[PATCH net 0/2] bnx2x: bug fixes patch series

[PATCH net 0/2] bnx2x: bug fixes patch series

[Yuval Mintz]

Hi Dave,

This series contains fixes for bnx2x's 'unload previous driver' flow
and its recovery flow.

Please consider applying these patches to 'net'.

Thanks,
Yuval Mintz

[PATCH net 1/2] bnx2x: fix unload previous driver flow when flr-capable

[Yuval Mintz]

The existing previous driver unload flow is flawed, causing the probe of 
functions reaching the 'uncommon fork' in flr-capable devices to fail.

This patch resolves this, as well as fixing the flow for hypervisors which 
disable flr capabilities from functions as they pass them as PDA to VMs, 
as we cannot base the flow on the pci configuration space.

Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c |   55 +++++++++++-----------
 1 files changed, 28 insertions(+), 27 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index dd451c3..5384a6d 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -9384,32 +9384,24 @@ static int __devinit bnx2x_prev_mark_path(struct bnx2x *bp)
 	return rc;
 }
 
-static bool __devinit bnx2x_can_flr(struct bnx2x *bp)
-{
-	int pos;
-	u32 cap;
-	struct pci_dev *dev = bp->pdev;
-
-	pos = pci_pcie_cap(dev);
-	if (!pos)
-		return false;
-
-	pci_read_config_dword(dev, pos + PCI_EXP_DEVCAP, &cap);
-	if (!(cap & PCI_EXP_DEVCAP_FLR))
-		return false;
-
-	return true;
-}
-
 static int __devinit bnx2x_do_flr(struct bnx2x *bp)
 {
 	int i, pos;
 	u16 status;
 	struct pci_dev *dev = bp->pdev;
 
-	/* probe the capability first */
-	if (bnx2x_can_flr(bp))
-		return -ENOTTY;
+
+	if (CHIP_IS_E1x(bp)) {
+		BNX2X_DEV_INFO("FLR not supported in E1/E1H\n");
+		return -EINVAL;
+	}
+
+	/* only bootcode REQ_BC_VER_4_INITIATE_FLR and onwards support flr */
+	if (bp->common.bc_ver < REQ_BC_VER_4_INITIATE_FLR) {
+		BNX2X_ERR("FLR not supported by BC_VER: 0x%x\n",
+			  bp->common.bc_ver);
+		return -EINVAL;
+	}
 
 	pos = pci_pcie_cap(dev);
 	if (!pos)
@@ -9429,12 +9421,8 @@ static int __devinit bnx2x_do_flr(struct bnx2x *bp)
 		"transaction is not cleared; proceeding with reset anyway\n");
 
 clear:
-	if (bp->common.bc_ver < REQ_BC_VER_4_INITIATE_FLR) {
-		BNX2X_ERR("FLR not supported by BC_VER: 0x%x\n",
-			  bp->common.bc_ver);
-		return -EINVAL;
-	}
 
+	BNX2X_DEV_INFO("Initiating FLR\n");
 	bnx2x_fw_command(bp, DRV_MSG_CODE_INITIATE_FLR, 0);
 
 	return 0;
@@ -9454,8 +9442,21 @@ static int __devinit bnx2x_prev_unload_uncommon(struct bnx2x *bp)
 	 * the one required, then FLR will be sufficient to clean any residue
 	 * left by previous driver
 	 */
-	if (bnx2x_test_firmware_version(bp, false) && bnx2x_can_flr(bp))
-		return bnx2x_do_flr(bp);
+	rc = bnx2x_test_firmware_version(bp, false);
+
+	if (!rc) {
+		/* fw version is good */
+		BNX2X_DEV_INFO("FW version matches our own. Attempting FLR\n");
+		rc = bnx2x_do_flr(bp);
+	}
+
+	if (!rc) {
+		/* FLR was performed */
+		BNX2X_DEV_INFO("FLR successful\n");
+		return 0;
+	}
+
+	BNX2X_DEV_INFO("Could not FLR\n");
 
 	/* Close the MCP request, return failure*/
 	rc = bnx2x_prev_mcp_done(bp);
-- 
1.7.9.rc2

[PATCH net 2/2] bnx2x: Fix recovery flow cleanup during probe

[Yuval Mintz]

During probe, every function probed clears the recovery registers from
all functions on its path - thus signaling that given a future recovery
event, there will be no need to wait for those functions.

This is a flawed behaviour - each function should only be responsible
for its own bit.

Since this registers are handled during the load/unload routines,
this cleanup is removed altogether.

Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com>
Signed-off-by: Ariel Elior <ariele@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c |   17 -----------------
 1 files changed, 0 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 5384a6d..02b5a34 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -4041,20 +4041,6 @@ static bool bnx2x_get_load_status(struct bnx2x *bp, int engine)
 	return val != 0;
 }
 
-/*
- * Reset the load status for the current engine.
- */
-static void bnx2x_clear_load_status(struct bnx2x *bp)
-{
-	u32 val;
-	u32 mask = (BP_PATH(bp) ? BNX2X_PATH1_LOAD_CNT_MASK :
-		    BNX2X_PATH0_LOAD_CNT_MASK);
-	bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_RECOVERY_REG);
-	val = REG_RD(bp, BNX2X_RECOVERY_GLOB_REG);
-	REG_WR(bp, BNX2X_RECOVERY_GLOB_REG, val & (~mask));
-	bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_RECOVERY_REG);
-}
-
 static void _print_next_block(int idx, const char *blk)
 {
 	pr_cont("%s%s", idx ? ", " : "", blk);
@@ -11428,9 +11414,6 @@ static int __devinit bnx2x_init_dev(struct pci_dev *pdev,
 	if (!chip_is_e1x)
 		REG_WR(bp, PGLUE_B_REG_INTERNAL_PFID_ENABLE_TARGET_READ, 1);
 
-	/* Reset the load counter */
-	bnx2x_clear_load_status(bp);
-
 	dev->watchdog_timeo = TX_TIMEOUT;
 
 	dev->netdev_ops = &bnx2x_netdev_ops;
-- 
1.7.9.rc2

Re: [PATCH net 0/2] bnx2x: bug fixes patch series

[David Miller]

From: "Yuval Mintz" <yuvalmin@broadcom.com>
Date: Thu, 9 Aug 2012 17:37:24 +0300

> This series contains fixes for bnx2x's 'unload previous driver' flow
> and its recovery flow.
> 
> Please consider applying these patches to 'net'.

Both patches applied, thanks.

[PATCH net 0/2] bnx2x: Bug fixes patch series

[Yuval Mintz]

Hi Dave,

This patch contain fixes in two areas -

  1. Preventing UNDI FW from illegally accessing memory when it shouldn't.

  2. Fix interrupt coalescing initialization and configuration when working
     with multiple CoS.

Please consider applying these patches to `net'.

Thanks,
Yuval Mintz

Re: [PATCH net 0/2] bnx2x: Bug fixes patch series

[David Miller]

From: "Yuval Mintz" <yuvalmin@broadcom.com>
Date: Thu, 18 Apr 2013 11:49:04 +0300

> Hi Dave,
> 
> This patch contain fixes in two areas -
> 
>   1. Preventing UNDI FW from illegally accessing memory when it shouldn't.
> 
>   2. Fix interrupt coalescing initialization and configuration when working
>      with multiple CoS.
> 
> Please consider applying these patches to `net'.

Both applied, thanks.

[PATCH net 0/2] bnx2x: Bug fixes patch series

[Yuval Mintz]

Hi Dave,

This series contains 2 unrelated fixes:

 1. FW asserts during unload might be encountered if specific memory
    allocations fail during load.

 2. SR-IOV related, fixes a crash that will occour if the driver were to
    be rmmoded and then re-probed while there are both assigned and unassigned
    VFs originating from the same PF.

Please consider applying these patches to `net'.

Thanks,
Yuval Mintz

Re: [PATCH net 0/2] bnx2x: Bug fixes patch series

[David Miller]

From: "Yuval Mintz" <yuvalmin@broadcom.com>
Date: Sun, 27 Oct 2013 13:06:47 +0200

> This series contains 2 unrelated fixes:
> 
>  1. FW asserts during unload might be encountered if specific memory
>     allocations fail during load.
> 
>  2. SR-IOV related, fixes a crash that will occour if the driver were to
>     be rmmoded and then re-probed while there are both assigned and unassigned
>     VFs originating from the same PF.
> 
> Please consider applying these patches to `net'.

Both applied, thanks.

[PATCH net 0/2] bnx2x: Bug fixes patch series

[Yuval Mintz]

Hi Dave,

This series contains 2 unrelated fixes:

 1. FW asserts during unload might be encountered if specific memory
    allocations fail during load.

 2. SR-IOV related, fixes a crash that will occour if the driver were to
    be rmmoded and then re-probed while there are both assigned and unassigned
    VFs originating from the same PF.

Please consider applying these patches to `net'.

Thanks,
Yuval Mintz