* [PATCH net-next 0/6] sfc: SFN8000 support improvements
[not found] <1137d3d8-0b1b-d5f5-149a-1771bbac7aa6@solarflare.com>
@ 2016-08-11 11:59 ` Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 1/6] sfc: update MCDI protocol headers Bert Kenward
` (6 more replies)
0 siblings, 7 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 11:59 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
This series improves support for the recently released SFN8000 series
of adapters. Specifically, it retrieves interrupt moderation timer
settings directly from the adapter and uses those settings. It also
uses a new event queue initialisation interface, allowing specification
of a performance objective rather than enabling individual flags.
Bert Kenward (6):
sfc: update MCDI protocol headers
sfc: allow asynchronous MCDI without completion function
sfc: retrieve second word of datapath capabilities
sfc: use new performance based event queue init
sfc: set interrupt moderation via MCDI
sfc: get timer configuration from adapter
drivers/net/ethernet/sfc/ef10.c | 221 +++++++++++---
drivers/net/ethernet/sfc/efx.c | 96 +++---
drivers/net/ethernet/sfc/efx.h | 2 +
drivers/net/ethernet/sfc/falcon.c | 9 +-
drivers/net/ethernet/sfc/mcdi.c | 5 +-
drivers/net/ethernet/sfc/mcdi_pcol.h | 530 ++++++++++++++++++++++++++++++++-
drivers/net/ethernet/sfc/net_driver.h | 12 +-
drivers/net/ethernet/sfc/nic.h | 5 +
drivers/net/ethernet/sfc/ptp.c | 2 +-
drivers/net/ethernet/sfc/siena.c | 14 +-
drivers/net/ethernet/sfc/workarounds.h | 4 +
11 files changed, 782 insertions(+), 118 deletions(-)
--
2.7.4
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH net-next 1/6] sfc: update MCDI protocol headers
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
@ 2016-08-11 12:01 ` Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 2/6] sfc: allow asynchronous MCDI without completion function Bert Kenward
` (5 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:01 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/mcdi_pcol.h | 530 ++++++++++++++++++++++++++++++++++-
1 file changed, 515 insertions(+), 15 deletions(-)
diff --git a/drivers/net/ethernet/sfc/mcdi_pcol.h b/drivers/net/ethernet/sfc/mcdi_pcol.h
index c9a5b00..ccceafc 100644
--- a/drivers/net/ethernet/sfc/mcdi_pcol.h
+++ b/drivers/net/ethernet/sfc/mcdi_pcol.h
@@ -2645,16 +2645,20 @@
#define MC_CMD_POLL_BIST_MEM_BUS_MC 0x0
/* enum: CSR IREG bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_CSR 0x1
-/* enum: RX DPCPU bus. */
+/* enum: RX0 DPCPU bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_RX 0x2
/* enum: TX0 DPCPU bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_TX0 0x3
/* enum: TX1 DPCPU bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_TX1 0x4
-/* enum: RX DICPU bus. */
+/* enum: RX0 DICPU bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_DICPU_RX 0x5
/* enum: TX DICPU bus. */
#define MC_CMD_POLL_BIST_MEM_BUS_DICPU_TX 0x6
+/* enum: RX1 DPCPU bus. */
+#define MC_CMD_POLL_BIST_MEM_BUS_DPCPU_RX1 0x7
+/* enum: RX1 DICPU bus. */
+#define MC_CMD_POLL_BIST_MEM_BUS_DICPU_RX1 0x8
/* Pattern written to RAM / register */
#define MC_CMD_POLL_BIST_OUT_MEM_EXPECT_OFST 16
/* Actual value read from RAM / register */
@@ -3612,6 +3616,8 @@
#define MC_CMD_NVRAM_INFO_OUT_PROTECTED_WIDTH 1
#define MC_CMD_NVRAM_INFO_OUT_TLV_LBN 1
#define MC_CMD_NVRAM_INFO_OUT_TLV_WIDTH 1
+#define MC_CMD_NVRAM_INFO_OUT_CMAC_LBN 6
+#define MC_CMD_NVRAM_INFO_OUT_CMAC_WIDTH 1
#define MC_CMD_NVRAM_INFO_OUT_A_B_LBN 7
#define MC_CMD_NVRAM_INFO_OUT_A_B_WIDTH 1
#define MC_CMD_NVRAM_INFO_OUT_PHYSDEV_OFST 16
@@ -4389,6 +4395,8 @@
* the command will fail with MC_CMD_ERR_FILTERS_PRESENT.
*/
#define MC_CMD_WORKAROUND_BUG26807 0x6
+/* enum: Bug 61265 work around (broken EVQ TMR writes). */
+#define MC_CMD_WORKAROUND_BUG61265 0x7
/* 0 = disable the workaround indicated by TYPE; any non-zero value = enable
* the workaround
*/
@@ -4413,7 +4421,6 @@
* (GET_PHY_CFG_OUT_MEDIA_TYPE); the valid 'page number' input values, and the
* output data, are interpreted on a per-type basis. For SFP+: PAGE=0 or 1
* returns a 128-byte block read from module I2C address 0xA0 offset 0 or 0x80.
- * Anything else: currently undefined. Locks required: None. Return code: 0.
*/
#define MC_CMD_GET_PHY_MEDIA_INFO 0x4b
@@ -5479,6 +5486,8 @@
#define LICENSED_V3_FEATURES_TX_SNIFF_WIDTH 1
#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_LBN 8
#define LICENSED_V3_FEATURES_PROXY_FILTER_OPS_WIDTH 1
+#define LICENSED_V3_FEATURES_EVENT_CUT_THROUGH_LBN 9
+#define LICENSED_V3_FEATURES_EVENT_CUT_THROUGH_WIDTH 1
#define LICENSED_V3_FEATURES_MASK_LBN 0
#define LICENSED_V3_FEATURES_MASK_WIDTH 64
@@ -5634,6 +5643,109 @@
/* Only valid if INTRFLAG was true */
#define MC_CMD_INIT_EVQ_OUT_IRQ_OFST 0
+/* MC_CMD_INIT_EVQ_V2_IN msgrequest */
+#define MC_CMD_INIT_EVQ_V2_IN_LENMIN 44
+#define MC_CMD_INIT_EVQ_V2_IN_LENMAX 548
+#define MC_CMD_INIT_EVQ_V2_IN_LEN(num) (36+8*(num))
+/* Size, in entries */
+#define MC_CMD_INIT_EVQ_V2_IN_SIZE_OFST 0
+/* Desired instance. Must be set to a specific instance, which is a function
+ * local queue index.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_INSTANCE_OFST 4
+/* The initial timer value. The load value is ignored if the timer mode is DIS.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_LOAD_OFST 8
+/* The reload value is ignored in one-shot modes */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_RELOAD_OFST 12
+/* tbd */
+#define MC_CMD_INIT_EVQ_V2_IN_FLAGS_OFST 16
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INTERRUPTING_LBN 0
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INTERRUPTING_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RPTR_DOS_LBN 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RPTR_DOS_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INT_ARMD_LBN 2
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_INT_ARMD_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_CUT_THRU_LBN 3
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_CUT_THRU_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RX_MERGE_LBN 4
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_RX_MERGE_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TX_MERGE_LBN 5
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TX_MERGE_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_USE_TIMER_LBN 6
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_USE_TIMER_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_LBN 7
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_WIDTH 4
+/* enum: All initialisation flags specified by host. */
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_MANUAL 0x0
+/* enum: MEDFORD only. Certain initialisation flags specified by host may be
+ * over-ridden by firmware based on licenses and firmware variant in order to
+ * provide the lowest latency achievable. See
+ * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_LOW_LATENCY 0x1
+/* enum: MEDFORD only. Certain initialisation flags specified by host may be
+ * over-ridden by firmware based on licenses and firmware variant in order to
+ * provide the best throughput achievable. See
+ * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_THROUGHPUT 0x2
+/* enum: MEDFORD only. Certain initialisation flags may be over-ridden by
+ * firmware based on licenses and firmware variant. See
+ * MC_CMD_INIT_EVQ_V2/MC_CMD_INIT_EVQ_V2_OUT/FLAGS for list of affected flags.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO 0x3
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_MODE_OFST 20
+/* enum: Disabled */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_MODE_DIS 0x0
+/* enum: Immediate */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_IMMED_START 0x1
+/* enum: Triggered */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_TRIG_START 0x2
+/* enum: Hold-off */
+#define MC_CMD_INIT_EVQ_V2_IN_TMR_INT_HLDOFF 0x3
+/* Target EVQ for wakeups if in wakeup mode. */
+#define MC_CMD_INIT_EVQ_V2_IN_TARGET_EVQ_OFST 24
+/* Target interrupt if in interrupting mode (note union with target EVQ). Use
+ * MC_CMD_RESOURCE_INSTANCE_ANY unless a specific one required for test
+ * purposes.
