patch #1 ~ #9 prepare for fix CVEs
patch #10 ~ #14 fixes: CVE-2021-28711 CVE-2021-28712 CVE-2021-28713 CVE-2021-28714 CVE-2021-28715
Juergen Gross (14): xen/netback: avoid race in xenvif_rx_ring_slots_available() xen: sync include/xen/interface/io/ring.h with Xen's newest version xen/blkfront: read response from backend only once xen/blkfront: don't take local copy of a request from the ring page xen/blkfront: don't trust the backend response data blindly xen/netfront: read response from backend only once xen/netfront: don't read data from request on the ring page xen/netfront: disentangle tx_skb_freelist xen/netfront: don't trust the backend response data blindly xen/blkfront: harden blkfront against event channel storms xen/netfront: harden netfront against event channel storms xen/console: harden hvc_xen against event channel storms xen/netback: fix rx queue stall detection xen/netback: don't queue unlimited number of packages
drivers/block/xen-blkfront.c | 141 ++++++++---- drivers/net/xen-netback/common.h | 1 + drivers/net/xen-netback/rx.c | 70 ++++-- drivers/net/xen-netfront.c | 372 ++++++++++++++++++++----------- drivers/tty/hvc/hvc_xen.c | 30 ++- include/xen/interface/io/ring.h | 293 +++++++++++++----------- 6 files changed, 575 insertions(+), 332 deletions(-)
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.177 commit 7740c404d0680293e42a0a17e4af1f5952165bdd
--------------------------------
[ Upstream commit ec7d8e7dd3a59528e305a18e93f1cb98f7faf83b ]
Since commit 23025393dbeb3b8b3 ("xen/netback: use lateeoi irq binding") xenvif_rx_ring_slots_available() is no longer called only from the rx queue kernel thread, so it needs to access the rx queue with the associated queue held.
Reported-by: Igor Druzhinin igor.druzhinin@citrix.com Fixes: 23025393dbeb3b8b3 ("xen/netback: use lateeoi irq binding") Signed-off-by: Juergen Gross jgross@suse.com Acked-by: Wei Liu wl@xen.org Link: https://lore.kernel.org/r/20210202070938.7863-1-jgross@suse.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netback/rx.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/xen-netback/rx.c b/drivers/net/xen-netback/rx.c index 9b62f65b630e4..48e2006f96ce6 100644 --- a/drivers/net/xen-netback/rx.c +++ b/drivers/net/xen-netback/rx.c @@ -38,10 +38,15 @@ static bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue) RING_IDX prod, cons; struct sk_buff *skb; int needed; + unsigned long flags; + + spin_lock_irqsave(&queue->rx_queue.lock, flags);
skb = skb_peek(&queue->rx_queue); - if (!skb) + if (!skb) { + spin_unlock_irqrestore(&queue->rx_queue.lock, flags); return false; + }
needed = DIV_ROUND_UP(skb->len, XEN_PAGE_SIZE); if (skb_is_gso(skb)) @@ -49,6 +54,8 @@ static bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue) if (skb->sw_hash) needed++;
+ spin_unlock_irqrestore(&queue->rx_queue.lock, flags); + do { prod = queue->rx.sring->req_prod; cons = queue->rx.req_cons;
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit dcbfd3f13b1363a358e65b1800aac840d176eacc
--------------------------------
commit 629a5d87e26fe96bcaab44cbb81f5866af6f7008 upstream.
Sync include/xen/interface/io/ring.h with Xen's newest version in order to get the RING_COPY_RESPONSE() and RING_RESPONSE_PROD_OVERFLOW() macros.
Note that this will correct the wrong license info by adding the missing original copyright notice.
Signed-off-by: Juergen Gross jgross@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- include/xen/interface/io/ring.h | 293 +++++++++++++++++--------------- 1 file changed, 158 insertions(+), 135 deletions(-)
diff --git a/include/xen/interface/io/ring.h b/include/xen/interface/io/ring.h index 3f40501fc60b1..b39cdbc522ec7 100644 --- a/include/xen/interface/io/ring.h +++ b/include/xen/interface/io/ring.h @@ -1,21 +1,53 @@ -/* SPDX-License-Identifier: GPL-2.0 */ /****************************************************************************** * ring.h * * Shared producer-consumer ring macros. * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to + * deal in the Software without restriction, including without limitation the + * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or + * sell copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER + * DEALINGS IN THE SOFTWARE. + * * Tim Deegan and Andrew Warfield November 2004. */
#ifndef __XEN_PUBLIC_IO_RING_H__ #define __XEN_PUBLIC_IO_RING_H__
+/* + * When #include'ing this header, you need to provide the following + * declaration upfront: + * - standard integers types (uint8_t, uint16_t, etc) + * They are provided by stdint.h of the standard headers. + * + * In addition, if you intend to use the FLEX macros, you also need to + * provide the following, before invoking the FLEX macros: + * - size_t + * - memcpy + * - grant_ref_t + * These declarations are provided by string.h of the standard headers, + * and grant_table.h from the Xen public headers. + */ + #include <xen/interface/grant_table.h>
typedef unsigned int RING_IDX;
/* Round a 32-bit unsigned constant down to the nearest power of two. */ -#define __RD2(_x) (((_x) & 0x00000002) ? 0x2 : ((_x) & 0x1)) +#define __RD2(_x) (((_x) & 0x00000002) ? 0x2 : ((_x) & 0x1)) #define __RD4(_x) (((_x) & 0x0000000c) ? __RD2((_x)>>2)<<2 : __RD2(_x)) #define __RD8(_x) (((_x) & 0x000000f0) ? __RD4((_x)>>4)<<4 : __RD4(_x)) #define __RD16(_x) (((_x) & 0x0000ff00) ? __RD8((_x)>>8)<<8 : __RD8(_x)) @@ -27,82 +59,79 @@ typedef unsigned int RING_IDX; * A ring contains as many entries as will fit, rounded down to the nearest * power of two (so we can mask with (size-1) to loop around). */ -#define __CONST_RING_SIZE(_s, _sz) \ - (__RD32(((_sz) - offsetof(struct _s##_sring, ring)) / \ - sizeof(((struct _s##_sring *)0)->ring[0]))) - +#define __CONST_RING_SIZE(_s, _sz) \ + (__RD32(((_sz) - offsetof(struct _s##_sring, ring)) / \ + sizeof(((struct _s##_sring *)0)->ring[0]))) /* * The same for passing in an actual pointer instead of a name tag. */ -#define __RING_SIZE(_s, _sz) \ - (__RD32(((_sz) - (long)&(_s)->ring + (long)(_s)) / sizeof((_s)->ring[0]))) +#define __RING_SIZE(_s, _sz) \ + (__RD32(((_sz) - (long)(_s)->ring + (long)(_s)) / sizeof((_s)->ring[0])))
/* * Macros to make the correct C datatypes for a new kind of ring. * * To make a new ring datatype, you need to have two message structures, - * let's say struct request, and struct response already defined. + * let's say request_t, and response_t already defined. * * In a header where you want the ring datatype declared, you then do: * - * DEFINE_RING_TYPES(mytag, struct request, struct response); + * DEFINE_RING_TYPES(mytag, request_t, response_t); * * These expand out to give you a set of types, as you can see below. * The most important of these are: * - * struct mytag_sring - The shared ring. - * struct mytag_front_ring - The 'front' half of the ring. - * struct mytag_back_ring - The 'back' half of the ring. + * mytag_sring_t - The shared ring. + * mytag_front_ring_t - The 'front' half of the ring. + * mytag_back_ring_t - The 'back' half of the ring. * * To initialize a ring in your code you need to know the location and size * of the shared memory area (PAGE_SIZE, for instance). To initialise * the front half: * - * struct mytag_front_ring front_ring; - * SHARED_RING_INIT((struct mytag_sring *)shared_page); - * FRONT_RING_INIT(&front_ring, (struct mytag_sring *)shared_page, - * PAGE_SIZE); + * mytag_front_ring_t front_ring; + * SHARED_RING_INIT((mytag_sring_t *)shared_page); + * FRONT_RING_INIT(&front_ring, (mytag_sring_t *)shared_page, PAGE_SIZE); * * Initializing the back follows similarly (note that only the front * initializes the shared ring): * - * struct mytag_back_ring back_ring; - * BACK_RING_INIT(&back_ring, (struct mytag_sring *)shared_page, - * PAGE_SIZE); + * mytag_back_ring_t back_ring; + * BACK_RING_INIT(&back_ring, (mytag_sring_t *)shared_page, PAGE_SIZE); */
-#define DEFINE_RING_TYPES(__name, __req_t, __rsp_t) \ - \ -/* Shared ring entry */ \ -union __name##_sring_entry { \ - __req_t req; \ - __rsp_t rsp; \ -}; \ - \ -/* Shared ring page */ \ -struct __name##_sring { \ - RING_IDX req_prod, req_event; \ - RING_IDX rsp_prod, rsp_event; \ - uint8_t pad[48]; \ - union __name##_sring_entry ring[1]; /* variable-length */ \ -}; \ - \ -/* "Front" end's private variables */ \ -struct __name##_front_ring { \ - RING_IDX req_prod_pvt; \ - RING_IDX rsp_cons; \ - unsigned int nr_ents; \ - struct __name##_sring *sring; \ -}; \ - \ -/* "Back" end's private variables */ \ -struct __name##_back_ring { \ - RING_IDX rsp_prod_pvt; \ - RING_IDX req_cons; \ - unsigned int nr_ents; \ - struct __name##_sring *sring; \ -}; - +#define DEFINE_RING_TYPES(__name, __req_t, __rsp_t) \ + \ +/* Shared ring entry */ \ +union __name##_sring_entry { \ + __req_t req; \ + __rsp_t rsp; \ +}; \ + \ +/* Shared ring page */ \ +struct __name##_sring { \ + RING_IDX req_prod, req_event; \ + RING_IDX rsp_prod, rsp_event; \ + uint8_t __pad[48]; \ + union __name##_sring_entry ring[1]; /* variable-length */ \ +}; \ + \ +/* "Front" end's private variables */ \ +struct __name##_front_ring { \ + RING_IDX req_prod_pvt; \ + RING_IDX rsp_cons; \ + unsigned int nr_ents; \ + struct __name##_sring *sring; \ +}; \ + \ +/* "Back" end's private variables */ \ +struct __name##_back_ring { \ + RING_IDX rsp_prod_pvt; \ + RING_IDX req_cons; \ + unsigned int nr_ents; \ + struct __name##_sring *sring; \ +}; \ + \ /* * Macros for manipulating rings. * @@ -119,105 +148,99 @@ struct __name##_back_ring { \ */
