April 2019 - Linux-stable-mirror

FAILED: patch "[PATCH] cifs: fix kref underflow in close_shroot()" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 2f94a3125b8742b05a011d62b16f52eb8f9ebe1c Mon Sep 17 00:00:00 2001 From: Ronnie Sahlberg <lsahlber(a)redhat.com> Date: Thu, 28 Mar 2019 11:20:02 +1000 Subject: [PATCH] cifs: fix kref underflow in close_shroot() Fix a bug where we used to not initialize the cached fid structure at all in open_shroot() if the open was successful but we did not get a lease. This would leave the structure uninitialized and later when we close the handle we would in close_shroot() try to kref_put() an uninitialized refcount. Fix this by always initializing this structure if the open was successful but only do the extra get() if we got a lease. This extra get() is only used to hold the structure until we get a lease break from the server at which point we will kref_put() it during lease processing. Signed-off-by: Ronnie Sahlberg <lsahlber(a)redhat.com> Signed-off-by: Steve French <stfrench(a)microsoft.com> CC: Stable <stable(a)vger.kernel.org> diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 1022a3771e14..7cfafac255aa 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -717,20 +717,18 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid) oparms.fid->mid = le64_to_cpu(o_rsp->sync_hdr.MessageId); #endif /* CIFS_DEBUG2 */ - if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) - oplock = smb2_parse_lease_state(server, o_rsp, - &oparms.fid->epoch, - oparms.fid->lease_key); - else - goto oshr_exit; - - memcpy(tcon->crfid.fid, pfid, sizeof(struct cifs_fid)); tcon->crfid.tcon = tcon; tcon->crfid.is_valid = true; kref_init(&tcon->crfid.refcount); - kref_get(&tcon->crfid.refcount); + if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) { + kref_get(&tcon->crfid.refcount); + oplock = smb2_parse_lease_state(server, o_rsp, + &oparms.fid->epoch, + oparms.fid->lease_key); + } else + goto oshr_exit; qi_rsp = (struct smb2_query_info_rsp *)rsp_iov[1].iov_base; if (le32_to_cpu(qi_rsp->OutputBufferLength) < sizeof(struct smb2_file_all_info))

6 years

1
0
0 0

FAILED: patch "[PATCH] cifs: fix kref underflow in close_shroot()" failed to apply to 4.19-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.19-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 2f94a3125b8742b05a011d62b16f52eb8f9ebe1c Mon Sep 17 00:00:00 2001 From: Ronnie Sahlberg <lsahlber(a)redhat.com> Date: Thu, 28 Mar 2019 11:20:02 +1000 Subject: [PATCH] cifs: fix kref underflow in close_shroot() Fix a bug where we used to not initialize the cached fid structure at all in open_shroot() if the open was successful but we did not get a lease. This would leave the structure uninitialized and later when we close the handle we would in close_shroot() try to kref_put() an uninitialized refcount. Fix this by always initializing this structure if the open was successful but only do the extra get() if we got a lease. This extra get() is only used to hold the structure until we get a lease break from the server at which point we will kref_put() it during lease processing. Signed-off-by: Ronnie Sahlberg <lsahlber(a)redhat.com> Signed-off-by: Steve French <stfrench(a)microsoft.com> CC: Stable <stable(a)vger.kernel.org> diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 1022a3771e14..7cfafac255aa 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -717,20 +717,18 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid) oparms.fid->mid = le64_to_cpu(o_rsp->sync_hdr.MessageId); #endif /* CIFS_DEBUG2 */ - if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) - oplock = smb2_parse_lease_state(server, o_rsp, - &oparms.fid->epoch, - oparms.fid->lease_key); - else - goto oshr_exit; - - memcpy(tcon->crfid.fid, pfid, sizeof(struct cifs_fid)); tcon->crfid.tcon = tcon; tcon->crfid.is_valid = true; kref_init(&tcon->crfid.refcount); - kref_get(&tcon->crfid.refcount); + if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) { + kref_get(&tcon->crfid.refcount); + oplock = smb2_parse_lease_state(server, o_rsp, + &oparms.fid->epoch, + oparms.fid->lease_key); + } else + goto oshr_exit; qi_rsp = (struct smb2_query_info_rsp *)rsp_iov[1].iov_base; if (le32_to_cpu(qi_rsp->OutputBufferLength) < sizeof(struct smb2_file_all_info))

