* [PATCH net 1/4] bnxt_en: Initialize CP doorbell value before ring allocation
2016-03-28 23:46 [PATCH net 0/4] misc. small fixes Michael Chan
@ 2016-03-28 23:46 ` Michael Chan
2016-03-28 23:46 ` [PATCH net 2/4] bnxt_en: Implement proper firmware message padding Michael Chan
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Michael Chan @ 2016-03-28 23:46 UTC (permalink / raw)
To: davem; +Cc: netdev
From: Prashant Sreedharan <prashant@broadcom.com>
The existing code does the following:
allocate completion ring
initialize completion ring doorbell
disable interrupts on this completion ring by writing to the doorbell
We can have a race where firmware sends an asynchronous event to the host
after completion ring allocation and before doorbell is initialized.
When this happens driver can crash while ringing the doorbell using
uninitialized value as part of handling the IRQ/napi request.
Signed-off-by: Prashant Sreedharan <prashant.sreedharan@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index aabbd51..c92053c 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -3391,11 +3391,11 @@ static int bnxt_hwrm_ring_alloc(struct bnxt *bp)
struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
struct bnxt_ring_struct *ring = &cpr->cp_ring_struct;
+ cpr->cp_doorbell = bp->bar1 + i * 0x80;
rc = hwrm_ring_alloc_send_msg(bp, ring, HWRM_RING_ALLOC_CMPL, i,
INVALID_STATS_CTX_ID);
if (rc)
goto err_out;
- cpr->cp_doorbell = bp->bar1 + i * 0x80;
BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);
bp->grp_info[i].cp_fw_ring_id = ring->fw_ring_id;
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 2/4] bnxt_en: Implement proper firmware message padding.
2016-03-28 23:46 [PATCH net 0/4] misc. small fixes Michael Chan
2016-03-28 23:46 ` [PATCH net 1/4] bnxt_en: Initialize CP doorbell value before ring allocation Michael Chan
@ 2016-03-28 23:46 ` Michael Chan
2016-03-28 23:46 ` [PATCH net 3/4] bnxt_en: Fix typo in bnxt_hwrm_set_pause_common() Michael Chan
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Michael Chan @ 2016-03-28 23:46 UTC (permalink / raw)
To: davem; +Cc: netdev
The size of every padded firmware message is specified in the first
HWRM_VER_GET response message. Use this value to pad every message
after that.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +++++-
drivers/net/ethernet/broadcom/bnxt/bnxt.h | 2 ++
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c92053c..4600a05 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -2653,7 +2653,7 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len,
/* Write request msg to hwrm channel */
__iowrite32_copy(bp->bar0, data, msg_len / 4);
- for (i = msg_len; i < HWRM_MAX_REQ_LEN; i += 4)
+ for (i = msg_len; i < BNXT_HWRM_MAX_REQ_LEN; i += 4)
writel(0, bp->bar0 + i);
/* currently supports only one outstanding message */
@@ -3830,6 +3830,7 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
struct hwrm_ver_get_input req = {0};
struct hwrm_ver_get_output *resp = bp->hwrm_cmd_resp_addr;
+ bp->hwrm_max_req_len = HWRM_MAX_REQ_LEN;
bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_VER_GET, -1, -1);
req.hwrm_intf_maj = HWRM_VERSION_MAJOR;
req.hwrm_intf_min = HWRM_VERSION_MINOR;
@@ -3855,6 +3856,9 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
if (!bp->hwrm_cmd_timeout)
bp->hwrm_cmd_timeout = DFLT_HWRM_CMD_TIMEOUT;
+ if (resp->hwrm_intf_maj >= 1)
+ bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len);
+
hwrm_ver_get_exit:
mutex_unlock(&bp->hwrm_cmd_lock);
return rc;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index ec04c47..709b95b 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -477,6 +477,7 @@ struct rx_tpa_end_cmp_ext {
#define RING_CMP(idx) ((idx) & bp->cp_ring_mask)
#define NEXT_CMP(idx) RING_CMP(ADV_RAW_CMP(idx, 1))
+#define BNXT_HWRM_MAX_REQ_LEN (bp->hwrm_max_req_len)
#define DFLT_HWRM_CMD_TIMEOUT 500
#define HWRM_CMD_TIMEOUT (bp->hwrm_cmd_timeout)
#define HWRM_RESET_TIMEOUT ((HWRM_CMD_TIMEOUT) * 4)
@@ -953,6 +954,7 @@ struct bnxt {
dma_addr_t hw_tx_port_stats_map;
int hw_port_stats_size;
+ u16 hwrm_max_req_len;
int hwrm_cmd_timeout;
struct mutex hwrm_cmd_lock; /* serialize hwrm messages */
struct hwrm_ver_get_output ver_resp;
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 3/4] bnxt_en: Fix typo in bnxt_hwrm_set_pause_common().