+ */
+#define MC_CMD_INIT_EVQ_V2_IN_IRQ_NUM_OFST 24
+/* Event Counter Mode. */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_OFST 28
+/* enum: Disabled */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_DIS 0x0
+/* enum: Disabled */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_RX 0x1
+/* enum: Disabled */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_TX 0x2
+/* enum: Disabled */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_MODE_RXTX 0x3
+/* Event queue packet count threshold. */
+#define MC_CMD_INIT_EVQ_V2_IN_COUNT_THRSHLD_OFST 32
+/* 64-bit address of 4k of 4k-aligned host memory buffer */
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_OFST 36
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_LEN 8
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_LO_OFST 36
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_HI_OFST 40
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_MINNUM 1
+#define MC_CMD_INIT_EVQ_V2_IN_DMA_ADDR_MAXNUM 64
+
+/* MC_CMD_INIT_EVQ_V2_OUT msgresponse */
+#define MC_CMD_INIT_EVQ_V2_OUT_LEN 8
+/* Only valid if INTRFLAG was true */
+#define MC_CMD_INIT_EVQ_V2_OUT_IRQ_OFST 0
+/* Actual configuration applied on the card */
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAGS_OFST 4
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_CUT_THRU_LBN 0
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_CUT_THRU_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RX_MERGE_LBN 1
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RX_MERGE_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_TX_MERGE_LBN 2
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_TX_MERGE_WIDTH 1
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RXQ_FORCE_EV_MERGING_LBN 3
+#define MC_CMD_INIT_EVQ_V2_OUT_FLAG_RXQ_FORCE_EV_MERGING_WIDTH 1
+
/* QUEUE_CRC_MODE structuredef */
#define QUEUE_CRC_MODE_LEN 1
#define QUEUE_CRC_MODE_MODE_LBN 0
@@ -5697,8 +5809,8 @@
#define MC_CMD_INIT_RXQ_IN_FLAG_PREFIX_WIDTH 1
#define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_LBN 9
#define MC_CMD_INIT_RXQ_IN_FLAG_DISABLE_SCATTER_WIDTH 1
-#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_LBN 10
-#define MC_CMD_INIT_RXQ_IN_FLAG_FORCE_EV_MERGING_WIDTH 1
+#define MC_CMD_INIT_RXQ_IN_UNUSED_LBN 10
+#define MC_CMD_INIT_RXQ_IN_UNUSED_WIDTH 1
/* Owner ID to use if in buffer mode (zero if physical) */
#define MC_CMD_INIT_RXQ_IN_OWNER_ID_OFST 20
/* The port ID associated with the v-adaptor which should contain this DMAQ. */
@@ -7854,6 +7966,20 @@
#define MC_CMD_GET_CAPABILITIES_V2_OUT_EVENT_CUT_THROUGH_WIDTH 1
#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_LBN 4
#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_CUT_THROUGH_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VFIFO_ULL_MODE_LBN 5
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_VFIFO_ULL_MODE_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_MAC_STATS_40G_TX_SIZE_BINS_LBN 6
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_MAC_STATS_40G_TX_SIZE_BINS_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN 7
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_TIMESTAMPING_LBN 8
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_TIMESTAMPING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TIMESTAMP_LBN 9
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_TIMESTAMP_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_SNIFF_LBN 10
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_RX_SNIFF_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_SNIFF_LBN 11
+#define MC_CMD_GET_CAPABILITIES_V2_OUT_TX_SNIFF_WIDTH 1
/* Number of FATSOv2 contexts per datapath supported by this NIC. Not present
* on older firmware (check the length).
*/
@@ -7910,6 +8036,288 @@
#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_OFST 70
#define MC_CMD_GET_CAPABILITIES_V2_OUT_SIZE_PIO_BUFF_LEN 2
+/* MC_CMD_GET_CAPABILITIES_V3_OUT msgresponse */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_LEN 73
+/* First word of flags. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_FLAGS1_OFST 0
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VPORT_RECONFIGURE_LBN 3
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VPORT_RECONFIGURE_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_STRIPING_LBN 4
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_STRIPING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_QUERY_LBN 5
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_QUERY_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_PORT_VLAN_RESTRICT_LBN 6
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_PORT_VLAN_RESTRICT_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_DRV_ATTACH_PREBOOT_LBN 7
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_DRV_ATTACH_PREBOOT_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_FORCE_EVENT_MERGING_LBN 8
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_FORCE_EVENT_MERGING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_SET_MAC_ENHANCED_LBN 9
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_SET_MAC_ENHANCED_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_LBN 10
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_UNKNOWN_UCAST_DST_FILTER_ALWAYS_MULTI_RECIPIENT_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_LBN 11
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VADAPTOR_PERMIT_SET_MAC_WHEN_FILTERS_INSTALLED_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_SECURITY_FILTERING_LBN 12
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_SECURITY_FILTERING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_ADDITIONAL_RSS_MODES_LBN 13
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_ADDITIONAL_RSS_MODES_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_QBB_LBN 14
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_QBB_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_VAR_BUFFERS_LBN 15
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_VAR_BUFFERS_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_RSS_LIMITED_LBN 16
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_RSS_LIMITED_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_LBN 17
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PACKED_STREAM_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_INCLUDE_FCS_LBN 18
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_INCLUDE_FCS_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VLAN_INSERTION_LBN 19
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VLAN_INSERTION_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_VLAN_STRIPPING_LBN 20
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_VLAN_STRIPPING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_LBN 21
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_0_LBN 22
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_0_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_14_LBN 23
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_PREFIX_LEN_14_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_TIMESTAMP_LBN 24
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_TIMESTAMP_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_BATCHING_LBN 25
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_BATCHING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_MCAST_FILTER_CHAINING_LBN 26
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_MCAST_FILTER_CHAINING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PM_AND_RXDP_COUNTERS_LBN 27
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PM_AND_RXDP_COUNTERS_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DISABLE_SCATTER_LBN 28
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DISABLE_SCATTER_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MCAST_UDP_LOOPBACK_LBN 29
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MCAST_UDP_LOOPBACK_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_LBN 30
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVB_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VXLAN_NVGRE_LBN 31
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VXLAN_NVGRE_WIDTH 1
+/* RxDPCPU firmware id. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DPCPU_FW_ID_OFST 4
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DPCPU_FW_ID_LEN 2
+/* enum: Standard RXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP 0x0
+/* enum: Low latency RXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_LOW_LATENCY 0x1
+/* enum: Packed stream RXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_PACKED_STREAM 0x2
+/* enum: BIST RXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_BIST 0x10a
+/* enum: RXDP Test firmware image 1 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_CUT_THROUGH 0x101
+/* enum: RXDP Test firmware image 2 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD 0x102
+/* enum: RXDP Test firmware image 3 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_TO_MC_STORE_FORWARD_FIRST 0x103
+/* enum: RXDP Test firmware image 4 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_EVERY_EVENT_BATCHABLE 0x104
+/* enum: RXDP Test firmware image 5 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_BACKPRESSURE 0x105
+/* enum: RXDP Test firmware image 6 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_PACKET_EDITS 0x106
+/* enum: RXDP Test firmware image 7 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_RX_HDR_SPLIT 0x107
+/* enum: RXDP Test firmware image 8 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_DISABLE_DL 0x108
+/* enum: RXDP Test firmware image 9 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXDP_TEST_FW_DOORBELL_DELAY 0x10b
+/* TxDPCPU firmware id. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DPCPU_FW_ID_OFST 6
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DPCPU_FW_ID_LEN 2
+/* enum: Standard TXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP 0x0
+/* enum: Low latency TXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_LOW_LATENCY 0x1
+/* enum: High packet rate TXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_HIGH_PACKET_RATE 0x3
+/* enum: BIST TXDP firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_BIST 0x12d
+/* enum: TXDP Test firmware image 1 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_TSO_EDIT 0x101
+/* enum: TXDP Test firmware image 2 */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_PACKET_EDITS 0x102
+/* enum: TXDP CSR bus test firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXDP_TEST_FW_CSR 0x103
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_OFST 8
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_LEN 2
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_REV_LBN 0
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_REV_WIDTH 12
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_TYPE_LBN 12
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_VERSION_TYPE_WIDTH 4
+/* enum: reserved value - do not use (may indicate alternative interpretation
+ * of REV field in future)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_RESERVED 0x0
+/* enum: Trivial RX PD firmware for early Huntington development (Huntington
+ * development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_FIRST_PKT 0x1
