
In file included from drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:60:0:
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c: In function 'ixgbe_can_reuse_rx_page':
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:1960:46: error: 'rx_ring' undeclared (first use in this function)
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
^
drivers/net/ethernet/intel/ixgbe/ixgbe.h:371:65: note: in definition of macro 'ixgbe_rx_pg_size'
#define ixgbe_rx_pg_size(_ring) (PAGE_SIZE << ixgbe_rx_pg_order(_ring))
^~~~~
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:1960:46: note: each undeclared identifier is reported only once for each function it appears in
unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
^
drivers/net/ethernet/intel/ixgbe/ixgbe.h:371:65: note: in definition of macro 'ixgbe_rx_pg_size'
#define ixgbe_rx_pg_size(_ring) (PAGE_SIZE << ixgbe_rx_pg_order(_ring))
^~~~~
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c: In function 'ixgbe_fetch_rx_buffer':
>> drivers/net/ethernet/intel/ixgbe/ixgbe_main.c:2124:3: error: implicit declaration of function '__page_frag_cache_drain' [-Werror=implicit-function-declaration]
__page_frag_cache_drain(page,
^~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/__page_frag_cache_drain +2124 drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
1954
1955 static bool ixgbe_can_reuse_rx_page(struct ixgbe_rx_buffer *rx_buffer,
1956 struct page *page,
1957 const unsigned int truesize)
1958 {
1959 #if (PAGE_SIZE >= 8192)
> 1960 unsigned int last_offset = ixgbe_rx_pg_size(rx_ring) -
1961 ixgbe_rx_bufsz(rx_ring);
1962 #endif
1963 unsigned int pagecnt_bias = rx_buffer->pagecnt_bias--;
1964
1965 /* avoid re-using remote pages */
1966 if (unlikely(ixgbe_page_is_reserved(page)))
1967 return false;
1968
1969 #if (PAGE_SIZE < 8192)
1970 /* if we are only owner of page we can reuse it */
1971 if (unlikely(page_count(page) != pagecnt_bias))
1972 return false;
1973
1974 /* flip page offset to other buffer */
1975 rx_buffer->page_offset ^= truesize;
1976 #else
1977 /* move offset up to the next cache line */
1978 rx_buffer->page_offset += truesize;
1979
1980 if (rx_buffer->page_offset > last_offset)
1981 return false;
1982 #endif
1983
1984 /* If we have drained the page fragment pool we need to update
1985 * the pagecnt_bias and page count so that we fully restock the
1986 * number of references the driver holds.
1987 */
1988 if (unlikely(pagecnt_bias == 1)) {
1989 page_ref_add(page, USHRT_MAX);
1990 rx_buffer->pagecnt_bias = USHRT_MAX;
1991 }
1992
1993 return true;
1994 }
1995
1996 /**
1997 * ixgbe_add_rx_frag - Add contents of Rx buffer to sk_buff
1998 * @rx_ring: rx descriptor ring to transact packets on
1999 * @rx_buffer: buffer containing page to add
2000 * @rx_desc: descriptor containing length of buffer written by hardware
2001 * @skb: sk_buff to place the data into
2002 *
2003 * This function will add the data contained in rx_buffer->page to the skb.
2004 * This is done either through a direct copy if the data in the buffer is
2005 * less than the skb header size, otherwise it will just attach the page as
2006 * a frag to the skb.
2007 *
2008 * The function will then update the page offset if necessary and return
2009 * true if the buffer can be reused by the adapter.
2010 **/
2011 static bool ixgbe_add_rx_frag(struct ixgbe_ring *rx_ring,
2012 struct ixgbe_rx_buffer *rx_buffer,
2013 unsigned int size,
2014 struct sk_buff *skb)
2015 {
2016 struct page *page = rx_buffer->page;
2017 unsigned char *va = page_address(page) + rx_buffer->page_offset;
2018 #if (PAGE_SIZE < 8192)
2019 unsigned int truesize = ixgbe_rx_bufsz(rx_ring);
2020 #else
2021 unsigned int truesize = ALIGN(size, L1_CACHE_BYTES);
2022 #endif
2023
2024 if (unlikely(skb_is_nonlinear(skb)))
2025 goto add_tail_frag;
2026
2027 if (size <= IXGBE_RX_HDR_SIZE) {
2028 memcpy(__skb_put(skb, size), va, ALIGN(size, sizeof(long)));
2029
2030 /* page is not reserved, we can reuse buffer as-is */
2031 if (likely(!ixgbe_page_is_reserved(page)))
2032 return true;
2033
2034 /* this page cannot be reused so discard it */
2035 return false;
2036 }
2037
2038 add_tail_frag:
2039 skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, page,
2040 rx_buffer->page_offset, size, truesize);
2041
2042 return ixgbe_can_reuse_rx_page(rx_buffer, page, truesize);
2043 }
2044
2045 static struct sk_buff *ixgbe_fetch_rx_buffer(struct ixgbe_ring *rx_ring,
2046 union ixgbe_adv_rx_desc *rx_desc)
2047 {
2048 unsigned int size = le16_to_cpu(rx_desc->wb.upper.length);
2049 struct ixgbe_rx_buffer *rx_buffer;
2050 struct sk_buff *skb;
2051 struct page *page;
2052
