netdev - [PATCH] bnxt_en: Fix RSS logic in __bnxt_reserve

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20240724222106.147744-1-michael.chan@broadcom.com>
Date: Wed, 24 Jul 2024 15:21:06 -0700
From: Michael Chan <michael.chan@...adcom.com>
To: davem@...emloft.net
Cc: netdev@...r.kernel.org,
	edumazet@...gle.com,
	kuba@...nel.org,
	pabeni@...hat.com,
	pavan.chebbi@...adcom.com,
	andrew.gospodarek@...adcom.com
Subject: [PATCH] bnxt_en: Fix RSS logic in __bnxt_reserve_rings()

From: Pavan Chebbi <pavan.chebbi@...adcom.com>

In __bnxt_reserve_rings(), the existing code unconditionally sets the
default RSS indirection table to default if netif_is_rxfh_configured()
returns false.  This used to be correct before we added RSS contexts
support.  For example, if the user is changing the number of ethtool
channels, we will enter this path to reserve the new number of rings.
We will then set the RSS indirection table to default to cover the new
number of rings if netif_is_rxfh_configured() is false.

Now, with RSS contexts support, if the user has added or deleted RSS
contexts, we may now enter this path to reserve the new number of VNICs.
However, netif_is_rxfh_configured() will not return the correct state if
we are still in the middle of set_rxfh().  So the existing code may
set the indirection table of the default RSS context to default by
mistake.

Fix it to check if the reservation of the RX rings is changing.  Only
check netif_is_rxfh_configured() if it is changing.  RX rings will not
change in the middle of set_rxfh() and this will fix the issue.

Fixes: b3d0083caf9a ("bnxt_en: Support RSS contexts in ethtool .{get|set}_rxfh()")
Reviewed-by: Andy Gospodarek <andrew.gospodarek@...adcom.com>
Signed-off-by: Pavan Chebbi <pavan.chebbi@...adcom.com>
Signed-off-by: Michael Chan <michael.chan@...adcom.com>
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index bb3be33c1bbd..f788f114e430 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -7648,8 +7648,8 @@ static int bnxt_get_avail_msix(struct bnxt *bp, int num);
 static int __bnxt_reserve_rings(struct bnxt *bp)
 {
 	struct bnxt_hw_rings hwr = {0};
+	int rx_rings, old_rx_rings, rc;
 	int cp = bp->cp_nr_rings;
-	int rx_rings, rc;
 	int ulp_msix = 0;
 	bool sh = false;
 	int tx_cp;
@@ -7683,6 +7683,7 @@ static int __bnxt_reserve_rings(struct bnxt *bp)
 	hwr.grp = bp->rx_nr_rings;
 	hwr.rss_ctx = bnxt_get_total_rss_ctxs(bp, &hwr);
 	hwr.stat = bnxt_get_func_stat_ctxs(bp);
+	old_rx_rings = bp->hw_resc.resv_rx_rings;

 	rc = bnxt_hwrm_reserve_rings(bp, &hwr);
 	if (rc)
@@ -7737,7 +7738,8 @@ static int __bnxt_reserve_rings(struct bnxt *bp)
 	if (!bnxt_rings_ok(bp, &hwr))
 		return -ENOMEM;

-	if (!netif_is_rxfh_configured(bp->dev))
+	if (old_rx_rings != bp->hw_resc.resv_rx_rings &&
+	    !netif_is_rxfh_configured(bp->dev))
 		bnxt_set_dflt_rss_indir_tbl(bp, NULL);

 	if (!bnxt_ulp_registered(bp->edev) && BNXT_NEW_RM(bp)) {
-- 
2.30.1