+/* enum: RX PD firmware with approximately Siena-compatible behaviour
+ * (Huntington development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_SIENA_COMPAT 0x2
+/* enum: Virtual switching (full feature) RX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_VSWITCH 0x3
+/* enum: siena_compat variant RX PD firmware using PM rather than MAC
+ * (Huntington development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_SIENA_COMPAT_PM 0x4
+/* enum: Low latency RX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_LOW_LATENCY 0x5
+/* enum: Packed stream RX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_PACKED_STREAM 0x6
+/* enum: RX PD firmware handling layer 2 only for high packet rate performance
+ * tests (Medford development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_LAYER2_PERF 0x7
+/* enum: Rules engine RX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_RULES_ENGINE 0x8
+/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe
+/* enum: RX PD firmware parsing but not filtering network overlay tunnel
+ * encapsulations (Medford development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RXPD_FW_TYPE_TESTFW_ENCAP_PARSING_ONLY 0xf
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_OFST 10
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_LEN 2
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_REV_LBN 0
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_REV_WIDTH 12
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_TYPE_LBN 12
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_VERSION_TYPE_WIDTH 4
+/* enum: reserved value - do not use (may indicate alternative interpretation
+ * of REV field in future)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_RESERVED 0x0
+/* enum: Trivial TX PD firmware for early Huntington development (Huntington
+ * development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_FIRST_PKT 0x1
+/* enum: TX PD firmware with approximately Siena-compatible behaviour
+ * (Huntington development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_SIENA_COMPAT 0x2
+/* enum: Virtual switching (full feature) TX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_VSWITCH 0x3
+/* enum: siena_compat variant TX PD firmware using PM rather than MAC
+ * (Huntington development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_SIENA_COMPAT_PM 0x4
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_LOW_LATENCY 0x5 /* enum */
+/* enum: TX PD firmware handling layer 2 only for high packet rate performance
+ * tests (Medford development only)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_LAYER2_PERF 0x7
+/* enum: Rules engine TX PD production firmware */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_RULES_ENGINE 0x8
+/* enum: RX PD firmware for GUE parsing prototype (Medford development only) */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TXPD_FW_TYPE_TESTFW_GUE_PROTOTYPE 0xe
+/* Hardware capabilities of NIC */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_HW_CAPABILITIES_OFST 12
+/* Licensed capabilities */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_LICENSE_CAPABILITIES_OFST 16
+/* Second word of flags. Not present on older firmware (check the length). */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_FLAGS2_OFST 20
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_LBN 0
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_ENCAP_LBN 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_ENCAP_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVQ_TIMER_CTRL_LBN 2
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVQ_TIMER_CTRL_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVENT_CUT_THROUGH_LBN 3
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_EVENT_CUT_THROUGH_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_CUT_THROUGH_LBN 4
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_CUT_THROUGH_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VFIFO_ULL_MODE_LBN 5
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_VFIFO_ULL_MODE_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_MAC_STATS_40G_TX_SIZE_BINS_LBN 6
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_MAC_STATS_40G_TX_SIZE_BINS_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_INIT_EVQ_V2_LBN 7
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_INIT_EVQ_V2_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_TIMESTAMPING_LBN 8
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_MAC_TIMESTAMPING_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TIMESTAMP_LBN 9
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TIMESTAMP_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_SNIFF_LBN 10
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_SNIFF_WIDTH 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_SNIFF_LBN 11
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_SNIFF_WIDTH 1
+/* Number of FATSOv2 contexts per datapath supported by this NIC. Not present
+ * on older firmware (check the length).
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_N_CONTEXTS_OFST 24
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_TSO_V2_N_CONTEXTS_LEN 2
+/* One byte per PF containing the number of the external port assigned to this
+ * PF, indexed by PF number. Special values indicate that a PF is either not
+ * present or not assigned.
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_OFST 26
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_LEN 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PFS_TO_PORTS_ASSIGNMENT_NUM 16
+/* enum: The caller is not permitted to access information on this PF. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_ACCESS_NOT_PERMITTED 0xff
+/* enum: PF does not exist. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_PRESENT 0xfe
+/* enum: PF does exist but is not assigned to any external port. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_ASSIGNED 0xfd
+/* enum: This value indicates that PF is assigned, but it cannot be expressed
+ * in this field. It is intended for a possible future situation where a more
+ * complex scheme of PFs to ports mapping is being used. The future driver
+ * should look for a new field supporting the new scheme. The current/old
+ * driver should treat this value as PF_NOT_ASSIGNED.
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_INCOMPATIBLE_ASSIGNMENT 0xfc
+/* One byte per PF containing the number of its VFs, indexed by PF number. A
+ * special value indicates that a PF is not present.
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_OFST 42
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_LEN 1
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VFS_PER_PF_NUM 16
+/* enum: The caller is not permitted to access information on this PF. */
+/* MC_CMD_GET_CAPABILITIES_V3_OUT_ACCESS_NOT_PERMITTED 0xff */
+/* enum: PF does not exist. */
+/* MC_CMD_GET_CAPABILITIES_V3_OUT_PF_NOT_PRESENT 0xfe */
+/* Number of VIs available for each external port */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_OFST 58
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_LEN 2
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_VIS_PER_PORT_NUM 4
+/* Size of RX descriptor cache expressed as binary logarithm The actual size
+ * equals (2 ^ RX_DESC_CACHE_SIZE)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DESC_CACHE_SIZE_OFST 66
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_RX_DESC_CACHE_SIZE_LEN 1
+/* Size of TX descriptor cache expressed as binary logarithm The actual size
+ * equals (2 ^ TX_DESC_CACHE_SIZE)
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DESC_CACHE_SIZE_OFST 67
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_TX_DESC_CACHE_SIZE_LEN 1
+/* Total number of available PIO buffers */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_PIO_BUFFS_OFST 68
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_NUM_PIO_BUFFS_LEN 2
+/* Size of a single PIO buffer */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_SIZE_PIO_BUFF_OFST 70
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_SIZE_PIO_BUFF_LEN 2
+/* On chips later than Medford the amount of address space assigned to each VI
+ * is configurable. This is a global setting that the driver must query to
+ * discover the VI to address mapping. Cut-through PIO (CTPIO) in not available
+ * with 8k VI windows.
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_OFST 72
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_LEN 1
+/* enum: Each VI occupies 8k as on Huntington and Medford. PIO is at offset 4k.
+ * CTPIO is not mapped.
+ */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_8K 0x0
+/* enum: Each VI occupies 16k. PIO is at offset 4k. CTPIO is at offset 12k. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_16K 0x1
+/* enum: Each VI occupies 64k. PIO is at offset 4k. CTPIO is at offset 12k. */
+#define MC_CMD_GET_CAPABILITIES_V3_OUT_VI_WINDOW_MODE_64K 0x2
+
/***********************************/
/* MC_CMD_V2_EXTN
@@ -9026,7 +9434,7 @@
*/
#define MC_CMD_GET_RXDP_CONFIG 0xc2
-#define MC_CMD_0xc2_PRIVILEGE_CTG SRIOV_CTG_ADMIN
+#define MC_CMD_0xc2_PRIVILEGE_CTG SRIOV_CTG_GENERAL
/* MC_CMD_GET_RXDP_CONFIG_IN msgrequest */
#define MC_CMD_GET_RXDP_CONFIG_IN_LEN 0
@@ -10125,7 +10533,9 @@
* that this operation returns a zero-length response
*/
#define MC_CMD_LICENSING_V3_IN_OP_UPDATE_LICENSE 0x0
-/* enum: report counts of installed licenses */
+/* enum: report counts of installed licenses Returns EAGAIN if license
+ * processing (updating) has been started but not yet completed.
+ */
#define MC_CMD_LICENSING_V3_IN_OP_REPORT_LICENSE 0x1
/* MC_CMD_LICENSING_V3_OUT msgresponse */
@@ -10763,6 +11173,8 @@
#define MC_CMD_GET_WORKAROUNDS_OUT_BUG42008 0x20
/* enum: Bug 26807 features present in firmware (multicast filter chaining) */
#define MC_CMD_GET_WORKAROUNDS_OUT_BUG26807 0x40
+/* enum: Bug 61265 work around (broken EVQ TMR writes). */
+#define MC_CMD_GET_WORKAROUNDS_OUT_BUG61265 0x80
/***********************************/
@@ -11280,22 +11692,110 @@
#define MC_CMD_0x118_PRIVILEGE_CTG SRIOV_CTG_ADMIN
/* MC_CMD_RX_BALANCING_IN msgrequest */
-#define MC_CMD_RX_BALANCING_IN_LEN 4
+#define MC_CMD_RX_BALANCING_IN_LEN 16
/* The RX port whose upconverter table will be modified */
#define MC_CMD_RX_BALANCING_IN_PORT_OFST 0
-#define MC_CMD_RX_BALANCING_IN_PORT_LEN 1
/* The VLAN priority associated to the table index and vFIFO */
-#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 1
-#define MC_CMD_RX_BALANCING_IN_PRIORITY_LEN 1
+#define MC_CMD_RX_BALANCING_IN_PRIORITY_OFST 4
/* The resulting bit of SRC^DST for indexing the table */
-#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 2
-#define MC_CMD_RX_BALANCING_IN_SRC_DST_LEN 1
+#define MC_CMD_RX_BALANCING_IN_SRC_DST_OFST 8
/* The RX engine to which the vFIFO in the table entry will point to */
-#define MC_CMD_RX_BALANCING_IN_ENG_OFST 3
-#define MC_CMD_RX_BALANCING_IN_ENG_LEN 1
+#define MC_CMD_RX_BALANCING_IN_ENG_OFST 12
/* MC_CMD_RX_BALANCING_OUT msgresponse */
#define MC_CMD_RX_BALANCING_OUT_LEN 0
+/***********************************/
+/* MC_CMD_SET_EVQ_TMR
+ * Update the timer load, timer reload and timer mode values for a given EVQ.