/* Initialising empty rings */ -#define SHARED_RING_INIT(_s) do { \ - (_s)->req_prod = (_s)->rsp_prod = 0; \ - (_s)->req_event = (_s)->rsp_event = 1; \ - memset((_s)->pad, 0, sizeof((_s)->pad)); \ +#define SHARED_RING_INIT(_s) do { \ + (_s)->req_prod = (_s)->rsp_prod = 0; \ + (_s)->req_event = (_s)->rsp_event = 1; \ + (void)memset((_s)->__pad, 0, sizeof((_s)->__pad)); \ } while(0)
-#define FRONT_RING_INIT(_r, _s, __size) do { \ - (_r)->req_prod_pvt = 0; \ - (_r)->rsp_cons = 0; \ - (_r)->nr_ents = __RING_SIZE(_s, __size); \ - (_r)->sring = (_s); \ +#define FRONT_RING_ATTACH(_r, _s, _i, __size) do { \ + (_r)->req_prod_pvt = (_i); \ + (_r)->rsp_cons = (_i); \ + (_r)->nr_ents = __RING_SIZE(_s, __size); \ + (_r)->sring = (_s); \ } while (0)
-#define BACK_RING_INIT(_r, _s, __size) do { \ - (_r)->rsp_prod_pvt = 0; \ - (_r)->req_cons = 0; \ - (_r)->nr_ents = __RING_SIZE(_s, __size); \ - (_r)->sring = (_s); \ -} while (0) +#define FRONT_RING_INIT(_r, _s, __size) FRONT_RING_ATTACH(_r, _s, 0, __size)
-/* Initialize to existing shared indexes -- for recovery */ -#define FRONT_RING_ATTACH(_r, _s, __size) do { \ - (_r)->sring = (_s); \ - (_r)->req_prod_pvt = (_s)->req_prod; \ - (_r)->rsp_cons = (_s)->rsp_prod; \ - (_r)->nr_ents = __RING_SIZE(_s, __size); \ +#define BACK_RING_ATTACH(_r, _s, _i, __size) do { \ + (_r)->rsp_prod_pvt = (_i); \ + (_r)->req_cons = (_i); \ + (_r)->nr_ents = __RING_SIZE(_s, __size); \ + (_r)->sring = (_s); \ } while (0)
-#define BACK_RING_ATTACH(_r, _s, __size) do { \ - (_r)->sring = (_s); \ - (_r)->rsp_prod_pvt = (_s)->rsp_prod; \ - (_r)->req_cons = (_s)->req_prod; \ - (_r)->nr_ents = __RING_SIZE(_s, __size); \ -} while (0) +#define BACK_RING_INIT(_r, _s, __size) BACK_RING_ATTACH(_r, _s, 0, __size)
/* How big is this ring? */ -#define RING_SIZE(_r) \ +#define RING_SIZE(_r) \ ((_r)->nr_ents)
/* Number of free requests (for use on front side only). */ -#define RING_FREE_REQUESTS(_r) \ +#define RING_FREE_REQUESTS(_r) \ (RING_SIZE(_r) - ((_r)->req_prod_pvt - (_r)->rsp_cons))
/* Test if there is an empty slot available on the front ring. * (This is only meaningful from the front. ) */ -#define RING_FULL(_r) \ +#define RING_FULL(_r) \ (RING_FREE_REQUESTS(_r) == 0)
/* Test if there are outstanding messages to be processed on a ring. */ -#define RING_HAS_UNCONSUMED_RESPONSES(_r) \ +#define RING_HAS_UNCONSUMED_RESPONSES(_r) \ ((_r)->sring->rsp_prod - (_r)->rsp_cons)
-#define RING_HAS_UNCONSUMED_REQUESTS(_r) \ - ({ \ - unsigned int req = (_r)->sring->req_prod - (_r)->req_cons; \ - unsigned int rsp = RING_SIZE(_r) - \ - ((_r)->req_cons - (_r)->rsp_prod_pvt); \ - req < rsp ? req : rsp; \ - }) +#define RING_HAS_UNCONSUMED_REQUESTS(_r) ({ \ + unsigned int req = (_r)->sring->req_prod - (_r)->req_cons; \ + unsigned int rsp = RING_SIZE(_r) - \ + ((_r)->req_cons - (_r)->rsp_prod_pvt); \ + req < rsp ? req : rsp; \ +})
/* Direct access to individual ring elements, by index. */ -#define RING_GET_REQUEST(_r, _idx) \ +#define RING_GET_REQUEST(_r, _idx) \ (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].req))
+#define RING_GET_RESPONSE(_r, _idx) \ + (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].rsp)) + /* - * Get a local copy of a request. + * Get a local copy of a request/response. * - * Use this in preference to RING_GET_REQUEST() so all processing is + * Use this in preference to RING_GET_{REQUEST,RESPONSE}() so all processing is * done on a local copy that cannot be modified by the other end. * * Note that https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145 may cause this - * to be ineffective where _req is a struct which consists of only bitfields. + * to be ineffective where dest is a struct which consists of only bitfields. */ -#define RING_COPY_REQUEST(_r, _idx, _req) do { \ - /* Use volatile to force the copy into _req. */ \ - *(_req) = *(volatile typeof(_req))RING_GET_REQUEST(_r, _idx); \ +#define RING_COPY_(type, r, idx, dest) do { \ + /* Use volatile to force the copy into dest. */ \ + *(dest) = *(volatile typeof(dest))RING_GET_##type(r, idx); \ } while (0)
-#define RING_GET_RESPONSE(_r, _idx) \ - (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].rsp)) +#define RING_COPY_REQUEST(r, idx, req) RING_COPY_(REQUEST, r, idx, req) +#define RING_COPY_RESPONSE(r, idx, rsp) RING_COPY_(RESPONSE, r, idx, rsp)
/* Loop termination condition: Would the specified index overflow the ring? */ -#define RING_REQUEST_CONS_OVERFLOW(_r, _cons) \ +#define RING_REQUEST_CONS_OVERFLOW(_r, _cons) \ (((_cons) - (_r)->rsp_prod_pvt) >= RING_SIZE(_r))
/* Ill-behaved frontend determination: Can there be this many requests? */ -#define RING_REQUEST_PROD_OVERFLOW(_r, _prod) \ +#define RING_REQUEST_PROD_OVERFLOW(_r, _prod) \ (((_prod) - (_r)->rsp_prod_pvt) > RING_SIZE(_r))
+/* Ill-behaved backend determination: Can there be this many responses? */ +#define RING_RESPONSE_PROD_OVERFLOW(_r, _prod) \ + (((_prod) - (_r)->rsp_cons) > RING_SIZE(_r))
-#define RING_PUSH_REQUESTS(_r) do { \ - virt_wmb(); /* back sees requests /before/ updated producer index */ \ - (_r)->sring->req_prod = (_r)->req_prod_pvt; \ +#define RING_PUSH_REQUESTS(_r) do { \ + virt_wmb(); /* back sees requests /before/ updated producer index */\ + (_r)->sring->req_prod = (_r)->req_prod_pvt; \ } while (0)
-#define RING_PUSH_RESPONSES(_r) do { \ - virt_wmb(); /* front sees responses /before/ updated producer index */ \ - (_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \ +#define RING_PUSH_RESPONSES(_r) do { \ + virt_wmb(); /* front sees resps /before/ updated producer index */ \ + (_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \ } while (0)
/* @@ -250,40 +273,40 @@ struct __name##_back_ring { \ * field appropriately. */
-#define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \ - RING_IDX __old = (_r)->sring->req_prod; \ - RING_IDX __new = (_r)->req_prod_pvt; \ - virt_wmb(); /* back sees requests /before/ updated producer index */ \ - (_r)->sring->req_prod = __new; \ - virt_mb(); /* back sees new requests /before/ we check req_event */ \ - (_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \ - (RING_IDX)(__new - __old)); \ +#define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \ + RING_IDX __old = (_r)->sring->req_prod; \ + RING_IDX __new = (_r)->req_prod_pvt; \ + virt_wmb(); /* back sees requests /before/ updated producer index */\ + (_r)->sring->req_prod = __new; \ + virt_mb(); /* back sees new requests /before/ we check req_event */ \ + (_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \ + (RING_IDX)(__new - __old)); \ } while (0)
-#define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \ - RING_IDX __old = (_r)->sring->rsp_prod; \ - RING_IDX __new = (_r)->rsp_prod_pvt; \ - virt_wmb(); /* front sees responses /before/ updated producer index */ \ - (_r)->sring->rsp_prod = __new; \ - virt_mb(); /* front sees new responses /before/ we check rsp_event */ \ - (_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \ - (RING_IDX)(__new - __old)); \ +#define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \ + RING_IDX __old = (_r)->sring->rsp_prod; \ + RING_IDX __new = (_r)->rsp_prod_pvt; \ + virt_wmb(); /* front sees resps /before/ updated producer index */ \ + (_r)->sring->rsp_prod = __new; \ + virt_mb(); /* front sees new resps /before/ we check rsp_event */ \ + (_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \ + (RING_IDX)(__new - __old)); \ } while (0)
-#define RING_FINAL_CHECK_FOR_REQUESTS(_r, _work_to_do) do { \ - (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ - if (_work_to_do) break; \ - (_r)->sring->req_event = (_r)->req_cons + 1; \ - virt_mb(); \ - (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ +#define RING_FINAL_CHECK_FOR_REQUESTS(_r, _work_to_do) do { \ + (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ + if (_work_to_do) break; \ + (_r)->sring->req_event = (_r)->req_cons + 1; \ + virt_mb(); \ + (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \ } while (0)
-#define RING_FINAL_CHECK_FOR_RESPONSES(_r, _work_to_do) do { \ - (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ - if (_work_to_do) break; \ - (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \ - virt_mb(); \ - (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ +#define RING_FINAL_CHECK_FOR_RESPONSES(_r, _work_to_do) do { \ + (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ + if (_work_to_do) break; \ + (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \ + virt_mb(); \ + (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \ } while (0)
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit b647c449f166a99fd5a011e4a2bc6f8b9fcd3222
--------------------------------
commit 71b66243f9898d0e54296b4e7035fb33cdcb0707 upstream.
