* [PATCH net-next 09/16] sfc: use MCDI epoch flag to improve MC reboot detection in the driver
From: Ben Hutchings @ 2013-08-28 15:19 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
From: Daniel Pieczko <dpieczko@solarflare.com>
The Huntington MC will reject all MCDI requests after an MC reboot until it sees
one with the NOT_EPOCH flag clear. This flag is set by default for all requests,
and then cleared on the first request after we detect that an MC reboot has
occurred.
The old MCDI_STATUS_DELAY_COUNT gave a timeout of 10ms, which was not long enough
for the driver to detect that a reboot had occurred based on the warm boot count
while calling efx_mcdi_poll_reboot() from the loop in efx_mcdi_ev_death().
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/mcdi.c | 19 +++++++++++++------
drivers/net/ethernet/sfc/mcdi.h | 2 ++
2 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
index 3810634..63121ad 100644
--- a/drivers/net/ethernet/sfc/mcdi.c
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -26,9 +26,10 @@
/* A reboot/assertion causes the MCDI status word to be set after the
* command word is set or a REBOOT event is sent. If we notice a reboot
- * via these mechanisms then wait 10ms for the status word to be set. */
+ * via these mechanisms then wait 20ms for the status word to be set.
+ */
#define MCDI_STATUS_DELAY_US 100
-#define MCDI_STATUS_DELAY_COUNT 100
+#define MCDI_STATUS_DELAY_COUNT 200
#define MCDI_STATUS_SLEEP_MS \
(MCDI_STATUS_DELAY_US * MCDI_STATUS_DELAY_COUNT / 1000)
@@ -56,6 +57,7 @@ int efx_mcdi_init(struct efx_nic *efx)
mcdi->mode = MCDI_MODE_POLL;
(void) efx_mcdi_poll_reboot(efx);
+ mcdi->new_epoch = true;
/* Recover from a failed assertion before probing */
return efx_mcdi_handle_assertion(efx);
@@ -85,24 +87,26 @@ static void efx_mcdi_copyin(struct efx_nic *efx, unsigned cmd,
if (efx->type->mcdi_max_ver == 1) {
/* MCDI v1 */
- EFX_POPULATE_DWORD_6(hdr[0],
+ EFX_POPULATE_DWORD_7(hdr[0],
MCDI_HEADER_RESPONSE, 0,
MCDI_HEADER_RESYNC, 1,
MCDI_HEADER_CODE, cmd,
MCDI_HEADER_DATALEN, inlen,
MCDI_HEADER_SEQ, seqno,
- MCDI_HEADER_XFLAGS, xflags);
+ MCDI_HEADER_XFLAGS, xflags,
+ MCDI_HEADER_NOT_EPOCH, !mcdi->new_epoch);
hdr_len = 4;
} else {
/* MCDI v2 */
BUG_ON(inlen > MCDI_CTL_SDU_LEN_MAX_V2);
- EFX_POPULATE_DWORD_6(hdr[0],
+ EFX_POPULATE_DWORD_7(hdr[0],
MCDI_HEADER_RESPONSE, 0,
MCDI_HEADER_RESYNC, 1,
MCDI_HEADER_CODE, MC_CMD_V2_EXTN,
MCDI_HEADER_DATALEN, 0,
MCDI_HEADER_SEQ, seqno,
- MCDI_HEADER_XFLAGS, xflags);
+ MCDI_HEADER_XFLAGS, xflags,
+ MCDI_HEADER_NOT_EPOCH, !mcdi->new_epoch);
EFX_POPULATE_DWORD_2(hdr[1],
MC_CMD_V2_EXTN_IN_EXTENDED_CMD, cmd,
MC_CMD_V2_EXTN_IN_ACTUAL_LEN, inlen);
@@ -373,6 +377,7 @@ int efx_mcdi_rpc_start(struct efx_nic *efx, unsigned cmd,
spin_unlock_bh(&mcdi->iface_lock);
efx_mcdi_copyin(efx, cmd, inbuf, inlen);
+ mcdi->new_epoch = false;
return 0;
}
@@ -435,6 +440,7 @@ int efx_mcdi_rpc_finish(struct efx_nic *efx, unsigned cmd, size_t inlen,
if (rc == -EIO || rc == -EINTR) {
msleep(MCDI_STATUS_SLEEP_MS);
efx_mcdi_poll_reboot(efx);
+ mcdi->new_epoch = true;
}
}
@@ -530,6 +536,7 @@ static void efx_mcdi_ev_death(struct efx_nic *efx, int rc)
break;
udelay(MCDI_STATUS_DELAY_US);
}
+ mcdi->new_epoch = true;
}
spin_unlock(&mcdi->iface_lock);
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
index 5f67ac3..303d9e8 100644
--- a/drivers/net/ethernet/sfc/mcdi.h
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -36,6 +36,7 @@ enum efx_mcdi_mode {
* @state: Request handling state. Waited for by @wq.
* @mode: Poll for mcdi completion, or wait for an mcdi_event.
* @wq: Wait queue for threads waiting for @state != %MCDI_STATE_RUNNING
+ * @new_epoch: Indicates start of day or start of MC reboot recovery
* @iface_lock: Serialises access to all the following fields
* @seqno: The next sequence number to use for mcdi requests.
* @credits: Number of spurious MCDI completion events allowed before we
@@ -49,6 +50,7 @@ struct efx_mcdi_iface {
enum efx_mcdi_mode mode;
wait_queue_head_t wq;
spinlock_t iface_lock;
+ bool new_epoch;
unsigned int credits;
unsigned int seqno;
int resprc;
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 08/16] sfc: Add EF10 support for TX/RX DMA error events handling.
From: Ben Hutchings @ 2013-08-28 15:19 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
From: Alexandre Rames <arames@solarflare.com>
Also, since we handle all DMA errors in the same way, merge
RESET_TYPE_(RX|TX)_DESC_FETCH into RESET_TYPE_DMA_ERROR.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/efx.c | 3 +--
drivers/net/ethernet/sfc/enum.h | 6 ++----
drivers/net/ethernet/sfc/falcon.c | 3 +--
drivers/net/ethernet/sfc/farch.c | 6 +++---
drivers/net/ethernet/sfc/mcdi.c | 8 ++++++++
5 files changed, 15 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 9fe375f..efad5f7 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -80,8 +80,7 @@ const char *const efx_reset_type_names[] = {
[RESET_TYPE_TX_WATCHDOG] = "TX_WATCHDOG",
[RESET_TYPE_INT_ERROR] = "INT_ERROR",
[RESET_TYPE_RX_RECOVERY] = "RX_RECOVERY",
- [RESET_TYPE_RX_DESC_FETCH] = "RX_DESC_FETCH",
- [RESET_TYPE_TX_DESC_FETCH] = "TX_DESC_FETCH",
+ [RESET_TYPE_DMA_ERROR] = "DMA_ERROR",
[RESET_TYPE_TX_SKIP] = "TX_SKIP",
[RESET_TYPE_MC_FAILURE] = "MC_FAILURE",
};
diff --git a/drivers/net/ethernet/sfc/enum.h b/drivers/net/ethernet/sfc/enum.h
index ab8fb58..8665921 100644
--- a/drivers/net/ethernet/sfc/enum.h
+++ b/drivers/net/ethernet/sfc/enum.h
@@ -147,8 +147,7 @@ enum efx_loopback_mode {
* @RESET_TYPE_TX_WATCHDOG: reset due to TX watchdog
* @RESET_TYPE_INT_ERROR: reset due to internal error
* @RESET_TYPE_RX_RECOVERY: reset to recover from RX datapath errors
- * @RESET_TYPE_RX_DESC_FETCH: pcie error during rx descriptor fetch
- * @RESET_TYPE_TX_DESC_FETCH: pcie error during tx descriptor fetch
+ * @RESET_TYPE_DMA_ERROR: DMA error
* @RESET_TYPE_TX_SKIP: hardware completed empty tx descriptors
* @RESET_TYPE_MC_FAILURE: MC reboot/assertion
*/
@@ -163,8 +162,7 @@ enum reset_type {
RESET_TYPE_TX_WATCHDOG,
RESET_TYPE_INT_ERROR,
RESET_TYPE_RX_RECOVERY,
- RESET_TYPE_RX_DESC_FETCH,
- RESET_TYPE_TX_DESC_FETCH,
+ RESET_TYPE_DMA_ERROR,
RESET_TYPE_TX_SKIP,
RESET_TYPE_MC_FAILURE,
RESET_TYPE_MAX,
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index 0be1f37..1140b83 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -1893,8 +1893,7 @@ static enum reset_type falcon_map_reset_reason(enum reset_type reason)
{
switch (reason) {
case RESET_TYPE_RX_RECOVERY:
- case RESET_TYPE_RX_DESC_FETCH:
- case RESET_TYPE_TX_DESC_FETCH:
+ case RESET_TYPE_DMA_ERROR:
case RESET_TYPE_TX_SKIP:
/* These can occasionally occur due to hardware bugs.
* We try to reset without disrupting the link.
diff --git a/drivers/net/ethernet/sfc/farch.c b/drivers/net/ethernet/sfc/farch.c
index b6af8f4..d21483d 100644
--- a/drivers/net/ethernet/sfc/farch.c
+++ b/drivers/net/ethernet/sfc/farch.c
@@ -832,7 +832,7 @@ efx_farch_handle_tx_event(struct efx_channel *channel, efx_qword_t *event)
efx_farch_notify_tx_desc(tx_queue);
netif_tx_unlock(efx->net_dev);
} else if (EFX_QWORD_FIELD(*event, FSF_AZ_TX_EV_PKT_ERR)) {
- efx_schedule_reset(efx, RESET_TYPE_TX_DESC_FETCH);
+ efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR);
} else {
netif_err(efx, tx_err, efx->net_dev,
"channel %d unexpected TX event "
@@ -1217,7 +1217,7 @@ efx_farch_handle_driver_event(struct efx_channel *channel, efx_qword_t *event)
"RX DMA Q %d reports descriptor fetch error."
" RX Q %d is disabled.\n", ev_sub_data,
ev_sub_data);
- efx_schedule_reset(efx, RESET_TYPE_RX_DESC_FETCH);
+ efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR);
} else
efx_sriov_desc_fetch_err(efx, ev_sub_data);
break;
@@ -1227,7 +1227,7 @@ efx_farch_handle_driver_event(struct efx_channel *channel, efx_qword_t *event)
"TX DMA Q %d reports descriptor fetch error."
" TX Q %d is disabled.\n", ev_sub_data,
ev_sub_data);
- efx_schedule_reset(efx, RESET_TYPE_TX_DESC_FETCH);
+ efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR);
} else
efx_sriov_desc_fetch_err(efx, ev_sub_data);
break;
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
index ff90ebf..3810634 100644
--- a/drivers/net/ethernet/sfc/mcdi.c
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -587,6 +587,14 @@ void efx_mcdi_process_event(struct efx_channel *channel,
efx_ptp_event(efx, event);
break;
+ case MCDI_EVENT_CODE_TX_ERR:
+ case MCDI_EVENT_CODE_RX_ERR:
+ netif_err(efx, hw, efx->net_dev,
+ "%s DMA error (event: "EFX_QWORD_FMT")\n",
+ code == MCDI_EVENT_CODE_TX_ERR ? "TX" : "RX",
+ EFX_QWORD_VAL(*event));
+ efx_schedule_reset(efx, RESET_TYPE_DMA_ERROR);
+ break;
default:
netif_err(efx, hw, efx->net_dev, "Unknown MCDI event 0x%x\n",
code);
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 07/16] sfc: Add a function pointer to abstract write of host time into NIC shared memory
From: Ben Hutchings @ 2013-08-28 15:18 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
From: Laurence Evans <levans@solarflare.com>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/net_driver.h | 1 +
drivers/net/ethernet/sfc/ptp.c | 3 +--
drivers/net/ethernet/sfc/siena.c | 14 ++++++++++++++
3 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index d33656b..21b1c2a 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -1117,6 +1117,7 @@ struct efx_nic_type {
size_t *retlen, const u8 *buffer);
int (*mtd_sync)(struct mtd_info *mtd);
#endif
+ void (*ptp_write_host_time)(struct efx_nic *efx, u32 host_time);
int revision;
unsigned int txd_ptr_tbl_base;
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index 4de2f8f..c60cabb 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -390,8 +390,7 @@ static void efx_ptp_send_times(struct efx_nic *efx,
host_time = (now.ts_real.tv_sec << MC_NANOSECOND_BITS |
now.ts_real.tv_nsec);
/* Update host time in NIC memory */
- _efx_writed(efx, cpu_to_le32(host_time),
- FR_CZ_MC_TREG_SMEM + MC_SMEM_P0_PTP_TIME_OFST);
+ efx->type->ptp_write_host_time(efx, host_time);
}
*last_time = now;
}
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index a51d90c..3c0a544 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -855,6 +855,19 @@ fail:
/**************************************************************************
*
+ * PTP
+ *
+ **************************************************************************
+ */
+
+static void siena_ptp_write_host_time(struct efx_nic *efx, u32 host_time)
+{
+ _efx_writed(efx, cpu_to_le32(host_time),
+ FR_CZ_MC_TREG_SMEM + MC_SMEM_P0_PTP_TIME_OFST);
+}
+
+/**************************************************************************
+ *
* Revision-dependent attributes used by efx.c and nic.c
*
**************************************************************************
@@ -943,6 +956,7 @@ const struct efx_nic_type siena_a0_nic_type = {
.mtd_write = efx_mcdi_mtd_write,
.mtd_sync = efx_mcdi_mtd_sync,
#endif
+ .ptp_write_host_time = siena_ptp_write_host_time,
.revision = EFX_REV_SIENA_A0,
.txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL,
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 06/16] sfc: PTP MCDI requests need to initialise periph ID field
From: Ben Hutchings @ 2013-08-28 15:18 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
From: Laurence Evans <levans@solarflare.com>
This field is ignored by Siena firmware but is significant to EF10 firmware.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/ptp.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index b386901..4de2f8f 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -313,6 +313,7 @@ static int efx_ptp_enable(struct efx_nic *efx)
MCDI_DECLARE_BUF(inbuf, MC_CMD_PTP_IN_ENABLE_LEN);
MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_ENABLE);
+ MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0);
MCDI_SET_DWORD(inbuf, PTP_IN_ENABLE_QUEUE,
efx->ptp_data->channel->channel);
MCDI_SET_DWORD(inbuf, PTP_IN_ENABLE_MODE, efx->ptp_data->mode);
@@ -331,6 +332,7 @@ static int efx_ptp_disable(struct efx_nic *efx)
MCDI_DECLARE_BUF(inbuf, MC_CMD_PTP_IN_DISABLE_LEN);
MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_DISABLE);
+ MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0);
return efx_mcdi_rpc(efx, MC_CMD_PTP, inbuf, sizeof(inbuf),
NULL, 0, NULL);
}
@@ -531,6 +533,7 @@ static int efx_ptp_synchronize(struct efx_nic *efx, unsigned int num_readings)
int *start = ptp->start.addr;
MCDI_SET_DWORD(synch_buf, PTP_IN_OP, MC_CMD_PTP_OP_SYNCHRONIZE);
+ MCDI_SET_DWORD(synch_buf, PTP_IN_PERIPH_ID, 0);
MCDI_SET_DWORD(synch_buf, PTP_IN_SYNCHRONIZE_NUMTIMESETS,
num_readings);
MCDI_SET_QWORD(synch_buf, PTP_IN_SYNCHRONIZE_START_ADDR,
@@ -574,6 +577,7 @@ static int efx_ptp_xmit_skb(struct efx_nic *efx, struct sk_buff *skb)
size_t len;
MCDI_SET_DWORD(ptp_data->txbuf, PTP_IN_OP, MC_CMD_PTP_OP_TRANSMIT);
+ MCDI_SET_DWORD(ptp_data->txbuf, PTP_IN_PERIPH_ID, 0);
MCDI_SET_DWORD(ptp_data->txbuf, PTP_IN_TRANSMIT_LENGTH, skb->len);
if (skb_shinfo(skb)->nr_frags != 0) {
rc = skb_linearize(skb);
@@ -1377,6 +1381,7 @@ static int efx_phc_adjfreq(struct ptp_clock_info *ptp, s32 delta)
(PPB_EXTRA_BITS + MAX_PPB_BITS));
MCDI_SET_DWORD(inadj, PTP_IN_OP, MC_CMD_PTP_OP_ADJUST);
+ MCDI_SET_DWORD(inadj, PTP_IN_PERIPH_ID, 0);
MCDI_SET_QWORD(inadj, PTP_IN_ADJUST_FREQ, adjustment_ns);
MCDI_SET_DWORD(inadj, PTP_IN_ADJUST_SECONDS, 0);
MCDI_SET_DWORD(inadj, PTP_IN_ADJUST_NANOSECONDS, 0);
@@ -1399,6 +1404,7 @@ static int efx_phc_adjtime(struct ptp_clock_info *ptp, s64 delta)
MCDI_DECLARE_BUF(inbuf, MC_CMD_PTP_IN_ADJUST_LEN);
MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_ADJUST);
+ MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0);
MCDI_SET_QWORD(inbuf, PTP_IN_ADJUST_FREQ, 0);
MCDI_SET_DWORD(inbuf, PTP_IN_ADJUST_SECONDS, (u32)delta_ts.tv_sec);
MCDI_SET_DWORD(inbuf, PTP_IN_ADJUST_NANOSECONDS, (u32)delta_ts.tv_nsec);
@@ -1417,6 +1423,7 @@ static int efx_phc_gettime(struct ptp_clock_info *ptp, struct timespec *ts)
int rc;
MCDI_SET_DWORD(inbuf, PTP_IN_OP, MC_CMD_PTP_OP_READ_NIC_TIME);
+ MCDI_SET_DWORD(inbuf, PTP_IN_PERIPH_ID, 0);
rc = efx_mcdi_rpc(efx, MC_CMD_PTP, inbuf, sizeof(inbuf),
outbuf, sizeof(outbuf), NULL);
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 05/16] sfc: Delegate MAC/NIC statistic description to efx_nic_type
From: Ben Hutchings @ 2013-08-28 15:18 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
Various hardware statistics that are available for Siena are
unavailable or meaningless for Falcon. Huntington adds further to the
NIC-type-specific statistics, as it has different MAC blocks from
Falcon/Siena.
All NIC types still provide most statistics by DMA, and use
little-endian byte order.
Therefore:
1. Add some general utility functions for reporting hardware statistics,
efx_nic_describe_stats() and efx_nic_update_stats().
2. Add an efx_nic_type::describe_stats operation to get the number and
names of statistics, implemented using efx_nic_describe_stats()
3. Change efx_nic_type::update_stats to store the core statistics
(struct rtnl_link_stats64) or full statistics (array of u64) in a
caller-provided buffer. Use efx_nic_update_stats() to aid in the
implementation.
4. Rename struct efx_ethtool_stat to struct efx_sw_stat_desc and
EFX_ETHTOOL_NUM_STATS to EFX_ETHTOOL_SW_STAT_COUNT.
5. Remove efx_nic::mac_stats and struct efx_mac_stats.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/efx.c | 27 +----
drivers/net/ethernet/sfc/ethtool.c | 117 +++---------------
drivers/net/ethernet/sfc/falcon.c | 212 ++++++++++++++++++---------------
drivers/net/ethernet/sfc/mcdi.c | 13 +-
drivers/net/ethernet/sfc/net_driver.h | 96 +++------------
drivers/net/ethernet/sfc/nic.c | 83 +++++++++++++
drivers/net/ethernet/sfc/nic.h | 124 ++++++++++++++++++++
drivers/net/ethernet/sfc/siena.c | 215 +++++++++++++++++++++-------------
8 files changed, 498 insertions(+), 389 deletions(-)
diff --git a/drivers/net/ethernet/sfc/efx.c b/drivers/net/ethernet/sfc/efx.c
index 1d4f388..9fe375f 100644
--- a/drivers/net/ethernet/sfc/efx.c
+++ b/drivers/net/ethernet/sfc/efx.c
@@ -1920,34 +1920,9 @@ static struct rtnl_link_stats64 *efx_net_stats(struct net_device *net_dev,
struct rtnl_link_stats64 *stats)
{
struct efx_nic *efx = netdev_priv(net_dev);
- struct efx_mac_stats *mac_stats = &efx->mac_stats;
spin_lock_bh(&efx->stats_lock);
-
- efx->type->update_stats(efx);
-
- stats->rx_packets = mac_stats->rx_packets;
- stats->tx_packets = mac_stats->tx_packets;
- stats->rx_bytes = mac_stats->rx_bytes;
- stats->tx_bytes = mac_stats->tx_bytes;
- stats->rx_dropped = efx->n_rx_nodesc_drop_cnt;
- stats->multicast = mac_stats->rx_multicast;
- stats->collisions = mac_stats->tx_collision;
- stats->rx_length_errors = (mac_stats->rx_gtjumbo +
- mac_stats->rx_length_error);
- stats->rx_crc_errors = mac_stats->rx_bad;
- stats->rx_frame_errors = mac_stats->rx_align_error;
- stats->rx_fifo_errors = mac_stats->rx_overflow;
- stats->rx_missed_errors = mac_stats->rx_missed;
- stats->tx_window_errors = mac_stats->tx_late_collision;
-
- stats->rx_errors = (stats->rx_length_errors +
- stats->rx_crc_errors +
- stats->rx_frame_errors +
- mac_stats->rx_symbol_error);
- stats->tx_errors = (stats->tx_window_errors +
- mac_stats->tx_bad);
-
+ efx->type->update_stats(efx, NULL, stats);
spin_unlock_bh(&efx->stats_lock);
return stats;
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 86b4391..0b8ffdf 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -19,10 +19,9 @@
#include "filter.h"
#include "nic.h"
-struct efx_ethtool_stat {
+struct efx_sw_stat_desc {
const char *name;
enum {
- EFX_ETHTOOL_STAT_SOURCE_mac_stats,
EFX_ETHTOOL_STAT_SOURCE_nic,
EFX_ETHTOOL_STAT_SOURCE_channel,
EFX_ETHTOOL_STAT_SOURCE_tx_queue
@@ -31,7 +30,7 @@ struct efx_ethtool_stat {
u64(*get_stat) (void *field); /* Reader function */
};
-/* Initialiser for a struct #efx_ethtool_stat with type-checking */
+/* Initialiser for a struct efx_sw_stat_desc with type-checking */
#define EFX_ETHTOOL_STAT(stat_name, source_name, field, field_type, \
get_stat_function) { \
.name = #stat_name, \
@@ -48,24 +47,11 @@ static u64 efx_get_uint_stat(void *field)
return *(unsigned int *)field;
}
-static u64 efx_get_u64_stat(void *field)
-{
- return *(u64 *) field;
-}
-
static u64 efx_get_atomic_stat(void *field)
{
return atomic_read((atomic_t *) field);
}
-#define EFX_ETHTOOL_U64_MAC_STAT(field) \
- EFX_ETHTOOL_STAT(field, mac_stats, field, \
- u64, efx_get_u64_stat)
-
-#define EFX_ETHTOOL_UINT_NIC_STAT(name) \
- EFX_ETHTOOL_STAT(name, nic, n_##name, \
- unsigned int, efx_get_uint_stat)
-
#define EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(field) \
EFX_ETHTOOL_STAT(field, nic, field, \
atomic_t, efx_get_atomic_stat)
@@ -78,72 +64,11 @@ static u64 efx_get_atomic_stat(void *field)
EFX_ETHTOOL_STAT(tx_##field, tx_queue, field, \
unsigned int, efx_get_uint_stat)
-static const struct efx_ethtool_stat efx_ethtool_stats[] = {
- EFX_ETHTOOL_U64_MAC_STAT(tx_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(tx_good_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(tx_bad_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(tx_packets),
- EFX_ETHTOOL_U64_MAC_STAT(tx_bad),
- EFX_ETHTOOL_U64_MAC_STAT(tx_pause),
- EFX_ETHTOOL_U64_MAC_STAT(tx_control),
- EFX_ETHTOOL_U64_MAC_STAT(tx_unicast),
- EFX_ETHTOOL_U64_MAC_STAT(tx_multicast),
- EFX_ETHTOOL_U64_MAC_STAT(tx_broadcast),
- EFX_ETHTOOL_U64_MAC_STAT(tx_lt64),
- EFX_ETHTOOL_U64_MAC_STAT(tx_64),
- EFX_ETHTOOL_U64_MAC_STAT(tx_65_to_127),
- EFX_ETHTOOL_U64_MAC_STAT(tx_128_to_255),
- EFX_ETHTOOL_U64_MAC_STAT(tx_256_to_511),
- EFX_ETHTOOL_U64_MAC_STAT(tx_512_to_1023),
- EFX_ETHTOOL_U64_MAC_STAT(tx_1024_to_15xx),
- EFX_ETHTOOL_U64_MAC_STAT(tx_15xx_to_jumbo),
- EFX_ETHTOOL_U64_MAC_STAT(tx_gtjumbo),
- EFX_ETHTOOL_U64_MAC_STAT(tx_collision),
- EFX_ETHTOOL_U64_MAC_STAT(tx_single_collision),
- EFX_ETHTOOL_U64_MAC_STAT(tx_multiple_collision),
- EFX_ETHTOOL_U64_MAC_STAT(tx_excessive_collision),
- EFX_ETHTOOL_U64_MAC_STAT(tx_deferred),
- EFX_ETHTOOL_U64_MAC_STAT(tx_late_collision),
- EFX_ETHTOOL_U64_MAC_STAT(tx_excessive_deferred),
- EFX_ETHTOOL_U64_MAC_STAT(tx_non_tcpudp),
- EFX_ETHTOOL_U64_MAC_STAT(tx_mac_src_error),
- EFX_ETHTOOL_U64_MAC_STAT(tx_ip_src_error),
+static const struct efx_sw_stat_desc efx_sw_stat_desc[] = {
EFX_ETHTOOL_UINT_TXQ_STAT(tso_bursts),
EFX_ETHTOOL_UINT_TXQ_STAT(tso_long_headers),
EFX_ETHTOOL_UINT_TXQ_STAT(tso_packets),
EFX_ETHTOOL_UINT_TXQ_STAT(pushes),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(rx_good_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad_bytes),
- EFX_ETHTOOL_U64_MAC_STAT(rx_packets),
- EFX_ETHTOOL_U64_MAC_STAT(rx_good),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad),
- EFX_ETHTOOL_U64_MAC_STAT(rx_pause),
- EFX_ETHTOOL_U64_MAC_STAT(rx_control),
- EFX_ETHTOOL_U64_MAC_STAT(rx_unicast),
- EFX_ETHTOOL_U64_MAC_STAT(rx_multicast),
- EFX_ETHTOOL_U64_MAC_STAT(rx_broadcast),
- EFX_ETHTOOL_U64_MAC_STAT(rx_lt64),
- EFX_ETHTOOL_U64_MAC_STAT(rx_64),
- EFX_ETHTOOL_U64_MAC_STAT(rx_65_to_127),