+ * The requested timer values (in TMR_LOAD_REQ_NS and TMR_RELOAD_REQ_NS) will
+ * be rounded up to the granularity supported by the hardware, then truncated
+ * to the range supported by the hardware. The resulting value after the
+ * rounding and truncation will be returned to the caller (in TMR_LOAD_ACT_NS
+ * and TMR_RELOAD_ACT_NS).
+ */
+#define MC_CMD_SET_EVQ_TMR 0x120
+
+#define MC_CMD_0x120_PRIVILEGE_CTG SRIOV_CTG_GENERAL
+
+/* MC_CMD_SET_EVQ_TMR_IN msgrequest */
+#define MC_CMD_SET_EVQ_TMR_IN_LEN 16
+/* Function-relative queue instance */
+#define MC_CMD_SET_EVQ_TMR_IN_INSTANCE_OFST 0
+/* Requested value for timer load (in nanoseconds) */
+#define MC_CMD_SET_EVQ_TMR_IN_TMR_LOAD_REQ_NS_OFST 4
+/* Requested value for timer reload (in nanoseconds) */
+#define MC_CMD_SET_EVQ_TMR_IN_TMR_RELOAD_REQ_NS_OFST 8
+/* Timer mode. Meanings as per EVQ_TMR_REG.TC_TIMER_VAL */
+#define MC_CMD_SET_EVQ_TMR_IN_TMR_MODE_OFST 12
+#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_DIS 0x0 /* enum */
+#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_IMMED_START 0x1 /* enum */
+#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_TRIG_START 0x2 /* enum */
+#define MC_CMD_SET_EVQ_TMR_IN_TIMER_MODE_INT_HLDOFF 0x3 /* enum */
+
+/* MC_CMD_SET_EVQ_TMR_OUT msgresponse */
+#define MC_CMD_SET_EVQ_TMR_OUT_LEN 8
+/* Actual value for timer load (in nanoseconds) */
+#define MC_CMD_SET_EVQ_TMR_OUT_TMR_LOAD_ACT_NS_OFST 0
+/* Actual value for timer reload (in nanoseconds) */
+#define MC_CMD_SET_EVQ_TMR_OUT_TMR_RELOAD_ACT_NS_OFST 4
+
+
+/***********************************/
+/* MC_CMD_GET_EVQ_TMR_PROPERTIES
+ * Query properties about the event queue timers.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES 0x122
+
+#define MC_CMD_0x122_PRIVILEGE_CTG SRIOV_CTG_GENERAL
+
+/* MC_CMD_GET_EVQ_TMR_PROPERTIES_IN msgrequest */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_IN_LEN 0
+
+/* MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT msgresponse */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN 36
+/* Reserved for future use. */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_FLAGS_OFST 0
+/* For timers updated via writes to EVQ_TMR_REG, this is the time interval (in
+ * nanoseconds) for each increment of the timer load/reload count. The
+ * requested duration of a timer is this value multiplied by the timer
+ * load/reload count.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_NS_PER_COUNT_OFST 4
+/* For timers updated via writes to EVQ_TMR_REG, this is the maximum value
+ * allowed for timer load/reload counts.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_MAX_COUNT_OFST 8
+/* For timers updated via writes to EVQ_TMR_REG, timer load/reload counts not a
+ * multiple of this step size will be rounded in an implementation defined
+ * manner.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_STEP_OFST 12
+/* Maximum timer duration (in nanoseconds) for timers updated via MCDI. Only
+ * meaningful if MC_CMD_SET_EVQ_TMR is implemented.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_MAX_NS_OFST 16
+/* Timer durations requested via MCDI that are not a multiple of this step size
+ * will be rounded up. Only meaningful if MC_CMD_SET_EVQ_TMR is implemented.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_STEP_NS_OFST 20
+/* For timers updated using the bug35388 workaround, this is the time interval
+ * (in nanoseconds) for each increment of the timer load/reload count. The
+ * requested duration of a timer is this value multiplied by the timer
+ * load/reload count. This field is only meaningful if the bug35388 workaround
+ * is enabled.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_NS_PER_COUNT_OFST 24
+/* For timers updated using the bug35388 workaround, this is the maximum value
+ * allowed for timer load/reload counts. This field is only meaningful if the
+ * bug35388 workaround is enabled.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_MAX_COUNT_OFST 28
+/* For timers updated using the bug35388 workaround, timer load/reload counts
+ * not a multiple of this step size will be rounded in an implementation
+ * defined manner. This field is only meaningful if the bug35388 workaround is
+ * enabled.
+ */
+#define MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_STEP_OFST 32
#endif /* MCDI_PCOL_H */
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 2/6] sfc: allow asynchronous MCDI without completion function
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 1/6] sfc: update MCDI protocol headers Bert Kenward
@ 2016-08-11 12:01 ` Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 3/6] sfc: retrieve second word of datapath capabilities Bert Kenward
` (4 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:01 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/mcdi.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
index d28e7dd..9fbc12a 100644
--- a/drivers/net/ethernet/sfc/mcdi.c
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -548,7 +548,10 @@ static bool efx_mcdi_complete_async(struct efx_mcdi_iface *mcdi, bool timeout)
efx_mcdi_display_error(efx, async->cmd, async->inlen, errbuf,
err_len, rc);
}
- async->complete(efx, async->cookie, rc, outbuf, data_len);
+
+ if (async->complete)
+ async->complete(efx, async->cookie, rc, outbuf,
+ min(async->outlen, data_len));
kfree(async);
efx_mcdi_release(mcdi);
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 3/6] sfc: retrieve second word of datapath capabilities
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 1/6] sfc: update MCDI protocol headers Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 2/6] sfc: allow asynchronous MCDI without completion function Bert Kenward
@ 2016-08-11 12:01 ` Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 4/6] sfc: use new performance based event queue init Bert Kenward
` (3 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:01 UTC (permalink / raw)
To: Dave Miller; +Cc: linux-net-drivers, netdev
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 10 ++++++++--
drivers/net/ethernet/sfc/nic.h | 3 +++
2 files changed, 11 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index f658fee..fd5d086 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -177,7 +177,7 @@ static int efx_ef10_get_vf_index(struct efx_nic *efx)
static int efx_ef10_init_datapath_caps(struct efx_nic *efx)
{
- MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_CAPABILITIES_OUT_LEN);
+ MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_CAPABILITIES_V2_OUT_LEN);
struct efx_ef10_nic_data *nic_data = efx->nic_data;
size_t outlen;
int rc;
@@ -188,7 +188,7 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx)
outbuf, sizeof(outbuf), &outlen);
if (rc)
return rc;
- if (outlen < sizeof(outbuf)) {
+ if (outlen < MC_CMD_GET_CAPABILITIES_OUT_LEN) {
netif_err(efx, drv, efx->net_dev,
"unable to read datapath firmware capabilities\n");
return -EIO;
@@ -197,6 +197,12 @@ static int efx_ef10_init_datapath_caps(struct efx_nic *efx)
nic_data->datapath_caps =
MCDI_DWORD(outbuf, GET_CAPABILITIES_OUT_FLAGS1);
+ if (outlen >= MC_CMD_GET_CAPABILITIES_V2_OUT_LEN)
+ nic_data->datapath_caps2 = MCDI_DWORD(outbuf,
+ GET_CAPABILITIES_V2_OUT_FLAGS2);
+ else
+ nic_data->datapath_caps2 = 0;
+
/* record the DPCPU firmware IDs to determine VEB vswitching support.