In order to avoid problems in case the backend is modifying a response on the ring page while the frontend has already seen it, just read the response into a local buffer in one go and then operate on that buffer only.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Acked-by: Roger Pau Monné roger.pau@citrix.com Link: https://lore.kernel.org/r/20210730103854.12681-2-jgross@suse.com Signed-off-by: Juergen Gross jgross@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/block/xen-blkfront.c | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-)
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 5830d94178862..1cd470cbf29d3 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1547,7 +1547,7 @@ static bool blkif_completion(unsigned long *id, static irqreturn_t blkif_interrupt(int irq, void *dev_id) { struct request *req; - struct blkif_response *bret; + struct blkif_response bret; RING_IDX i, rp; unsigned long flags; struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)dev_id; @@ -1564,8 +1564,9 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) for (i = rinfo->ring.rsp_cons; i != rp; i++) { unsigned long id;
- bret = RING_GET_RESPONSE(&rinfo->ring, i); - id = bret->id; + RING_COPY_RESPONSE(&rinfo->ring, i, &bret); + id = bret.id; + /* * The backend has messed up and given us an id that we would * never have given to it (we stamp it up to BLK_RING_SIZE - @@ -1573,39 +1574,39 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) */ if (id >= BLK_RING_SIZE(info)) { WARN(1, "%s: response to %s has incorrect id (%ld)\n", - info->gd->disk_name, op_name(bret->operation), id); + info->gd->disk_name, op_name(bret.operation), id); /* We can't safely get the 'struct request' as * the id is busted. */ continue; } req = rinfo->shadow[id].request;
- if (bret->operation != BLKIF_OP_DISCARD) { + if (bret.operation != BLKIF_OP_DISCARD) { /* * We may need to wait for an extra response if the * I/O request is split in 2 */ - if (!blkif_completion(&id, rinfo, bret)) + if (!blkif_completion(&id, rinfo, &bret)) continue; }
if (add_id_to_freelist(rinfo, id)) { WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n", - info->gd->disk_name, op_name(bret->operation), id); + info->gd->disk_name, op_name(bret.operation), id); continue; }
- if (bret->status == BLKIF_RSP_OKAY) + if (bret.status == BLKIF_RSP_OKAY) blkif_req(req)->error = BLK_STS_OK; else blkif_req(req)->error = BLK_STS_IOERR;
- switch (bret->operation) { + switch (bret.operation) { case BLKIF_OP_DISCARD: - if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { + if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) { struct request_queue *rq = info->rq; printk(KERN_WARNING "blkfront: %s: %s op failed\n", - info->gd->disk_name, op_name(bret->operation)); + info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; info->feature_discard = 0; info->feature_secdiscard = 0; @@ -1615,15 +1616,15 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) break; case BLKIF_OP_FLUSH_DISKCACHE: case BLKIF_OP_WRITE_BARRIER: - if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { + if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) { printk(KERN_WARNING "blkfront: %s: %s op failed\n", - info->gd->disk_name, op_name(bret->operation)); + info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; } - if (unlikely(bret->status == BLKIF_RSP_ERROR && + if (unlikely(bret.status == BLKIF_RSP_ERROR && rinfo->shadow[id].req.u.rw.nr_segments == 0)) { printk(KERN_WARNING "blkfront: %s: empty %s op failed\n", - info->gd->disk_name, op_name(bret->operation)); + info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; } if (unlikely(blkif_req(req)->error)) { @@ -1636,9 +1637,9 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) /* fall through */ case BLKIF_OP_READ: case BLKIF_OP_WRITE: - if (unlikely(bret->status != BLKIF_RSP_OKAY)) + if (unlikely(bret.status != BLKIF_RSP_OKAY)) dev_dbg(&info->xbdev->dev, "Bad return from blkdev data " - "request: %x\n", bret->status); + "request: %x\n", bret.status);
break; default:
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit 80baa77511eee363140b76265086f7f88f872877
--------------------------------
commit 8f5a695d99000fc3aa73934d7ced33cfc64dcdab upstream.
In order to avoid a malicious backend being able to influence the local copy of a request build the request locally first and then copy it to the ring page instead of doing it the other way round as today.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Acked-by: Roger Pau Monné roger.pau@citrix.com Link: https://lore.kernel.org/r/20210730103854.12681-3-jgross@suse.com Signed-off-by: Juergen Gross jgross@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/block/xen-blkfront.c | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 1cd470cbf29d3..7942f70cc2e9e 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -535,7 +535,7 @@ static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo, rinfo->shadow[id].status = REQ_WAITING; rinfo->shadow[id].associated_id = NO_ASSOCIATED_ID;
- (*ring_req)->u.rw.id = id; + rinfo->shadow[id].req.u.rw.id = id;
return id; } @@ -543,11 +543,12 @@ static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo, static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_info *rinfo) { struct blkfront_info *info = rinfo->dev_info; - struct blkif_request *ring_req; + struct blkif_request *ring_req, *final_ring_req; unsigned long id;
/* Fill out a communications ring structure. */ - id = blkif_ring_get_request(rinfo, req, &ring_req); + id = blkif_ring_get_request(rinfo, req, &final_ring_req); + ring_req = &rinfo->shadow[id].req;
ring_req->operation = BLKIF_OP_DISCARD; ring_req->u.discard.nr_sectors = blk_rq_sectors(req); @@ -558,8 +559,8 @@ static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_inf else ring_req->u.discard.flag = 0;
- /* Keep a private copy so we can reissue requests when recovering. */ - rinfo->shadow[id].req = *ring_req; + /* Copy the request to the ring page. */ + *final_ring_req = *ring_req;
return 0; } @@ -692,6 +693,7 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri { struct blkfront_info *info = rinfo->dev_info; struct blkif_request *ring_req, *extra_ring_req = NULL; + struct blkif_request *final_ring_req, *final_extra_ring_req = NULL; unsigned long id, extra_id = NO_ASSOCIATED_ID; bool require_extra_req = false; int i; @@ -736,7 +738,8 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri }
/* Fill out a communications ring structure. */ - id = blkif_ring_get_request(rinfo, req, &ring_req); + id = blkif_ring_get_request(rinfo, req, &final_ring_req); + ring_req = &rinfo->shadow[id].req;
num_sg = blk_rq_map_sg(req->q, req, rinfo->shadow[id].sg); num_grant = 0; @@ -787,7 +790,9 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri ring_req->u.rw.nr_segments = num_grant; if (unlikely(require_extra_req)) { extra_id = blkif_ring_get_request(rinfo, req, - &extra_ring_req); + &final_extra_ring_req); + extra_ring_req = &rinfo->shadow[extra_id].req; + /* * Only the first request contains the scatter-gather * list. @@ -829,10 +834,10 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri if (setup.segments) kunmap_atomic(setup.segments);
- /* Keep a private copy so we can reissue requests when recovering. */ - rinfo->shadow[id].req = *ring_req; + /* Copy request(s) to the ring page. */ + *final_ring_req = *ring_req; if (unlikely(require_extra_req)) - rinfo->shadow[extra_id].req = *extra_ring_req; + *final_extra_ring_req = *extra_ring_req;
if (new_persistent_gnts) gnttab_free_grant_references(setup.gref_head);
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit f89a05402f7403b2b9e5ef588a16c29f0d5c72a1
--------------------------------
commit b94e4b147fd1992ad450e1fea1fdaa3738753373 upstream.
Today blkfront will trust the backend to send only sane response data. In order to avoid privilege escalations or crashes in case of malicious backends verify the data to be within expected limits. Especially make sure that the response always references an outstanding request.
Introduce a new state of the ring BLKIF_STATE_ERROR which will be switched to in case an inconsistency is being detected. Recovering from this state is possible only via removing and adding the virtual device again (e.g. via a suspend/resume cycle).