- EFX_ETHTOOL_U64_MAC_STAT(rx_128_to_255),
- EFX_ETHTOOL_U64_MAC_STAT(rx_256_to_511),
- EFX_ETHTOOL_U64_MAC_STAT(rx_512_to_1023),
- EFX_ETHTOOL_U64_MAC_STAT(rx_1024_to_15xx),
- EFX_ETHTOOL_U64_MAC_STAT(rx_15xx_to_jumbo),
- EFX_ETHTOOL_U64_MAC_STAT(rx_gtjumbo),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad_lt64),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad_64_to_15xx),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad_15xx_to_jumbo),
- EFX_ETHTOOL_U64_MAC_STAT(rx_bad_gtjumbo),
- EFX_ETHTOOL_U64_MAC_STAT(rx_overflow),
- EFX_ETHTOOL_U64_MAC_STAT(rx_missed),
- EFX_ETHTOOL_U64_MAC_STAT(rx_false_carrier),
- EFX_ETHTOOL_U64_MAC_STAT(rx_symbol_error),
- EFX_ETHTOOL_U64_MAC_STAT(rx_align_error),
- EFX_ETHTOOL_U64_MAC_STAT(rx_length_error),
- EFX_ETHTOOL_U64_MAC_STAT(rx_internal_error),
- EFX_ETHTOOL_UINT_NIC_STAT(rx_nodesc_drop_cnt),
EFX_ETHTOOL_ATOMIC_NIC_ERROR_STAT(rx_reset),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_tobe_disc),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_ip_hdr_chksum_err),
@@ -153,8 +78,7 @@ static const struct efx_ethtool_stat efx_ethtool_stats[] = {
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_nodesc_trunc),
};
-/* Number of ethtool statistics */
-#define EFX_ETHTOOL_NUM_STATS ARRAY_SIZE(efx_ethtool_stats)
+#define EFX_ETHTOOL_SW_STAT_COUNT ARRAY_SIZE(efx_sw_stat_desc)
#define EFX_ETHTOOL_EEPROM_MAGIC 0xEFAB
@@ -424,12 +348,14 @@ static int efx_ethtool_fill_self_tests(struct efx_nic *efx,
static int efx_ethtool_get_sset_count(struct net_device *net_dev,
int string_set)
{
+ struct efx_nic *efx = netdev_priv(net_dev);
+
switch (string_set) {
case ETH_SS_STATS:
- return EFX_ETHTOOL_NUM_STATS;
+ return efx->type->describe_stats(efx, NULL) +
+ EFX_ETHTOOL_SW_STAT_COUNT;
case ETH_SS_TEST:
- return efx_ethtool_fill_self_tests(netdev_priv(net_dev),
- NULL, NULL, NULL);
+ return efx_ethtool_fill_self_tests(efx, NULL, NULL, NULL);
default:
return -EINVAL;
}
@@ -443,9 +369,11 @@ static void efx_ethtool_get_strings(struct net_device *net_dev,
switch (string_set) {
case ETH_SS_STATS:
- for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++)
+ strings += (efx->type->describe_stats(efx, strings) *
+ ETH_GSTRING_LEN);
+ for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++)
strlcpy(strings + i * ETH_GSTRING_LEN,
- efx_ethtool_stats[i].name, ETH_GSTRING_LEN);
+ efx_sw_stat_desc[i].name, ETH_GSTRING_LEN);
break;
case ETH_SS_TEST:
efx_ethtool_fill_self_tests(efx, NULL, strings, NULL);
@@ -461,27 +389,20 @@ static void efx_ethtool_get_stats(struct net_device *net_dev,
u64 *data)
{
struct efx_nic *efx = netdev_priv(net_dev);
- struct efx_mac_stats *mac_stats = &efx->mac_stats;
- const struct efx_ethtool_stat *stat;
+ const struct efx_sw_stat_desc *stat;
struct efx_channel *channel;
struct efx_tx_queue *tx_queue;
int i;
- EFX_BUG_ON_PARANOID(stats->n_stats != EFX_ETHTOOL_NUM_STATS);
-
spin_lock_bh(&efx->stats_lock);
- /* Update MAC and NIC statistics */
- efx->type->update_stats(efx);
+ /* Get NIC statistics */
+ data += efx->type->update_stats(efx, data, NULL);
- /* Fill detailed statistics buffer */
- for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++) {
- stat = &efx_ethtool_stats[i];
+ /* Get software statistics */
+ for (i = 0; i < EFX_ETHTOOL_SW_STAT_COUNT; i++) {
+ stat = &efx_sw_stat_desc[i];
switch (stat->source) {
- case EFX_ETHTOOL_STAT_SOURCE_mac_stats:
- data[i] = stat->get_stat((void *)mac_stats +
- stat->offset);
- break;
case EFX_ETHTOOL_STAT_SOURCE_nic:
data[i] = stat->get_stat((void *)efx + stat->offset);
break;
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index 76a9e0d..0be1f37 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -31,7 +31,7 @@
/**************************************************************************
*
- * MAC stats DMA format
+ * NIC stats
*
**************************************************************************
*/
@@ -134,27 +134,67 @@
#define FALCON_XMAC_STATS_DMA_FLAG(efx) \
(*(u32 *)((efx)->stats_buffer.addr + XgDmaDone_offset))
-#define FALCON_STAT_OFFSET(falcon_stat) EFX_VAL(falcon_stat, offset)
-#define FALCON_STAT_WIDTH(falcon_stat) EFX_VAL(falcon_stat, WIDTH)
-
-/* Retrieve statistic from statistics block */
-#define FALCON_STAT(efx, falcon_stat, efx_stat) do { \
- if (FALCON_STAT_WIDTH(falcon_stat) == 16) \
- (efx)->mac_stats.efx_stat += le16_to_cpu( \
- *((__force __le16 *) \
- (efx->stats_buffer.addr + \
- FALCON_STAT_OFFSET(falcon_stat)))); \
- else if (FALCON_STAT_WIDTH(falcon_stat) == 32) \
- (efx)->mac_stats.efx_stat += le32_to_cpu( \
- *((__force __le32 *) \
- (efx->stats_buffer.addr + \
- FALCON_STAT_OFFSET(falcon_stat)))); \
- else \
- (efx)->mac_stats.efx_stat += le64_to_cpu( \
- *((__force __le64 *) \
- (efx->stats_buffer.addr + \
- FALCON_STAT_OFFSET(falcon_stat)))); \
- } while (0)
+#define FALCON_DMA_STAT(ext_name, hw_name) \
+ [FALCON_STAT_ ## ext_name] = \
+ { #ext_name, \
+ /* 48-bit stats are zero-padded to 64 on DMA */ \
+ hw_name ## _ ## WIDTH == 48 ? 64 : hw_name ## _ ## WIDTH, \
+ hw_name ## _ ## offset }
+#define FALCON_OTHER_STAT(ext_name) \
+ [FALCON_STAT_ ## ext_name] = { #ext_name, 0, 0 }
+
+static const struct efx_hw_stat_desc falcon_stat_desc[FALCON_STAT_COUNT] = {
+ FALCON_DMA_STAT(tx_bytes, XgTxOctets),
+ FALCON_DMA_STAT(tx_packets, XgTxPkts),
+ FALCON_DMA_STAT(tx_pause, XgTxPausePkts),
+ FALCON_DMA_STAT(tx_control, XgTxControlPkts),
+ FALCON_DMA_STAT(tx_unicast, XgTxUnicastPkts),
+ FALCON_DMA_STAT(tx_multicast, XgTxMulticastPkts),
+ FALCON_DMA_STAT(tx_broadcast, XgTxBroadcastPkts),
+ FALCON_DMA_STAT(tx_lt64, XgTxUndersizePkts),
+ FALCON_DMA_STAT(tx_64, XgTxPkts64Octets),
+ FALCON_DMA_STAT(tx_65_to_127, XgTxPkts65to127Octets),
+ FALCON_DMA_STAT(tx_128_to_255, XgTxPkts128to255Octets),
+ FALCON_DMA_STAT(tx_256_to_511, XgTxPkts256to511Octets),
+ FALCON_DMA_STAT(tx_512_to_1023, XgTxPkts512to1023Octets),
+ FALCON_DMA_STAT(tx_1024_to_15xx, XgTxPkts1024to15xxOctets),
+ FALCON_DMA_STAT(tx_15xx_to_jumbo, XgTxPkts1519toMaxOctets),
+ FALCON_DMA_STAT(tx_gtjumbo, XgTxOversizePkts),
+ FALCON_DMA_STAT(tx_non_tcpudp, XgTxNonTcpUdpPkt),
+ FALCON_DMA_STAT(tx_mac_src_error, XgTxMacSrcErrPkt),
+ FALCON_DMA_STAT(tx_ip_src_error, XgTxIpSrcErrPkt),
+ FALCON_DMA_STAT(rx_bytes, XgRxOctets),
+ FALCON_DMA_STAT(rx_good_bytes, XgRxOctetsOK),
+ FALCON_OTHER_STAT(rx_bad_bytes),
+ FALCON_DMA_STAT(rx_packets, XgRxPkts),
+ FALCON_DMA_STAT(rx_good, XgRxPktsOK),
+ FALCON_DMA_STAT(rx_bad, XgRxFCSerrorPkts),
+ FALCON_DMA_STAT(rx_pause, XgRxPausePkts),
+ FALCON_DMA_STAT(rx_control, XgRxControlPkts),
+ FALCON_DMA_STAT(rx_unicast, XgRxUnicastPkts),
+ FALCON_DMA_STAT(rx_multicast, XgRxMulticastPkts),
+ FALCON_DMA_STAT(rx_broadcast, XgRxBroadcastPkts),
+ FALCON_DMA_STAT(rx_lt64, XgRxUndersizePkts),
+ FALCON_DMA_STAT(rx_64, XgRxPkts64Octets),
+ FALCON_DMA_STAT(rx_65_to_127, XgRxPkts65to127Octets),
+ FALCON_DMA_STAT(rx_128_to_255, XgRxPkts128to255Octets),
+ FALCON_DMA_STAT(rx_256_to_511, XgRxPkts256to511Octets),
+ FALCON_DMA_STAT(rx_512_to_1023, XgRxPkts512to1023Octets),
+ FALCON_DMA_STAT(rx_1024_to_15xx, XgRxPkts1024to15xxOctets),
+ FALCON_DMA_STAT(rx_15xx_to_jumbo, XgRxPkts15xxtoMaxOctets),
+ FALCON_DMA_STAT(rx_gtjumbo, XgRxOversizePkts),
+ FALCON_DMA_STAT(rx_bad_lt64, XgRxUndersizeFCSerrorPkts),
+ FALCON_DMA_STAT(rx_bad_gtjumbo, XgRxJabberPkts),
+ FALCON_DMA_STAT(rx_overflow, XgRxDropEvents),
+ FALCON_DMA_STAT(rx_symbol_error, XgRxSymbolError),
+ FALCON_DMA_STAT(rx_align_error, XgRxAlignError),
+ FALCON_DMA_STAT(rx_length_error, XgRxLengthError),
+ FALCON_DMA_STAT(rx_internal_error, XgRxInternalMACError),
+ FALCON_OTHER_STAT(rx_nodesc_drop_cnt),
+};
+static const unsigned long falcon_stat_mask[] = {
+ [0 ... BITS_TO_LONGS(FALCON_STAT_COUNT) - 1] = ~0UL,
+};
/**************************************************************************
*
@@ -1159,66 +1199,6 @@ static int falcon_reconfigure_xmac(struct efx_nic *efx)
return 0;
}
-static void falcon_update_stats_xmac(struct efx_nic *efx)
-{
- struct efx_mac_stats *mac_stats = &efx->mac_stats;
-
- /* Update MAC stats from DMAed values */
- FALCON_STAT(efx, XgRxOctets, rx_bytes);
- FALCON_STAT(efx, XgRxOctetsOK, rx_good_bytes);
- FALCON_STAT(efx, XgRxPkts, rx_packets);
- FALCON_STAT(efx, XgRxPktsOK, rx_good);
- FALCON_STAT(efx, XgRxBroadcastPkts, rx_broadcast);
- FALCON_STAT(efx, XgRxMulticastPkts, rx_multicast);
- FALCON_STAT(efx, XgRxUnicastPkts, rx_unicast);
- FALCON_STAT(efx, XgRxUndersizePkts, rx_lt64);
- FALCON_STAT(efx, XgRxOversizePkts, rx_gtjumbo);
- FALCON_STAT(efx, XgRxJabberPkts, rx_bad_gtjumbo);
- FALCON_STAT(efx, XgRxUndersizeFCSerrorPkts, rx_bad_lt64);
- FALCON_STAT(efx, XgRxDropEvents, rx_overflow);
- FALCON_STAT(efx, XgRxFCSerrorPkts, rx_bad);
- FALCON_STAT(efx, XgRxAlignError, rx_align_error);
- FALCON_STAT(efx, XgRxSymbolError, rx_symbol_error);
- FALCON_STAT(efx, XgRxInternalMACError, rx_internal_error);
- FALCON_STAT(efx, XgRxControlPkts, rx_control);
- FALCON_STAT(efx, XgRxPausePkts, rx_pause);
- FALCON_STAT(efx, XgRxPkts64Octets, rx_64);
- FALCON_STAT(efx, XgRxPkts65to127Octets, rx_65_to_127);
- FALCON_STAT(efx, XgRxPkts128to255Octets, rx_128_to_255);
- FALCON_STAT(efx, XgRxPkts256to511Octets, rx_256_to_511);
- FALCON_STAT(efx, XgRxPkts512to1023Octets, rx_512_to_1023);
- FALCON_STAT(efx, XgRxPkts1024to15xxOctets, rx_1024_to_15xx);
- FALCON_STAT(efx, XgRxPkts15xxtoMaxOctets, rx_15xx_to_jumbo);
- FALCON_STAT(efx, XgRxLengthError, rx_length_error);
- FALCON_STAT(efx, XgTxPkts, tx_packets);
- FALCON_STAT(efx, XgTxOctets, tx_bytes);
- FALCON_STAT(efx, XgTxMulticastPkts, tx_multicast);
- FALCON_STAT(efx, XgTxBroadcastPkts, tx_broadcast);
- FALCON_STAT(efx, XgTxUnicastPkts, tx_unicast);
- FALCON_STAT(efx, XgTxControlPkts, tx_control);
- FALCON_STAT(efx, XgTxPausePkts, tx_pause);
- FALCON_STAT(efx, XgTxPkts64Octets, tx_64);
- FALCON_STAT(efx, XgTxPkts65to127Octets, tx_65_to_127);
- FALCON_STAT(efx, XgTxPkts128to255Octets, tx_128_to_255);
- FALCON_STAT(efx, XgTxPkts256to511Octets, tx_256_to_511);
- FALCON_STAT(efx, XgTxPkts512to1023Octets, tx_512_to_1023);
- FALCON_STAT(efx, XgTxPkts1024to15xxOctets, tx_1024_to_15xx);
- FALCON_STAT(efx, XgTxPkts1519toMaxOctets, tx_15xx_to_jumbo);
- FALCON_STAT(efx, XgTxUndersizePkts, tx_lt64);
- FALCON_STAT(efx, XgTxOversizePkts, tx_gtjumbo);
- FALCON_STAT(efx, XgTxNonTcpUdpPkt, tx_non_tcpudp);
- FALCON_STAT(efx, XgTxMacSrcErrPkt, tx_mac_src_error);
- FALCON_STAT(efx, XgTxIpSrcErrPkt, tx_ip_src_error);
-
- /* Update derived statistics */
- efx_update_diff_stat(&mac_stats->tx_good_bytes,
- mac_stats->tx_bytes - mac_stats->tx_bad_bytes -
- mac_stats->tx_control * 64);
- efx_update_diff_stat(&mac_stats->rx_bad_bytes,
- mac_stats->rx_bytes - mac_stats->rx_good_bytes -
- mac_stats->rx_control * 64);
-}
-
static void falcon_poll_xmac(struct efx_nic *efx)
{
struct falcon_nic_data *nic_data = efx->nic_data;
@@ -1422,7 +1402,9 @@ static void falcon_stats_complete(struct efx_nic *efx)
nic_data->stats_pending = false;
if (FALCON_XMAC_STATS_DMA_FLAG(efx)) {
rmb(); /* read the done flag before the stats */
- falcon_update_stats_xmac(efx);
+ efx_nic_update_stats(falcon_stat_desc, FALCON_STAT_COUNT,
+ falcon_stat_mask, nic_data->stats,
+ efx->stats_buffer.addr, true);
} else {
netif_err(efx, hw, efx->net_dev,
"timed out waiting for statistics\n");
@@ -2538,23 +2520,65 @@ static void falcon_remove_nic(struct efx_nic *efx)
efx->nic_data = NULL;
}
-static void falcon_update_nic_stats(struct efx_nic *efx)
+static size_t falcon_describe_nic_stats(struct efx_nic *efx, u8 *names)
+{
+ return efx_nic_describe_stats(falcon_stat_desc, FALCON_STAT_COUNT,
+ falcon_stat_mask, names);
+}
+
+static size_t falcon_update_nic_stats(struct efx_nic *efx, u64 *full_stats,
+ struct rtnl_link_stats64 *core_stats)
{
struct falcon_nic_data *nic_data = efx->nic_data;
+ u64 *stats = nic_data->stats;
efx_oword_t cnt;
- if (nic_data->stats_disable_count)
- return;
+ if (!nic_data->stats_disable_count) {
+ efx_reado(efx, &cnt, FR_AZ_RX_NODESC_DROP);
+ stats[FALCON_STAT_rx_nodesc_drop_cnt] +=
+ EFX_OWORD_FIELD(cnt, FRF_AB_RX_NODESC_DROP_CNT);
+
+ if (nic_data->stats_pending &&
+ FALCON_XMAC_STATS_DMA_FLAG(efx)) {
+ nic_data->stats_pending = false;
+ rmb(); /* read the done flag before the stats */
+ efx_nic_update_stats(
+ falcon_stat_desc, FALCON_STAT_COUNT,
+ falcon_stat_mask,
+ stats, efx->stats_buffer.addr, true);
+ }
- efx_reado(efx, &cnt, FR_AZ_RX_NODESC_DROP);
- efx->n_rx_nodesc_drop_cnt +=
- EFX_OWORD_FIELD(cnt, FRF_AB_RX_NODESC_DROP_CNT);
+ /* Update derived statistic */
+ efx_update_diff_stat(&stats[FALCON_STAT_rx_bad_bytes],
+ stats[FALCON_STAT_rx_bytes] -
+ stats[FALCON_STAT_rx_good_bytes] -
+ stats[FALCON_STAT_rx_control] * 64);
+ }
- if (nic_data->stats_pending && FALCON_XMAC_STATS_DMA_FLAG(efx)) {
- nic_data->stats_pending = false;
- rmb(); /* read the done flag before the stats */
- falcon_update_stats_xmac(efx);
+ if (full_stats)
+ memcpy(full_stats, stats, sizeof(u64) * FALCON_STAT_COUNT);
+
+ if (core_stats) {
+ core_stats->rx_packets = stats[FALCON_STAT_rx_packets];
+ core_stats->tx_packets = stats[FALCON_STAT_tx_packets];
+ core_stats->rx_bytes = stats[FALCON_STAT_rx_bytes];
+ core_stats->tx_bytes = stats[FALCON_STAT_tx_bytes];
+ core_stats->rx_dropped = stats[FALCON_STAT_rx_nodesc_drop_cnt];
+ core_stats->multicast = stats[FALCON_STAT_rx_multicast];
+ core_stats->rx_length_errors =
+ stats[FALCON_STAT_rx_gtjumbo] +
+ stats[FALCON_STAT_rx_length_error];
+ core_stats->rx_crc_errors = stats[FALCON_STAT_rx_bad];
+ core_stats->rx_frame_errors = stats[FALCON_STAT_rx_align_error];
+ core_stats->rx_fifo_errors = stats[FALCON_STAT_rx_overflow];
+
+ core_stats->rx_errors = (core_stats->rx_length_errors +
+ core_stats->rx_crc_errors +
+ core_stats->rx_frame_errors +
+ stats[FALCON_STAT_rx_symbol_error]);
}
+
+ return FALCON_STAT_COUNT;
}
void falcon_start_nic_stats(struct efx_nic *efx)
@@ -2643,6 +2667,7 @@ const struct efx_nic_type falcon_a1_nic_type = {
.fini_dmaq = efx_farch_fini_dmaq,
.prepare_flush = falcon_prepare_flush,
.finish_flush = efx_port_dummy_op_void,
+ .describe_stats = falcon_describe_nic_stats,
.update_stats = falcon_update_nic_stats,
.start_stats = falcon_start_nic_stats,
.stop_stats = falcon_stop_nic_stats,
@@ -2735,6 +2760,7 @@ const struct efx_nic_type falcon_b0_nic_type = {
.fini_dmaq = efx_farch_fini_dmaq,
.prepare_flush = falcon_prepare_flush,
.finish_flush = efx_port_dummy_op_void,
+ .describe_stats = falcon_describe_nic_stats,
.update_stats = falcon_update_nic_stats,
.start_stats = falcon_start_nic_stats,
.stop_stats = falcon_stop_nic_stats,
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
index 67b07de..ff90ebf 100644
--- a/drivers/net/ethernet/sfc/mcdi.c
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -236,21 +236,10 @@ static int efx_mcdi_poll(struct efx_nic *efx)
*/
int efx_mcdi_poll_reboot(struct efx_nic *efx)
{
- int rc;
-
if (!efx->mcdi)
return 0;
- rc = efx->type->mcdi_poll_reboot(efx);
- if (!rc)
- return 0;
-
- /* MAC statistics have been cleared on the NIC; clear our copy
- * so that efx_update_diff_stat() can continue to work.
- */
- memset(&efx->mac_stats, 0, sizeof(efx->mac_stats));
-
- return rc;
+ return efx->type->mcdi_poll_reboot(efx);
}
static void efx_mcdi_acquire(struct efx_mcdi_iface *mcdi)
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index fa881cd..d33656b 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -595,75 +595,17 @@ static inline bool efx_phy_mode_disabled(enum efx_phy_mode mode)
return !!(mode & ~PHY_MODE_TX_DISABLED);
}
-/*
- * Efx extended statistics
- *
- * Not all statistics are provided by all supported MACs. The purpose
- * is this structure is to contain the raw statistics provided by each
- * MAC.
+/**
+ * struct efx_hw_stat_desc - Description of a hardware statistic
+ * @name: Name of the statistic as visible through ethtool, or %NULL if
+ * it should not be exposed
+ * @dma_width: Width in bits (0 for non-DMA statistics)
+ * @offset: Offset within stats (ignored for non-DMA statistics)
*/
-struct efx_mac_stats {
- u64 tx_bytes;
- u64 tx_good_bytes;
- u64 tx_bad_bytes;
- u64 tx_packets;
- u64 tx_bad;
- u64 tx_pause;
- u64 tx_control;
- u64 tx_unicast;
- u64 tx_multicast;
- u64 tx_broadcast;
- u64 tx_lt64;
- u64 tx_64;
- u64 tx_65_to_127;
- u64 tx_128_to_255;
- u64 tx_256_to_511;
- u64 tx_512_to_1023;
- u64 tx_1024_to_15xx;
- u64 tx_15xx_to_jumbo;
- u64 tx_gtjumbo;
- u64 tx_collision;
- u64 tx_single_collision;
- u64 tx_multiple_collision;
- u64 tx_excessive_collision;
- u64 tx_deferred;
- u64 tx_late_collision;
- u64 tx_excessive_deferred;
- u64 tx_non_tcpudp;
- u64 tx_mac_src_error;
- u64 tx_ip_src_error;
- u64 rx_bytes;
- u64 rx_good_bytes;
- u64 rx_bad_bytes;
- u64 rx_packets;
- u64 rx_good;
- u64 rx_bad;
- u64 rx_pause;
- u64 rx_control;
- u64 rx_unicast;
- u64 rx_multicast;
- u64 rx_broadcast;
- u64 rx_lt64;
- u64 rx_64;
- u64 rx_65_to_127;
- u64 rx_128_to_255;
- u64 rx_256_to_511;
- u64 rx_512_to_1023;
- u64 rx_1024_to_15xx;
- u64 rx_15xx_to_jumbo;
- u64 rx_gtjumbo;
- u64 rx_bad_lt64;
- u64 rx_bad_64_to_15xx;
- u64 rx_bad_15xx_to_jumbo;
- u64 rx_bad_gtjumbo;
- u64 rx_overflow;
- u64 rx_missed;
- u64 rx_false_carrier;
- u64 rx_symbol_error;
- u64 rx_align_error;
- u64 rx_length_error;
- u64 rx_internal_error;
- u64 rx_good_lt64;
+struct efx_hw_stat_desc {
+ const char *name;
+ u16 dma_width;
+ u16 offset;
};
/* Number of bits used in a multicast filter hash address */
@@ -795,12 +737,8 @@ struct vfdi_status;
* @last_irq_cpu: Last CPU to handle a possible test interrupt. This
* field is used by efx_test_interrupts() to verify that an
* interrupt has occurred.
- * @n_rx_nodesc_drop_cnt: RX no descriptor drop count
- * @mac_stats: MAC statistics. These include all statistics the MACs
- * can provide. Generic code converts these into a standard
- * &struct net_device_stats.
- * @stats_lock: Statistics update lock. Serialises statistics fetches
- * and access to @mac_stats.
+ * @stats_lock: Statistics update lock. Must be held when calling
+ * efx_nic_type::{update,start,stop}_stats.
*
* This is stored in the private area of the &struct net_device.
*/
@@ -939,8 +877,6 @@ struct efx_nic {
struct delayed_work monitor_work ____cacheline_aligned_in_smp;
spinlock_t biu_lock;
int last_irq_cpu;
- unsigned n_rx_nodesc_drop_cnt;
- struct efx_mac_stats mac_stats;
spinlock_t stats_lock;
};
@@ -984,7 +920,9 @@ struct efx_mtd_partition {
* (for Falcon architecture)
* @finish_flush: Clean up after flushing the DMA queues (for Falcon
* architecture)
- * @update_stats: Update statistics not provided by event handling
+ * @describe_stats: Describe statistics for ethtool
+ * @update_stats: Update statistics not provided by event handling.
+ * Either argument may be %NULL.
* @start_stats: Start the regular fetching of statistics
* @stop_stats: Stop the regular fetching of statistics
* @set_id_led: Set state of identifying LED or revert to automatic function
@@ -1098,7 +1036,9 @@ struct efx_nic_type {
int (*fini_dmaq)(struct efx_nic *efx);
void (*prepare_flush)(struct efx_nic *efx);
void (*finish_flush)(struct efx_nic *efx);
- void (*update_stats)(struct efx_nic *efx);
+ size_t (*describe_stats)(struct efx_nic *efx, u8 *names);
+ size_t (*update_stats)(struct efx_nic *efx, u64 *full_stats,
+ struct rtnl_link_stats64 *core_stats);
void (*start_stats)(struct efx_nic *efx);
void (*stop_stats)(struct efx_nic *efx);
void (*set_id_led)(struct efx_nic *efx, enum efx_led_mode mode);
diff --git a/drivers/net/ethernet/sfc/nic.c b/drivers/net/ethernet/sfc/nic.c
index 66c71ed..b9b1277 100644
--- a/drivers/net/ethernet/sfc/nic.c
+++ b/drivers/net/ethernet/sfc/nic.c
@@ -429,3 +429,86 @@ void efx_nic_get_regs(struct efx_nic *efx, void *buf)
}
}
}
+
+/**
+ * efx_nic_describe_stats - Describe supported statistics for ethtool
+ * @desc: Array of &struct efx_hw_stat_desc describing the statistics
+ * @count: Length of the @desc array
+ * @mask: Bitmask of which elements of @desc are enabled
+ * @names: Buffer to copy names to, or %NULL. The names are copied
+ * starting at intervals of %ETH_GSTRING_LEN bytes.
+ *
+ * Returns the number of visible statistics, i.e. the number of set
+ * bits in the first @count bits of @mask for which a name is defined.
+ */
+size_t efx_nic_describe_stats(const struct efx_hw_stat_desc *desc, size_t count,
+ const unsigned long *mask, u8 *names)
+{
+ size_t visible = 0;
+ size_t index;
+
+ for_each_set_bit(index, mask, count) {
+ if (desc[index].name) {
+ if (names) {
+ strlcpy(names, desc[index].name,
+ ETH_GSTRING_LEN);
+ names += ETH_GSTRING_LEN;
+ }
+ ++visible;
+ }
+ }
+
+ return visible;
+}
+
+/**
+ * efx_nic_update_stats - Convert statistics DMA buffer to array of u64
+ * @desc: Array of &struct efx_hw_stat_desc describing the DMA buffer
+ * layout. DMA widths of 0, 16, 32 and 64 are supported; where
+ * the width is specified as 0 the corresponding element of
+ * @stats is not updated.
