From: Dan Carpenter <dan.carpenter@oracle.com>
To: kxie@chelsio.com
Cc: linux-scsi@vger.kernel.org
Subject: re: [SCSI] cxgb4i: add support for T5 adapter
Date: Thu, 27 Jun 2013 23:47:02 +0300 [thread overview]
Message-ID: <20130627204627.GA13082@elgon.mountain> (raw)
Hello Karen Xie,
The patch 3bd3e8bf6250: "[SCSI] cxgb4i: add support for T5 adapter"
from May 29, 2013, leads to the following Sparse warning:
drivers/scsi/cxgbi/cxgb4i/cxgb4i.c:284:29: warning: incorrect type in assignment (different base types)
drivers/scsi/cxgbi/cxgb4i/cxgb4i.c:284:29: expected restricted __be64 [usertype] params
drivers/scsi/cxgbi/cxgb4i/cxgb4i.c:284:29: got restricted __be32 [usertype] <noident>
drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
271 struct cpl_t5_act_open_req *req =
272 (struct cpl_t5_act_open_req *)skb->head;
273
274 req = (struct cpl_t5_act_open_req *)skb->head;
275
276 INIT_TP_WR(req, 0);
277 OPCODE_TID(req) = cpu_to_be32(MK_OPCODE_TID(CPL_ACT_OPEN_REQ,
278 qid_atid));
279 req->local_port = csk->saddr.sin_port;
280 req->peer_port = csk->daddr.sin_port;
281 req->local_ip = csk->saddr.sin_addr.s_addr;
282 req->peer_ip = csk->daddr.sin_addr.s_addr;
283 req->opt0 = cpu_to_be64(opt0);
284 req->params = cpu_to_be32(select_ntuple(csk->cdev, csk->l2t));
^^^^^^^^^^^
For cpl_t5_act_open_req ->params is a __be64. For other types it is a
__be32 so I suspect this is a copy and paste error.
285 opt2 |= 1 << 31;
286 req->opt2 = cpu_to_be32(opt2);
Here is an article on how to use Sparse for endian bugs:
http://lwn.net/Articles/205624/
Also I'm on the topic, I had another concern later in the file:
drivers/scsi/cxgbi/cxgb4i/cxgb4i.c
1629 rpl = (struct cpl_act_establish *)skb->data;
1630 opc = rpl->ot.opcode;
1631 log_debug(1 << CXGBI_DBG_TOE,
1632 "cdev %p, opcode 0x%x(0x%x,0x%x), skb %p.\n",
1633 cdev, opc, rpl->ot.opcode_tid, ntohl(rpl->ot.opcode_tid), skb);
1634 if (cxgb4i_cplhandlers[opc])
1635 cxgb4i_cplhandlers[opc](cdev, skb);
^^^
Since "opc" comes from skb->data shouldn't we check that it is less than
ARRAY_SIZE(cxgb4i_cplhandlers) (239 elements)? Another solution that's
used in drivers/net/ethernet/chelsio/cxgb/cpl5_cmd.h would be to define
the array as having 256 elements.
1636 else {
1637 pr_err("No handler for opcode 0x%x.\n", opc);
1638 __kfree_skb(skb);
1639 }
regards,
dan carpenter
reply other threads:[~2013-06-27 20:47 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130627204627.GA13082@elgon.mountain \
--to=dan.carpenter@oracle.com \
--cc=kxie@chelsio.com \
--cc=linux-scsi@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox