From 0be4c96e7c57a5359e35eba32fd3559824f8c751 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 10 Feb 2017 10:31:49 -0800 Subject: [PATCH 01/58] net_sched: fix error recovery at qdisc creation commit 87b60cfacf9f17cf71933c6e33b66e68160af71d upstream. Dmitry reported uses after free in qdisc code [1] The problem here is that ops->init() can return an error. qdisc_create_dflt() then call ops->destroy(), while qdisc_create() does _not_ call it. Four qdisc chose to call their own ops->destroy(), assuming their caller would not. This patch makes sure qdisc_create() calls ops->destroy() and fixes the four qdisc to avoid double free. [1] BUG: KASAN: use-after-free in mq_destroy+0x242/0x290 net/sched/sch_mq.c:33 at addr ffff8801d415d440 Read of size 8 by task syz-executor2/5030 CPU: 0 PID: 5030 Comm: syz-executor2 Not tainted 4.3.5-smp-DEV #119 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 0000000000000046 ffff8801b435b870 ffffffff81bbbed4 ffff8801db000400 ffff8801d415d440 ffff8801d415dc40 ffff8801c4988510 ffff8801b435b898 ffffffff816682b1 ffff8801b435b928 ffff8801d415d440 ffff8801c49880c0 Call Trace: [] __dump_stack lib/dump_stack.c:15 [inline] [] dump_stack+0x6c/0x98 lib/dump_stack.c:51 [] kasan_object_err+0x21/0x70 mm/kasan/report.c:158 [] print_address_description mm/kasan/report.c:196 [inline] [] kasan_report_error+0x1b4/0x4b0 mm/kasan/report.c:285 [] kasan_report mm/kasan/report.c:305 [inline] [] __asan_report_load8_noabort+0x43/0x50 mm/kasan/report.c:326 [] mq_destroy+0x242/0x290 net/sched/sch_mq.c:33 [] qdisc_destroy+0x12d/0x290 net/sched/sch_generic.c:953 [] qdisc_create_dflt+0xf0/0x120 net/sched/sch_generic.c:848 [] attach_default_qdiscs net/sched/sch_generic.c:1029 [inline] [] dev_activate+0x6ad/0x880 net/sched/sch_generic.c:1064 [] __dev_open+0x221/0x320 net/core/dev.c:1403 [] __dev_change_flags+0x15e/0x3e0 net/core/dev.c:6858 [] dev_change_flags+0x8e/0x140 net/core/dev.c:6926 [] dev_ifsioc+0x446/0x890 net/core/dev_ioctl.c:260 [] dev_ioctl+0x1ba/0xb80 net/core/dev_ioctl.c:546 [] sock_do_ioctl+0x99/0xb0 net/socket.c:879 [] sock_ioctl+0x2a0/0x390 net/socket.c:958 [] vfs_ioctl fs/ioctl.c:44 [inline] [] do_vfs_ioctl+0x8a8/0xe50 fs/ioctl.c:611 [] SYSC_ioctl fs/ioctl.c:626 [inline] [] SyS_ioctl+0x94/0xc0 fs/ioctl.c:617 [] entry_SYSCALL_64_fastpath+0x12/0x17 Signed-off-by: Eric Dumazet Reported-by: Dmitry Vyukov Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/sched/sch_api.c | 2 ++ net/sched/sch_hhf.c | 8 ++++++-- net/sched/sch_mq.c | 10 +++------- net/sched/sch_mqprio.c | 19 ++++++------------- net/sched/sch_sfq.c | 3 ++- 5 files changed, 19 insertions(+), 23 deletions(-) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 95b560f0b253..35ed0d8c53b0 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1004,6 +1004,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, return sch; } + /* ops->init() failed, we call ->destroy() like qdisc_create_dflt() */ + ops->destroy(sch); err_out3: dev_put(dev); kfree((char *) sch - sch->padded); diff --git a/net/sched/sch_hhf.c b/net/sched/sch_hhf.c index 13d6f83ec491..45d4b2f22f62 100644 --- a/net/sched/sch_hhf.c +++ b/net/sched/sch_hhf.c @@ -636,7 +636,9 @@ static int hhf_init(struct Qdisc *sch, struct nlattr *opt) q->hhf_arrays[i] = hhf_zalloc(HHF_ARRAYS_LEN * sizeof(u32)); if (!q->hhf_arrays[i]) { - hhf_destroy(sch); + /* Note: hhf_destroy() will be called + * by our caller. + */ return -ENOMEM; } } @@ -647,7 +649,9 @@ static int hhf_init(struct Qdisc *sch, struct nlattr *opt) q->hhf_valid_bits[i] = hhf_zalloc(HHF_ARRAYS_LEN / BITS_PER_BYTE); if (!q->hhf_valid_bits[i]) { - hhf_destroy(sch); + /* Note: hhf_destroy() will be called + * by our caller. + */ return -ENOMEM; } } diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c index 3e82f047caaf..d9c84328e7eb 100644 --- a/net/sched/sch_mq.c +++ b/net/sched/sch_mq.c @@ -52,7 +52,7 @@ static int mq_init(struct Qdisc *sch, struct nlattr *opt) /* pre-allocate qdiscs, attachment can't fail */ priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]), GFP_KERNEL); - if (priv->qdiscs == NULL) + if (!priv->qdiscs) return -ENOMEM; for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { @@ -60,18 +60,14 @@ static int mq_init(struct Qdisc *sch, struct nlattr *opt) qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops, TC_H_MAKE(TC_H_MAJ(sch->handle), TC_H_MIN(ntx + 1))); - if (qdisc == NULL) - goto err; + if (!qdisc) + return -ENOMEM; priv->qdiscs[ntx] = qdisc; qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; } sch->flags |= TCQ_F_MQROOT; return 0; - -err: - mq_destroy(sch); - return -ENOMEM; } static void mq_attach(struct Qdisc *sch) diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c index ad70ecf57ce7..66bccc5ff4ea 100644 --- a/net/sched/sch_mqprio.c +++ b/net/sched/sch_mqprio.c @@ -117,20 +117,17 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) /* pre-allocate qdisc, attachment can't fail */ priv->qdiscs = kcalloc(dev->num_tx_queues, sizeof(priv->qdiscs[0]), GFP_KERNEL); - if (priv->qdiscs == NULL) { - err = -ENOMEM; - goto err; - } + if (!priv->qdiscs) + return -ENOMEM; for (i = 0; i < dev->num_tx_queues; i++) { dev_queue = netdev_get_tx_queue(dev, i); qdisc = qdisc_create_dflt(dev_queue, default_qdisc_ops, TC_H_MAKE(TC_H_MAJ(sch->handle), TC_H_MIN(i + 1))); - if (qdisc == NULL) { - err = -ENOMEM; - goto err; - } + if (!qdisc) + return -ENOMEM; + priv->qdiscs[i] = qdisc; qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; } @@ -143,7 +140,7 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) priv->hw_owned = 1; err = dev->netdev_ops->ndo_setup_tc(dev, qopt->num_tc); if (err) - goto err; + return err; } else { netdev_set_num_tc(dev, qopt->num_tc); for (i = 0; i < qopt->num_tc; i++) @@ -157,10 +154,6 @@ static int mqprio_init(struct Qdisc *sch, struct nlattr *opt) sch->flags |= TCQ_F_MQROOT; return 0; - -err: - mqprio_destroy(sch); - return err; } static void mqprio_attach(struct Qdisc *sch) diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c index 498f0a2cb47f..4431e2833e45 100644 --- a/net/sched/sch_sfq.c +++ b/net/sched/sch_sfq.c @@ -742,9 +742,10 @@ static int sfq_init(struct Qdisc *sch, struct nlattr *opt) q->ht = sfq_alloc(sizeof(q->ht[0]) * q->divisor); q->slots = sfq_alloc(sizeof(q->slots[0]) * q->maxflows); if (!q->ht || !q->slots) { - sfq_destroy(sch); + /* Note: sfq_destroy() will be called by our caller */ return -ENOMEM; } + for (i = 0; i < q->divisor; i++) q->ht[i] = SFQ_EMPTY_SLOT; From c485792ed6491865037abc1e18bbd66240112c93 Mon Sep 17 00:00:00 2001 From: Gao Feng Date: Wed, 28 Jun 2017 12:53:54 +0800 Subject: [PATCH 02/58] net: sched: Fix one possible panic when no destroy callback commit c1a4872ebfb83b1af7144f7b29ac8c4b344a12a8 upstream. When qdisc fail to init, qdisc_create would invoke the destroy callback to cleanup. But there is no check if the callback exists really. So it would cause the panic if there is no real destroy callback like the qdisc codel, fq, and so on. Take codel as an example following: When a malicious user constructs one invalid netlink msg, it would cause codel_init->codel_change->nla_parse_nested failed. Then kernel would invoke the destroy callback directly but qdisc codel doesn't define one. It causes one panic as a result. Now add one the check for destroy to avoid the possible panic. Fixes: 87b60cfacf9f ("net_sched: fix error recovery at qdisc creation") Signed-off-by: Gao Feng Acked-by: Eric Dumazet Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/sched/sch_api.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 35ed0d8c53b0..6d340cd6e2a7 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -1005,7 +1005,8 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, return sch; } /* ops->init() failed, we call ->destroy() like qdisc_create_dflt() */ - ops->destroy(sch); + if (ops->destroy) + ops->destroy(sch); err_out3: dev_put(dev); kfree((char *) sch - sch->padded); From f71e5140242ae9268cb44ccaa4118e8cda69910c Mon Sep 17 00:00:00 2001 From: Zach Brown Date: Tue, 20 Jun 2017 12:48:11 -0500 Subject: [PATCH 03/58] net/phy: micrel: configure intterupts after autoneg workaround commit b866203d872d5deeafcecd25ea429d6748b5bd56 upstream. The commit ("net/phy: micrel: Add workaround for bad autoneg") fixes an autoneg failure case by resetting the hardware. This turns off intterupts. Things will work themselves out if the phy polls, as it will figure out it's state during a poll. However if the phy uses only intterupts, the phy will stall, since interrupts are off. This patch fixes the issue by calling config_intr after resetting the phy. Fixes: d2fd719bcb0e ("net/phy: micrel: Add workaround for bad autoneg ") Signed-off-by: Zach Brown Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/phy/micrel.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index e13ad6cdcc22..c8b85f1069ff 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -539,6 +539,8 @@ static int ksz9031_read_status(struct phy_device *phydev) if ((regval & 0xFF) == 0xFF) { phy_init_hw(phydev); phydev->link = 0; + if (phydev->drv->config_intr && phy_interrupt_is_valid(phydev)) + phydev->drv->config_intr(phydev); } return 0; From 6a87cca39f29aa9a3f2ea03ea4ed909656ce3980 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 21 Jun 2017 14:34:58 -0700 Subject: [PATCH 04/58] ipv6: avoid unregistering inet6_dev for loopback commit 60abc0be96e00ca71bac083215ac91ad2e575096 upstream. The per netns loopback_dev->ip6_ptr is unregistered and set to NULL when its mtu is set to smaller than IPV6_MIN_MTU, this leads to that we could set rt->rt6i_idev NULL after a rt6_uncached_list_flush_dev() and then crash after another call. In this case we should just bring its inet6_dev down, rather than unregistering it, at least prior to commit 176c39af29bc ("netns: fix addrconf_ifdown kernel panic") we always override the case for loopback. Thanks a lot to Andrey for finding a reliable reproducer. Fixes: 176c39af29bc ("netns: fix addrconf_ifdown kernel panic") Reported-by: Andrey Konovalov Cc: Andrey Konovalov Cc: Daniel Lezcano Cc: David Ahern Signed-off-by: Cong Wang Acked-by: David Ahern Tested-by: Andrey Konovalov Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv6/addrconf.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 03dadbf6cc5e..56627f0f4088 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3143,6 +3143,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct inet6_dev *idev = __in6_dev_get(dev); + struct net *net = dev_net(dev); int run_pending = 0; int err; @@ -3158,7 +3159,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, case NETDEV_CHANGEMTU: /* if MTU under IPV6_MIN_MTU stop IPv6 on this interface. */ if (dev->mtu < IPV6_MIN_MTU) { - addrconf_ifdown(dev, 1); + addrconf_ifdown(dev, dev != net->loopback_dev); break; } @@ -3271,7 +3272,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event, * IPV6_MIN_MTU stop IPv6 on this interface. */ if (dev->mtu < IPV6_MIN_MTU) - addrconf_ifdown(dev, 1); + addrconf_ifdown(dev, dev != net->loopback_dev); } break; From ccff2f4afbc183a163b2feb83132ead13d9283ed Mon Sep 17 00:00:00 2001 From: Richard Cochran Date: Fri, 23 Jun 2017 17:51:31 +0200 Subject: [PATCH 05/58] net: dp83640: Avoid NULL pointer dereference. commit db9d8b29d19d2801793e4419f4c6272bf8951c62 upstream. The function, skb_complete_tx_timestamp(), used to allow passing in a NULL pointer for the time stamps, but that was changed in commit 62bccb8cdb69051b95a55ab0c489e3cab261c8ef ("net-timestamp: Make the clone operation stand-alone from phy timestamping"), and the existing call sites, all of which are in the dp83640 driver, were fixed up. Even though the kernel-doc was subsequently updated in commit 7a76a021cd5a292be875fbc616daf03eab1e6996 ("net-timestamp: Update skb_complete_tx_timestamp comment"), still a bug fix from Manfred Rudigier came into the driver using the old semantics. Probably Manfred derived that patch from an older kernel version. This fix should be applied to the stable trees as well. Fixes: 81e8f2e930fe ("net: dp83640: Fix tx timestamp overflow handling.") Signed-off-by: Richard Cochran Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/phy/dp83640.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c index 84b9cca152eb..e83acc608678 100644 --- a/drivers/net/phy/dp83640.c +++ b/drivers/net/phy/dp83640.c @@ -907,7 +907,7 @@ static void decode_txts(struct dp83640_private *dp83640, if (overflow) { pr_debug("tx timestamp queue overflow, count %d\n", overflow); while (skb) { - skb_complete_tx_timestamp(skb, NULL); + kfree_skb(skb); skb = skb_dequeue(&dp83640->tx_queue); } return; From 32a44f1ba7edf3d58afb1fdce852f800f476b622 Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Sat, 24 Jun 2017 23:50:30 -0700 Subject: [PATCH 06/58] tcp: reset sk_rx_dst in tcp_disconnect() commit d747a7a51b00984127a88113cdbbc26f91e9d815 upstream. We have to reset the sk->sk_rx_dst when we disconnect a TCP connection, because otherwise when we re-connect it this dst reference is simply overridden in tcp_finish_connect(). This fixes a dst leak which leads to a loopback dev refcnt leak. It is a long-standing bug, Kevin reported a very similar (if not same) bug before. Thanks to Andrei for providing such a reliable reproducer which greatly narrows down the problem. Fixes: 41063e9dd119 ("ipv4: Early TCP socket demux.") Reported-by: Andrei Vagin Reported-by: Kevin Xu Signed-off-by: Cong Wang Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv4/tcp.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 4bd8678329d6..0870a86e9d96 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2263,6 +2263,8 @@ int tcp_disconnect(struct sock *sk, int flags) tcp_init_send_head(sk); memset(&tp->rx_opt, 0, sizeof(tp->rx_opt)); __sk_dst_reset(sk); + dst_release(sk->sk_rx_dst); + sk->sk_rx_dst = NULL; tcp_saved_syn_free(tp); WARN_ON(inet->inet_num && !icsk->icsk_bind_hash); From d598f7ff24db728c7417fb02871aa6578d97e224 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 27 Jun 2017 07:02:20 -0700 Subject: [PATCH 07/58] net: prevent sign extension in dev_get_stats() commit 6f64ec74515925cced6df4571638b5a099a49aae upstream. Similar to the fix provided by Dominik Heidler in commit 9b3dc0a17d73 ("l2tp: cast l2tp traffic counter to unsigned") we need to take care of 32bit kernels in dev_get_stats(). When using atomic_long_read(), we add a 'long' to u64 and might misinterpret high order bit, unless we cast to unsigned. Fixes: caf586e5f23ce ("net: add a core netdev->rx_dropped counter") Fixes: 015f0688f57ca ("net: net: add a core netdev->tx_dropped counter") Fixes: 6e7333d315a76 ("net: add rx_nohandler stat counter") Signed-off-by: Eric Dumazet Cc: Jarod Wilson Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/core/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 524d8b28e690..c8f0385ea492 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -7052,8 +7052,8 @@ struct rtnl_link_stats64 *dev_get_stats(struct net_device *dev, } else { netdev_stats_to_stats64(storage, &dev->stats); } - storage->rx_dropped += atomic_long_read(&dev->rx_dropped); - storage->tx_dropped += atomic_long_read(&dev->tx_dropped); + storage->rx_dropped += (unsigned long)atomic_long_read(&dev->rx_dropped); + storage->tx_dropped += (unsigned long)atomic_long_read(&dev->tx_dropped); return storage; } EXPORT_SYMBOL(dev_get_stats); From 1a4f13e0a99a85c455ff2f6dc117f6f049c039fa Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 29 Jun 2017 03:04:59 +0200 Subject: [PATCH 08/58] bpf: prevent leaking pointer via xadd on unpriviledged commit 6bdf6abc56b53103324dfd270a86580306e1a232 upstream. Leaking kernel addresses on unpriviledged is generally disallowed, for example, verifier rejects the following: 0: (b7) r0 = 0 1: (18) r2 = 0xffff897e82304400 3: (7b) *(u64 *)(r1 +48) = r2 R2 leaks addr into ctx Doing pointer arithmetic on them is also forbidden, so that they don't turn into unknown value and then get leaked out. However, there's xadd as a special case, where we don't check the src reg for being a pointer register, e.g. the following will pass: 0: (b7) r0 = 0 1: (7b) *(u64 *)(r1 +48) = r0 2: (18) r2 = 0xffff897e82304400 ; map 4: (db) lock *(u64 *)(r1 +48) += r2 5: (95) exit We could store the pointer into skb->cb, loose the type context, and then read it out from there again to leak it eventually out of a map value. Or more easily in a different variant, too: 0: (bf) r6 = r1 1: (7a) *(u64 *)(r10 -8) = 0 2: (bf) r2 = r10 3: (07) r2 += -8 4: (18) r1 = 0x0 6: (85) call bpf_map_lookup_elem#1 7: (15) if r0 == 0x0 goto pc+3 R0=map_value(ks=8,vs=8,id=0),min_value=0,max_value=0 R6=ctx R10=fp 8: (b7) r3 = 0 9: (7b) *(u64 *)(r0 +0) = r3 10: (db) lock *(u64 *)(r0 +0) += r6 11: (b7) r0 = 0 12: (95) exit from 7 to 11: R0=inv,min_value=0,max_value=0 R6=ctx R10=fp 11: (b7) r0 = 0 12: (95) exit Prevent this by checking xadd src reg for pointer types. Also add a couple of test cases related to this. Fixes: 1be7f75d1668 ("bpf: enable non-root eBPF programs") Fixes: 17a5267067f3 ("bpf: verifier (add verifier core)") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Acked-by: Martin KaFai Lau Acked-by: Edward Cree Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- kernel/bpf/verifier.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 2cbfba78d3db..863e24f1e62e 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -754,6 +754,11 @@ static int check_xadd(struct verifier_env *env, struct bpf_insn *insn) if (err) return err; + if (is_pointer_value(env, insn->src_reg)) { + verbose("R%d leaks addr into mem\n", insn->src_reg); + return -EACCES; + } + /* check whether atomic_add can read the memory */ err = check_mem_access(env, insn->dst_reg, insn->off, BPF_SIZE(insn->code), BPF_READ, -1); From 38ae32c9f13d3f0c642d373f35489752aa60d466 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Michal=20Kube=C4=8Dek?= Date: Thu, 29 Jun 2017 11:13:36 +0200 Subject: [PATCH 09/58] net: handle NAPI_GRO_FREE_STOLEN_HEAD case also in napi_frags_finish() commit e44699d2c28067f69698ccb68dd3ddeacfebc434 upstream. Recently I started seeing warnings about pages with refcount -1. The problem was traced to packets being reused after their head was merged into a GRO packet by skb_gro_receive(). While bisecting the issue pointed to commit c21b48cc1bbf ("net: adjust skb->truesize in ___pskb_trim()") and I have never seen it on a kernel with it reverted, I believe the real problem appeared earlier when the option to merge head frag in GRO was implemented. Handling NAPI_GRO_FREE_STOLEN_HEAD state was only added to GRO_MERGED_FREE branch of napi_skb_finish() so that if the driver uses napi_gro_frags() and head is merged (which in my case happens after the skb_condense() call added by the commit mentioned above), the skb is reused including the head that has been merged. As a result, we release the page reference twice and eventually end up with negative page refcount. To fix the problem, handle NAPI_GRO_FREE_STOLEN_HEAD in napi_frags_finish() the same way it's done in napi_skb_finish(). Fixes: d7e8883cfcf4 ("net: make GRO aware of skb->head_frag") Signed-off-by: Michal Kubecek Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/core/dev.c | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index c8f0385ea492..dc5d3d546150 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4375,6 +4375,12 @@ struct packet_offload *gro_find_complete_by_type(__be16 type) } EXPORT_SYMBOL(gro_find_complete_by_type); +static void napi_skb_free_stolen_head(struct sk_buff *skb) +{ + skb_dst_drop(skb); + kmem_cache_free(skbuff_head_cache, skb); +} + static gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb) { switch (ret) { @@ -4388,12 +4394,10 @@ static gro_result_t napi_skb_finish(gro_result_t ret, struct sk_buff *skb) break; case GRO_MERGED_FREE: - if (NAPI_GRO_CB(skb)->free == NAPI_GRO_FREE_STOLEN_HEAD) { - skb_dst_drop(skb); - kmem_cache_free(skbuff_head_cache, skb); - } else { + if (NAPI_GRO_CB(skb)->free == NAPI_GRO_FREE_STOLEN_HEAD) + napi_skb_free_stolen_head(skb); + else __kfree_skb(skb); - } break; case GRO_HELD: @@ -4459,10 +4463,16 @@ static gro_result_t napi_frags_finish(struct napi_struct *napi, break; case GRO_DROP: - case GRO_MERGED_FREE: napi_reuse_skb(napi, skb); break; + case GRO_MERGED_FREE: + if (NAPI_GRO_CB(skb)->free == NAPI_GRO_FREE_STOLEN_HEAD) + napi_skb_free_stolen_head(skb); + else + napi_reuse_skb(napi, skb); + break; + case GRO_MERGED: break; } From 0c32b01ee49d4784ac85fb4a4ee903ac86e7d183 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Thu, 29 Jun 2017 16:56:54 +0200 Subject: [PATCH 10/58] ipv6: dad: don't remove dynamic addresses if link is down commit ec8add2a4c9df723c94a863b8fcd6d93c472deed upstream. Currently, when the link for $DEV is down, this command succeeds but the address is removed immediately by DAD (1): ip addr add 1111::12/64 dev $DEV valid_lft 3600 preferred_lft 1800 In the same situation, this will succeed and not remove the address (2): ip addr add 1111::12/64 dev $DEV ip addr change 1111::12/64 dev $DEV valid_lft 3600 preferred_lft 1800 The comment in addrconf_dad_begin() when !IF_READY makes it look like this is the intended behavior, but doesn't explain why: * If the device is not ready: * - keep it tentative if it is a permanent address. * - otherwise, kill it. We clearly cannot prevent userspace from doing (2), but we can make (1) work consistently with (2). addrconf_dad_stop() is only called in two cases: if DAD failed, or to skip DAD when the link is down. In that second case, the fix is to avoid deleting the address, like we already do for permanent addresses. Fixes: 3c21edbd1137 ("[IPV6]: Defer IPv6 device initialization until the link becomes ready.") Signed-off-by: Sabrina Dubroca Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/ipv6/addrconf.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 56627f0f4088..735b22b1b4ea 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1772,17 +1772,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, const struct in6_addr *add static void addrconf_dad_stop(struct inet6_ifaddr *ifp, int dad_failed) { - if (ifp->flags&IFA_F_PERMANENT) { - spin_lock_bh(&ifp->lock); - addrconf_del_dad_work(ifp); - ifp->flags |= IFA_F_TENTATIVE; - if (dad_failed) - ifp->flags |= IFA_F_DADFAILED; - spin_unlock_bh(&ifp->lock); - if (dad_failed) - ipv6_ifa_notify(0, ifp); - in6_ifa_put(ifp); - } else if (ifp->flags&IFA_F_TEMPORARY) { + if (ifp->flags&IFA_F_TEMPORARY) { struct inet6_ifaddr *ifpub; spin_lock_bh(&ifp->lock); ifpub = ifp->ifpub; @@ -1795,6 +1785,16 @@ static void addrconf_dad_stop(struct inet6_ifaddr *ifp, int dad_failed) spin_unlock_bh(&ifp->lock); } ipv6_del_addr(ifp); + } else if (ifp->flags&IFA_F_PERMANENT || !dad_failed) { + spin_lock_bh(&ifp->lock); + addrconf_del_dad_work(ifp); + ifp->flags |= IFA_F_TENTATIVE; + if (dad_failed) + ifp->flags |= IFA_F_DADFAILED; + spin_unlock_bh(&ifp->lock); + if (dad_failed) + ipv6_ifa_notify(0, ifp); + in6_ifa_put(ifp); } else { ipv6_del_addr(ifp); } From eb7bef1db6f31cdba61a0d8794839fab57da5ed8 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 5 Jul 2017 14:41:46 -0600 Subject: [PATCH 11/58] net: ipv6: Compare lwstate in detecting duplicate nexthops MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit f06b7549b79e29a672336d4e134524373fb7a232 upstream. Lennert reported a failure to add different mpls encaps in a multipath route: $ ip -6 route add 1234::/16 \ nexthop encap mpls 10 via fe80::1 dev ens3 \ nexthop encap mpls 20 via fe80::1 dev ens3 RTNETLINK answers: File exists The problem is that the duplicate nexthop detection does not compare lwtunnel configuration. Add it. Fixes: 19e42e451506 ("ipv6: support for fib route lwtunnel encap attributes") Signed-off-by: David Ahern Reported-by: João Taveira Araújo Reported-by: Lennert Buytenhek Acked-by: Roopa Prabhu Tested-by: Lennert Buytenhek Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- include/net/ip6_route.h | 8 ++++++++ net/ipv6/ip6_fib.c | 5 +---- net/ipv6/route.c | 8 +------- 3 files changed, 10 insertions(+), 11 deletions(-) diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 6275d651f76e..b8a8d4239e85 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -21,6 +21,7 @@ struct route_info { #include #include #include +#include #include #include #include @@ -208,4 +209,11 @@ static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt, return daddr; } +static inline bool rt6_duplicate_nexthop(struct rt6_info *a, struct rt6_info *b) +{ + return a->dst.dev == b->dst.dev && + a->rt6i_idev == b->rt6i_idev && + ipv6_addr_equal(&a->rt6i_gateway, &b->rt6i_gateway) && + !lwtunnel_cmp_encap(a->dst.lwtstate, b->dst.lwtstate); +} #endif diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 1ac06723f0d7..f60e8caea767 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -767,10 +767,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, goto next_iter; } - if (iter->dst.dev == rt->dst.dev && - iter->rt6i_idev == rt->rt6i_idev && - ipv6_addr_equal(&iter->rt6i_gateway, - &rt->rt6i_gateway)) { + if (rt6_duplicate_nexthop(iter, rt)) { if (rt->rt6i_nsiblings) rt->rt6i_nsiblings = 0; if (!(iter->rt6i_flags & RTF_EXPIRES)) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8f4177a1d4f5..ef335070e98a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2833,17 +2833,11 @@ static int ip6_route_info_append(struct list_head *rt6_nh_list, struct rt6_info *rt, struct fib6_config *r_cfg) { struct rt6_nh *nh; - struct rt6_info *rtnh; int err = -EEXIST; list_for_each_entry(nh, rt6_nh_list, next) { /* check if rt6_info already exists */ - rtnh = nh->rt6_info; - - if (rtnh->dst.dev == rt->dst.dev && - rtnh->rt6i_idev == rt->rt6i_idev && - ipv6_addr_equal(&rtnh->rt6i_gateway, - &rt->rt6i_gateway)) + if (rt6_duplicate_nexthop(nh->rt6_info, rt)) return err; } From 89e7f17f8f64763394c164ffe218c599cda4ab70 Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 6 Jul 2017 15:24:40 +0300 Subject: [PATCH 12/58] vrf: fix bug_on triggered by rx when destroying a vrf commit f630c38ef0d785101363a8992bbd4f302180f86f upstream. When destroying a VRF device we cleanup the slaves in its ndo_uninit() function, but that causes packets to be switched (skb->dev == vrf being destroyed) even though we're pass the point where the VRF should be receiving any packets while it is being dismantled. This causes a BUG_ON to trigger if we have raw sockets (trace below). The reason is that the inetdev of the VRF has been destroyed but we're still sending packets up the stack with it, so let's free the slaves in the dellink callback as David Ahern suggested. Note that this fix doesn't prevent packets from going up when the VRF device is admin down. [ 35.631371] ------------[ cut here ]------------ [ 35.631603] kernel BUG at net/ipv4/fib_frontend.c:285! [ 35.631854] invalid opcode: 0000 [#1] SMP [ 35.631977] Modules linked in: [ 35.632081] CPU: 2 PID: 22 Comm: ksoftirqd/2 Not tainted 4.12.0-rc7+ #45 [ 35.632247] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [ 35.632477] task: ffff88005ad68000 task.stack: ffff88005ad64000 [ 35.632632] RIP: 0010:fib_compute_spec_dst+0xfc/0x1ee [ 35.632769] RSP: 0018:ffff88005ad67978 EFLAGS: 00010202 [ 35.632910] RAX: 0000000000000001 RBX: ffff880059a7f200 RCX: 0000000000000000 [ 35.633084] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffffffff82274af0 [ 35.633256] RBP: ffff88005ad679f8 R08: 000000000001ef70 R09: 0000000000000046 [ 35.633430] R10: ffff88005ad679f8 R11: ffff880037731cb0 R12: 0000000000000001 [ 35.633603] R13: ffff8800599e3000 R14: 0000000000000000 R15: ffff8800599cb852 [ 35.634114] FS: 0000000000000000(0000) GS:ffff88005d900000(0000) knlGS:0000000000000000 [ 35.634306] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 35.634456] CR2: 00007f3563227095 CR3: 000000000201d000 CR4: 00000000000406e0 [ 35.634632] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 35.634865] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 35.635055] Call Trace: [ 35.635271] ? __lock_acquire+0xf0d/0x1117 [ 35.635522] ipv4_pktinfo_prepare+0x82/0x151 [ 35.635831] raw_rcv_skb+0x17/0x3c [ 35.636062] raw_rcv+0xe5/0xf7 [ 35.636287] raw_local_deliver+0x169/0x1d9 [ 35.636534] ip_local_deliver_finish+0x87/0x1c4 [ 35.636820] ip_local_deliver+0x63/0x7f [ 35.637058] ip_rcv_finish+0x340/0x3a1 [ 35.637295] ip_rcv+0x314/0x34a [ 35.637525] __netif_receive_skb_core+0x49f/0x7c5 [ 35.637780] ? lock_acquire+0x13f/0x1d7 [ 35.638018] ? lock_acquire+0x15e/0x1d7 [ 35.638259] __netif_receive_skb+0x1e/0x94 [ 35.638502] ? __netif_receive_skb+0x1e/0x94 [ 35.638748] netif_receive_skb_internal+0x74/0x300 [ 35.639002] ? dev_gro_receive+0x2ed/0x411 [ 35.639246] ? lock_is_held_type+0xc4/0xd2 [ 35.639491] napi_gro_receive+0x105/0x1a0 [ 35.639736] receive_buf+0xc32/0xc74 [ 35.639965] ? detach_buf+0x67/0x153 [ 35.640201] ? virtqueue_get_buf_ctx+0x120/0x176 [ 35.640453] virtnet_poll+0x128/0x1c5 [ 35.640690] net_rx_action+0x103/0x343 [ 35.640932] __do_softirq+0x1c7/0x4b7 [ 35.641171] run_ksoftirqd+0x23/0x5c [ 35.641403] smpboot_thread_fn+0x24f/0x26d [ 35.641646] ? sort_range+0x22/0x22 [ 35.641878] kthread+0x129/0x131 [ 35.642104] ? __list_add+0x31/0x31 [ 35.642335] ? __list_add+0x31/0x31 [ 35.642568] ret_from_fork+0x2a/0x40 [ 35.642804] Code: 05 bd 87 a3 00 01 e8 1f ef 98 ff 4d 85 f6 48 c7 c7 f0 4a 27 82 41 0f 94 c4 31 c9 31 d2 41 0f b6 f4 e8 04 71 a1 ff 45 84 e4 74 02 <0f> 0b 0f b7 93 c4 00 00 00 4d 8b a5 80 05 00 00 48 03 93 d0 00 [ 35.644342] RIP: fib_compute_spec_dst+0xfc/0x1ee RSP: ffff88005ad67978 Fixes: 193125dbd8eb ("net: Introduce VRF device driver") Reported-by: Chris Cormier Signed-off-by: Nikolay Aleksandrov Acked-by: David Ahern Signed-off-by: David S. Miller [backport to 4.4 - gregkh] Signed-off-by: Greg Kroah-Hartman --- drivers/net/vrf.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index 349aecbc210a..ac945f8781ac 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -733,15 +733,15 @@ static int vrf_del_slave(struct net_device *dev, struct net_device *port_dev) static void vrf_dev_uninit(struct net_device *dev) { struct net_vrf *vrf = netdev_priv(dev); - struct slave_queue *queue = &vrf->queue; - struct list_head *head = &queue->all_slaves; - struct slave *slave, *next; +// struct slave_queue *queue = &vrf->queue; +// struct list_head *head = &queue->all_slaves; +// struct slave *slave, *next; vrf_rtable_destroy(vrf); vrf_rt6_destroy(vrf); - list_for_each_entry_safe(slave, next, head, list) - vrf_del_slave(dev, slave->dev); +// list_for_each_entry_safe(slave, next, head, list) +// vrf_del_slave(dev, slave->dev); free_percpu(dev->dstats); dev->dstats = NULL; @@ -914,6 +914,14 @@ static int vrf_validate(struct nlattr *tb[], struct nlattr *data[]) static void vrf_dellink(struct net_device *dev, struct list_head *head) { + struct net_vrf *vrf = netdev_priv(dev); + struct slave_queue *queue = &vrf->queue; + struct list_head *all_slaves = &queue->all_slaves; + struct slave *slave, *next; + + list_for_each_entry_safe(slave, next, all_slaves, list) + vrf_del_slave(dev, slave->dev); + unregister_netdevice_queue(dev, head); } From 9618eb4af306e7f21330a6acf22cbec037d17f22 Mon Sep 17 00:00:00 2001 From: Sowmini Varadhan Date: Thu, 6 Jul 2017 08:15:06 -0700 Subject: [PATCH 13/58] rds: tcp: use sock_create_lite() to create the accept socket commit 0933a578cd55b02dc80f219dc8f2efb17ec61c9a upstream. There are two problems with calling sock_create_kern() from rds_tcp_accept_one() 1. it sets up a new_sock->sk that is wasteful, because this ->sk is going to get replaced by inet_accept() in the subsequent ->accept() 2. The new_sock->sk is a leaked reference in sock_graft() which expects to find a null parent->sk Avoid these problems by calling sock_create_lite(). Signed-off-by: Sowmini Varadhan Acked-by: Santosh Shilimkar Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/rds/tcp_listen.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/rds/tcp_listen.c b/net/rds/tcp_listen.c index 0936a4a32b47..e353e3255206 100644 --- a/net/rds/tcp_listen.c +++ b/net/rds/tcp_listen.c @@ -78,7 +78,7 @@ int rds_tcp_accept_one(struct socket *sock) struct inet_sock *inet; struct rds_tcp_connection *rs_tcp; - ret = sock_create_kern(sock_net(sock->sk), sock->sk->sk_family, + ret = sock_create_lite(sock->sk->sk_family, sock->sk->sk_type, sock->sk->sk_protocol, &new_sock); if (ret) From 4c7021c2fb74047649c03845ce6fd13626a5a418 Mon Sep 17 00:00:00 2001 From: Arend van Spriel Date: Fri, 7 Jul 2017 21:09:06 +0100 Subject: [PATCH 14/58] brcmfmac: fix possible buffer overflow in brcmf_cfg80211_mgmt_tx() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 8f44c9a41386729fea410e688959ddaa9d51be7c upstream. The lower level nl80211 code in cfg80211 ensures that "len" is between 25 and NL80211_ATTR_FRAME (2304). We subtract DOT11_MGMT_HDR_LEN (24) from "len" so thats's max of 2280. However, the action_frame->data[] buffer is only BRCMF_FIL_ACTION_FRAME_SIZE (1800) bytes long so this memcpy() can overflow. memcpy(action_frame->data, &buf[DOT11_MGMT_HDR_LEN], le16_to_cpu(action_frame->len)); Fixes: 18e2f61db3b70 ("brcmfmac: P2P action frame tx.") Reported-by: "freenerguo(郭大兴)" Signed-off-by: Arend van Spriel Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- drivers/net/wireless/brcm80211/brcmfmac/cfg80211.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/wireless/brcm80211/brcmfmac/cfg80211.c b/drivers/net/wireless/brcm80211/brcmfmac/cfg80211.c index 70a6985334d5..da5826d788d6 100644 --- a/drivers/net/wireless/brcm80211/brcmfmac/cfg80211.c +++ b/drivers/net/wireless/brcm80211/brcmfmac/cfg80211.c @@ -4472,6 +4472,11 @@ brcmf_cfg80211_mgmt_tx(struct wiphy *wiphy, struct wireless_dev *wdev, cfg80211_mgmt_tx_status(wdev, *cookie, buf, len, true, GFP_KERNEL); } else if (ieee80211_is_action(mgmt->frame_control)) { + if (len > BRCMF_FIL_ACTION_FRAME_SIZE + DOT11_MGMT_HDR_LEN) { + brcmf_err("invalid action frame length\n"); + err = -EINVAL; + goto exit; + } af_params = kzalloc(sizeof(*af_params), GFP_KERNEL); if (af_params == NULL) { brcmf_err("unable to allocate frame\n"); From 05bf0b6ef9ce7e8967c96fd419ad0ee5d7fe5418 Mon Sep 17 00:00:00 2001 From: Srinivas Dasari Date: Fri, 7 Jul 2017 01:43:41 +0300 Subject: [PATCH 15/58] cfg80211: Define nla_policy for NL80211_ATTR_LOCAL_MESH_POWER_MODE commit 8feb69c7bd89513be80eb19198d48f154b254021 upstream. Buffer overread may happen as nl80211_set_station() reads 4 bytes from the attribute NL80211_ATTR_LOCAL_MESH_POWER_MODE without validating the size of data received when userspace sends less than 4 bytes of data with NL80211_ATTR_LOCAL_MESH_POWER_MODE. Define nla_policy for NL80211_ATTR_LOCAL_MESH_POWER_MODE to avoid the buffer overread. Fixes: 3b1c5a5307f ("{cfg,nl}80211: mesh power mode primitives and userspace access") Signed-off-by: Srinivas Dasari Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg Signed-off-by: Greg Kroah-Hartman --- net/wireless/nl80211.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 9d0953e5734f..c1dec30a226b 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -359,6 +359,7 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_SCAN_FLAGS] = { .type = NLA_U32 }, [NL80211_ATTR_P2P_CTWINDOW] = { .type = NLA_U8 }, [NL80211_ATTR_P2P_OPPPS] = { .type = NLA_U8 }, + [NL80211_ATTR_LOCAL_MESH_POWER_MODE] = {. type = NLA_U32 }, [NL80211_ATTR_ACL_POLICY] = {. type = NLA_U32 }, [NL80211_ATTR_MAC_ADDRS] = { .type = NLA_NESTED }, [NL80211_ATTR_STA_CAPABILITY] = { .type = NLA_U16 }, From 24d0410712b06e102d330c6ccb817e56b3c85ec9 Mon Sep 17 00:00:00 2001 From: Srinivas Dasari Date: Fri, 7 Jul 2017 01:43:42 +0300 Subject: [PATCH 16/58] cfg80211: Validate frequencies nested in NL80211_ATTR_SCAN_FREQUENCIES commit d7f13f7450369281a5d0ea463cc69890a15923ae upstream. validate_scan_freqs() retrieves frequencies from attributes nested in the attribute NL80211_ATTR_SCAN_FREQUENCIES with nla_get_u32(), which reads 4 bytes from each attribute without validating the size of data received. Attributes nested in NL80211_ATTR_SCAN_FREQUENCIES don't have an nla policy. Validate size of each attribute before parsing to avoid potential buffer overread. Fixes: 2a519311926 ("cfg80211/nl80211: scanning (and mac80211 update to use it)") Signed-off-by: Srinivas Dasari Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg Signed-off-by: Greg Kroah-Hartman --- net/wireless/nl80211.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index c1dec30a226b..793291cabb44 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -5706,6 +5706,10 @@ static int validate_scan_freqs(struct nlattr *freqs) struct nlattr *attr1, *attr2; int n_channels = 0, tmp1, tmp2; + nla_for_each_nested(attr1, freqs, tmp1) + if (nla_len(attr1) != sizeof(u32)) + return 0; + nla_for_each_nested(attr1, freqs, tmp1) { n_channels++; /* From 2d3c10e215713f242e0e0d67a38a2deb00e92a1b Mon Sep 17 00:00:00 2001 From: Srinivas Dasari Date: Fri, 7 Jul 2017 01:43:39 +0300 Subject: [PATCH 17/58] cfg80211: Check if PMKID attribute is of expected size commit 9361df14d1cbf966409d5d6f48bb334384fbe138 upstream. nla policy checks for only maximum length of the attribute data when the attribute type is NLA_BINARY. If userspace sends less data than specified, the wireless drivers may access illegal memory. When type is NLA_UNSPEC, nla policy check ensures that userspace sends minimum specified length number of bytes. Remove type assignment to NLA_BINARY from nla_policy of NL80211_ATTR_PMKID to make this NLA_UNSPEC and to make sure minimum WLAN_PMKID_LEN bytes are received from userspace with NL80211_ATTR_PMKID. Fixes: 67fbb16be69d ("nl80211: PMKSA caching support") Signed-off-by: Srinivas Dasari Signed-off-by: Jouni Malinen Signed-off-by: Johannes Berg Signed-off-by: Greg Kroah-Hartman --- net/wireless/nl80211.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 793291cabb44..de10e3c0e2a4 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -302,8 +302,7 @@ static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = { [NL80211_ATTR_WPA_VERSIONS] = { .type = NLA_U32 }, [NL80211_ATTR_PID] = { .type = NLA_U32 }, [NL80211_ATTR_4ADDR] = { .type = NLA_U8 }, - [NL80211_ATTR_PMKID] = { .type = NLA_BINARY, - .len = WLAN_PMKID_LEN }, + [NL80211_ATTR_PMKID] = { .len = WLAN_PMKID_LEN }, [NL80211_ATTR_DURATION] = { .type = NLA_U32 }, [NL80211_ATTR_COOKIE] = { .type = NLA_U64 }, [NL80211_ATTR_TX_RATES] = { .type = NLA_NESTED }, From 970616464f43b33e309996dc58bbf49a8a29f5e4 Mon Sep 17 00:00:00 2001 From: Suzuki K Poulose Date: Fri, 30 Jun 2017 10:58:28 +0100 Subject: [PATCH 18/58] irqchip/gic-v3: Fix out-of-bound access in gic_set_affinity commit 866d7c1b0a3c70387646c4e455e727a58c5d465a upstream. The GICv3 driver doesn't check if the target CPU for gic_set_affinity is valid before going ahead and making the changes. This triggers the following splat with KASAN: [ 141.189434] BUG: KASAN: global-out-of-bounds in gic_set_affinity+0x8c/0x140 [ 141.189704] Read of size 8 at addr ffff200009741d20 by task swapper/1/0 [ 141.189958] [ 141.190158] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 4.12.0-rc7 [ 141.190458] Hardware name: Foundation-v8A (DT) [ 141.190658] Call trace: [ 141.190908] [] dump_backtrace+0x0/0x328 [ 141.191224] [] show_stack+0x14/0x20 [ 141.191507] [] dump_stack+0xa4/0xc8 [ 141.191858] [] print_address_description+0x13c/0x250 [ 141.192219] [] kasan_report+0x210/0x300 [ 141.192547] [] __asan_load8+0x84/0x98 [ 141.192874] [] gic_set_affinity+0x8c/0x140 [ 141.193158] [] irq_do_set_affinity+0x54/0xb8 [ 141.193473] [] irq_set_affinity_locked+0x64/0xf0 [ 141.193828] [] __irq_set_affinity+0x48/0x78 [ 141.194158] [] arm_perf_starting_cpu+0x104/0x150 [ 141.194513] [] cpuhp_invoke_callback+0x17c/0x1f8 [ 141.194783] [] notify_cpu_starting+0x8c/0xb8 [ 141.195130] [] secondary_start_kernel+0x15c/0x200 [ 141.195390] [<0000000080db81b4>] 0x80db81b4 [ 141.195603] [ 141.195685] The buggy address belongs to the variable: [ 141.196012] __cpu_logical_map+0x200/0x220 [ 141.196176] [ 141.196315] Memory state around the buggy address: [ 141.196586] ffff200009741c00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 141.196913] ffff200009741c80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 141.197158] >ffff200009741d00: 00 00 00 00 fa fa fa fa 00 00 00 00 00 00 00 00 [ 141.197487] ^ [ 141.197758] ffff200009741d80: 00 00 00 00 00 00 00 00 fa fa fa fa 00 00 00 00 [ 141.198060] ffff200009741e00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 [ 141.198358] ================================================================== [ 141.198609] Disabling lock debugging due to kernel taint [ 141.198961] CPU1: Booted secondary processor [410fd051] This patch adds the check to make sure the cpu is valid. Fixes: commit 021f653791ad17e03f98 ("irqchip: gic-v3: Initial support for GICv3") Signed-off-by: Suzuki K Poulose Signed-off-by: Marc Zyngier Signed-off-by: Greg Kroah-Hartman --- drivers/irqchip/irq-gic-v3.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index e33c729b9f48..5a1490b046ac 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -632,6 +632,9 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *mask_val, int enabled; u64 val; + if (cpu >= nr_cpu_ids) + return -EINVAL; + if (gic_irq_in_rdist(d)) return -EINVAL; From e18ca17b9040bedd3768a257d46537471eff45a1 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Sun, 2 Jul 2017 22:00:41 +0200 Subject: [PATCH 19/58] parisc: Report SIGSEGV instead of SIGBUS when running out of stack commit 247462316f85a9e0479445c1a4223950b68ffac1 upstream. When a process runs out of stack the parisc kernel wrongly faults with SIGBUS instead of the expected SIGSEGV signal. This example shows how the kernel faults: do_page_fault() command='a.out' type=15 address=0xfaac2000 in libc-2.24.so[f8308000+16c000] trap #15: Data TLB miss fault, vm_start = 0xfa2c2000, vm_end = 0xfaac2000 The vma->vm_end value is the first address which does not belong to the vma, so adjust the check to include vma->vm_end to the range for which to send the SIGSEGV signal. This patch unbreaks building the debian libsigsegv package. Signed-off-by: Helge Deller Signed-off-by: Greg Kroah-Hartman --- arch/parisc/mm/fault.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c index f9064449908a..d8c2f3bcfc18 100644 --- a/arch/parisc/mm/fault.c +++ b/arch/parisc/mm/fault.c @@ -303,7 +303,7 @@ bad_area: case 15: /* Data TLB miss fault/Data page fault */ /* send SIGSEGV when outside of vma */ if (!vma || - address < vma->vm_start || address > vma->vm_end) { + address < vma->vm_start || address >= vma->vm_end) { si.si_signo = SIGSEGV; si.si_code = SEGV_MAPERR; break; From f265641dc874b01f816fadd70f175ee8de806266 Mon Sep 17 00:00:00 2001 From: Eric Biggers Date: Mon, 12 Jun 2017 23:18:30 -0700 Subject: [PATCH 20/58] parisc: use compat_sys_keyctl() commit b0f94efd5aa8daa8a07d7601714c2573266cd4c9 upstream. Architectures with a compat syscall table must put compat_sys_keyctl() in it, not sys_keyctl(). The parisc architecture was not doing this; fix it. Signed-off-by: Eric Biggers Acked-by: Helge Deller Signed-off-by: Helge Deller Signed-off-by: Greg Kroah-Hartman --- arch/parisc/kernel/syscall_table.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S index d4ffcfbc9885..041e1f9ec129 100644 --- a/arch/parisc/kernel/syscall_table.S +++ b/arch/parisc/kernel/syscall_table.S @@ -361,7 +361,7 @@ ENTRY_SAME(ni_syscall) /* 263: reserved for vserver */ ENTRY_SAME(add_key) ENTRY_SAME(request_key) /* 265 */ - ENTRY_SAME(keyctl) + ENTRY_COMP(keyctl) ENTRY_SAME(ioprio_set) ENTRY_SAME(ioprio_get) ENTRY_SAME(inotify_init) From 635a58225c360e293154a05fc66843dc686989fe Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Mon, 3 Jul 2017 10:38:05 +0200 Subject: [PATCH 21/58] parisc: DMA API: return error instead of BUG_ON for dma ops on non dma devs commit 33f9e02495d15a061f0c94ef46f5103a2d0c20f3 upstream. Enabling parport pc driver on a B2600 (and probably other 64bit PARISC systems) produced following BUG: CPU: 0 PID: 1 Comm: swapper Not tainted 4.12.0-rc5-30198-g1132d5e #156 task: 000000009e050000 task.stack: 000000009e04c000 YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI PSW: 00001000000001101111111100001111 Not tainted r00-03 000000ff0806ff0f 000000009e04c990 0000000040871b78 000000009e04cac0 r04-07 0000000040c14de0 ffffffffffffffff 000000009e07f098 000000009d82d200 r08-11 000000009d82d210 0000000000000378 0000000000000000 0000000040c345e0 r12-15 0000000000000005 0000000040c345e0 0000000000000000 0000000040c9d5e0 r16-19 0000000040c345e0 00000000f00001c4 00000000f00001bc 0000000000000061 r20-23 000000009e04ce28 0000000000000010 0000000000000010 0000000040b89e40 r24-27 0000000000000003 0000000000ffffff 000000009d82d210 0000000040c14de0 r28-31 0000000000000000 000000009e04ca90 000000009e04cb40 0000000000000000 sr00-03 0000000000000000 0000000000000000 0000000000000000 0000000000000000 sr04-07 0000000000000000 0000000000000000 0000000000000000 0000000000000000 IASQ: 0000000000000000 0000000000000000 IAOQ: 00000000404aece0 00000000404aece4 IIR: 03ffe01f ISR: 0000000010340000 IOR: 000001781304cac8 CPU: 0 CR30: 000000009e04c000 CR31: 00000000e2976de2 ORIG_R28: 0000000000000200 IAOQ[0]: sba_dma_supported+0x80/0xd0 IAOQ[1]: sba_dma_supported+0x84/0xd0 RP(r2): parport_pc_probe_port+0x178/0x1200 Cause is a call to dma_coerce_mask_and_coherenet in parport_pc_probe_port, which PARISC DMA API doesn't handle very nicely. This commit gives back DMA_ERROR_CODE for DMA API calls, if device isn't capable of DMA transaction. Signed-off-by: Thomas Bogendoerfer Signed-off-by: Helge Deller Signed-off-by: Greg Kroah-Hartman --- arch/parisc/include/asm/dma-mapping.h | 11 +++++++---- drivers/parisc/ccio-dma.c | 12 ++++++++++++ drivers/parisc/dino.c | 5 ++++- drivers/parisc/lba_pci.c | 6 ++++-- drivers/parisc/sba_iommu.c | 14 ++++++++++++++ 5 files changed, 41 insertions(+), 7 deletions(-) diff --git a/arch/parisc/include/asm/dma-mapping.h b/arch/parisc/include/asm/dma-mapping.h index d8d60a57183f..f53725202955 100644 --- a/arch/parisc/include/asm/dma-mapping.h +++ b/arch/parisc/include/asm/dma-mapping.h @@ -39,6 +39,8 @@ struct hppa_dma_ops { ** flush/purge and allocate "regular" cacheable pages for everything. */ +#define DMA_ERROR_CODE (~(dma_addr_t)0) + #ifdef CONFIG_PA11 extern struct hppa_dma_ops pcxl_dma_ops; extern struct hppa_dma_ops pcx_dma_ops; @@ -209,12 +211,13 @@ parisc_walk_tree(struct device *dev) break; } } - BUG_ON(!dev->platform_data); return dev->platform_data; } - -#define GET_IOC(dev) (HBA_DATA(parisc_walk_tree(dev))->iommu) - + +#define GET_IOC(dev) ({ \ + void *__pdata = parisc_walk_tree(dev); \ + __pdata ? HBA_DATA(__pdata)->iommu : NULL; \ +}) #ifdef CONFIG_IOMMU_CCIO struct parisc_device; diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c index 8e11fb2831cd..34f1d6b41fb9 100644 --- a/drivers/parisc/ccio-dma.c +++ b/drivers/parisc/ccio-dma.c @@ -741,6 +741,8 @@ ccio_map_single(struct device *dev, void *addr, size_t size, BUG_ON(!dev); ioc = GET_IOC(dev); + if (!ioc) + return DMA_ERROR_CODE; BUG_ON(size <= 0); @@ -805,6 +807,10 @@ ccio_unmap_single(struct device *dev, dma_addr_t iova, size_t size, BUG_ON(!dev); ioc = GET_IOC(dev); + if (!ioc) { + WARN_ON(!ioc); + return; + } DBG_RUN("%s() iovp 0x%lx/%x\n", __func__, (long)iova, size); @@ -908,6 +914,8 @@ ccio_map_sg(struct device *dev, struct scatterlist *sglist, int nents, BUG_ON(!dev); ioc = GET_IOC(dev); + if (!ioc) + return 0; DBG_RUN_SG("%s() START %d entries\n", __func__, nents); @@ -980,6 +988,10 @@ ccio_unmap_sg(struct device *dev, struct scatterlist *sglist, int nents, BUG_ON(!dev); ioc = GET_IOC(dev); + if (!ioc) { + WARN_ON(!ioc); + return; + } DBG_RUN_SG("%s() START %d entries, %p,%x\n", __func__, nents, sg_virt(sglist), sglist->length); diff --git a/drivers/parisc/dino.c b/drivers/parisc/dino.c index a0580afe1713..7b0ca1551d7b 100644 --- a/drivers/parisc/dino.c +++ b/drivers/parisc/dino.c @@ -154,7 +154,10 @@ struct dino_device }; /* Looks nice and keeps the compiler happy */ -#define DINO_DEV(d) ((struct dino_device *) d) +#define DINO_DEV(d) ({ \ + void *__pdata = d; \ + BUG_ON(!__pdata); \ + (struct dino_device *)__pdata; }) /* diff --git a/drivers/parisc/lba_pci.c b/drivers/parisc/lba_pci.c index 42844c2bc065..d0c2759076a2 100644 --- a/drivers/parisc/lba_pci.c +++ b/drivers/parisc/lba_pci.c @@ -111,8 +111,10 @@ static u32 lba_t32; /* Looks nice and keeps the compiler happy */ -#define LBA_DEV(d) ((struct lba_device *) (d)) - +#define LBA_DEV(d) ({ \ + void *__pdata = d; \ + BUG_ON(!__pdata); \ + (struct lba_device *)__pdata; }) /* ** Only allow 8 subsidiary busses per LBA diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c index 225049b492e5..d6326144ce01 100644 --- a/drivers/parisc/sba_iommu.c +++ b/drivers/parisc/sba_iommu.c @@ -691,6 +691,8 @@ static int sba_dma_supported( struct device *dev, u64 mask) return 0; ioc = GET_IOC(dev); + if (!ioc) + return 0; /* * check if mask is >= than the current max IO Virt Address @@ -722,6 +724,8 @@ sba_map_single(struct device *dev, void *addr, size_t size, int pide; ioc = GET_IOC(dev); + if (!ioc) + return DMA_ERROR_CODE; /* save offset bits */ offset = ((dma_addr_t) (long) addr) & ~IOVP_MASK; @@ -803,6 +807,10 @@ sba_unmap_single(struct device *dev, dma_addr_t iova, size_t size, DBG_RUN("%s() iovp 0x%lx/%x\n", __func__, (long) iova, size); ioc = GET_IOC(dev); + if (!ioc) { + WARN_ON(!ioc); + return; + } offset = iova & ~IOVP_MASK; iova ^= offset; /* clear offset bits */ size += offset; @@ -942,6 +950,8 @@ sba_map_sg(struct device *dev, struct scatterlist *sglist, int nents, DBG_RUN_SG("%s() START %d entries\n", __func__, nents); ioc = GET_IOC(dev); + if (!ioc) + return 0; /* Fast path single entry scatterlists. */ if (nents == 1) { @@ -1027,6 +1037,10 @@ sba_unmap_sg(struct device *dev, struct scatterlist *sglist, int nents, __func__, nents, sg_virt(sglist), sglist->length); ioc = GET_IOC(dev); + if (!ioc) { + WARN_ON(!ioc); + return; + } #ifdef SBA_COLLECT_STATS ioc->usg_calls++; From b29145746576a638f141e16f9438f0e98a8e3ea0 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Mon, 29 May 2017 17:14:16 +0200 Subject: [PATCH 22/58] parisc/mm: Ensure IRQs are off in switch_mm() commit 649aa24254e85bf6bd7807dd372d083707852b1f upstream. This is because of commit f98db6013c55 ("sched/core: Add switch_mm_irqs_off() and use it in the scheduler") in which switch_mm_irqs_off() is called by the scheduler, vs switch_mm() which is used by use_mm(). This patch lets the parisc code mirror the x86 and powerpc code, ie. it disables interrupts in switch_mm(), and optimises the scheduler case by defining switch_mm_irqs_off(). Signed-off-by: Helge Deller Signed-off-by: Helge Deller Signed-off-by: Greg Kroah-Hartman --- arch/parisc/include/asm/mmu_context.h | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/arch/parisc/include/asm/mmu_context.h b/arch/parisc/include/asm/mmu_context.h index 59be25764433..a81226257878 100644 --- a/arch/parisc/include/asm/mmu_context.h +++ b/arch/parisc/include/asm/mmu_context.h @@ -49,15 +49,26 @@ static inline void load_context(mm_context_t context) mtctl(__space_to_prot(context), 8); } -static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next, struct task_struct *tsk) +static inline void switch_mm_irqs_off(struct mm_struct *prev, + struct mm_struct *next, struct task_struct *tsk) { - if (prev != next) { mtctl(__pa(next->pgd), 25); load_context(next->context); } } +static inline void switch_mm(struct mm_struct *prev, + struct mm_struct *next, struct task_struct *tsk) +{ + unsigned long flags; + + local_irq_save(flags); + switch_mm_irqs_off(prev, next, tsk); + local_irq_restore(flags); +} +#define switch_mm_irqs_off switch_mm_irqs_off + #define deactivate_mm(tsk,mm) do { } while (0) static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next) From 0d6ee0bea866a40c537a0d2277e0bf12430e1334 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 25 May 2017 12:58:33 +0000 Subject: [PATCH 23/58] tools/lib/lockdep: Reduce MAX_LOCK_DEPTH to avoid overflowing lock_chain/: Depth commit 98dcea0cfd04e083ac74137ceb9a632604740e2d upstream. liblockdep has been broken since commit 75dd602a5198 ("lockdep: Fix lock_chain::base size"), as that adds a check that MAX_LOCK_DEPTH is within the range of lock_chain::depth and in liblockdep it is much too large. That should have resulted in a compiler error, but didn't because: - the check uses ARRAY_SIZE(), which isn't yet defined in liblockdep so is assumed to be an (undeclared) function - putting a function call inside a BUILD_BUG_ON() expression quietly turns it into some nonsense involving a variable-length array It did produce a compiler warning, but I didn't notice because liblockdep already produces too many warnings if -Wall is enabled (which I'll fix shortly). Even before that commit, which reduced lock_chain::depth from 8 bits to 6, MAX_LOCK_DEPTH was too large. Signed-off-by: Ben Hutchings Signed-off-by: Sasha Levin Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: a.p.zijlstra@chello.nl Link: http://lkml.kernel.org/r/20170525130005.5947-3-alexander.levin@verizon.com Signed-off-by: Ingo Molnar Signed-off-by: Greg Kroah-Hartman --- tools/lib/lockdep/uinclude/linux/lockdep.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/lib/lockdep/uinclude/linux/lockdep.h b/tools/lib/lockdep/uinclude/linux/lockdep.h index c808c7d02d21..e69118b2077e 100644 --- a/tools/lib/lockdep/uinclude/linux/lockdep.h +++ b/tools/lib/lockdep/uinclude/linux/lockdep.h @@ -8,7 +8,7 @@ #include #include -#define MAX_LOCK_DEPTH 2000UL +#define MAX_LOCK_DEPTH 255UL #define asmlinkage #define __visible From 717ce69e47f0d77571b5a564784d16c1d920ee9b Mon Sep 17 00:00:00 2001 From: Marcin Nowakowski Date: Thu, 6 Jul 2017 15:35:31 -0700 Subject: [PATCH 24/58] kernel/extable.c: mark core_kernel_text notrace commit c0d80ddab89916273cb97114889d3f337bc370ae upstream. core_kernel_text is used by MIPS in its function graph trace processing, so having this method traced leads to an infinite set of recursive calls such as: Call Trace: ftrace_return_to_handler+0x50/0x128 core_kernel_text+0x10/0x1b8 prepare_ftrace_return+0x6c/0x114 ftrace_graph_caller+0x20/0x44 return_to_handler+0x10/0x30 return_to_handler+0x0/0x30 return_to_handler+0x0/0x30 ftrace_ops_no_ops+0x114/0x1bc core_kernel_text+0x10/0x1b8 core_kernel_text+0x10/0x1b8 core_kernel_text+0x10/0x1b8 ftrace_ops_no_ops+0x114/0x1bc core_kernel_text+0x10/0x1b8 prepare_ftrace_return+0x6c/0x114 ftrace_graph_caller+0x20/0x44 (...) Mark the function notrace to avoid it being traced. Link: http://lkml.kernel.org/r/1498028607-6765-1-git-send-email-marcin.nowakowski@imgtec.com Signed-off-by: Marcin Nowakowski Reviewed-by: Masami Hiramatsu Cc: Peter Zijlstra Cc: Thomas Meyer Cc: Ingo Molnar Cc: Steven Rostedt Cc: Daniel Borkmann Cc: Paul Gortmaker Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- kernel/extable.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/extable.c b/kernel/extable.c index e820ccee9846..4f06fc34313f 100644 --- a/kernel/extable.c +++ b/kernel/extable.c @@ -66,7 +66,7 @@ static inline int init_kernel_text(unsigned long addr) return 0; } -int core_kernel_text(unsigned long addr) +int notrace core_kernel_text(unsigned long addr) { if (addr >= (unsigned long)_stext && addr < (unsigned long)_etext) From 2d0db02d2e8f45e215fd2e646820c669c6e70159 Mon Sep 17 00:00:00 2001 From: Sahitya Tummala Date: Mon, 10 Jul 2017 15:49:57 -0700 Subject: [PATCH 25/58] mm/list_lru.c: fix list_lru_count_node() to be race free commit 2c80cd57c74339889a8752b20862a16c28929c3a upstream. list_lru_count_node() iterates over all memcgs to get the total number of entries on the node but it can race with memcg_drain_all_list_lrus(), which migrates the entries from a dead cgroup to another. This can return incorrect number of entries from list_lru_count_node(). Fix this by keeping track of entries per node and simply return it in list_lru_count_node(). Link: http://lkml.kernel.org/r/1498707555-30525-1-git-send-email-stummala@codeaurora.org Signed-off-by: Sahitya Tummala Acked-by: Vladimir Davydov Cc: Jan Kara Cc: Alexander Polakov Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- include/linux/list_lru.h | 1 + mm/list_lru.c | 14 ++++++-------- 2 files changed, 7 insertions(+), 8 deletions(-) diff --git a/include/linux/list_lru.h b/include/linux/list_lru.h index 2a6b9947aaa3..743b34f56f2b 100644 --- a/include/linux/list_lru.h +++ b/include/linux/list_lru.h @@ -44,6 +44,7 @@ struct list_lru_node { /* for cgroup aware lrus points to per cgroup lists, otherwise NULL */ struct list_lru_memcg *memcg_lrus; #endif + long nr_items; } ____cacheline_aligned_in_smp; struct list_lru { diff --git a/mm/list_lru.c b/mm/list_lru.c index 5d8dffd5b57c..786176b1a0ee 100644 --- a/mm/list_lru.c +++ b/mm/list_lru.c @@ -117,6 +117,7 @@ bool list_lru_add(struct list_lru *lru, struct list_head *item) l = list_lru_from_kmem(nlru, item); list_add_tail(item, &l->list); l->nr_items++; + nlru->nr_items++; spin_unlock(&nlru->lock); return true; } @@ -136,6 +137,7 @@ bool list_lru_del(struct list_lru *lru, struct list_head *item) l = list_lru_from_kmem(nlru, item); list_del_init(item); l->nr_items--; + nlru->nr_items--; spin_unlock(&nlru->lock); return true; } @@ -183,15 +185,10 @@ EXPORT_SYMBOL_GPL(list_lru_count_one); unsigned long list_lru_count_node(struct list_lru *lru, int nid) { - long count = 0; - int memcg_idx; + struct list_lru_node *nlru; - count += __list_lru_count_one(lru, nid, -1); - if (list_lru_memcg_aware(lru)) { - for_each_memcg_cache_index(memcg_idx) - count += __list_lru_count_one(lru, nid, memcg_idx); - } - return count; + nlru = &lru->node[nid]; + return nlru->nr_items; } EXPORT_SYMBOL_GPL(list_lru_count_node); @@ -226,6 +223,7 @@ restart: assert_spin_locked(&nlru->lock); case LRU_REMOVED: isolated++; + nlru->nr_items--; /* * If the lru lock has been dropped, our list * traversal is now invalid and so we have to From 68b0f5d85b3a25ee452424e18fe06b93da8dd4d2 Mon Sep 17 00:00:00 2001 From: Sahitya Tummala Date: Mon, 10 Jul 2017 15:50:00 -0700 Subject: [PATCH 26/58] fs/dcache.c: fix spin lockup issue on nlru->lock commit b17c070fb624cf10162cf92ea5e1ec25cd8ac176 upstream. __list_lru_walk_one() acquires nlru spin lock (nlru->lock) for longer duration if there are more number of items in the lru list. As per the current code, it can hold the spin lock for upto maximum UINT_MAX entries at a time. So if there are more number of items in the lru list, then "BUG: spinlock lockup suspected" is observed in the below path: spin_bug+0x90 do_raw_spin_lock+0xfc _raw_spin_lock+0x28 list_lru_add+0x28 dput+0x1c8 path_put+0x20 terminate_walk+0x3c path_lookupat+0x100 filename_lookup+0x6c user_path_at_empty+0x54 SyS_faccessat+0xd0 el0_svc_naked+0x24 This nlru->lock is acquired by another CPU in this path - d_lru_shrink_move+0x34 dentry_lru_isolate_shrink+0x48 __list_lru_walk_one.isra.10+0x94 list_lru_walk_node+0x40 shrink_dcache_sb+0x60 do_remount_sb+0xbc do_emergency_remount+0xb0 process_one_work+0x228 worker_thread+0x2e0 kthread+0xf4 ret_from_fork+0x10 Fix this lockup by reducing the number of entries to be shrinked from the lru list to 1024 at once. Also, add cond_resched() before processing the lru list again. Link: http://marc.info/?t=149722864900001&r=1&w=2 Link: http://lkml.kernel.org/r/1498707575-2472-1-git-send-email-stummala@codeaurora.org Signed-off-by: Sahitya Tummala Suggested-by: Jan Kara Suggested-by: Vladimir Davydov Acked-by: Vladimir Davydov Cc: Alexander Polakov Cc: Al Viro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- fs/dcache.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index 849c1c1e787b..3000cbb54949 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1128,11 +1128,12 @@ void shrink_dcache_sb(struct super_block *sb) LIST_HEAD(dispose); freed = list_lru_walk(&sb->s_dentry_lru, - dentry_lru_isolate_shrink, &dispose, UINT_MAX); + dentry_lru_isolate_shrink, &dispose, 1024); this_cpu_sub(nr_dentry_unused, freed); shrink_dentry_list(&dispose); - } while (freed > 0); + cond_resched(); + } while (list_lru_count(&sb->s_dentry_lru) > 0); } EXPORT_SYMBOL(shrink_dcache_sb); From 4544e9ebef4c2f2536d3c7ca6c6a22d0f9f13d2f Mon Sep 17 00:00:00 2001 From: Cyril Bur Date: Mon, 10 Jul 2017 15:52:21 -0700 Subject: [PATCH 27/58] checkpatch: silence perl 5.26.0 unescaped left brace warnings commit 8d81ae05d0176da1c54aeaed697fa34be5c5575e upstream. As of perl 5, version 26, subversion 0 (v5.26.0) some new warnings have occurred when running checkpatch. Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through in regex; marked by <-- HERE in m/^(.\s*){ <-- HERE \s*/ at scripts/checkpatch.pl line 3544. Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through in regex; marked by <-- HERE in m/^(.\s*){ <-- HERE \s*/ at scripts/checkpatch.pl line 3885. Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through in regex; marked by <-- HERE in m/^(\+.*(?:do|\))){ <-- HERE / at scripts/checkpatch.pl line 4374. It seems perfectly reasonable to do as the warning suggests and simply escape the left brace in these three locations. Link: http://lkml.kernel.org/r/20170607060135.17384-1-cyrilbur@gmail.com Signed-off-by: Cyril Bur Acked-by: Joe Perches Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- scripts/checkpatch.pl | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl index 2b3c22808c3b..6ac6550d751c 100755 --- a/scripts/checkpatch.pl +++ b/scripts/checkpatch.pl @@ -3252,7 +3252,7 @@ sub process { $fixedline =~ s/\s*=\s*$/ = {/; fix_insert_line($fixlinenr, $fixedline); $fixedline = $line; - $fixedline =~ s/^(.\s*){\s*/$1/; + $fixedline =~ s/^(.\s*)\{\s*/$1/; fix_insert_line($fixlinenr, $fixedline); } } @@ -3602,7 +3602,7 @@ sub process { my $fixedline = rtrim($prevrawline) . " {"; fix_insert_line($fixlinenr, $fixedline); $fixedline = $rawline; - $fixedline =~ s/^(.\s*){\s*/$1\t/; + $fixedline =~ s/^(.\s*)\{\s*/$1\t/; if ($fixedline !~ /^\+\s*$/) { fix_insert_line($fixlinenr, $fixedline); } @@ -4091,7 +4091,7 @@ sub process { if (ERROR("SPACING", "space required before the open brace '{'\n" . $herecurr) && $fix) { - $fixed[$fixlinenr] =~ s/^(\+.*(?:do|\))){/$1 {/; + $fixed[$fixlinenr] =~ s/^(\+.*(?:do|\)))\{/$1 {/; } } From 7eb968cd04d404e6c73cd82c1122f6e06ad2d1e8 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 10 Jul 2017 15:52:37 -0700 Subject: [PATCH 28/58] binfmt_elf: use ELF_ET_DYN_BASE only for PIE commit eab09532d40090698b05a07c1c87f39fdbc5fab5 upstream. The ELF_ET_DYN_BASE position was originally intended to keep loaders away from ET_EXEC binaries. (For example, running "/lib/ld-linux.so.2 /bin/cat" might cause the subsequent load of /bin/cat into where the loader had been loaded.) With the advent of PIE (ET_DYN binaries with an INTERP Program Header), ELF_ET_DYN_BASE continued to be used since the kernel was only looking at ET_DYN. However, since ELF_ET_DYN_BASE is traditionally set at the top 1/3rd of the TASK_SIZE, a substantial portion of the address space is unused. For 32-bit tasks when RLIMIT_STACK is set to RLIM_INFINITY, programs are loaded above the mmap region. This means they can be made to collide (CVE-2017-1000370) or nearly collide (CVE-2017-1000371) with pathological stack regions. Lowering ELF_ET_DYN_BASE solves both by moving programs below the mmap region in all cases, and will now additionally avoid programs falling back to the mmap region by enforcing MAP_FIXED for program loads (i.e. if it would have collided with the stack, now it will fail to load instead of falling back to the mmap region). To allow for a lower ELF_ET_DYN_BASE, loaders (ET_DYN without INTERP) are loaded into the mmap region, leaving space available for either an ET_EXEC binary with a fixed location or PIE being loaded into mmap by the loader. Only PIE programs are loaded offset from ELF_ET_DYN_BASE, which means architectures can now safely lower their values without risk of loaders colliding with their subsequently loaded programs. For 64-bit, ELF_ET_DYN_BASE is best set to 4GB to allow runtimes to use the entire 32-bit address space for 32-bit pointers. Thanks to PaX Team, Daniel Micay, and Rik van Riel for inspiration and suggestions on how to implement this solution. Fixes: d1fd836dcf00 ("mm: split ET_DYN ASLR from mmap ASLR") Link: http://lkml.kernel.org/r/20170621173201.GA114489@beast Signed-off-by: Kees Cook Acked-by: Rik van Riel Cc: Daniel Micay Cc: Qualys Security Advisory Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Alexander Viro Cc: Dmitry Safonov Cc: Andy Lutomirski Cc: Grzegorz Andrejczuk Cc: Masahiro Yamada Cc: Benjamin Herrenschmidt Cc: Catalin Marinas Cc: Heiko Carstens Cc: James Hogan Cc: Martin Schwidefsky Cc: Michael Ellerman Cc: Paul Mackerras Cc: Pratyush Anand Cc: Russell King Cc: Will Deacon Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- arch/x86/include/asm/elf.h | 13 +++++---- fs/binfmt_elf.c | 59 ++++++++++++++++++++++++++++++++------ 2 files changed, 58 insertions(+), 14 deletions(-) diff --git a/arch/x86/include/asm/elf.h b/arch/x86/include/asm/elf.h index d262f985bbc8..07cf288b692e 100644 --- a/arch/x86/include/asm/elf.h +++ b/arch/x86/include/asm/elf.h @@ -245,12 +245,13 @@ extern int force_personality32; #define CORE_DUMP_USE_REGSET #define ELF_EXEC_PAGESIZE 4096 -/* This is the location that an ET_DYN program is loaded if exec'ed. Typical - use of this is to invoke "./ld.so someprog" to test out a new version of - the loader. We need to make sure that it is out of the way of the program - that it will "exec", and that there is sufficient room for the brk. */ - -#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) +/* + * This is the base location for PIE (ET_DYN with INTERP) loads. On + * 64-bit, this is raised to 4GB to leave the entire 32-bit address + * space open for things that want to use the area for 32-bit pointers. + */ +#define ELF_ET_DYN_BASE (mmap_is_ia32() ? 0x000400000UL : \ + 0x100000000UL) /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. This could be done in user space, diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 29ef427c0652..f44e93d2650d 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -905,17 +905,60 @@ static int load_elf_binary(struct linux_binprm *bprm) elf_flags = MAP_PRIVATE | MAP_DENYWRITE | MAP_EXECUTABLE; vaddr = elf_ppnt->p_vaddr; + /* + * If we are loading ET_EXEC or we have already performed + * the ET_DYN load_addr calculations, proceed normally. + */ if (loc->elf_ex.e_type == ET_EXEC || load_addr_set) { elf_flags |= MAP_FIXED; } else if (loc->elf_ex.e_type == ET_DYN) { - /* Try and get dynamic programs out of the way of the - * default mmap base, as well as whatever program they - * might try to exec. This is because the brk will - * follow the loader, and is not movable. */ - load_bias = ELF_ET_DYN_BASE - vaddr; - if (current->flags & PF_RANDOMIZE) - load_bias += arch_mmap_rnd(); - load_bias = ELF_PAGESTART(load_bias); + /* + * This logic is run once for the first LOAD Program + * Header for ET_DYN binaries to calculate the + * randomization (load_bias) for all the LOAD + * Program Headers, and to calculate the entire + * size of the ELF mapping (total_size). (Note that + * load_addr_set is set to true later once the + * initial mapping is performed.) + * + * There are effectively two types of ET_DYN + * binaries: programs (i.e. PIE: ET_DYN with INTERP) + * and loaders (ET_DYN without INTERP, since they + * _are_ the ELF interpreter). The loaders must + * be loaded away from programs since the program + * may otherwise collide with the loader (especially + * for ET_EXEC which does not have a randomized + * position). For example to handle invocations of + * "./ld.so someprog" to test out a new version of + * the loader, the subsequent program that the + * loader loads must avoid the loader itself, so + * they cannot share the same load range. Sufficient + * room for the brk must be allocated with the + * loader as well, since brk must be available with + * the loader. + * + * Therefore, programs are loaded offset from + * ELF_ET_DYN_BASE and loaders are loaded into the + * independently randomized mmap region (0 load_bias + * without MAP_FIXED). + */ + if (elf_interpreter) { + load_bias = ELF_ET_DYN_BASE; + if (current->flags & PF_RANDOMIZE) + load_bias += arch_mmap_rnd(); + elf_flags |= MAP_FIXED; + } else + load_bias = 0; + + /* + * Since load_bias is used for all subsequent loading + * calculations, we must lower it by the first vaddr + * so that the remaining calculations based on the + * ELF vaddrs will be correctly offset. The result + * is then page aligned. + */ + load_bias = ELF_PAGESTART(load_bias - vaddr); + total_size = total_mapping_size(elf_phdata, loc->elf_ex.e_phnum); if (!total_size) { From d2471b5e84f32de4e09b58f5436a4ce3ee935e32 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 10 Jul 2017 15:52:40 -0700 Subject: [PATCH 29/58] arm: move ELF_ET_DYN_BASE to 4MB commit 6a9af90a3bcde217a1c053e135f5f43e5d5fafbd upstream. Now that explicitly executed loaders are loaded in the mmap region, we have more freedom to decide where we position PIE binaries in the address space to avoid possible collisions with mmap or stack regions. 4MB is chosen here mainly to have parity with x86, where this is the traditional minimum load location, likely to avoid historically requiring a 4MB page table entry when only a portion of the first 4MB would be used (since the NULL address is avoided). For ARM the position could be 0x8000, the standard ET_EXEC load address, but that is needlessly close to the NULL address, and anyone running PIE on 32-bit ARM will have an MMU, so the tight mapping is not needed. Link: http://lkml.kernel.org/r/1498154792-49952-2-git-send-email-keescook@chromium.org Signed-off-by: Kees Cook Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: James Hogan Cc: Pratyush Anand Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Alexander Viro Cc: Andy Lutomirski Cc: Daniel Micay Cc: Dmitry Safonov Cc: Grzegorz Andrejczuk Cc: Kees Cook Cc: Masahiro Yamada Cc: Qualys Security Advisory Cc: Rik van Riel Cc: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- arch/arm/include/asm/elf.h | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/arch/arm/include/asm/elf.h b/arch/arm/include/asm/elf.h index d2315ffd8f12..f13ae153fb24 100644 --- a/arch/arm/include/asm/elf.h +++ b/arch/arm/include/asm/elf.h @@ -112,12 +112,8 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs); #define CORE_DUMP_USE_REGSET #define ELF_EXEC_PAGESIZE 4096 -/* This is the location that an ET_DYN program is loaded if exec'ed. Typical - use of this is to invoke "./ld.so someprog" to test out a new version of - the loader. We need to make sure that it is out of the way of the program - that it will "exec", and that there is sufficient room for the brk. */ - -#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) +/* This is the base location for PIE (ET_DYN with INTERP) loads. */ +#define ELF_ET_DYN_BASE 0x400000UL /* When the program starts, a1 contains a pointer to a function to be registered with atexit, as per the SVR4 ABI. A value of 0 means we From 43cf90f788aca3fc66d3cf5b03827bafecd2de24 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 10 Jul 2017 15:52:44 -0700 Subject: [PATCH 30/58] arm64: move ELF_ET_DYN_BASE to 4GB / 4MB commit 02445990a96e60a67526510d8b00f7e3d14101c3 upstream. Now that explicitly executed loaders are loaded in the mmap region, we have more freedom to decide where we position PIE binaries in the address space to avoid possible collisions with mmap or stack regions. For 64-bit, align to 4GB to allow runtimes to use the entire 32-bit address space for 32-bit pointers. On 32-bit use 4MB, to match ARM. This could be 0x8000, the standard ET_EXEC load address, but that is needlessly close to the NULL address, and anyone running arm compat PIE will have an MMU, so the tight mapping is not needed. Link: http://lkml.kernel.org/r/1498251600-132458-4-git-send-email-keescook@chromium.org Signed-off-by: Kees Cook Cc: Ard Biesheuvel Cc: Catalin Marinas Cc: Mark Rutland Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- arch/arm64/include/asm/elf.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm64/include/asm/elf.h b/arch/arm64/include/asm/elf.h index 44dd892a4bbe..9e11dbe1cec3 100644 --- a/arch/arm64/include/asm/elf.h +++ b/arch/arm64/include/asm/elf.h @@ -120,12 +120,11 @@ typedef struct user_fpsimd_state elf_fpregset_t; #define ELF_EXEC_PAGESIZE PAGE_SIZE /* - * This is the location that an ET_DYN program is loaded if exec'ed. Typical - * use of this is to invoke "./ld.so someprog" to test out a new version of - * the loader. We need to make sure that it is out of the way of the program - * that it will "exec", and that there is sufficient room for the brk. + * This is the base location for PIE (ET_DYN with INTERP) loads. On + * 64-bit, this is raised to 4GB to leave the entire 32-bit address + * space open for things that want to use the area for 32-bit pointers. */ -#define ELF_ET_DYN_BASE (2 * TASK_SIZE_64 / 3) +#define ELF_ET_DYN_BASE 0x100000000UL /* * When the program starts, a1 contains a pointer to a function to be @@ -165,7 +164,8 @@ extern int arch_setup_additional_pages(struct linux_binprm *bprm, #ifdef CONFIG_COMPAT -#define COMPAT_ELF_ET_DYN_BASE (2 * TASK_SIZE_32 / 3) +/* PIE load location for compat arm. Must match ARM ELF_ET_DYN_BASE. */ +#define COMPAT_ELF_ET_DYN_BASE 0x000400000UL /* AArch32 registers. */ #define COMPAT_ELF_NGREG 18 From 72a333a0468ed9c492c4da6b017aac9456695c7a Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 10 Jul 2017 15:52:47 -0700 Subject: [PATCH 31/58] powerpc: move ELF_ET_DYN_BASE to 4GB / 4MB commit 47ebb09d54856500c5a5e14824781902b3bb738e upstream. Now that explicitly executed loaders are loaded in the mmap region, we have more freedom to decide where we position PIE binaries in the address space to avoid possible collisions with mmap or stack regions. For 64-bit, align to 4GB to allow runtimes to use the entire 32-bit address space for 32-bit pointers. On 32-bit use 4MB, which is the traditional x86 minimum load location, likely to avoid historically requiring a 4MB page table entry when only a portion of the first 4MB would be used (since the NULL address is avoided). Link: http://lkml.kernel.org/r/1498154792-49952-4-git-send-email-keescook@chromium.org Signed-off-by: Kees Cook Tested-by: Michael Ellerman Acked-by: Michael Ellerman Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: James Hogan Cc: Pratyush Anand Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- arch/powerpc/include/asm/elf.h | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/arch/powerpc/include/asm/elf.h b/arch/powerpc/include/asm/elf.h index ee46ffef608e..743ad7a400d6 100644 --- a/arch/powerpc/include/asm/elf.h +++ b/arch/powerpc/include/asm/elf.h @@ -23,12 +23,13 @@ #define CORE_DUMP_USE_REGSET #define ELF_EXEC_PAGESIZE PAGE_SIZE -/* This is the location that an ET_DYN program is loaded if exec'ed. Typical - use of this is to invoke "./ld.so someprog" to test out a new version of - the loader. We need to make sure that it is out of the way of the program - that it will "exec", and that there is sufficient room for the brk. */ - -#define ELF_ET_DYN_BASE 0x20000000 +/* + * This is the base location for PIE (ET_DYN with INTERP) loads. On + * 64-bit, this is raised to 4GB to leave the entire 32-bit address + * space open for things that want to use the area for 32-bit pointers. + */ +#define ELF_ET_DYN_BASE (is_32bit_task() ? 0x000400000UL : \ + 0x100000000UL) #define ELF_CORE_EFLAGS (is_elf2_task() ? 2 : 0) From 7888c0296c87a6c83bf2a0311c8e929f5f11832f Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 10 Jul 2017 15:52:51 -0700 Subject: [PATCH 32/58] s390: reduce ELF_ET_DYN_BASE commit a73dc5370e153ac63718d850bddf0c9aa9d871e6 upstream. Now that explicitly executed loaders are loaded in the mmap region, we have more freedom to decide where we position PIE binaries in the address space to avoid possible collisions with mmap or stack regions. For 64-bit, align to 4GB to allow runtimes to use the entire 32-bit address space for 32-bit pointers. On 32-bit use 4MB, which is the traditional x86 minimum load location, likely to avoid historically requiring a 4MB page table entry when only a portion of the first 4MB would be used (since the NULL address is avoided). For s390 the position could be 0x10000, but that is needlessly close to the NULL address. Link: http://lkml.kernel.org/r/1498154792-49952-5-git-send-email-keescook@chromium.org Signed-off-by: Kees Cook Cc: Russell King Cc: Catalin Marinas Cc: Will Deacon Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Michael Ellerman Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: James Hogan Cc: Pratyush Anand Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- arch/s390/include/asm/elf.h | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/arch/s390/include/asm/elf.h b/arch/s390/include/asm/elf.h index bab6739a1154..b9eb7b1a49d2 100644 --- a/arch/s390/include/asm/elf.h +++ b/arch/s390/include/asm/elf.h @@ -154,14 +154,13 @@ extern unsigned int vdso_enabled; #define CORE_DUMP_USE_REGSET #define ELF_EXEC_PAGESIZE 4096 -/* This is the location that an ET_DYN program is loaded if exec'ed. Typical - use of this is to invoke "./ld.so someprog" to test out a new version of - the loader. We need to make sure that it is out of the way of the program - that it will "exec", and that there is sufficient room for the brk. 64-bit - tasks are aligned to 4GB. */ -#define ELF_ET_DYN_BASE (is_32bit_task() ? \ - (STACK_TOP / 3 * 2) : \ - (STACK_TOP / 3 * 2) & ~((1UL << 32) - 1)) +/* + * This is the base location for PIE (ET_DYN with INTERP) loads. On + * 64-bit, this is raised to 4GB to leave the entire 32-bit address + * space open for things that want to use the area for 32-bit pointers. + */ +#define ELF_ET_DYN_BASE (is_compat_task() ? 0x000400000UL : \ + 0x100000000UL) /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. */ From 86949eb96466ec182303afc3c386bd70cc67b991 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 7 Jul 2017 11:57:29 -0700 Subject: [PATCH 33/58] exec: Limit arg stack to at most 75% of _STK_LIM commit da029c11e6b12f321f36dac8771e833b65cec962 upstream. To avoid pathological stack usage or the need to special-case setuid execs, just limit all arg stack usage to at most 75% of _STK_LIM (6MB). Signed-off-by: Kees Cook Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- fs/exec.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/fs/exec.c b/fs/exec.c index 02153068a694..9c5ee2a880aa 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -206,8 +206,7 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, if (write) { unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; - unsigned long ptr_size; - struct rlimit *rlim; + unsigned long ptr_size, limit; /* * Since the stack will hold pointers to the strings, we @@ -236,14 +235,16 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, return page; /* - * Limit to 1/4-th the stack size for the argv+env strings. + * Limit to 1/4 of the max stack size or 3/4 of _STK_LIM + * (whichever is smaller) for the argv+env strings. * This ensures that: * - the remaining binfmt code will not run out of stack space, * - the program will have a reasonable amount of stack left * to work from. */ - rlim = current->signal->rlim; - if (size > READ_ONCE(rlim[RLIMIT_STACK].rlim_cur) / 4) + limit = _STK_LIM / 4 * 3; + limit = min(limit, rlimit(RLIMIT_STACK) / 4); + if (size > limit) goto fail; } From 050b074e228523d239e47576606be3c0de5dfa00 Mon Sep 17 00:00:00 2001 From: Adam Borowski Date: Sat, 3 Jun 2017 09:35:06 +0200 Subject: [PATCH 34/58] vt: fix unchecked __put_user() in tioclinux ioctls commit 6987dc8a70976561d22450b5858fc9767788cc1c upstream. Only read access is checked before this call. Actually, at the moment this is not an issue, as every in-tree arch does the same manual checks for VERIFY_READ vs VERIFY_WRITE, relying on the MMU to tell them apart, but this wasn't the case in the past and may happen again on some odd arch in the future. If anyone cares about 3.7 and earlier, this is a security hole (untested) on real 80386 CPUs. Signed-off-by: Adam Borowski Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/vt.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index 5ab54ef4f304..e4f69bddcfb1 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -2708,13 +2708,13 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) * related to the kernel should not use this. */ data = vt_get_shift_state(); - ret = __put_user(data, p); + ret = put_user(data, p); break; case TIOCL_GETMOUSEREPORTING: console_lock(); /* May be overkill */ data = mouse_reporting(); console_unlock(); - ret = __put_user(data, p); + ret = put_user(data, p); break; case TIOCL_SETVESABLANK: console_lock(); @@ -2723,7 +2723,7 @@ int tioclinux(struct tty_struct *tty, unsigned long arg) break; case TIOCL_GETKMSGREDIRECT: data = vt_get_kmsg_redirect(); - ret = __put_user(data, p); + ret = put_user(data, p); break; case TIOCL_SETKMSGREDIRECT: if (!capable(CAP_SYS_ADMIN)) { From 7cbc3955ef3b5bda9336654f1b10a803202e7ed0 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 15 May 2017 14:42:07 -0500 Subject: [PATCH 35/58] mnt: In umount propagation reparent in a separate pass commit 570487d3faf2a1d8a220e6ee10f472163123d7da upstream. It was observed that in some pathlogical cases that the current code does not unmount everything it should. After investigation it was determined that the issue is that mnt_change_mntpoint can can change which mounts are available to be unmounted during mount propagation which is wrong. The trivial reproducer is: $ cat ./pathological.sh mount -t tmpfs test-base /mnt cd /mnt mkdir 1 2 1/1 mount --bind 1 1 mount --make-shared 1 mount --bind 1 2 mount --bind 1/1 1/1 mount --bind 1/1 1/1 echo grep test-base /proc/self/mountinfo umount 1/1 echo grep test-base /proc/self/mountinfo $ unshare -Urm ./pathological.sh The expected output looks like: 46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000 47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 49 54 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 50 53 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 51 49 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 54 47 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 53 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 52 50 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000 47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 The output without the fix looks like: 46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000 47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 49 54 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 50 53 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 51 49 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 54 47 0:25 /1/1 /mnt/1/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 53 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 52 50 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 46 31 0:25 / /mnt rw,relatime - tmpfs test-base rw,uid=1000,gid=1000 47 46 0:25 /1 /mnt/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 48 46 0:25 /1 /mnt/2 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 52 48 0:25 /1/1 /mnt/2/1 rw,relatime shared:1 - tmpfs test-base rw,uid=1000,gid=1000 That last mount in the output was in the propgation tree to be unmounted but was missed because the mnt_change_mountpoint changed it's parent before the walk through the mount propagation tree observed it. Fixes: 1064f874abc0 ("mnt: Tuck mounts under others instead of creating shadow/side mounts.") Acked-by: Andrei Vagin Reviewed-by: Ram Pai Signed-off-by: "Eric W. Biederman" Signed-off-by: Greg Kroah-Hartman --- fs/mount.h | 1 + fs/namespace.c | 1 + fs/pnode.c | 35 ++++++++++++++++++++++++++++++----- 3 files changed, 32 insertions(+), 5 deletions(-) diff --git a/fs/mount.h b/fs/mount.h index 13a4ebbbaa74..2352231d4f0f 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -57,6 +57,7 @@ struct mount { struct mnt_namespace *mnt_ns; /* containing namespace */ struct mountpoint *mnt_mp; /* where is it mounted */ struct hlist_node mnt_mp_list; /* list mounts with the same mountpoint */ + struct list_head mnt_reparent; /* reparent list entry */ #ifdef CONFIG_FSNOTIFY struct hlist_head mnt_fsnotify_marks; __u32 mnt_fsnotify_mask; diff --git a/fs/namespace.c b/fs/namespace.c index f26d18d69712..4c7174d2041e 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -237,6 +237,7 @@ static struct mount *alloc_vfsmnt(const char *name) INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); INIT_HLIST_NODE(&mnt->mnt_mp_list); + INIT_LIST_HEAD(&mnt->mnt_reparent); #ifdef CONFIG_FSNOTIFY INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks); #endif diff --git a/fs/pnode.c b/fs/pnode.c index b394ca5307ec..948e85ad0374 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -441,7 +441,7 @@ static void mark_umount_candidates(struct mount *mnt) * NOTE: unmounting 'mnt' naturally propagates to all other mounts its * parent propagates to. */ -static void __propagate_umount(struct mount *mnt) +static void __propagate_umount(struct mount *mnt, struct list_head *to_reparent) { struct mount *parent = mnt->mnt_parent; struct mount *m; @@ -466,17 +466,38 @@ static void __propagate_umount(struct mount *mnt) */ topper = find_topper(child); if (topper) - mnt_change_mountpoint(child->mnt_parent, child->mnt_mp, - topper); + list_add_tail(&topper->mnt_reparent, to_reparent); - if (list_empty(&child->mnt_mounts)) { + if (topper || list_empty(&child->mnt_mounts)) { list_del_init(&child->mnt_child); + list_del_init(&child->mnt_reparent); child->mnt.mnt_flags |= MNT_UMOUNT; list_move_tail(&child->mnt_list, &mnt->mnt_list); } } } +static void reparent_mounts(struct list_head *to_reparent) +{ + while (!list_empty(to_reparent)) { + struct mount *mnt, *parent; + struct mountpoint *mp; + + mnt = list_first_entry(to_reparent, struct mount, mnt_reparent); + list_del_init(&mnt->mnt_reparent); + + /* Where should this mount be reparented to? */ + mp = mnt->mnt_mp; + parent = mnt->mnt_parent; + while (parent->mnt.mnt_flags & MNT_UMOUNT) { + mp = parent->mnt_mp; + parent = parent->mnt_parent; + } + + mnt_change_mountpoint(parent, mp, mnt); + } +} + /* * collect all mounts that receive propagation from the mount in @list, * and return these additional mounts in the same list. @@ -487,11 +508,15 @@ static void __propagate_umount(struct mount *mnt) int propagate_umount(struct list_head *list) { struct mount *mnt; + LIST_HEAD(to_reparent); list_for_each_entry_reverse(mnt, list, mnt_list) mark_umount_candidates(mnt); list_for_each_entry(mnt, list, mnt_list) - __propagate_umount(mnt); + __propagate_umount(mnt, &to_reparent); + + reparent_mounts(&to_reparent); + return 0; } From fdb8f10499924d96b8eb60cc1354b278f6fcf0e9 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 24 Oct 2016 16:16:13 -0500 Subject: [PATCH 36/58] mnt: In propgate_umount handle visiting mounts in any order commit 99b19d16471e9c3faa85cad38abc9cbbe04c6d55 upstream. While investigating some poor umount performance I realized that in the case of overlapping mount trees where some of the mounts are locked the code has been failing to unmount all of the mounts it should have been unmounting. This failure to unmount all of the necessary mounts can be reproduced with: $ cat locked_mounts_test.sh mount -t tmpfs test-base /mnt mount --make-shared /mnt mkdir -p /mnt/b mount -t tmpfs test1 /mnt/b mount --make-shared /mnt/b mkdir -p /mnt/b/10 mount -t tmpfs test2 /mnt/b/10 mount --make-shared /mnt/b/10 mkdir -p /mnt/b/10/20 mount --rbind /mnt/b /mnt/b/10/20 unshare -Urm --propagation unchaged /bin/sh -c 'sleep 5; if [ $(grep test /proc/self/mountinfo | wc -l) -eq 1 ] ; then echo SUCCESS ; else echo FAILURE ; fi' sleep 1 umount -l /mnt/b wait %% $ unshare -Urm ./locked_mounts_test.sh This failure is corrected by removing the prepass that marks mounts that may be umounted. A first pass is added that umounts mounts if possible and if not sets mount mark if they could be unmounted if they weren't locked and adds them to a list to umount possibilities. This first pass reconsiders the mounts parent if it is on the list of umount possibilities, ensuring that information of umoutability will pass from child to mount parent. A second pass then walks through all mounts that are umounted and processes their children unmounting them or marking them for reparenting. A last pass cleans up the state on the mounts that could not be umounted and if applicable reparents them to their first parent that remained mounted. While a bit longer than the old code this code is much more robust as it allows information to flow up from the leaves and down from the trunk making the order in which mounts are encountered in the umount propgation tree irrelevant. Fixes: 0c56fe31420c ("mnt: Don't propagate unmounts to locked mounts") Reviewed-by: Andrei Vagin Signed-off-by: "Eric W. Biederman" Signed-off-by: Greg Kroah-Hartman --- fs/mount.h | 2 +- fs/namespace.c | 2 +- fs/pnode.c | 144 +++++++++++++++++++++++++++++-------------------- 3 files changed, 88 insertions(+), 60 deletions(-) diff --git a/fs/mount.h b/fs/mount.h index 2352231d4f0f..37c64bbe840c 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -57,7 +57,7 @@ struct mount { struct mnt_namespace *mnt_ns; /* containing namespace */ struct mountpoint *mnt_mp; /* where is it mounted */ struct hlist_node mnt_mp_list; /* list mounts with the same mountpoint */ - struct list_head mnt_reparent; /* reparent list entry */ + struct list_head mnt_umounting; /* list entry for umount propagation */ #ifdef CONFIG_FSNOTIFY struct hlist_head mnt_fsnotify_marks; __u32 mnt_fsnotify_mask; diff --git a/fs/namespace.c b/fs/namespace.c index 4c7174d2041e..ec4078d16eb7 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -237,7 +237,7 @@ static struct mount *alloc_vfsmnt(const char *name) INIT_LIST_HEAD(&mnt->mnt_slave_list); INIT_LIST_HEAD(&mnt->mnt_slave); INIT_HLIST_NODE(&mnt->mnt_mp_list); - INIT_LIST_HEAD(&mnt->mnt_reparent); + INIT_LIST_HEAD(&mnt->mnt_umounting); #ifdef CONFIG_FSNOTIFY INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks); #endif diff --git a/fs/pnode.c b/fs/pnode.c index 948e85ad0374..acc2eefbc4ff 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -415,86 +415,95 @@ void propagate_mount_unlock(struct mount *mnt) } } -/* - * Mark all mounts that the MNT_LOCKED logic will allow to be unmounted. - */ -static void mark_umount_candidates(struct mount *mnt) +static void umount_one(struct mount *mnt, struct list_head *to_umount) { - struct mount *parent = mnt->mnt_parent; - struct mount *m; - - BUG_ON(parent == mnt); - - for (m = propagation_next(parent, parent); m; - m = propagation_next(m, parent)) { - struct mount *child = __lookup_mnt(&m->mnt, - mnt->mnt_mountpoint); - if (!child || (child->mnt.mnt_flags & MNT_UMOUNT)) - continue; - if (!IS_MNT_LOCKED(child) || IS_MNT_MARKED(m)) { - SET_MNT_MARK(child); - } - } + CLEAR_MNT_MARK(mnt); + mnt->mnt.mnt_flags |= MNT_UMOUNT; + list_del_init(&mnt->mnt_child); + list_del_init(&mnt->mnt_umounting); + list_move_tail(&mnt->mnt_list, to_umount); } /* * NOTE: unmounting 'mnt' naturally propagates to all other mounts its * parent propagates to. */ -static void __propagate_umount(struct mount *mnt, struct list_head *to_reparent) +static bool __propagate_umount(struct mount *mnt, + struct list_head *to_umount, + struct list_head *to_restore) { - struct mount *parent = mnt->mnt_parent; - struct mount *m; + bool progress = false; + struct mount *child; - BUG_ON(parent == mnt); + /* + * The state of the parent won't change if this mount is + * already unmounted or marked as without children. + */ + if (mnt->mnt.mnt_flags & (MNT_UMOUNT | MNT_MARKED)) + goto out; - for (m = propagation_next(parent, parent); m; - m = propagation_next(m, parent)) { - struct mount *topper; - struct mount *child = __lookup_mnt(&m->mnt, - mnt->mnt_mountpoint); - /* - * umount the child only if the child has no children - * and the child is marked safe to unmount. - */ - if (!child || !IS_MNT_MARKED(child)) + /* Verify topper is the only grandchild that has not been + * speculatively unmounted. + */ + list_for_each_entry(child, &mnt->mnt_mounts, mnt_child) { + if (child->mnt_mountpoint == mnt->mnt.mnt_root) continue; - CLEAR_MNT_MARK(child); + if (!list_empty(&child->mnt_umounting) && IS_MNT_MARKED(child)) + continue; + /* Found a mounted child */ + goto children; + } - /* If there is exactly one mount covering all of child - * replace child with that mount. - */ - topper = find_topper(child); - if (topper) - list_add_tail(&topper->mnt_reparent, to_reparent); + /* Mark mounts that can be unmounted if not locked */ + SET_MNT_MARK(mnt); + progress = true; - if (topper || list_empty(&child->mnt_mounts)) { - list_del_init(&child->mnt_child); - list_del_init(&child->mnt_reparent); - child->mnt.mnt_flags |= MNT_UMOUNT; - list_move_tail(&child->mnt_list, &mnt->mnt_list); + /* If a mount is without children and not locked umount it. */ + if (!IS_MNT_LOCKED(mnt)) { + umount_one(mnt, to_umount); + } else { +children: + list_move_tail(&mnt->mnt_umounting, to_restore); + } +out: + return progress; +} + +static void umount_list(struct list_head *to_umount, + struct list_head *to_restore) +{ + struct mount *mnt, *child, *tmp; + list_for_each_entry(mnt, to_umount, mnt_list) { + list_for_each_entry_safe(child, tmp, &mnt->mnt_mounts, mnt_child) { + /* topper? */ + if (child->mnt_mountpoint == mnt->mnt.mnt_root) + list_move_tail(&child->mnt_umounting, to_restore); + else + umount_one(child, to_umount); } } } -static void reparent_mounts(struct list_head *to_reparent) +static void restore_mounts(struct list_head *to_restore) { - while (!list_empty(to_reparent)) { + /* Restore mounts to a clean working state */ + while (!list_empty(to_restore)) { struct mount *mnt, *parent; struct mountpoint *mp; - mnt = list_first_entry(to_reparent, struct mount, mnt_reparent); - list_del_init(&mnt->mnt_reparent); + mnt = list_first_entry(to_restore, struct mount, mnt_umounting); + CLEAR_MNT_MARK(mnt); + list_del_init(&mnt->mnt_umounting); - /* Where should this mount be reparented to? */ + /* Should this mount be reparented? */ mp = mnt->mnt_mp; parent = mnt->mnt_parent; while (parent->mnt.mnt_flags & MNT_UMOUNT) { mp = parent->mnt_mp; parent = parent->mnt_parent; } - - mnt_change_mountpoint(parent, mp, mnt); + if (parent != mnt->mnt_parent) + mnt_change_mountpoint(parent, mp, mnt); } } @@ -508,15 +517,34 @@ static void reparent_mounts(struct list_head *to_reparent) int propagate_umount(struct list_head *list) { struct mount *mnt; - LIST_HEAD(to_reparent); + LIST_HEAD(to_restore); + LIST_HEAD(to_umount); - list_for_each_entry_reverse(mnt, list, mnt_list) - mark_umount_candidates(mnt); + list_for_each_entry(mnt, list, mnt_list) { + struct mount *parent = mnt->mnt_parent; + struct mount *m; - list_for_each_entry(mnt, list, mnt_list) - __propagate_umount(mnt, &to_reparent); + for (m = propagation_next(parent, parent); m; + m = propagation_next(m, parent)) { + struct mount *child = __lookup_mnt(&m->mnt, + mnt->mnt_mountpoint); + if (!child) + continue; - reparent_mounts(&to_reparent); + /* Check the child and parents while progress is made */ + while (__propagate_umount(child, + &to_umount, &to_restore)) { + /* Is the parent a umount candidate? */ + child = child->mnt_parent; + if (list_empty(&child->mnt_umounting)) + break; + } + } + } + + umount_list(&to_umount, &to_restore); + restore_mounts(&to_restore); + list_splice_tail(&to_umount, list); return 0; } From f07288cfb0f7091566fb8750672157eb10be53f4 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Mon, 24 Oct 2016 17:25:19 -0500 Subject: [PATCH 37/58] mnt: Make propagate_umount less slow for overlapping mount propagation trees commit 296990deb389c7da21c78030376ba244dc1badf5 upstream. Andrei Vagin pointed out that time to executue propagate_umount can go non-linear (and take a ludicrious amount of time) when the mount propogation trees of the mounts to be unmunted by a lazy unmount overlap. Make the walk of the mount propagation trees nearly linear by remembering which mounts have already been visited, allowing subsequent walks to detect when walking a mount propgation tree or a subtree of a mount propgation tree would be duplicate work and to skip them entirely. Walk the list of mounts whose propgatation trees need to be traversed from the mount highest in the mount tree to mounts lower in the mount tree so that odds are higher that the code will walk the largest trees first, allowing later tree walks to be skipped entirely. Add cleanup_umount_visitation to remover the code's memory of which mounts have been visited. Add the functions last_slave and skip_propagation_subtree to allow skipping appropriate parts of the mount propagation tree without needing to change the logic of the rest of the code. A script to generate overlapping mount propagation trees: $ cat runs.h set -e mount -t tmpfs zdtm /mnt mkdir -p /mnt/1 /mnt/2 mount -t tmpfs zdtm /mnt/1 mount --make-shared /mnt/1 mkdir /mnt/1/1 iteration=10 if [ -n "$1" ] ; then iteration=$1 fi for i in $(seq $iteration); do mount --bind /mnt/1/1 /mnt/1/1 done mount --rbind /mnt/1 /mnt/2 TIMEFORMAT='%Rs' nr=$(( ( 2 ** ( $iteration + 1 ) ) + 1 )) echo -n "umount -l /mnt/1 -> $nr " time umount -l /mnt/1 nr=$(cat /proc/self/mountinfo | grep zdtm | wc -l ) time umount -l /mnt/2 $ for i in $(seq 9 19); do echo $i; unshare -Urm bash ./run.sh $i; done Here are the performance numbers with and without the patch: mhash | 8192 | 8192 | 1048576 | 1048576 mounts | before | after | before | after ------------------------------------------------ 1025 | 0.040s | 0.016s | 0.038s | 0.019s 2049 | 0.094s | 0.017s | 0.080s | 0.018s 4097 | 0.243s | 0.019s | 0.206s | 0.023s 8193 | 1.202s | 0.028s | 1.562s | 0.032s 16385 | 9.635s | 0.036s | 9.952s | 0.041s 32769 | 60.928s | 0.063s | 44.321s | 0.064s 65537 | | 0.097s | | 0.097s 131073 | | 0.233s | | 0.176s 262145 | | 0.653s | | 0.344s 524289 | | 2.305s | | 0.735s 1048577 | | 7.107s | | 2.603s Andrei Vagin reports fixing the performance problem is part of the work to fix CVE-2016-6213. Fixes: a05964f3917c ("[PATCH] shared mounts handling: umount") Reported-by: Andrei Vagin Reviewed-by: Andrei Vagin Signed-off-by: "Eric W. Biederman" Signed-off-by: Greg Kroah-Hartman --- fs/pnode.c | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 62 insertions(+), 1 deletion(-) diff --git a/fs/pnode.c b/fs/pnode.c index acc2eefbc4ff..d15c63e97ef1 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -24,6 +24,11 @@ static inline struct mount *first_slave(struct mount *p) return list_entry(p->mnt_slave_list.next, struct mount, mnt_slave); } +static inline struct mount *last_slave(struct mount *p) +{ + return list_entry(p->mnt_slave_list.prev, struct mount, mnt_slave); +} + static inline struct mount *next_slave(struct mount *p) { return list_entry(p->mnt_slave.next, struct mount, mnt_slave); @@ -164,6 +169,19 @@ static struct mount *propagation_next(struct mount *m, } } +static struct mount *skip_propagation_subtree(struct mount *m, + struct mount *origin) +{ + /* + * Advance m such that propagation_next will not return + * the slaves of m. + */ + if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) + m = last_slave(m); + + return m; +} + static struct mount *next_group(struct mount *m, struct mount *origin) { while (1) { @@ -507,6 +525,15 @@ static void restore_mounts(struct list_head *to_restore) } } +static void cleanup_umount_visitations(struct list_head *visited) +{ + while (!list_empty(visited)) { + struct mount *mnt = + list_first_entry(visited, struct mount, mnt_umounting); + list_del_init(&mnt->mnt_umounting); + } +} + /* * collect all mounts that receive propagation from the mount in @list, * and return these additional mounts in the same list. @@ -519,11 +546,23 @@ int propagate_umount(struct list_head *list) struct mount *mnt; LIST_HEAD(to_restore); LIST_HEAD(to_umount); + LIST_HEAD(visited); - list_for_each_entry(mnt, list, mnt_list) { + /* Find candidates for unmounting */ + list_for_each_entry_reverse(mnt, list, mnt_list) { struct mount *parent = mnt->mnt_parent; struct mount *m; + /* + * If this mount has already been visited it is known that it's + * entire peer group and all of their slaves in the propagation + * tree for the mountpoint has already been visited and there is + * no need to visit them again. + */ + if (!list_empty(&mnt->mnt_umounting)) + continue; + + list_add_tail(&mnt->mnt_umounting, &visited); for (m = propagation_next(parent, parent); m; m = propagation_next(m, parent)) { struct mount *child = __lookup_mnt(&m->mnt, @@ -531,6 +570,27 @@ int propagate_umount(struct list_head *list) if (!child) continue; + if (!list_empty(&child->mnt_umounting)) { + /* + * If the child has already been visited it is + * know that it's entire peer group and all of + * their slaves in the propgation tree for the + * mountpoint has already been visited and there + * is no need to visit this subtree again. + */ + m = skip_propagation_subtree(m, parent); + continue; + } else if (child->mnt.mnt_flags & MNT_UMOUNT) { + /* + * We have come accross an partially unmounted + * mount in list that has not been visited yet. + * Remember it has been visited and continue + * about our merry way. + */ + list_add_tail(&child->mnt_umounting, &visited); + continue; + } + /* Check the child and parents while progress is made */ while (__propagate_umount(child, &to_umount, &to_restore)) { @@ -544,6 +604,7 @@ int propagate_umount(struct list_head *list) umount_list(&to_umount, &to_restore); restore_mounts(&to_restore); + cleanup_umount_visitations(&visited); list_splice_tail(&to_umount, list); return 0; From a2e0b1c18c0904b4f1a2de620ad282b5d01710bb Mon Sep 17 00:00:00 2001 From: Andy Lutomirski Date: Thu, 29 Jun 2017 08:46:12 -0700 Subject: [PATCH 38/58] selftests/capabilities: Fix the test_execve test commit 796a3bae2fba6810427efdb314a1c126c9490fb3 upstream. test_execve does rather odd mount manipulations to safely create temporary setuid and setgid executables that aren't visible to the rest of the system. Those executables end up in the test's cwd, but that cwd is MNT_DETACHed. The core namespace code considers MNT_DETACHed trees to belong to no mount namespace at all and, in general, MNT_DETACHed trees are only barely function. This interacted with commit 380cf5ba6b0a ("fs: Treat foreign mounts as nosuid") to cause all MNT_DETACHed trees to act as though they're nosuid, breaking the test. Fix it by just not detaching the tree. It's still in a private mount namespace and is therefore still invisible to the rest of the system (except via /proc, and the same nosuid logic will protect all other programs on the system from believing in test_execve's setuid bits). While we're at it, fix some blatant whitespace problems. Reported-by: Naresh Kamboju Fixes: 380cf5ba6b0a ("fs: Treat foreign mounts as nosuid") Cc: "Eric W. Biederman" Cc: Kees Cook Cc: Shuah Khan Cc: Greg KH Cc: linux-kselftest@vger.kernel.org Signed-off-by: Andy Lutomirski Acked-by: Greg Kroah-Hartman Signed-off-by: Shuah Khan Signed-off-by: Greg Kroah-Hartman --- tools/testing/selftests/capabilities/test_execve.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/tools/testing/selftests/capabilities/test_execve.c b/tools/testing/selftests/capabilities/test_execve.c index 10a21a958aaf..763f37fecfb8 100644 --- a/tools/testing/selftests/capabilities/test_execve.c +++ b/tools/testing/selftests/capabilities/test_execve.c @@ -138,9 +138,6 @@ static void chdir_to_tmpfs(void) if (chdir(cwd) != 0) err(1, "chdir to private tmpfs"); - - if (umount2(".", MNT_DETACH) != 0) - err(1, "detach private tmpfs"); } static void copy_fromat_to(int fromfd, const char *fromname, const char *toname) @@ -248,7 +245,7 @@ static int do_tests(int uid, const char *our_path) err(1, "chown"); if (chmod("validate_cap_sgidnonroot", S_ISGID | 0710) != 0) err(1, "chmod"); -} + } capng_get_caps_process(); @@ -384,7 +381,7 @@ static int do_tests(int uid, const char *our_path) } else { printf("[RUN]\tNon-root +ia, sgidnonroot => i\n"); exec_other_validate_cap("./validate_cap_sgidnonroot", - false, false, true, false); + false, false, true, false); if (fork_wait()) { printf("[RUN]\tNon-root +ia, sgidroot => i\n"); From 5e07de5b5a4ea5c5b56ff73f359a248d97926629 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 29 Feb 2016 12:29:47 -0500 Subject: [PATCH 39/58] tpm: Get rid of chip->pdev commit 8cfffc9d4d3786d3b496a021d7224e06328bac7d upstream. This is a hold over from before the struct device conversion. - All prints should be using &chip->dev, which is the Linux standard. This changes prints to use tpm0 as the device name, not the PnP/etc ID. - The few places involving sysfs/modules that really do need the parent just use chip->dev.parent instead - We no longer need to get_device(pdev) in any places since it is no longer used by any of the code. The kref on the parent is held by the device core during device_add and dropped in device_del Signed-off-by: Jason Gunthorpe Signed-off-by: Stefan Berger Tested-by: Stefan Berger Reviewed-by: Jarkko Sakkinen Tested-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen Signed-off-by: Greg Kroah-Hartman --- drivers/char/tpm/tpm-chip.c | 15 ++++++--------- drivers/char/tpm/tpm-dev.c | 4 +--- drivers/char/tpm/tpm-interface.c | 30 +++++++++++++++-------------- drivers/char/tpm/tpm-sysfs.c | 6 +++--- drivers/char/tpm/tpm.h | 3 +-- drivers/char/tpm/tpm2-cmd.c | 8 ++++---- drivers/char/tpm/tpm_atmel.c | 14 +++++++------- drivers/char/tpm/tpm_i2c_atmel.c | 16 +++++++-------- drivers/char/tpm/tpm_i2c_infineon.c | 6 +++--- drivers/char/tpm/tpm_i2c_nuvoton.c | 22 ++++++++++----------- drivers/char/tpm/tpm_infineon.c | 22 ++++++++++----------- drivers/char/tpm/tpm_nsc.c | 20 +++++++++---------- drivers/char/tpm/tpm_tis.c | 16 +++++++-------- 13 files changed, 89 insertions(+), 93 deletions(-) diff --git a/drivers/char/tpm/tpm-chip.c b/drivers/char/tpm/tpm-chip.c index 252142524ff2..f55b4921c723 100644 --- a/drivers/char/tpm/tpm-chip.c +++ b/drivers/char/tpm/tpm-chip.c @@ -49,7 +49,7 @@ struct tpm_chip *tpm_chip_find_get(int chip_num) if (chip_num != TPM_ANY_NUM && chip_num != pos->dev_num) continue; - if (try_module_get(pos->pdev->driver->owner)) { + if (try_module_get(pos->dev.parent->driver->owner)) { chip = pos; break; } @@ -112,13 +112,11 @@ struct tpm_chip *tpmm_chip_alloc(struct device *dev, scnprintf(chip->devname, sizeof(chip->devname), "tpm%d", chip->dev_num); - chip->pdev = dev; - dev_set_drvdata(dev, chip); chip->dev.class = tpm_class; chip->dev.release = tpm_dev_release; - chip->dev.parent = chip->pdev; + chip->dev.parent = dev; #ifdef CONFIG_ACPI chip->dev.groups = chip->groups; #endif @@ -133,7 +131,7 @@ struct tpm_chip *tpmm_chip_alloc(struct device *dev, device_initialize(&chip->dev); cdev_init(&chip->cdev, &tpm_fops); - chip->cdev.owner = chip->pdev->driver->owner; + chip->cdev.owner = dev->driver->owner; chip->cdev.kobj.parent = &chip->dev.kobj; devm_add_action(dev, (void (*)(void *)) put_device, &chip->dev); @@ -236,9 +234,8 @@ int tpm_chip_register(struct tpm_chip *chip) chip->flags |= TPM_CHIP_FLAG_REGISTERED; if (!(chip->flags & TPM_CHIP_FLAG_TPM2)) { - rc = __compat_only_sysfs_link_entry_to_kobj(&chip->pdev->kobj, - &chip->dev.kobj, - "ppi"); + rc = __compat_only_sysfs_link_entry_to_kobj( + &chip->dev.parent->kobj, &chip->dev.kobj, "ppi"); if (rc && rc != -ENOENT) { tpm_chip_unregister(chip); return rc; @@ -273,7 +270,7 @@ void tpm_chip_unregister(struct tpm_chip *chip) synchronize_rcu(); if (!(chip->flags & TPM_CHIP_FLAG_TPM2)) - sysfs_remove_link(&chip->pdev->kobj, "ppi"); + sysfs_remove_link(&chip->dev.parent->kobj, "ppi"); tpm1_chip_unregister(chip); tpm_del_char_device(chip); diff --git a/drivers/char/tpm/tpm-dev.c b/drivers/char/tpm/tpm-dev.c index 4f3137d9a35e..6ed0651cbe58 100644 --- a/drivers/char/tpm/tpm-dev.c +++ b/drivers/char/tpm/tpm-dev.c @@ -61,7 +61,7 @@ static int tpm_open(struct inode *inode, struct file *file) * by the check of is_open variable, which is protected * by driver_lock. */ if (test_and_set_bit(0, &chip->is_open)) { - dev_dbg(chip->pdev, "Another process owns this TPM\n"); + dev_dbg(&chip->dev, "Another process owns this TPM\n"); return -EBUSY; } @@ -79,7 +79,6 @@ static int tpm_open(struct inode *inode, struct file *file) INIT_WORK(&priv->work, timeout_work); file->private_data = priv; - get_device(chip->pdev); return 0; } @@ -166,7 +165,6 @@ static int tpm_release(struct inode *inode, struct file *file) file->private_data = NULL; atomic_set(&priv->data_pending, 0); clear_bit(0, &priv->chip->is_open); - put_device(priv->chip->pdev); kfree(priv); return 0; } diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index 17abe52e6365..4391953a7711 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -343,7 +343,7 @@ ssize_t tpm_transmit(struct tpm_chip *chip, const u8 *buf, size_t bufsiz, if (count == 0) return -ENODATA; if (count > bufsiz) { - dev_err(chip->pdev, + dev_err(&chip->dev, "invalid count value %x %zx\n", count, bufsiz); return -E2BIG; } @@ -353,7 +353,7 @@ ssize_t tpm_transmit(struct tpm_chip *chip, const u8 *buf, size_t bufsiz, rc = chip->ops->send(chip, (u8 *) buf, count); if (rc < 0) { - dev_err(chip->pdev, + dev_err(&chip->dev, "tpm_transmit: tpm_send: error %zd\n", rc); goto out; } @@ -372,7 +372,7 @@ ssize_t tpm_transmit(struct tpm_chip *chip, const u8 *buf, size_t bufsiz, goto out_recv; if (chip->ops->req_canceled(chip, status)) { - dev_err(chip->pdev, "Operation Canceled\n"); + dev_err(&chip->dev, "Operation Canceled\n"); rc = -ECANCELED; goto out; } @@ -382,14 +382,14 @@ ssize_t tpm_transmit(struct tpm_chip *chip, const u8 *buf, size_t bufsiz, } while (time_before(jiffies, stop)); chip->ops->cancel(chip); - dev_err(chip->pdev, "Operation Timed out\n"); + dev_err(&chip->dev, "Operation Timed out\n"); rc = -ETIME; goto out; out_recv: rc = chip->ops->recv(chip, (u8 *) buf, bufsiz); if (rc < 0) - dev_err(chip->pdev, + dev_err(&chip->dev, "tpm_transmit: tpm_recv: error %zd\n", rc); out: if (!(flags & TPM_TRANSMIT_UNLOCKED)) @@ -416,7 +416,7 @@ ssize_t tpm_transmit_cmd(struct tpm_chip *chip, const void *cmd, err = be32_to_cpu(header->return_code); if (err != 0 && desc) - dev_err(chip->pdev, "A TPM error (%d) occurred %s\n", err, + dev_err(&chip->dev, "A TPM error (%d) occurred %s\n", err, desc); return err; @@ -514,7 +514,7 @@ int tpm_get_timeouts(struct tpm_chip *chip) if (rc == TPM_ERR_INVALID_POSTINIT) { /* The TPM is not started, we are the first to talk to it. Execute a startup command. */ - dev_info(chip->pdev, "Issuing TPM_STARTUP"); + dev_info(&chip->dev, "Issuing TPM_STARTUP"); if (tpm_startup(chip, TPM_ST_CLEAR)) return rc; @@ -526,7 +526,7 @@ int tpm_get_timeouts(struct tpm_chip *chip) 0, NULL); } if (rc) { - dev_err(chip->pdev, + dev_err(&chip->dev, "A TPM error (%zd) occurred attempting to determine the timeouts\n", rc); goto duration; @@ -565,7 +565,7 @@ int tpm_get_timeouts(struct tpm_chip *chip) /* Report adjusted timeouts */ if (chip->vendor.timeout_adjusted) { - dev_info(chip->pdev, + dev_info(&chip->dev, HW_ERR "Adjusting reported timeouts: A %lu->%luus B %lu->%luus C %lu->%luus D %lu->%luus\n", old_timeout[0], new_timeout[0], old_timeout[1], new_timeout[1], @@ -612,7 +612,7 @@ duration: chip->vendor.duration[TPM_MEDIUM] *= 1000; chip->vendor.duration[TPM_LONG] *= 1000; chip->vendor.duration_adjusted = true; - dev_info(chip->pdev, "Adjusting TPM timeout parameters."); + dev_info(&chip->dev, "Adjusting TPM timeout parameters."); } return 0; } @@ -802,7 +802,9 @@ int tpm_do_selftest(struct tpm_chip *chip) * around 300ms while the self test is ongoing, keep trying * until the self test duration expires. */ if (rc == -ETIME) { - dev_info(chip->pdev, HW_ERR "TPM command timed out during continue self test"); + dev_info( + &chip->dev, HW_ERR + "TPM command timed out during continue self test"); msleep(delay_msec); continue; } @@ -812,7 +814,7 @@ int tpm_do_selftest(struct tpm_chip *chip) rc = be32_to_cpu(cmd.header.out.return_code); if (rc == TPM_ERR_DISABLED || rc == TPM_ERR_DEACTIVATED) { - dev_info(chip->pdev, + dev_info(&chip->dev, "TPM is disabled/deactivated (0x%X)\n", rc); /* TPM is disabled and/or deactivated; driver can * proceed and TPM does handle commands for @@ -966,10 +968,10 @@ int tpm_pm_suspend(struct device *dev) } if (rc) - dev_err(chip->pdev, + dev_err(&chip->dev, "Error (%d) sending savestate before suspend\n", rc); else if (try > 0) - dev_warn(chip->pdev, "TPM savestate took %dms\n", + dev_warn(&chip->dev, "TPM savestate took %dms\n", try * TPM_TIMEOUT_RETRY); return rc; diff --git a/drivers/char/tpm/tpm-sysfs.c b/drivers/char/tpm/tpm-sysfs.c index f880856aa75e..10370c22e98b 100644 --- a/drivers/char/tpm/tpm-sysfs.c +++ b/drivers/char/tpm/tpm-sysfs.c @@ -284,16 +284,16 @@ static const struct attribute_group tpm_dev_group = { int tpm_sysfs_add_device(struct tpm_chip *chip) { int err; - err = sysfs_create_group(&chip->pdev->kobj, + err = sysfs_create_group(&chip->dev.parent->kobj, &tpm_dev_group); if (err) - dev_err(chip->pdev, + dev_err(&chip->dev, "failed to create sysfs attributes, %d\n", err); return err; } void tpm_sysfs_del_device(struct tpm_chip *chip) { - sysfs_remove_group(&chip->pdev->kobj, &tpm_dev_group); + sysfs_remove_group(&chip->dev.parent->kobj, &tpm_dev_group); } diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 2216861f89f1..57c4c26c38ea 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -171,7 +171,6 @@ enum tpm_chip_flags { }; struct tpm_chip { - struct device *pdev; /* Device stuff */ struct device dev; struct cdev cdev; @@ -203,7 +202,7 @@ struct tpm_chip { static inline void tpm_chip_put(struct tpm_chip *chip) { - module_put(chip->pdev->driver->owner); + module_put(chip->dev.parent->driver->owner); } static inline int tpm_read_index(int base, int index) diff --git a/drivers/char/tpm/tpm2-cmd.c b/drivers/char/tpm/tpm2-cmd.c index cb7e4f6b70ba..286bd090a488 100644 --- a/drivers/char/tpm/tpm2-cmd.c +++ b/drivers/char/tpm/tpm2-cmd.c @@ -570,7 +570,7 @@ static void tpm2_flush_context_cmd(struct tpm_chip *chip, u32 handle, rc = tpm_buf_init(&buf, TPM2_ST_NO_SESSIONS, TPM2_CC_FLUSH_CONTEXT); if (rc) { - dev_warn(chip->pdev, "0x%08x was not flushed, out of memory\n", + dev_warn(&chip->dev, "0x%08x was not flushed, out of memory\n", handle); return; } @@ -580,7 +580,7 @@ static void tpm2_flush_context_cmd(struct tpm_chip *chip, u32 handle, rc = tpm_transmit_cmd(chip, buf.data, PAGE_SIZE, flags, "flushing context"); if (rc) - dev_warn(chip->pdev, "0x%08x was not flushed, rc=%d\n", handle, + dev_warn(&chip->dev, "0x%08x was not flushed, rc=%d\n", handle, rc); tpm_buf_destroy(&buf); @@ -753,7 +753,7 @@ void tpm2_shutdown(struct tpm_chip *chip, u16 shutdown_type) * except print the error code on a system failure. */ if (rc < 0) - dev_warn(chip->pdev, "transmit returned %d while stopping the TPM", + dev_warn(&chip->dev, "transmit returned %d while stopping the TPM", rc); } EXPORT_SYMBOL_GPL(tpm2_shutdown); @@ -820,7 +820,7 @@ static int tpm2_start_selftest(struct tpm_chip *chip, bool full) * immediately. This is a workaround for that. */ if (rc == TPM2_RC_TESTING) { - dev_warn(chip->pdev, "Got RC_TESTING, ignoring\n"); + dev_warn(&chip->dev, "Got RC_TESTING, ignoring\n"); rc = 0; } diff --git a/drivers/char/tpm/tpm_atmel.c b/drivers/char/tpm/tpm_atmel.c index dfadad0916a1..a48a878f791d 100644 --- a/drivers/char/tpm/tpm_atmel.c +++ b/drivers/char/tpm/tpm_atmel.c @@ -49,7 +49,7 @@ static int tpm_atml_recv(struct tpm_chip *chip, u8 *buf, size_t count) for (i = 0; i < 6; i++) { status = ioread8(chip->vendor.iobase + 1); if ((status & ATML_STATUS_DATA_AVAIL) == 0) { - dev_err(chip->pdev, "error reading header\n"); + dev_err(&chip->dev, "error reading header\n"); return -EIO; } *buf++ = ioread8(chip->vendor.iobase); @@ -60,12 +60,12 @@ static int tpm_atml_recv(struct tpm_chip *chip, u8 *buf, size_t count) size = be32_to_cpu(*native_size); if (count < size) { - dev_err(chip->pdev, + dev_err(&chip->dev, "Recv size(%d) less than available space\n", size); for (; i < size; i++) { /* clear the waiting data anyway */ status = ioread8(chip->vendor.iobase + 1); if ((status & ATML_STATUS_DATA_AVAIL) == 0) { - dev_err(chip->pdev, "error reading data\n"); + dev_err(&chip->dev, "error reading data\n"); return -EIO; } } @@ -76,7 +76,7 @@ static int tpm_atml_recv(struct tpm_chip *chip, u8 *buf, size_t count) for (; i < size; i++) { status = ioread8(chip->vendor.iobase + 1); if ((status & ATML_STATUS_DATA_AVAIL) == 0) { - dev_err(chip->pdev, "error reading data\n"); + dev_err(&chip->dev, "error reading data\n"); return -EIO; } *buf++ = ioread8(chip->vendor.iobase); @@ -86,7 +86,7 @@ static int tpm_atml_recv(struct tpm_chip *chip, u8 *buf, size_t count) status = ioread8(chip->vendor.iobase + 1); if (status & ATML_STATUS_DATA_AVAIL) { - dev_err(chip->pdev, "data available is stuck\n"); + dev_err(&chip->dev, "data available is stuck\n"); return -EIO; } @@ -97,9 +97,9 @@ static int tpm_atml_send(struct tpm_chip *chip, u8 *buf, size_t count) { int i; - dev_dbg(chip->pdev, "tpm_atml_send:\n"); + dev_dbg(&chip->dev, "tpm_atml_send:\n"); for (i = 0; i < count; i++) { - dev_dbg(chip->pdev, "%d 0x%x(%d)\n", i, buf[i], buf[i]); + dev_dbg(&chip->dev, "%d 0x%x(%d)\n", i, buf[i], buf[i]); iowrite8(buf[i], chip->vendor.iobase); } diff --git a/drivers/char/tpm/tpm_i2c_atmel.c b/drivers/char/tpm/tpm_i2c_atmel.c index 8dfb88b9739c..dd8f0eb3170a 100644 --- a/drivers/char/tpm/tpm_i2c_atmel.c +++ b/drivers/char/tpm/tpm_i2c_atmel.c @@ -52,7 +52,7 @@ struct priv_data { static int i2c_atmel_send(struct tpm_chip *chip, u8 *buf, size_t len) { struct priv_data *priv = chip->vendor.priv; - struct i2c_client *client = to_i2c_client(chip->pdev); + struct i2c_client *client = to_i2c_client(chip->dev.parent); s32 status; priv->len = 0; @@ -62,7 +62,7 @@ static int i2c_atmel_send(struct tpm_chip *chip, u8 *buf, size_t len) status = i2c_master_send(client, buf, len); - dev_dbg(chip->pdev, + dev_dbg(&chip->dev, "%s(buf=%*ph len=%0zx) -> sts=%d\n", __func__, (int)min_t(size_t, 64, len), buf, len, status); return status; @@ -71,7 +71,7 @@ static int i2c_atmel_send(struct tpm_chip *chip, u8 *buf, size_t len) static int i2c_atmel_recv(struct tpm_chip *chip, u8 *buf, size_t count) { struct priv_data *priv = chip->vendor.priv; - struct i2c_client *client = to_i2c_client(chip->pdev); + struct i2c_client *client = to_i2c_client(chip->dev.parent); struct tpm_output_header *hdr = (struct tpm_output_header *)priv->buffer; u32 expected_len; @@ -88,7 +88,7 @@ static int i2c_atmel_recv(struct tpm_chip *chip, u8 *buf, size_t count) return -ENOMEM; if (priv->len >= expected_len) { - dev_dbg(chip->pdev, + dev_dbg(&chip->dev, "%s early(buf=%*ph count=%0zx) -> ret=%d\n", __func__, (int)min_t(size_t, 64, expected_len), buf, count, expected_len); @@ -97,7 +97,7 @@ static int i2c_atmel_recv(struct tpm_chip *chip, u8 *buf, size_t count) } rc = i2c_master_recv(client, buf, expected_len); - dev_dbg(chip->pdev, + dev_dbg(&chip->dev, "%s reread(buf=%*ph count=%0zx) -> ret=%d\n", __func__, (int)min_t(size_t, 64, expected_len), buf, count, expected_len); @@ -106,13 +106,13 @@ static int i2c_atmel_recv(struct tpm_chip *chip, u8 *buf, size_t count) static void i2c_atmel_cancel(struct tpm_chip *chip) { - dev_err(chip->pdev, "TPM operation cancellation was requested, but is not supported"); + dev_err(&chip->dev, "TPM operation cancellation was requested, but is not supported"); } static u8 i2c_atmel_read_status(struct tpm_chip *chip) { struct priv_data *priv = chip->vendor.priv; - struct i2c_client *client = to_i2c_client(chip->pdev); + struct i2c_client *client = to_i2c_client(chip->dev.parent); int rc; /* The TPM fails the I2C read until it is ready, so we do the entire @@ -125,7 +125,7 @@ static u8 i2c_atmel_read_status(struct tpm_chip *chip) /* Once the TPM has completed the command the command remains readable * until another command is issued. */ rc = i2c_master_recv(client, priv->buffer, sizeof(priv->buffer)); - dev_dbg(chip->pdev, + dev_dbg(&chip->dev, "%s: sts=%d", __func__, rc); if (rc <= 0) return 0; diff --git a/drivers/char/tpm/tpm_i2c_infineon.c b/drivers/char/tpm/tpm_i2c_infineon.c index 63d5d22e9e60..f2aa99e34b4b 100644 --- a/drivers/char/tpm/tpm_i2c_infineon.c +++ b/drivers/char/tpm/tpm_i2c_infineon.c @@ -446,7 +446,7 @@ static int tpm_tis_i2c_recv(struct tpm_chip *chip, u8 *buf, size_t count) /* read first 10 bytes, including tag, paramsize, and result */ size = recv_data(chip, buf, TPM_HEADER_SIZE); if (size < TPM_HEADER_SIZE) { - dev_err(chip->pdev, "Unable to read header\n"); + dev_err(&chip->dev, "Unable to read header\n"); goto out; } @@ -459,14 +459,14 @@ static int tpm_tis_i2c_recv(struct tpm_chip *chip, u8 *buf, size_t count) size += recv_data(chip, &buf[TPM_HEADER_SIZE], expected - TPM_HEADER_SIZE); if (size < expected) { - dev_err(chip->pdev, "Unable to read remainder of result\n"); + dev_err(&chip->dev, "Unable to read remainder of result\n"); size = -ETIME; goto out; } wait_for_stat(chip, TPM_STS_VALID, chip->vendor.timeout_c, &status); if (status & TPM_STS_DATA_AVAIL) { /* retry? */ - dev_err(chip->pdev, "Error left over data\n"); + dev_err(&chip->dev, "Error left over data\n"); size = -EIO; goto out; } diff --git a/drivers/char/tpm/tpm_i2c_nuvoton.c b/drivers/char/tpm/tpm_i2c_nuvoton.c index 847f1597fe9b..a1e1474dda30 100644 --- a/drivers/char/tpm/tpm_i2c_nuvoton.c +++ b/drivers/char/tpm/tpm_i2c_nuvoton.c @@ -96,13 +96,13 @@ static s32 i2c_nuvoton_write_buf(struct i2c_client *client, u8 offset, u8 size, /* read TPM_STS register */ static u8 i2c_nuvoton_read_status(struct tpm_chip *chip) { - struct i2c_client *client = to_i2c_client(chip->pdev); + struct i2c_client *client = to_i2c_client(chip->dev.parent); s32 status; u8 data; status = i2c_nuvoton_read_buf(client, TPM_STS, 1, &data); if (status <= 0) { - dev_err(chip->pdev, "%s() error return %d\n", __func__, + dev_err(&chip->dev, "%s() error return %d\n", __func__, status); data = TPM_STS_ERR_VAL; } @@ -127,13 +127,13 @@ static s32 i2c_nuvoton_write_status(struct i2c_client *client, u8 data) /* write commandReady to TPM_STS register */ static void i2c_nuvoton_ready(struct tpm_chip *chip) { - struct i2c_client *client = to_i2c_client(chip->pdev); + struct i2c_client *client = to_i2c_client(chip->dev.parent); s32 status; /* this causes the current command to be aborted */ status = i2c_nuvoton_write_status(client, TPM_STS_COMMAND_READY); if (status < 0) - dev_err(chip->pdev, + dev_err(&chip->dev, "%s() fail to write TPM_STS.commandReady\n", __func__); } @@ -212,7 +212,7 @@ static int i2c_nuvoton_wait_for_stat(struct tpm_chip *chip, u8 mask, u8 value, return 0; } while (time_before(jiffies, stop)); } - dev_err(chip->pdev, "%s(%02x, %02x) -> timeout\n", __func__, mask, + dev_err(&chip->dev, "%s(%02x, %02x) -> timeout\n", __func__, mask, value); return -ETIMEDOUT; } @@ -240,7 +240,7 @@ static int i2c_nuvoton_recv_data(struct i2c_client *client, &chip->vendor.read_queue) == 0) { burst_count = i2c_nuvoton_get_burstcount(client, chip); if (burst_count < 0) { - dev_err(chip->pdev, + dev_err(&chip->dev, "%s() fail to read burstCount=%d\n", __func__, burst_count); return -EIO; @@ -249,12 +249,12 @@ static int i2c_nuvoton_recv_data(struct i2c_client *client, rc = i2c_nuvoton_read_buf(client, TPM_DATA_FIFO_R, bytes2read, &buf[size]); if (rc < 0) { - dev_err(chip->pdev, + dev_err(&chip->dev, "%s() fail on i2c_nuvoton_read_buf()=%d\n", __func__, rc); return -EIO; } - dev_dbg(chip->pdev, "%s(%d):", __func__, bytes2read); + dev_dbg(&chip->dev, "%s(%d):", __func__, bytes2read); size += bytes2read; } @@ -264,7 +264,7 @@ static int i2c_nuvoton_recv_data(struct i2c_client *client, /* Read TPM command results */ static int i2c_nuvoton_recv(struct tpm_chip *chip, u8 *buf, size_t count) { - struct device *dev = chip->pdev; + struct device *dev = chip->dev.parent; struct i2c_client *client = to_i2c_client(dev); s32 rc; int expected, status, burst_count, retries, size = 0; @@ -334,7 +334,7 @@ static int i2c_nuvoton_recv(struct tpm_chip *chip, u8 *buf, size_t count) break; } i2c_nuvoton_ready(chip); - dev_dbg(chip->pdev, "%s() -> %d\n", __func__, size); + dev_dbg(&chip->dev, "%s() -> %d\n", __func__, size); return size; } @@ -347,7 +347,7 @@ static int i2c_nuvoton_recv(struct tpm_chip *chip, u8 *buf, size_t count) */ static int i2c_nuvoton_send(struct tpm_chip *chip, u8 *buf, size_t len) { - struct device *dev = chip->pdev; + struct device *dev = chip->dev.parent; struct i2c_client *client = to_i2c_client(dev); u32 ordinal; size_t count = 0; diff --git a/drivers/char/tpm/tpm_infineon.c b/drivers/char/tpm/tpm_infineon.c index 6c488e635fdd..e3cf9f3545c5 100644 --- a/drivers/char/tpm/tpm_infineon.c +++ b/drivers/char/tpm/tpm_infineon.c @@ -195,9 +195,9 @@ static int wait(struct tpm_chip *chip, int wait_for_bit) } if (i == TPM_MAX_TRIES) { /* timeout occurs */ if (wait_for_bit == STAT_XFE) - dev_err(chip->pdev, "Timeout in wait(STAT_XFE)\n"); + dev_err(&chip->dev, "Timeout in wait(STAT_XFE)\n"); if (wait_for_bit == STAT_RDA) - dev_err(chip->pdev, "Timeout in wait(STAT_RDA)\n"); + dev_err(&chip->dev, "Timeout in wait(STAT_RDA)\n"); return -EIO; } return 0; @@ -220,7 +220,7 @@ static void wait_and_send(struct tpm_chip *chip, u8 sendbyte) static void tpm_wtx(struct tpm_chip *chip) { number_of_wtx++; - dev_info(chip->pdev, "Granting WTX (%02d / %02d)\n", + dev_info(&chip->dev, "Granting WTX (%02d / %02d)\n", number_of_wtx, TPM_MAX_WTX_PACKAGES); wait_and_send(chip, TPM_VL_VER); wait_and_send(chip, TPM_CTRL_WTX); @@ -231,7 +231,7 @@ static void tpm_wtx(struct tpm_chip *chip) static void tpm_wtx_abort(struct tpm_chip *chip) { - dev_info(chip->pdev, "Aborting WTX\n"); + dev_info(&chip->dev, "Aborting WTX\n"); wait_and_send(chip, TPM_VL_VER); wait_and_send(chip, TPM_CTRL_WTX_ABORT); wait_and_send(chip, 0x00); @@ -257,7 +257,7 @@ recv_begin: } if (buf[0] != TPM_VL_VER) { - dev_err(chip->pdev, + dev_err(&chip->dev, "Wrong transport protocol implementation!\n"); return -EIO; } @@ -272,7 +272,7 @@ recv_begin: } if ((size == 0x6D00) && (buf[1] == 0x80)) { - dev_err(chip->pdev, "Error handling on vendor layer!\n"); + dev_err(&chip->dev, "Error handling on vendor layer!\n"); return -EIO; } @@ -284,7 +284,7 @@ recv_begin: } if (buf[1] == TPM_CTRL_WTX) { - dev_info(chip->pdev, "WTX-package received\n"); + dev_info(&chip->dev, "WTX-package received\n"); if (number_of_wtx < TPM_MAX_WTX_PACKAGES) { tpm_wtx(chip); goto recv_begin; @@ -295,14 +295,14 @@ recv_begin: } if (buf[1] == TPM_CTRL_WTX_ABORT_ACK) { - dev_info(chip->pdev, "WTX-abort acknowledged\n"); + dev_info(&chip->dev, "WTX-abort acknowledged\n"); return size; } if (buf[1] == TPM_CTRL_ERROR) { - dev_err(chip->pdev, "ERROR-package received:\n"); + dev_err(&chip->dev, "ERROR-package received:\n"); if (buf[4] == TPM_INF_NAK) - dev_err(chip->pdev, + dev_err(&chip->dev, "-> Negative acknowledgement" " - retransmit command!\n"); return -EIO; @@ -321,7 +321,7 @@ static int tpm_inf_send(struct tpm_chip *chip, u8 * buf, size_t count) ret = empty_fifo(chip, 1); if (ret) { - dev_err(chip->pdev, "Timeout while clearing FIFO\n"); + dev_err(&chip->dev, "Timeout while clearing FIFO\n"); return -EIO; } diff --git a/drivers/char/tpm/tpm_nsc.c b/drivers/char/tpm/tpm_nsc.c index 289389ecef84..766370bed60c 100644 --- a/drivers/char/tpm/tpm_nsc.c +++ b/drivers/char/tpm/tpm_nsc.c @@ -113,7 +113,7 @@ static int nsc_wait_for_ready(struct tpm_chip *chip) } while (time_before(jiffies, stop)); - dev_info(chip->pdev, "wait for ready failed\n"); + dev_info(&chip->dev, "wait for ready failed\n"); return -EBUSY; } @@ -129,12 +129,12 @@ static int tpm_nsc_recv(struct tpm_chip *chip, u8 * buf, size_t count) return -EIO; if (wait_for_stat(chip, NSC_STATUS_F0, NSC_STATUS_F0, &data) < 0) { - dev_err(chip->pdev, "F0 timeout\n"); + dev_err(&chip->dev, "F0 timeout\n"); return -EIO; } if ((data = inb(chip->vendor.base + NSC_DATA)) != NSC_COMMAND_NORMAL) { - dev_err(chip->pdev, "not in normal mode (0x%x)\n", + dev_err(&chip->dev, "not in normal mode (0x%x)\n", data); return -EIO; } @@ -143,7 +143,7 @@ static int tpm_nsc_recv(struct tpm_chip *chip, u8 * buf, size_t count) for (p = buffer; p < &buffer[count]; p++) { if (wait_for_stat (chip, NSC_STATUS_OBF, NSC_STATUS_OBF, &data) < 0) { - dev_err(chip->pdev, + dev_err(&chip->dev, "OBF timeout (while reading data)\n"); return -EIO; } @@ -154,11 +154,11 @@ static int tpm_nsc_recv(struct tpm_chip *chip, u8 * buf, size_t count) if ((data & NSC_STATUS_F0) == 0 && (wait_for_stat(chip, NSC_STATUS_F0, NSC_STATUS_F0, &data) < 0)) { - dev_err(chip->pdev, "F0 not set\n"); + dev_err(&chip->dev, "F0 not set\n"); return -EIO; } if ((data = inb(chip->vendor.base + NSC_DATA)) != NSC_COMMAND_EOC) { - dev_err(chip->pdev, + dev_err(&chip->dev, "expected end of command(0x%x)\n", data); return -EIO; } @@ -189,19 +189,19 @@ static int tpm_nsc_send(struct tpm_chip *chip, u8 * buf, size_t count) return -EIO; if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) { - dev_err(chip->pdev, "IBF timeout\n"); + dev_err(&chip->dev, "IBF timeout\n"); return -EIO; } outb(NSC_COMMAND_NORMAL, chip->vendor.base + NSC_COMMAND); if (wait_for_stat(chip, NSC_STATUS_IBR, NSC_STATUS_IBR, &data) < 0) { - dev_err(chip->pdev, "IBR timeout\n"); + dev_err(&chip->dev, "IBR timeout\n"); return -EIO; } for (i = 0; i < count; i++) { if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) { - dev_err(chip->pdev, + dev_err(&chip->dev, "IBF timeout (while writing data)\n"); return -EIO; } @@ -209,7 +209,7 @@ static int tpm_nsc_send(struct tpm_chip *chip, u8 * buf, size_t count) } if (wait_for_stat(chip, NSC_STATUS_IBF, 0, &data) < 0) { - dev_err(chip->pdev, "IBF timeout\n"); + dev_err(&chip->dev, "IBF timeout\n"); return -EIO; } outb(NSC_COMMAND_EOC, chip->vendor.base + NSC_COMMAND); diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index f10a107614b4..7f13221aeb30 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -293,7 +293,7 @@ static int tpm_tis_recv(struct tpm_chip *chip, u8 *buf, size_t count) /* read first 10 bytes, including tag, paramsize, and result */ if ((size = recv_data(chip, buf, TPM_HEADER_SIZE)) < TPM_HEADER_SIZE) { - dev_err(chip->pdev, "Unable to read header\n"); + dev_err(&chip->dev, "Unable to read header\n"); goto out; } @@ -306,7 +306,7 @@ static int tpm_tis_recv(struct tpm_chip *chip, u8 *buf, size_t count) if ((size += recv_data(chip, &buf[TPM_HEADER_SIZE], expected - TPM_HEADER_SIZE)) < expected) { - dev_err(chip->pdev, "Unable to read remainder of result\n"); + dev_err(&chip->dev, "Unable to read remainder of result\n"); size = -ETIME; goto out; } @@ -315,7 +315,7 @@ static int tpm_tis_recv(struct tpm_chip *chip, u8 *buf, size_t count) &chip->vendor.int_queue, false); status = tpm_tis_status(chip); if (status & TPM_STS_DATA_AVAIL) { /* retry? */ - dev_err(chip->pdev, "Error left over data\n"); + dev_err(&chip->dev, "Error left over data\n"); size = -EIO; goto out; } @@ -401,7 +401,7 @@ static void disable_interrupts(struct tpm_chip *chip) iowrite32(intmask, chip->vendor.iobase + TPM_INT_ENABLE(chip->vendor.locality)); - devm_free_irq(chip->pdev, chip->vendor.irq, chip); + devm_free_irq(&chip->dev, chip->vendor.irq, chip); chip->vendor.irq = 0; } @@ -463,7 +463,7 @@ static int tpm_tis_send(struct tpm_chip *chip, u8 *buf, size_t len) msleep(1); if (!priv->irq_tested) { disable_interrupts(chip); - dev_err(chip->pdev, + dev_err(&chip->dev, FW_BUG "TPM interrupt not working, polling instead\n"); } priv->irq_tested = true; @@ -533,7 +533,7 @@ static int probe_itpm(struct tpm_chip *chip) rc = tpm_tis_send_data(chip, cmd_getticks, len); if (rc == 0) { - dev_info(chip->pdev, "Detected an iTPM.\n"); + dev_info(&chip->dev, "Detected an iTPM.\n"); rc = 1; } else rc = -EFAULT; @@ -766,7 +766,7 @@ static int tpm_tis_init(struct device *dev, struct tpm_info *tpm_info, if (devm_request_irq (dev, i, tis_int_probe, IRQF_SHARED, chip->devname, chip) != 0) { - dev_info(chip->pdev, + dev_info(&chip->dev, "Unable to request irq: %d for probe\n", i); continue; @@ -818,7 +818,7 @@ static int tpm_tis_init(struct device *dev, struct tpm_info *tpm_info, if (devm_request_irq (dev, chip->vendor.irq, tis_int_handler, IRQF_SHARED, chip->devname, chip) != 0) { - dev_info(chip->pdev, + dev_info(&chip->dev, "Unable to request irq: %d for use\n", chip->vendor.irq); chip->vendor.irq = 0; From 5ec5771bcb2b4c45771f3f750701ab79dd4cb21a Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Fri, 12 Feb 2016 20:29:53 -0700 Subject: [PATCH 40/58] tpm: Provide strong locking for device removal commit 4e26195f240d73150e8308ae42874702e3df8d2c upstream. Add a read/write semaphore around the ops function pointers so ops can be set to null when the driver un-registers. Previously the tpm core expected module locking to be enough to ensure that tpm_unregister could not be called during certain times, however that hasn't been sufficient for a long time. Introduce a read/write semaphore around 'ops' so the core can set it to null when unregistering. This provides a strong fence around the driver callbacks, guaranteeing to the driver that no callbacks are running or will run again. For now the ops_lock is placed very high in the call stack, it could be pushed down and made more granular in future if necessary. Signed-off-by: Jason Gunthorpe Reviewed-by: Stefan Berger Reviewed-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen Signed-off-by: Greg Kroah-Hartman --- drivers/char/tpm/tpm-chip.c | 72 +++++++++++++++++++++++++++++--- drivers/char/tpm/tpm-dev.c | 11 ++++- drivers/char/tpm/tpm-interface.c | 19 +++++---- drivers/char/tpm/tpm-sysfs.c | 5 +++ drivers/char/tpm/tpm.h | 14 ++++--- 5 files changed, 100 insertions(+), 21 deletions(-) diff --git a/drivers/char/tpm/tpm-chip.c b/drivers/char/tpm/tpm-chip.c index f55b4921c723..f3a887e4f692 100644 --- a/drivers/char/tpm/tpm-chip.c +++ b/drivers/char/tpm/tpm-chip.c @@ -36,10 +36,60 @@ static DEFINE_SPINLOCK(driver_lock); struct class *tpm_class; dev_t tpm_devt; -/* - * tpm_chip_find_get - return tpm_chip for a given chip number - * @chip_num the device number for the chip +/** + * tpm_try_get_ops() - Get a ref to the tpm_chip + * @chip: Chip to ref + * + * The caller must already have some kind of locking to ensure that chip is + * valid. This function will lock the chip so that the ops member can be + * accessed safely. The locking prevents tpm_chip_unregister from + * completing, so it should not be held for long periods. + * + * Returns -ERRNO if the chip could not be got. */ +int tpm_try_get_ops(struct tpm_chip *chip) +{ + int rc = -EIO; + + get_device(&chip->dev); + + down_read(&chip->ops_sem); + if (!chip->ops) + goto out_lock; + + if (!try_module_get(chip->dev.parent->driver->owner)) + goto out_lock; + + return 0; +out_lock: + up_read(&chip->ops_sem); + put_device(&chip->dev); + return rc; +} +EXPORT_SYMBOL_GPL(tpm_try_get_ops); + +/** + * tpm_put_ops() - Release a ref to the tpm_chip + * @chip: Chip to put + * + * This is the opposite pair to tpm_try_get_ops(). After this returns chip may + * be kfree'd. + */ +void tpm_put_ops(struct tpm_chip *chip) +{ + module_put(chip->dev.parent->driver->owner); + up_read(&chip->ops_sem); + put_device(&chip->dev); +} +EXPORT_SYMBOL_GPL(tpm_put_ops); + +/** + * tpm_chip_find_get() - return tpm_chip for a given chip number + * @chip_num: id to find + * + * The return'd chip has been tpm_try_get_ops'd and must be released via + * tpm_put_ops + */ struct tpm_chip *tpm_chip_find_get(int chip_num) { struct tpm_chip *pos, *chip = NULL; @@ -49,10 +99,10 @@ struct tpm_chip *tpm_chip_find_get(int chip_num) if (chip_num != TPM_ANY_NUM && chip_num != pos->dev_num) continue; - if (try_module_get(pos->dev.parent->driver->owner)) { + /* rcu prevents chip from being free'd */ + if (!tpm_try_get_ops(pos)) chip = pos; - break; - } + break; } rcu_read_unlock(); return chip; @@ -94,6 +144,7 @@ struct tpm_chip *tpmm_chip_alloc(struct device *dev, return ERR_PTR(-ENOMEM); mutex_init(&chip->tpm_mutex); + init_rwsem(&chip->ops_sem); INIT_LIST_HEAD(&chip->list); chip->ops = ops; @@ -171,6 +222,12 @@ static int tpm_add_char_device(struct tpm_chip *chip) static void tpm_del_char_device(struct tpm_chip *chip) { cdev_del(&chip->cdev); + + /* Make the driver uncallable. */ + down_write(&chip->ops_sem); + chip->ops = NULL; + up_write(&chip->ops_sem); + device_del(&chip->dev); } @@ -256,6 +313,9 @@ EXPORT_SYMBOL_GPL(tpm_chip_register); * Takes the chip first away from the list of available TPM chips and then * cleans up all the resources reserved by tpm_chip_register(). * + * Once this function returns the driver call backs in 'op's will not be + * running and will no longer start. + * * NOTE: This function should be only called before deinitializing chip * resources. */ diff --git a/drivers/char/tpm/tpm-dev.c b/drivers/char/tpm/tpm-dev.c index 6ed0651cbe58..912ad30be585 100644 --- a/drivers/char/tpm/tpm-dev.c +++ b/drivers/char/tpm/tpm-dev.c @@ -136,9 +136,18 @@ static ssize_t tpm_write(struct file *file, const char __user *buf, return -EFAULT; } - /* atomic tpm command send and result receive */ + /* atomic tpm command send and result receive. We only hold the ops + * lock during this period so that the tpm can be unregistered even if + * the char dev is held open. + */ + if (tpm_try_get_ops(priv->chip)) { + mutex_unlock(&priv->buffer_mutex); + return -EPIPE; + } out_size = tpm_transmit(priv->chip, priv->data_buffer, sizeof(priv->data_buffer), 0); + + tpm_put_ops(priv->chip); if (out_size < 0) { mutex_unlock(&priv->buffer_mutex); return out_size; diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index 4391953a7711..8588f2e4b9af 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -687,7 +687,7 @@ int tpm_is_tpm2(u32 chip_num) rc = (chip->flags & TPM_CHIP_FLAG_TPM2) != 0; - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } @@ -716,7 +716,7 @@ int tpm_pcr_read(u32 chip_num, int pcr_idx, u8 *res_buf) rc = tpm2_pcr_read(chip, pcr_idx, res_buf); else rc = tpm_pcr_read_dev(chip, pcr_idx, res_buf); - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } EXPORT_SYMBOL_GPL(tpm_pcr_read); @@ -751,7 +751,7 @@ int tpm_pcr_extend(u32 chip_num, int pcr_idx, const u8 *hash) if (chip->flags & TPM_CHIP_FLAG_TPM2) { rc = tpm2_pcr_extend(chip, pcr_idx, hash); - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } @@ -761,7 +761,7 @@ int tpm_pcr_extend(u32 chip_num, int pcr_idx, const u8 *hash) rc = tpm_transmit_cmd(chip, &cmd, EXTEND_PCR_RESULT_SIZE, 0, "attempting extend a PCR value"); - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } EXPORT_SYMBOL_GPL(tpm_pcr_extend); @@ -842,7 +842,7 @@ int tpm_send(u32 chip_num, void *cmd, size_t buflen) rc = tpm_transmit_cmd(chip, cmd, buflen, 0, "attempting tpm_cmd"); - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } EXPORT_SYMBOL_GPL(tpm_send); @@ -1025,7 +1025,7 @@ int tpm_get_random(u32 chip_num, u8 *out, size_t max) if (chip->flags & TPM_CHIP_FLAG_TPM2) { err = tpm2_get_random(chip, out, max); - tpm_chip_put(chip); + tpm_put_ops(chip); return err; } @@ -1047,7 +1047,7 @@ int tpm_get_random(u32 chip_num, u8 *out, size_t max) num_bytes -= recd; } while (retries-- && total < max); - tpm_chip_put(chip); + tpm_put_ops(chip); return total ? total : -EIO; } EXPORT_SYMBOL_GPL(tpm_get_random); @@ -1073,7 +1073,7 @@ int tpm_seal_trusted(u32 chip_num, struct trusted_key_payload *payload, rc = tpm2_seal_trusted(chip, payload, options); - tpm_chip_put(chip); + tpm_put_ops(chip); return rc; } EXPORT_SYMBOL_GPL(tpm_seal_trusted); @@ -1099,7 +1099,8 @@ int tpm_unseal_trusted(u32 chip_num, struct trusted_key_payload *payload, rc = tpm2_unseal_trusted(chip, payload, options); - tpm_chip_put(chip); + tpm_put_ops(chip); + return rc; } EXPORT_SYMBOL_GPL(tpm_unseal_trusted); diff --git a/drivers/char/tpm/tpm-sysfs.c b/drivers/char/tpm/tpm-sysfs.c index 10370c22e98b..8af4145d10c7 100644 --- a/drivers/char/tpm/tpm-sysfs.c +++ b/drivers/char/tpm/tpm-sysfs.c @@ -295,5 +295,10 @@ int tpm_sysfs_add_device(struct tpm_chip *chip) void tpm_sysfs_del_device(struct tpm_chip *chip) { + /* The sysfs routines rely on an implicit tpm_try_get_ops, this + * function is called before ops is null'd and the sysfs core + * synchronizes this removal so that no callbacks are running or can + * run again + */ sysfs_remove_group(&chip->dev.parent->kobj, &tpm_dev_group); } diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 57c4c26c38ea..e21e2c599e66 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -174,7 +174,13 @@ struct tpm_chip { struct device dev; struct cdev cdev; + /* A driver callback under ops cannot be run unless ops_sem is held + * (sometimes implicitly, eg for the sysfs code). ops becomes null + * when the driver is unregistered, see tpm_try_get_ops. + */ + struct rw_semaphore ops_sem; const struct tpm_class_ops *ops; + unsigned int flags; int dev_num; /* /dev/tpm# */ @@ -200,11 +206,6 @@ struct tpm_chip { #define to_tpm_chip(d) container_of(d, struct tpm_chip, dev) -static inline void tpm_chip_put(struct tpm_chip *chip) -{ - module_put(chip->dev.parent->driver->owner); -} - static inline int tpm_read_index(int base, int index) { outb(index, base); @@ -516,6 +517,9 @@ extern int wait_for_tpm_stat(struct tpm_chip *, u8, unsigned long, wait_queue_head_t *, bool); struct tpm_chip *tpm_chip_find_get(int chip_num); +__must_check int tpm_try_get_ops(struct tpm_chip *chip); +void tpm_put_ops(struct tpm_chip *chip); + extern struct tpm_chip *tpmm_chip_alloc(struct device *dev, const struct tpm_class_ops *ops); extern int tpm_chip_register(struct tpm_chip *chip); From 5c9a2972983fca37e73648d9a3aa62a9ad048c3c Mon Sep 17 00:00:00 2001 From: Josh Zimmerman Date: Sun, 25 Jun 2017 14:53:23 -0700 Subject: [PATCH 41/58] Add "shutdown" to "struct class". commit f77af15165847406b15d8f70c382c4cb15846b2a upstream. The TPM class has some common shutdown code that must be executed for all drivers. This adds some needed functionality for that. Signed-off-by: Josh Zimmerman Acked-by: Greg Kroah-Hartman Fixes: 74d6b3ceaa17 ("tpm: fix suspend/resume paths for TPM 2.0") Reviewed-by: Jarkko Sakkinen Tested-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen Signed-off-by: James Morris Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 6 +++++- include/linux/device.h | 2 ++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index f18856f5954b..afe045792796 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -2094,7 +2094,11 @@ void device_shutdown(void) pm_runtime_get_noresume(dev); pm_runtime_barrier(dev); - if (dev->bus && dev->bus->shutdown) { + if (dev->class && dev->class->shutdown) { + if (initcall_debug) + dev_info(dev, "shutdown\n"); + dev->class->shutdown(dev); + } else if (dev->bus && dev->bus->shutdown) { if (initcall_debug) dev_info(dev, "shutdown\n"); dev->bus->shutdown(dev); diff --git a/include/linux/device.h b/include/linux/device.h index b8f411b57dcb..7075a2485ed3 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -368,6 +368,7 @@ int subsys_virtual_register(struct bus_type *subsys, * @suspend: Used to put the device to sleep mode, usually to a low power * state. * @resume: Used to bring the device from the sleep mode. + * @shutdown: Called at shut-down time to quiesce the device. * @ns_type: Callbacks so sysfs can detemine namespaces. * @namespace: Namespace of the device belongs to this class. * @pm: The default device power management operations of this class. @@ -396,6 +397,7 @@ struct class { int (*suspend)(struct device *dev, pm_message_t state); int (*resume)(struct device *dev); + int (*shutdown)(struct device *dev); const struct kobj_ns_type_operations *ns_type; const void *(*namespace)(struct device *dev); From 0f4fa2138e83d33468baac218c4be652f4619cb7 Mon Sep 17 00:00:00 2001 From: Josh Zimmerman Date: Sun, 25 Jun 2017 14:53:24 -0700 Subject: [PATCH 42/58] tpm: Issue a TPM2_Shutdown for TPM2 devices. commit d1bd4a792d3961a04e6154118816b00167aad91a upstream. If a TPM2 loses power without a TPM2_Shutdown command being issued (a "disorderly reboot"), it may lose some state that has yet to be persisted to NVRam, and will increment the DA counter. After the DA counter gets sufficiently large, the TPM will lock the user out. NOTE: This only changes behavior on TPM2 devices. Since TPM1 uses sysfs, and sysfs relies on implicit locking on chip->ops, it is not safe to allow this code to run in TPM1, or to add sysfs support to TPM2, until that locking is made explicit. Signed-off-by: Josh Zimmerman Cc: stable@vger.kernel.org Fixes: 74d6b3ceaa17 ("tpm: fix suspend/resume paths for TPM 2.0") Reviewed-by: Jarkko Sakkinen Tested-by: Jarkko Sakkinen Signed-off-by: Jarkko Sakkinen Signed-off-by: James Morris Signed-off-by: Greg Kroah-Hartman --- drivers/char/tpm/tpm-chip.c | 36 ++++++++++++++++++++++++++++++++++++ drivers/char/tpm/tpm-sysfs.c | 7 +++++++ 2 files changed, 43 insertions(+) diff --git a/drivers/char/tpm/tpm-chip.c b/drivers/char/tpm/tpm-chip.c index f3a887e4f692..6d56877b2e0a 100644 --- a/drivers/char/tpm/tpm-chip.c +++ b/drivers/char/tpm/tpm-chip.c @@ -124,6 +124,41 @@ static void tpm_dev_release(struct device *dev) kfree(chip); } + +/** + * tpm_class_shutdown() - prepare the TPM device for loss of power. + * @dev: device to which the chip is associated. + * + * Issues a TPM2_Shutdown command prior to loss of power, as required by the + * TPM 2.0 spec. + * Then, calls bus- and device- specific shutdown code. + * + * XXX: This codepath relies on the fact that sysfs is not enabled for + * TPM2: sysfs uses an implicit lock on chip->ops, so this could race if TPM2 + * has sysfs support enabled before TPM sysfs's implicit locking is fixed. + */ +static int tpm_class_shutdown(struct device *dev) +{ + struct tpm_chip *chip = container_of(dev, struct tpm_chip, dev); + + if (chip->flags & TPM_CHIP_FLAG_TPM2) { + down_write(&chip->ops_sem); + tpm2_shutdown(chip, TPM2_SU_CLEAR); + chip->ops = NULL; + up_write(&chip->ops_sem); + } + /* Allow bus- and device-specific code to run. Note: since chip->ops + * is NULL, more-specific shutdown code will not be able to issue TPM + * commands. + */ + if (dev->bus && dev->bus->shutdown) + dev->bus->shutdown(dev); + else if (dev->driver && dev->driver->shutdown) + dev->driver->shutdown(dev); + return 0; +} + + /** * tpmm_chip_alloc() - allocate a new struct tpm_chip instance * @dev: device to which the chip is associated @@ -166,6 +201,7 @@ struct tpm_chip *tpmm_chip_alloc(struct device *dev, dev_set_drvdata(dev, chip); chip->dev.class = tpm_class; + chip->dev.class->shutdown = tpm_class_shutdown; chip->dev.release = tpm_dev_release; chip->dev.parent = dev; #ifdef CONFIG_ACPI diff --git a/drivers/char/tpm/tpm-sysfs.c b/drivers/char/tpm/tpm-sysfs.c index 8af4145d10c7..6a4056a3f7ee 100644 --- a/drivers/char/tpm/tpm-sysfs.c +++ b/drivers/char/tpm/tpm-sysfs.c @@ -284,6 +284,13 @@ static const struct attribute_group tpm_dev_group = { int tpm_sysfs_add_device(struct tpm_chip *chip) { int err; + + /* XXX: If you wish to remove this restriction, you must first update + * tpm_sysfs to explicitly lock chip->ops. + */ + if (chip->flags & TPM_CHIP_FLAG_TPM2) + return 0; + err = sysfs_create_group(&chip->dev.parent->kobj, &tpm_dev_group); From 8f93a9aa1d654c691602359d426ab66997947b1c Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Fri, 14 Jul 2017 14:49:38 -0700 Subject: [PATCH 43/58] mm: fix overflow check in expand_upwards() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 37511fb5c91db93d8bd6e3f52f86e5a7ff7cfcdf upstream. Jörn Engel noticed that the expand_upwards() function might not return -ENOMEM in case the requested address is (unsigned long)-PAGE_SIZE and if the architecture didn't defined TASK_SIZE as multiple of PAGE_SIZE. Affected architectures are arm, frv, m68k, blackfin, h8300 and xtensa which all define TASK_SIZE as 0xffffffff, but since none of those have an upwards-growing stack we currently have no actual issue. Nevertheless let's fix this just in case any of the architectures with an upward-growing stack (currently parisc, metag and partly ia64) define TASK_SIZE similar. Link: http://lkml.kernel.org/r/20170702192452.GA11868@p100.box Fixes: bd726c90b6b8 ("Allow stack to grow up to address space limit") Signed-off-by: Helge Deller Reported-by: Jörn Engel Cc: Hugh Dickins Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- mm/mmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/mmap.c b/mm/mmap.c index 0990f8bc0fbe..eaa460ddcaf9 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -2176,7 +2176,7 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address) /* Guard against exceeding limits of the address space. */ address &= PAGE_MASK; - if (address >= TASK_SIZE) + if (address >= (TASK_SIZE & PAGE_MASK)) return -ENOMEM; address += PAGE_SIZE; From d05fbdbe6d1c2b00f1d2c3010a4a4c710fb4097a Mon Sep 17 00:00:00 2001 From: Martin Hicks Date: Tue, 2 May 2017 09:38:35 -0400 Subject: [PATCH 44/58] crypto: talitos - Extend max key length for SHA384/512-HMAC and AEAD MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 03d2c5114c95797c0aa7d9f463348b171a274fd4 upstream. An updated patch that also handles the additional key length requirements for the AEAD algorithms. The max keysize is not 96. For SHA384/512 it's 128, and for the AEAD algorithms it's longer still. Extend the max keysize for the AEAD size for AES256 + HMAC(SHA512). Fixes: 357fb60502ede ("crypto: talitos - add sha224, sha384 and sha512 to existing AEAD algorithms") Signed-off-by: Martin Hicks Acked-by: Horia Geantă Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- drivers/crypto/talitos.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/talitos.c b/drivers/crypto/talitos.c index 9a8a18aafd5c..6a60936b46e0 100644 --- a/drivers/crypto/talitos.c +++ b/drivers/crypto/talitos.c @@ -804,7 +804,7 @@ static void talitos_unregister_rng(struct device *dev) * crypto alg */ #define TALITOS_CRA_PRIORITY 3000 -#define TALITOS_MAX_KEY_SIZE 96 +#define TALITOS_MAX_KEY_SIZE (AES_MAX_KEY_SIZE + SHA512_BLOCK_SIZE) #define TALITOS_MAX_IV_LENGTH 16 /* max of AES_BLOCK_SIZE, DES3_EDE_BLOCK_SIZE */ struct talitos_ctx { @@ -1388,6 +1388,11 @@ static int ablkcipher_setkey(struct crypto_ablkcipher *cipher, { struct talitos_ctx *ctx = crypto_ablkcipher_ctx(cipher); + if (keylen > TALITOS_MAX_KEY_SIZE) { + crypto_ablkcipher_set_flags(cipher, CRYPTO_TFM_RES_BAD_KEY_LEN); + return -EINVAL; + } + memcpy(&ctx->key, key, keylen); ctx->keylen = keylen; From fc43f0935035008a65e1a8ed50c99679dce65377 Mon Sep 17 00:00:00 2001 From: Gilad Ben-Yossef Date: Wed, 28 Jun 2017 10:22:03 +0300 Subject: [PATCH 45/58] crypto: atmel - only treat EBUSY as transient if backlog commit 1606043f214f912a52195293614935811a6e3e53 upstream. The Atmel SHA driver was treating -EBUSY as indication of queueing to backlog without checking that backlog is enabled for the request. Fix it by checking request flags. Signed-off-by: Gilad Ben-Yossef Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- drivers/crypto/atmel-sha.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/crypto/atmel-sha.c b/drivers/crypto/atmel-sha.c index 0dadb6332f0e..7abe908427df 100644 --- a/drivers/crypto/atmel-sha.c +++ b/drivers/crypto/atmel-sha.c @@ -963,7 +963,9 @@ static int atmel_sha_finup(struct ahash_request *req) ctx->flags |= SHA_FLAGS_FINUP; err1 = atmel_sha_update(req); - if (err1 == -EINPROGRESS || err1 == -EBUSY) + if (err1 == -EINPROGRESS || + (err1 == -EBUSY && (ahash_request_flags(req) & + CRYPTO_TFM_REQ_MAY_BACKLOG))) return err1; /* From f1bf5d83a9c17b739f2a944e3686d6b97693e67f Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Tue, 4 Jul 2017 12:21:12 +0800 Subject: [PATCH 46/58] crypto: sha1-ssse3 - Disable avx2 commit b82ce24426a4071da9529d726057e4e642948667 upstream. It has been reported that sha1-avx2 can cause page faults by reading beyond the end of the input. This patch disables it until it can be fixed. Fixes: 7c1da8d0d046 ("crypto: sha - SHA1 transform x86_64 AVX2") Reported-by: Jan Stancek Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- arch/x86/crypto/sha1_ssse3_glue.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/crypto/sha1_ssse3_glue.c b/arch/x86/crypto/sha1_ssse3_glue.c index dd14616b7739..7de207a11014 100644 --- a/arch/x86/crypto/sha1_ssse3_glue.c +++ b/arch/x86/crypto/sha1_ssse3_glue.c @@ -201,7 +201,7 @@ asmlinkage void sha1_transform_avx2(u32 *digest, const char *data, static bool avx2_usable(void) { - if (avx_usable() && boot_cpu_has(X86_FEATURE_AVX2) + if (false && avx_usable() && boot_cpu_has(X86_FEATURE_AVX2) && boot_cpu_has(X86_FEATURE_BMI1) && boot_cpu_has(X86_FEATURE_BMI2)) return true; From 4e3c1188521a62ccaa1a7e41498be24a886c502b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Horia=20Geant=C4=83?= Date: Fri, 7 Jul 2017 16:57:06 +0300 Subject: [PATCH 47/58] crypto: caam - fix signals handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 7459e1d25ffefa2b1be799477fcc1f6c62f6cec7 upstream. Driver does not properly handle the case when signals interrupt wait_for_completion_interruptible(): -it does not check for return value -completion structure is allocated on stack; in case a signal interrupts the sleep, it will go out of scope, causing the worker thread (caam_jr_dequeue) to fail when it accesses it wait_for_completion_interruptible() is replaced with uninterruptable wait_for_completion(). We choose to block all signals while waiting for I/O (device executing the split key generation job descriptor) since the alternative - in order to have a deterministic device state - would be to flush the job ring (aborting *all* in-progress jobs). Fixes: 045e36780f115 ("crypto: caam - ahash hmac support") Fixes: 4c1ec1f930154 ("crypto: caam - refactor key_gen, sg") Signed-off-by: Horia Geantă Signed-off-by: Herbert Xu Signed-off-by: Greg Kroah-Hartman --- drivers/crypto/caam/caamhash.c | 2 +- drivers/crypto/caam/key_gen.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/crypto/caam/caamhash.c b/drivers/crypto/caam/caamhash.c index 99d5e11db194..e06cc5df30be 100644 --- a/drivers/crypto/caam/caamhash.c +++ b/drivers/crypto/caam/caamhash.c @@ -498,7 +498,7 @@ static int hash_digest_key(struct caam_hash_ctx *ctx, const u8 *key_in, ret = caam_jr_enqueue(jrdev, desc, split_key_done, &result); if (!ret) { /* in progress */ - wait_for_completion_interruptible(&result.completion); + wait_for_completion(&result.completion); ret = result.err; #ifdef DEBUG print_hex_dump(KERN_ERR, diff --git a/drivers/crypto/caam/key_gen.c b/drivers/crypto/caam/key_gen.c index e1eaf4ff9762..3ce1d5cdcbd2 100644 --- a/drivers/crypto/caam/key_gen.c +++ b/drivers/crypto/caam/key_gen.c @@ -103,7 +103,7 @@ int gen_split_key(struct device *jrdev, u8 *key_out, int split_key_len, ret = caam_jr_enqueue(jrdev, desc, split_key_done, &result); if (!ret) { /* in progress */ - wait_for_completion_interruptible(&result.completion); + wait_for_completion(&result.completion); ret = result.err; #ifdef DEBUG print_hex_dump(KERN_ERR, "ctx.key@"__stringify(__LINE__)": ", From 5c34f49776b8d556bb83d31945cfdb5340792bf2 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 25 Apr 2017 14:00:49 +0200 Subject: [PATCH 48/58] sched/topology: Fix overlapping sched_group_mask commit 73bb059f9b8a00c5e1bf2f7ca83138c05d05e600 upstream. The point of sched_group_mask is to select those CPUs from sched_group_cpus that can actually arrive at this balance domain. The current code gets it wrong, as can be readily demonstrated with a topology like: node 0 1 2 3 0: 10 20 30 20 1: 20 10 20 30 2: 30 20 10 20 3: 20 30 20 10 Where (for example) domain 1 on CPU1 ends up with a mask that includes CPU0: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 (mask: 1), 2, 0 [] domain 1: span 0-3 level NUMA [] groups: 0-2 (mask: 0-2) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072) This causes sched_balance_cpu() to compute the wrong CPU and consequently should_we_balance() will terminate early resulting in missed load-balance opportunities. The fixed topology looks like: [] CPU1 attaching sched-domain: [] domain 0: span 0-2 level NUMA [] groups: 1 (mask: 1), 2, 0 [] domain 1: span 0-3 level NUMA [] groups: 0-2 (mask: 1) (cpu_capacity: 3072), 0,2-3 (cpu_capacity: 3072) (note: this relies on OVERLAP domains to always have children, this is true because the regular topology domains are still here -- this is before degenerate trimming) Debugged-by: Lauro Ramos Venancio Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: linux-kernel@vger.kernel.org Fixes: e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans") Signed-off-by: Ingo Molnar Signed-off-by: Greg Kroah-Hartman --- kernel/sched/core.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 20253dbc8610..62030b63b671 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6123,6 +6123,9 @@ enum s_alloc { * Build an iteration mask that can exclude certain CPUs from the upwards * domain traversal. * + * Only CPUs that can arrive at this group should be considered to continue + * balancing. + * * Asymmetric node setups can result in situations where the domain tree is of * unequal depth, make sure to skip domains that already cover the entire * range. @@ -6141,11 +6144,24 @@ static void build_group_mask(struct sched_domain *sd, struct sched_group *sg) for_each_cpu(i, span) { sibling = *per_cpu_ptr(sdd->sd, i); - if (!cpumask_test_cpu(i, sched_domain_span(sibling))) + + /* + * Can happen in the asymmetric case, where these siblings are + * unused. The mask will not be empty because those CPUs that + * do have the top domain _should_ span the domain. + */ + if (!sibling->child) + continue; + + /* If we would not end up here, we can't continue from here */ + if (!cpumask_equal(sg_span, sched_domain_span(sibling->child))) continue; cpumask_set_cpu(i, sched_group_mask(sg)); } + + /* We must not have empty masks here */ + WARN_ON_ONCE(cpumask_empty(sched_group_mask(sg))); } /* From 988067ec9606e3d7bd7e1125118e77e06c2eda4a Mon Sep 17 00:00:00 2001 From: Lauro Ramos Venancio Date: Thu, 20 Apr 2017 16:51:40 -0300 Subject: [PATCH 49/58] sched/topology: Optimize build_group_mask() commit f32d782e31bf079f600dcec126ed117b0577e85c upstream. The group mask is always used in intersection with the group CPUs. So, when building the group mask, we don't have to care about CPUs that are not part of the group. Signed-off-by: Lauro Ramos Venancio Signed-off-by: Peter Zijlstra (Intel) Cc: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: lwang@redhat.com Cc: riel@redhat.com Link: http://lkml.kernel.org/r/1492717903-5195-2-git-send-email-lvenanci@redhat.com Signed-off-by: Ingo Molnar Signed-off-by: Greg Kroah-Hartman --- kernel/sched/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 62030b63b671..c436426a80dd 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6137,12 +6137,12 @@ enum s_alloc { */ static void build_group_mask(struct sched_domain *sd, struct sched_group *sg) { - const struct cpumask *span = sched_domain_span(sd); + const struct cpumask *sg_span = sched_group_cpus(sg); struct sd_data *sdd = sd->private; struct sched_domain *sibling; int i; - for_each_cpu(i, span) { + for_each_cpu(i, sg_span) { sibling = *per_cpu_ptr(sdd->sd, i); /* From 275d4be9d39f91ddbaa2f42b69535683bb4d0573 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sun, 25 Jun 2017 19:31:13 +0200 Subject: [PATCH 50/58] PM / wakeirq: Convert to SRCU commit ea0212f40c6bc0594c8eff79266759e3ecd4bacc upstream. The wakeirq infrastructure uses RCU to protect the list of wakeirqs. That breaks the irq bus locking infrastructure, which is allows sleeping functions to be called so interrupt controllers behind slow busses, e.g. i2c, can be handled. The wakeirq functions hold rcu_read_lock and call into irq functions, which in case of interrupts using the irq bus locking will trigger a might_sleep() splat. Convert the wakeirq infrastructure to Sleepable RCU and unbreak it. Fixes: 4990d4fe327b (PM / Wakeirq: Add automated device wake IRQ handling) Reported-by: Brian Norris Suggested-by: Paul E. McKenney Signed-off-by: Thomas Gleixner Reviewed-by: Paul E. McKenney Tested-by: Tony Lindgren Tested-by: Brian Norris Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/wakeup.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index a1e0b9ab847a..e613633ffe9c 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -60,6 +60,8 @@ static LIST_HEAD(wakeup_sources); static DECLARE_WAIT_QUEUE_HEAD(wakeup_count_wait_queue); +DEFINE_STATIC_SRCU(wakeup_srcu); + static struct wakeup_source deleted_ws = { .name = "deleted", .lock = __SPIN_LOCK_UNLOCKED(deleted_ws.lock), @@ -198,7 +200,7 @@ void wakeup_source_remove(struct wakeup_source *ws) spin_lock_irqsave(&events_lock, flags); list_del_rcu(&ws->entry); spin_unlock_irqrestore(&events_lock, flags); - synchronize_rcu(); + synchronize_srcu(&wakeup_srcu); } EXPORT_SYMBOL_GPL(wakeup_source_remove); @@ -330,13 +332,14 @@ void device_wakeup_detach_irq(struct device *dev) void device_wakeup_arm_wake_irqs(void) { struct wakeup_source *ws; + int srcuidx; - rcu_read_lock(); + srcuidx = srcu_read_lock(&wakeup_srcu); list_for_each_entry_rcu(ws, &wakeup_sources, entry) { if (ws->wakeirq) dev_pm_arm_wake_irq(ws->wakeirq); } - rcu_read_unlock(); + srcu_read_unlock(&wakeup_srcu, srcuidx); } /** @@ -347,13 +350,14 @@ void device_wakeup_arm_wake_irqs(void) void device_wakeup_disarm_wake_irqs(void) { struct wakeup_source *ws; + int srcuidx; - rcu_read_lock(); + srcuidx = srcu_read_lock(&wakeup_srcu); list_for_each_entry_rcu(ws, &wakeup_sources, entry) { if (ws->wakeirq) dev_pm_disarm_wake_irq(ws->wakeirq); } - rcu_read_unlock(); + srcu_read_unlock(&wakeup_srcu, srcuidx); } /** @@ -807,10 +811,10 @@ EXPORT_SYMBOL_GPL(pm_wakeup_event); void pm_print_active_wakeup_sources(void) { struct wakeup_source *ws; - int active = 0; + int srcuidx, active = 0; struct wakeup_source *last_activity_ws = NULL; - rcu_read_lock(); + srcuidx = srcu_read_lock(&wakeup_srcu); list_for_each_entry_rcu(ws, &wakeup_sources, entry) { if (ws->active) { pr_info("active wakeup source: %s\n", ws->name); @@ -826,7 +830,7 @@ void pm_print_active_wakeup_sources(void) if (!active && last_activity_ws) pr_info("last active wakeup source: %s\n", last_activity_ws->name); - rcu_read_unlock(); + srcu_read_unlock(&wakeup_srcu, srcuidx); } EXPORT_SYMBOL_GPL(pm_print_active_wakeup_sources); @@ -953,8 +957,9 @@ void pm_wakep_autosleep_enabled(bool set) { struct wakeup_source *ws; ktime_t now = ktime_get(); + int srcuidx; - rcu_read_lock(); + srcuidx = srcu_read_lock(&wakeup_srcu); list_for_each_entry_rcu(ws, &wakeup_sources, entry) { spin_lock_irq(&ws->lock); if (ws->autosleep_enabled != set) { @@ -968,7 +973,7 @@ void pm_wakep_autosleep_enabled(bool set) } spin_unlock_irq(&ws->lock); } - rcu_read_unlock(); + srcu_read_unlock(&wakeup_srcu, srcuidx); } #endif /* CONFIG_PM_AUTOSLEEP */ @@ -1029,15 +1034,16 @@ static int print_wakeup_source_stats(struct seq_file *m, static int wakeup_sources_stats_show(struct seq_file *m, void *unused) { struct wakeup_source *ws; + int srcuidx; seq_puts(m, "name\t\tactive_count\tevent_count\twakeup_count\t" "expire_count\tactive_since\ttotal_time\tmax_time\t" "last_change\tprevent_suspend_time\n"); - rcu_read_lock(); + srcuidx = srcu_read_lock(&wakeup_srcu); list_for_each_entry_rcu(ws, &wakeup_sources, entry) print_wakeup_source_stats(m, ws); - rcu_read_unlock(); + srcu_read_unlock(&wakeup_srcu, srcuidx); print_wakeup_source_stats(m, &deleted_ws); From c17f6512463e0675fb066affc41bc0b1c6dfb60b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 10 Jul 2017 10:21:40 +0300 Subject: [PATCH 51/58] PM / QoS: return -EINVAL for bogus strings commit 2ca30331c156ca9e97643ad05dd8930b8fe78b01 upstream. In the current code, if the user accidentally writes a bogus command to this sysfs file, then we set the latency tolerance to an uninitialized variable. Fixes: 2d984ad132a8 (PM / QoS: Introcuce latency tolerance device PM QoS type) Signed-off-by: Dan Carpenter Acked-by: Pavel Machek Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/sysfs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index a7b46798c81d..39efa7e6c0c0 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -268,6 +268,8 @@ static ssize_t pm_qos_latency_tolerance_store(struct device *dev, value = PM_QOS_LATENCY_TOLERANCE_NO_CONSTRAINT; else if (!strcmp(buf, "any") || !strcmp(buf, "any\n")) value = PM_QOS_LATENCY_ANY; + else + return -EINVAL; } ret = dev_pm_qos_update_user_latency_tolerance(dev, value); return ret < 0 ? ret : n; From 999b96b4de81a984bc94302b9ab7cad9f45eb6df Mon Sep 17 00:00:00 2001 From: Pavankumar Kondeti Date: Fri, 9 Dec 2016 21:50:17 +0530 Subject: [PATCH 52/58] tracing: Use SOFTIRQ_OFFSET for softirq dectection for more accurate results commit c59f29cb144a6a0dfac16ede9dc8eafc02dc56ca upstream. The 's' flag is supposed to indicate that a softirq is running. This can be detected by testing the preempt_count with SOFTIRQ_OFFSET. The current code tests the preempt_count with SOFTIRQ_MASK, which would be true even when softirqs are disabled but not serving a softirq. Link: http://lkml.kernel.org/r/1481300417-3564-1-git-send-email-pkondeti@codeaurora.org Signed-off-by: Pavankumar Kondeti Signed-off-by: Steven Rostedt Signed-off-by: Amit Pundir Signed-off-by: Greg Kroah-Hartman --- kernel/trace/trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 4c21c0b7dc91..c83d59913d78 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1660,7 +1660,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags, TRACE_FLAG_IRQS_NOSUPPORT | #endif ((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) | - ((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) | + ((pc & SOFTIRQ_OFFSET) ? TRACE_FLAG_SOFTIRQ : 0) | (tif_need_resched() ? TRACE_FLAG_NEED_RESCHED : 0) | (test_preempt_need_resched() ? TRACE_FLAG_PREEMPT_RESCHED : 0); } From c6f3576ed0f21058629d5461eb014c1e0e3f7c4d Mon Sep 17 00:00:00 2001 From: Paolo Bonzini Date: Tue, 8 Mar 2016 09:52:13 +0100 Subject: [PATCH 53/58] KVM: x86: disable MPX if host did not enable MPX XSAVE features commit a87036add09283e6c4f4103a15c596c67b86ab86 upstream. When eager FPU is disabled, KVM will still see the MPX bit in CPUID and presumably the MPX vmentry and vmexit controls. However, it will not be able to expose the MPX XSAVE features to the guest, because the guest's accessible XSAVE features are always a subset of host_xcr0. In this case, we should disable the MPX CPUID bit, the BNDCFGS MSR, and the MPX vmentry and vmexit controls for nested virtualization. It is then unnecessary to enable guest eager FPU if the guest has the MPX CPUID bit set. Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman --- arch/x86/kvm/cpuid.c | 13 ++++++++++--- arch/x86/kvm/cpuid.h | 9 +-------- arch/x86/kvm/vmx.c | 13 ++++++------- 3 files changed, 17 insertions(+), 18 deletions(-) diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 9357b29de9bc..83d6369c45f5 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -46,11 +46,18 @@ static u32 xstate_required_size(u64 xstate_bv, bool compacted) return ret; } +bool kvm_mpx_supported(void) +{ + return ((host_xcr0 & (XFEATURE_MASK_BNDREGS | XFEATURE_MASK_BNDCSR)) + && kvm_x86_ops->mpx_supported()); +} +EXPORT_SYMBOL_GPL(kvm_mpx_supported); + u64 kvm_supported_xcr0(void) { u64 xcr0 = KVM_SUPPORTED_XCR0 & host_xcr0; - if (!kvm_x86_ops->mpx_supported()) + if (!kvm_mpx_supported()) xcr0 &= ~(XFEATURE_MASK_BNDREGS | XFEATURE_MASK_BNDCSR); return xcr0; @@ -97,7 +104,7 @@ int kvm_update_cpuid(struct kvm_vcpu *vcpu) if (best && (best->eax & (F(XSAVES) | F(XSAVEC)))) best->ebx = xstate_required_size(vcpu->arch.xcr0, true); - vcpu->arch.eager_fpu = use_eager_fpu() || guest_cpuid_has_mpx(vcpu); + vcpu->arch.eager_fpu = use_eager_fpu(); if (vcpu->arch.eager_fpu) kvm_x86_ops->fpu_activate(vcpu); @@ -295,7 +302,7 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, #endif unsigned f_rdtscp = kvm_x86_ops->rdtscp_supported() ? F(RDTSCP) : 0; unsigned f_invpcid = kvm_x86_ops->invpcid_supported() ? F(INVPCID) : 0; - unsigned f_mpx = kvm_x86_ops->mpx_supported() ? F(MPX) : 0; + unsigned f_mpx = kvm_mpx_supported() ? F(MPX) : 0; unsigned f_xsaves = kvm_x86_ops->xsaves_supported() ? F(XSAVES) : 0; /* cpuid 1.edx */ diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h index 3f5c48ddba45..33d574fb2445 100644 --- a/arch/x86/kvm/cpuid.h +++ b/arch/x86/kvm/cpuid.h @@ -4,6 +4,7 @@ #include "x86.h" int kvm_update_cpuid(struct kvm_vcpu *vcpu); +bool kvm_mpx_supported(void); struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, u32 function, u32 index); int kvm_dev_ioctl_get_cpuid(struct kvm_cpuid2 *cpuid, @@ -134,14 +135,6 @@ static inline bool guest_cpuid_has_rtm(struct kvm_vcpu *vcpu) return best && (best->ebx & bit(X86_FEATURE_RTM)); } -static inline bool guest_cpuid_has_mpx(struct kvm_vcpu *vcpu) -{ - struct kvm_cpuid_entry2 *best; - - best = kvm_find_cpuid_entry(vcpu, 7, 0); - return best && (best->ebx & bit(X86_FEATURE_MPX)); -} - static inline bool guest_cpuid_has_pcommit(struct kvm_vcpu *vcpu) { struct kvm_cpuid_entry2 *best; diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index bbaa11f4e74b..5579ae069dee 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -863,7 +863,6 @@ static unsigned long nested_ept_get_cr3(struct kvm_vcpu *vcpu); static u64 construct_eptp(unsigned long root_hpa); static void kvm_cpu_vmxon(u64 addr); static void kvm_cpu_vmxoff(void); -static bool vmx_mpx_supported(void); static bool vmx_xsaves_supported(void); static int vmx_cpu_uses_apicv(struct kvm_vcpu *vcpu); static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr); @@ -2541,7 +2540,7 @@ static void nested_vmx_setup_ctls_msrs(struct vcpu_vmx *vmx) VM_EXIT_LOAD_IA32_EFER | VM_EXIT_SAVE_IA32_EFER | VM_EXIT_SAVE_VMX_PREEMPTION_TIMER | VM_EXIT_ACK_INTR_ON_EXIT; - if (vmx_mpx_supported()) + if (kvm_mpx_supported()) vmx->nested.nested_vmx_exit_ctls_high |= VM_EXIT_CLEAR_BNDCFGS; /* We support free control of debug control saving. */ @@ -2562,7 +2561,7 @@ static void nested_vmx_setup_ctls_msrs(struct vcpu_vmx *vmx) VM_ENTRY_LOAD_IA32_PAT; vmx->nested.nested_vmx_entry_ctls_high |= (VM_ENTRY_ALWAYSON_WITHOUT_TRUE_MSR | VM_ENTRY_LOAD_IA32_EFER); - if (vmx_mpx_supported()) + if (kvm_mpx_supported()) vmx->nested.nested_vmx_entry_ctls_high |= VM_ENTRY_LOAD_BNDCFGS; /* We support free control of debug control loading. */ @@ -2813,7 +2812,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = vmcs_readl(GUEST_SYSENTER_ESP); break; case MSR_IA32_BNDCFGS: - if (!vmx_mpx_supported()) + if (!kvm_mpx_supported()) return 1; msr_info->data = vmcs_read64(GUEST_BNDCFGS); break; @@ -2890,7 +2889,7 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) vmcs_writel(GUEST_SYSENTER_ESP, data); break; case MSR_IA32_BNDCFGS: - if (!vmx_mpx_supported()) + if (!kvm_mpx_supported()) return 1; vmcs_write64(GUEST_BNDCFGS, data); break; @@ -3363,7 +3362,7 @@ static void init_vmcs_shadow_fields(void) for (i = j = 0; i < max_shadow_read_write_fields; i++) { switch (shadow_read_write_fields[i]) { case GUEST_BNDCFGS: - if (!vmx_mpx_supported()) + if (!kvm_mpx_supported()) continue; break; default: @@ -10265,7 +10264,7 @@ static void prepare_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12, vmcs12->guest_sysenter_cs = vmcs_read32(GUEST_SYSENTER_CS); vmcs12->guest_sysenter_esp = vmcs_readl(GUEST_SYSENTER_ESP); vmcs12->guest_sysenter_eip = vmcs_readl(GUEST_SYSENTER_EIP); - if (vmx_mpx_supported()) + if (kvm_mpx_supported()) vmcs12->guest_bndcfgs = vmcs_read64(GUEST_BNDCFGS); if (nested_cpu_has_xsaves(vmcs12)) vmcs12->xss_exit_bitmap = vmcs_read64(XSS_EXIT_BITMAP); From be6f33a61f335cb06b3c8c2cee90214f97c4bda3 Mon Sep 17 00:00:00 2001 From: Jim Mattson Date: Tue, 23 May 2017 11:52:52 -0700 Subject: [PATCH 54/58] kvm: vmx: Do not disable intercepts for BNDCFGS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit a8b6fda38f80e75afa3b125c9e7f2550b579454b upstream. The MSR permission bitmaps are shared by all VMs. However, some VMs may not be configured to support MPX, even when the host does. If the host supports VMX and the guest does not, we should intercept accesses to the BNDCFGS MSR, so that we can synthesize a #GP fault. Furthermore, if the host does not support MPX and the "ignore_msrs" kvm kernel parameter is set, then we should intercept accesses to the BNDCFGS MSR, so that we can skip over the rdmsr/wrmsr without raising a #GP fault. Fixes: da8999d31818fdc8 ("KVM: x86: Intel MPX vmx and msr handle") Signed-off-by: Jim Mattson Signed-off-by: Radim Krčmář Signed-off-by: Greg Kroah-Hartman --- arch/x86/kvm/vmx.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 5579ae069dee..0e0c52e57e90 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -6252,7 +6252,6 @@ static __init int hardware_setup(void) vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_CS, false); vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_ESP, false); vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_EIP, false); - vmx_disable_intercept_for_msr(MSR_IA32_BNDCFGS, true); memcpy(vmx_msr_bitmap_legacy_x2apic, vmx_msr_bitmap_legacy, PAGE_SIZE); From a40f0ccd4f1c2751d5df5d9a63b9cf93ff70d343 Mon Sep 17 00:00:00 2001 From: Jim Mattson Date: Wed, 24 May 2017 10:49:25 -0700 Subject: [PATCH 55/58] kvm: x86: Guest BNDCFGS requires guest MPX support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 4439af9f911ae0243ffe4e2dfc12bace49605d8b upstream. The BNDCFGS MSR should only be exposed to the guest if the guest supports MPX. (cf. the TSC_AUX MSR and RDTSCP.) Fixes: 0dd376e709975779 ("KVM: x86: add MSR_IA32_BNDCFGS to msrs_to_save") Change-Id: I3ad7c01bda616715137ceac878f3fa7e66b6b387 Signed-off-by: Jim Mattson Signed-off-by: Radim Krčmář Signed-off-by: Greg Kroah-Hartman --- arch/x86/kvm/cpuid.h | 8 ++++++++ arch/x86/kvm/vmx.c | 4 ++-- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h index 33d574fb2445..d1534feefcfe 100644 --- a/arch/x86/kvm/cpuid.h +++ b/arch/x86/kvm/cpuid.h @@ -143,6 +143,14 @@ static inline bool guest_cpuid_has_pcommit(struct kvm_vcpu *vcpu) return best && (best->ebx & bit(X86_FEATURE_PCOMMIT)); } +static inline bool guest_cpuid_has_mpx(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 7, 0); + return best && (best->ebx & bit(X86_FEATURE_MPX)); +} + static inline bool guest_cpuid_has_rdtscp(struct kvm_vcpu *vcpu) { struct kvm_cpuid_entry2 *best; diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 0e0c52e57e90..3d776b625766 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2812,7 +2812,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = vmcs_readl(GUEST_SYSENTER_ESP); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported()) + if (!kvm_mpx_supported() || !guest_cpuid_has_mpx(vcpu)) return 1; msr_info->data = vmcs_read64(GUEST_BNDCFGS); break; @@ -2889,7 +2889,7 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) vmcs_writel(GUEST_SYSENTER_ESP, data); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported()) + if (!kvm_mpx_supported() || !guest_cpuid_has_mpx(vcpu)) return 1; vmcs_write64(GUEST_BNDCFGS, data); break; From 6ffa92ffaed817cea7625ebf9c7f217e293923be Mon Sep 17 00:00:00 2001 From: Jim Mattson Date: Tue, 23 May 2017 11:52:54 -0700 Subject: [PATCH 56/58] kvm: vmx: Check value written to IA32_BNDCFGS MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 4531662d1abf6c1f0e5c2b86ddb60e61509786c8 upstream. Bits 11:2 must be zero and the linear addess in bits 63:12 must be canonical. Otherwise, WRMSR(BNDCFGS) should raise #GP. Fixes: 0dd376e709975779 ("KVM: x86: add MSR_IA32_BNDCFGS to msrs_to_save") Signed-off-by: Jim Mattson Signed-off-by: Radim Krčmář Signed-off-by: Greg Kroah-Hartman --- arch/x86/include/asm/msr-index.h | 2 ++ arch/x86/kvm/vmx.c | 3 +++ 2 files changed, 5 insertions(+) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 690b4027e17c..37db36fddc88 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -405,6 +405,8 @@ #define MSR_IA32_TSC_ADJUST 0x0000003b #define MSR_IA32_BNDCFGS 0x00000d90 +#define MSR_IA32_BNDCFGS_RSVD 0x00000ffc + #define MSR_IA32_XSS 0x00000da0 #define FEATURE_CONTROL_LOCKED (1<<0) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 3d776b625766..a2dd5a5ee472 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2891,6 +2891,9 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) case MSR_IA32_BNDCFGS: if (!kvm_mpx_supported() || !guest_cpuid_has_mpx(vcpu)) return 1; + if (is_noncanonical_address(data & PAGE_MASK) || + (data & MSR_IA32_BNDCFGS_RSVD)) + return 1; vmcs_write64(GUEST_BNDCFGS, data); break; case MSR_IA32_TSC: From 3ed43caedf69a6fd417bdda174eb8fa06b3b5dab Mon Sep 17 00:00:00 2001 From: Haozhong Zhang Date: Tue, 4 Jul 2017 10:27:41 +0800 Subject: [PATCH 57/58] kvm: vmx: allow host to access guest MSR_IA32_BNDCFGS commit 691bd4340bef49cf7e5855d06cf24444b5bf2d85 upstream. It's easier for host applications, such as QEMU, if they can always access guest MSR_IA32_BNDCFGS in VMCS, even though MPX is disabled in guest cpuid. Signed-off-by: Haozhong Zhang Signed-off-by: Paolo Bonzini Signed-off-by: Greg Kroah-Hartman --- arch/x86/kvm/vmx.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index a2dd5a5ee472..b12391119ce8 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2812,7 +2812,8 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = vmcs_readl(GUEST_SYSENTER_ESP); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported() || !guest_cpuid_has_mpx(vcpu)) + if (!kvm_mpx_supported() || + (!msr_info->host_initiated && !guest_cpuid_has_mpx(vcpu))) return 1; msr_info->data = vmcs_read64(GUEST_BNDCFGS); break; @@ -2889,7 +2890,8 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) vmcs_writel(GUEST_SYSENTER_ESP, data); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported() || !guest_cpuid_has_mpx(vcpu)) + if (!kvm_mpx_supported() || + (!msr_info->host_initiated && !guest_cpuid_has_mpx(vcpu))) return 1; if (is_noncanonical_address(data & PAGE_MASK) || (data & MSR_IA32_BNDCFGS_RSVD)) From ece78cd779f4f306a869f217bc0236507b0fe5f5 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 21 Jul 2017 07:45:10 +0200 Subject: [PATCH 58/58] Linux 4.4.78 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index bf49a61d02e2..ac77ae8ee0b1 100644 --- a/Makefile +++ b/Makefile @@ -1,6 +1,6 @@ VERSION = 4 PATCHLEVEL = 4 -SUBLEVEL = 77 +SUBLEVEL = 78 EXTRAVERSION = NAME = Blurry Fish Butt