+ * @count: Length of the @desc array
+ * @mask: Bitmask of which elements of @desc are enabled
+ * @stats: Buffer to update with the converted statistics. The length
+ * of this array must be at least the number of set bits in the
+ * first @count bits of @mask.
+ * @dma_buf: DMA buffer containing hardware statistics
+ * @accumulate: If set, the converted values will be added rather than
+ * directly stored to the corresponding elements of @stats
+ */
+void efx_nic_update_stats(const struct efx_hw_stat_desc *desc, size_t count,
+ const unsigned long *mask,
+ u64 *stats, const void *dma_buf, bool accumulate)
+{
+ size_t index;
+
+ for_each_set_bit(index, mask, count) {
+ if (desc[index].dma_width) {
+ const void *addr = dma_buf + desc[index].offset;
+ u64 val;
+
+ switch (desc[index].dma_width) {
+ case 16:
+ val = le16_to_cpup((__le16 *)addr);
+ break;
+ case 32:
+ val = le32_to_cpup((__le32 *)addr);
+ break;
+ case 64:
+ val = le64_to_cpup((__le64 *)addr);
+ break;
+ default:
+ WARN_ON(1);
+ val = 0;
+ break;
+ }
+
+ if (accumulate)
+ *stats += val;
+ else
+ *stats = val;
+ }
+
+ ++stats;
+ }
+}
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index c3e0f1f..9afbf36 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -191,10 +191,62 @@ static inline bool falcon_spi_present(const struct falcon_spi_device *spi)
return spi->size != 0;
}
+enum {
+ FALCON_STAT_tx_bytes,
+ FALCON_STAT_tx_packets,
+ FALCON_STAT_tx_pause,
+ FALCON_STAT_tx_control,
+ FALCON_STAT_tx_unicast,
+ FALCON_STAT_tx_multicast,
+ FALCON_STAT_tx_broadcast,
+ FALCON_STAT_tx_lt64,
+ FALCON_STAT_tx_64,
+ FALCON_STAT_tx_65_to_127,
+ FALCON_STAT_tx_128_to_255,
+ FALCON_STAT_tx_256_to_511,
+ FALCON_STAT_tx_512_to_1023,
+ FALCON_STAT_tx_1024_to_15xx,
+ FALCON_STAT_tx_15xx_to_jumbo,
+ FALCON_STAT_tx_gtjumbo,
+ FALCON_STAT_tx_non_tcpudp,
+ FALCON_STAT_tx_mac_src_error,
+ FALCON_STAT_tx_ip_src_error,
+ FALCON_STAT_rx_bytes,
+ FALCON_STAT_rx_good_bytes,
+ FALCON_STAT_rx_bad_bytes,
+ FALCON_STAT_rx_packets,
+ FALCON_STAT_rx_good,
+ FALCON_STAT_rx_bad,
+ FALCON_STAT_rx_pause,
+ FALCON_STAT_rx_control,
+ FALCON_STAT_rx_unicast,
+ FALCON_STAT_rx_multicast,
+ FALCON_STAT_rx_broadcast,
+ FALCON_STAT_rx_lt64,
+ FALCON_STAT_rx_64,
+ FALCON_STAT_rx_65_to_127,
+ FALCON_STAT_rx_128_to_255,
+ FALCON_STAT_rx_256_to_511,
+ FALCON_STAT_rx_512_to_1023,
+ FALCON_STAT_rx_1024_to_15xx,
+ FALCON_STAT_rx_15xx_to_jumbo,
+ FALCON_STAT_rx_gtjumbo,
+ FALCON_STAT_rx_bad_lt64,
+ FALCON_STAT_rx_bad_gtjumbo,
+ FALCON_STAT_rx_overflow,
+ FALCON_STAT_rx_symbol_error,
+ FALCON_STAT_rx_align_error,
+ FALCON_STAT_rx_length_error,
+ FALCON_STAT_rx_internal_error,
+ FALCON_STAT_rx_nodesc_drop_cnt,
+ FALCON_STAT_COUNT
+};
+
/**
* struct falcon_nic_data - Falcon NIC state
* @pci_dev2: Secondary function of Falcon A
* @board: Board state and functions
+ * @stats: Hardware statistics
* @stats_disable_count: Nest count for disabling statistics fetches
* @stats_pending: Is there a pending DMA of MAC statistics.
* @stats_timer: A timer for regularly fetching MAC statistics.
@@ -207,6 +259,7 @@ static inline bool falcon_spi_present(const struct falcon_spi_device *spi)
struct falcon_nic_data {
struct pci_dev *pci_dev2;
struct falcon_board board;
+ u64 stats[FALCON_STAT_COUNT];
unsigned int stats_disable_count;
bool stats_pending;
struct timer_list stats_timer;
@@ -223,12 +276,75 @@ static inline struct falcon_board *falcon_board(struct efx_nic *efx)
return &data->board;
}
+enum {
+ SIENA_STAT_tx_bytes,
+ SIENA_STAT_tx_good_bytes,
+ SIENA_STAT_tx_bad_bytes,
+ SIENA_STAT_tx_packets,
+ SIENA_STAT_tx_bad,
+ SIENA_STAT_tx_pause,
+ SIENA_STAT_tx_control,
+ SIENA_STAT_tx_unicast,
+ SIENA_STAT_tx_multicast,
+ SIENA_STAT_tx_broadcast,
+ SIENA_STAT_tx_lt64,
+ SIENA_STAT_tx_64,
+ SIENA_STAT_tx_65_to_127,
+ SIENA_STAT_tx_128_to_255,
+ SIENA_STAT_tx_256_to_511,
+ SIENA_STAT_tx_512_to_1023,
+ SIENA_STAT_tx_1024_to_15xx,
+ SIENA_STAT_tx_15xx_to_jumbo,
+ SIENA_STAT_tx_gtjumbo,
+ SIENA_STAT_tx_collision,
+ SIENA_STAT_tx_single_collision,
+ SIENA_STAT_tx_multiple_collision,
+ SIENA_STAT_tx_excessive_collision,
+ SIENA_STAT_tx_deferred,
+ SIENA_STAT_tx_late_collision,
+ SIENA_STAT_tx_excessive_deferred,
+ SIENA_STAT_tx_non_tcpudp,
+ SIENA_STAT_tx_mac_src_error,
+ SIENA_STAT_tx_ip_src_error,
+ SIENA_STAT_rx_bytes,
+ SIENA_STAT_rx_good_bytes,
+ SIENA_STAT_rx_bad_bytes,
+ SIENA_STAT_rx_packets,
+ SIENA_STAT_rx_good,
+ SIENA_STAT_rx_bad,
+ SIENA_STAT_rx_pause,
+ SIENA_STAT_rx_control,
+ SIENA_STAT_rx_unicast,
+ SIENA_STAT_rx_multicast,
+ SIENA_STAT_rx_broadcast,
+ SIENA_STAT_rx_lt64,
+ SIENA_STAT_rx_64,
+ SIENA_STAT_rx_65_to_127,
+ SIENA_STAT_rx_128_to_255,
+ SIENA_STAT_rx_256_to_511,
+ SIENA_STAT_rx_512_to_1023,
+ SIENA_STAT_rx_1024_to_15xx,
+ SIENA_STAT_rx_15xx_to_jumbo,
+ SIENA_STAT_rx_gtjumbo,
+ SIENA_STAT_rx_bad_gtjumbo,
+ SIENA_STAT_rx_overflow,
+ SIENA_STAT_rx_false_carrier,
+ SIENA_STAT_rx_symbol_error,
+ SIENA_STAT_rx_align_error,
+ SIENA_STAT_rx_length_error,
+ SIENA_STAT_rx_internal_error,
+ SIENA_STAT_rx_nodesc_drop_cnt,
+ SIENA_STAT_COUNT
+};
+
/**
* struct siena_nic_data - Siena NIC state
* @wol_filter_id: Wake-on-LAN packet filter id
+ * @stats: Hardware statistics
*/
struct siena_nic_data {
int wol_filter_id;
+ u64 stats[SIENA_STAT_COUNT];
};
/*
@@ -533,6 +649,14 @@ extern int efx_farch_test_registers(struct efx_nic *efx,
extern size_t efx_nic_get_regs_len(struct efx_nic *efx);
extern void efx_nic_get_regs(struct efx_nic *efx, void *buf);
+extern size_t
+efx_nic_describe_stats(const struct efx_hw_stat_desc *desc, size_t count,
+ const unsigned long *mask, u8 *names);
+extern void
+efx_nic_update_stats(const struct efx_hw_stat_desc *desc, size_t count,
+ const unsigned long *mask,
+ u64 *stats, const void *dma_buf, bool accumulate);
+
#define EFX_MAX_FLUSH_TIME 5000
extern void efx_farch_generate_event(struct efx_nic *efx, unsigned int evq,
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 11b4739..a51d90c 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -380,117 +380,160 @@ static void siena_remove_nic(struct efx_nic *efx)
efx_mcdi_fini(efx);
}
+#define SIENA_DMA_STAT(ext_name, mcdi_name) \
+ [SIENA_STAT_ ## ext_name] = \
+ { #ext_name, 64, 8 * MC_CMD_MAC_ ## mcdi_name }
+#define SIENA_OTHER_STAT(ext_name) \
+ [SIENA_STAT_ ## ext_name] = { #ext_name, 0, 0 }
+
+static const struct efx_hw_stat_desc siena_stat_desc[SIENA_STAT_COUNT] = {
+ SIENA_DMA_STAT(tx_bytes, TX_BYTES),
+ SIENA_OTHER_STAT(tx_good_bytes),
+ SIENA_DMA_STAT(tx_bad_bytes, TX_BAD_BYTES),
+ SIENA_DMA_STAT(tx_packets, TX_PKTS),
+ SIENA_DMA_STAT(tx_bad, TX_BAD_FCS_PKTS),
+ SIENA_DMA_STAT(tx_pause, TX_PAUSE_PKTS),
+ SIENA_DMA_STAT(tx_control, TX_CONTROL_PKTS),
+ SIENA_DMA_STAT(tx_unicast, TX_UNICAST_PKTS),
+ SIENA_DMA_STAT(tx_multicast, TX_MULTICAST_PKTS),
+ SIENA_DMA_STAT(tx_broadcast, TX_BROADCAST_PKTS),
+ SIENA_DMA_STAT(tx_lt64, TX_LT64_PKTS),
+ SIENA_DMA_STAT(tx_64, TX_64_PKTS),
+ SIENA_DMA_STAT(tx_65_to_127, TX_65_TO_127_PKTS),
+ SIENA_DMA_STAT(tx_128_to_255, TX_128_TO_255_PKTS),
+ SIENA_DMA_STAT(tx_256_to_511, TX_256_TO_511_PKTS),
+ SIENA_DMA_STAT(tx_512_to_1023, TX_512_TO_1023_PKTS),
+ SIENA_DMA_STAT(tx_1024_to_15xx, TX_1024_TO_15XX_PKTS),
+ SIENA_DMA_STAT(tx_15xx_to_jumbo, TX_15XX_TO_JUMBO_PKTS),
+ SIENA_DMA_STAT(tx_gtjumbo, TX_GTJUMBO_PKTS),
+ SIENA_OTHER_STAT(tx_collision),
+ SIENA_DMA_STAT(tx_single_collision, TX_SINGLE_COLLISION_PKTS),
+ SIENA_DMA_STAT(tx_multiple_collision, TX_MULTIPLE_COLLISION_PKTS),
+ SIENA_DMA_STAT(tx_excessive_collision, TX_EXCESSIVE_COLLISION_PKTS),
+ SIENA_DMA_STAT(tx_deferred, TX_DEFERRED_PKTS),
+ SIENA_DMA_STAT(tx_late_collision, TX_LATE_COLLISION_PKTS),
+ SIENA_DMA_STAT(tx_excessive_deferred, TX_EXCESSIVE_DEFERRED_PKTS),
+ SIENA_DMA_STAT(tx_non_tcpudp, TX_NON_TCPUDP_PKTS),
+ SIENA_DMA_STAT(tx_mac_src_error, TX_MAC_SRC_ERR_PKTS),
+ SIENA_DMA_STAT(tx_ip_src_error, TX_IP_SRC_ERR_PKTS),
+ SIENA_DMA_STAT(rx_bytes, RX_BYTES),
+ SIENA_OTHER_STAT(rx_good_bytes),
+ SIENA_DMA_STAT(rx_bad_bytes, RX_BAD_BYTES),
+ SIENA_DMA_STAT(rx_packets, RX_PKTS),
+ SIENA_DMA_STAT(rx_good, RX_GOOD_PKTS),
+ SIENA_DMA_STAT(rx_bad, RX_BAD_FCS_PKTS),
+ SIENA_DMA_STAT(rx_pause, RX_PAUSE_PKTS),
+ SIENA_DMA_STAT(rx_control, RX_CONTROL_PKTS),
+ SIENA_DMA_STAT(rx_unicast, RX_UNICAST_PKTS),
+ SIENA_DMA_STAT(rx_multicast, RX_MULTICAST_PKTS),
+ SIENA_DMA_STAT(rx_broadcast, RX_BROADCAST_PKTS),
+ SIENA_DMA_STAT(rx_lt64, RX_UNDERSIZE_PKTS),
+ SIENA_DMA_STAT(rx_64, RX_64_PKTS),
+ SIENA_DMA_STAT(rx_65_to_127, RX_65_TO_127_PKTS),
+ SIENA_DMA_STAT(rx_128_to_255, RX_128_TO_255_PKTS),
+ SIENA_DMA_STAT(rx_256_to_511, RX_256_TO_511_PKTS),
+ SIENA_DMA_STAT(rx_512_to_1023, RX_512_TO_1023_PKTS),
+ SIENA_DMA_STAT(rx_1024_to_15xx, RX_1024_TO_15XX_PKTS),
+ SIENA_DMA_STAT(rx_15xx_to_jumbo, RX_15XX_TO_JUMBO_PKTS),
+ SIENA_DMA_STAT(rx_gtjumbo, RX_GTJUMBO_PKTS),
+ SIENA_DMA_STAT(rx_bad_gtjumbo, RX_JABBER_PKTS),
+ SIENA_DMA_STAT(rx_overflow, RX_OVERFLOW_PKTS),
+ SIENA_DMA_STAT(rx_false_carrier, RX_FALSE_CARRIER_PKTS),
+ SIENA_DMA_STAT(rx_symbol_error, RX_SYMBOL_ERROR_PKTS),
+ SIENA_DMA_STAT(rx_align_error, RX_ALIGN_ERROR_PKTS),
+ SIENA_DMA_STAT(rx_length_error, RX_LENGTH_ERROR_PKTS),
+ SIENA_DMA_STAT(rx_internal_error, RX_INTERNAL_ERROR_PKTS),
+ SIENA_DMA_STAT(rx_nodesc_drop_cnt, RX_NODESC_DROPS),
+};
+static const unsigned long siena_stat_mask[] = {
+ [0 ... BITS_TO_LONGS(SIENA_STAT_COUNT) - 1] = ~0UL,
+};
+
+static size_t siena_describe_nic_stats(struct efx_nic *efx, u8 *names)
+{
+ return efx_nic_describe_stats(siena_stat_desc, SIENA_STAT_COUNT,
+ siena_stat_mask, names);
+}
+
static int siena_try_update_nic_stats(struct efx_nic *efx)
{
+ struct siena_nic_data *nic_data = efx->nic_data;
+ u64 *stats = nic_data->stats;
__le64 *dma_stats;
- struct efx_mac_stats *mac_stats;
__le64 generation_start, generation_end;
- mac_stats = &efx->mac_stats;
dma_stats = efx->stats_buffer.addr;
generation_end = dma_stats[MC_CMD_MAC_GENERATION_END];
if (generation_end == EFX_MC_STATS_GENERATION_INVALID)
return 0;
rmb();
-
-#define MAC_STAT(M, D) \
- mac_stats->M = le64_to_cpu(dma_stats[MC_CMD_MAC_ ## D])
-
- MAC_STAT(tx_bytes, TX_BYTES);
- MAC_STAT(tx_bad_bytes, TX_BAD_BYTES);
- efx_update_diff_stat(&mac_stats->tx_good_bytes,
- mac_stats->tx_bytes - mac_stats->tx_bad_bytes);
- MAC_STAT(tx_packets, TX_PKTS);
- MAC_STAT(tx_bad, TX_BAD_FCS_PKTS);
- MAC_STAT(tx_pause, TX_PAUSE_PKTS);
- MAC_STAT(tx_control, TX_CONTROL_PKTS);
- MAC_STAT(tx_unicast, TX_UNICAST_PKTS);
- MAC_STAT(tx_multicast, TX_MULTICAST_PKTS);
- MAC_STAT(tx_broadcast, TX_BROADCAST_PKTS);
- MAC_STAT(tx_lt64, TX_LT64_PKTS);
- MAC_STAT(tx_64, TX_64_PKTS);
- MAC_STAT(tx_65_to_127, TX_65_TO_127_PKTS);
- MAC_STAT(tx_128_to_255, TX_128_TO_255_PKTS);
- MAC_STAT(tx_256_to_511, TX_256_TO_511_PKTS);
- MAC_STAT(tx_512_to_1023, TX_512_TO_1023_PKTS);
- MAC_STAT(tx_1024_to_15xx, TX_1024_TO_15XX_PKTS);
- MAC_STAT(tx_15xx_to_jumbo, TX_15XX_TO_JUMBO_PKTS);
- MAC_STAT(tx_gtjumbo, TX_GTJUMBO_PKTS);
- mac_stats->tx_collision = 0;
- MAC_STAT(tx_single_collision, TX_SINGLE_COLLISION_PKTS);
- MAC_STAT(tx_multiple_collision, TX_MULTIPLE_COLLISION_PKTS);
- MAC_STAT(tx_excessive_collision, TX_EXCESSIVE_COLLISION_PKTS);
- MAC_STAT(tx_deferred, TX_DEFERRED_PKTS);
- MAC_STAT(tx_late_collision, TX_LATE_COLLISION_PKTS);
- mac_stats->tx_collision = (mac_stats->tx_single_collision +
- mac_stats->tx_multiple_collision +
- mac_stats->tx_excessive_collision +
- mac_stats->tx_late_collision);
- MAC_STAT(tx_excessive_deferred, TX_EXCESSIVE_DEFERRED_PKTS);
- MAC_STAT(tx_non_tcpudp, TX_NON_TCPUDP_PKTS);
- MAC_STAT(tx_mac_src_error, TX_MAC_SRC_ERR_PKTS);
- MAC_STAT(tx_ip_src_error, TX_IP_SRC_ERR_PKTS);
- MAC_STAT(rx_bytes, RX_BYTES);
- MAC_STAT(rx_bad_bytes, RX_BAD_BYTES);
- efx_update_diff_stat(&mac_stats->rx_good_bytes,
- mac_stats->rx_bytes - mac_stats->rx_bad_bytes);
- MAC_STAT(rx_packets, RX_PKTS);
- MAC_STAT(rx_good, RX_GOOD_PKTS);
- MAC_STAT(rx_bad, RX_BAD_FCS_PKTS);
- MAC_STAT(rx_pause, RX_PAUSE_PKTS);
- MAC_STAT(rx_control, RX_CONTROL_PKTS);
- MAC_STAT(rx_unicast, RX_UNICAST_PKTS);
- MAC_STAT(rx_multicast, RX_MULTICAST_PKTS);
- MAC_STAT(rx_broadcast, RX_BROADCAST_PKTS);
- MAC_STAT(rx_lt64, RX_UNDERSIZE_PKTS);
- MAC_STAT(rx_64, RX_64_PKTS);
- MAC_STAT(rx_65_to_127, RX_65_TO_127_PKTS);
- MAC_STAT(rx_128_to_255, RX_128_TO_255_PKTS);
- MAC_STAT(rx_256_to_511, RX_256_TO_511_PKTS);
- MAC_STAT(rx_512_to_1023, RX_512_TO_1023_PKTS);
- MAC_STAT(rx_1024_to_15xx, RX_1024_TO_15XX_PKTS);
- MAC_STAT(rx_15xx_to_jumbo, RX_15XX_TO_JUMBO_PKTS);
- MAC_STAT(rx_gtjumbo, RX_GTJUMBO_PKTS);
- mac_stats->rx_bad_lt64 = 0;
- mac_stats->rx_bad_64_to_15xx = 0;
- mac_stats->rx_bad_15xx_to_jumbo = 0;
- MAC_STAT(rx_bad_gtjumbo, RX_JABBER_PKTS);
- MAC_STAT(rx_overflow, RX_OVERFLOW_PKTS);
- mac_stats->rx_missed = 0;
- MAC_STAT(rx_false_carrier, RX_FALSE_CARRIER_PKTS);
- MAC_STAT(rx_symbol_error, RX_SYMBOL_ERROR_PKTS);
- MAC_STAT(rx_align_error, RX_ALIGN_ERROR_PKTS);
- MAC_STAT(rx_length_error, RX_LENGTH_ERROR_PKTS);
- MAC_STAT(rx_internal_error, RX_INTERNAL_ERROR_PKTS);
- mac_stats->rx_good_lt64 = 0;
-
- efx->n_rx_nodesc_drop_cnt =
- le64_to_cpu(dma_stats[MC_CMD_MAC_RX_NODESC_DROPS]);
-
-#undef MAC_STAT
-
+ efx_nic_update_stats(siena_stat_desc, SIENA_STAT_COUNT, siena_stat_mask,
+ stats, efx->stats_buffer.addr, false);
rmb();
generation_start = dma_stats[MC_CMD_MAC_GENERATION_START];
if (generation_end != generation_start)
return -EAGAIN;
+ /* Update derived statistics */
+ efx_update_diff_stat(&stats[SIENA_STAT_tx_good_bytes],
+ stats[SIENA_STAT_tx_bytes] -
+ stats[SIENA_STAT_tx_bad_bytes]);
+ stats[SIENA_STAT_tx_collision] =
+ stats[SIENA_STAT_tx_single_collision] +
+ stats[SIENA_STAT_tx_multiple_collision] +
+ stats[SIENA_STAT_tx_excessive_collision] +
+ stats[SIENA_STAT_tx_late_collision];
+ efx_update_diff_stat(&stats[SIENA_STAT_rx_good_bytes],
+ stats[SIENA_STAT_rx_bytes] -
+ stats[SIENA_STAT_rx_bad_bytes]);
return 0;
}
-static void siena_update_nic_stats(struct efx_nic *efx)
+static size_t siena_update_nic_stats(struct efx_nic *efx, u64 *full_stats,
+ struct rtnl_link_stats64 *core_stats)
{
+ struct siena_nic_data *nic_data = efx->nic_data;
+ u64 *stats = nic_data->stats;
int retry;
/* If we're unlucky enough to read statistics wduring the DMA, wait
* up to 10ms for it to finish (typically takes <500us) */
for (retry = 0; retry < 100; ++retry) {
if (siena_try_update_nic_stats(efx) == 0)
- return;
+ break;
udelay(100);
}
- /* Use the old values instead */
+ if (full_stats)
+ memcpy(full_stats, stats, sizeof(u64) * SIENA_STAT_COUNT);
+
+ if (core_stats) {
+ core_stats->rx_packets = stats[SIENA_STAT_rx_packets];
+ core_stats->tx_packets = stats[SIENA_STAT_tx_packets];
+ core_stats->rx_bytes = stats[SIENA_STAT_rx_bytes];
+ core_stats->tx_bytes = stats[SIENA_STAT_tx_bytes];
+ core_stats->rx_dropped = stats[SIENA_STAT_rx_nodesc_drop_cnt];
+ core_stats->multicast = stats[SIENA_STAT_rx_multicast];
+ core_stats->collisions = stats[SIENA_STAT_tx_collision];
+ core_stats->rx_length_errors =
+ stats[SIENA_STAT_rx_gtjumbo] +
+ stats[SIENA_STAT_rx_length_error];
+ core_stats->rx_crc_errors = stats[SIENA_STAT_rx_bad];
+ core_stats->rx_frame_errors = stats[SIENA_STAT_rx_align_error];
+ core_stats->rx_fifo_errors = stats[SIENA_STAT_rx_overflow];
+ core_stats->tx_window_errors =
+ stats[SIENA_STAT_tx_late_collision];
+
+ core_stats->rx_errors = (core_stats->rx_length_errors +
+ core_stats->rx_crc_errors +
+ core_stats->rx_frame_errors +
+ stats[SIENA_STAT_rx_symbol_error]);
+ core_stats->tx_errors = (core_stats->tx_window_errors +
+ stats[SIENA_STAT_tx_bad]);
+ }
+
+ return SIENA_STAT_COUNT;
}
static int siena_mac_reconfigure(struct efx_nic *efx)
@@ -652,6 +695,7 @@ static void siena_mcdi_read_response(struct efx_nic *efx, efx_dword_t *outbuf,
static int siena_mcdi_poll_reboot(struct efx_nic *efx)
{
+ struct siena_nic_data *nic_data = efx->nic_data;
unsigned int addr = FR_CZ_MC_TREG_SMEM + MCDI_STATUS(efx);
efx_dword_t reg;
u32 value;
@@ -665,6 +709,12 @@ static int siena_mcdi_poll_reboot(struct efx_nic *efx)
EFX_ZERO_DWORD(reg);
efx_writed(efx, ®, addr);
+ /* MAC statistics have been cleared on the NIC; clear the local
+ * copies that we update with efx_update_diff_stat().
+ */
+ nic_data->stats[SIENA_STAT_tx_good_bytes] = 0;
+ nic_data->stats[SIENA_STAT_rx_good_bytes] = 0;
+
if (value == MC_STATUS_DWORD_ASSERT)
return -EINTR;
else
@@ -830,6 +880,7 @@ const struct efx_nic_type siena_a0_nic_type = {
.fini_dmaq = efx_farch_fini_dmaq,
.prepare_flush = siena_prepare_flush,
.finish_flush = siena_finish_flush,
+ .describe_stats = siena_describe_nic_stats,
.update_stats = siena_update_nic_stats,
.start_stats = efx_mcdi_mac_start_stats,
.stop_stats = efx_mcdi_mac_stop_stats,
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 04/16] sfc: Remove driver-local struct ethtool_string
From: Ben Hutchings @ 2013-08-28 15:18 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
It's not really helpful to pretend ethtool string arrays are
structured.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/ethtool.c | 37 +++++++++++++------------------------
1 file changed, 13 insertions(+), 24 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 55c3826..86b4391 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -19,10 +19,6 @@
#include "filter.h"
#include "nic.h"
-struct ethtool_string {
- char name[ETH_GSTRING_LEN];
-};
-
struct efx_ethtool_stat {
const char *name;
enum {
@@ -289,12 +285,11 @@ static void efx_ethtool_set_msglevel(struct net_device *net_dev, u32 msg_enable)
*
* Fill in an individual self-test entry.
