bnxt_en: Skip firmware fatal error recovery if chip is not accessible

[ Upstream commit 83474a9b252ab23e6003865c2775024344cb9c09 ]

Driver starts firmware fatal error recovery by detecting
heartbeat failure or fw reset count register changing.  But
these checks are not reliable if the device is not accessible.
This can happen while DPC (Downstream Port containment) is in
progress.  Skip firmware fatal recovery if pci_device_is_present()
returns false.

Fixes: acfb50e4e7 ("bnxt_en: Add FW fatal devlink_health_reporter.")
Reviewed-by: Somnath Kotur <somnath.kotur@broadcom.com>
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Signed-off-by: Vikas Gupta <vikas.gupta@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
This commit is contained in:
Vikas Gupta 2023-06-07 00:54:07 -07:00 committed by Greg Kroah-Hartman
parent 5fc86a4580
commit 5ce24936d5

View File

@ -11576,6 +11576,7 @@ static void bnxt_tx_timeout(struct net_device *dev, unsigned int txqueue)
static void bnxt_fw_health_check(struct bnxt *bp)
{
struct bnxt_fw_health *fw_health = bp->fw_health;
struct pci_dev *pdev = bp->pdev;
u32 val;
if (!fw_health->enabled || test_bit(BNXT_STATE_IN_FW_RESET, &bp->state))
@ -11589,7 +11590,7 @@ static void bnxt_fw_health_check(struct bnxt *bp)
}
val = bnxt_fw_health_readl(bp, BNXT_FW_HEARTBEAT_REG);
if (val == fw_health->last_fw_heartbeat) {
if (val == fw_health->last_fw_heartbeat && pci_device_is_present(pdev)) {
fw_health->arrests++;
goto fw_reset;
}
@ -11597,7 +11598,7 @@ static void bnxt_fw_health_check(struct bnxt *bp)
fw_health->last_fw_heartbeat = val;
val = bnxt_fw_health_readl(bp, BNXT_FW_RESET_CNT_REG);
if (val != fw_health->last_fw_reset_cnt) {
if (val != fw_health->last_fw_reset_cnt && pci_device_is_present(pdev)) {
fw_health->discoveries++;
goto fw_reset;
}