[PATCH net 0/2] bnx2x: kdump related fixes

[PATCH net 0/2] bnx2x: kdump related fixes

[Yuval Mintz]

This patch series aims to fix bnx2x driver issues when loading in kdump kernel.
Both issues fixed here would be fatal to the device, requiring full reset of
the system in order to recover, preventing the device from serving its purpose
in the kdump environment.

Dave,

Please consider applying this series to `net'.

Thanks,
Yuval
-- 
1.8.3.1

[PATCH net 1/2] bnx2x: Fix kdump on 4-port device

[Yuval Mintz]

When running in a kdump kernel, it's very likely that due to sync. loss with
management firmware the first PCI function to probe and reach the previous
unload flow would decide it can reset the chip and continue onward. While doing
so, it will only close its own Rx port.

On a 4-port device where 2nd port on engine is a 1g-port, the 2nd port would
allow ingress traffic after the chip is reset [assuming it was active on the
first kernel]. This would later cause a HW attention.

This changes driver flow to close both ports' 1g capabilities during the
previous driver unload flow prior to the chip reset.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 53 ++++++++++++++++--------
 1 file changed, 35 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index ae571a1..f391edb 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -129,8 +129,8 @@ struct bnx2x_mac_vals {
 	u32 xmac_val;
 	u32 emac_addr;
 	u32 emac_val;
-	u32 umac_addr;
-	u32 umac_val;
+	u32 umac_addr[2];
+	u32 umac_val[2];
 	u32 bmac_addr;
 	u32 bmac_val[2];
 };
@@ -10141,6 +10141,25 @@ static u32 bnx2x_get_pretend_reg(struct bnx2x *bp)
 	return base + (BP_ABS_FUNC(bp)) * stride;
 }
 
+static bool bnx2x_prev_unload_close_umac(struct bnx2x *bp,
+					 u8 port, u32 reset_reg,
+					 struct bnx2x_mac_vals *vals)
+{
+	u32 mask = MISC_REGISTERS_RESET_REG_2_UMAC0 << port;
+	u32 base_addr;
+
+	if (!(mask & reset_reg))
+		return false;
+
+	BNX2X_DEV_INFO("Disable umac Rx %02x\n", port);
+	base_addr = port ? GRCBASE_UMAC1 : GRCBASE_UMAC0;
+	vals->umac_addr[port] = base_addr + UMAC_REG_COMMAND_CONFIG;
+	vals->umac_val[port] = REG_RD(bp, vals->umac_addr[port]);
+	REG_WR(bp, vals->umac_addr[port], 0);
+
+	return true;
+}
+
 static void bnx2x_prev_unload_close_mac(struct bnx2x *bp,
 					struct bnx2x_mac_vals *vals)
 {
@@ -10149,10 +10168,7 @@ static void bnx2x_prev_unload_close_mac(struct bnx2x *bp,
 	u8 port = BP_PORT(bp);
 
 	/* reset addresses as they also mark which values were changed */
-	vals->bmac_addr = 0;
-	vals->umac_addr = 0;
-	vals->xmac_addr = 0;
-	vals->emac_addr = 0;
+	memset(vals, 0, sizeof(*vals));
 
 	reset_reg = REG_RD(bp, MISC_REG_RESET_REG_2);
 
@@ -10201,15 +10217,11 @@ static void bnx2x_prev_unload_close_mac(struct bnx2x *bp,
 			REG_WR(bp, vals->xmac_addr, 0);
 			mac_stopped = true;
 		}
-		mask = MISC_REGISTERS_RESET_REG_2_UMAC0 << port;
-		if (mask & reset_reg) {
-			BNX2X_DEV_INFO("Disable umac Rx\n");
-			base_addr = BP_PORT(bp) ? GRCBASE_UMAC1 : GRCBASE_UMAC0;
-			vals->umac_addr = base_addr + UMAC_REG_COMMAND_CONFIG;
-			vals->umac_val = REG_RD(bp, vals->umac_addr);
-			REG_WR(bp, vals->umac_addr, 0);
-			mac_stopped = true;
-		}
+
+		mac_stopped |= bnx2x_prev_unload_close_umac(bp, 0,
+							    reset_reg, vals);
+		mac_stopped |= bnx2x_prev_unload_close_umac(bp, 1,
+							    reset_reg, vals);
 	}
 
 	if (mac_stopped)
@@ -10505,8 +10517,11 @@ static int bnx2x_prev_unload_common(struct bnx2x *bp)
 		/* Close the MAC Rx to prevent BRB from filling up */
 		bnx2x_prev_unload_close_mac(bp, &mac_vals);
 
-		/* close LLH filters towards the BRB */
+		/* close LLH filters for both ports towards the BRB */
+		bnx2x_set_rx_filter(&bp->link_params, 0);
+		bp->link_params.port ^= 1;
 		bnx2x_set_rx_filter(&bp->link_params, 0);
+		bp->link_params.port ^= 1;
 