Make all warning messages issued due to valid error responses rate limited in order to avoid message floods being triggered by a malicious backend.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Acked-by: Roger Pau Monné roger.pau@citrix.com Link: https://lore.kernel.org/r/20210730103854.12681-4-jgross@suse.com Signed-off-by: Juergen Gross jgross@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/block/xen-blkfront.c | 70 +++++++++++++++++++++++++++--------- 1 file changed, 53 insertions(+), 17 deletions(-)
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 7942f70cc2e9e..90d6e13192bda 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -79,6 +79,7 @@ enum blkif_state { BLKIF_STATE_DISCONNECTED, BLKIF_STATE_CONNECTED, BLKIF_STATE_SUSPENDED, + BLKIF_STATE_ERROR, };
struct grant { @@ -88,6 +89,7 @@ struct grant { };
enum blk_req_status { + REQ_PROCESSING, REQ_WAITING, REQ_DONE, REQ_ERROR, @@ -532,7 +534,7 @@ static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo,
id = get_id_from_freelist(rinfo); rinfo->shadow[id].request = req; - rinfo->shadow[id].status = REQ_WAITING; + rinfo->shadow[id].status = REQ_PROCESSING; rinfo->shadow[id].associated_id = NO_ASSOCIATED_ID;
rinfo->shadow[id].req.u.rw.id = id; @@ -561,6 +563,7 @@ static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_inf
/* Copy the request to the ring page. */ *final_ring_req = *ring_req; + rinfo->shadow[id].status = REQ_WAITING;
return 0; } @@ -836,8 +839,11 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri
/* Copy request(s) to the ring page. */ *final_ring_req = *ring_req; - if (unlikely(require_extra_req)) + rinfo->shadow[id].status = REQ_WAITING; + if (unlikely(require_extra_req)) { *final_extra_ring_req = *extra_ring_req; + rinfo->shadow[extra_id].status = REQ_WAITING; + }
if (new_persistent_gnts) gnttab_free_grant_references(setup.gref_head); @@ -1410,8 +1416,8 @@ static enum blk_req_status blkif_rsp_to_req_status(int rsp) static int blkif_get_final_status(enum blk_req_status s1, enum blk_req_status s2) { - BUG_ON(s1 == REQ_WAITING); - BUG_ON(s2 == REQ_WAITING); + BUG_ON(s1 < REQ_DONE); + BUG_ON(s2 < REQ_DONE);
if (s1 == REQ_ERROR || s2 == REQ_ERROR) return BLKIF_RSP_ERROR; @@ -1444,7 +1450,7 @@ static bool blkif_completion(unsigned long *id, s->status = blkif_rsp_to_req_status(bret->status);
/* Wait the second response if not yet here. */ - if (s2->status == REQ_WAITING) + if (s2->status < REQ_DONE) return false;
bret->status = blkif_get_final_status(s->status, @@ -1563,11 +1569,17 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
spin_lock_irqsave(&rinfo->ring_lock, flags); again: - rp = rinfo->ring.sring->rsp_prod; - rmb(); /* Ensure we see queued responses up to 'rp'. */ + rp = READ_ONCE(rinfo->ring.sring->rsp_prod); + virt_rmb(); /* Ensure we see queued responses up to 'rp'. */ + if (RING_RESPONSE_PROD_OVERFLOW(&rinfo->ring, rp)) { + pr_alert("%s: illegal number of responses %u\n", + info->gd->disk_name, rp - rinfo->ring.rsp_cons); + goto err; + }
for (i = rinfo->ring.rsp_cons; i != rp; i++) { unsigned long id; + unsigned int op;
RING_COPY_RESPONSE(&rinfo->ring, i, &bret); id = bret.id; @@ -1578,14 +1590,28 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) * look in get_id_from_freelist. */ if (id >= BLK_RING_SIZE(info)) { - WARN(1, "%s: response to %s has incorrect id (%ld)\n", - info->gd->disk_name, op_name(bret.operation), id); - /* We can't safely get the 'struct request' as - * the id is busted. */ - continue; + pr_alert("%s: response has incorrect id (%ld)\n", + info->gd->disk_name, id); + goto err; + } + if (rinfo->shadow[id].status != REQ_WAITING) { + pr_alert("%s: response references no pending request\n", + info->gd->disk_name); + goto err; } + + rinfo->shadow[id].status = REQ_PROCESSING; req = rinfo->shadow[id].request;
+ op = rinfo->shadow[id].req.operation; + if (op == BLKIF_OP_INDIRECT) + op = rinfo->shadow[id].req.u.indirect.indirect_op; + if (bret.operation != op) { + pr_alert("%s: response has wrong operation (%u instead of %u)\n", + info->gd->disk_name, bret.operation, op); + goto err; + } + if (bret.operation != BLKIF_OP_DISCARD) { /* * We may need to wait for an extra response if the @@ -1610,7 +1636,8 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) case BLKIF_OP_DISCARD: if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) { struct request_queue *rq = info->rq; - printk(KERN_WARNING "blkfront: %s: %s op failed\n", + + pr_warn_ratelimited("blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; info->feature_discard = 0; @@ -1622,13 +1649,13 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) case BLKIF_OP_FLUSH_DISKCACHE: case BLKIF_OP_WRITE_BARRIER: if (unlikely(bret.status == BLKIF_RSP_EOPNOTSUPP)) { - printk(KERN_WARNING "blkfront: %s: %s op failed\n", + pr_warn_ratelimited("blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; } if (unlikely(bret.status == BLKIF_RSP_ERROR && rinfo->shadow[id].req.u.rw.nr_segments == 0)) { - printk(KERN_WARNING "blkfront: %s: empty %s op failed\n", + pr_warn_ratelimited("blkfront: %s: empty %s op failed\n", info->gd->disk_name, op_name(bret.operation)); blkif_req(req)->error = BLK_STS_NOTSUPP; } @@ -1643,8 +1670,9 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) case BLKIF_OP_READ: case BLKIF_OP_WRITE: if (unlikely(bret.status != BLKIF_RSP_OKAY)) - dev_dbg(&info->xbdev->dev, "Bad return from blkdev data " - "request: %x\n", bret.status); + dev_dbg_ratelimited(&info->xbdev->dev, + "Bad return from blkdev data request: %#x\n", + bret.status);
break; default: @@ -1669,6 +1697,14 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) spin_unlock_irqrestore(&rinfo->ring_lock, flags);
return IRQ_HANDLED; + + err: + info->connected = BLKIF_STATE_ERROR; + + spin_unlock_irqrestore(&rinfo->ring_lock, flags); + + pr_alert("%s disabled for further use\n", info->gd->disk_name); + return IRQ_HANDLED; }
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit e7d1024f5b19e76ae7ff6789998cefb91c2b6a61
--------------------------------
commit 8446066bf8c1f9f7b7412c43fbea0fb87464d75b upstream.
In order to avoid problems in case the backend is modifying a response on the ring page while the frontend has already seen it, just read the response into a local buffer in one go and then operate on that buffer only.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netfront.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 6b4675a9494b2..424d35b3842ca 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -387,13 +387,13 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) rmb(); /* Ensure we see responses up to 'rp'. */
for (cons = queue->tx.rsp_cons; cons != prod; cons++) { - struct xen_netif_tx_response *txrsp; + struct xen_netif_tx_response txrsp;
- txrsp = RING_GET_RESPONSE(&queue->tx, cons); - if (txrsp->status == XEN_NETIF_RSP_NULL) + RING_COPY_RESPONSE(&queue->tx, cons, &txrsp); + if (txrsp.status == XEN_NETIF_RSP_NULL) continue;
- id = txrsp->id; + id = txrsp.id; skb = queue->tx_skbs[id].skb; if (unlikely(gnttab_query_foreign_access( queue->grant_tx_ref[id]) != 0)) { @@ -742,7 +742,7 @@ static int xennet_get_extras(struct netfront_queue *queue, RING_IDX rp)
{ - struct xen_netif_extra_info *extra; + struct xen_netif_extra_info extra; struct device *dev = &queue->info->netdev->dev; RING_IDX cons = queue->rx.rsp_cons; int err = 0; @@ -758,24 +758,22 @@ static int xennet_get_extras(struct netfront_queue *queue, break; }
- extra = (struct xen_netif_extra_info *) - RING_GET_RESPONSE(&queue->rx, ++cons); + RING_COPY_RESPONSE(&queue->rx, ++cons, &extra);
- if (unlikely(!extra->type || - extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) { + if (unlikely(!extra.type || + extra.type >= XEN_NETIF_EXTRA_TYPE_MAX)) { if (net_ratelimit()) dev_warn(dev, "Invalid extra type: %d\n", - extra->type); + extra.type); err = -EINVAL; } else { - memcpy(&extras[extra->type - 1], extra, - sizeof(*extra)); + extras[extra.type - 1] = extra; }
skb = xennet_get_rx_skb(queue, cons); ref = xennet_get_rx_ref(queue, cons); xennet_move_rx_slot(queue, skb, ref); - } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE); + } while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE);
queue->rx.rsp_cons = cons; return err; @@ -785,7 +783,7 @@ static int xennet_get_responses(struct netfront_queue *queue, struct netfront_rx_info *rinfo, RING_IDX rp, struct sk_buff_head *list) { - struct xen_netif_rx_response *rx = &rinfo->rx; + struct xen_netif_rx_response *rx = &rinfo->rx, rx_local; struct xen_netif_extra_info *extras = rinfo->extras; struct device *dev = &queue->info->netdev->dev; RING_IDX cons = queue->rx.rsp_cons; @@ -843,7 +841,8 @@ static int xennet_get_responses(struct netfront_queue *queue, break; }
- rx = RING_GET_RESPONSE(&queue->rx, cons + slots); + RING_COPY_RESPONSE(&queue->rx, cons + slots, &rx_local); + rx = &rx_local; skb = xennet_get_rx_skb(queue, cons + slots); ref = xennet_get_rx_ref(queue, cons + slots); slots++; @@ -898,10 +897,11 @@ static int xennet_fill_frags(struct netfront_queue *queue, struct sk_buff *nskb;
while ((nskb = __skb_dequeue(list))) { - struct xen_netif_rx_response *rx = - RING_GET_RESPONSE(&queue->rx, ++cons); + struct xen_netif_rx_response rx; skb_frag_t *nfrag = &skb_shinfo(nskb)->frags[0];
+ RING_COPY_RESPONSE(&queue->rx, ++cons, &rx); + if (skb_shinfo(skb)->nr_frags == MAX_SKB_FRAGS) { unsigned int pull_to = NETFRONT_SKB_CB(skb)->pull_to;
@@ -916,7 +916,7 @@ static int xennet_fill_frags(struct netfront_queue *queue,
skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, skb_frag_page(nfrag), - rx->offset, rx->status, PAGE_SIZE); + rx.offset, rx.status, PAGE_SIZE);
skb_shinfo(nskb)->nr_frags = 0; kfree_skb(nskb); @@ -1014,7 +1014,7 @@ static int xennet_poll(struct napi_struct *napi, int budget) i = queue->rx.rsp_cons; work_done = 0; while ((i != rp) && (work_done < budget)) { - memcpy(rx, RING_GET_RESPONSE(&queue->rx, i), sizeof(*rx)); + RING_COPY_RESPONSE(&queue->rx, i, rx); memset(extras, 0, sizeof(rinfo.extras));
err = xennet_get_responses(queue, &rinfo, rp, &tmpq);
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit 26509bb5dd2fc9177fe612a778cc4b8f305ea89f
--------------------------------
commit 162081ec33c2686afa29d91bf8d302824aa846c7 upstream.