6 years

1
0
0 0

FAILED: patch "[PATCH] cifs: fix kref underflow in close_shroot()" failed to apply to 5.0-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 5.0-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 2f94a3125b8742b05a011d62b16f52eb8f9ebe1c Mon Sep 17 00:00:00 2001 From: Ronnie Sahlberg <lsahlber(a)redhat.com> Date: Thu, 28 Mar 2019 11:20:02 +1000 Subject: [PATCH] cifs: fix kref underflow in close_shroot() Fix a bug where we used to not initialize the cached fid structure at all in open_shroot() if the open was successful but we did not get a lease. This would leave the structure uninitialized and later when we close the handle we would in close_shroot() try to kref_put() an uninitialized refcount. Fix this by always initializing this structure if the open was successful but only do the extra get() if we got a lease. This extra get() is only used to hold the structure until we get a lease break from the server at which point we will kref_put() it during lease processing. Signed-off-by: Ronnie Sahlberg <lsahlber(a)redhat.com> Signed-off-by: Steve French <stfrench(a)microsoft.com> CC: Stable <stable(a)vger.kernel.org> diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 1022a3771e14..7cfafac255aa 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -717,20 +717,18 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, struct cifs_fid *pfid) oparms.fid->mid = le64_to_cpu(o_rsp->sync_hdr.MessageId); #endif /* CIFS_DEBUG2 */ - if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) - oplock = smb2_parse_lease_state(server, o_rsp, - &oparms.fid->epoch, - oparms.fid->lease_key); - else - goto oshr_exit; - - memcpy(tcon->crfid.fid, pfid, sizeof(struct cifs_fid)); tcon->crfid.tcon = tcon; tcon->crfid.is_valid = true; kref_init(&tcon->crfid.refcount); - kref_get(&tcon->crfid.refcount); + if (o_rsp->OplockLevel == SMB2_OPLOCK_LEVEL_LEASE) { + kref_get(&tcon->crfid.refcount); + oplock = smb2_parse_lease_state(server, o_rsp, + &oparms.fid->epoch, + oparms.fid->lease_key); + } else + goto oshr_exit; qi_rsp = (struct smb2_query_info_rsp *)rsp_iov[1].iov_base; if (le32_to_cpu(qi_rsp->OutputBufferLength) < sizeof(struct smb2_file_all_info))

6 years

1
0
0 0

[PATCH stable 4.9 v2] net: stmmac: Set dma ring length before enabling the DMA

by Lars Persson

This was fixed in upstream by commit 7d9e6c5afab6 ("net: stmmac: Integrate XGMAC into main driver flow") that is a new feature commit. We found a race condition in the DMA init sequence that hits if the PHY already has link up during stmmac_hw_setup. Since the ring length was programmed after enabling the RX path, we might receive a packet before the correct ring length is programmed. When that happened we could not get reliable interrupts for DMA RX and the MTL complained about RX FIFO overrun. Signed-off-by: Lars Persson <larper(a)axis.com> Cc: stable(a)vger.kernel.org # 4.9.x Cc: Giuseppe Cavallaro <peppe.cavallaro(a)st.com> Cc: Alexandre Torgue <alexandre.torgue(a)st.com> Cc: Jose Abreu <joabreu(a)synopsys.com> --- v2: Dropped changes not present in 7d9e6c5afab6. A patch for strict data book complicance is net-next material. --- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index fc437d75ac76..b46b56ad7517 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1747,11 +1747,6 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) if (ret < 0) pr_warn("%s: failed debugFS registration\n", __func__); #endif - /* Start the ball rolling... */ - pr_debug("%s: DMA RX/TX processes started...\n", dev->name); - priv->hw->dma->start_tx(priv->ioaddr); - priv->hw->dma->start_rx(priv->ioaddr); - /* Dump DMA/MAC registers */ if (netif_msg_hw(priv)) { priv->hw->mac->dump_regs(priv->hw); @@ -1779,6 +1774,11 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) if (priv->tso) priv->hw->dma->enable_tso(priv->ioaddr, 1, STMMAC_CHAN0); + /* Start the ball rolling... */ + pr_debug("%s: DMA RX/TX processes started...\n", dev->name); + priv->hw->dma->start_tx(priv->ioaddr); + priv->hw->dma->start_rx(priv->ioaddr); + return 0; } -- 2.11.0

