[<prev] [next>] [day] [month] [year] [list]
Message-ID: <8931251f-b24a-5043-3bdd-f6ea810759f3@canonical.com>
Date: Fri, 2 Jul 2021 11:42:17 +0100
From: Colin Ian King <colin.king@...onical.com>
To: Jes Sorensen <jes@...ined-monkey.org>,
Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
"open list:HIPPI" <linux-hippi@...site.dk>
Cc: "David S. Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: hippi: incorrect address masking and compare operation
Hi,
Static analysis with Coverity has detected an issue in
drivers/net/hippi/rrunner.c where a masking operation and a comparison
is always false.
The analysis is as follows:
656 /*
657 * Sanity test to see if we conflict with the DMA
658 * limitations of the Roadrunner.
659 */
Operands don't affect result (CONSTANT_EXPRESSION_RESULT)
dead_error_condition: The condition ((unsigned long)skb->data & 0xfffUL)
> 18446744073709486295UL cannot be true.
660 if ((((unsigned long)skb->data) & 0xfff) > ~65320)
Logically dead code (DEADCODE)dead_error_line: Execution cannot reach
this statement: printk("skb alloc error\n");.
661 printk("skb alloc error\n");
662
I suspect the masking 0xfff is incorrect here, I think it be ~0xfff but
I'm not 100% sure.
Colin
Powered by blists - more mailing lists