2053 rx_buffer = &rx_ring->rx_buffer_info[rx_ring->next_to_clean];
2054 page = rx_buffer->page;
2055 prefetchw(page);
2056
2057 skb = rx_buffer->skb;
2058
2059 if (likely(!skb)) {
2060 void *page_addr = page_address(page) +
2061 rx_buffer->page_offset;
2062
2063 /* prefetch first cache line of first page */
2064 prefetch(page_addr);
2065 #if L1_CACHE_BYTES < 128
2066 prefetch(page_addr + L1_CACHE_BYTES);
2067 #endif
2068
2069 /* allocate a skb to store the frags */
2070 skb = napi_alloc_skb(&rx_ring->q_vector->napi,
2071 IXGBE_RX_HDR_SIZE);
2072 if (unlikely(!skb)) {
2073 rx_ring->rx_stats.alloc_rx_buff_failed++;
2074 return NULL;
2075 }
2076
2077 /*
2078 * we will be copying header into skb->data in
2079 * pskb_may_pull so it is in our interest to prefetch
2080 * it now to avoid a possible cache miss
2081 */
2082 prefetchw(skb->data);
2083
2084 /*
2085 * Delay unmapping of the first packet. It carries the
2086 * header information, HW may still access the header
2087 * after the writeback. Only unmap it when EOP is
2088 * reached
2089 */
2090 if (likely(ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP)))
2091 goto dma_sync;
2092
2093 IXGBE_CB(skb)->dma = rx_buffer->dma;
2094 } else {
2095 if (ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))
2096 ixgbe_dma_sync_frag(rx_ring, skb);
2097
2098 dma_sync:
2099 /* we are reusing so sync this buffer for CPU use */
2100 dma_sync_single_range_for_cpu(rx_ring->dev,
2101 rx_buffer->dma,
2102 rx_buffer->page_offset,
2103 size,
2104 DMA_FROM_DEVICE);
2105
2106 rx_buffer->skb = NULL;
2107 }
2108
2109 /* pull page into skb */
2110 if (ixgbe_add_rx_frag(rx_ring, rx_buffer, size, skb)) {
2111 /* hand second half of page back to the ring */
2112 ixgbe_reuse_rx_page(rx_ring, rx_buffer);
2113 } else {
2114 if (IXGBE_CB(skb)->dma == rx_buffer->dma) {
2115 /* the page has been released from the ring */
2116 IXGBE_CB(skb)->page_released = true;
2117 } else {
2118 /* we are not reusing the buffer so unmap it */
2119 dma_unmap_page_attrs(rx_ring->dev, rx_buffer->dma,
2120 ixgbe_rx_pg_size(rx_ring),
2121 DMA_FROM_DEVICE,
2122 IXGBE_RX_DMA_ATTR);
2123 }
> 2124 __page_frag_cache_drain(page,
2125 rx_buffer->pagecnt_bias);
2126 }
2127
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
-------------- next part --------------
A non-text attachment was scrubbed...
Name: .config.gz
Type: application/gzip
Size: 48633 bytes
Desc: not available
URL: <http://lists.osuosl.org/pipermail/intel-wired-lan/attachments/20170107/ab8fc1ad/attachment-0001.bin>
^ permalink raw reply [flat|nested] 30+ messages in thread
end of thread, other threads:[~2017-01-07 13:06 UTC | newest]
Thread overview: 30+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-01-06 16:06 [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb Alexander Duyck
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:06 ` [next PATCH 01/11] ixgbe: Add function for checking to see if we can reuse page Alexander Duyck
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:06 ` [next PATCH 02/11] ixgbe: Only DMA sync frame length Alexander Duyck
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:06 ` [next PATCH 03/11] ixgbe: Update driver to make use of DMA attributes in Rx path Alexander Duyck
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:06 ` [next PATCH 04/11] ixgbe: Update code to better handle incrementing page count Alexander Duyck
2017-01-06 16:06 ` [Intel-wired-lan] " Alexander Duyck
2017-01-07 13:06 ` kbuild test robot
2017-01-07 13:06 ` [Intel-wired-lan] " kbuild test robot
2017-01-06 16:07 ` [next PATCH 05/11] ixgbe: Make use of order 1 pages and 3K buffers independent of FCoE Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 06/11] ixgbe: Use length to determine if descriptor is done Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 07/11] ixgbe: Break out Rx buffer page management Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 08/11] ixgbe: Add support for padding packet Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 09/11] ixgbe: Add private flag to control buffer mode Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 10/11] ixgbe: Add support for build_skb Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 16:07 ` [next PATCH 11/11] ixgbe: Don't bother clearing buffer memory for descriptor rings Alexander Duyck
2017-01-06 16:07 ` [Intel-wired-lan] " Alexander Duyck
2017-01-06 21:41 ` [next PATCH 00/11] ixgbe: Add support for writable pages and build_skb David Miller
2017-01-06 21:41 ` [Intel-wired-lan] " David Miller
2017-01-07 1:12 ` Jeff Kirsher
2017-01-07 1:12 ` [Intel-wired-lan] " Jeff Kirsher
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.