6 years

1
0
0 0

[PATCH -tip] kprobes: Fix an inverted result check for reusing optimized probe

by Masami Hiramatsu

Fix an inverted result check for reusing unused kprobe correctly. This has been introduced by commit 819319fc9346 ("kprobes: Return error if we fail to reuse kprobe instead of BUG_ON()"), which missed to handle the return value of kprobe_optready() as error-value. In reality, the kprobe_optready() returns a bool result, so "true" case must be passed instead of 0. This causes some errors on kprobe boot-time selftests on arm. [ 4.563544] Beginning kprobe tests... [ 4.563648] Probe ARM code [ 4.563733] kprobe [ 4.564700] kretprobe [ 4.565538] ARM instruction simulation [ 4.565671] Check decoding tables [ 4.565883] Run test cases [ 5.070700] FAIL: test_case_handler not run [ 5.070938] FAIL: Test andge r10, r11, r14, asr r7 [ 5.071118] FAIL: Scenario 11 ... [ 74.174729] FAIL: Scenario 7 [ 74.211776] Total instruction simulation tests=1631, pass=1433 fail=198 [ 74.212168] kprobe tests failed This can happen if an optimized probe is unregistered and next kprobe is registered on same address until the previous probe is not reclaimed. If this happens, a hidden aggregated probe may be kept in memory, and no new kprobe can probe same address. Also, in that case register_kprobe() will return "1" instead of minus error value, which can mislead caller logic. Signed-off-by: Masami Hiramatsu <mhiramat(a)kernel.org> Fixes: 819319fc9346 ("kprobes: Return error if we fail to reuse kprobe instead of BUG_ON()") Cc: stable(a)vger.kernel.org # v5.0+ --- kernel/kprobes.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/kernel/kprobes.c b/kernel/kprobes.c index c83e54727131..b1ea30a5540e 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -709,7 +709,6 @@ static void unoptimize_kprobe(struct kprobe *p, bool force) static int reuse_unused_kprobe(struct kprobe *ap) { struct optimized_kprobe *op; - int ret; /* * Unused kprobe MUST be on the way of delayed unoptimizing (means @@ -720,9 +719,8 @@ static int reuse_unused_kprobe(struct kprobe *ap) /* Enable the probe again */ ap->flags &= ~KPROBE_FLAG_DISABLED; /* Optimize it again (remove from op->list) */ - ret = kprobe_optready(ap); - if (ret) - return ret; + if (!kprobe_optready(ap)) + return -EINVAL; optimize_kprobe(ap); return 0;