*/
nic_data->rx_dpcpu_fw_id =
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index 96944c3..531c5a8 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -511,6 +511,8 @@ enum {
* after MC reboot
* @datapath_caps: Capabilities of datapath firmware (FLAGS1 field of
* %MC_CMD_GET_CAPABILITIES response)
+ * @datapath_caps2: Further Capabilities of datapath firmware (FLAGS2 field of
+ * %MC_CMD_GET_CAPABILITIES response)
* @rx_dpcpu_fw_id: Firmware ID of the RxDPCPU
* @tx_dpcpu_fw_id: Firmware ID of the TxDPCPU
* @vport_id: The function's vport ID, only relevant for PFs
@@ -542,6 +544,7 @@ struct efx_ef10_nic_data {
bool workaround_26807;
bool must_check_datapath_caps;
u32 datapath_caps;
+ u32 datapath_caps2;
unsigned int rx_dpcpu_fw_id;
unsigned int tx_dpcpu_fw_id;
unsigned int vport_id;
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 4/6] sfc: use new performance based event queue init
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
` (2 preceding siblings ...)
2016-08-11 12:01 ` [PATCH net-next 3/6] sfc: retrieve second word of datapath capabilities Bert Kenward
@ 2016-08-11 12:01 ` Bert Kenward
2016-08-11 12:02 ` [PATCH net-next 5/6] sfc: set interrupt moderation via MCDI Bert Kenward
` (2 subsequent siblings)
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:01 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
Rather than explicitly specifying flags we can now specify a desired
performance target to the firmware, ie higher throughput or lower latency.
For now we use the default "auto" configuration.
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 43 +++++++++++++++++++++++++++++------------
1 file changed, 31 insertions(+), 12 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index fd5d086..e57fa2e 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -2541,13 +2541,12 @@ fail:
static int efx_ef10_ev_init(struct efx_channel *channel)
{
MCDI_DECLARE_BUF(inbuf,
- MC_CMD_INIT_EVQ_IN_LEN(EFX_MAX_EVQ_SIZE * 8 /
- EFX_BUF_SIZE));
- MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_OUT_LEN);
+ MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_MAX_EVQ_SIZE * 8 /
+ EFX_BUF_SIZE));
+ MCDI_DECLARE_BUF(outbuf, MC_CMD_INIT_EVQ_V2_OUT_LEN);
size_t entries = channel->eventq.buf.len / EFX_BUF_SIZE;
struct efx_nic *efx = channel->efx;
struct efx_ef10_nic_data *nic_data;
- bool supports_rx_merge;
size_t inlen, outlen;
unsigned int enabled, implemented;
dma_addr_t dma_addr;
@@ -2555,9 +2554,6 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
int i;
nic_data = efx->nic_data;
- supports_rx_merge =
- !!(nic_data->datapath_caps &
- 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN);
/* Fill event queue with all ones (i.e. empty events) */
memset(channel->eventq.buf.addr, 0xff, channel->eventq.buf.len);
@@ -2566,11 +2562,6 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_INSTANCE, channel->channel);
/* INIT_EVQ expects index in vector table, not absolute */
MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_IRQ_NUM, channel->channel);
- MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS,
- INIT_EVQ_IN_FLAG_INTERRUPTING, 1,
- INIT_EVQ_IN_FLAG_RX_MERGE, 1,
- INIT_EVQ_IN_FLAG_TX_MERGE, 1,
- INIT_EVQ_IN_FLAG_CUT_THRU, !supports_rx_merge);
MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_MODE,
MC_CMD_INIT_EVQ_IN_TMR_MODE_DIS);
MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_TMR_LOAD, 0);
@@ -2579,6 +2570,27 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
MC_CMD_INIT_EVQ_IN_COUNT_MODE_DIS);
MCDI_SET_DWORD(inbuf, INIT_EVQ_IN_COUNT_THRSHLD, 0);
+ if (nic_data->datapath_caps2 &
+ 1 << MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN) {
+ /* Use the new generic approach to specifying event queue
+ * configuration, requesting lower latency or higher throughput.
+ * The options that actually get used appear in the output.
+ */
+ MCDI_POPULATE_DWORD_2(inbuf, INIT_EVQ_V2_IN_FLAGS,
+ INIT_EVQ_V2_IN_FLAG_INTERRUPTING, 1,
+ INIT_EVQ_V2_IN_FLAG_TYPE,
+ MC_CMD_INIT_EVQ_V2_IN_FLAG_TYPE_AUTO);
+ } else {
+ bool cut_thru = !(nic_data->datapath_caps &
+ 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN);
+
+ MCDI_POPULATE_DWORD_4(inbuf, INIT_EVQ_IN_FLAGS,
+ INIT_EVQ_IN_FLAG_INTERRUPTING, 1,
+ INIT_EVQ_IN_FLAG_RX_MERGE, 1,
+ INIT_EVQ_IN_FLAG_TX_MERGE, 1,
+ INIT_EVQ_IN_FLAG_CUT_THRU, cut_thru);
+ }
+
dma_addr = channel->eventq.buf.dma_addr;
for (i = 0; i < entries; ++i) {
MCDI_SET_ARRAY_QWORD(inbuf, INIT_EVQ_IN_DMA_ADDR, i, dma_addr);
@@ -2589,6 +2601,13 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
rc = efx_mcdi_rpc(efx, MC_CMD_INIT_EVQ, inbuf, inlen,
outbuf, sizeof(outbuf), &outlen);
+
+ if (outlen >= MC_CMD_INIT_EVQ_V2_OUT_LEN)
+ netif_dbg(efx, drv, efx->net_dev,
+ "Channel %d using event queue flags %08x\n",
+ channel->channel,
+ MCDI_DWORD(outbuf, INIT_EVQ_V2_OUT_FLAGS));
+
/* IRQ return is ignored */
if (channel->channel || rc)
return rc;
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 5/6] sfc: set interrupt moderation via MCDI
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
` (3 preceding siblings ...)
2016-08-11 12:01 ` [PATCH net-next 4/6] sfc: use new performance based event queue init Bert Kenward
@ 2016-08-11 12:02 ` Bert Kenward
2016-08-11 12:02 ` [PATCH net-next 6/6] sfc: get timer configuration from adapter Bert Kenward
2016-08-13 3:51 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements David Miller
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:02 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
SFN8000-series NICs require a new method of setting interrupt moderation,
via MCDI. This is indicated by a workaround flag. This new MCDI command
takes an explicit time value rather than a number of ticks. It therefore
makes sense to also store the moderation values in terms of time, since
that is what the ethtool interface is interested in.
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 38 +++++++++++----
drivers/net/ethernet/sfc/efx.c | 89 ++++++++++++++++++----------------
drivers/net/ethernet/sfc/efx.h | 2 +
drivers/net/ethernet/sfc/falcon.c | 7 ++-
drivers/net/ethernet/sfc/net_driver.h | 10 ++--
drivers/net/ethernet/sfc/nic.h | 2 +
drivers/net/ethernet/sfc/ptp.c | 2 +-
drivers/net/ethernet/sfc/siena.c | 11 +++--
drivers/net/ethernet/sfc/workarounds.h | 4 ++
9 files changed, 103 insertions(+), 62 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index e57fa2e..5f6c4fa 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -1749,27 +1749,43 @@ static size_t efx_ef10_update_stats_vf(struct efx_nic *efx, u64 *full_stats,
static void efx_ef10_push_irq_moderation(struct efx_channel *channel)
{
struct efx_nic *efx = channel->efx;
- unsigned int mode, value;
+ unsigned int mode, usecs;
efx_dword_t timer_cmd;
- if (channel->irq_moderation) {
+ if (channel->irq_moderation_us) {
mode = 3;
- value = channel->irq_moderation - 1;
+ usecs = channel->irq_moderation_us;
} else {
mode = 0;
- value = 0;
+ usecs = 0;
}
- if (EFX_EF10_WORKAROUND_35388(efx)) {
+ if (EFX_EF10_WORKAROUND_61265(efx)) {
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_SET_EVQ_TMR_IN_LEN);
+ unsigned int ns = usecs * 1000;
+
+ MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_INSTANCE,
+ channel->channel);
+ MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_LOAD_REQ_NS, ns);
+ MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_RELOAD_REQ_NS, ns);
+ MCDI_SET_DWORD(inbuf, SET_EVQ_TMR_IN_TMR_MODE, mode);
+
+ efx_mcdi_rpc_async(efx, MC_CMD_SET_EVQ_TMR,
+ inbuf, sizeof(inbuf), 0, NULL, 0);
+ } else if (EFX_EF10_WORKAROUND_35388(efx)) {
+ unsigned int ticks = efx_usecs_to_ticks(efx, usecs);
+
EFX_POPULATE_DWORD_3(timer_cmd, ERF_DD_EVQ_IND_TIMER_FLAGS,
EFE_DD_EVQ_IND_TIMER_FLAGS,
ERF_DD_EVQ_IND_TIMER_MODE, mode,
- ERF_DD_EVQ_IND_TIMER_VAL, value);
+ ERF_DD_EVQ_IND_TIMER_VAL, ticks);
efx_writed_page(efx, &timer_cmd, ER_DD_EVQ_INDIRECT,
channel->channel);
} else {
+ unsigned int ticks = efx_usecs_to_ticks(efx, usecs);
+
EFX_POPULATE_DWORD_2(timer_cmd, ERF_DZ_TC_TIMER_MODE, mode,
- ERF_DZ_TC_TIMER_VAL, value);
+ ERF_DZ_TC_TIMER_VAL, ticks);
efx_writed_page(efx, &timer_cmd, ER_DZ_EVQ_TMR,
channel->channel);
}
@@ -2615,10 +2631,11 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
/* Successfully created event queue on channel 0 */
rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled);
if (rc == -ENOSYS) {
- /* GET_WORKAROUNDS was implemented before the bug26807
- * workaround, thus the latter must be unavailable in this fw
+ /* GET_WORKAROUNDS was implemented before these workarounds,
+ * thus they must be unavailable in this firmware.