 		/* Check if the UNDI driver was previously loaded */
 		if (bnx2x_prev_is_after_undi(bp)) {
@@ -10553,8 +10568,10 @@ static int bnx2x_prev_unload_common(struct bnx2x *bp)
 
 	if (mac_vals.xmac_addr)
 		REG_WR(bp, mac_vals.xmac_addr, mac_vals.xmac_val);
-	if (mac_vals.umac_addr)
-		REG_WR(bp, mac_vals.umac_addr, mac_vals.umac_val);
+	if (mac_vals.umac_addr[0])
+		REG_WR(bp, mac_vals.umac_addr[0], mac_vals.umac_val[0]);
+	if (mac_vals.umac_addr[1])
+		REG_WR(bp, mac_vals.umac_addr[1], mac_vals.umac_val[1]);
 	if (mac_vals.emac_addr)
 		REG_WR(bp, mac_vals.emac_addr, mac_vals.emac_val);
 	if (mac_vals.bmac_addr) {
-- 
1.9.3

[PATCH net 2/2] bnx2x: Fix kdump when iommu=on

[Yuval Mintz]

When IOMM-vtd is active, once main kernel crashes unfinished DMAE transactions
will be blocked, putting the HW in an error state which will cause further
transactions to timeout.

Current employed logic uses wrong macros, causing the first function to be the
only function that cleanups that error state during its probe/load.

This patch allows all the functions to successfully re-load in kdump kernel.

Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Ariel Elior <Ariel.Elior@qlogic.com>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 39 ++++++++++--------------
 1 file changed, 16 insertions(+), 23 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index f391edb..16b56b1 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -7866,6 +7866,20 @@ int bnx2x_init_hw_func_cnic(struct bnx2x *bp)
 	return 0;
 }
 
+/* previous driver DMAE transaction may have occurred when pre-boot stage ended
+ * and boot began, or when kdump kernel was loaded. Either case would invalidate
+ * the addresses of the transaction, resulting in was-error bit set in the pci
+ * causing all hw-to-host pcie transactions to timeout. If this happened we want
+ * to clear the interrupt which detected this from the pglueb and the was done
+ * bit
+ */
+static void bnx2x_clean_pglue_errors(struct bnx2x *bp)
+{
+	if (!CHIP_IS_E1x(bp))
+		REG_WR(bp, PGLUE_B_REG_WAS_ERROR_PF_7_0_CLR,
+		       1 << BP_ABS_FUNC(bp));
+}
+
 static int bnx2x_init_hw_func(struct bnx2x *bp)
 {
 	int port = BP_PORT(bp);
@@ -7958,8 +7972,7 @@ static int bnx2x_init_hw_func(struct bnx2x *bp)
 
 	bnx2x_init_block(bp, BLOCK_PGLUE_B, init_phase);
 
-	if (!CHIP_IS_E1x(bp))
-		REG_WR(bp, PGLUE_B_REG_WAS_ERROR_PF_7_0_CLR, func);
+	bnx2x_clean_pglue_errors(bp);
 
 	bnx2x_init_block(bp, BLOCK_ATC, init_phase);
 	bnx2x_init_block(bp, BLOCK_DMAE, init_phase);
@@ -10588,26 +10601,6 @@ static int bnx2x_prev_unload_common(struct bnx2x *bp)
 	return bnx2x_prev_mcp_done(bp);
 }
 
-/* previous driver DMAE transaction may have occurred when pre-boot stage ended
- * and boot began, or when kdump kernel was loaded. Either case would invalidate
- * the addresses of the transaction, resulting in was-error bit set in the pci
- * causing all hw-to-host pcie transactions to timeout. If this happened we want
- * to clear the interrupt which detected this from the pglueb and the was done
- * bit
- */
-static void bnx2x_prev_interrupted_dmae(struct bnx2x *bp)
-{
-	if (!CHIP_IS_E1x(bp)) {
-		u32 val = REG_RD(bp, PGLUE_B_REG_PGLUE_B_INT_STS);
-		if (val & PGLUE_B_PGLUE_B_INT_STS_REG_WAS_ERROR_ATTN) {
-			DP(BNX2X_MSG_SP,
-			   "'was error' bit was found to be set in pglueb upon startup. Clearing\n");
-			REG_WR(bp, PGLUE_B_REG_WAS_ERROR_PF_7_0_CLR,
-			       1 << BP_FUNC(bp));
-		}
-	}
-}
-
 static int bnx2x_prev_unload(struct bnx2x *bp)
 {
 	int time_counter = 10;
@@ -10617,7 +10610,7 @@ static int bnx2x_prev_unload(struct bnx2x *bp)
 	/* clear hw from errors which may have resulted from an interrupted
 	 * dmae transaction.
 	 */
-	bnx2x_prev_interrupted_dmae(bp);
+	bnx2x_clean_pglue_errors(bp);
 
 	/* Release previously held locks */
 	hw_lock_reg = (BP_FUNC(bp) <= 5) ?
-- 
1.9.3

Re: [PATCH net 0/2] bnx2x: kdump related fixes

[David Miller]

From: Yuval Mintz <Yuval.Mintz@qlogic.com>
Date: Wed, 1 Apr 2015 10:02:18 +0300

> This patch series aims to fix bnx2x driver issues when loading in kdump kernel.
> Both issues fixed here would be fatal to the device, requiring full reset of
> the system in order to recover, preventing the device from serving its purpose
> in the kdump environment.

Series applied, thank you.