2016-03-28 23:46 [PATCH net 0/4] misc. small fixes Michael Chan
2016-03-28 23:46 ` [PATCH net 1/4] bnxt_en: Initialize CP doorbell value before ring allocation Michael Chan
2016-03-28 23:46 ` [PATCH net 2/4] bnxt_en: Implement proper firmware message padding Michael Chan
@ 2016-03-28 23:46 ` Michael Chan
2016-03-28 23:46 ` [PATCH net 4/4] bnxt_en: Fix ethtool -a reporting Michael Chan
2016-03-29 4:10 ` [PATCH net 0/4] misc. small fixes David Miller
4 siblings, 0 replies; 6+ messages in thread
From: Michael Chan @ 2016-03-28 23:46 UTC (permalink / raw)
To: davem; +Cc: netdev
The typo caused the wrong flow control bit to be set.
Reported by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 4600a05..12a009d 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4559,7 +4559,7 @@ bnxt_hwrm_set_pause_common(struct bnxt *bp, struct hwrm_port_phy_cfg_input *req)
if (bp->link_info.req_flow_ctrl & BNXT_LINK_PAUSE_RX)
req->auto_pause |= PORT_PHY_CFG_REQ_AUTO_PAUSE_RX;
if (bp->link_info.req_flow_ctrl & BNXT_LINK_PAUSE_TX)
- req->auto_pause |= PORT_PHY_CFG_REQ_AUTO_PAUSE_RX;
+ req->auto_pause |= PORT_PHY_CFG_REQ_AUTO_PAUSE_TX;
req->enables |=
cpu_to_le32(PORT_PHY_CFG_REQ_ENABLES_AUTO_PAUSE);
} else {
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH net 4/4] bnxt_en: Fix ethtool -a reporting.
2016-03-28 23:46 [PATCH net 0/4] misc. small fixes Michael Chan
` (2 preceding siblings ...)
2016-03-28 23:46 ` [PATCH net 3/4] bnxt_en: Fix typo in bnxt_hwrm_set_pause_common() Michael Chan
@ 2016-03-28 23:46 ` Michael Chan
2016-03-29 4:10 ` [PATCH net 0/4] misc. small fixes David Miller
4 siblings, 0 replies; 6+ messages in thread
From: Michael Chan @ 2016-03-28 23:46 UTC (permalink / raw)
To: davem; +Cc: netdev
To report flow control tx/rx settings accurately regardless of autoneg
setting, we should use link_info->req_flow_ctrl. Before this patch,
the reported settings were only correct when autoneg was on.
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 9ada166..2e472f6 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -855,10 +855,8 @@ static void bnxt_get_pauseparam(struct net_device *dev,
if (BNXT_VF(bp))
return;
epause->autoneg = !!(link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL);
- epause->rx_pause =
- ((link_info->auto_pause_setting & BNXT_LINK_PAUSE_RX) != 0);
- epause->tx_pause =
- ((link_info->auto_pause_setting & BNXT_LINK_PAUSE_TX) != 0);
+ epause->rx_pause = !!(link_info->req_flow_ctrl & BNXT_LINK_PAUSE_RX);
+ epause->tx_pause = !!(link_info->req_flow_ctrl & BNXT_LINK_PAUSE_TX);
}
static int bnxt_set_pauseparam(struct net_device *dev,
--
1.8.3.1
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH net 0/4] misc. small fixes.
2016-03-28 23:46 [PATCH net 0/4] misc. small fixes Michael Chan
` (3 preceding siblings ...)
2016-03-28 23:46 ` [PATCH net 4/4] bnxt_en: Fix ethtool -a reporting Michael Chan
@ 2016-03-29 4:10 ` David Miller
4 siblings, 0 replies; 6+ messages in thread
From: David Miller @ 2016-03-29 4:10 UTC (permalink / raw)
To: michael.chan; +Cc: netdev
From: Michael Chan <michael.chan@broadcom.com>
Date: Mon, 28 Mar 2016 19:46:03 -0400
> Misc. small fixes for net.
Series applied, thanks Michael.
^ permalink raw reply [flat|nested] 6+ messages in thread