*/
nic_data->workaround_26807 = false;
+ nic_data->workaround_61265 = false;
rc = 0;
} else if (rc) {
goto fail;
@@ -2658,6 +2675,9 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
rc = 0;
}
}
+
+ nic_data->workaround_61265 =
+ !!(implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG61265);
}
if (!rc)
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 14b821b..b26e1f9 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -281,6 +281,27 @@ static int efx_process_channel(struct efx_channel *channel, int budget)
* NAPI guarantees serialisation of polls of the same device, which
* provides the guarantee required by efx_process_channel().
*/
+static void efx_update_irq_mod(struct efx_nic *efx, struct efx_channel *channel)
+{
+ int step = efx->irq_mod_step_us;
+
+ if (channel->irq_mod_score < irq_adapt_low_thresh) {
+ if (channel->irq_moderation_us > step) {
+ channel->irq_moderation_us -= step;
+ efx->type->push_irq_moderation(channel);
+ }
+ } else if (channel->irq_mod_score > irq_adapt_high_thresh) {
+ if (channel->irq_moderation_us <
+ efx->irq_rx_moderation_us) {
+ channel->irq_moderation_us += step;
+ efx->type->push_irq_moderation(channel);
+ }
+ }
+
+ channel->irq_count = 0;
+ channel->irq_mod_score = 0;
+}
+
static int efx_poll(struct napi_struct *napi, int budget)
{
struct efx_channel *channel =
@@ -301,22 +322,7 @@ static int efx_poll(struct napi_struct *napi, int budget)
if (efx_channel_has_rx_queue(channel) &&
efx->irq_rx_adaptive &&
unlikely(++channel->irq_count == 1000)) {
- if (unlikely(channel->irq_mod_score <
- irq_adapt_low_thresh)) {
- if (channel->irq_moderation > 1) {
- channel->irq_moderation -= 1;
- efx->type->push_irq_moderation(channel);
- }
- } else if (unlikely(channel->irq_mod_score >
- irq_adapt_high_thresh)) {
- if (channel->irq_moderation <
- efx->irq_rx_moderation) {
- channel->irq_moderation += 1;
- efx->type->push_irq_moderation(channel);
- }
- }
- channel->irq_count = 0;
- channel->irq_mod_score = 0;
+ efx_update_irq_mod(efx, channel);
}
efx_filter_rfs_expire(channel);
@@ -1703,6 +1709,7 @@ static int efx_probe_nic(struct efx_nic *efx)
netif_set_real_num_rx_queues(efx->net_dev, efx->n_rx_channels);
/* Initialise the interrupt moderation settings */
+ efx->irq_mod_step_us = DIV_ROUND_UP(efx->timer_quantum_ns, 1000);
efx_init_irq_moderation(efx, tx_irq_mod_usec, rx_irq_mod_usec, true,
true);
@@ -1949,14 +1956,21 @@ static void efx_remove_all(struct efx_nic *efx)
* Interrupt moderation
*
**************************************************************************/
-
-static unsigned int irq_mod_ticks(unsigned int usecs, unsigned int quantum_ns)
+unsigned int efx_usecs_to_ticks(struct efx_nic *efx, unsigned int usecs)
{
if (usecs == 0)
return 0;
- if (usecs * 1000 < quantum_ns)
+ if (usecs * 1000 < efx->timer_quantum_ns)
return 1; /* never round down to 0 */
- return usecs * 1000 / quantum_ns;
+ return usecs * 1000 / efx->timer_quantum_ns;
+}
+
+unsigned int efx_ticks_to_usecs(struct efx_nic *efx, unsigned int ticks)
+{
+ /* We must round up when converting ticks to microseconds
+ * because we round down when converting the other way.
+ */
+ return DIV_ROUND_UP(ticks * efx->timer_quantum_ns, 1000);
}
/* Set interrupt moderation parameters */
@@ -1968,18 +1982,12 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
unsigned int irq_mod_max = DIV_ROUND_UP(efx->type->timer_period_max *
efx->timer_quantum_ns,
1000);
- unsigned int tx_ticks;
- unsigned int rx_ticks;
-
EFX_ASSERT_RESET_SERIALISED(efx);
if (tx_usecs > irq_mod_max || rx_usecs > irq_mod_max)
return -EINVAL;
- tx_ticks = irq_mod_ticks(tx_usecs, efx->timer_quantum_ns);
- rx_ticks = irq_mod_ticks(rx_usecs, efx->timer_quantum_ns);
-
- if (tx_ticks != rx_ticks && efx->tx_channel_offset == 0 &&
+ if (tx_usecs != rx_usecs && efx->tx_channel_offset == 0 &&
!rx_may_override_tx) {
netif_err(efx, drv, efx->net_dev, "Channels are shared. "
"RX and TX IRQ moderation must be equal\n");
@@ -1987,12 +1995,12 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
}
efx->irq_rx_adaptive = rx_adaptive;
- efx->irq_rx_moderation = rx_ticks;
+ efx->irq_rx_moderation_us = rx_usecs;
efx_for_each_channel(channel, efx) {
if (efx_channel_has_rx_queue(channel))
- channel->irq_moderation = rx_ticks;
+ channel->irq_moderation_us = rx_usecs;
else if (efx_channel_has_tx_queues(channel))
- channel->irq_moderation = tx_ticks;
+ channel->irq_moderation_us = tx_usecs;
}
return 0;
@@ -2001,26 +2009,21 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
void efx_get_irq_moderation(struct efx_nic *efx, unsigned int *tx_usecs,
unsigned int *rx_usecs, bool *rx_adaptive)
{
- /* We must round up when converting ticks to microseconds
- * because we round down when converting the other way.
- */
-
*rx_adaptive = efx->irq_rx_adaptive;
- *rx_usecs = DIV_ROUND_UP(efx->irq_rx_moderation *
- efx->timer_quantum_ns,
- 1000);
+ *rx_usecs = efx->irq_rx_moderation_us;
/* If channels are shared between RX and TX, so is IRQ
* moderation. Otherwise, IRQ moderation is the same for all
* TX channels and is not adaptive.