*/
-static void efx_fill_test(unsigned int test_index,
- struct ethtool_string *strings, u64 *data,
+static void efx_fill_test(unsigned int test_index, u8 *strings, u64 *data,
int *test, const char *unit_format, int unit_id,
const char *test_format, const char *test_id)
{
- struct ethtool_string unit_str, test_str;
+ char unit_str[ETH_GSTRING_LEN], test_str[ETH_GSTRING_LEN];
/* Fill data value, if applicable */
if (data)
@@ -303,15 +298,14 @@ static void efx_fill_test(unsigned int test_index,
/* Fill string, if applicable */
if (strings) {
if (strchr(unit_format, '%'))
- snprintf(unit_str.name, sizeof(unit_str.name),
+ snprintf(unit_str, sizeof(unit_str),
unit_format, unit_id);
else
- strcpy(unit_str.name, unit_format);
- snprintf(test_str.name, sizeof(test_str.name),
- test_format, test_id);
- snprintf(strings[test_index].name,
- sizeof(strings[test_index].name),
- "%-6s %-24s", unit_str.name, test_str.name);
+ strcpy(unit_str, unit_format);
+ snprintf(test_str, sizeof(test_str), test_format, test_id);
+ snprintf(strings + test_index * ETH_GSTRING_LEN,
+ ETH_GSTRING_LEN,
+ "%-6s %-24s", unit_str, test_str);
}
}
@@ -334,7 +328,7 @@ static int efx_fill_loopback_test(struct efx_nic *efx,
struct efx_loopback_self_tests *lb_tests,
enum efx_loopback_mode mode,
unsigned int test_index,
- struct ethtool_string *strings, u64 *data)
+ u8 *strings, u64 *data)
{
struct efx_channel *channel =
efx_get_channel(efx, efx->tx_channel_offset);
@@ -371,8 +365,7 @@ static int efx_fill_loopback_test(struct efx_nic *efx,
*/
static int efx_ethtool_fill_self_tests(struct efx_nic *efx,
struct efx_self_tests *tests,
- struct ethtool_string *strings,
- u64 *data)
+ u8 *strings, u64 *data)
{
struct efx_channel *channel;
unsigned int n = 0, i;
@@ -446,20 +439,16 @@ static void efx_ethtool_get_strings(struct net_device *net_dev,
u32 string_set, u8 *strings)
{
struct efx_nic *efx = netdev_priv(net_dev);
- struct ethtool_string *ethtool_strings =
- (struct ethtool_string *)strings;
int i;
switch (string_set) {
case ETH_SS_STATS:
for (i = 0; i < EFX_ETHTOOL_NUM_STATS; i++)
- strlcpy(ethtool_strings[i].name,
- efx_ethtool_stats[i].name,
- sizeof(ethtool_strings[i].name));
+ strlcpy(strings + i * ETH_GSTRING_LEN,
+ efx_ethtool_stats[i].name, ETH_GSTRING_LEN);
break;
case ETH_SS_TEST:
- efx_ethtool_fill_self_tests(efx, NULL,
- ethtool_strings, NULL);
+ efx_ethtool_fill_self_tests(efx, NULL, strings, NULL);
break;
default:
/* No other string sets */
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 03/16] sfc: Remove more left-overs from Falcon GMAC support
From: Ben Hutchings @ 2013-08-28 15:17 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
We only ever used the XMAC (10G link speed) in production.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/ethtool.c | 2 --
drivers/net/ethernet/sfc/falcon.c | 17 ++++++-----------
drivers/net/ethernet/sfc/nic.h | 5 -----
3 files changed, 6 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 58ae28b..55c3826 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -205,8 +205,6 @@ static int efx_ethtool_get_settings(struct net_device *net_dev,
efx->phy_op->get_settings(efx, ecmd);
mutex_unlock(&efx->mac_lock);
- /* GMAC does not support 1000Mbps HD */
- ecmd->supported &= ~SUPPORTED_1000baseT_Half;
/* Both MACs support pause frames (bidirectional and respond-only) */
ecmd->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index 3c99b6c..76a9e0d 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -131,8 +131,8 @@
#define XgDmaDone_offset 0xD4
#define XgDmaDone_WIDTH 32
-#define FALCON_STATS_NOT_DONE 0x00000000
-#define FALCON_STATS_DONE 0xffffffff
+#define FALCON_XMAC_STATS_DMA_FLAG(efx) \
+ (*(u32 *)((efx)->stats_buffer.addr + XgDmaDone_offset))
#define FALCON_STAT_OFFSET(falcon_stat) EFX_VAL(falcon_stat, offset)
#define FALCON_STAT_WIDTH(falcon_stat) EFX_VAL(falcon_stat, WIDTH)
@@ -1398,10 +1398,7 @@ static void falcon_stats_request(struct efx_nic *efx)
WARN_ON(nic_data->stats_pending);
WARN_ON(nic_data->stats_disable_count);
- if (nic_data->stats_dma_done == NULL)
- return; /* no mac selected */
-
- *nic_data->stats_dma_done = FALCON_STATS_NOT_DONE;
+ FALCON_XMAC_STATS_DMA_FLAG(efx) = 0;
nic_data->stats_pending = true;
wmb(); /* ensure done flag is clear */
@@ -1423,7 +1420,7 @@ static void falcon_stats_complete(struct efx_nic *efx)
return;
nic_data->stats_pending = false;
- if (*nic_data->stats_dma_done == FALCON_STATS_DONE) {
+ if (FALCON_XMAC_STATS_DMA_FLAG(efx)) {
rmb(); /* read the done flag before the stats */
falcon_update_stats_xmac(efx);
} else {
@@ -1706,7 +1703,6 @@ static int falcon_probe_port(struct efx_nic *efx)
(u64)efx->stats_buffer.dma_addr,
efx->stats_buffer.addr,
(u64)virt_to_phys(efx->stats_buffer.addr));
- nic_data->stats_dma_done = efx->stats_buffer.addr + XgDmaDone_offset;
return 0;
}
@@ -2554,8 +2550,7 @@ static void falcon_update_nic_stats(struct efx_nic *efx)
efx->n_rx_nodesc_drop_cnt +=
EFX_OWORD_FIELD(cnt, FRF_AB_RX_NODESC_DROP_CNT);
- if (nic_data->stats_pending &&
- *nic_data->stats_dma_done == FALCON_STATS_DONE) {
+ if (nic_data->stats_pending && FALCON_XMAC_STATS_DMA_FLAG(efx)) {
nic_data->stats_pending = false;
rmb(); /* read the done flag before the stats */
falcon_update_stats_xmac(efx);
@@ -2588,7 +2583,7 @@ void falcon_stop_nic_stats(struct efx_nic *efx)
/* Wait enough time for the most recent transfer to
* complete. */
for (i = 0; i < 4 && nic_data->stats_pending; i++) {
- if (*nic_data->stats_dma_done == FALCON_STATS_DONE)
+ if (FALCON_XMAC_STATS_DMA_FLAG(efx))
break;
msleep(1);
}
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index 2b84aeb..c3e0f1f 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -117,9 +117,6 @@ enum {
(1 << LOOPBACK_XGXS) | \
(1 << LOOPBACK_XAUI))
-#define FALCON_GMAC_LOOPBACKS \
- (1 << LOOPBACK_GMAC)
-
/* Alignment of PCIe DMA boundaries (4KB) */
#define EFX_PAGE_SIZE 4096
/* Size and alignment of buffer table entries (same) */
@@ -201,7 +198,6 @@ static inline bool falcon_spi_present(const struct falcon_spi_device *spi)
* @stats_disable_count: Nest count for disabling statistics fetches
* @stats_pending: Is there a pending DMA of MAC statistics.
* @stats_timer: A timer for regularly fetching MAC statistics.
- * @stats_dma_done: Pointer to the flag which indicates DMA completion.
* @spi_flash: SPI flash device
* @spi_eeprom: SPI EEPROM device
* @spi_lock: SPI bus lock
@@ -214,7 +210,6 @@ struct falcon_nic_data {
unsigned int stats_disable_count;
bool stats_pending;
struct timer_list stats_timer;
- u32 *stats_dma_done;
struct falcon_spi_device spi_flash;
struct falcon_spi_device spi_eeprom;
struct mutex spi_lock;
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 02/16] sfc: Move MTD operations into efx_nic_type
From: Ben Hutchings @ 2013-08-28 15:17 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
Merge the per-NIC-type MTD probe selection and struct efx_mtd_ops into
struct efx_nic_type. Move the implementations into the appropriate
source files.
Several NVRAM functions are now only called from MTD operations which
are now implemented in the same file (falcon.c or mcdi.c). There is no
need for them to be extern, or to be defined at all if CONFIG_SFC_MTD
is not enabled, so move them into the #ifdef CONFIG_SFC_MTD sections
in those files.
Most of the SPI-related definitions are also only used in falcon.c,
so move them there. Put the remainder of spi.h into nic.h (which
previously included it).
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/efx.h | 7 +-
drivers/net/ethernet/sfc/falcon.c | 381 ++++++++++++++++++++---
drivers/net/ethernet/sfc/mcdi.c | 352 ++++++++++++++-------
drivers/net/ethernet/sfc/mcdi.h | 32 +-
drivers/net/ethernet/sfc/mtd.c | 564 +---------------------------------
drivers/net/ethernet/sfc/net_driver.h | 29 +-
drivers/net/ethernet/sfc/nic.h | 33 +-
drivers/net/ethernet/sfc/siena.c | 141 ++++++++-
drivers/net/ethernet/sfc/spi.h | 99 ------
9 files changed, 807 insertions(+), 831 deletions(-)
delete mode 100644 drivers/net/ethernet/sfc/spi.h
diff --git a/drivers/net/ethernet/sfc/efx.h b/drivers/net/ethernet/sfc/efx.h
index 9e35738..3bbc047 100644
--- a/drivers/net/ethernet/sfc/efx.h
+++ b/drivers/net/ethernet/sfc/efx.h
@@ -204,7 +204,12 @@ extern void efx_port_dummy_op_void(struct efx_nic *efx);
/* MTD */
#ifdef CONFIG_SFC_MTD
-extern int efx_mtd_probe(struct efx_nic *efx);
+extern int efx_mtd_add(struct efx_nic *efx, struct efx_mtd_partition *parts,
+ size_t n_parts, size_t sizeof_part);
+static inline int efx_mtd_probe(struct efx_nic *efx)
+{
+ return efx->type->mtd_probe(efx);
+}
extern void efx_mtd_rename(struct efx_nic *efx);
extern void efx_mtd_remove(struct efx_nic *efx);
#else
diff --git a/drivers/net/ethernet/sfc/falcon.c b/drivers/net/ethernet/sfc/falcon.c
index 304131b..3c99b6c 100644
--- a/drivers/net/ethernet/sfc/falcon.c
+++ b/drivers/net/ethernet/sfc/falcon.c
@@ -19,7 +19,6 @@
#include "net_driver.h"
#include "bitfield.h"
#include "efx.h"
-#include "spi.h"
#include "nic.h"
#include "farch_regs.h"
#include "io.h"
@@ -159,11 +158,49 @@
/**************************************************************************
*
- * Non-volatile configuration
+ * Basic SPI command set and bit definitions
+ *
+ *************************************************************************/
+
+#define SPI_WRSR 0x01 /* Write status register */
+#define SPI_WRITE 0x02 /* Write data to memory array */
+#define SPI_READ 0x03 /* Read data from memory array */
+#define SPI_WRDI 0x04 /* Reset write enable latch */
+#define SPI_RDSR 0x05 /* Read status register */
+#define SPI_WREN 0x06 /* Set write enable latch */
+#define SPI_SST_EWSR 0x50 /* SST: Enable write to status register */
+
+#define SPI_STATUS_WPEN 0x80 /* Write-protect pin enabled */
+#define SPI_STATUS_BP2 0x10 /* Block protection bit 2 */
+#define SPI_STATUS_BP1 0x08 /* Block protection bit 1 */
+#define SPI_STATUS_BP0 0x04 /* Block protection bit 0 */
+#define SPI_STATUS_WEN 0x02 /* State of the write enable latch */
+#define SPI_STATUS_NRDY 0x01 /* Device busy flag */
+
+/**************************************************************************
+ *
+ * Non-volatile memory layout
*
**************************************************************************
*/
+/* SFC4000 flash is partitioned into:
+ * 0-0x400 chip and board config (see struct falcon_nvconfig)
+ * 0x400-0x8000 unused (or may contain VPD if EEPROM not present)
+ * 0x8000-end boot code (mapped to PCI expansion ROM)
+ * SFC4000 small EEPROM (size < 0x400) is used for VPD only.
+ * SFC4000 large EEPROM (size >= 0x400) is partitioned into:
+ * 0-0x400 chip and board config
+ * configurable VPD
+ * 0x800-0x1800 boot config
+ * Aside from the chip and board config, all of these are optional and may
+ * be absent or truncated depending on the devices used.
+ */
+#define FALCON_NVCONFIG_END 0x400U
+#define FALCON_FLASH_BOOTCODE_START 0x8000U
+#define FALCON_EEPROM_BOOTCONFIG_START 0x800U
+#define FALCON_EEPROM_BOOTCONFIG_END 0x1800U
+
/* Board configuration v2 (v1 is obsolete; later versions are compatible) */
struct falcon_nvconfig_board_v2 {
__le16 nports;
@@ -434,9 +471,10 @@ static int falcon_spi_wait(struct efx_nic *efx)
}
}
-int falcon_spi_cmd(struct efx_nic *efx, const struct falcon_spi_device *spi,
- unsigned int command, int address,
- const void *in, void *out, size_t len)
+static int
+falcon_spi_cmd(struct efx_nic *efx, const struct falcon_spi_device *spi,
+ unsigned int command, int address,
+ const void *in, void *out, size_t len)
{
bool addressed = (address >= 0);
bool reading = (out != NULL);
@@ -490,13 +528,6 @@ int falcon_spi_cmd(struct efx_nic *efx, const struct falcon_spi_device *spi,
return 0;
}
-static size_t
-falcon_spi_write_limit(const struct falcon_spi_device *spi, size_t start)
-{
- return min(FALCON_SPI_MAX_LEN,
- (spi->block_size - (start & (spi->block_size - 1))));
-}
-
static inline u8
falcon_spi_munge_command(const struct falcon_spi_device *spi,
const u8 command, const unsigned int address)
@@ -504,34 +535,9 @@ falcon_spi_munge_command(const struct falcon_spi_device *spi,
return command | (((address >> 8) & spi->munge_address) << 3);
}
-/* Wait up to 10 ms for buffered write completion */
-int
-falcon_spi_wait_write(struct efx_nic *efx, const struct falcon_spi_device *spi)
-{
- unsigned long timeout = jiffies + 1 + DIV_ROUND_UP(HZ, 100);
- u8 status;
- int rc;
-
- for (;;) {
- rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
- &status, sizeof(status));
- if (rc)
- return rc;
- if (!(status & SPI_STATUS_NRDY))
- return 0;
- if (time_after_eq(jiffies, timeout)) {
- netif_err(efx, hw, efx->net_dev,
- "SPI write timeout on device %d"
- " last status=0x%02x\n",
- spi->device_id, status);
- return -ETIMEDOUT;
- }
- schedule_timeout_uninterruptible(1);
- }
-}
-
-int falcon_spi_read(struct efx_nic *efx, const struct falcon_spi_device *spi,
- loff_t start, size_t len, size_t *retlen, u8 *buffer)
+static int
+falcon_spi_read(struct efx_nic *efx, const struct falcon_spi_device *spi,
+ loff_t start, size_t len, size_t *retlen, u8 *buffer)
{
size_t block_len, pos = 0;
unsigned int command;
@@ -560,7 +566,51 @@ int falcon_spi_read(struct efx_nic *efx, const struct falcon_spi_device *spi,
return rc;
}
-int
+#ifdef CONFIG_SFC_MTD
+
+struct falcon_mtd_partition {
+ struct efx_mtd_partition common;
+ const struct falcon_spi_device *spi;
+ size_t offset;
+};
+
+#define to_falcon_mtd_partition(mtd) \
+ container_of(mtd, struct falcon_mtd_partition, common.mtd)
+
+static size_t
+falcon_spi_write_limit(const struct falcon_spi_device *spi, size_t start)
+{
+ return min(FALCON_SPI_MAX_LEN,
+ (spi->block_size - (start & (spi->block_size - 1))));
+}
+
+/* Wait up to 10 ms for buffered write completion */
+static int
+falcon_spi_wait_write(struct efx_nic *efx, const struct falcon_spi_device *spi)
+{
+ unsigned long timeout = jiffies + 1 + DIV_ROUND_UP(HZ, 100);
+ u8 status;
+ int rc;
+
+ for (;;) {
+ rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+ &status, sizeof(status));
+ if (rc)
+ return rc;
+ if (!(status & SPI_STATUS_NRDY))
+ return 0;
+ if (time_after_eq(jiffies, timeout)) {
+ netif_err(efx, hw, efx->net_dev,
+ "SPI write timeout on device %d"
+ " last status=0x%02x\n",
+ spi->device_id, status);
+ return -ETIMEDOUT;
+ }
+ schedule_timeout_uninterruptible(1);
+ }
+}
+
+static int
falcon_spi_write(struct efx_nic *efx, const struct falcon_spi_device *spi,
loff_t start, size_t len, size_t *retlen, const u8 *buffer)
{
@@ -609,6 +659,238 @@ falcon_spi_write(struct efx_nic *efx, const struct falcon_spi_device *spi,
return rc;
}
+static int
+falcon_spi_slow_wait(struct falcon_mtd_partition *part, bool uninterruptible)
+{
+ const struct falcon_spi_device *spi = part->spi;
+ struct efx_nic *efx = part->common.mtd.priv;
+ u8 status;
+ int rc, i;
+
+ /* Wait up to 4s for flash/EEPROM to finish a slow operation. */
+ for (i = 0; i < 40; i++) {
+ __set_current_state(uninterruptible ?
+ TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE);
+ schedule_timeout(HZ / 10);
+ rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+ &status, sizeof(status));
+ if (rc)
+ return rc;
+ if (!(status & SPI_STATUS_NRDY))
+ return 0;
+ if (signal_pending(current))
+ return -EINTR;
+ }
+ pr_err("%s: timed out waiting for %s\n",
+ part->common.name, part->common.dev_type_name);
+ return -ETIMEDOUT;
+}
+
+static int
+falcon_spi_unlock(struct efx_nic *efx, const struct falcon_spi_device *spi)
+{
+ const u8 unlock_mask = (SPI_STATUS_BP2 | SPI_STATUS_BP1 |
+ SPI_STATUS_BP0);
+ u8 status;
+ int rc;
+
+ rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
+ &status, sizeof(status));
+ if (rc)
+ return rc;
+
+ if (!(status & unlock_mask))
+ return 0; /* already unlocked */
+
+ rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
+ if (rc)
+ return rc;
+ rc = falcon_spi_cmd(efx, spi, SPI_SST_EWSR, -1, NULL, NULL, 0);
+ if (rc)
+ return rc;
+
+ status &= ~unlock_mask;
+ rc = falcon_spi_cmd(efx, spi, SPI_WRSR, -1, &status,
+ NULL, sizeof(status));
+ if (rc)
+ return rc;
+ rc = falcon_spi_wait_write(efx, spi);
+ if (rc)
+ return rc;
+
+ return 0;
+}
+
+#define FALCON_SPI_VERIFY_BUF_LEN 16
+
+static int
+falcon_spi_erase(struct falcon_mtd_partition *part, loff_t start, size_t len)
+{
+ const struct falcon_spi_device *spi = part->spi;
+ struct efx_nic *efx = part->common.mtd.priv;
+ unsigned pos, block_len;
+ u8 empty[FALCON_SPI_VERIFY_BUF_LEN];
+ u8 buffer[FALCON_SPI_VERIFY_BUF_LEN];
+ int rc;
+
+ if (len != spi->erase_size)
+ return -EINVAL;
+
+ if (spi->erase_command == 0)
+ return -EOPNOTSUPP;
+
+ rc = falcon_spi_unlock(efx, spi);
+ if (rc)
+ return rc;
+ rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
+ if (rc)
+ return rc;
+ rc = falcon_spi_cmd(efx, spi, spi->erase_command, start, NULL,
+ NULL, 0);
+ if (rc)
+ return rc;
+ rc = falcon_spi_slow_wait(part, false);
+
+ /* Verify the entire region has been wiped */
+ memset(empty, 0xff, sizeof(empty));
+ for (pos = 0; pos < len; pos += block_len) {
+ block_len = min(len - pos, sizeof(buffer));
+ rc = falcon_spi_read(efx, spi, start + pos, block_len,
+ NULL, buffer);
+ if (rc)
+ return rc;
+ if (memcmp(empty, buffer, block_len))
+ return -EIO;
+
+ /* Avoid locking up the system */
+ cond_resched();
+ if (signal_pending(current))
+ return -EINTR;
+ }
+
+ return rc;
+}
+
+static void falcon_mtd_rename(struct efx_mtd_partition *part)
+{
+ struct efx_nic *efx = part->mtd.priv;
+
+ snprintf(part->name, sizeof(part->name), "%s %s",
+ efx->name, part->type_name);
+}
+
+static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, u8 *buffer)
+{
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ struct falcon_nic_data *nic_data = efx->nic_data;
+ int rc;
+
+ rc = mutex_lock_interruptible(&nic_data->spi_lock);
+ if (rc)
+ return rc;
+ rc = falcon_spi_read(efx, part->spi, part->offset + start,
+ len, retlen, buffer);
+ mutex_unlock(&nic_data->spi_lock);
+ return rc;
+}
+
+static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
+{
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ struct falcon_nic_data *nic_data = efx->nic_data;
+ int rc;
+
+ rc = mutex_lock_interruptible(&nic_data->spi_lock);
+ if (rc)
+ return rc;
+ rc = falcon_spi_erase(part, part->offset + start, len);
+ mutex_unlock(&nic_data->spi_lock);
+ return rc;
+}
+
+static int falcon_mtd_write(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, const u8 *buffer)
+{
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ struct falcon_nic_data *nic_data = efx->nic_data;
+ int rc;
+
+ rc = mutex_lock_interruptible(&nic_data->spi_lock);
+ if (rc)
+ return rc;
+ rc = falcon_spi_write(efx, part->spi, part->offset + start,
+ len, retlen, buffer);
+ mutex_unlock(&nic_data->spi_lock);
+ return rc;
+}
+
+static int falcon_mtd_sync(struct mtd_info *mtd)
+{
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ struct falcon_nic_data *nic_data = efx->nic_data;
+ int rc;
+
+ mutex_lock(&nic_data->spi_lock);
+ rc = falcon_spi_slow_wait(part, true);
+ mutex_unlock(&nic_data->spi_lock);
+ return rc;
+}
+
+static int falcon_mtd_probe(struct efx_nic *efx)
+{
+ struct falcon_nic_data *nic_data = efx->nic_data;
+ struct falcon_mtd_partition *parts;
+ struct falcon_spi_device *spi;
+ size_t n_parts;
+ int rc = -ENODEV;
+
+ ASSERT_RTNL();
+
+ /* Allocate space for maximum number of partitions */
+ parts = kcalloc(2, sizeof(*parts), GFP_KERNEL);
+ n_parts = 0;
+
+ spi = &nic_data->spi_flash;
+ if (falcon_spi_present(spi) && spi->size > FALCON_FLASH_BOOTCODE_START) {
+ parts[n_parts].spi = spi;
+ parts[n_parts].offset = FALCON_FLASH_BOOTCODE_START;
+ parts[n_parts].common.dev_type_name = "flash";
+ parts[n_parts].common.type_name = "sfc_flash_bootrom";
+ parts[n_parts].common.mtd.type = MTD_NORFLASH;
+ parts[n_parts].common.mtd.flags = MTD_CAP_NORFLASH;
+ parts[n_parts].common.mtd.size = spi->size - FALCON_FLASH_BOOTCODE_START;
+ parts[n_parts].common.mtd.erasesize = spi->erase_size;
+ n_parts++;
+ }
+
+ spi = &nic_data->spi_eeprom;
+ if (falcon_spi_present(spi) && spi->size > FALCON_EEPROM_BOOTCONFIG_START) {
+ parts[n_parts].spi = spi;
+ parts[n_parts].offset = FALCON_EEPROM_BOOTCONFIG_START;
+ parts[n_parts].common.dev_type_name = "EEPROM";
+ parts[n_parts].common.type_name = "sfc_bootconfig";
+ parts[n_parts].common.mtd.type = MTD_RAM;
+ parts[n_parts].common.mtd.flags = MTD_CAP_RAM;
+ parts[n_parts].common.mtd.size =
+ min(spi->size, FALCON_EEPROM_BOOTCONFIG_END) -
+ FALCON_EEPROM_BOOTCONFIG_START;
+ parts[n_parts].common.mtd.erasesize = spi->erase_size;
+ n_parts++;
+ }
+
+ rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
+ if (rc)
+ kfree(parts);
+ return rc;
+}
+
+#endif /* CONFIG_SFC_MTD */
+
/**************************************************************************
*
* XMAC operations
@@ -2417,6 +2699,15 @@ const struct efx_nic_type falcon_a1_nic_type = {
.filter_get_rx_id_limit = efx_farch_filter_get_rx_id_limit,
.filter_get_rx_ids = efx_farch_filter_get_rx_ids,
+#ifdef CONFIG_SFC_MTD
+ .mtd_probe = falcon_mtd_probe,
+ .mtd_rename = falcon_mtd_rename,
+ .mtd_read = falcon_mtd_read,
+ .mtd_erase = falcon_mtd_erase,
+ .mtd_write = falcon_mtd_write,
+ .mtd_sync = falcon_mtd_sync,
+#endif
+
.revision = EFX_REV_FALCON_A1,
.txd_ptr_tbl_base = FR_AA_TX_DESC_PTR_TBL_KER,
.rxd_ptr_tbl_base = FR_AA_RX_DESC_PTR_TBL_KER,
@@ -2500,6 +2791,14 @@ const struct efx_nic_type falcon_b0_nic_type = {
.filter_rfs_insert = efx_farch_filter_rfs_insert,
.filter_rfs_expire_one = efx_farch_filter_rfs_expire_one,
#endif
+#ifdef CONFIG_SFC_MTD
+ .mtd_probe = falcon_mtd_probe,
+ .mtd_rename = falcon_mtd_rename,
+ .mtd_read = falcon_mtd_read,
+ .mtd_erase = falcon_mtd_erase,
+ .mtd_write = falcon_mtd_write,
+ .mtd_sync = falcon_mtd_sync,
+#endif
.revision = EFX_REV_FALCON_B0,
.txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL,
diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c
index 1c8bf81..67b07de 100644
--- a/drivers/net/ethernet/sfc/mcdi.c
+++ b/drivers/net/ethernet/sfc/mcdi.c
@@ -811,125 +811,6 @@ fail:
return rc;
}
-int efx_mcdi_nvram_update_start(struct efx_nic *efx, unsigned int type)
-{
- MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_UPDATE_START_IN_LEN);
- int rc;
-
- MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_START_IN_TYPE, type);
-
- BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_START_OUT_LEN != 0);
-
- rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_START, inbuf, sizeof(inbuf),
- NULL, 0, NULL);
- if (rc)
- goto fail;
-
- return 0;
-
-fail:
- netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
- return rc;
-}
-
-int efx_mcdi_nvram_read(struct efx_nic *efx, unsigned int type,
- loff_t offset, u8 *buffer, size_t length)
-{
- MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_READ_IN_LEN);
- MCDI_DECLARE_BUF(outbuf,
- MC_CMD_NVRAM_READ_OUT_LEN(EFX_MCDI_NVRAM_LEN_MAX));
- size_t outlen;
- int rc;
-
- MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_TYPE, type);
- MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_OFFSET, offset);
- MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_LENGTH, length);
-
- rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_READ, inbuf, sizeof(inbuf),
- outbuf, sizeof(outbuf), &outlen);
- if (rc)
- goto fail;
-
- memcpy(buffer, MCDI_PTR(outbuf, NVRAM_READ_OUT_READ_BUFFER), length);
- return 0;
-
-fail:
- netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
- return rc;
-}
-
-int efx_mcdi_nvram_write(struct efx_nic *efx, unsigned int type,
- loff_t offset, const u8 *buffer, size_t length)
-{
- MCDI_DECLARE_BUF(inbuf,
- MC_CMD_NVRAM_WRITE_IN_LEN(EFX_MCDI_NVRAM_LEN_MAX));
- int rc;
-
- MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_TYPE, type);
- MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_OFFSET, offset);
- MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_LENGTH, length);
- memcpy(MCDI_PTR(inbuf, NVRAM_WRITE_IN_WRITE_BUFFER), buffer, length);
-
- BUILD_BUG_ON(MC_CMD_NVRAM_WRITE_OUT_LEN != 0);
-
- rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_WRITE, inbuf,
- ALIGN(MC_CMD_NVRAM_WRITE_IN_LEN(length), 4),
- NULL, 0, NULL);
- if (rc)
- goto fail;
-
- return 0;
-
-fail:
- netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
- return rc;
-}
-
-int efx_mcdi_nvram_erase(struct efx_nic *efx, unsigned int type,
- loff_t offset, size_t length)
-{
- MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_ERASE_IN_LEN);
- int rc;
-
- MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_TYPE, type);
- MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_OFFSET, offset);
- MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_LENGTH, length);
-
- BUILD_BUG_ON(MC_CMD_NVRAM_ERASE_OUT_LEN != 0);
-
- rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_ERASE, inbuf, sizeof(inbuf),
- NULL, 0, NULL);
- if (rc)
- goto fail;
-
- return 0;
-
-fail:
- netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
- return rc;
-}
-
-int efx_mcdi_nvram_update_finish(struct efx_nic *efx, unsigned int type)
-{
- MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_UPDATE_FINISH_IN_LEN);
- int rc;
-
- MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_FINISH_IN_TYPE, type);
-
- BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_FINISH_OUT_LEN != 0);
-
- rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_FINISH, inbuf, sizeof(inbuf),
- NULL, 0, NULL);
- if (rc)
- goto fail;
-
- return 0;
-
-fail:
- netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
- return rc;
-}
-
static int efx_mcdi_nvram_test(struct efx_nic *efx, unsigned int type)
{
MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_TEST_IN_LEN);
@@ -1272,3 +1153,236 @@ fail:
return rc;
}
+#ifdef CONFIG_SFC_MTD
+
+#define EFX_MCDI_NVRAM_LEN_MAX 128
+
+static int efx_mcdi_nvram_update_start(struct efx_nic *efx, unsigned int type)
+{
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_UPDATE_START_IN_LEN);
+ int rc;
+
+ MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_START_IN_TYPE, type);
+
+ BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_START_OUT_LEN != 0);
+
+ rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_START, inbuf, sizeof(inbuf),
+ NULL, 0, NULL);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+ return rc;
+}
+
+static int efx_mcdi_nvram_read(struct efx_nic *efx, unsigned int type,
+ loff_t offset, u8 *buffer, size_t length)
+{
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_READ_IN_LEN);
+ MCDI_DECLARE_BUF(outbuf,
+ MC_CMD_NVRAM_READ_OUT_LEN(EFX_MCDI_NVRAM_LEN_MAX));
+ size_t outlen;
+ int rc;
+
+ MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_TYPE, type);
+ MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_OFFSET, offset);
+ MCDI_SET_DWORD(inbuf, NVRAM_READ_IN_LENGTH, length);
+
+ rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_READ, inbuf, sizeof(inbuf),
+ outbuf, sizeof(outbuf), &outlen);
+ if (rc)
+ goto fail;
+
+ memcpy(buffer, MCDI_PTR(outbuf, NVRAM_READ_OUT_READ_BUFFER), length);
+ return 0;
+
+fail:
+ netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+ return rc;
+}
+
+static int efx_mcdi_nvram_write(struct efx_nic *efx, unsigned int type,
+ loff_t offset, const u8 *buffer, size_t length)
+{
+ MCDI_DECLARE_BUF(inbuf,
+ MC_CMD_NVRAM_WRITE_IN_LEN(EFX_MCDI_NVRAM_LEN_MAX));
+ int rc;
+
+ MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_TYPE, type);
+ MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_OFFSET, offset);
+ MCDI_SET_DWORD(inbuf, NVRAM_WRITE_IN_LENGTH, length);
+ memcpy(MCDI_PTR(inbuf, NVRAM_WRITE_IN_WRITE_BUFFER), buffer, length);
+
+ BUILD_BUG_ON(MC_CMD_NVRAM_WRITE_OUT_LEN != 0);
+
+ rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_WRITE, inbuf,
+ ALIGN(MC_CMD_NVRAM_WRITE_IN_LEN(length), 4),
+ NULL, 0, NULL);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+ return rc;
+}
+
+static int efx_mcdi_nvram_erase(struct efx_nic *efx, unsigned int type,
+ loff_t offset, size_t length)
+{
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_ERASE_IN_LEN);
+ int rc;
+
+ MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_TYPE, type);
+ MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_OFFSET, offset);
+ MCDI_SET_DWORD(inbuf, NVRAM_ERASE_IN_LENGTH, length);
+
+ BUILD_BUG_ON(MC_CMD_NVRAM_ERASE_OUT_LEN != 0);
+
+ rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_ERASE, inbuf, sizeof(inbuf),
+ NULL, 0, NULL);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+ return rc;
+}
+
+static int efx_mcdi_nvram_update_finish(struct efx_nic *efx, unsigned int type)
+{
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_NVRAM_UPDATE_FINISH_IN_LEN);
+ int rc;
+
+ MCDI_SET_DWORD(inbuf, NVRAM_UPDATE_FINISH_IN_TYPE, type);
+
+ BUILD_BUG_ON(MC_CMD_NVRAM_UPDATE_FINISH_OUT_LEN != 0);
+
+ rc = efx_mcdi_rpc(efx, MC_CMD_NVRAM_UPDATE_FINISH, inbuf, sizeof(inbuf),
+ NULL, 0, NULL);
+ if (rc)
+ goto fail;
+
+ return 0;
+
+fail:
+ netif_err(efx, hw, efx->net_dev, "%s: failed rc=%d\n", __func__, rc);
+ return rc;
+}
+
+int efx_mcdi_mtd_read(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, u8 *buffer)
+{
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ loff_t offset = start;
+ loff_t end = min_t(loff_t, start + len, mtd->size);
+ size_t chunk;
+ int rc = 0;
+
+ while (offset < end) {
+ chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
+ rc = efx_mcdi_nvram_read(efx, part->nvram_type, offset,
+ buffer, chunk);
+ if (rc)
+ goto out;
+ offset += chunk;
+ buffer += chunk;
+ }
+out:
+ *retlen = offset - start;
+ return rc;
+}
+
+int efx_mcdi_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
+{
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ loff_t offset = start & ~((loff_t)(mtd->erasesize - 1));
+ loff_t end = min_t(loff_t, start + len, mtd->size);
+ size_t chunk = part->common.mtd.erasesize;
+ int rc = 0;
+
+ if (!part->updating) {
+ rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
+ if (rc)
+ goto out;
+ part->updating = true;
+ }
+
+ /* The MCDI interface can in fact do multiple erase blocks at once;
+ * but erasing may be slow, so we make multiple calls here to avoid
+ * tripping the MCDI RPC timeout. */
+ while (offset < end) {
+ rc = efx_mcdi_nvram_erase(efx, part->nvram_type, offset,
+ chunk);
+ if (rc)
+ goto out;
+ offset += chunk;
+ }
+out:
+ return rc;
+}
+
+int efx_mcdi_mtd_write(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, const u8 *buffer)
+{
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ loff_t offset = start;
+ loff_t end = min_t(loff_t, start + len, mtd->size);
+ size_t chunk;
+ int rc = 0;
+
+ if (!part->updating) {
+ rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
+ if (rc)
+ goto out;
+ part->updating = true;
+ }
+
+ while (offset < end) {
+ chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
+ rc = efx_mcdi_nvram_write(efx, part->nvram_type, offset,
+ buffer, chunk);
+ if (rc)
+ goto out;
+ offset += chunk;
+ buffer += chunk;
+ }
+out:
+ *retlen = offset - start;
+ return rc;
+}
+
+int efx_mcdi_mtd_sync(struct mtd_info *mtd)
+{
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
+ struct efx_nic *efx = mtd->priv;
+ int rc = 0;
+
+ if (part->updating) {
+ part->updating = false;
+ rc = efx_mcdi_nvram_update_finish(efx, part->nvram_type);
+ }
+
+ return rc;
+}
+
+void efx_mcdi_mtd_rename(struct efx_mtd_partition *part)
+{
+ struct efx_mcdi_mtd_partition *mcdi_part =
+ container_of(part, struct efx_mcdi_mtd_partition, common);
+ struct efx_nic *efx = part->mtd.priv;
+
+ snprintf(part->name, sizeof(part->name), "%s %s:%02x",
+ efx->name, part->type_name, mcdi_part->fw_subtype);
+}
+
+#endif /* CONFIG_SFC_MTD */
diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h
index a465cc1..5f67ac3 100644
--- a/drivers/net/ethernet/sfc/mcdi.h
+++ b/drivers/net/ethernet/sfc/mcdi.h
@@ -65,6 +65,16 @@ struct efx_mcdi_mon {
unsigned int n_attrs;
};
+struct efx_mcdi_mtd_partition {
+ struct efx_mtd_partition common;
+ bool updating;
+ u8 nvram_type;
+ u16 fw_subtype;
+};
+
+#define to_efx_mcdi_mtd_partition(mtd) \
+ container_of(mtd, struct efx_mcdi_mtd_partition, common.mtd)
+
/**
* struct efx_mcdi_data - extra state for NICs that implement MCDI
* @iface: Interface/protocol state
@@ -250,18 +260,6 @@ extern int efx_mcdi_nvram_types(struct efx_nic *efx, u32 *nvram_types_out);
extern int efx_mcdi_nvram_info(struct efx_nic *efx, unsigned int type,
size_t *size_out, size_t *erase_size_out,
bool *protected_out);
-extern int efx_mcdi_nvram_update_start(struct efx_nic *efx,
- unsigned int type);
-extern int efx_mcdi_nvram_read(struct efx_nic *efx, unsigned int type,
- loff_t offset, u8 *buffer, size_t length);
-extern int efx_mcdi_nvram_write(struct efx_nic *efx, unsigned int type,
- loff_t offset, const u8 *buffer,
- size_t length);
-#define EFX_MCDI_NVRAM_LEN_MAX 128
-extern int efx_mcdi_nvram_erase(struct efx_nic *efx, unsigned int type,
- loff_t offset, size_t length);
-extern int efx_mcdi_nvram_update_finish(struct efx_nic *efx,
- unsigned int type);
extern int efx_mcdi_nvram_test_all(struct efx_nic *efx);
extern int efx_mcdi_handle_assertion(struct efx_nic *efx);
extern void efx_mcdi_set_id_led(struct efx_nic *efx, enum efx_led_mode mode);
@@ -291,4 +289,14 @@ static inline int efx_mcdi_mon_probe(struct efx_nic *efx) { return 0; }
static inline void efx_mcdi_mon_remove(struct efx_nic *efx) {}
#endif
+#ifdef CONFIG_SFC_MTD
+extern int efx_mcdi_mtd_read(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, u8 *buffer);
+extern int efx_mcdi_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len);
+extern int efx_mcdi_mtd_write(struct mtd_info *mtd, loff_t start,
+ size_t len, size_t *retlen, const u8 *buffer);
+extern int efx_mcdi_mtd_sync(struct mtd_info *mtd);
+extern void efx_mcdi_mtd_rename(struct efx_mtd_partition *part);
+#endif
+
#endif /* EFX_MCDI_H */
diff --git a/drivers/net/ethernet/sfc/mtd.c b/drivers/net/ethernet/sfc/mtd.c
index f5819c7..8be9a69 100644
--- a/drivers/net/ethernet/sfc/mtd.c
+++ b/drivers/net/ethernet/sfc/mtd.c
@@ -8,167 +8,17 @@
* by the Free Software Foundation, incorporated herein by reference.
*/
-#include <linux/bitops.h>
#include <linux/module.h>
#include <linux/mtd/mtd.h>
-#include <linux/delay.h>
#include <linux/slab.h>
#include <linux/rtnetlink.h>
#include "net_driver.h"
-#include "spi.h"
#include "efx.h"
-#include "nic.h"
-#include "mcdi.h"
-#include "mcdi_pcol.h"
-
-#define FALCON_SPI_VERIFY_BUF_LEN 16
-
-struct efx_mtd_partition {
- struct list_head node;
- struct mtd_info mtd;
- const char *dev_type_name;
- const char *type_name;
- char name[IFNAMSIZ + 20];
-};
-
-struct falcon_mtd_partition {
- struct efx_mtd_partition common;
- const struct falcon_spi_device *spi;
- size_t offset;
-};
-
-struct efx_mtd_ops {
- void (*rename)(struct efx_mtd_partition *part);
- int (*read)(struct mtd_info *mtd, loff_t start, size_t len,
- size_t *retlen, u8 *buffer);
- int (*erase)(struct mtd_info *mtd, loff_t start, size_t len);
- int (*write)(struct mtd_info *mtd, loff_t start, size_t len,
- size_t *retlen, const u8 *buffer);
- int (*sync)(struct mtd_info *mtd);
-};
#define to_efx_mtd_partition(mtd) \
container_of(mtd, struct efx_mtd_partition, mtd)
-#define to_falcon_mtd_partition(mtd) \
- container_of(mtd, struct falcon_mtd_partition, common.mtd)
-
-static int falcon_mtd_probe(struct efx_nic *efx);
-static int siena_mtd_probe(struct efx_nic *efx);
-
-/* SPI utilities */
-
-static int
-falcon_spi_slow_wait(struct falcon_mtd_partition *part, bool uninterruptible)
-{
- const struct falcon_spi_device *spi = part->spi;
- struct efx_nic *efx = part->common.mtd.priv;
- u8 status;
- int rc, i;
-
- /* Wait up to 4s for flash/EEPROM to finish a slow operation. */
- for (i = 0; i < 40; i++) {
- __set_current_state(uninterruptible ?
- TASK_UNINTERRUPTIBLE : TASK_INTERRUPTIBLE);
- schedule_timeout(HZ / 10);
- rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
- &status, sizeof(status));
- if (rc)
- return rc;
- if (!(status & SPI_STATUS_NRDY))
- return 0;
- if (signal_pending(current))
- return -EINTR;
- }
- pr_err("%s: timed out waiting for %s\n",
- part->common.name, part->common.dev_type_name);
- return -ETIMEDOUT;
-}
-
-static int
-falcon_spi_unlock(struct efx_nic *efx, const struct falcon_spi_device *spi)
-{
- const u8 unlock_mask = (SPI_STATUS_BP2 | SPI_STATUS_BP1 |
- SPI_STATUS_BP0);
- u8 status;
- int rc;
-
- rc = falcon_spi_cmd(efx, spi, SPI_RDSR, -1, NULL,
- &status, sizeof(status));
- if (rc)
- return rc;
-
- if (!(status & unlock_mask))
- return 0; /* already unlocked */
-
- rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
- if (rc)
- return rc;
- rc = falcon_spi_cmd(efx, spi, SPI_SST_EWSR, -1, NULL, NULL, 0);
- if (rc)
- return rc;
-
- status &= ~unlock_mask;
- rc = falcon_spi_cmd(efx, spi, SPI_WRSR, -1, &status,
- NULL, sizeof(status));
- if (rc)
- return rc;
- rc = falcon_spi_wait_write(efx, spi);
- if (rc)
- return rc;
-
- return 0;
-}
-
-static int
-falcon_spi_erase(struct falcon_mtd_partition *part, loff_t start, size_t len)
-{
- const struct falcon_spi_device *spi = part->spi;
- struct efx_nic *efx = part->common.mtd.priv;
- unsigned pos, block_len;
- u8 empty[FALCON_SPI_VERIFY_BUF_LEN];
- u8 buffer[FALCON_SPI_VERIFY_BUF_LEN];
- int rc;
-
- if (len != spi->erase_size)
- return -EINVAL;
-
- if (spi->erase_command == 0)
- return -EOPNOTSUPP;
-
- rc = falcon_spi_unlock(efx, spi);
- if (rc)
- return rc;
- rc = falcon_spi_cmd(efx, spi, SPI_WREN, -1, NULL, NULL, 0);
- if (rc)
- return rc;
- rc = falcon_spi_cmd(efx, spi, spi->erase_command, start, NULL,
- NULL, 0);
- if (rc)
- return rc;
- rc = falcon_spi_slow_wait(part, false);
-
- /* Verify the entire region has been wiped */
- memset(empty, 0xff, sizeof(empty));
- for (pos = 0; pos < len; pos += block_len) {
- block_len = min(len - pos, sizeof(buffer));
- rc = falcon_spi_read(efx, spi, start + pos, block_len,
- NULL, buffer);
- if (rc)
- return rc;
- if (memcmp(empty, buffer, block_len))
- return -EIO;
-
- /* Avoid locking up the system */
- cond_resched();
- if (signal_pending(current))
- return -EINTR;
- }
-
- return rc;
-}
-
/* MTD interface */
static int efx_mtd_erase(struct mtd_info *mtd, struct erase_info *erase)
@@ -176,7 +26,7 @@ static int efx_mtd_erase(struct mtd_info *mtd, struct erase_info *erase)
struct efx_nic *efx = mtd->priv;
int rc;
- rc = efx->mtd_ops->erase(mtd, erase->addr, erase->len);
+ rc = efx->type->mtd_erase(mtd, erase->addr, erase->len);
if (rc == 0) {
erase->state = MTD_ERASE_DONE;
} else {
@@ -193,7 +43,7 @@ static void efx_mtd_sync(struct mtd_info *mtd)
struct efx_nic *efx = mtd->priv;
int rc;
- rc = efx->mtd_ops->sync(mtd);
+ rc = efx->type->mtd_sync(mtd);
if (rc)
pr_err("%s: %s sync failed (%d)\n",
part->name, part->dev_type_name, rc);
@@ -213,15 +63,8 @@ static void efx_mtd_remove_partition(struct efx_mtd_partition *part)
list_del(&part->node);
}
-static void efx_mtd_rename_partition(struct efx_mtd_partition *part)
-{
- struct efx_nic *efx = part->mtd.priv;
-
- efx->mtd_ops->rename(part);
-}
-
-static int efx_mtd_add(struct efx_nic *efx, struct efx_mtd_partition *parts,
- size_t n_parts, size_t sizeof_part)
+int efx_mtd_add(struct efx_nic *efx, struct efx_mtd_partition *parts,
+ size_t n_parts, size_t sizeof_part)
{
struct efx_mtd_partition *part;
size_t i;
@@ -236,11 +79,11 @@ static int efx_mtd_add(struct efx_nic *efx, struct efx_mtd_partition *parts,
part->mtd.priv = efx;
part->mtd.name = part->name;
part->mtd._erase = efx_mtd_erase;
- part->mtd._read = efx->mtd_ops->read;
- part->mtd._write = efx->mtd_ops->write;
+ part->mtd._read = efx->type->mtd_read;
+ part->mtd._write = efx->type->mtd_write;
part->mtd._sync = efx_mtd_sync;
- efx_mtd_rename_partition(part);
+ efx->type->mtd_rename(part);
if (mtd_device_register(&part->mtd, NULL, 0))
goto fail;
@@ -286,396 +129,5 @@ void efx_mtd_rename(struct efx_nic *efx)
ASSERT_RTNL();
list_for_each_entry(part, &efx->mtd_list, node)
- efx_mtd_rename_partition(part);
-}
-
-int efx_mtd_probe(struct efx_nic *efx)
-{
- if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
- return siena_mtd_probe(efx);
- else
- return falcon_mtd_probe(efx);
-}
-
-/* Implementation of MTD operations for Falcon */
-
-static void falcon_mtd_rename(struct efx_mtd_partition *part)
-{
- struct efx_nic *efx = part->mtd.priv;
-
- snprintf(part->name, sizeof(part->name), "%s %s",
- efx->name, part->type_name);
-}
-
-static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
- size_t len, size_t *retlen, u8 *buffer)
-{
- struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- struct falcon_nic_data *nic_data = efx->nic_data;
- int rc;
-
- rc = mutex_lock_interruptible(&nic_data->spi_lock);
- if (rc)
- return rc;
- rc = falcon_spi_read(efx, part->spi, part->offset + start,
- len, retlen, buffer);
- mutex_unlock(&nic_data->spi_lock);
- return rc;
-}
-
-static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
-{
- struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- struct falcon_nic_data *nic_data = efx->nic_data;
- int rc;
-
- rc = mutex_lock_interruptible(&nic_data->spi_lock);
- if (rc)
- return rc;
- rc = falcon_spi_erase(part, part->offset + start, len);
- mutex_unlock(&nic_data->spi_lock);
- return rc;
-}
-
-static int falcon_mtd_write(struct mtd_info *mtd, loff_t start,
- size_t len, size_t *retlen, const u8 *buffer)
-{
- struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- struct falcon_nic_data *nic_data = efx->nic_data;
- int rc;
-
- rc = mutex_lock_interruptible(&nic_data->spi_lock);
- if (rc)
- return rc;
- rc = falcon_spi_write(efx, part->spi, part->offset + start,
- len, retlen, buffer);
- mutex_unlock(&nic_data->spi_lock);
- return rc;
-}
-
-static int falcon_mtd_sync(struct mtd_info *mtd)
-{
- struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- struct falcon_nic_data *nic_data = efx->nic_data;
- int rc;
-
- mutex_lock(&nic_data->spi_lock);
- rc = falcon_spi_slow_wait(part, true);
- mutex_unlock(&nic_data->spi_lock);
- return rc;
-}
-
-static const struct efx_mtd_ops falcon_mtd_ops = {
- .rename = falcon_mtd_rename,
- .read = falcon_mtd_read,
- .erase = falcon_mtd_erase,
- .write = falcon_mtd_write,
- .sync = falcon_mtd_sync,
-};
-
-static int falcon_mtd_probe(struct efx_nic *efx)
-{
- struct falcon_nic_data *nic_data = efx->nic_data;
- struct falcon_mtd_partition *parts;
- struct falcon_spi_device *spi;
- size_t n_parts;
- int rc = -ENODEV;
-
- ASSERT_RTNL();
-
- efx->mtd_ops = &falcon_mtd_ops;
-
- /* Allocate space for maximum number of partitions */
- parts = kcalloc(2, sizeof(*parts), GFP_KERNEL);
- n_parts = 0;
-
- spi = &nic_data->spi_flash;
- if (falcon_spi_present(spi) && spi->size > FALCON_FLASH_BOOTCODE_START) {
- parts[n_parts].spi = spi;
- parts[n_parts].offset = FALCON_FLASH_BOOTCODE_START;
- parts[n_parts].common.dev_type_name = "flash";
- parts[n_parts].common.type_name = "sfc_flash_bootrom";
- parts[n_parts].common.mtd.type = MTD_NORFLASH;
- parts[n_parts].common.mtd.flags = MTD_CAP_NORFLASH;
- parts[n_parts].common.mtd.size = spi->size - FALCON_FLASH_BOOTCODE_START;
- parts[n_parts].common.mtd.erasesize = spi->erase_size;
- n_parts++;
- }
-
- spi = &nic_data->spi_eeprom;
- if (falcon_spi_present(spi) && spi->size > FALCON_EEPROM_BOOTCONFIG_START) {
- parts[n_parts].spi = spi;
- parts[n_parts].offset = FALCON_EEPROM_BOOTCONFIG_START;
- parts[n_parts].common.dev_type_name = "EEPROM";
- parts[n_parts].common.type_name = "sfc_bootconfig";
- parts[n_parts].common.mtd.type = MTD_RAM;
- parts[n_parts].common.mtd.flags = MTD_CAP_RAM;
- parts[n_parts].common.mtd.size =
- min(spi->size, FALCON_EEPROM_BOOTCONFIG_END) -
- FALCON_EEPROM_BOOTCONFIG_START;
- parts[n_parts].common.mtd.erasesize = spi->erase_size;
- n_parts++;
- }
-
- rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
- if (rc)
- kfree(parts);
- return rc;
-}
-
-/* Implementation of MTD operations for Siena */
-
-struct efx_mcdi_mtd_partition {
- struct efx_mtd_partition common;
- bool updating;
- u8 nvram_type;
- u16 fw_subtype;
-};
-
-#define to_efx_mcdi_mtd_partition(mtd) \
- container_of(mtd, struct efx_mcdi_mtd_partition, common.mtd)
-
-static void siena_mtd_rename(struct efx_mtd_partition *part)
-{
- struct efx_mcdi_mtd_partition *mcdi_part =
- container_of(part, struct efx_mcdi_mtd_partition, common);
- struct efx_nic *efx = part->mtd.priv;
-
- snprintf(part->name, sizeof(part->name), "%s %s:%02x",
- efx->name, part->type_name, mcdi_part->fw_subtype);
-}
-
-static int siena_mtd_read(struct mtd_info *mtd, loff_t start,
- size_t len, size_t *retlen, u8 *buffer)
-{
- struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- loff_t offset = start;
- loff_t end = min_t(loff_t, start + len, mtd->size);
- size_t chunk;
- int rc = 0;
-
- while (offset < end) {
- chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
- rc = efx_mcdi_nvram_read(efx, part->nvram_type, offset,
- buffer, chunk);
- if (rc)
- goto out;
- offset += chunk;
- buffer += chunk;
- }
-out:
- *retlen = offset - start;
- return rc;
-}
-
-static int siena_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
-{
- struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- loff_t offset = start & ~((loff_t)(mtd->erasesize - 1));
- loff_t end = min_t(loff_t, start + len, mtd->size);
- size_t chunk = part->common.mtd.erasesize;
- int rc = 0;
-
- if (!part->updating) {
- rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
- if (rc)
- goto out;
- part->updating = true;
- }
-
- /* The MCDI interface can in fact do multiple erase blocks at once;
- * but erasing may be slow, so we make multiple calls here to avoid
- * tripping the MCDI RPC timeout. */
- while (offset < end) {
- rc = efx_mcdi_nvram_erase(efx, part->nvram_type, offset,
- chunk);
- if (rc)
- goto out;
- offset += chunk;
- }
-out:
- return rc;
-}
-
-static int siena_mtd_write(struct mtd_info *mtd, loff_t start,
- size_t len, size_t *retlen, const u8 *buffer)
-{
- struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- loff_t offset = start;
- loff_t end = min_t(loff_t, start + len, mtd->size);
- size_t chunk;
- int rc = 0;
-
- if (!part->updating) {
- rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
- if (rc)
- goto out;
- part->updating = true;
- }
-
- while (offset < end) {
- chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
- rc = efx_mcdi_nvram_write(efx, part->nvram_type, offset,
- buffer, chunk);
- if (rc)
- goto out;
- offset += chunk;
- buffer += chunk;
- }
-out:
- *retlen = offset - start;
- return rc;
-}
-
-static int siena_mtd_sync(struct mtd_info *mtd)
-{
- struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
- struct efx_nic *efx = mtd->priv;
- int rc = 0;
-
- if (part->updating) {
- part->updating = false;
- rc = efx_mcdi_nvram_update_finish(efx, part->nvram_type);
- }
-
- return rc;
+ efx->type->mtd_rename(part);
}
-
-static const struct efx_mtd_ops siena_mtd_ops = {
- .rename = siena_mtd_rename,
- .read = siena_mtd_read,
- .erase = siena_mtd_erase,
- .write = siena_mtd_write,
- .sync = siena_mtd_sync,
-};
-
-struct siena_nvram_type_info {
- int port;
- const char *name;
-};
-
-static const struct siena_nvram_type_info siena_nvram_types[] = {
- [MC_CMD_NVRAM_TYPE_DISABLED_CALLISTO] = { 0, "sfc_dummy_phy" },
- [MC_CMD_NVRAM_TYPE_MC_FW] = { 0, "sfc_mcfw" },
- [MC_CMD_NVRAM_TYPE_MC_FW_BACKUP] = { 0, "sfc_mcfw_backup" },
- [MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT0] = { 0, "sfc_static_cfg" },
- [MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT1] = { 1, "sfc_static_cfg" },
- [MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT0] = { 0, "sfc_dynamic_cfg" },
- [MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT1] = { 1, "sfc_dynamic_cfg" },
- [MC_CMD_NVRAM_TYPE_EXP_ROM] = { 0, "sfc_exp_rom" },
- [MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT0] = { 0, "sfc_exp_rom_cfg" },
- [MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT1] = { 1, "sfc_exp_rom_cfg" },
- [MC_CMD_NVRAM_TYPE_PHY_PORT0] = { 0, "sfc_phy_fw" },
- [MC_CMD_NVRAM_TYPE_PHY_PORT1] = { 1, "sfc_phy_fw" },
- [MC_CMD_NVRAM_TYPE_FPGA] = { 0, "sfc_fpga" },
-};
-
-static int siena_mtd_probe_partition(struct efx_nic *efx,
- struct efx_mcdi_mtd_partition *part,
- unsigned int type)
-{
- const struct siena_nvram_type_info *info;
- size_t size, erase_size;
- bool protected;
- int rc;
-
- if (type >= ARRAY_SIZE(siena_nvram_types) ||
- siena_nvram_types[type].name == NULL)
- return -ENODEV;
-
- info = &siena_nvram_types[type];
-
- if (info->port != efx_port_num(efx))
- return -ENODEV;
-
- rc = efx_mcdi_nvram_info(efx, type, &size, &erase_size, &protected);
- if (rc)
- return rc;
- if (protected)
- return -ENODEV; /* hide it */
-
- part->nvram_type = type;
- part->common.dev_type_name = "Siena NVRAM manager";
- part->common.type_name = info->name;
-
- part->common.mtd.type = MTD_NORFLASH;
- part->common.mtd.flags = MTD_CAP_NORFLASH;
- part->common.mtd.size = size;
- part->common.mtd.erasesize = erase_size;
-
- return 0;
-}
-
-static int siena_mtd_get_fw_subtypes(struct efx_nic *efx,
- struct efx_mcdi_mtd_partition *parts,
- size_t n_parts)
-{
- uint16_t fw_subtype_list[
- MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_MAXNUM];
- size_t i;
- int rc;
-
- rc = efx_mcdi_get_board_cfg(efx, NULL, fw_subtype_list, NULL);
- if (rc)
- return rc;
-
- for (i = 0; i < n_parts; i++)
- parts[i].fw_subtype = fw_subtype_list[parts[i].nvram_type];
-
- return 0;
-}
-
-static int siena_mtd_probe(struct efx_nic *efx)
-{
- struct efx_mcdi_mtd_partition *parts;
- u32 nvram_types;
- unsigned int type;
- size_t n_parts;
- int rc;
-
- ASSERT_RTNL();
-
- efx->mtd_ops = &siena_mtd_ops;
-
- rc = efx_mcdi_nvram_types(efx, &nvram_types);
- if (rc)
- return rc;
-
- parts = kcalloc(hweight32(nvram_types), sizeof(*parts), GFP_KERNEL);
- if (!parts)
- return -ENOMEM;
-
- type = 0;
- n_parts = 0;
-
- while (nvram_types != 0) {
- if (nvram_types & 1) {
- rc = siena_mtd_probe_partition(efx, &parts[n_parts],
- type);
- if (rc == 0)
- n_parts++;
- else if (rc != -ENODEV)
- goto fail;
- }
- type++;
- nvram_types >>= 1;
- }
-
- rc = siena_mtd_get_fw_subtypes(efx, parts, n_parts);
- if (rc)
- goto fail;
-
- rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
-fail:
- if (rc)
- kfree(parts);
- return rc;
-}
-
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 958ede1..fa881cd 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -27,6 +27,7 @@
#include <linux/mutex.h>
#include <linux/vmalloc.h>
#include <linux/i2c.h>
+#include <linux/mtd/mtd.h>
#include "enum.h"
#include "bitfield.h"
@@ -868,7 +869,6 @@ struct efx_nic {
struct delayed_work selftest_work;
#ifdef CONFIG_SFC_MTD
- const struct efx_mtd_ops *mtd_ops;
struct list_head mtd_list;
#endif
@@ -954,6 +954,14 @@ static inline unsigned int efx_port_num(struct efx_nic *efx)
return efx->port_num;
}
+struct efx_mtd_partition {
+ struct list_head node;
+ struct mtd_info mtd;
+ const char *dev_type_name;
+ const char *type_name;
+ char name[IFNAMSIZ + 20];
+};
+
/**
* struct efx_nic_type - Efx device type definition
* @mem_map_size: Get memory BAR mapped size
@@ -1047,6 +1055,15 @@ static inline unsigned int efx_port_num(struct efx_nic *efx)
* @filter_rfs_expire_one: Consider expiring a filter inserted for RFS.