6 years

1
0
0 0

Re: [PATCH] arm64: kaslr: Reserve size of ARM64_MEMSTART_ALIGN in linear region

by Nicolas Boichat

Dear stable maintainers, I encountered a similar issue on a 4.19.33 kernel (Chromium OS). On my board, the system would not even be able to boot if KASLR decides to map the linear region to the top of the virtual address space. This happens every 253 boots on average (there are 0xfd possible random offsets, and only the top one fails). I tried to debug the issue, and it appears physical memory allocated for vmemmap and mem_section array would end up at the same location, corrupting each other early on boot. I could not figure out exactly why this is happening, but in any case, this patch fixes my issue (no failure in 744 reboots with 240 unique offsets, and counting...), and IMHO the ERR_PTR justification in the commit message is enough to warrant inclusion in -stable branches. The patch below was committed to mainline as: commit c8a43c18a97845e7f94ed7d181c11f41964976a2 arm64: kaslr: Reserve size of ARM64_MEMSTART_ALIGN in linear region and should be included in stable branches after this commit: Fixes: c031a4213c11a5db ("arm64: kaslr: randomize the linear region") i.e. anything after kernel 4.5 (git describe says v4.5-rc4-62-gc031a4213c11a5d). Thanks, Nicolas On Wed, Jan 16, 2019 at 4:38 PM Yueyi Li <liyueyi(a)live.com> wrote: > > > > On 2019/1/16 15:51, Ard Biesheuvel wrote: > > On Wed, 16 Jan 2019 at 04:37, Yueyi Li <liyueyi(a)live.com> wrote: > >> OK, thanks. But seems this mail be ignored, do i need re-sent the patch? > >> > >> On 2018/12/26 21:49, Ard Biesheuvel wrote: > >>> On Tue, 25 Dec 2018 at 03:30, Yueyi Li <liyueyi(a)live.com> wrote: > >>>> Hi Ard, > >>>> > >>>> > >>>> On 2018/12/24 17:45, Ard Biesheuvel wrote: > >>>>> Does the following change fix your issue as well? > >>>>> > >>>>> index 9b432d9fcada..9dcf0ff75a11 100644 > >>>>> --- a/arch/arm64/mm/init.c > >>>>> +++ b/arch/arm64/mm/init.c > >>>>> @@ -447,7 +447,7 @@ void __init arm64_memblock_init(void) > >>>>> * memory spans, randomize the linear region as well. > >>>>> */ > >>>>> if (memstart_offset_seed > 0 && range >= ARM64_MEMSTART_ALIGN) { > >>>>> - range = range / ARM64_MEMSTART_ALIGN + 1; > >>>>> + range /= ARM64_MEMSTART_ALIGN; > >>>>> memstart_addr -= ARM64_MEMSTART_ALIGN * > >>>>> ((range * memstart_offset_seed) >> 16); > >>>>> } > >>>> Yes, it can fix this also. I just think modify the first *range* > >>>> calculation would be easier to grasp, what do you think? > >>>> > >>> I don't think there is a difference, to be honest, but I will leave it > >>> up to the maintainers to decide which approach they prefer. > > No it has been merged already. It is in v5.0-rc2 I think. > > OK, thanks. :-)

6 years

2
1
0 0

[PATCH] vmbus: Remove the undesired put_cpu_ptr() in hv_synic_cleanup()

by Dexuan Cui

With CONFIG_DEBUG_PREEMPT=y, the put_cpu_ptr() triggiers an underflow warning in preempt_count_sub(). Fixes: 37cdd991fac8 ("vmbus: put related per-cpu variable together") Cc: stable(a)vger.kernel.org Cc: Stephen Hemminger <sthemmin(a)microsoft.com> Signed-off-by: Dexuan Cui <decui(a)microsoft.com> --- drivers/hv/hv.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c index 632d25674e7f..45653029ee18 100644 --- a/drivers/hv/hv.c +++ b/drivers/hv/hv.c @@ -408,7 +408,6 @@ int hv_synic_cleanup(unsigned int cpu) clockevents_unbind_device(hv_cpu->clk_evt, cpu); hv_ce_shutdown(hv_cpu->clk_evt); - put_cpu_ptr(hv_cpu); } hv_get_synint_state(VMBUS_MESSAGE_SINT, shared_sint.as_uint64); -- 2.19.1