*/
- if (efx->tx_channel_offset == 0)
+ if (efx->tx_channel_offset == 0) {
*tx_usecs = *rx_usecs;
- else
- *tx_usecs = DIV_ROUND_UP(
- efx->channel[efx->tx_channel_offset]->irq_moderation *
- efx->timer_quantum_ns,
- 1000);
+ } else {
+ struct efx_channel *tx_channel;
+
+ tx_channel = efx->channel[efx->tx_channel_offset];
+ *tx_usecs = tx_channel->irq_moderation_us;
+ }
}
/**************************************************************************
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index c3ae739..342ae16 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -204,6 +204,8 @@ int efx_try_recovery(struct efx_nic *efx);
/* Global */
void efx_schedule_reset(struct efx_nic *efx, enum reset_type type);
+unsigned int efx_usecs_to_ticks(struct efx_nic *efx, unsigned int usecs);
+unsigned int efx_ticks_to_usecs(struct efx_nic *efx, unsigned int ticks);
int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
unsigned int rx_usecs, bool rx_adaptive,
bool rx_may_override_tx);
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index d790cb8..d3d0bcc 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -378,12 +378,15 @@ static void falcon_push_irq_moderation(struct efx_channel *channel)
struct efx_nic *efx = channel->efx;
/* Set timer register */
- if (channel->irq_moderation) {
+ if (channel->irq_moderation_us) {
+ unsigned int ticks;
+
+ ticks = efx_usecs_to_ticks(efx, channel->irq_moderation_us);
EFX_POPULATE_DWORD_2(timer_cmd,
FRF_AB_TC_TIMER_MODE,
FFE_BB_TIMER_MODE_INT_HLDOFF,
FRF_AB_TC_TIMER_VAL,
- channel->irq_moderation - 1);
+ ticks - 1);
} else {
EFX_POPULATE_DWORD_2(timer_cmd,
FRF_AB_TC_TIMER_MODE,
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 9ff062a..3e8c6fb 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -392,7 +392,7 @@ enum efx_sync_events_state {
* @eventq_init: Event queue initialised flag
* @enabled: Channel enabled indicator
* @irq: IRQ number (MSI and MSI-X only)
- * @irq_moderation: IRQ moderation value (in hardware ticks)
+ * @irq_moderation_us: IRQ moderation value (in microseconds)
* @napi_dev: Net device used with NAPI
* @napi_str: NAPI control structure
* @state: state for NAPI vs busy polling
@@ -433,7 +433,7 @@ struct efx_channel {
bool eventq_init;
bool enabled;
int irq;
- unsigned int irq_moderation;
+ unsigned int irq_moderation_us;
struct net_device *napi_dev;
struct napi_struct napi_str;
#ifdef CONFIG_NET_RX_BUSY_POLL
@@ -811,7 +811,8 @@ struct vfdi_status;
* @interrupt_mode: Interrupt mode
* @timer_quantum_ns: Interrupt timer quantum, in nanoseconds
* @irq_rx_adaptive: Adaptive IRQ moderation enabled for RX event queues
- * @irq_rx_moderation: IRQ moderation time for RX event queues
+ * @irq_rx_mod_step_us: Step size for IRQ moderation for RX event queues
+ * @irq_rx_moderation_us: IRQ moderation time for RX event queues
* @msg_enable: Log message enable flags
* @state: Device state number (%STATE_*). Serialised by the rtnl_lock.
* @reset_pending: Bitmask for pending resets
@@ -941,7 +942,8 @@ struct efx_nic {
enum efx_int_mode interrupt_mode;
unsigned int timer_quantum_ns;
bool irq_rx_adaptive;
- unsigned int irq_rx_moderation;
+ unsigned int irq_mod_step_us;
+ unsigned int irq_rx_moderation_us;
u32 msg_enable;
enum nic_state state;
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index 531c5a8..d8b1694 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -507,6 +507,7 @@ enum {
* @stats: Hardware statistics
* @workaround_35388: Flag: firmware supports workaround for bug 35388
* @workaround_26807: Flag: firmware supports workaround for bug 26807
+ * @workaround_61265: Flag: firmware supports workaround for bug 61265
* @must_check_datapath_caps: Flag: @datapath_caps needs to be revalidated
* after MC reboot
* @datapath_caps: Capabilities of datapath firmware (FLAGS1 field of
@@ -542,6 +543,7 @@ struct efx_ef10_nic_data {
u64 stats[EF10_STAT_COUNT];
bool workaround_35388;
bool workaround_26807;
+ bool workaround_61265;
bool must_check_datapath_caps;
u32 datapath_caps;
u32 datapath_caps2;
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index c771e0a..dd204d9 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -1306,7 +1306,7 @@ static int efx_ptp_probe_channel(struct efx_channel *channel)
{
struct efx_nic *efx = channel->efx;
- channel->irq_moderation = 0;
+ channel->irq_moderation_us = 0;
channel->rx_queue.core_index = 0;
return efx_ptp_probe(efx, channel);
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 2219b54..994e33f 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -34,19 +34,24 @@ static void siena_init_wol(struct efx_nic *efx);
static void siena_push_irq_moderation(struct efx_channel *channel)
{
+ struct efx_nic *efx = channel->efx;
efx_dword_t timer_cmd;
- if (channel->irq_moderation)
+ if (channel->irq_moderation_us) {
+ unsigned int ticks;
+
+ ticks = efx_usecs_to_ticks(efx, channel->irq_moderation_us);
EFX_POPULATE_DWORD_2(timer_cmd,
FRF_CZ_TC_TIMER_MODE,
FFE_CZ_TIMER_MODE_INT_HLDOFF,
FRF_CZ_TC_TIMER_VAL,
- channel->irq_moderation - 1);
- else
+ ticks - 1);
+ } else {
EFX_POPULATE_DWORD_2(timer_cmd,
FRF_CZ_TC_TIMER_MODE,
FFE_CZ_TIMER_MODE_DIS,
FRF_CZ_TC_TIMER_VAL, 0);
+ }
efx_writed_page_locked(channel->efx, &timer_cmd, FR_BZ_TIMER_COMMAND_P0,
channel->channel);
}
diff --git a/drivers/net/ethernet/sfc/workarounds.h b/drivers/net/ethernet/sfc/workarounds.h
index 2310b75..351cd14 100644
--- a/drivers/net/ethernet/sfc/workarounds.h
+++ b/drivers/net/ethernet/sfc/workarounds.h
@@ -50,4 +50,8 @@
#define EFX_WORKAROUND_35388(efx) \
(efx_nic_rev(efx) == EFX_REV_HUNT_A0 && EFX_EF10_WORKAROUND_35388(efx))
+/* Moderation timer access must go through MCDI */
+#define EFX_EF10_WORKAROUND_61265(efx) \
+ (((struct efx_ef10_nic_data *)efx->nic_data)->workaround_61265)
+
#endif /* EFX_WORKAROUNDS_H */
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* [PATCH net-next 6/6] sfc: get timer configuration from adapter
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
` (4 preceding siblings ...)
2016-08-11 12:02 ` [PATCH net-next 5/6] sfc: set interrupt moderation via MCDI Bert Kenward
@ 2016-08-11 12:02 ` Bert Kenward
2016-08-13 3:51 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements David Miller
6 siblings, 0 replies; 8+ messages in thread
From: Bert Kenward @ 2016-08-11 12:02 UTC (permalink / raw)
To: Dave Miller; +Cc: netdev, linux-net-drivers
On SFN8000 series adapters the MC provides a method to get the timer
quantum and the maximum timer setting. We revert to the old values if the
new call is unavailable.
Signed-off-by: Bert Kenward <bkenward@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 142 +++++++++++++++++++++++++++-------
drivers/net/ethernet/sfc/efx.c | 9 ++-
drivers/net/ethernet/sfc/falcon.c | 2 +
drivers/net/ethernet/sfc/net_driver.h | 2 +
drivers/net/ethernet/sfc/siena.c | 3 +
5 files changed, 125 insertions(+), 33 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 5f6c4fa..b8c9f18 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -233,6 +233,116 @@ static int efx_ef10_get_sysclk_freq(struct efx_nic *efx)
return rc > 0 ? rc : -ERANGE;
}
+static int efx_ef10_get_timer_workarounds(struct efx_nic *efx)
+{
+ struct efx_ef10_nic_data *nic_data = efx->nic_data;
+ unsigned int implemented;
+ unsigned int enabled;
+ int rc;
+
+ nic_data->workaround_35388 = false;
+ nic_data->workaround_61265 = false;
+
+ rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled);
+
+ if (rc == -ENOSYS) {
+ /* Firmware without GET_WORKAROUNDS - not a problem. */
+ rc = 0;
+ } else if (rc == 0) {
+ /* Bug61265 workaround is always enabled if implemented. */
+ if (enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG61265)
+ nic_data->workaround_61265 = true;
+
+ if (enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG35388) {
+ nic_data->workaround_35388 = true;
+ } else if (implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG35388) {
+ /* Workaround is implemented but not enabled.
+ * Try to enable it.