In order to avoid a malicious backend being able to influence the local processing of a request build the request locally first and then copy it to the ring page. Any reading from the request influencing the processing in the frontend needs to be done on the local instance.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netfront.c | 80 ++++++++++++++++++-------------------- 1 file changed, 37 insertions(+), 43 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 424d35b3842ca..4f0d13a49b17e 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -423,7 +423,8 @@ struct xennet_gnttab_make_txreq { struct netfront_queue *queue; struct sk_buff *skb; struct page *page; - struct xen_netif_tx_request *tx; /* Last request */ + struct xen_netif_tx_request *tx; /* Last request on ring page */ + struct xen_netif_tx_request tx_local; /* Last request local copy*/ unsigned int size; };
@@ -451,30 +452,27 @@ static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset, queue->grant_tx_page[id] = page; queue->grant_tx_ref[id] = ref;
- tx->id = id; - tx->gref = ref; - tx->offset = offset; - tx->size = len; - tx->flags = 0; + info->tx_local.id = id; + info->tx_local.gref = ref; + info->tx_local.offset = offset; + info->tx_local.size = len; + info->tx_local.flags = 0; + + *tx = info->tx_local;
info->tx = tx; - info->size += tx->size; + info->size += info->tx_local.size; }
static struct xen_netif_tx_request *xennet_make_first_txreq( - struct netfront_queue *queue, struct sk_buff *skb, - struct page *page, unsigned int offset, unsigned int len) + struct xennet_gnttab_make_txreq *info, + unsigned int offset, unsigned int len) { - struct xennet_gnttab_make_txreq info = { - .queue = queue, - .skb = skb, - .page = page, - .size = 0, - }; + info->size = 0;
- gnttab_for_one_grant(page, offset, len, xennet_tx_setup_grant, &info); + gnttab_for_one_grant(info->page, offset, len, xennet_tx_setup_grant, info);
- return info.tx; + return info->tx; }
static void xennet_make_one_txreq(unsigned long gfn, unsigned int offset, @@ -487,35 +485,27 @@ static void xennet_make_one_txreq(unsigned long gfn, unsigned int offset, xennet_tx_setup_grant(gfn, offset, len, data); }
-static struct xen_netif_tx_request *xennet_make_txreqs( - struct netfront_queue *queue, struct xen_netif_tx_request *tx, - struct sk_buff *skb, struct page *page, +static void xennet_make_txreqs( + struct xennet_gnttab_make_txreq *info, + struct page *page, unsigned int offset, unsigned int len) { - struct xennet_gnttab_make_txreq info = { - .queue = queue, - .skb = skb, - .tx = tx, - }; - /* Skip unused frames from start of page */ page += offset >> PAGE_SHIFT; offset &= ~PAGE_MASK;
while (len) { - info.page = page; - info.size = 0; + info->page = page; + info->size = 0;
gnttab_foreach_grant_in_range(page, offset, len, xennet_make_one_txreq, - &info); + info);
page++; offset = 0; - len -= info.size; + len -= info->size; } - - return info.tx; }
/* @@ -569,7 +559,7 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev { struct netfront_info *np = netdev_priv(dev); struct netfront_stats *tx_stats = this_cpu_ptr(np->tx_stats); - struct xen_netif_tx_request *tx, *first_tx; + struct xen_netif_tx_request *first_tx; unsigned int i; int notify; int slots; @@ -578,6 +568,7 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev unsigned int len; unsigned long flags; struct netfront_queue *queue = NULL; + struct xennet_gnttab_make_txreq info = { }; unsigned int num_queues = dev->real_num_tx_queues; u16 queue_index; struct sk_buff *nskb; @@ -635,21 +626,24 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev }
/* First request for the linear area. */ - first_tx = tx = xennet_make_first_txreq(queue, skb, - page, offset, len); - offset += tx->size; + info.queue = queue; + info.skb = skb; + info.page = page; + first_tx = xennet_make_first_txreq(&info, offset, len); + offset += info.tx_local.size; if (offset == PAGE_SIZE) { page++; offset = 0; } - len -= tx->size; + len -= info.tx_local.size;
if (skb->ip_summed == CHECKSUM_PARTIAL) /* local packet? */ - tx->flags |= XEN_NETTXF_csum_blank | XEN_NETTXF_data_validated; + first_tx->flags |= XEN_NETTXF_csum_blank | + XEN_NETTXF_data_validated; else if (skb->ip_summed == CHECKSUM_UNNECESSARY) /* remote but checksummed. */ - tx->flags |= XEN_NETTXF_data_validated; + first_tx->flags |= XEN_NETTXF_data_validated;
/* Optional extra info after the first request. */ if (skb_shinfo(skb)->gso_size) { @@ -658,7 +652,7 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev gso = (struct xen_netif_extra_info *) RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++);
- tx->flags |= XEN_NETTXF_extra_info; + first_tx->flags |= XEN_NETTXF_extra_info;
gso->u.gso.size = skb_shinfo(skb)->gso_size; gso->u.gso.type = (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) ? @@ -672,13 +666,13 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev }
/* Requests for the rest of the linear area. */ - tx = xennet_make_txreqs(queue, tx, skb, page, offset, len); + xennet_make_txreqs(&info, page, offset, len);
/* Requests for all the frags. */ for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; - tx = xennet_make_txreqs(queue, tx, skb, - skb_frag_page(frag), frag->page_offset, + xennet_make_txreqs(&info, skb_frag_page(frag), + frag->page_offset, skb_frag_size(frag)); }
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit e52c0efbd23ca50e4afa76e6ec3c8c83b032c823
--------------------------------
commit 21631d2d741a64a073e167c27769e73bc7844a2f upstream.
The tx_skb_freelist elements are in a single linked list with the request id used as link reference. The per element link field is in a union with the skb pointer of an in use request.
Move the link reference out of the union in order to enable a later reuse of it for requests which need a populated skb pointer.
Rename add_id_to_freelist() and get_id_from_freelist() to add_id_to_list() and get_id_from_list() in order to prepare using those for other lists as well. Define ~0 as value to indicate the end of a list and place that value into the link for a request not being on the list.
When freeing a skb zero the skb pointer in the request. Use a NULL value of the skb pointer instead of skb_entry_is_link() for deciding whether a request has a skb linked to it.
Remove skb_entry_set_link() and open code it instead as it is really trivial now.
Signed-off-by: Juergen Gross jgross@suse.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netfront.c | 61 ++++++++++++++++---------------------- 1 file changed, 25 insertions(+), 36 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index 4f0d13a49b17e..e5815e2b55d55 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -119,17 +119,11 @@ struct netfront_queue {
/* * {tx,rx}_skbs store outstanding skbuffs. Free tx_skb entries - * are linked from tx_skb_freelist through skb_entry.link. - * - * NB. Freelist index entries are always going to be less than - * PAGE_OFFSET, whereas pointers to skbs will always be equal or - * greater than PAGE_OFFSET: we use this property to distinguish - * them. + * are linked from tx_skb_freelist through tx_link. */ - union skb_entry { - struct sk_buff *skb; - unsigned long link; - } tx_skbs[NET_TX_RING_SIZE]; + struct sk_buff *tx_skbs[NET_TX_RING_SIZE]; + unsigned short tx_link[NET_TX_RING_SIZE]; +#define TX_LINK_NONE 0xffff grant_ref_t gref_tx_head; grant_ref_t grant_tx_ref[NET_TX_RING_SIZE]; struct page *grant_tx_page[NET_TX_RING_SIZE]; @@ -167,33 +161,25 @@ struct netfront_rx_info { struct xen_netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1]; };
-static void skb_entry_set_link(union skb_entry *list, unsigned short id) -{ - list->link = id; -} - -static int skb_entry_is_link(const union skb_entry *list) -{ - BUILD_BUG_ON(sizeof(list->skb) != sizeof(list->link)); - return (unsigned long)list->skb < PAGE_OFFSET; -} - /* * Access macros for acquiring freeing slots in tx_skbs[]. */
-static void add_id_to_freelist(unsigned *head, union skb_entry *list, - unsigned short id) +static void add_id_to_list(unsigned *head, unsigned short *list, + unsigned short id) { - skb_entry_set_link(&list[id], *head); + list[id] = *head; *head = id; }
-static unsigned short get_id_from_freelist(unsigned *head, - union skb_entry *list) +static unsigned short get_id_from_list(unsigned *head, unsigned short *list) { unsigned int id = *head; - *head = list[id].link; + + if (id != TX_LINK_NONE) { + *head = list[id]; + list[id] = TX_LINK_NONE; + } return id; }
@@ -394,7 +380,8 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) continue;
id = txrsp.id; - skb = queue->tx_skbs[id].skb; + skb = queue->tx_skbs[id]; + queue->tx_skbs[id] = NULL; if (unlikely(gnttab_query_foreign_access( queue->grant_tx_ref[id]) != 0)) { pr_alert("%s: warning -- grant still in use by backend domain\n", @@ -407,7 +394,7 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) &queue->gref_tx_head, queue->grant_tx_ref[id]); queue->grant_tx_ref[id] = GRANT_INVALID_REF; queue->grant_tx_page[id] = NULL; - add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, id); + add_id_to_list(&queue->tx_skb_freelist, queue->tx_link, id); dev_kfree_skb_irq(skb); }
@@ -440,7 +427,7 @@ static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset, struct netfront_queue *queue = info->queue; struct sk_buff *skb = info->skb;
- id = get_id_from_freelist(&queue->tx_skb_freelist, queue->tx_skbs); + id = get_id_from_list(&queue->tx_skb_freelist, queue->tx_link); tx = RING_GET_REQUEST(&queue->tx, queue->tx.req_prod_pvt++); ref = gnttab_claim_grant_reference(&queue->gref_tx_head); WARN_ON_ONCE(IS_ERR_VALUE((unsigned long)(int)ref)); @@ -448,7 +435,7 @@ static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset, gnttab_grant_foreign_access_ref(ref, queue->info->xbdev->otherend_id, gfn, GNTMAP_readonly);
- queue->tx_skbs[id].skb = skb; + queue->tx_skbs[id] = skb; queue->grant_tx_page[id] = page; queue->grant_tx_ref[id] = ref;
@@ -1130,17 +1117,18 @@ static void xennet_release_tx_bufs(struct netfront_queue *queue)
for (i = 0; i < NET_TX_RING_SIZE; i++) { /* Skip over entries which are actually freelist references */ - if (skb_entry_is_link(&queue->tx_skbs[i])) + if (!queue->tx_skbs[i]) continue;
- skb = queue->tx_skbs[i].skb; + skb = queue->tx_skbs[i]; + queue->tx_skbs[i] = NULL; get_page(queue->grant_tx_page[i]); gnttab_end_foreign_access(queue->grant_tx_ref[i], GNTMAP_readonly, (unsigned long)page_address(queue->grant_tx_page[i])); queue->grant_tx_page[i] = NULL; queue->grant_tx_ref[i] = GRANT_INVALID_REF; - add_id_to_freelist(&queue->tx_skb_freelist, queue->tx_skbs, i); + add_id_to_list(&queue->tx_skb_freelist, queue->tx_link, i); dev_kfree_skb_irq(skb); } } @@ -1615,13 +1603,14 @@ static int xennet_init_queue(struct netfront_queue *queue) snprintf(queue->name, sizeof(queue->name), "vif%s-q%u", devid, queue->id);
- /* Initialise tx_skbs as a free chain containing every entry. */ + /* Initialise tx_skb_freelist as a free chain containing every entry. */ queue->tx_skb_freelist = 0; for (i = 0; i < NET_TX_RING_SIZE; i++) { - skb_entry_set_link(&queue->tx_skbs[i], i+1); + queue->tx_link[i] = i + 1; queue->grant_tx_ref[i] = GRANT_INVALID_REF; queue->grant_tx_page[i] = NULL; } + queue->tx_link[NET_TX_RING_SIZE - 1] = TX_LINK_NONE;
/* Clear out rx_skbs */ for (i = 0; i < NET_RX_RING_SIZE; i++) {
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.219 commit e4e01b0e4b7fca40586639647f83eaa21a0a8203
--------------------------------
commit a884daa61a7d91650987e855464526aef219590f upstream.
Today netfront will trust the backend to send only sane response data. In order to avoid privilege escalations or crashes in case of malicious backends verify the data to be within expected limits. Especially make sure that the response always references an outstanding request.
Note that only the tx queue needs special id handling, as for the rx queue the id is equal to the index in the ring page.
Introduce a new indicator for the device whether it is broken and let the device stop working when it is set. Set this indicator in case the backend sets any weird data.
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netfront.c | 80 +++++++++++++++++++++++++++++++++++--- 1 file changed, 75 insertions(+), 5 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index e5815e2b55d55..e5bbd2d4252dc 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -124,10 +124,12 @@ struct netfront_queue { struct sk_buff *tx_skbs[NET_TX_RING_SIZE]; unsigned short tx_link[NET_TX_RING_SIZE]; #define TX_LINK_NONE 0xffff +#define TX_PENDING 0xfffe grant_ref_t gref_tx_head; grant_ref_t grant_tx_ref[NET_TX_RING_SIZE]; struct page *grant_tx_page[NET_TX_RING_SIZE]; unsigned tx_skb_freelist; + unsigned int tx_pend_queue;
spinlock_t rx_lock ____cacheline_aligned_in_smp; struct xen_netif_rx_front_ring rx; @@ -153,6 +155,9 @@ struct netfront_info { struct netfront_stats __percpu *rx_stats; struct netfront_stats __percpu *tx_stats;
+ /* Is device behaving sane? */ + bool broken; + atomic_t rx_gso_checksum_fixup; };
@@ -337,7 +342,7 @@ static int xennet_open(struct net_device *dev) unsigned int i = 0; struct netfront_queue *queue = NULL;
- if (!np->queues) + if (!np->queues || np->broken) return -ENODEV;
for (i = 0; i < num_queues; ++i) { @@ -365,11 +370,17 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) unsigned short id; struct sk_buff *skb; bool more_to_do; + const struct device *dev = &queue->info->netdev->dev;
BUG_ON(!netif_carrier_ok(queue->info->netdev));
do { prod = queue->tx.sring->rsp_prod; + if (RING_RESPONSE_PROD_OVERFLOW(&queue->tx, prod)) { + dev_alert(dev, "Illegal number of responses %u\n", + prod - queue->tx.rsp_cons); + goto err; + } rmb(); /* Ensure we see responses up to 'rp'. */
for (cons = queue->tx.rsp_cons; cons != prod; cons++) { @@ -379,14 +390,27 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) if (txrsp.status == XEN_NETIF_RSP_NULL) continue;
- id = txrsp.id; + id = txrsp.id; + if (id >= RING_SIZE(&queue->tx)) { + dev_alert(dev, + "Response has incorrect id (%u)\n", + id); + goto err; + } + if (queue->tx_link[id] != TX_PENDING) { + dev_alert(dev, + "Response for inactive request\n"); + goto err; + } + + queue->tx_link[id] = TX_LINK_NONE; skb = queue->tx_skbs[id]; queue->tx_skbs[id] = NULL; if (unlikely(gnttab_query_foreign_access( queue->grant_tx_ref[id]) != 0)) { - pr_alert("%s: warning -- grant still in use by backend domain\n", - __func__); - BUG(); + dev_alert(dev, + "Grant still in use by backend domain\n"); + goto err; } gnttab_end_foreign_access_ref( queue->grant_tx_ref[id], GNTMAP_readonly); @@ -404,6 +428,12 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) } while (more_to_do);
xennet_maybe_wake_tx(queue); + + return; + + err: + queue->info->broken = true; + dev_alert(dev, "Disabled for further use\n"); }
struct xennet_gnttab_make_txreq { @@ -447,6 +477,12 @@ static void xennet_tx_setup_grant(unsigned long gfn, unsigned int offset,
*tx = info->tx_local;
+ /* + * Put the request in the pending queue, it will be set to be pending + * when the producer index is about to be raised. + */ + add_id_to_list(&queue->tx_pend_queue, queue->tx_link, id); + info->tx = tx; info->size += info->tx_local.size; } @@ -540,6 +576,15 @@ static u16 xennet_select_queue(struct net_device *dev, struct sk_buff *skb, return queue_idx; }
+static void xennet_mark_tx_pending(struct netfront_queue *queue) +{ + unsigned int i; + + while ((i = get_id_from_list(&queue->tx_pend_queue, queue->tx_link)) != + TX_LINK_NONE) + queue->tx_link[i] = TX_PENDING; +} + #define MAX_XEN_SKB_FRAGS (65536 / XEN_PAGE_SIZE + 1)
static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev) @@ -563,6 +608,8 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev /* Drop the packet if no queues are set up */ if (num_queues < 1) goto drop; + if (unlikely(np->broken)) + goto drop; /* Determine which queue to transmit this SKB on */ queue_index = skb_get_queue_mapping(skb); queue = &np->queues[queue_index]; @@ -666,6 +713,8 @@ static netdev_tx_t xennet_start_xmit(struct sk_buff *skb, struct net_device *dev /* First request has the packet length. */ first_tx->size = skb->len;
+ xennet_mark_tx_pending(queue); + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&queue->tx, notify); if (notify) notify_remote_via_irq(queue->tx_irq); @@ -990,6 +1039,13 @@ static int xennet_poll(struct napi_struct *napi, int budget) skb_queue_head_init(&tmpq);
rp = queue->rx.sring->rsp_prod; + if (RING_RESPONSE_PROD_OVERFLOW(&queue->rx, rp)) { + dev_alert(&dev->dev, "Illegal number of responses %u\n", + rp - queue->rx.rsp_cons); + queue->info->broken = true; + spin_unlock(&queue->rx_lock); + return 0; + } rmb(); /* Ensure we see queued responses up to 'rp'. */
i = queue->rx.rsp_cons; @@ -1208,6 +1264,9 @@ static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) struct netfront_queue *queue = dev_id; unsigned long flags;
+ if (queue->info->broken) + return IRQ_HANDLED; + spin_lock_irqsave(&queue->tx_lock, flags); xennet_tx_buf_gc(queue); spin_unlock_irqrestore(&queue->tx_lock, flags); @@ -1220,6 +1279,9 @@ static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id) struct netfront_queue *queue = dev_id; struct net_device *dev = queue->info->netdev;
+ if (queue->info->broken) + return IRQ_HANDLED; + if (likely(netif_carrier_ok(dev) && RING_HAS_UNCONSUMED_RESPONSES(&queue->rx))) napi_schedule(&queue->napi); @@ -1241,6 +1303,10 @@ static void xennet_poll_controller(struct net_device *dev) struct netfront_info *info = netdev_priv(dev); unsigned int num_queues = dev->real_num_tx_queues; unsigned int i; + + if (info->broken) + return; + for (i = 0; i < num_queues; ++i) xennet_interrupt(0, &info->queues[i]); } @@ -1605,6 +1671,7 @@ static int xennet_init_queue(struct netfront_queue *queue)
/* Initialise tx_skb_freelist as a free chain containing every entry. */ queue->tx_skb_freelist = 0; + queue->tx_pend_queue = TX_LINK_NONE; for (i = 0; i < NET_TX_RING_SIZE; i++) { queue->tx_link[i] = i + 1; queue->grant_tx_ref[i] = GRANT_INVALID_REF; @@ -1815,6 +1882,9 @@ static int talk_to_netback(struct xenbus_device *dev, if (info->queues) xennet_destroy_queues(info);
+ /* For the case of a reconnect reset the "broken" indicator. */ + info->broken = false; + err = xennet_create_queues(info, &num_queues); if (err < 0) { xenbus_dev_fatal(dev, err, "creating queues");
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.222 commit 269d7124bcfad2558d2329d0fe603ca20b20d3f4 CVE: CVE-2021-28711
--------------------------------
commit 0fd08a34e8e3b67ec9bd8287ac0facf8374b844a upstream.
The Xen blkfront driver is still vulnerable for an attack via excessive number of events sent by the backend. Fix that by using lateeoi event channels.
This is part of XSA-391
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/block/xen-blkfront.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-)
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 90d6e13192bda..4e52917ebd311 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1563,9 +1563,12 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) unsigned long flags; struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)dev_id; struct blkfront_info *info = rinfo->dev_info; + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS;
- if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) + if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) { + xen_irq_lateeoi(irq, XEN_EOI_FLAG_SPURIOUS); return IRQ_HANDLED; + }
spin_lock_irqsave(&rinfo->ring_lock, flags); again: @@ -1581,6 +1584,8 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) unsigned long id; unsigned int op;
+ eoiflag = 0; + RING_COPY_RESPONSE(&rinfo->ring, i, &bret); id = bret.id;
@@ -1696,6 +1701,8 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
spin_unlock_irqrestore(&rinfo->ring_lock, flags);
+ xen_irq_lateeoi(irq, eoiflag); + return IRQ_HANDLED;
err: @@ -1703,6 +1710,8 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id)
spin_unlock_irqrestore(&rinfo->ring_lock, flags);
+ /* No EOI in order to avoid further interrupts. */ + pr_alert("%s disabled for further use\n", info->gd->disk_name); return IRQ_HANDLED; } @@ -1742,8 +1751,8 @@ static int setup_blkring(struct xenbus_device *dev, if (err) goto fail;
- err = bind_evtchn_to_irqhandler(rinfo->evtchn, blkif_interrupt, 0, - "blkif", rinfo); + err = bind_evtchn_to_irqhandler_lateeoi(rinfo->evtchn, blkif_interrupt, + 0, "blkif", rinfo); if (err <= 0) { xenbus_dev_fatal(dev, err, "bind_evtchn_to_irqhandler failed");
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.222 commit 3559ca594f15fcd23ed10c0056d40d71e5dab8e5 CVE: CVE-2021-28712
--------------------------------
commit b27d47950e481f292c0a5ad57357edb9d95d03ba upstream.
The Xen netfront driver is still vulnerable for an attack via excessive number of events sent by the backend. Fix that by using lateeoi event channels.
For being able to detect the case of no rx responses being added while the carrier is down a new lock is needed in order to update and test rsp_cons and the number of seen unconsumed responses atomically.
This is part of XSA-391
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netfront.c | 125 ++++++++++++++++++++++++++++--------- 1 file changed, 94 insertions(+), 31 deletions(-)
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c index e5bbd2d4252dc..da1b93f98cd1a 100644 --- a/drivers/net/xen-netfront.c +++ b/drivers/net/xen-netfront.c @@ -140,6 +140,9 @@ struct netfront_queue { struct sk_buff *rx_skbs[NET_RX_RING_SIZE]; grant_ref_t gref_rx_head; grant_ref_t grant_rx_ref[NET_RX_RING_SIZE]; + + unsigned int rx_rsp_unconsumed; + spinlock_t rx_cons_lock; };
struct netfront_info { @@ -364,12 +367,13 @@ static int xennet_open(struct net_device *dev) return 0; }
-static void xennet_tx_buf_gc(struct netfront_queue *queue) +static bool xennet_tx_buf_gc(struct netfront_queue *queue) { RING_IDX cons, prod; unsigned short id; struct sk_buff *skb; bool more_to_do; + bool work_done = false; const struct device *dev = &queue->info->netdev->dev;
BUG_ON(!netif_carrier_ok(queue->info->netdev)); @@ -386,6 +390,8 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue) for (cons = queue->tx.rsp_cons; cons != prod; cons++) { struct xen_netif_tx_response txrsp;
+ work_done = true; + RING_COPY_RESPONSE(&queue->tx, cons, &txrsp); if (txrsp.status == XEN_NETIF_RSP_NULL) continue; @@ -429,11 +435,13 @@ static void xennet_tx_buf_gc(struct netfront_queue *queue)
xennet_maybe_wake_tx(queue);
- return; + return work_done;
err: queue->info->broken = true; dev_alert(dev, "Disabled for further use\n"); + + return work_done; }
struct xennet_gnttab_make_txreq { @@ -754,6 +762,16 @@ static int xennet_close(struct net_device *dev) return 0; }
+static void xennet_set_rx_rsp_cons(struct netfront_queue *queue, RING_IDX val) +{ + unsigned long flags; + + spin_lock_irqsave(&queue->rx_cons_lock, flags); + queue->rx.rsp_cons = val; + queue->rx_rsp_unconsumed = RING_HAS_UNCONSUMED_RESPONSES(&queue->rx); + spin_unlock_irqrestore(&queue->rx_cons_lock, flags); +} + static void xennet_move_rx_slot(struct netfront_queue *queue, struct sk_buff *skb, grant_ref_t ref) { @@ -805,7 +823,7 @@ static int xennet_get_extras(struct netfront_queue *queue, xennet_move_rx_slot(queue, skb, ref); } while (extra.flags & XEN_NETIF_EXTRA_FLAG_MORE);
- queue->rx.rsp_cons = cons; + xennet_set_rx_rsp_cons(queue, cons); return err; }
@@ -885,7 +903,7 @@ static int xennet_get_responses(struct netfront_queue *queue, }
if (unlikely(err)) - queue->rx.rsp_cons = cons + slots; + xennet_set_rx_rsp_cons(queue, cons + slots);
return err; } @@ -939,7 +957,8 @@ static int xennet_fill_frags(struct netfront_queue *queue, __pskb_pull_tail(skb, pull_to - skb_headlen(skb)); } if (unlikely(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS)) { - queue->rx.rsp_cons = ++cons + skb_queue_len(list); + xennet_set_rx_rsp_cons(queue, + ++cons + skb_queue_len(list)); kfree_skb(nskb); return -ENOENT; } @@ -952,7 +971,7 @@ static int xennet_fill_frags(struct netfront_queue *queue, kfree_skb(nskb); }
- queue->rx.rsp_cons = cons; + xennet_set_rx_rsp_cons(queue, cons);
return 0; } @@ -1073,7 +1092,9 @@ static int xennet_poll(struct napi_struct *napi, int budget)
if (unlikely(xennet_set_skb_gso(skb, gso))) { __skb_queue_head(&tmpq, skb); - queue->rx.rsp_cons += skb_queue_len(&tmpq); + xennet_set_rx_rsp_cons(queue, + queue->rx.rsp_cons + + skb_queue_len(&tmpq)); goto err; } } @@ -1097,7 +1118,8 @@ static int xennet_poll(struct napi_struct *napi, int budget)
__skb_queue_tail(&rxq, skb);
- i = ++queue->rx.rsp_cons; + i = queue->rx.rsp_cons + 1; + xennet_set_rx_rsp_cons(queue, i); work_done++; }
@@ -1259,40 +1281,79 @@ static int xennet_set_features(struct net_device *dev, return 0; }
-static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) +static bool xennet_handle_tx(struct netfront_queue *queue, unsigned int *eoi) { - struct netfront_queue *queue = dev_id; unsigned long flags;
- if (queue->info->broken) - return IRQ_HANDLED; + if (unlikely(queue->info->broken)) + return false;
spin_lock_irqsave(&queue->tx_lock, flags); - xennet_tx_buf_gc(queue); + if (xennet_tx_buf_gc(queue)) + *eoi = 0; spin_unlock_irqrestore(&queue->tx_lock, flags);
+ return true; +} + +static irqreturn_t xennet_tx_interrupt(int irq, void *dev_id) +{ + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; + + if (likely(xennet_handle_tx(dev_id, &eoiflag))) + xen_irq_lateeoi(irq, eoiflag); + return IRQ_HANDLED; }
-static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id) +static bool xennet_handle_rx(struct netfront_queue *queue, unsigned int *eoi) { - struct netfront_queue *queue = dev_id; - struct net_device *dev = queue->info->netdev; + unsigned int work_queued; + unsigned long flags;
- if (queue->info->broken) - return IRQ_HANDLED; + if (unlikely(queue->info->broken)) + return false; + + spin_lock_irqsave(&queue->rx_cons_lock, flags); + work_queued = RING_HAS_UNCONSUMED_RESPONSES(&queue->rx); + if (work_queued > queue->rx_rsp_unconsumed) { + queue->rx_rsp_unconsumed = work_queued; + *eoi = 0; + } else if (unlikely(work_queued < queue->rx_rsp_unconsumed)) { + const struct device *dev = &queue->info->netdev->dev; + + spin_unlock_irqrestore(&queue->rx_cons_lock, flags); + dev_alert(dev, "RX producer index going backwards\n"); + dev_alert(dev, "Disabled for further use\n"); + queue->info->broken = true; + return false; + } + spin_unlock_irqrestore(&queue->rx_cons_lock, flags);
- if (likely(netif_carrier_ok(dev) && - RING_HAS_UNCONSUMED_RESPONSES(&queue->rx))) + if (likely(netif_carrier_ok(queue->info->netdev) && work_queued)) napi_schedule(&queue->napi);
+ return true; +} + +static irqreturn_t xennet_rx_interrupt(int irq, void *dev_id) +{ + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; + + if (likely(xennet_handle_rx(dev_id, &eoiflag))) + xen_irq_lateeoi(irq, eoiflag); + return IRQ_HANDLED; }
static irqreturn_t xennet_interrupt(int irq, void *dev_id) { - xennet_tx_interrupt(irq, dev_id); - xennet_rx_interrupt(irq, dev_id); + unsigned int eoiflag = XEN_EOI_FLAG_SPURIOUS; + + if (xennet_handle_tx(dev_id, &eoiflag) && + xennet_handle_rx(dev_id, &eoiflag)) + xen_irq_lateeoi(irq, eoiflag); + return IRQ_HANDLED; }
@@ -1519,9 +1580,10 @@ static int setup_netfront_single(struct netfront_queue *queue) if (err < 0) goto fail;
- err = bind_evtchn_to_irqhandler(queue->tx_evtchn, - xennet_interrupt, - 0, queue->info->netdev->name, queue); + err = bind_evtchn_to_irqhandler_lateeoi(queue->tx_evtchn, + xennet_interrupt, 0, + queue->info->netdev->name, + queue); if (err < 0) goto bind_fail; queue->rx_evtchn = queue->tx_evtchn; @@ -1549,18 +1611,18 @@ static int setup_netfront_split(struct netfront_queue *queue)
snprintf(queue->tx_irq_name, sizeof(queue->tx_irq_name), "%s-tx", queue->name); - err = bind_evtchn_to_irqhandler(queue->tx_evtchn, - xennet_tx_interrupt, - 0, queue->tx_irq_name, queue); + err = bind_evtchn_to_irqhandler_lateeoi(queue->tx_evtchn, + xennet_tx_interrupt, 0, + queue->tx_irq_name, queue); if (err < 0) goto bind_tx_fail; queue->tx_irq = err;
snprintf(queue->rx_irq_name, sizeof(queue->rx_irq_name), "%s-rx", queue->name); - err = bind_evtchn_to_irqhandler(queue->rx_evtchn, - xennet_rx_interrupt, - 0, queue->rx_irq_name, queue); + err = bind_evtchn_to_irqhandler_lateeoi(queue->rx_evtchn, + xennet_rx_interrupt, 0, + queue->rx_irq_name, queue); if (err < 0) goto bind_rx_fail; queue->rx_irq = err; @@ -1662,6 +1724,7 @@ static int xennet_init_queue(struct netfront_queue *queue)
spin_lock_init(&queue->tx_lock); spin_lock_init(&queue->rx_lock); + spin_lock_init(&queue->rx_cons_lock);
timer_setup(&queue->rx_refill_timer, rx_refill_timeout, 0);
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.222 commit 57e46acb3b48ea4e8efb1e1bea2e89e0c6cc43e2 CVE: CVE-2021-28713
--------------------------------
This is commit fe415186b43df0db1f17fa3a46275fd92107fe71 upstream.
The Xen console driver is still vulnerable for an attack via excessive number of events sent by the backend. Fix that by using a lateeoi event channel.
For the normal domU initial console this requires the introduction of bind_evtchn_to_irq_lateeoi() as there is no xenbus device available at the time the event channel is bound to the irq.
As the decision whether an interrupt was spurious or not requires to test for bytes having been read from the backend, move sending the event into the if statement, as sending an event without having found any bytes to be read is making no sense at all.
This is part of XSA-391
Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/tty/hvc/hvc_xen.c | 30 +++++++++++++++++++++++++++--- 1 file changed, 27 insertions(+), 3 deletions(-)
diff --git a/drivers/tty/hvc/hvc_xen.c b/drivers/tty/hvc/hvc_xen.c index dc43fa96c3de7..62df09f7cee65 100644 --- a/drivers/tty/hvc/hvc_xen.c +++ b/drivers/tty/hvc/hvc_xen.c @@ -37,6 +37,8 @@ struct xencons_info { struct xenbus_device *xbdev; struct xencons_interface *intf; unsigned int evtchn; + XENCONS_RING_IDX out_cons; + unsigned int out_cons_same; struct hvc_struct *hvc; int irq; int vtermno; @@ -131,6 +133,8 @@ static int domU_read_console(uint32_t vtermno, char *buf, int len) XENCONS_RING_IDX cons, prod; int recv = 0; struct xencons_info *xencons = vtermno_to_xencons(vtermno); + unsigned int eoiflag = 0; + if (xencons == NULL) return -EINVAL; intf = xencons->intf; @@ -146,7 +150,27 @@ static int domU_read_console(uint32_t vtermno, char *buf, int len) mb(); /* read ring before consuming */ intf->in_cons = cons;
- notify_daemon(xencons); + /* + * When to mark interrupt having been spurious: + * - there was no new data to be read, and + * - the backend did not consume some output bytes, and + * - the previous round with no read data didn't see consumed bytes + * (we might have a race with an interrupt being in flight while + * updating xencons->out_cons, so account for that by allowing one + * round without any visible reason) + */ + if (intf->out_cons != xencons->out_cons) { + xencons->out_cons = intf->out_cons; + xencons->out_cons_same = 0; + } + if (recv) { + notify_daemon(xencons); + } else if (xencons->out_cons_same++ > 1) { + eoiflag = XEN_EOI_FLAG_SPURIOUS; + } + + xen_irq_lateeoi(xencons->irq, eoiflag); + return recv; }
@@ -375,7 +399,7 @@ static int xencons_connect_backend(struct xenbus_device *dev, if (ret) return ret; info->evtchn = evtchn; - irq = bind_evtchn_to_irq(evtchn); + irq = bind_interdomain_evtchn_to_irq_lateeoi(dev->otherend_id, evtchn); if (irq < 0) return irq; info->irq = irq; @@ -539,7 +563,7 @@ static int __init xen_hvc_init(void) return r;
info = vtermno_to_xencons(HVC_COOKIE); - info->irq = bind_evtchn_to_irq(info->evtchn); + info->irq = bind_evtchn_to_irq_lateeoi(info->evtchn); } if (info->irq < 0) info->irq = 0; /* NO_IRQ */
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.222 commit 1de7644eac41981817fb66b74e0f82ca4477dc9d CVE: CVE-2021-28714
--------------------------------
commit 6032046ec4b70176d247a71836186d47b25d1684 upstream.
Commit 1d5d48523900a4b ("xen-netback: require fewer guest Rx slots when not using GSO") introduced a security problem in netback, as an interface would only be regarded to be stalled if no slot is available in the rx queue ring page. In case the SKB at the head of the queued requests will need more than one rx slot and only one slot is free the stall detection logic will never trigger, as the test for that is only looking for at least one slot to be free.
Fix that by testing for the needed number of slots instead of only one slot being available.
In order to not have to take the rx queue lock that often, store the number of needed slots in the queue data. As all SKB dequeue operations happen in the rx queue kernel thread this is safe, as long as the number of needed slots is accessed via READ/WRITE_ONCE() only and updates are always done with the rx queue lock held.
Add a small helper for obtaining the number of free slots.
This is part of XSA-392
Fixes: 1d5d48523900a4b ("xen-netback: require fewer guest Rx slots when not using GSO") Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netback/common.h | 1 + drivers/net/xen-netback/rx.c | 65 ++++++++++++++++++++------------ 2 files changed, 42 insertions(+), 24 deletions(-)
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h index 86d23d0f563c4..751254dcee3b5 100644 --- a/drivers/net/xen-netback/common.h +++ b/drivers/net/xen-netback/common.h @@ -203,6 +203,7 @@ struct xenvif_queue { /* Per-queue data for xenvif */ unsigned int rx_queue_max; unsigned int rx_queue_len; unsigned long last_rx_time; + unsigned int rx_slots_needed; bool stalled;
struct xenvif_copy_state rx_copy; diff --git a/drivers/net/xen-netback/rx.c b/drivers/net/xen-netback/rx.c index 48e2006f96ce6..04353df147266 100644 --- a/drivers/net/xen-netback/rx.c +++ b/drivers/net/xen-netback/rx.c @@ -33,28 +33,36 @@ #include <xen/xen.h> #include <xen/events.h>
-static bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue) +/* + * Update the needed ring page slots for the first SKB queued. + * Note that any call sequence outside the RX thread calling this function + * needs to wake up the RX thread via a call of xenvif_kick_thread() + * afterwards in order to avoid a race with putting the thread to sleep. + */ +static void xenvif_update_needed_slots(struct xenvif_queue *queue, + const struct sk_buff *skb) { - RING_IDX prod, cons; - struct sk_buff *skb; - int needed; - unsigned long flags; - - spin_lock_irqsave(&queue->rx_queue.lock, flags); + unsigned int needed = 0;
- skb = skb_peek(&queue->rx_queue); - if (!skb) { - spin_unlock_irqrestore(&queue->rx_queue.lock, flags); - return false; + if (skb) { + needed = DIV_ROUND_UP(skb->len, XEN_PAGE_SIZE); + if (skb_is_gso(skb)) + needed++; + if (skb->sw_hash) + needed++; }
- needed = DIV_ROUND_UP(skb->len, XEN_PAGE_SIZE); - if (skb_is_gso(skb)) - needed++; - if (skb->sw_hash) - needed++; + WRITE_ONCE(queue->rx_slots_needed, needed); +}
- spin_unlock_irqrestore(&queue->rx_queue.lock, flags); +static bool xenvif_rx_ring_slots_available(struct xenvif_queue *queue) +{ + RING_IDX prod, cons; + unsigned int needed; + + needed = READ_ONCE(queue->rx_slots_needed); + if (!needed) + return false;
do { prod = queue->rx.sring->req_prod; @@ -80,6 +88,9 @@ void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
spin_lock_irqsave(&queue->rx_queue.lock, flags);
+ if (skb_queue_empty(&queue->rx_queue)) + xenvif_update_needed_slots(queue, skb); + __skb_queue_tail(&queue->rx_queue, skb);
queue->rx_queue_len += skb->len; @@ -100,6 +111,8 @@ static struct sk_buff *xenvif_rx_dequeue(struct xenvif_queue *queue)
skb = __skb_dequeue(&queue->rx_queue); if (skb) { + xenvif_update_needed_slots(queue, skb_peek(&queue->rx_queue)); + queue->rx_queue_len -= skb->len; if (queue->rx_queue_len < queue->rx_queue_max) { struct netdev_queue *txq; @@ -474,27 +487,31 @@ void xenvif_rx_action(struct xenvif_queue *queue) xenvif_rx_copy_flush(queue); }
-static bool xenvif_rx_queue_stalled(struct xenvif_queue *queue) +static RING_IDX xenvif_rx_queue_slots(const struct xenvif_queue *queue) { RING_IDX prod, cons;
prod = queue->rx.sring->req_prod; cons = queue->rx.req_cons;
+ return prod - cons; +} + +static bool xenvif_rx_queue_stalled(const struct xenvif_queue *queue) +{ + unsigned int needed = READ_ONCE(queue->rx_slots_needed); + return !queue->stalled && - prod - cons < 1 && + xenvif_rx_queue_slots(queue) < needed && time_after(jiffies, queue->last_rx_time + queue->vif->stall_timeout); }
static bool xenvif_rx_queue_ready(struct xenvif_queue *queue) { - RING_IDX prod, cons; - - prod = queue->rx.sring->req_prod; - cons = queue->rx.req_cons; + unsigned int needed = READ_ONCE(queue->rx_slots_needed);
- return queue->stalled && prod - cons >= 1; + return queue->stalled && xenvif_rx_queue_slots(queue) >= needed; }
bool xenvif_have_rx_work(struct xenvif_queue *queue, bool test_kthread)
From: Juergen Gross jgross@suse.com
stable inclusion from linux-4.19.222 commit c9f17e92917fd5786be872626a3928979ecc4c39 CVE: CVE-2021-28715
--------------------------------
commit be81992f9086b230623ae3ebbc85ecee4d00a3d3 upstream.
In case a guest isn't consuming incoming network traffic as fast as it is coming in, xen-netback is buffering network packages in unlimited numbers today. This can result in host OOM situations.
Commit f48da8b14d04ca8 ("xen-netback: fix unlimited guest Rx internal queue and carrier flapping") meant to introduce a mechanism to limit the amount of buffered data by stopping the Tx queue when reaching the data limit, but this doesn't work for cases like UDP.
When hitting the limit don't queue further SKBs, but drop them instead. In order to be able to tell Rx packages have been dropped increment the rx_dropped statistics counter in this case.
It should be noted that the old solution to continue queueing SKBs had the additional problem of an overflow of the 32-bit rx_queue_len value would result in intermittent Tx queue enabling.
This is part of XSA-392
Fixes: f48da8b14d04ca8 ("xen-netback: fix unlimited guest Rx internal queue and carrier flapping") Signed-off-by: Juergen Gross jgross@suse.com Reviewed-by: Jan Beulich jbeulich@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org Signed-off-by: Yang Yingliang yangyingliang@huawei.com Reviewed-by: Xiu Jianfeng xiujianfeng@huawei.com Signed-off-by: Yang Yingliang yangyingliang@huawei.com --- drivers/net/xen-netback/rx.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/drivers/net/xen-netback/rx.c b/drivers/net/xen-netback/rx.c index 04353df147266..7f68067c01745 100644 --- a/drivers/net/xen-netback/rx.c +++ b/drivers/net/xen-netback/rx.c @@ -88,16 +88,19 @@ void xenvif_rx_queue_tail(struct xenvif_queue *queue, struct sk_buff *skb)
spin_lock_irqsave(&queue->rx_queue.lock, flags);
- if (skb_queue_empty(&queue->rx_queue)) - xenvif_update_needed_slots(queue, skb); - - __skb_queue_tail(&queue->rx_queue, skb); - - queue->rx_queue_len += skb->len; - if (queue->rx_queue_len > queue->rx_queue_max) { + if (queue->rx_queue_len >= queue->rx_queue_max) { struct net_device *dev = queue->vif->dev;
netif_tx_stop_queue(netdev_get_tx_queue(dev, queue->id)); + kfree_skb(skb); + queue->vif->dev->stats.rx_dropped++; + } else { + if (skb_queue_empty(&queue->rx_queue)) + xenvif_update_needed_slots(queue, skb); + + __skb_queue_tail(&queue->rx_queue, skb); + + queue->rx_queue_len += skb->len; }
spin_unlock_irqrestore(&queue->rx_queue.lock, flags); @@ -147,6 +150,7 @@ static void xenvif_rx_queue_drop_expired(struct xenvif_queue *queue) break; xenvif_rx_dequeue(queue); kfree_skb(skb); + queue->vif->dev->stats.rx_dropped++; } }
-
Yang Yingliang