xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>,
	xen-devel@lists.xensource.com,
	Ian Campbell <ian.campbell@citrix.com>,
	"Xu, Dongxiao" <dongxiao.xu@intel.com>
Subject: [PATCH] xen: netback: check if foreign pages are actually netback-created foreign pages.
Date: Fri, 30 Jul 2010 15:16:47 +0100	[thread overview]
Message-ID: <1280499407-7496-2-git-send-email-ian.campbell@citrix.com> (raw)
In-Reply-To: <1280499392.24292.1964.camel@zakaz.uk.xensource.com>

020ba906 "xen/netback: Multiple tasklets support." changed
netbk_gop_frag_copy to attempt to lookup a pending_tx_info for any
foreign page, regardless of whether the page was a netback-foreign
page.

In the case of non-netback pages this can lead to dereferencing a NULL
src_pend->netif.

Restore the behaviour of netif_page_index prior toa3031942
"xen/netback: Introduce a new struct type page_ext" by performing
tests to ensure that page is a netback page and extend the same checks
to netif_page_group.

Actually combine netif_page_{index,group} in to a single function
since they are always called together and it saves duplicating all the
checks.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Xu, Dongxiao <dongxiao.xu@intel.com>
---
 drivers/xen/netback/netback.c |   56 ++++++++++++++++++++++++++++------------
 1 files changed, 39 insertions(+), 17 deletions(-)

diff --git a/drivers/xen/netback/netback.c b/drivers/xen/netback/netback.c
index 0ed7e61..ed7cd65 100644
--- a/drivers/xen/netback/netback.c
+++ b/drivers/xen/netback/netback.c
@@ -89,18 +89,37 @@ static inline void netif_set_page_ext(struct page *pg, unsigned int group,
 	pg->mapping = ext.mapping;
 }
 
-static inline unsigned int netif_page_group(const struct page *pg)
+static inline int netif_get_page_ext(struct page *pg, unsigned int *_group, unsigned int *_idx)
 {
 	union page_ext ext = { .mapping = pg->mapping };
+	struct xen_netbk *netbk;
+	unsigned int group, idx;
 
-	return ext.e.group - 1;
-}
+	if (!PageForeign(pg))
+		return 0;
 
-static inline unsigned int netif_page_index(const struct page *pg)
-{
-	union page_ext ext = { .mapping = pg->mapping };
+	group = ext.e.group - 1;
+
+	if (group < 0 || group >= xen_netbk_group_nr)
+		return 0;
+
+	netbk = &xen_netbk[group];
+
+	if (netbk->mmap_pages == NULL)
+		return 0;
 
-	return ext.e.idx;
+	idx = ext.e.idx;
+
+	if ((idx < 0) || (idx >= MAX_PENDING_REQS))
+		return 0;
+
+	if (netbk->mmap_pages[idx] != pg)
+		return 0;
+
+	*_group = group;
+	*_idx = idx;
+
+	return 1;
 }
 
 /*
@@ -386,8 +405,12 @@ static void netbk_gop_frag_copy(struct xen_netif *netif,
 {
 	struct gnttab_copy *copy_gop;
 	struct netbk_rx_meta *meta;
-	int group = netif_page_group(page);
-	int idx = netif_page_index(page);
+	/*
+	 * These variables a used iff netif_get_page_ext returns true,
+	 * in which case they are guaranteed to be initialized.
+         */
+	unsigned int uninitialized_var(group), uninitialized_var(idx);
+	int foreign = netif_get_page_ext(page, &group, &idx);
 	unsigned long bytes;
 
 	/* Data must not cross a page boundary. */
@@ -445,7 +468,7 @@ static void netbk_gop_frag_copy(struct xen_netif *netif,
 
 		copy_gop = npo->copy + npo->copy_prod++;
 		copy_gop->flags = GNTCOPY_dest_gref;
-		if (PageForeign(page)) {
+		if (foreign) {
 			struct xen_netbk *netbk = &xen_netbk[group];
 			struct pending_tx_info *src_pend;
 
@@ -1546,14 +1569,13 @@ static void netif_idx_release(struct xen_netbk *netbk, u16 pending_idx)
 
 static void netif_page_release(struct page *page, unsigned int order)
 {
-	int group = netif_page_group(page);
-	int idx = netif_page_index(page);
-	struct xen_netbk *netbk = &xen_netbk[group];
+	unsigned int group, idx;
+	int foreign = netif_get_page_ext(page, &group, &idx);
+
+	BUG_ON(!foreign);
 	BUG_ON(order);
-	BUG_ON(group < 0 || group >= xen_netbk_group_nr);
-	BUG_ON(idx < 0 || idx >= MAX_PENDING_REQS);
-	BUG_ON(netbk->mmap_pages[idx] != page);
-	netif_idx_release(netbk, idx);
+
+	netif_idx_release(&xen_netbk[group], idx);
 }
 
 irqreturn_t netif_be_int(int irq, void *dev_id)
-- 
1.5.6.5

      parent reply	other threads:[~2010-07-30 14:16 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-30 14:16 [GIT/PATCH 0/2] pair of netback fixes Ian Campbell
2010-07-30 14:16 ` [PATCH] xen: netback: do not unleash netback threads until initialisation is complete Ian Campbell
2010-07-30 14:16 ` Ian Campbell [this message]

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=1280499407-7496-2-git-send-email-ian.campbell@citrix.com \
    --to=ian.campbell@citrix.com \
    --cc=dongxiao.xu@intel.com \
    --cc=jeremy@goop.org \
    --cc=xen-devel@lists.xensource.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;
as well as URLs for NNTP newsgroup(s).