+ */
+ rc = efx_mcdi_set_workaround(efx,
+ MC_CMD_WORKAROUND_BUG35388,
+ true, NULL);
+ if (rc == 0)
+ nic_data->workaround_35388 = true;
+ /* If we failed to set the workaround just carry on. */
+ rc = 0;
+ }
+ }
+
+ netif_dbg(efx, probe, efx->net_dev,
+ "workaround for bug 35388 is %sabled\n",
+ nic_data->workaround_35388 ? "en" : "dis");
+ netif_dbg(efx, probe, efx->net_dev,
+ "workaround for bug 61265 is %sabled\n",
+ nic_data->workaround_61265 ? "en" : "dis");
+
+ return rc;
+}
+
+static void efx_ef10_process_timer_config(struct efx_nic *efx,
+ const efx_dword_t *data)
+{
+ unsigned int max_count;
+
+ if (EFX_EF10_WORKAROUND_61265(efx)) {
+ efx->timer_quantum_ns = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_STEP_NS);
+ efx->timer_max_ns = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_MCDI_TMR_MAX_NS);
+ } else if (EFX_EF10_WORKAROUND_35388(efx)) {
+ efx->timer_quantum_ns = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_NS_PER_COUNT);
+ max_count = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_BUG35388_TMR_MAX_COUNT);
+ efx->timer_max_ns = max_count * efx->timer_quantum_ns;
+ } else {
+ efx->timer_quantum_ns = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_NS_PER_COUNT);
+ max_count = MCDI_DWORD(data,
+ GET_EVQ_TMR_PROPERTIES_OUT_TMR_REG_MAX_COUNT);
+ efx->timer_max_ns = max_count * efx->timer_quantum_ns;
+ }
+
+ netif_dbg(efx, probe, efx->net_dev,
+ "got timer properties from MC: quantum %u ns; max %u ns\n",
+ efx->timer_quantum_ns, efx->timer_max_ns);
+}
+
+static int efx_ef10_get_timer_config(struct efx_nic *efx)
+{
+ MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN);
+ int rc;
+
+ rc = efx_ef10_get_timer_workarounds(efx);
+ if (rc)
+ return rc;
+
+ rc = efx_mcdi_rpc_quiet(efx, MC_CMD_GET_EVQ_TMR_PROPERTIES, NULL, 0,
+ outbuf, sizeof(outbuf), NULL);
+
+ if (rc == 0) {
+ efx_ef10_process_timer_config(efx, outbuf);
+ } else if (rc == -ENOSYS || rc == -EPERM) {
+ /* Not available - fall back to Huntington defaults. */
+ unsigned int quantum;
+
+ rc = efx_ef10_get_sysclk_freq(efx);
+ if (rc < 0)
+ return rc;
+
+ quantum = 1536000 / rc; /* 1536 cycles */
+ efx->timer_quantum_ns = quantum;
+ efx->timer_max_ns = efx->type->timer_period_max * quantum;
+ rc = 0;
+ } else {
+ efx_mcdi_display_error(efx, MC_CMD_GET_EVQ_TMR_PROPERTIES,
+ MC_CMD_GET_EVQ_TMR_PROPERTIES_OUT_LEN,
+ NULL, 0, rc);
+ }
+
+ return rc;
+}
+
static int efx_ef10_get_mac_address_pf(struct efx_nic *efx, u8 *mac_address)
{
MCDI_DECLARE_BUF(outbuf, MC_CMD_GET_MAC_ADDRESSES_OUT_LEN);
@@ -533,33 +643,11 @@ static int efx_ef10_probe(struct efx_nic *efx)
if (rc)
goto fail5;
- rc = efx_ef10_get_sysclk_freq(efx);
+ rc = efx_ef10_get_timer_config(efx);
if (rc < 0)
goto fail5;
efx->timer_quantum_ns = 1536000 / rc; /* 1536 cycles */
- /* Check whether firmware supports bug 35388 workaround.
- * First try to enable it, then if we get EPERM, just
- * ask if it's already enabled
- */
- rc = efx_mcdi_set_workaround(efx, MC_CMD_WORKAROUND_BUG35388, true, NULL);
- if (rc == 0) {
- nic_data->workaround_35388 = true;
- } else if (rc == -EPERM) {
- unsigned int enabled;
-
- rc = efx_mcdi_get_workarounds(efx, NULL, &enabled);
- if (rc)
- goto fail3;
- nic_data->workaround_35388 = enabled &
- MC_CMD_GET_WORKAROUNDS_OUT_BUG35388;
- } else if (rc != -ENOSYS && rc != -ENOENT) {
- goto fail5;
- }
- netif_dbg(efx, probe, efx->net_dev,
- "workaround for bug 35388 is %sabled\n",
- nic_data->workaround_35388 ? "en" : "dis");
-
rc = efx_mcdi_mon_probe(efx);
if (rc && rc != -EPERM)
goto fail5;
@@ -2631,11 +2719,10 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
/* Successfully created event queue on channel 0 */
rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled);
if (rc == -ENOSYS) {
- /* GET_WORKAROUNDS was implemented before these workarounds,
- * thus they must be unavailable in this firmware.
+ /* GET_WORKAROUNDS was implemented before this workaround,
+ * thus it must be unavailable in this firmware.
*/
nic_data->workaround_26807 = false;
- nic_data->workaround_61265 = false;
rc = 0;
} else if (rc) {
goto fail;
@@ -2675,9 +2762,6 @@ static int efx_ef10_ev_init(struct efx_channel *channel)
rc = 0;
}
}
-
- nic_data->workaround_61265 =
- !!(implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG61265);
}
if (!rc)
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index b26e1f9..f3826ae 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -1979,12 +1979,13 @@ int efx_init_irq_moderation(struct efx_nic *efx, unsigned int tx_usecs,
bool rx_may_override_tx)
{
struct efx_channel *channel;
- unsigned int irq_mod_max = DIV_ROUND_UP(efx->type->timer_period_max *
- efx->timer_quantum_ns,
- 1000);
+ unsigned int timer_max_us;
+
EFX_ASSERT_RESET_SERIALISED(efx);
- if (tx_usecs > irq_mod_max || rx_usecs > irq_mod_max)
+ timer_max_us = efx->timer_max_ns / 1000;
+
+ if (tx_usecs > timer_max_us || rx_usecs > timer_max_us)
return -EINVAL;
if (tx_usecs != rx_usecs && efx->tx_channel_offset == 0 &&
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index d3d0bcc..1a70926 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -2376,6 +2376,8 @@ static int falcon_probe_nic(struct efx_nic *efx)
EFX_MAX_CHANNELS);
efx->max_tx_channels = efx->max_channels;
efx->timer_quantum_ns = 4968; /* 621 cycles */
+ efx->timer_max_ns = efx->type->timer_period_max *
+ efx->timer_quantum_ns;
/* Initialise I2C adapter */
board = falcon_board(efx);
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 3e8c6fb..13b7f52 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -810,6 +810,7 @@ struct vfdi_status;
* @membase: Memory BAR value
* @interrupt_mode: Interrupt mode
* @timer_quantum_ns: Interrupt timer quantum, in nanoseconds
+ * @timer_max_ns: Interrupt timer maximum value, in nanoseconds
* @irq_rx_adaptive: Adaptive IRQ moderation enabled for RX event queues
* @irq_rx_mod_step_us: Step size for IRQ moderation for RX event queues
* @irq_rx_moderation_us: IRQ moderation time for RX event queues
@@ -941,6 +942,7 @@ struct efx_nic {
enum efx_int_mode interrupt_mode;
unsigned int timer_quantum_ns;
+ unsigned int timer_max_ns;
bool irq_rx_adaptive;
unsigned int irq_mod_step_us;
unsigned int irq_rx_moderation_us;
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 994e33f..04ed1b4 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -227,6 +227,9 @@ static int siena_probe_nvconfig(struct efx_nic *efx)
efx->timer_quantum_ns =
(caps & (1 << MC_CMD_CAPABILITIES_TURBO_ACTIVE_LBN)) ?
3072 : 6144; /* 768 cycles */
+ efx->timer_max_ns = efx->type->timer_period_max *
+ efx->timer_quantum_ns;
+
return rc;
}
--
2.7.4
^ permalink raw reply related [flat|nested] 8+ messages in thread
* Re: [PATCH net-next 0/6] sfc: SFN8000 support improvements
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
` (5 preceding siblings ...)
2016-08-11 12:02 ` [PATCH net-next 6/6] sfc: get timer configuration from adapter Bert Kenward
@ 2016-08-13 3:51 ` David Miller
6 siblings, 0 replies; 8+ messages in thread
From: David Miller @ 2016-08-13 3:51 UTC (permalink / raw)
To: bkenward; +Cc: netdev, linux-net-drivers
From: Bert Kenward <bkenward@solarflare.com>
Date: Thu, 11 Aug 2016 12:59:20 +0100
> This series improves support for the recently released SFN8000 series
> of adapters. Specifically, it retrieves interrupt moderation timer
> settings directly from the adapter and uses those settings. It also
> uses a new event queue initialisation interface, allowing specification
> of a performance objective rather than enabling individual flags.
Series applied, thanks.
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2016-08-13 3:51 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <1137d3d8-0b1b-d5f5-149a-1771bbac7aa6@solarflare.com>
2016-08-11 11:59 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 1/6] sfc: update MCDI protocol headers Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 2/6] sfc: allow asynchronous MCDI without completion function Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 3/6] sfc: retrieve second word of datapath capabilities Bert Kenward
2016-08-11 12:01 ` [PATCH net-next 4/6] sfc: use new performance based event queue init Bert Kenward
2016-08-11 12:02 ` [PATCH net-next 5/6] sfc: set interrupt moderation via MCDI Bert Kenward
2016-08-11 12:02 ` [PATCH net-next 6/6] sfc: get timer configuration from adapter Bert Kenward
2016-08-13 3:51 ` [PATCH net-next 0/6] sfc: SFN8000 support improvements David Miller
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.