* This must check whether the specified table entry is used by RFS
* and that rps_may_expire_flow() returns true for it.
+ * @mtd_probe: Probe and add MTD partitions associated with this net device,
+ * using efx_mtd_add()
+ * @mtd_rename: Set an MTD partition name using the net device name
+ * @mtd_read: Read from an MTD partition
+ * @mtd_erase: Erase part of an MTD partition
+ * @mtd_write: Write to an MTD partition
+ * @mtd_sync: Wait for write-back to complete on MTD partition. This
+ * also notifies the driver that a writer has finished using this
+ * partition.
* @revision: Hardware architecture revision
* @txd_ptr_tbl_base: TX descriptor ring base address
* @rxd_ptr_tbl_base: RX descriptor ring base address
@@ -1150,6 +1167,16 @@ struct efx_nic_type {
bool (*filter_rfs_expire_one)(struct efx_nic *efx, u32 flow_id,
unsigned int index);
#endif
+#ifdef CONFIG_SFC_MTD
+ int (*mtd_probe)(struct efx_nic *efx);
+ void (*mtd_rename)(struct efx_mtd_partition *part);
+ int (*mtd_read)(struct mtd_info *mtd, loff_t start, size_t len,
+ size_t *retlen, u8 *buffer);
+ int (*mtd_erase)(struct mtd_info *mtd, loff_t start, size_t len);
+ int (*mtd_write)(struct mtd_info *mtd, loff_t start, size_t len,
+ size_t *retlen, const u8 *buffer);
+ int (*mtd_sync)(struct mtd_info *mtd);
+#endif
int revision;
unsigned int txd_ptr_tbl_base;
diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h
index b90dc8a..2b84aeb 100644
--- a/drivers/net/ethernet/sfc/nic.h
+++ b/drivers/net/ethernet/sfc/nic.h
@@ -16,7 +16,6 @@
#include "net_driver.h"
#include "efx.h"
#include "mcdi.h"
-#include "spi.h"
/*
* Falcon hardware control
@@ -164,6 +163,38 @@ struct falcon_board {
};
/**
+ * struct falcon_spi_device - a Falcon SPI (Serial Peripheral Interface) device
+ * @device_id: Controller's id for the device
+ * @size: Size (in bytes)
+ * @addr_len: Number of address bytes in read/write commands
+ * @munge_address: Flag whether addresses should be munged.
+ * Some devices with 9-bit addresses (e.g. AT25040A EEPROM)
+ * use bit 3 of the command byte as address bit A8, rather
+ * than having a two-byte address. If this flag is set, then
+ * commands should be munged in this way.
+ * @erase_command: Erase command (or 0 if sector erase not needed).
+ * @erase_size: Erase sector size (in bytes)
+ * Erase commands affect sectors with this size and alignment.
+ * This must be a power of two.
+ * @block_size: Write block size (in bytes).
+ * Write commands are limited to blocks with this size and alignment.
+ */
+struct falcon_spi_device {
+ int device_id;
+ unsigned int size;
+ unsigned int addr_len;
+ unsigned int munge_address:1;
+ u8 erase_command;
+ unsigned int erase_size;
+ unsigned int block_size;
+};
+
+static inline bool falcon_spi_present(const struct falcon_spi_device *spi)
+{
+ return spi->size != 0;
+}
+
+/**
* struct falcon_nic_data - Falcon NIC state
* @pci_dev2: Secondary function of Falcon A
* @board: Board state and functions
diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c
index 1be81e4..11b4739 100644
--- a/drivers/net/ethernet/sfc/siena.c
+++ b/drivers/net/ethernet/sfc/siena.c
@@ -18,7 +18,6 @@
#include "bitfield.h"
#include "efx.h"
#include "nic.h"
-#include "spi.h"
#include "farch_regs.h"
#include "io.h"
#include "phy.h"
@@ -674,6 +673,138 @@ static int siena_mcdi_poll_reboot(struct efx_nic *efx)
/**************************************************************************
*
+ * MTD
+ *
+ **************************************************************************
+ */
+
+#ifdef CONFIG_SFC_MTD
+
+struct siena_nvram_type_info {
+ int port;
+ const char *name;
+};
+
+static const struct siena_nvram_type_info siena_nvram_types[] = {
+ [MC_CMD_NVRAM_TYPE_DISABLED_CALLISTO] = { 0, "sfc_dummy_phy" },
+ [MC_CMD_NVRAM_TYPE_MC_FW] = { 0, "sfc_mcfw" },
+ [MC_CMD_NVRAM_TYPE_MC_FW_BACKUP] = { 0, "sfc_mcfw_backup" },
+ [MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT0] = { 0, "sfc_static_cfg" },
+ [MC_CMD_NVRAM_TYPE_STATIC_CFG_PORT1] = { 1, "sfc_static_cfg" },
+ [MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT0] = { 0, "sfc_dynamic_cfg" },
+ [MC_CMD_NVRAM_TYPE_DYNAMIC_CFG_PORT1] = { 1, "sfc_dynamic_cfg" },
+ [MC_CMD_NVRAM_TYPE_EXP_ROM] = { 0, "sfc_exp_rom" },
+ [MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT0] = { 0, "sfc_exp_rom_cfg" },
+ [MC_CMD_NVRAM_TYPE_EXP_ROM_CFG_PORT1] = { 1, "sfc_exp_rom_cfg" },
+ [MC_CMD_NVRAM_TYPE_PHY_PORT0] = { 0, "sfc_phy_fw" },
+ [MC_CMD_NVRAM_TYPE_PHY_PORT1] = { 1, "sfc_phy_fw" },
+ [MC_CMD_NVRAM_TYPE_FPGA] = { 0, "sfc_fpga" },
+};
+
+static int siena_mtd_probe_partition(struct efx_nic *efx,
+ struct efx_mcdi_mtd_partition *part,
+ unsigned int type)
+{
+ const struct siena_nvram_type_info *info;
+ size_t size, erase_size;
+ bool protected;
+ int rc;
+
+ if (type >= ARRAY_SIZE(siena_nvram_types) ||
+ siena_nvram_types[type].name == NULL)
+ return -ENODEV;
+
+ info = &siena_nvram_types[type];
+
+ if (info->port != efx_port_num(efx))
+ return -ENODEV;
+
+ rc = efx_mcdi_nvram_info(efx, type, &size, &erase_size, &protected);
+ if (rc)
+ return rc;
+ if (protected)
+ return -ENODEV; /* hide it */
+
+ part->nvram_type = type;
+ part->common.dev_type_name = "Siena NVRAM manager";
+ part->common.type_name = info->name;
+
+ part->common.mtd.type = MTD_NORFLASH;
+ part->common.mtd.flags = MTD_CAP_NORFLASH;
+ part->common.mtd.size = size;
+ part->common.mtd.erasesize = erase_size;
+
+ return 0;
+}
+
+static int siena_mtd_get_fw_subtypes(struct efx_nic *efx,
+ struct efx_mcdi_mtd_partition *parts,
+ size_t n_parts)
+{
+ uint16_t fw_subtype_list[
+ MC_CMD_GET_BOARD_CFG_OUT_FW_SUBTYPE_LIST_MAXNUM];
+ size_t i;
+ int rc;
+
+ rc = efx_mcdi_get_board_cfg(efx, NULL, fw_subtype_list, NULL);
+ if (rc)
+ return rc;
+
+ for (i = 0; i < n_parts; i++)
+ parts[i].fw_subtype = fw_subtype_list[parts[i].nvram_type];
+
+ return 0;
+}
+
+static int siena_mtd_probe(struct efx_nic *efx)
+{
+ struct efx_mcdi_mtd_partition *parts;
+ u32 nvram_types;
+ unsigned int type;
+ size_t n_parts;
+ int rc;
+
+ ASSERT_RTNL();
+
+ rc = efx_mcdi_nvram_types(efx, &nvram_types);
+ if (rc)
+ return rc;
+
+ parts = kcalloc(hweight32(nvram_types), sizeof(*parts), GFP_KERNEL);
+ if (!parts)
+ return -ENOMEM;
+
+ type = 0;
+ n_parts = 0;
+
+ while (nvram_types != 0) {
+ if (nvram_types & 1) {
+ rc = siena_mtd_probe_partition(efx, &parts[n_parts],
+ type);
+ if (rc == 0)
+ n_parts++;
+ else if (rc != -ENODEV)
+ goto fail;
+ }
+ type++;
+ nvram_types >>= 1;
+ }
+
+ rc = siena_mtd_get_fw_subtypes(efx, parts, n_parts);
+ if (rc)
+ goto fail;
+
+ rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
+fail:
+ if (rc)
+ kfree(parts);
+ return rc;
+}
+
+#endif /* CONFIG_SFC_MTD */
+
+/**************************************************************************
+ *
* Revision-dependent attributes used by efx.c and nic.c
*
**************************************************************************
@@ -753,6 +884,14 @@ const struct efx_nic_type siena_a0_nic_type = {
.filter_rfs_insert = efx_farch_filter_rfs_insert,
.filter_rfs_expire_one = efx_farch_filter_rfs_expire_one,
#endif
+#ifdef CONFIG_SFC_MTD
+ .mtd_probe = siena_mtd_probe,
+ .mtd_rename = efx_mcdi_mtd_rename,
+ .mtd_read = efx_mcdi_mtd_read,
+ .mtd_erase = efx_mcdi_mtd_erase,
+ .mtd_write = efx_mcdi_mtd_write,
+ .mtd_sync = efx_mcdi_mtd_sync,
+#endif
.revision = EFX_REV_SIENA_A0,
.txd_ptr_tbl_base = FR_BZ_TX_DESC_PTR_TBL,
diff --git a/drivers/net/ethernet/sfc/spi.h b/drivers/net/ethernet/sfc/spi.h
deleted file mode 100644
index ee951feb..0000000
--- a/drivers/net/ethernet/sfc/spi.h
+++ /dev/null
@@ -1,99 +0,0 @@
-/****************************************************************************
- * Driver for Solarflare Solarstorm network controllers and boards
- * Copyright 2005 Fen Systems Ltd.
- * Copyright 2006-2010 Solarflare Communications Inc.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published
- * by the Free Software Foundation, incorporated herein by reference.
- */
-
-#ifndef EFX_SPI_H
-#define EFX_SPI_H
-
-#include "net_driver.h"
-
-/**************************************************************************
- *
- * Basic SPI command set and bit definitions
- *
- *************************************************************************/
-
-#define SPI_WRSR 0x01 /* Write status register */
-#define SPI_WRITE 0x02 /* Write data to memory array */
-#define SPI_READ 0x03 /* Read data from memory array */
-#define SPI_WRDI 0x04 /* Reset write enable latch */
-#define SPI_RDSR 0x05 /* Read status register */
-#define SPI_WREN 0x06 /* Set write enable latch */
-#define SPI_SST_EWSR 0x50 /* SST: Enable write to status register */
-
-#define SPI_STATUS_WPEN 0x80 /* Write-protect pin enabled */
-#define SPI_STATUS_BP2 0x10 /* Block protection bit 2 */
-#define SPI_STATUS_BP1 0x08 /* Block protection bit 1 */
-#define SPI_STATUS_BP0 0x04 /* Block protection bit 0 */
-#define SPI_STATUS_WEN 0x02 /* State of the write enable latch */
-#define SPI_STATUS_NRDY 0x01 /* Device busy flag */
-
-/**
- * struct falcon_spi_device - a Falcon SPI (Serial Peripheral Interface) device
- * @device_id: Controller's id for the device
- * @size: Size (in bytes)
- * @addr_len: Number of address bytes in read/write commands
- * @munge_address: Flag whether addresses should be munged.
- * Some devices with 9-bit addresses (e.g. AT25040A EEPROM)
- * use bit 3 of the command byte as address bit A8, rather
- * than having a two-byte address. If this flag is set, then
- * commands should be munged in this way.
- * @erase_command: Erase command (or 0 if sector erase not needed).
- * @erase_size: Erase sector size (in bytes)
- * Erase commands affect sectors with this size and alignment.
- * This must be a power of two.
- * @block_size: Write block size (in bytes).
- * Write commands are limited to blocks with this size and alignment.
- */
-struct falcon_spi_device {
- int device_id;
- unsigned int size;
- unsigned int addr_len;
- unsigned int munge_address:1;
- u8 erase_command;
- unsigned int erase_size;
- unsigned int block_size;
-};
-
-static inline bool falcon_spi_present(const struct falcon_spi_device *spi)
-{
- return spi->size != 0;
-}
-
-int falcon_spi_cmd(struct efx_nic *efx,
- const struct falcon_spi_device *spi, unsigned int command,
- int address, const void *in, void *out, size_t len);
-int falcon_spi_wait_write(struct efx_nic *efx,
- const struct falcon_spi_device *spi);
-int falcon_spi_read(struct efx_nic *efx,
- const struct falcon_spi_device *spi, loff_t start,
- size_t len, size_t *retlen, u8 *buffer);
-int falcon_spi_write(struct efx_nic *efx,
- const struct falcon_spi_device *spi, loff_t start,
- size_t len, size_t *retlen, const u8 *buffer);
-
-/*
- * SFC4000 flash is partitioned into:
- * 0-0x400 chip and board config (see falcon_hwdefs.h)
- * 0x400-0x8000 unused (or may contain VPD if EEPROM not present)
- * 0x8000-end boot code (mapped to PCI expansion ROM)
- * SFC4000 small EEPROM (size < 0x400) is used for VPD only.
- * SFC4000 large EEPROM (size >= 0x400) is partitioned into:
- * 0-0x400 chip and board config
- * configurable VPD
- * 0x800-0x1800 boot config
- * Aside from the chip and board config, all of these are optional and may
- * be absent or truncated depending on the devices used.
- */
-#define FALCON_NVCONFIG_END 0x400U
-#define FALCON_FLASH_BOOTCODE_START 0x8000U
-#define FALCON_EEPROM_BOOTCONFIG_START 0x800U
-#define FALCON_EEPROM_BOOTCONFIG_END 0x1800U
-
-#endif /* EFX_SPI_H */
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* [PATCH net-next 01/16] sfc: Move NIC-type-specific MTD partition date into separate structures
From: Ben Hutchings @ 2013-08-28 15:16 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
In-Reply-To: <1377702837.2264.21.camel@bwh-desktop.uk.level5networks.com>
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
---
drivers/net/ethernet/sfc/mtd.c | 197 +++++++++++++++++++++++------------------
1 file changed, 112 insertions(+), 85 deletions(-)
diff --git a/drivers/net/ethernet/sfc/mtd.c b/drivers/net/ethernet/sfc/mtd.c
index ba6c87a..f5819c7 100644
--- a/drivers/net/ethernet/sfc/mtd.c
+++ b/drivers/net/ethernet/sfc/mtd.c
@@ -27,23 +27,19 @@
struct efx_mtd_partition {
struct list_head node;
struct mtd_info mtd;
- union {
- struct {
- bool updating;
- u8 nvram_type;
- u16 fw_subtype;
- } mcdi;
- struct {
- const struct falcon_spi_device *spi;
- size_t offset;
- } falcon;
- };
const char *dev_type_name;
const char *type_name;
char name[IFNAMSIZ + 20];
};
+struct falcon_mtd_partition {
+ struct efx_mtd_partition common;
+ const struct falcon_spi_device *spi;
+ size_t offset;
+};
+
struct efx_mtd_ops {
+ void (*rename)(struct efx_mtd_partition *part);
int (*read)(struct mtd_info *mtd, loff_t start, size_t len,
size_t *retlen, u8 *buffer);
int (*erase)(struct mtd_info *mtd, loff_t start, size_t len);
@@ -55,16 +51,19 @@ struct efx_mtd_ops {
#define to_efx_mtd_partition(mtd) \
container_of(mtd, struct efx_mtd_partition, mtd)
+#define to_falcon_mtd_partition(mtd) \
+ container_of(mtd, struct falcon_mtd_partition, common.mtd)
+
static int falcon_mtd_probe(struct efx_nic *efx);
static int siena_mtd_probe(struct efx_nic *efx);
/* SPI utilities */
static int
-falcon_spi_slow_wait(struct efx_mtd_partition *part, bool uninterruptible)
+falcon_spi_slow_wait(struct falcon_mtd_partition *part, bool uninterruptible)
{
- const struct falcon_spi_device *spi = part->falcon.spi;
- struct efx_nic *efx = part->mtd.priv;
+ const struct falcon_spi_device *spi = part->spi;
+ struct efx_nic *efx = part->common.mtd.priv;
u8 status;
int rc, i;
@@ -83,7 +82,7 @@ falcon_spi_slow_wait(struct efx_mtd_partition *part, bool uninterruptible)
return -EINTR;
}
pr_err("%s: timed out waiting for %s\n",
- part->name, part->dev_type_name);
+ part->common.name, part->common.dev_type_name);
return -ETIMEDOUT;
}
@@ -123,10 +122,10 @@ falcon_spi_unlock(struct efx_nic *efx, const struct falcon_spi_device *spi)
}
static int
-falcon_spi_erase(struct efx_mtd_partition *part, loff_t start, size_t len)
+falcon_spi_erase(struct falcon_mtd_partition *part, loff_t start, size_t len)
{
- const struct falcon_spi_device *spi = part->falcon.spi;
- struct efx_nic *efx = part->mtd.priv;
+ const struct falcon_spi_device *spi = part->spi;
+ struct efx_nic *efx = part->common.mtd.priv;
unsigned pos, block_len;
u8 empty[FALCON_SPI_VERIFY_BUF_LEN];
u8 buffer[FALCON_SPI_VERIFY_BUF_LEN];
@@ -218,22 +217,18 @@ static void efx_mtd_rename_partition(struct efx_mtd_partition *part)
{
struct efx_nic *efx = part->mtd.priv;
- if (efx_nic_rev(efx) >= EFX_REV_SIENA_A0)
- snprintf(part->name, sizeof(part->name), "%s %s:%02x",
- efx->name, part->type_name, part->mcdi.fw_subtype);
- else
- snprintf(part->name, sizeof(part->name), "%s %s",
- efx->name, part->type_name);
+ efx->mtd_ops->rename(part);
}
-static int efx_mtd_add(struct efx_nic *efx,
- struct efx_mtd_partition *parts, size_t n_parts)
+static int efx_mtd_add(struct efx_nic *efx, struct efx_mtd_partition *parts,
+ size_t n_parts, size_t sizeof_part)
{
struct efx_mtd_partition *part;
size_t i;
for (i = 0; i < n_parts; i++) {
- part = &parts[i];
+ part = (struct efx_mtd_partition *)((char *)parts +
+ i * sizeof_part);
part->mtd.writesize = 1;
@@ -257,8 +252,11 @@ static int efx_mtd_add(struct efx_nic *efx,
return 0;
fail:
- while (i--)
- efx_mtd_remove_partition(&parts[i]);
+ while (i--) {
+ part = (struct efx_mtd_partition *)((char *)parts +
+ i * sizeof_part);
+ efx_mtd_remove_partition(part);
+ }
/* Failure is unlikely here, but probably means we're out of memory */
return -ENOMEM;
}
@@ -301,10 +299,18 @@ int efx_mtd_probe(struct efx_nic *efx)
/* Implementation of MTD operations for Falcon */
+static void falcon_mtd_rename(struct efx_mtd_partition *part)
+{
+ struct efx_nic *efx = part->mtd.priv;
+
+ snprintf(part->name, sizeof(part->name), "%s %s",
+ efx->name, part->type_name);
+}
+
static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
size_t len, size_t *retlen, u8 *buffer)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
struct falcon_nic_data *nic_data = efx->nic_data;
int rc;
@@ -312,7 +318,7 @@ static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
rc = mutex_lock_interruptible(&nic_data->spi_lock);
if (rc)
return rc;
- rc = falcon_spi_read(efx, part->falcon.spi, part->falcon.offset + start,
+ rc = falcon_spi_read(efx, part->spi, part->offset + start,
len, retlen, buffer);
mutex_unlock(&nic_data->spi_lock);
return rc;
@@ -320,7 +326,7 @@ static int falcon_mtd_read(struct mtd_info *mtd, loff_t start,
static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
struct falcon_nic_data *nic_data = efx->nic_data;
int rc;
@@ -328,7 +334,7 @@ static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
rc = mutex_lock_interruptible(&nic_data->spi_lock);
if (rc)
return rc;
- rc = falcon_spi_erase(part, part->falcon.offset + start, len);
+ rc = falcon_spi_erase(part, part->offset + start, len);
mutex_unlock(&nic_data->spi_lock);
return rc;
}
@@ -336,7 +342,7 @@ static int falcon_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
static int falcon_mtd_write(struct mtd_info *mtd, loff_t start,
size_t len, size_t *retlen, const u8 *buffer)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
struct falcon_nic_data *nic_data = efx->nic_data;
int rc;
@@ -344,15 +350,15 @@ static int falcon_mtd_write(struct mtd_info *mtd, loff_t start,
rc = mutex_lock_interruptible(&nic_data->spi_lock);
if (rc)
return rc;
- rc = falcon_spi_write(efx, part->falcon.spi,
- part->falcon.offset + start, len, retlen, buffer);
+ rc = falcon_spi_write(efx, part->spi, part->offset + start,
+ len, retlen, buffer);
mutex_unlock(&nic_data->spi_lock);
return rc;
}
static int falcon_mtd_sync(struct mtd_info *mtd)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct falcon_mtd_partition *part = to_falcon_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
struct falcon_nic_data *nic_data = efx->nic_data;
int rc;
@@ -364,6 +370,7 @@ static int falcon_mtd_sync(struct mtd_info *mtd)
}
static const struct efx_mtd_ops falcon_mtd_ops = {
+ .rename = falcon_mtd_rename,
.read = falcon_mtd_read,
.erase = falcon_mtd_erase,
.write = falcon_mtd_write,
@@ -373,7 +380,7 @@ static const struct efx_mtd_ops falcon_mtd_ops = {
static int falcon_mtd_probe(struct efx_nic *efx)
{
struct falcon_nic_data *nic_data = efx->nic_data;
- struct efx_mtd_partition *parts;
+ struct falcon_mtd_partition *parts;
struct falcon_spi_device *spi;
size_t n_parts;
int rc = -ENODEV;
@@ -388,33 +395,33 @@ static int falcon_mtd_probe(struct efx_nic *efx)
spi = &nic_data->spi_flash;
if (falcon_spi_present(spi) && spi->size > FALCON_FLASH_BOOTCODE_START) {
- parts[n_parts].falcon.spi = spi;
- parts[n_parts].falcon.offset = FALCON_FLASH_BOOTCODE_START;
- parts[n_parts].dev_type_name = "flash";
- parts[n_parts].type_name = "sfc_flash_bootrom";
- parts[n_parts].mtd.type = MTD_NORFLASH;
- parts[n_parts].mtd.flags = MTD_CAP_NORFLASH;
- parts[n_parts].mtd.size = spi->size - FALCON_FLASH_BOOTCODE_START;
- parts[n_parts].mtd.erasesize = spi->erase_size;
+ parts[n_parts].spi = spi;
+ parts[n_parts].offset = FALCON_FLASH_BOOTCODE_START;
+ parts[n_parts].common.dev_type_name = "flash";
+ parts[n_parts].common.type_name = "sfc_flash_bootrom";
+ parts[n_parts].common.mtd.type = MTD_NORFLASH;
+ parts[n_parts].common.mtd.flags = MTD_CAP_NORFLASH;
+ parts[n_parts].common.mtd.size = spi->size - FALCON_FLASH_BOOTCODE_START;
+ parts[n_parts].common.mtd.erasesize = spi->erase_size;
n_parts++;
}
spi = &nic_data->spi_eeprom;
if (falcon_spi_present(spi) && spi->size > FALCON_EEPROM_BOOTCONFIG_START) {
- parts[n_parts].falcon.spi = spi;
- parts[n_parts].falcon.offset = FALCON_EEPROM_BOOTCONFIG_START;
- parts[n_parts].dev_type_name = "EEPROM";
- parts[n_parts].type_name = "sfc_bootconfig";
- parts[n_parts].mtd.type = MTD_RAM;
- parts[n_parts].mtd.flags = MTD_CAP_RAM;
- parts[n_parts].mtd.size =
+ parts[n_parts].spi = spi;
+ parts[n_parts].offset = FALCON_EEPROM_BOOTCONFIG_START;
+ parts[n_parts].common.dev_type_name = "EEPROM";
+ parts[n_parts].common.type_name = "sfc_bootconfig";
+ parts[n_parts].common.mtd.type = MTD_RAM;
+ parts[n_parts].common.mtd.flags = MTD_CAP_RAM;
+ parts[n_parts].common.mtd.size =
min(spi->size, FALCON_EEPROM_BOOTCONFIG_END) -
FALCON_EEPROM_BOOTCONFIG_START;
- parts[n_parts].mtd.erasesize = spi->erase_size;
+ parts[n_parts].common.mtd.erasesize = spi->erase_size;
n_parts++;
}
- rc = efx_mtd_add(efx, parts, n_parts);
+ rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
if (rc)
kfree(parts);
return rc;
@@ -422,10 +429,30 @@ static int falcon_mtd_probe(struct efx_nic *efx)
/* Implementation of MTD operations for Siena */
+struct efx_mcdi_mtd_partition {
+ struct efx_mtd_partition common;
+ bool updating;
+ u8 nvram_type;
+ u16 fw_subtype;
+};
+
+#define to_efx_mcdi_mtd_partition(mtd) \
+ container_of(mtd, struct efx_mcdi_mtd_partition, common.mtd)
+
+static void siena_mtd_rename(struct efx_mtd_partition *part)
+{
+ struct efx_mcdi_mtd_partition *mcdi_part =
+ container_of(part, struct efx_mcdi_mtd_partition, common);
+ struct efx_nic *efx = part->mtd.priv;
+
+ snprintf(part->name, sizeof(part->name), "%s %s:%02x",
+ efx->name, part->type_name, mcdi_part->fw_subtype);
+}
+
static int siena_mtd_read(struct mtd_info *mtd, loff_t start,
size_t len, size_t *retlen, u8 *buffer)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
loff_t offset = start;
loff_t end = min_t(loff_t, start + len, mtd->size);
@@ -434,7 +461,7 @@ static int siena_mtd_read(struct mtd_info *mtd, loff_t start,
while (offset < end) {
chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
- rc = efx_mcdi_nvram_read(efx, part->mcdi.nvram_type, offset,
+ rc = efx_mcdi_nvram_read(efx, part->nvram_type, offset,
buffer, chunk);
if (rc)
goto out;
@@ -448,25 +475,25 @@ out:
static int siena_mtd_erase(struct mtd_info *mtd, loff_t start, size_t len)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
loff_t offset = start & ~((loff_t)(mtd->erasesize - 1));
loff_t end = min_t(loff_t, start + len, mtd->size);
- size_t chunk = part->mtd.erasesize;
+ size_t chunk = part->common.mtd.erasesize;
int rc = 0;
- if (!part->mcdi.updating) {
- rc = efx_mcdi_nvram_update_start(efx, part->mcdi.nvram_type);
+ if (!part->updating) {
+ rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
if (rc)
goto out;
- part->mcdi.updating = true;
+ part->updating = true;
}
/* The MCDI interface can in fact do multiple erase blocks at once;
* but erasing may be slow, so we make multiple calls here to avoid
* tripping the MCDI RPC timeout. */
while (offset < end) {
- rc = efx_mcdi_nvram_erase(efx, part->mcdi.nvram_type, offset,
+ rc = efx_mcdi_nvram_erase(efx, part->nvram_type, offset,
chunk);
if (rc)
goto out;
@@ -479,23 +506,23 @@ out:
static int siena_mtd_write(struct mtd_info *mtd, loff_t start,
size_t len, size_t *retlen, const u8 *buffer)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
loff_t offset = start;
loff_t end = min_t(loff_t, start + len, mtd->size);
size_t chunk;
int rc = 0;
- if (!part->mcdi.updating) {
- rc = efx_mcdi_nvram_update_start(efx, part->mcdi.nvram_type);
+ if (!part->updating) {
+ rc = efx_mcdi_nvram_update_start(efx, part->nvram_type);
if (rc)
goto out;
- part->mcdi.updating = true;
+ part->updating = true;
}
while (offset < end) {
chunk = min_t(size_t, end - offset, EFX_MCDI_NVRAM_LEN_MAX);
- rc = efx_mcdi_nvram_write(efx, part->mcdi.nvram_type, offset,
+ rc = efx_mcdi_nvram_write(efx, part->nvram_type, offset,
buffer, chunk);
if (rc)
goto out;
@@ -509,19 +536,20 @@ out:
static int siena_mtd_sync(struct mtd_info *mtd)
{
- struct efx_mtd_partition *part = to_efx_mtd_partition(mtd);
+ struct efx_mcdi_mtd_partition *part = to_efx_mcdi_mtd_partition(mtd);
struct efx_nic *efx = mtd->priv;
int rc = 0;
- if (part->mcdi.updating) {
- part->mcdi.updating = false;
- rc = efx_mcdi_nvram_update_finish(efx, part->mcdi.nvram_type);
+ if (part->updating) {
+ part->updating = false;
+ rc = efx_mcdi_nvram_update_finish(efx, part->nvram_type);
}
return rc;
}
static const struct efx_mtd_ops siena_mtd_ops = {
+ .rename = siena_mtd_rename,
.read = siena_mtd_read,
.erase = siena_mtd_erase,
.write = siena_mtd_write,
@@ -550,7 +578,7 @@ static const struct siena_nvram_type_info siena_nvram_types[] = {
};
static int siena_mtd_probe_partition(struct efx_nic *efx,
- struct efx_mtd_partition *part,
+ struct efx_mcdi_mtd_partition *part,
unsigned int type)
{
const struct siena_nvram_type_info *info;
@@ -573,20 +601,20 @@ static int siena_mtd_probe_partition(struct efx_nic *efx,
if (protected)
return -ENODEV; /* hide it */
- part->mcdi.nvram_type = type;
- part->dev_type_name = "Siena NVRAM manager";
- part->type_name = info->name;
+ part->nvram_type = type;
+ part->common.dev_type_name = "Siena NVRAM manager";
+ part->common.type_name = info->name;
- part->mtd.type = MTD_NORFLASH;
- part->mtd.flags = MTD_CAP_NORFLASH;
- part->mtd.size = size;
- part->mtd.erasesize = erase_size;
+ part->common.mtd.type = MTD_NORFLASH;
+ part->common.mtd.flags = MTD_CAP_NORFLASH;
+ part->common.mtd.size = size;
+ part->common.mtd.erasesize = erase_size;
return 0;
}
static int siena_mtd_get_fw_subtypes(struct efx_nic *efx,
- struct efx_mtd_partition *parts,
+ struct efx_mcdi_mtd_partition *parts,
size_t n_parts)
{
uint16_t fw_subtype_list[
@@ -599,15 +627,14 @@ static int siena_mtd_get_fw_subtypes(struct efx_nic *efx,
return rc;
for (i = 0; i < n_parts; i++)
- parts[i].mcdi.fw_subtype =
- fw_subtype_list[parts[i].mcdi.nvram_type];
+ parts[i].fw_subtype = fw_subtype_list[parts[i].nvram_type];
return 0;
}
static int siena_mtd_probe(struct efx_nic *efx)
{
- struct efx_mtd_partition *parts;
+ struct efx_mcdi_mtd_partition *parts;
u32 nvram_types;
unsigned int type;
size_t n_parts;
@@ -645,7 +672,7 @@ static int siena_mtd_probe(struct efx_nic *efx)
if (rc)
goto fail;
- rc = efx_mtd_add(efx, parts, n_parts);
+ rc = efx_mtd_add(efx, &parts[0].common, n_parts, sizeof(*parts));
fail:
if (rc)
kfree(parts);
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply related
* Pull request: sfc-next 2013-08-28
From: Ben Hutchings @ 2013-08-28 15:13 UTC (permalink / raw)
To: David Miller; +Cc: netdev, linux-net-drivers
[-- Attachment #1: Type: text/plain, Size: 3471 bytes --]
The following changes since commit b766630b351c68c0383831dba9b81a905e5e84c6:
sfc: Eliminate struct efx_mtd (2013-08-22 19:26:04 +0100)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/bwh/sfc-next.git for-davem
for you to fetch changes up to d4fbdcfe93928fbcb7374ea490e41f7b69d95380:
sfc: Use extended MC_CMD_SENSOR_INFO and MC_CMD_READ_SENSORS (2013-08-27 22:29:56 +0100)
1. Further cleanup and refactoring in preparation for EF10.
2. Remove ethtool stats that are always zero on Falcon boards.
3. Add an ethtool stat for merged TX completions.
4. Prepare to support merged RX completions.
5. Prepare to support more hwmon sensors.
6. Add support for new events that are generated by EF10 firmware.
7. Update MC reboot detection for EF10.
Ben.
----------------------------------------------------------------
Alexandre Rames (2):
sfc: Add EF10 support for TX/RX DMA error events handling.
sfc: Return an error code when a sensor is busy.
Ben Hutchings (10):
sfc: Move NIC-type-specific MTD partition date into separate structures
sfc: Move MTD operations into efx_nic_type
sfc: Remove more left-overs from Falcon GMAC support
sfc: Remove driver-local struct ethtool_string
sfc: Delegate MAC/NIC statistic description to efx_nic_type
sfc: Remove early call to efx_nic_type::reconfigure_mac in efx_reset_up()
sfc: Rename EFX_PAGE_BLOCK_SIZE to EFX_VI_PAGE_SIZE and adjust comments
sfc: Add TX merged completion counter
sfc: Add support for reading packet length from prefix
sfc: Use extended MC_CMD_SENSOR_INFO and MC_CMD_READ_SENSORS
Daniel Pieczko (1):
sfc: use MCDI epoch flag to improve MC reboot detection in the driver
Jon Cooper (1):
sfc: Generalise packet hash lookup to support EF10 RX prefix
Laurence Evans (2):
sfc: PTP MCDI requests need to initialise periph ID field
sfc: Add a function pointer to abstract write of host time into NIC shared memory
drivers/net/ethernet/sfc/efx.c | 37 +-
drivers/net/ethernet/sfc/efx.h | 7 +-
drivers/net/ethernet/sfc/enum.h | 6 +-
drivers/net/ethernet/sfc/ethtool.c | 155 ++-------
drivers/net/ethernet/sfc/falcon.c | 616 ++++++++++++++++++++++++++--------
drivers/net/ethernet/sfc/farch.c | 6 +-
drivers/net/ethernet/sfc/farch_regs.h | 4 +
drivers/net/ethernet/sfc/io.h | 8 +-
drivers/net/ethernet/sfc/mcdi.c | 392 ++++++++++++++--------
drivers/net/ethernet/sfc/mcdi.h | 34 +-
drivers/net/ethernet/sfc/mcdi_mon.c | 146 +++++---
drivers/net/ethernet/sfc/mtd.c | 551 +-----------------------------
drivers/net/ethernet/sfc/net_driver.h | 143 ++++----
drivers/net/ethernet/sfc/nic.c | 83 +++++
drivers/net/ethernet/sfc/nic.h | 162 ++++++++-
drivers/net/ethernet/sfc/ptp.c | 10 +-
drivers/net/ethernet/sfc/rx.c | 27 +-
drivers/net/ethernet/sfc/siena.c | 373 +++++++++++++++-----
drivers/net/ethernet/sfc/spi.h | 99 ------
drivers/net/ethernet/sfc/tx.c | 3 +
20 files changed, 1536 insertions(+), 1326 deletions(-)
delete mode 100644 drivers/net/ethernet/sfc/spi.h
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 490 bytes --]
^ permalink raw reply
* Re: [PATCH net-next 2/2] sh_eth: Enable Rx descriptor word 0 shift for r8a7790
From: Sergei Shtylyov @ 2013-08-28 15:05 UTC (permalink / raw)
To: Simon Horman; +Cc: David S. Miller, netdev, linux-sh, Magnus Damm, Kouei Abe
In-Reply-To: <1377650589-20960-3-git-send-email-horms+renesas@verge.net.au>
Hello.
On 08/28/2013 04:43 AM, Simon Horman wrote:
> From: Kouei Abe <kouei.abe.cp@renesas.com>
> This corrects an oversight when r8a7790 support was added to sh_eth.
No wonder it was overseen -- it's undocumented as of version 0.60 of the
manual.
> Signed-off-by: Kouei Abe <kouei.abe.cp@renesas.com>
> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> ---
> drivers/net/ethernet/renesas/sh_eth.c | 1 +
> 1 file changed, 1 insertion(+)
> diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
> index f386a8f..095ca8e 100644
> --- a/drivers/net/ethernet/renesas/sh_eth.c
> +++ b/drivers/net/ethernet/renesas/sh_eth.c
> @@ -416,6 +416,7 @@ static struct sh_eth_cpu_data r8a7790_data = {
> .tpauser = 1,
> .hw_swap = 1,
> .rmiimode = 1,
> + .shift_rd0 = 1,
I've seen the corresponding code in a Renesas BSP but wanted to double
check before sending...
WBR, Sergei
^ permalink raw reply
* Re: [PATCH net-next v1 5/9] bonding: convert bond_has_this_ip() to use upper devices
From: Vlad Yasevich @ 2013-08-28 14:56 UTC (permalink / raw)
To: Veaceslav Falico; +Cc: Jiri Pirko, netdev, Jay Vosburgh, Andy Gospodarek
In-Reply-To: <20130827181001.GD24836@redhat.com>
On 08/27/2013 02:10 PM, Veaceslav Falico wrote:
> On Tue, Aug 27, 2013 at 01:25:29PM +0200, Jiri Pirko wrote:
>> Tue, Aug 27, 2013 at 01:16:48PM CEST, vfalico@redhat.com wrote:
>>> On Mon, Aug 26, 2013 at 10:53:38PM +0200, Jiri Pirko wrote:
>>>> Mon, Aug 26, 2013 at 10:32:38PM CEST, vfalico@redhat.com wrote:
>>> ...snip...
>>>>> + rcu_read_lock();
>>>>> + netdev_for_each_upper_dev(bond->dev, upper, iter) {
>>>>> + if (ip == bond_confirm_addr(upper, 0, ip)) {
>>>>> + ret = true;
>>>>> + break;
>>>>> + }
>>>>
>>>> You need the same recursion __vlan_find_dev_deep() is doing. If you do
>>>> not do that, you will miss anything over the first upper level.
>>>
>>> Good point, and it's true for other uses also - bond_arp_send_all(), for
>>> example, will also miss anything that's higher than the first upper
>>> level.
>>>
>>> I can't think of a use case scenario when we would need only the first
>>> upper level - so maybe we should either make netdev_for_each_upper_dev()
>>> recursive by default (I don't know how it can be done easily, tbh,
>>> without
>>> modifying the existing code), or add something like:
>>>
>>> diff --git a/net/core/dev.c b/net/core/dev.c
>>> index 566e99a..4a4468f 100644
>>> --- a/net/core/dev.c
>>> +++ b/net/core/dev.c
>>> @@ -4387,6 +4387,31 @@ static void __append_search_uppers(struct
>>> list_head *search_list,
>>> }
>>> }
>>> +struct net_device *netdev_upper_recursive_do_rcu(struct net_device
>>> *dev,
>>> + struct net_device *orig_dev,
>>> + bool (*f)(struct net_device *,
>>> + struct net_device *))
>>> +{
>>> + struct netdev_upper *upper;
>>> + struct net_device *ret = NULL;
>>> +
>>> + list_for_each_entry_rcu(upper, &dev->upper_dev_list, list) {
>>> + if (f(orig_dev, upper->dev)) {
>>> + ret = upper->dev;
>>> + break;
>>> + }
>>> +
>>> + if (!list_empty(&upper->dev->upper_dev_list)) {
>>> + ret = netdev_upper_recursive_do_rcu(upper->dev,
>>> + orig_dev, f);
>>> + if (ret)
>>> + break;
>>> + }
>>> + }
>>> +
>>> + return ret;
>>> +}
>>> +
>>> static bool __netdev_search_upper_dev(struct net_device *dev,
>>> struct net_device *upper_dev)
>>> {
>>>
>>> How do you think?
>>
>> I do not like this. How about to put all levels to upper_dev list and
>> mark those who are not direct (not level1) ? Then we can use single list
>> for all purposes.
>
> I've looked at the code a bit more and I really don't see a way to do
> non-recursive, RCUed way to traverse the whole list of upper devices.
>
> I see three ways to handle this situation:
>
> 1) The one that I've posted, recursive search and calling a provided
> function (the function should also get as a parameter a user-provided void
> *pointer). It's, indeed, a bit hacky, however will work perfectly.
>
> 2) Implementing the search (recursive) in bonding (or any further device)
> itself. Less intrusive, however it'll be code duplication actually for
> point 1).
>
> 3) Adding lower_dev_list, populating it accordingly, and also adding an int
> distance to the netdev_upper (or, with this approach, rather
> netdev_adjacent
> or something like that), which will help to implement your idea - a device
> will have lower/upper_dev_list populated with all lower/upper devices and
> their distance (i.e. distance == 1 means that it's first level of
> lower/upper device). With this approach, we might also afterwards get rid
> of slave lists from 'grouping' devices like bonding/team/bridge/etc. and,
> thus, the locking.
>
> Now I'd rather go with 1), but if you don't like it - I can go with 2).
> And, if 3) sounds ok, I can implement it also and try to get rid of bonding
> slave list (hopefully).
>
> Do you have any other ideas/thoughts?
I've been playing with approach 2 for some work I am doing for
macvtap-over-bond, but I like you idea for 3 better.
It would make things simpler in the long run.
-vlad
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [net-next 06/10] ixgbe: cleanup some log messages
From: Joe Perches @ 2013-08-28 14:49 UTC (permalink / raw)
To: Jeff Kirsher; +Cc: davem, Don Skidmore, netdev, gospo, sassmann
In-Reply-To: <1377686028-30747-7-git-send-email-jeffrey.t.kirsher@intel.com>
On Wed, 2013-08-28 at 03:33 -0700, Jeff Kirsher wrote:
> From: Don Skidmore <donald.c.skidmore@intel.com>
> Some minor log messages cleanup, changing the level one message is logged,
> adding a bit of detail to another and put all the text on one line.
Not a comment about this patch but:
> diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
[]
> @@ -1282,7 +1282,7 @@ s32 ixgbe_identify_qsfp_module_generic(struct ixgbe_hw *hw)
> status = 0;
> } else {
> if (hw->allow_unsupported_sfp == true) {
At some point it might be nice to remove the
remaining bool direct comparisons to true/false
$ git grep -E -n "(==|!=)\s*(true|false)" drivers/net/ethernet/intel
drivers/net/ethernet/intel/e1000e/netdev.c:4871: (hw->mac.autoneg == true) &&
drivers/net/ethernet/intel/e1000e/phy.c:2651: * Assumes semaphore is already acquired. When page_set==true, assumes
drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c:763: if (hw->phy.reset_disable == false) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82598.c:819: if (hw->mac.orig_link_settings_stored == false) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:954: (hw->phy.smart_speed_active == false))
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:1099: if (hw->phy.reset_disable == false && hw->phy.ops.reset != NULL)
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:1167: if (hw->mac.orig_link_settings_stored == false) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:2371: if (hw->phy.qsfp_shared_i2c_bus == true) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:2397: if (hw->phy.qsfp_shared_i2c_bus == true) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:2424: if (hw->phy.qsfp_shared_i2c_bus == true) {
drivers/net/ethernet/intel/ixgbe/ixgbe_82599.c:2450: if (hw->phy.qsfp_shared_i2c_bus == true) {
drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c:1273: if (hw->allow_unsupported_sfp == true) {
^ permalink raw reply
* Re: [PATCH net-next 0/3] xen-netback: switch to NAPI + kthread 1:1 model
From: Wei Liu @ 2013-08-28 14:41 UTC (permalink / raw)
To: David Miller
Cc: wei.liu2, netdev, xen-devel, ian.campbell, msw, konrad.wilk,
annie.li
In-Reply-To: <20130828.102821.1417197411994987475.davem@davemloft.net>
On Wed, Aug 28, 2013 at 10:28:21AM -0400, David Miller wrote:
> From: Wei Liu <wei.liu2@citrix.com>
> Date: Wed, 28 Aug 2013 10:01:35 +0100
>
> > This series has been acked by maintainer. If there is no objection we
> > would like to see it in.
>
> It's in the patchwork queue and will be applied when I get to it.
>
> Sending queriest like this to me makes no sense at all when you can
> always look at the state of your patch at patchwork.
>
> http://patchwork.ozlabs.org/project/netdev/list/
Yes, I checked their state before querying. Just that you applied some
later posted patches and I thought that you didn't notice those acks.
Sorry for the noise.
Wei.
^ permalink raw reply
* Re: A question about include/linux/socket.h
From: Sinan Akpolat @ 2013-08-28 14:35 UTC (permalink / raw)
To: Ben Hutchings; +Cc: netdev
In-Reply-To: <1377700055.2264.11.camel@bwh-desktop.uk.level5networks.com>
28.8.2013 17:27 tarihinde, Ben Hutchings yazdı:
> On Wed, 2013-08-28 at 16:45 +0300, Sinan Akpolat wrote:
>> Hi,
>>
>> I am writing CAN bus userspace code and I get a compiler error stating
>> "sa_family_t is not defined" for "include/linux/can.h". sa_family_t is
>> defined in include/linux/socket.h but it is not carried to userspace
>> kernel headers.
>>
>> When I searched about it I saw that a similar problem in
>> include/linux/netlink.h and include/linux/socket.h is fixed by the patch:
>>
>> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/include?id=6602a4baf4d1a73cc4685a39ef859e1c5ddf654c.
> You probably want commit bcb949b8847655516ba499ca75cd8529f167e360, which
> fixed include/linux/can.h.
Yeah, that's the one. Thank you very much, I couldn't find it using cgit
interface in git.kernel.org.
>
>> I also read about UAPI but I can't use newer kernels.
>>
>> I searched the kernel git repo but couldn't find a similar patch for
>> can.h. When you search for "sa_family_t" in include/linux directory you
>> can see it is used in more than 10 files and I guess none of those files
>> will compile since they won't be able to find typedef sa_family_t (just
>> like can.h).
>>
>> So my question is:
>> - Does anyone remember a patch fixing this issue?
>> - Should the typedef sa_family_t moved to userspace kernel headers? Or
>> should structs containing sa_family_t type variables moved away from
>> user space?
> No, sa_family_t is normally defined by the C library headers. But
> kernel headers generally don't include those (for good reasons). So
> kernel headers should instead define and use __kernel_sa_family_t, and
> that's what they do now (mostly).
>
> Ben.
Thanks for the info.
>
>> I have been using the file manually defining sa_family_t but I wanted to
>> ask. I am using kernel version 2.6.37 but I guess can.h file was
>> unchanged until 3.5.
^ permalink raw reply
* Re: [PATCH net-next 0/3] xen-netback: switch to NAPI + kthread 1:1 model
From: David Miller @ 2013-08-28 14:28 UTC (permalink / raw)
To: wei.liu2; +Cc: netdev, xen-devel, ian.campbell, msw, konrad.wilk, annie.li
In-Reply-To: <20130828090135.GC15729@zion.uk.xensource.com>
From: Wei Liu <wei.liu2@citrix.com>
Date: Wed, 28 Aug 2013 10:01:35 +0100
> This series has been acked by maintainer. If there is no objection we
> would like to see it in.
It's in the patchwork queue and will be applied when I get to it.
Sending queriest like this to me makes no sense at all when you can
always look at the state of your patch at patchwork.
http://patchwork.ozlabs.org/project/netdev/list/
^ permalink raw reply
* Re: A question about include/linux/socket.h
From: Ben Hutchings @ 2013-08-28 14:27 UTC (permalink / raw)
To: Sinan Akpolat; +Cc: netdev
In-Reply-To: <521DFEF9.5040701@linkas.com.tr>
On Wed, 2013-08-28 at 16:45 +0300, Sinan Akpolat wrote:
> Hi,
>
> I am writing CAN bus userspace code and I get a compiler error stating
> "sa_family_t is not defined" for "include/linux/can.h". sa_family_t is
> defined in include/linux/socket.h but it is not carried to userspace
> kernel headers.
>
> When I searched about it I saw that a similar problem in
> include/linux/netlink.h and include/linux/socket.h is fixed by the patch:
>
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/include?id=6602a4baf4d1a73cc4685a39ef859e1c5ddf654c.
You probably want commit bcb949b8847655516ba499ca75cd8529f167e360, which
fixed include/linux/can.h.
> I also read about UAPI but I can't use newer kernels.
>
> I searched the kernel git repo but couldn't find a similar patch for
> can.h. When you search for "sa_family_t" in include/linux directory you
> can see it is used in more than 10 files and I guess none of those files
> will compile since they won't be able to find typedef sa_family_t (just
> like can.h).
>
> So my question is:
> - Does anyone remember a patch fixing this issue?
> - Should the typedef sa_family_t moved to userspace kernel headers? Or
> should structs containing sa_family_t type variables moved away from
> user space?
No, sa_family_t is normally defined by the C library headers. But
kernel headers generally don't include those (for good reasons). So
kernel headers should instead define and use __kernel_sa_family_t, and
that's what they do now (mostly).
Ben.
> I have been using the file manually defining sa_family_t but I wanted to
> ask. I am using kernel version 2.6.37 but I guess can.h file was
> unchanged until 3.5.
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] ipv6: Don't depend on per socket memory for neighbour discovery messages
From: Eric Dumazet @ 2013-08-28 14:23 UTC (permalink / raw)
To: Thomas Graf; +Cc: davem, netdev
In-Reply-To: <361d4275efc79e3048191306df25ad003ed497d7.1377644830.git.tgraf@suug.ch>
On Wed, 2013-08-28 at 01:07 +0200, Thomas Graf wrote:
> Allocating skbs when sending out neighbour discovery messages
> currently uses sock_alloc_send_skb() based on a per net namespace
> socket and thus share a socket wmem buffer space.
>
> If a netdevice is temporarily unable to transmit due to carrier
> loss or for other reasons, the queued up ndisc messages will cosnume
> all of the wmem space and will thus prevent from any more skbs to
> be allocated even for netdevices that are able to transmit packets.
>
> The number of neighbour discovery messages sent is very limited,
> simply use alloc_skb() and don't depend on any socket wmem space any
> longer.
>
> This patch has orginally been posted by Eric Dumazet in a modified
> form.
>
> Signed-off-by: Thomas Graf <tgraf@suug.ch>
> Cc: Eric Dumazet <eric.dumazet@gmail.com>
> ---
Seems good to me, thanks !
