[PATCH 0/4] Coverity buf fixes for bnx2x

[PATCH 0/4] Coverity buf fixes for bnx2x

[Stephen Hemminger]

These are obvious issues spotted several releases ago
by Coverity, and seem to have ended up assigned to me.

Hold off the merge until it is tested on real hardware;
sorry, I don't have this NIC.

Stephen Hemminger (4):
  net/bnx2x: remove dead conditional
  net/bnx2x: fix always true expression
  net/bnx2x: fix possible infinite loop at startup
  net/bnx2x: fix duplicate branch

 drivers/net/bnx2x/bnx2x.c       | 19 +++++++------------
 drivers/net/bnx2x/bnx2x_stats.c |  3 ---
 2 files changed, 7 insertions(+), 15 deletions(-)

-- 
2.45.2

[PATCH 1/4] net/bnx2x: remove dead conditional

[Stephen Hemminger]

The second if test here is impossible because it contradicts
previous line.

Coverity issue: 384428
Fixes: 540a211084a7 ("bnx2x: driver core")

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 drivers/net/bnx2x/bnx2x_stats.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x_stats.c b/drivers/net/bnx2x/bnx2x_stats.c
index d473c5e7ec..8adbe7e381 100644
--- a/drivers/net/bnx2x/bnx2x_stats.c
+++ b/drivers/net/bnx2x/bnx2x_stats.c
@@ -73,9 +73,6 @@ bnx2x_storm_stats_post(struct bnx2x_softc *sc)
 	int rc;
 
 	if (!sc->stats_pending) {
-		if (sc->stats_pending)
-			return;
-
 		sc->fw_stats_req->hdr.drv_stats_counter =
 			htole16(sc->stats_counter++);
 
-- 
2.45.2

[PATCH 2/4] net/bnx2x: fix always true expression

[Stephen Hemminger]

Coverity spotted that the check to enable single interrupt
mode would evaluate as always true since:

 The or condition sc->interrupt_mode != 2 || sc->interrupt_mode != 3
 will always be true because sc->interrupt_mode cannot be equal to
 two different values at the same time, so it must be not equal to
 at least one of them.

Coverity issue: 362046
Fixes: 540a211084a7 ("bnx2x: driver core")

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 drivers/net/bnx2x/bnx2x.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index 3153cc4d80..af31ac4604 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++ b/drivers/net/bnx2x/bnx2x.c
@@ -11189,11 +11189,9 @@ static int bnx2x_init_hw_func(struct bnx2x_softc *sc)
 /* Turn on a single ISR mode in IGU if driver is going to use
  * INT#x or MSI
  */
-		if ((sc->interrupt_mode != INTR_MODE_MSIX)
-		    || (sc->interrupt_mode != INTR_MODE_SINGLE_MSIX)) {
+		if (sc->interrupt_mode == INTR_MODE_INTX ||
+		    sc->interrupt_mode == INTR_MODE_MSI)
 			pf_conf |= IGU_PF_CONF_SINGLE_ISR_EN;
-		}
-
 /*
  * Timers workaround bug: function init part.
  * Need to wait 20msec after initializing ILT,
-- 
2.45.2

[PATCH 3/4] net/bnx2x: fix possible infinite loop at startup

[Stephen Hemminger]

Coverity spotted that one of the loop conditions was always true.
Fix by initializing the variable using same logic as Linux
kernel driver.

Coverity issue: 362057
Fixes: 540a211084a7 ("bnx2x: driver core")

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 drivers/net/bnx2x/bnx2x.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index af31ac4604..d96fcb55c9 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++ b/drivers/net/bnx2x/bnx2x.c
@@ -10331,12 +10331,13 @@ static int bnx2x_init_hw_common(struct bnx2x_softc *sc)
 	REG_WR(sc, PXP2_REG_RD_DISABLE_INPUTS, 0);
 
 	if (!CHIP_IS_E1x(sc)) {
-		int factor = 0;
+		int factor = CHIP_REV_IS_EMUL(sc) ? 1000 :
+				(CHIP_REV_IS_FPGA(sc) ? 400 : 0);
 
 		ecore_init_block(sc, BLOCK_PGLUE_B, PHASE_COMMON);
 		ecore_init_block(sc, BLOCK_ATC, PHASE_COMMON);
 
-/* let the HW do it's magic... */
+		/* let the HW do it's magic... */
 		do {
 			DELAY(200000);
 			val = REG_RD(sc, ATC_REG_ATC_INIT_DONE);
-- 
2.45.2

[PATCH 4/4] net/bnx2x: fix duplicate branch

[Stephen Hemminger]

Coverity spotted that both legs of the conditional are the same.
Looking at kernel driver there is additional code there, but the
kernel driver supports Wake On Lan, and DPDK does not.

Coverity issue: 362072
Fixes: 540a211084a7 ("bnx2x: driver core")

Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
 drivers/net/bnx2x/bnx2x.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c
index d96fcb55c9..51e5cabf7b 100644
--- a/drivers/net/bnx2x/bnx2x.c
+++ b/drivers/net/bnx2x/bnx2x.c
@@ -1623,16 +1623,12 @@ static int bnx2x_nic_unload_no_mcp(struct bnx2x_softc *sc)
 }
 
 /* request unload mode from the MCP: COMMON, PORT or FUNCTION */
-static uint32_t bnx2x_send_unload_req(struct bnx2x_softc *sc, int unload_mode)
+static uint32_t bnx2x_send_unload_req(struct bnx2x_softc *sc, int unload_mode __rte_unused)
 {
 	uint32_t reset_code = 0;
 
 	/* Select the UNLOAD request mode */
-	if (unload_mode == UNLOAD_NORMAL) {
-		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
-	} else {
-		reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
-	}
+	reset_code = DRV_MSG_CODE_UNLOAD_REQ_WOL_DIS;
 
 	/* Send the request to the MCP */
 	if (!BNX2X_NOMCP(sc)) {
-- 
2.45.2

RE: [EXTERNAL] [PATCH 4/4] net/bnx2x: fix duplicate branch

[Jerin Jacob]

> -----Original Message-----
> From: Stephen Hemminger <stephen@networkplumber.org>
> Sent: Tuesday, November 12, 2024 11:14 PM
> To: dev@dpdk.org
> Cc: Stephen Hemminger <stephen@networkplumber.org>; Julien Aube
> <julien_dpdk@jaube.fr>; Harish Patil <harish.patil@cavium.com>
> Subject: [EXTERNAL] [PATCH 4/4] net/bnx2x: fix duplicate branch
> 
> Coverity spotted that both legs of the conditional are the same. Looking at
> kernel driver there is additional code there, but the kernel driver supports Wake
> On Lan, and DPDK does not. Coverity issue: 362072 Fixes: 540a211084a7
> ("bnx2x: driver 
> Coverity spotted that both legs of the conditional are the same.
> Looking at kernel driver there is additional code there, but the kernel driver
> supports Wake On Lan, and DPDK does not.
> 
> Coverity issue: 362072
> Fixes: 540a211084a7 ("bnx2x: driver core")
> 
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>

Added Cc: stable@dpdk.org and  Series applied to dpdk-next-net-mrvl/for-main. Thanks