From: Dan Carpenter <dan.carpenter@oracle.com>
To: Roland Dreier <roland@kernel.org>
Cc: Sean Hefty <sean.hefty@intel.com>,
Hal Rosenstock <hal.rosenstock@gmail.com>,
Naresh Gottumukkala <bgottumukkala@emulex.com>,
Parav Pandit <parav.pandit@emulex.com>,
linux-rdma@vger.kernel.org, kernel-janitors@vger.kernel.org
Subject: [patch] RDMA/ocrdma: several stack info leaks
Date: Thu, 25 Jul 2013 20:05:32 +0300 [thread overview]
Message-ID: <20130725170532.GE7026@elgon.mountain> (raw)
A grab bag of places which don't properly initialize stack data. If the
struct member was at the end of the struct then I fixed the issue by not
passing the uninitialized bits to the user. This seems like the
prefered thing in infiniband drivers.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index dcfbab1..78bada7 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -256,7 +256,8 @@ struct ib_ucontext *ocrdma_alloc_ucontext(struct ib_device *ibdev,
resp.rsvd = 0;
memcpy(resp.fw_ver, dev->attr.fw_ver, sizeof(resp.fw_ver));
- status = ib_copy_to_udata(udata, &resp, sizeof(resp));
+ status = ib_copy_to_udata(udata, &resp,
+ sizeof(resp) - sizeof(resp.rsvd1));
if (status)
goto cpy_err;
return &ctx->ibucontext;
@@ -357,9 +358,12 @@ static int ocrdma_copy_pd_uresp(struct ocrdma_pd *pd,
goto dpp_map_err;
rsp.dpp_page_addr_hi = upper_32_bits(dpp_page_addr);
rsp.dpp_page_addr_lo = dpp_page_addr;
+ } else {
+ rsp.dpp_page_addr_hi = 0;
+ rsp.dpp_page_addr_lo = 0;
}
- status = ib_copy_to_udata(udata, &rsp, sizeof(rsp));
+ status = ib_copy_to_udata(udata, &rsp, sizeof(rsp) - sizeof(rsp.rsvd1));
if (status)
goto ucopy_err;
@@ -700,7 +704,8 @@ static int ocrdma_copy_cq_uresp(struct ocrdma_cq *cq, struct ib_udata *udata,
uresp.db_page_addr = cq->dev->nic_info.unmapped_db;
uresp.db_page_size = cq->dev->nic_info.db_page_size;
uresp.phase_change = cq->phase_change ? 1 : 0;
- status = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+ status = ib_copy_to_udata(udata, &uresp,
+ sizeof(uresp) - sizeof(u64) * 2);
if (status) {
pr_err("%s(%d) copy error cqid=0x%x.\n",
__func__, cq->dev->id, cq->id);
@@ -1461,6 +1466,8 @@ static int ocrdma_copy_srq_uresp(struct ocrdma_srq *srq, struct ib_udata *udata)
struct ocrdma_create_srq_uresp uresp;
uresp.rq_dbid = srq->rq.dbid;
+ uresp.resv0 = 0;
+ uresp.resv1 = 0;
uresp.num_rq_pages = 1;
uresp.rq_page_addr[0] = srq->rq.pa;
uresp.rq_page_size = srq->rq.len;
@@ -1476,7 +1483,8 @@ static int ocrdma_copy_srq_uresp(struct ocrdma_srq *srq, struct ib_udata *udata)
uresp.db_shift = 16;
}
- status = ib_copy_to_udata(udata, &uresp, sizeof(uresp));
+ status = ib_copy_to_udata(udata, &uresp,
+ sizeof(uresp) - sizeof(u64) * 2);
if (status)
return status;
status = ocrdma_add_mmap(srq->pd->uctx, uresp.rq_page_addr[0],
reply other threads:[~2013-07-25 17:05 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=20130725170532.GE7026@elgon.mountain \
--to=dan.carpenter@oracle.com \
--cc=bgottumukkala@emulex.com \
--cc=hal.rosenstock@gmail.com \
--cc=kernel-janitors@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=parav.pandit@emulex.com \
--cc=roland@kernel.org \
--cc=sean.hefty@intel.com \
/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