6 years

3
2
0 0

[PATCH v4.19.y 1/2] netfilter: nfnetlink_cttimeout: pass default timeout policy to obj_to_nlattr

by Zubin Mithra

From: Pablo Neira Ayuso <pablo(a)netfilter.org> commit 8866df9264a34e675b4ee8a151db819b87cce2d3 upstream Otherwise, we hit a NULL pointer deference since handlers always assume default timeout policy is passed. netlink: 24 bytes leftover after parsing attributes in process `syz-executor2'. kasan: CONFIG_KASAN_INLINE enabled kasan: GPF could be caused by NULL-ptr deref or user memory access general protection fault: 0000 [#1] PREEMPT SMP KASAN CPU: 0 PID: 9575 Comm: syz-executor1 Not tainted 4.19.0+ #312 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:icmp_timeout_obj_to_nlattr+0x77/0x170 net/netfilter/nf_conntrack_proto_icmp.c:297 Fixes: c779e849608a ("netfilter: conntrack: remove get_timeout() indirection") Reported-by: Eric Dumazet <eric.dumazet(a)gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Signed-off-by: Zubin Mithra <zsm(a)chromium.org> --- Notes: * Syzkaller reported a general protection fault in icmp_timeout_obj_to_nlattri when fuzzing a 4.19 kernel. Call Trace: cttimeout_default_fill_info net/netfilter/nfnetlink_cttimeout.c:424 [inline] cttimeout_default_get+0x574/0x766 net/netfilter/nfnetlink_cttimeout.c:471 nfnetlink_rcv_msg+0x544/0x5bd net/netfilter/nfnetlink.c:228 netlink_rcv_skb+0x1cf/0x2a1 net/netlink/af_netlink.c:2454 nfnetlink_rcv+0x30f/0x34a net/netfilter/nfnetlink.c:560 netlink_unicast_kernel net/netlink/af_netlink.c:1317 [inline] netlink_unicast+0x3c5/0x4c3 net/netlink/af_netlink.c:1343 netlink_sendmsg+0x829/0x88e net/netlink/af_netlink.c:1908 sock_sendmsg_nosec+0x8c/0xad net/socket.c:621 sock_sendmsg+0x4f/0x60 net/socket.c:631 ___sys_sendmsg+0x4d6/0x63f net/socket.c:2116 __sys_sendmsg+0xee/0x17b net/socket.c:2154 do_syscall_64+0xf8/0x133 arch/x86/entry/common.c:291 entry_SYSCALL_64_after_hwframe+0x49/0xbe * This patch resolves the following conflicts when applying to 4.19: - l3num is not used as an argument for cttimeout_default_fill_info() in 4.19 - nf_icmp_pernet, nf_tcp_pernet, nf_udp_pernet, nf_dccp_pernet, nf_icmpv6_pernet, nf_sctp_pernet, nf_generic_pernet do not exist in 4.19. Expand the usage of those functions into one-liners. * Tests run: Chrome OS tryjobs, Syzkaller reproducer net/netfilter/nfnetlink_cttimeout.c | 46 +++++++++++++++++++++++++---- 1 file changed, 40 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c index a30f8ba4b89a..1dc4ea327cbe 100644 --- a/net/netfilter/nfnetlink_cttimeout.c +++ b/net/netfilter/nfnetlink_cttimeout.c @@ -392,7 +392,8 @@ static int cttimeout_default_set(struct net *net, struct sock *ctnl, static int cttimeout_default_fill_info(struct net *net, struct sk_buff *skb, u32 portid, u32 seq, u32 type, int event, - const struct nf_conntrack_l4proto *l4proto) + const struct nf_conntrack_l4proto *l4proto, + const unsigned int *timeouts) { struct nlmsghdr *nlh; struct nfgenmsg *nfmsg; @@ -421,7 +422,7 @@ cttimeout_default_fill_info(struct net *net, struct sk_buff *skb, u32 portid, if (!nest_parms) goto nla_put_failure; - ret = l4proto->ctnl_timeout.obj_to_nlattr(skb, NULL); + ret = l4proto->ctnl_timeout.obj_to_nlattr(skb, timeouts); if (ret < 0) goto nla_put_failure; @@ -444,6 +445,7 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl, struct netlink_ext_ack *extack) { const struct nf_conntrack_l4proto *l4proto; + unsigned int *timeouts = NULL; struct sk_buff *skb2; int ret, err; __u16 l3num; @@ -456,12 +458,44 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl, l4num = nla_get_u8(cda[CTA_TIMEOUT_L4PROTO]); l4proto = nf_ct_l4proto_find_get(l3num, l4num); - /* This protocol is not supported, skip. */ - if (l4proto->l4proto != l4num) { - err = -EOPNOTSUPP; + err = -EOPNOTSUPP; + if (l4proto->l4proto != l4num) goto err; + + switch (l4proto->l4proto) { + case IPPROTO_ICMP: + timeouts = &net->ct.nf_ct_proto.icmp.timeout; + break; + case IPPROTO_TCP: + timeouts = net->ct.nf_ct_proto.tcp.timeouts; + break; + case IPPROTO_UDP: + timeouts = net->ct.nf_ct_proto.udp.timeouts; + break; + case IPPROTO_DCCP: +#ifdef CONFIG_NF_CT_PROTO_DCCP + timeouts = net->ct.nf_ct_proto.dccp.dccp_timeout; +#endif + break; + case IPPROTO_ICMPV6: + timeouts = &net->ct.nf_ct_proto.icmpv6.timeout; + break; + case IPPROTO_SCTP: +#ifdef CONFIG_NF_CT_PROTO_SCTP + timeouts = net->ct.nf_ct_proto.sctp.timeouts; +#endif + break; + case 255: + timeouts = &net->ct.nf_ct_proto.generic.timeout; + break; + default: + WARN_ON_ONCE(1); + break; } + if (!timeouts) + goto err; + skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (skb2 == NULL) { err = -ENOMEM; @@ -472,7 +506,7 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl, nlh->nlmsg_seq, NFNL_MSG_TYPE(nlh->nlmsg_type), IPCTNL_MSG_TIMEOUT_DEFAULT_SET, - l4proto); + l4proto, timeouts); if (ret <= 0) { kfree_skb(skb2); err = -ENOMEM; -- 2.21.0.392.gf8f6787159e-goog

6 years

2
2
0 0

[PATCH V5 3/9] blk-mq: free hw queue's resource in hctx's release handler

by Ming Lei

Once blk_cleanup_queue() returns, tags shouldn't be used any more, because blk_mq_free_tag_set() may be called. Commit 45a9c9d909b2 ("blk-mq: Fix a use-after-free") fixes this issue exactly. However, that commit introduces another issue. Before 45a9c9d909b2, we are allowed to run queue during cleaning up queue if the queue's kobj refcount is held. After that commit, queue can't be run during queue cleaning up, otherwise oops can be triggered easily because some fields of hctx are freed by blk_mq_free_queue() in blk_cleanup_queue(). We have invented ways for addressing this kind of issue before, such as: 8dc765d438f1 ("SCSI: fix queue cleanup race before queue initialization is done") c2856ae2f315 ("blk-mq: quiesce queue before freeing queue") But still can't cover all cases, recently James reports another such kind of issue: https://marc.info/?l=linux-scsi&m=155389088124782&w=2 This issue can be quite hard to address by previous way, given scsi_run_queue() may run requeues for other LUNs. Fixes the above issue by freeing hctx's resources in its release handler, and this way is safe becasue tags isn't needed for freeing such hctx resource. This approach follows typical design pattern wrt. kobject's release handler. Cc: Dongli Zhang <dongli.zhang(a)oracle.com> Cc: James Smart <james.smart(a)broadcom.com> Cc: Bart Van Assche <bart.vanassche(a)wdc.com> Cc: linux-scsi(a)vger.kernel.org, Cc: Martin K . Petersen <martin.petersen(a)oracle.com>, Cc: Christoph Hellwig <hch(a)lst.de>, Cc: James E . J . Bottomley <jejb(a)linux.vnet.ibm.com>, Cc: jianchao wang <jianchao.w.wang(a)oracle.com> Reported-by: James Smart <james.smart(a)broadcom.com> Fixes: 45a9c9d909b2 ("blk-mq: Fix a use-after-free") Cc: stable(a)vger.kernel.org Signed-off-by: Ming Lei <ming.lei(a)redhat.com> --- block/blk-core.c | 2 +- block/blk-mq-sysfs.c | 6 ++++++ block/blk-mq.c | 8 ++------ block/blk-mq.h | 2 +- 4 files changed, 10 insertions(+), 8 deletions(-) diff --git a/block/blk-core.c b/block/blk-core.c index 6583d67f3e34..20298aa5a77c 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -374,7 +374,7 @@ void blk_cleanup_queue(struct request_queue *q) blk_exit_queue(q); if (queue_is_mq(q)) - blk_mq_free_queue(q); + blk_mq_exit_queue(q); percpu_ref_exit(&q->q_usage_counter); diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 3f9c3f4ac44c..4040e62c3737 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -10,6 +10,7 @@ #include <linux/smp.h> #include <linux/blk-mq.h> +#include "blk.h" #include "blk-mq.h" #include "blk-mq-tag.h" @@ -33,6 +34,11 @@ static void blk_mq_hw_sysfs_release(struct kobject *kobj) { struct blk_mq_hw_ctx *hctx = container_of(kobj, struct blk_mq_hw_ctx, kobj); + + if (hctx->flags & BLK_MQ_F_BLOCKING) + cleanup_srcu_struct(hctx->srcu); + blk_free_flush_queue(hctx->fq); + sbitmap_free(&hctx->ctx_map); free_cpumask_var(hctx->cpumask); kfree(hctx->ctxs); kfree(hctx); diff --git a/block/blk-mq.c b/block/blk-mq.c index b512ba0cb359..afc9912e2e42 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2259,12 +2259,7 @@ static void blk_mq_exit_hctx(struct request_queue *q, if (set->ops->exit_hctx) set->ops->exit_hctx(hctx, hctx_idx); - if (hctx->flags & BLK_MQ_F_BLOCKING) - cleanup_srcu_struct(hctx->srcu); - blk_mq_remove_cpuhp(hctx); - blk_free_flush_queue(hctx->fq); - sbitmap_free(&hctx->ctx_map); } static void blk_mq_exit_hw_queues(struct request_queue *q, @@ -2899,7 +2894,8 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, } EXPORT_SYMBOL(blk_mq_init_allocated_queue); -void blk_mq_free_queue(struct request_queue *q) +/* tags can _not_ be used after returning from blk_mq_exit_queue */ +void blk_mq_exit_queue(struct request_queue *q) { struct blk_mq_tag_set *set = q->tag_set; diff --git a/block/blk-mq.h b/block/blk-mq.h index d704fc7766f4..c421e3a16e36 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -37,7 +37,7 @@ struct blk_mq_ctx { struct kobject kobj; } ____cacheline_aligned_in_smp; -void blk_mq_free_queue(struct request_queue *q); +void blk_mq_exit_queue(struct request_queue *q); int blk_mq_update_nr_requests(struct request_queue *q, unsigned int nr); void blk_mq_wake_waiters(struct request_queue *q); bool blk_mq_dispatch_rq_list(struct request_queue *, struct list_head *, bool); -- 2.9.5

6 years

2
2
0 0

[PATCH 5.1] rt2x00: do not increment sequence number while re-transmitting

by Stanislaw Gruszka

From: Vijayakumar Durai <vijayakumar.durai1(a)vivint.com> Currently rt2x00 devices retransmit the management frames with incremented sequence number if hardware is assigning the sequence. This is HW bug fixed already for non-QOS data frames, but it should be fixed for management frames except beacon. Without fix retransmitted frames have wrong SN: AlphaNet_e8:fb:36 Vivotek_52:31:51 Authentication, SN=1648, FN=0, Flags=........C Frame is not being retransmitted 1648 1 AlphaNet_e8:fb:36 Vivotek_52:31:51 Authentication, SN=1649, FN=0, Flags=....R...C Frame is being retransmitted 1649 1 AlphaNet_e8:fb:36 Vivotek_52:31:51 Authentication, SN=1650, FN=0, Flags=....R...C Frame is being retransmitted 1650 1 With the fix SN stays correctly the same: 88:6a:e3:e8:f9:a2 8c:f5:a3:88:76:87 Authentication, SN=1450, FN=0, Flags=........C 88:6a:e3:e8:f9:a2 8c:f5:a3:88:76:87 Authentication, SN=1450, FN=0, Flags=....R...C 88:6a:e3:e8:f9:a2 8c:f5:a3:88:76:87 Authentication, SN=1450, FN=0, Flags=....R...C Cc: stable(a)vger.kernel.org Signed-off-by: Vijayakumar Durai <vijayakumar.durai1(a)vivint.com> [sgruszka: simplify code, change comments and changelog] Signed-off-by: Stanislaw Gruszka <sgruszka(a)redhat.com> --- drivers/net/wireless/ralink/rt2x00/rt2x00.h | 1 - drivers/net/wireless/ralink/rt2x00/rt2x00mac.c | 10 ---------- drivers/net/wireless/ralink/rt2x00/rt2x00queue.c | 15 +++++++++------ 3 files changed, 9 insertions(+), 17 deletions(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00.h b/drivers/net/wireless/ralink/rt2x00/rt2x00.h index 4b1744e9fb78..50b92ca92bd7 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00.h @@ -673,7 +673,6 @@ enum rt2x00_state_flags { CONFIG_CHANNEL_HT40, CONFIG_POWERSAVING, CONFIG_HT_DISABLED, - CONFIG_QOS_DISABLED, CONFIG_MONITORING, /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c index 2825560e2424..e8462f25d252 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c @@ -642,19 +642,9 @@ void rt2x00mac_bss_info_changed(struct ieee80211_hw *hw, rt2x00dev->intf_associated--; rt2x00leds_led_assoc(rt2x00dev, !!rt2x00dev->intf_associated); - - clear_bit(CONFIG_QOS_DISABLED, &rt2x00dev->flags); } /* - * Check for access point which do not support 802.11e . We have to - * generate data frames sequence number in S/W for such AP, because - * of H/W bug. - */ - if (changes & BSS_CHANGED_QOS && !bss_conf->qos) - set_bit(CONFIG_QOS_DISABLED, &rt2x00dev->flags); - - /* * When the erp information has changed, we should perform * additional configuration steps. For all other changes we are done. */ diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00queue.c b/drivers/net/wireless/ralink/rt2x00/rt2x00queue.c index 92ddc19e7bf7..4834b4eb0206 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00queue.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00queue.c @@ -201,15 +201,18 @@ static void rt2x00queue_create_tx_descriptor_seq(struct rt2x00_dev *rt2x00dev, if (!rt2x00_has_cap_flag(rt2x00dev, REQUIRE_SW_SEQNO)) { /* * rt2800 has a H/W (or F/W) bug, device incorrectly increase - * seqno on retransmited data (non-QOS) frames. To workaround - * the problem let's generate seqno in software if QOS is - * disabled. + * seqno on retransmitted data (non-QOS) and management frames. + * To workaround the problem let's generate seqno in software. + * Except for beacons which are transmitted periodically by H/W + * hence hardware has to assign seqno for them. */ - if (test_bit(CONFIG_QOS_DISABLED, &rt2x00dev->flags)) - __clear_bit(ENTRY_TXD_GENERATE_SEQ, &txdesc->flags); - else + if (ieee80211_is_beacon(hdr->frame_control)) { + __set_bit(ENTRY_TXD_GENERATE_SEQ, &txdesc->flags); /* H/W will generate sequence number */ return; + } + + __clear_bit(ENTRY_TXD_GENERATE_SEQ, &txdesc->flags); } /* -- 2.7.5

6 years

3
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2019