Acked-by: Eric Dumazet <edumazet@google.com>
^ permalink raw reply
* Re: [PATCH net-next v2 02/13] net: add lower_dev_list to net_device and make a full mesh
From: Veaceslav Falico @ 2013-08-28 14:15 UTC (permalink / raw)
To: Jiri Pirko
Cc: netdev, David S. Miller, Eric Dumazet, Alexander Duyck, Cong Wang
In-Reply-To: <20130828135926.GA1403@minipsycho.brq.redhat.com>
On Wed, Aug 28, 2013 at 03:59:26PM +0200, Jiri Pirko wrote:
>Wed, Aug 28, 2013 at 02:02:21PM CEST, vfalico@redhat.com wrote:
>>This patch adds lower_dev_list list_head to net_device, which is the same
>>as upper_dev_list, only for lower devices, and begins to use it in the same
>>way as the upper list.
>>
>>It also changes the way the whole adjacent device lists work - now they
>>contain *all* of upper/lower devices, not only the first level. The first
>>level devices are distinguished by the bool neighbour field in
>>netdev_adjacent, also added by this patch.
>>
>>There are cases when a device can be added several times to the adjacent
>>list, the simplest would be:
>>
>> /---- eth0.10 ---\
>>eth0- --- bond0
>> \---- eth0.20 ---/
>>
>>where both bond0 and eth0 'see' each other in the adjacent lists two times.
>>To avoid duplication of netdev_adjacent structures ref_nr is being kept as
>>the number of times the device was added to the list.
>>
>>The 'full view' is achieved by adding, on link creation, all of the
>>upper_dev's upper_dev_list devices as upper devices to all of the
>>lower_dev's lower_dev_list devices (and to the lower_dev itself), and vice
>>versa. On unlink they are removed using the same logic.
>>
>>I've tested it with thousands vlans/bonds/bridges, everything works ok and
>>no observable lags even on a huge number of interfaces.
>>
>>Memory footprint for 128 devices interconnected with each other via both
>>upper and lower (which is impossible, but for the comparison) lists would be:
>>
>>128*128*2*sizeof(netdev_adjacent) = 1.5MB
>>
>>but in the real world we usualy have at most several devices with slaves
>>and a lot of vlans, so the footprint will be much lower.
>>
>>v2: new patch
>
>
>Hmm. I'm not sure if recursive searches in read paths wouldn't be better
>afterall. This looks a bit too complex
On the other hand it gives us linear speed and really easy access to upper
(and, if needed, lower) devices. And, till the end, it looks complex, but
at the bottom it's only messing with some lists :). I personally like this
approach a lot more than recursive, cause once done - it will be a pleasure
to work with.
...snip...
>> struct netdev_adjacent {
>> struct net_device *dev;
>>+ /* master device, we can only have one */
>> bool master;
>>+ /* is it first-level lower/upper or not */
>>+ bool neighbour;
>>+ /* how many times we've tried added this link. if it becomes 0 -
>>+ * we can remove the link and free the structure. For neighbour
>>+ * links it should always be 1.
>
> From this comment, I'm do not understand what is going on.
> Also for bool use "false" and "true" instead of "0" and "1".
>
>>+ */
>>+ u16 ref_nr;
This comment is about the ref_nr, I'll rephrase and rearrange it in the
next version.
...snip...
>>+#define __netdev_find_upper(d, ud) __netdev_find_adj(d, ud, true)
>>+#define __netdev_find_lower(d, ld) __netdev_find_adj(d, ld, false)
>
>Please make these 2 simple functions.
Ok.
...snip...
>> upper = list_first_entry(&dev->upper_dev_list,
>>- struct netdev_upper, list);
>>+ struct netdev_adjacent, list);
>
>I believe this should be part of previous patch. Just to be save, try to
>compile in between the patch application in series.
Jeez, I did compile it, seen this, however I've git commit --ammend-ed the
wrong commit :-/.
Sorry, will fix.
...snip...
>> upper = list_first_or_null_rcu(&dev->upper_dev_list,
>>- struct netdev_upper, list);
>>+ struct netdev_adjacent, list);
>
>Same as previous
Yup... :(
...snip...
>>+#define __netdev_upper_dev_insert(dev, adev, m, d) \
>>+ __netdev_adjacent_dev_insert(dev, adev, d, m, true)
>>+#define __netdev_lower_dev_insert(dev, adev, d) \
>>+ __netdev_adjacent_dev_insert(dev, adev, d, false, false)
>
>Also, make these functions.
Ok.
...snip...
>>+ /* try to maintain the mesh, sorry people, if you've paniced here -
>>+ * you've tried to remove an unexisting link, and that is bad.
>>+ */
>
>Unnecessary comment
Ok, will remove.
...snip...
>>+#define __netdev_upper_dev_remove(d, ud) \
>>+ __netdev_adjacent_dev_remove(d, ud, true)
>>+#define __netdev_lower_dev_remove(d, ld) \
>>+ __netdev_adjacent_dev_remove(d, ld, false)
>
>Also, make these functions.
Ok.
...snip...
>>+#define __netdev_adjacent_dev_link(d, ud) \
>>+ __netdev_adjacent_dev_insert_link(d, ud, false, false)
>>+#define __netdev_adjacent_dev_link_neighbour(d, ud, m) \
>>+ __netdev_adjacent_dev_insert_link(d, ud, m, true)
>
>Also, make these functions.
Ok.
...snip...
>>+ /* Now that we interconnected these devs, make all the upper_dev's
>>+ * upper_dev_list visible to every dev's lower_dev_list and vice
>>+ * versa, and don't forget the devices itself. All of these
>>+ * connections are non-neighbours.
>
> Please use same terms. ConnectionsXlinks
Ok, will rephrase.
Thanks a lot for the review!
^ permalink raw reply
* Re: [PATCH stable] ipv6: restrict neighbor entry creation to output flow
From: Hannes Frederic Sowa @ 2013-08-28 14:12 UTC (permalink / raw)
To: Jiri Pirko; +Cc: David Miller, mleitner, netdev, dbanerje, yoshfuji
In-Reply-To: <20130828140958.GA10015@minipsycho.brq.redhat.com>
On Wed, Aug 28, 2013 at 04:09:58PM +0200, Jiri Pirko wrote:
> Dave, is it possible to see the queue somewhere?
Here: <http://patchwork.ozlabs.org/bundle/davem/stable/?state=*>
Greetings,
Hannes
^ permalink raw reply
* Re: [PATCH stable] ipv6: restrict neighbor entry creation to output flow
From: Jiri Pirko @ 2013-08-28 14:09 UTC (permalink / raw)
To: David Miller; +Cc: hannes, mleitner, netdev, dbanerje, yoshfuji
In-Reply-To: <20130815.155454.417440388188230172.davem@davemloft.net>
Fri, Aug 16, 2013 at 12:54:54AM CEST, davem@davemloft.net wrote:
>From: Hannes Frederic Sowa <hannes@stressinduktion.org>
>Date: Wed, 14 Aug 2013 17:00:54 +0200
>
>> On Wed, Aug 14, 2013 at 10:53:27AM -0300, Marcelo Ricardo Leitner wrote:
>>> This patch is based on 3.2.y branch, the one used by reported. Please let me
>>> know if it should be different. Thanks.
>>>
>>> ---8<---
>>>
>>> Commit 0d6a77079c475033cb622c07c5a880b392ef664e introduced a regression on
>>> which routes to local delivery would not work anymore. Like this:
>>>
>>> $ ip -6 route add local 2001::/64 dev lo
>>> $ ping6 -c1 2001::9
>>> PING 2001::9(2001::9) 56 data bytes
>>> ping: sendmsg: Invalid argument
>>>
>>> As this is a local delivery, that commit would not allow the creation of a
>>> neighbor entry and thus the packet cannot be sent.
>>>
>>> But as TPROXY scenario actually needs to avoid the neighbor entry creation only
>>> for input flow, this patch now limits previous patch to input flow, keeping
>>> output as before that patch.
>>>
>>> Reported-by: Debabrata Banerjee <dbavatar@gmail.com>
>>> Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com>
>>> CC: Hannes Frederic Sowa <hannes@stressinduktion.org>
>>
>> Looks good, thanks Marcelo!
>>
>> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
>>
>> David, this patch is for all stable kernels except the 3.10 series.
>> It does not apply cleanly throughout the whole longterm kernels but the
>> changes should not be too difficult to adapt. Do you take care of this
>> or can we do something to ease this process?
>
>I've queued it up for -stable, thanks.
Dave, is it possible to see the queue somewhere?
Thanks.
Jiri
^ permalink raw reply
* Re: [PATCH 3/3] Send cgroup_path in SCM_CGROUP
From: Tejun Heo @ 2013-08-28 14:00 UTC (permalink / raw)
To: Jan Kaluza; +Cc: davem, LKML, netdev, eparis, rgb
In-Reply-To: <1377614400-27122-4-git-send-email-jkaluza@redhat.com>
Hello, Jan.
Can you please cc cgroup maintainers and mailing lists from
MAINTAINERS for cgroup related changes?
On Tue, Aug 27, 2013 at 04:40:00PM +0200, Jan Kaluza wrote:
Also, please describe what change is being made why on each patch.
Repeating the same description is fine but each patch descrption
should be more or less self-sufficient.
> +int scm_get_current_cgroup_path(char **cgroup_path)
> +{
> + int ret = 0;
> +
> + *cgroup_path = kmalloc(PAGE_SIZE, GFP_KERNEL);
PATH_MAX?
> + if (!(*cgroup_path))
> + return -ENOMEM;
> +
> + ret = task_cgroup_path(current, *cgroup_path, PAGE_SIZE);
> + if (ret < 0) {
> + kfree(*cgroup_path);
> + *cgroup_path = NULL;
> + }
> +
> + return ret;
> +}
Other than that, cgroup side looks fine to me. For cgroup bits,
Reviewed-by: Tejun Heo <tj@kernel.org>
Thanks.
--
tejun
^ permalink raw reply
* Re: [PATCH net-next v2 02/13] net: add lower_dev_list to net_device and make a full mesh
From: Jiri Pirko @ 2013-08-28 13:59 UTC (permalink / raw)
To: Veaceslav Falico
Cc: netdev, David S. Miller, Eric Dumazet, Alexander Duyck, Cong Wang
In-Reply-To: <1377691352-4146-3-git-send-email-vfalico@redhat.com>
Wed, Aug 28, 2013 at 02:02:21PM CEST, vfalico@redhat.com wrote:
>This patch adds lower_dev_list list_head to net_device, which is the same
>as upper_dev_list, only for lower devices, and begins to use it in the same
>way as the upper list.
>
>It also changes the way the whole adjacent device lists work - now they
>contain *all* of upper/lower devices, not only the first level. The first
>level devices are distinguished by the bool neighbour field in
>netdev_adjacent, also added by this patch.
>
>There are cases when a device can be added several times to the adjacent
>list, the simplest would be:
>
> /---- eth0.10 ---\
>eth0- --- bond0
> \---- eth0.20 ---/
>
>where both bond0 and eth0 'see' each other in the adjacent lists two times.
>To avoid duplication of netdev_adjacent structures ref_nr is being kept as
>the number of times the device was added to the list.
>
>The 'full view' is achieved by adding, on link creation, all of the
>upper_dev's upper_dev_list devices as upper devices to all of the
>lower_dev's lower_dev_list devices (and to the lower_dev itself), and vice
>versa. On unlink they are removed using the same logic.
>
>I've tested it with thousands vlans/bonds/bridges, everything works ok and
>no observable lags even on a huge number of interfaces.
>
>Memory footprint for 128 devices interconnected with each other via both
>upper and lower (which is impossible, but for the comparison) lists would be:
>
>128*128*2*sizeof(netdev_adjacent) = 1.5MB
>
>but in the real world we usualy have at most several devices with slaves
>and a lot of vlans, so the footprint will be much lower.
>
>v2: new patch
Hmm. I'm not sure if recursive searches in read paths wouldn't be better
afterall. This looks a bit too complex
>
>CC: "David S. Miller" <davem@davemloft.net>
>CC: Eric Dumazet <edumazet@google.com>
>CC: Jiri Pirko <jiri@resnulli.us>
>CC: Alexander Duyck <alexander.h.duyck@intel.com>
>CC: Cong Wang <amwang@redhat.com>
>Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
>---
> include/linux/netdevice.h | 1 +
> net/core/dev.c | 256 +++++++++++++++++++++++++++++++++++++++-----
> 2 files changed, 228 insertions(+), 29 deletions(-)
>
>diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>index 077363d..5ccf5b7 100644
>--- a/include/linux/netdevice.h
>+++ b/include/linux/netdevice.h
>@@ -1125,6 +1125,7 @@ struct net_device {
> struct list_head napi_list;
> struct list_head unreg_list;
> struct list_head upper_dev_list; /* List of upper devices */
>+ struct list_head lower_dev_list;
>
>
> /* currently active device features */
>diff --git a/net/core/dev.c b/net/core/dev.c
>index c2ac3b8..5d73f9f 100644
>--- a/net/core/dev.c
>+++ b/net/core/dev.c
>@@ -4369,7 +4369,15 @@ softnet_break:
>
> struct netdev_adjacent {
> struct net_device *dev;
>+ /* master device, we can only have one */
> bool master;
>+ /* is it first-level lower/upper or not */
>+ bool neighbour;
>+ /* how many times we've tried added this link. if it becomes 0 -
>+ * we can remove the link and free the structure. For neighbour
>+ * links it should always be 1.
From this comment, I'm do not understand what is going on.
Also for bool use "false" and "true" instead of "0" and "1".
>+ */
>+ u16 ref_nr;
> struct list_head list;
> struct rcu_head rcu;
> struct list_head search_list;
>@@ -4408,18 +4416,25 @@ static bool __netdev_search_upper_dev(struct net_device *dev,
> return ret;
> }
>
>-static struct netdev_adjacent *__netdev_find_upper(struct net_device *dev,
>- struct net_device *upper_dev)
>+static struct netdev_adjacent *__netdev_find_adj(struct net_device *dev,
>+ struct net_device *adj_dev,
>+ bool upper)
> {
>- struct netdev_adjacent *upper;
>+ struct netdev_adjacent *adj;
>+ struct list_head *dev_list;
>
>- list_for_each_entry(upper, &dev->upper_dev_list, list) {
>- if (upper->dev == upper_dev)
>- return upper;
>+ dev_list = upper ? &dev->upper_dev_list : &dev->lower_dev_list;
>+
>+ list_for_each_entry(adj, dev_list, list) {
>+ if (adj->dev == adj_dev)
>+ return adj;
> }
> return NULL;
> }
>
>+#define __netdev_find_upper(d, ud) __netdev_find_adj(d, ud, true)
>+#define __netdev_find_lower(d, ld) __netdev_find_adj(d, ld, false)
Please make these 2 simple functions.
>+
> /**
> * netdev_has_upper_dev - Check if device is linked to an upper device
> * @dev: device
>@@ -4470,7 +4485,7 @@ struct net_device *netdev_master_upper_dev_get(struct net_device *dev)
> return NULL;
>
> upper = list_first_entry(&dev->upper_dev_list,
>- struct netdev_upper, list);
>+ struct netdev_adjacent, list);
I believe this should be part of previous patch. Just to be save, try to
compile in between the patch application in series.
> if (likely(upper->master))
> return upper->dev;
> return NULL;
>@@ -4489,17 +4504,133 @@ struct net_device *netdev_master_upper_dev_get_rcu(struct net_device *dev)
> struct netdev_adjacent *upper;
>
> upper = list_first_or_null_rcu(&dev->upper_dev_list,
>- struct netdev_upper, list);
>+ struct netdev_adjacent, list);
Same as previous
> if (upper && likely(upper->master))
> return upper->dev;
> return NULL;
> }
> EXPORT_SYMBOL(netdev_master_upper_dev_get_rcu);
>
>+static int __netdev_adjacent_dev_insert(struct net_device *dev,
>+ struct net_device *adj_dev,
>+ bool neighbour, bool master,
>+ bool upper)
>+{
>+ struct netdev_adjacent *adj;
>+
>+ adj = __netdev_find_adj(dev, adj_dev, upper);
>+
>+ if (adj) {
>+ BUG_ON(neighbour);
>+ adj->ref_nr++;
>+ return 0;
>+ }
>+
>+ adj = kmalloc(sizeof(*adj), GFP_KERNEL);
>+ if (!adj)
>+ return -ENOMEM;
>+
>+ adj->dev = adj_dev;
>+ adj->master = master;
>+ adj->neighbour = neighbour;
>+ adj->ref_nr = 1;
>+ INIT_LIST_HEAD(&adj->search_list);
>+
>+ dev_hold(adj_dev);
>+ pr_debug("dev_hold for %s, because of %s link added from %s to %s\n",
>+ adj_dev->name, upper ? "upper" : "lower", dev->name,
>+ adj_dev->name);
>+
>+ if (!upper) {
>+ list_add_tail_rcu(&adj->list, &dev->lower_dev_list);
>+ return 0;
>+ }
>+
>+ /* Ensure that master upper link is always the first item in list. */
>+ if (master)
>+ list_add_rcu(&adj->list, &dev->upper_dev_list);
>+ else
>+ list_add_tail_rcu(&adj->list, &dev->upper_dev_list);
>+
>+ return 0;
>+}
>+
>+#define __netdev_upper_dev_insert(dev, adev, m, d) \
>+ __netdev_adjacent_dev_insert(dev, adev, d, m, true)
>+#define __netdev_lower_dev_insert(dev, adev, d) \
>+ __netdev_adjacent_dev_insert(dev, adev, d, false, false)
Also, make these functions.
>+
>+void __netdev_adjacent_dev_remove(struct net_device *dev,
>+ struct net_device *adj_dev, bool upper)
>+{
>+ struct netdev_adjacent *adj;
>+
>+ if (upper)
>+ adj = __netdev_find_upper(dev, adj_dev);
>+ else
>+ adj = __netdev_find_lower(dev, adj_dev);
>+
>+ /* try to maintain the mesh, sorry people, if you've paniced here -
>+ * you've tried to remove an unexisting link, and that is bad.
>+ */
Unnecessary comment
>+ if (!adj)
>+ BUG();
>+
>+ if (adj->ref_nr > 1) {
>+ adj->ref_nr--;
>+ return;
>+ }
>+
>+ list_del_rcu(&adj->list);
>+ pr_debug("dev_put for %s, because of %s link removed from %s to %s\n",
>+ adj_dev->name, upper ? "upper" : "lower", dev->name,
>+ adj_dev->name);
>+ dev_put(adj_dev);
>+ kfree_rcu(adj, rcu);
>+}
>+
>+#define __netdev_upper_dev_remove(d, ud) \
>+ __netdev_adjacent_dev_remove(d, ud, true)
>+#define __netdev_lower_dev_remove(d, ld) \
>+ __netdev_adjacent_dev_remove(d, ld, false)
Also, make these functions.
>+
>+int __netdev_adjacent_dev_insert_link(struct net_device *dev,
>+ struct net_device *upper_dev,
>+ bool master, bool neighbour)
>+{
>+ int ret;
>+
>+ ret = __netdev_upper_dev_insert(dev, upper_dev, master, neighbour);
>+ if (ret)
>+ return ret;
>+
>+ ret = __netdev_lower_dev_insert(upper_dev, dev, neighbour);
>+ if (ret) {
>+ __netdev_upper_dev_remove(dev, upper_dev);
>+ return ret;
>+ }
>+
>+ return 0;
>+}
>+
>+#define __netdev_adjacent_dev_link(d, ud) \
>+ __netdev_adjacent_dev_insert_link(d, ud, false, false)
>+#define __netdev_adjacent_dev_link_neighbour(d, ud, m) \
>+ __netdev_adjacent_dev_insert_link(d, ud, m, true)
Also, make these functions.
>+
>+void __netdev_adjacent_dev_unlink(struct net_device *dev,
>+ struct net_device *upper_dev)
>+{
>+ __netdev_upper_dev_remove(dev, upper_dev);
>+ __netdev_lower_dev_remove(upper_dev, dev);
>+}
>+
>+
> static int __netdev_upper_dev_link(struct net_device *dev,
> struct net_device *upper_dev, bool master)
> {
>- struct netdev_adjacent *upper;
>+ struct netdev_adjacent *i, *j, *to_i, *to_j;
>+ int ret = 0;
>
> ASSERT_RTNL();
>
>@@ -4516,22 +4647,76 @@ static int __netdev_upper_dev_link(struct net_device *dev,
> if (master && netdev_master_upper_dev_get(dev))
> return -EBUSY;
>
>- upper = kmalloc(sizeof(*upper), GFP_KERNEL);
>- if (!upper)
>- return -ENOMEM;
>+ ret = __netdev_adjacent_dev_link_neighbour(dev, upper_dev, master);
>+ if (ret)
>+ return ret;
>
>- upper->dev = upper_dev;
>- upper->master = master;
>- INIT_LIST_HEAD(&upper->search_list);
>+ /* Now that we interconnected these devs, make all the upper_dev's
>+ * upper_dev_list visible to every dev's lower_dev_list and vice
>+ * versa, and don't forget the devices itself. All of these
>+ * connections are non-neighbours.
Please use same terms. ConnectionsXlinks
>+ */
>+ list_for_each_entry(i, &upper_dev->upper_dev_list, list) {
>+ list_for_each_entry(j, &dev->lower_dev_list, list) {
>+ ret = __netdev_adjacent_dev_link(i->dev, j->dev);
>+ if (ret)
>+ goto rollback_mesh;
>+ }
>+ }
>+
>+ /* add dev to every upper_dev's upper device */
>+ list_for_each_entry(i, &upper_dev->upper_dev_list, list) {
>+ ret = __netdev_adjacent_dev_link(dev, i->dev);
>+ if (ret)
>+ goto rollback_upper_mesh;
>+ }
>+
>+ /* add upper_dev to every dev's lower device */
>+ list_for_each_entry(i, &dev->lower_dev_list, list) {
>+ ret = __netdev_adjacent_dev_link(i->dev, upper_dev);
>+ if (ret)
>+ goto rollback_lower_mesh;
>+ }
>
>- /* Ensure that master upper link is always the first item in list. */
>- if (master)
>- list_add_rcu(&upper->list, &dev->upper_dev_list);
>- else
>- list_add_tail_rcu(&upper->list, &dev->upper_dev_list);
>- dev_hold(upper_dev);
> call_netdevice_notifiers(NETDEV_CHANGEUPPER, dev);
> return 0;
>+
>+rollback_lower_mesh:
>+ to_i = i;
>+ list_for_each_entry(i, &dev->lower_dev_list, list) {
>+ if (i == to_i)
>+ break;
>+ __netdev_adjacent_dev_unlink(i->dev, upper_dev);
>+ }
>+
>+ i = NULL;
>+
>+rollback_upper_mesh:
>+ to_i = i;
>+ list_for_each_entry(i, &upper_dev->upper_dev_list, list) {
>+ if (i == to_i)
>+ break;
>+ __netdev_adjacent_dev_unlink(dev, i->dev);
>+ }
>+
>+ i = j = NULL;
>+
>+rollback_mesh:
>+ to_i = i;
>+ to_j = j;
>+ list_for_each_entry(i, &dev->lower_dev_list, list) {
>+ list_for_each_entry(j, &upper_dev->upper_dev_list, list) {
>+ if (i == to_i && j == to_j)
>+ break;
>+ __netdev_adjacent_dev_unlink(i->dev, j->dev);
>+ }
>+ if (i == to_i)
>+ break;
>+ }
>+
>+ __netdev_adjacent_dev_unlink(dev, upper_dev);
>+
>+ return ret;
> }
>
> /**
>@@ -4580,16 +4765,28 @@ EXPORT_SYMBOL(netdev_master_upper_dev_link);
> void netdev_upper_dev_unlink(struct net_device *dev,
> struct net_device *upper_dev)
> {
>- struct netdev_adjacent *upper;
>-
>+ struct netdev_adjacent *i, *j;
> ASSERT_RTNL();
>
>- upper = __netdev_find_upper(dev, upper_dev);
>- if (!upper)
>- return;
>- list_del_rcu(&upper->list);
>- dev_put(upper_dev);
>- kfree_rcu(upper, rcu);
>+ __netdev_adjacent_dev_unlink(dev, upper_dev);
>+
>+ /* Here is the tricky part. We must remove all dev's lower
>+ * devices from all upper_dev's upper devices and vice
>+ * versa, to maintain the graph relationship.
>+ */
>+ list_for_each_entry(i, &dev->lower_dev_list, list)
>+ list_for_each_entry(j, &upper_dev->upper_dev_list, list)
>+ __netdev_adjacent_dev_unlink(i->dev, j->dev);
>+
>+ /* remove also the devices itself from lower/upper device
>+ * list
>+ */
>+ list_for_each_entry(i, &dev->lower_dev_list, list)
>+ __netdev_adjacent_dev_unlink(i->dev, upper_dev);
>+
>+ list_for_each_entry(i, &upper_dev->upper_dev_list, list)
>+ __netdev_adjacent_dev_unlink(dev, i->dev);
>+
> call_netdevice_notifiers(NETDEV_CHANGEUPPER, dev);
> }
> EXPORT_SYMBOL(netdev_upper_dev_unlink);
>@@ -5850,6 +6047,7 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
> INIT_LIST_HEAD(&dev->unreg_list);
> INIT_LIST_HEAD(&dev->link_watch_list);
> INIT_LIST_HEAD(&dev->upper_dev_list);
>+ INIT_LIST_HEAD(&dev->lower_dev_list);
> dev->priv_flags = IFF_XMIT_DST_RELEASE;
> setup(dev);
>
>--
>1.7.1
>
^ permalink raw reply
* A question about include/linux/socket.h
From: Sinan Akpolat @ 2013-08-28 13:45 UTC (permalink / raw)
To: netdev
Hi,
I am writing CAN bus userspace code and I get a compiler error stating
"sa_family_t is not defined" for "include/linux/can.h". sa_family_t is
defined in include/linux/socket.h but it is not carried to userspace
kernel headers.
When I searched about it I saw that a similar problem in
include/linux/netlink.h and include/linux/socket.h is fixed by the patch:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/include?id=6602a4baf4d1a73cc4685a39ef859e1c5ddf654c.
I also read about UAPI but I can't use newer kernels.
I searched the kernel git repo but couldn't find a similar patch for
can.h. When you search for "sa_family_t" in include/linux directory you
can see it is used in more than 10 files and I guess none of those files
will compile since they won't be able to find typedef sa_family_t (just
like can.h).
So my question is:
- Does anyone remember a patch fixing this issue?
- Should the typedef sa_family_t moved to userspace kernel headers? Or
should structs containing sa_family_t type variables moved away from
user space?
I have been using the file manually defining sa_family_t but I wanted to
ask. I am using kernel version 2.6.37 but I guess can.h file was
unchanged until 3.5.
Thanks
^ permalink raw reply
* Re: [PATCH] ipv6: Don't depend on per socket memory for neighbour discovery messages
From: Hannes Frederic Sowa @ 2013-08-28 13:31 UTC (permalink / raw)
To: Thomas Graf; +Cc: davem, netdev, Eric Dumazet
In-Reply-To: <361d4275efc79e3048191306df25ad003ed497d7.1377644830.git.tgraf@suug.ch>
On Wed, Aug 28, 2013 at 01:07:25AM +0200, Thomas Graf wrote:
> Allocating skbs when sending out neighbour discovery messages
> currently uses sock_alloc_send_skb() based on a per net namespace
> socket and thus share a socket wmem buffer space.
>
> If a netdevice is temporarily unable to transmit due to carrier
> loss or for other reasons, the queued up ndisc messages will cosnume
> all of the wmem space and will thus prevent from any more skbs to
> be allocated even for netdevices that are able to transmit packets.
>
> The number of neighbour discovery messages sent is very limited,
> simply use alloc_skb() and don't depend on any socket wmem space any
> longer.
>
> This patch has orginally been posted by Eric Dumazet in a modified
> form.
>
> Signed-off-by: Thomas Graf <tgraf@suug.ch>
> Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
This fixes a real bug (I also had this issue) and thus would propose it
for stable, too.
Thanks!
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox