- Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS] openvswitch: fix GSO userspace truncation underflow
by Yi Yang 21 Sep '26

21 Sep '26

From: Kyle Zeng <kylebot(a)openai.com> stable inclusion from stable-v5.10.265 commit 50a6a85f3d6b1d22d8436848606cdef5d2c490b4 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16932 CVE: CVE-2026-68123 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4032f8ed10fcb84d41c508dfb04be96589f78dfe ] OVS_ACTION_ATTR_TRUNC currently stores a delta from the original skb length in OVS_CB(skb)->cutlen. When a later userspace action segments a GSO skb, queue_gso_packets() reuses that delta for each smaller segment. A segment can then reach queue_userspace_packet() with cutlen greater than skb->len, underflowing the length passed to skb_zerocopy(). Store the maximum preserved length instead and bound each consumer against the current skb length. Use U32_MAX as the no-truncation sentinel so the value remains valid if skb geometry changes before a consumer handles it. Fixes: f2a4d086ed4c ("openvswitch: Add packet truncation support.") Cc: stable(a)vger.kernel.org Assisted-by: Codex:gpt-5.5 Signed-off-by: Kyle Zeng <kylebot(a)openai.com> Reviewed-by: Ilya Maximets <i.maximets(a)ovn.org> Reviewed-by: Aaron Conole <aconole(a)redhat.com> Link: https://patch.msgid.link/20260707221635.27489-1-kylebot@openai.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [5.10.y supports neither OVS_ACTION_ATTR_PSAMPLE nor OVS drop reasons] Signed-off-by: Ilya Maximets <i.maximets(a)ovn.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: net/openvswitch/datapath.c [Context conflicts.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- net/openvswitch/actions.c | 15 +++++---------- net/openvswitch/datapath.c | 25 ++++++++++++++----------- net/openvswitch/datapath.h | 2 +- net/openvswitch/vport.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 7ac5f75e06a9..f3b3fd459600 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -949,12 +949,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; - if (unlikely(cutlen > 0)) { - if (skb->len - cutlen > ovs_mac_header_len(key)) - pskb_trim(skb, skb->len - cutlen); - else - pskb_trim(skb, ovs_mac_header_len(key)); - } + if (unlikely(cutlen < skb->len)) + pskb_trim(skb, max(cutlen, ovs_mac_header_len(key))); if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { @@ -1237,22 +1233,21 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, clone = skb_clone(skb, GFP_ATOMIC); if (clone) do_output(dp, clone, port, key); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; } case OVS_ACTION_ATTR_TRUNC: { struct ovs_action_trunc *trunc = nla_data(a); - if (skb->len > trunc->max_len) - OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + OVS_CB(skb)->cutlen = trunc->max_len; break; } case OVS_ACTION_ATTR_USERSPACE: output_userspace(dp, skb, key, a, attr, len, OVS_CB(skb)->cutlen); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index bd00e63603ca..1aeeb18c8812 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -248,7 +248,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall, 0); + error = ovs_dp_upcall(dp, skb, key, &upcall, U32_MAX); if (unlikely(error)) kfree_skb(skb); else @@ -399,7 +399,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, struct sk_buff *nskb = NULL; struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - size_t len; + size_t msg_size; + size_t skb_len; unsigned int hlen; int err, dp_ifindex; @@ -419,7 +420,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb = nskb; } - if (nla_attr_size(skb->len) > USHRT_MAX) { + skb_len = min(skb->len, cutlen); + if (nla_attr_size(skb_len) > USHRT_MAX) { err = -EFBIG; goto out; } @@ -434,13 +436,13 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, * padding logic. Only perform zerocopy if padding is not required. */ if (dp->user_features & OVS_DP_F_UNALIGNED) - hlen = skb_zerocopy_headlen(skb); + hlen = min(skb_zerocopy_headlen(skb), cutlen); else - hlen = skb->len; + hlen = skb_len; - len = upcall_msg_size(upcall_info, hlen - cutlen, - OVS_CB(skb)->acts_origlen); - user_skb = genlmsg_new(len, GFP_ATOMIC); + msg_size = upcall_msg_size(upcall_info, hlen, + OVS_CB(skb)->acts_origlen); + user_skb = genlmsg_new(msg_size, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -503,7 +505,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, } /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ - if (cutlen > 0) { + if (skb_len < skb->len) { if (nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) { err = -ENOBUFS; @@ -518,9 +520,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len - cutlen); + nla->nla_len = nla_attr_size(skb_len); - err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); + err = skb_zerocopy(user_skb, skb, skb_len, hlen); if (err) goto out; @@ -575,6 +577,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->ignore_df = 1; } OVS_CB(packet)->mru = mru; + OVS_CB(packet)->cutlen = U32_MAX; /* Build an sw_flow for sending this packet. */ flow = ovs_flow_alloc(); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index c9eb267c6f7e..9192c9970cc3 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -105,7 +105,7 @@ struct datapath { * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. * @acts_origlen: The netlink size of the flow actions applied to this skb. - * @cutlen: The number of bytes from the packet end to be removed. + * @cutlen: The number of bytes in the packet to preserve on output. */ struct ovs_skb_cb { struct vport *input_vport; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 5541f4c9a0a1..586428d2015d 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -447,7 +447,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] pppoe: reload header pointer after dev_hard_header()
by Yi Yang 21 Sep '26

21 Sep '26

From: Asim Viladi Oglu Manizada <manizada(a)pm.me> mainline inclusion from mainline-v7.2-rc5 commit e9c238f6fe42fb1b4dba3a578277de32cb487937 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16930 CVE: CVE-2026-68121 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- pppoe_sendmsg() saves a pointer to the PPPoE header before calling dev_hard_header(). Device header callbacks are allowed to reallocate the skb head, invalidating pointers into it. This can happen when a send is blocked in copy_from_user() while the first non-Ethernet port is added to an empty team device. The team's delegated GRE header callback then expands the skb head. PPPoE subsequently writes six bytes through the stale pointer into the freed head. Reload the PPPoE header through the skb's network-header offset after device header creation. pskb_expand_head() updates that offset when it relocates the head. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable(a)vger.kernel.org Signed-off-by: Asim Viladi Oglu Manizada <manizada(a)pm.me> Reviewed-by: Vadim Fedorenko <vadim.fedorenko(a)linux.dev> Reviewed-by: Eric Dumazet <edumazet(a)google.com> Link: https://patch.msgid.link/20260722093814.3017176-1-manizada@pm.me Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/net/ppp/pppoe.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c index c04f3dc17d76..86fc5f7551ff 100644 --- a/drivers/net/ppp/pppoe.c +++ b/drivers/net/ppp/pppoe.c @@ -897,6 +897,7 @@ static int pppoe_sendmsg(struct socket *sock, struct msghdr *m, dev_hard_header(skb, dev, ETH_P_PPP_SES, po->pppoe_pa.remote, NULL, total_len); + ph = pppoe_hdr(skb); memcpy(ph, &hdr, sizeof(struct pppoe_hdr)); ph->length = htons(total_len); -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] openvswitch: fix GSO userspace truncation underflow
by Yi Yang 21 Sep '26

21 Sep '26

From: Kyle Zeng <kylebot(a)openai.com> stable inclusion from stable-v5.10.265 commit 50a6a85f3d6b1d22d8436848606cdef5d2c490b4 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/16932 CVE: CVE-2026-68123 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4032f8ed10fcb84d41c508dfb04be96589f78dfe ] OVS_ACTION_ATTR_TRUNC currently stores a delta from the original skb length in OVS_CB(skb)->cutlen. When a later userspace action segments a GSO skb, queue_gso_packets() reuses that delta for each smaller segment. A segment can then reach queue_userspace_packet() with cutlen greater than skb->len, underflowing the length passed to skb_zerocopy(). Store the maximum preserved length instead and bound each consumer against the current skb length. Use U32_MAX as the no-truncation sentinel so the value remains valid if skb geometry changes before a consumer handles it. Fixes: f2a4d086ed4c ("openvswitch: Add packet truncation support.") Cc: stable(a)vger.kernel.org Assisted-by: Codex:gpt-5.5 Signed-off-by: Kyle Zeng <kylebot(a)openai.com> Reviewed-by: Ilya Maximets <i.maximets(a)ovn.org> Reviewed-by: Aaron Conole <aconole(a)redhat.com> Link: https://patch.msgid.link/20260707221635.27489-1-kylebot@openai.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> [5.10.y supports neither OVS_ACTION_ATTR_PSAMPLE nor OVS drop reasons] Signed-off-by: Ilya Maximets <i.maximets(a)ovn.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: net/openvswitch/datapath.c [Context conflicts.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- net/openvswitch/actions.c | 15 +++++---------- net/openvswitch/datapath.c | 25 ++++++++++++++----------- net/openvswitch/datapath.h | 2 +- net/openvswitch/vport.c | 2 +- 4 files changed, 21 insertions(+), 23 deletions(-) diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 7ac5f75e06a9..f3b3fd459600 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -949,12 +949,8 @@ static void do_output(struct datapath *dp, struct sk_buff *skb, int out_port, u16 mru = OVS_CB(skb)->mru; u32 cutlen = OVS_CB(skb)->cutlen; - if (unlikely(cutlen > 0)) { - if (skb->len - cutlen > ovs_mac_header_len(key)) - pskb_trim(skb, skb->len - cutlen); - else - pskb_trim(skb, ovs_mac_header_len(key)); - } + if (unlikely(cutlen < skb->len)) + pskb_trim(skb, max(cutlen, ovs_mac_header_len(key))); if (likely(!mru || (skb->len <= mru + vport->dev->hard_header_len))) { @@ -1237,22 +1233,21 @@ static int do_execute_actions(struct datapath *dp, struct sk_buff *skb, clone = skb_clone(skb, GFP_ATOMIC); if (clone) do_output(dp, clone, port, key); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; } case OVS_ACTION_ATTR_TRUNC: { struct ovs_action_trunc *trunc = nla_data(a); - if (skb->len > trunc->max_len) - OVS_CB(skb)->cutlen = skb->len - trunc->max_len; + OVS_CB(skb)->cutlen = trunc->max_len; break; } case OVS_ACTION_ATTR_USERSPACE: output_userspace(dp, skb, key, a, attr, len, OVS_CB(skb)->cutlen); - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; break; case OVS_ACTION_ATTR_HASH: diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index bd00e63603ca..1aeeb18c8812 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -248,7 +248,7 @@ void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key) upcall.cmd = OVS_PACKET_CMD_MISS; upcall.portid = ovs_vport_find_upcall_portid(p, skb); upcall.mru = OVS_CB(skb)->mru; - error = ovs_dp_upcall(dp, skb, key, &upcall, 0); + error = ovs_dp_upcall(dp, skb, key, &upcall, U32_MAX); if (unlikely(error)) kfree_skb(skb); else @@ -399,7 +399,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, struct sk_buff *nskb = NULL; struct sk_buff *user_skb = NULL; /* to be queued to userspace */ struct nlattr *nla; - size_t len; + size_t msg_size; + size_t skb_len; unsigned int hlen; int err, dp_ifindex; @@ -419,7 +420,8 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, skb = nskb; } - if (nla_attr_size(skb->len) > USHRT_MAX) { + skb_len = min(skb->len, cutlen); + if (nla_attr_size(skb_len) > USHRT_MAX) { err = -EFBIG; goto out; } @@ -434,13 +436,13 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, * padding logic. Only perform zerocopy if padding is not required. */ if (dp->user_features & OVS_DP_F_UNALIGNED) - hlen = skb_zerocopy_headlen(skb); + hlen = min(skb_zerocopy_headlen(skb), cutlen); else - hlen = skb->len; + hlen = skb_len; - len = upcall_msg_size(upcall_info, hlen - cutlen, - OVS_CB(skb)->acts_origlen); - user_skb = genlmsg_new(len, GFP_ATOMIC); + msg_size = upcall_msg_size(upcall_info, hlen, + OVS_CB(skb)->acts_origlen); + user_skb = genlmsg_new(msg_size, GFP_ATOMIC); if (!user_skb) { err = -ENOMEM; goto out; @@ -503,7 +505,7 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, } /* Add OVS_PACKET_ATTR_LEN when packet is truncated */ - if (cutlen > 0) { + if (skb_len < skb->len) { if (nla_put_u32(user_skb, OVS_PACKET_ATTR_LEN, skb->len)) { err = -ENOBUFS; @@ -518,9 +520,9 @@ static int queue_userspace_packet(struct datapath *dp, struct sk_buff *skb, err = -ENOBUFS; goto out; } - nla->nla_len = nla_attr_size(skb->len - cutlen); + nla->nla_len = nla_attr_size(skb_len); - err = skb_zerocopy(user_skb, skb, skb->len - cutlen, hlen); + err = skb_zerocopy(user_skb, skb, skb_len, hlen); if (err) goto out; @@ -575,6 +577,7 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) packet->ignore_df = 1; } OVS_CB(packet)->mru = mru; + OVS_CB(packet)->cutlen = U32_MAX; /* Build an sw_flow for sending this packet. */ flow = ovs_flow_alloc(); diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index c9eb267c6f7e..9192c9970cc3 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -105,7 +105,7 @@ struct datapath { * @mru: The maximum received fragement size; 0 if the packet is not * fragmented. * @acts_origlen: The netlink size of the flow actions applied to this skb. - * @cutlen: The number of bytes from the packet end to be removed. + * @cutlen: The number of bytes in the packet to preserve on output. */ struct ovs_skb_cb { struct vport *input_vport; diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c index 5541f4c9a0a1..586428d2015d 100644 --- a/net/openvswitch/vport.c +++ b/net/openvswitch/vport.c @@ -447,7 +447,7 @@ int ovs_vport_receive(struct vport *vport, struct sk_buff *skb, OVS_CB(skb)->input_vport = vport; OVS_CB(skb)->mru = 0; - OVS_CB(skb)->cutlen = 0; + OVS_CB(skb)->cutlen = U32_MAX; if (unlikely(dev_net(skb->dev) != ovs_dp_get_net(vport->dp))) { u32 mark; -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] selinux: check connect-related permissions on TCP Fast Open
by Yi Yang 20 Sep '26

20 Sep '26

From: Stephen Smalley <stephen.smalley.work(a)gmail.com> mainline inclusion from mainline-v7.2-rc3 commit 44c74d27d1b9aaa99fa8a83640c1223575262b80 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/17775 CVE: CVE-2026-72243 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Similar to Landlock, SELinux was not updated when TCP Fast Open support was introduced to ensure connect-related permissions are checked when using TCP Fast Open. Update its socket_sendmsg() hook to call selinux_socket_connect() when MSG_FASTOPEN is passed. Cc: stable(a)vger.kernel.org Link: https://lore.kernel.org/linux-security-module/20260616201615.275032-1-hexla… Link: https://lore.kernel.org/linux-security-module/20260617180526.15627-2-matthi… Reported-by: Bryam Vargas <hexlabsecurity(a)proton.me> Reported-by: Matthieu Buffet <matthieu(a)buffet.re> Reported-by: Mikhail Ivanov <ivanov.mikhail1(a)huawei-partners.com> Signed-off-by: Stephen Smalley <stephen.smalley.work(a)gmail.com> Tested-by: Bryam Vargas <hexlabsecurity(a)proton.me> Signed-off-by: Paul Moore <paul(a)paul-moore.com> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- security/selinux/hooks.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 3550ded07457..c0b5af6a82a7 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -4910,7 +4910,24 @@ static int selinux_socket_accept(struct socket *sock, struct socket *newsock) static int selinux_socket_sendmsg(struct socket *sock, struct msghdr *msg, int size) { - return sock_has_perm(sock->sk, SOCKET__WRITE); + int rc; + struct sockaddr *const addr = msg->msg_name; + const int addrlen = msg->msg_namelen; + + rc = sock_has_perm(sock->sk, SOCKET__WRITE); + if (rc) + return rc; + + if (addr && (msg->msg_flags & MSG_FASTOPEN) && + (sk_is_tcp(sock->sk) || + (sk_is_inet(sock->sk) && sock->sk->sk_type == SOCK_STREAM && + sock->sk->sk_protocol == IPPROTO_MPTCP))) { + rc = selinux_socket_connect(sock, addr, addrlen); + if (rc) + return rc; + } + + return 0; } static int selinux_socket_recvmsg(struct socket *sock, struct msghdr *msg, -- 2.25.1

2 1

[PATCH OLK-6.6 0/4] arm64: smt: Introduce VIP-SMT QoS mode for SMT
by Yipeng Zou 20 Sep '26

20 Sep '26

This patchset introduces VIP-SMT, an Hisilicon-specific SMT QoS enhancement for Arm64. It allows marking specific hardware threads as "VIP" (high-priority) and others as non-VIP within the same physical core, enabling differentiated resource allocation (instruction fetch, execution pipeline, and out-of-order resources) for latency-sensitive workloads while background tasks run on non-VIP threads. Use cases include running real-time/network packet processing on VIP threads, performance isolation for mixed-criticality tasks, and QoS enforcement for cloud/virtualization scenarios. The series is organized as follows: Patch 1 refactors arch_cpu_idle_{enter,exit}() so the SMT measurement and the VIP-SMT hooks can be shared cleanly across all configs; the ACTLR_XCALL_XINT register save/restore is moved under its own config guard. Patch 2 is the core feature. It adds CONFIG_ARM64_VIP_SMT and a new vip_smt.c driver exposing per-CPU sysfs entries under /sys/devices/system/cpu/cpuN/regs/vip-smt/ to configure three ACTLR system registers (IFU_ACTLR1_EL1, OOO_DEC_ROB_SHA_CTLR_EL1, and OOO_DEC_DSP_CTLR_EL1). Feature detection is based on MIDR (HIP13), restricted to EL2 at the moment, and requires SMT to be enabled on the physical core. Per-CPU register values are shadowed and restored on idle exit so they survive core power-down. Patch 3 adds a "novipsmt" kernel command line parameter (with an optional "force" value) that disables the feature at boot time and prevents the vip_smt sysfs directory from being created. Yipeng Zou (4): arm64: idle: make arch_cpu_idle_{enter,exit} more refactorable arm64: smt: Introduce VIP-SMT a QoS Mode for SMT arm64: smt: Add novipsmt kernel command line parameter arm64: configs: Enable ARM64_VIP_SMT in openeuler_defconfig .../admin-guide/kernel-parameters.txt | 4 + arch/arm64/Kconfig | 16 + arch/arm64/configs/openeuler_defconfig | 1 + arch/arm64/include/asm/vip_smt.h | 109 ++++ arch/arm64/kernel/Makefile | 1 + arch/arm64/kernel/cpufeature.c | 9 + arch/arm64/kernel/cpuinfo.c | 7 + arch/arm64/kernel/idle.c | 51 +- arch/arm64/kernel/vip_smt.c | 483 ++++++++++++++++++ arch/arm64/tools/cpucaps | 2 +- 10 files changed, 653 insertions(+), 30 deletions(-) create mode 100644 arch/arm64/include/asm/vip_smt.h create mode 100644 arch/arm64/kernel/vip_smt.c -- 2.34.1

2 5

[PATCH OLK-6.6] ceph: cap delegated inode count in ceph_parse_deleg_inos()
by Pan Taixi 20 Sep '26

20 Sep '26

From: Michael Bommarito <michael.bommarito(a)gmail.com> mainline inclusion from mainline-v7.3-rc1 commit 4bd3158bd62466d57ed72a3f7bc5f205fedd6919 category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/18927 CVE: CVE-2026-89648 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- ceph_parse_deleg_inos() decodes interval sets of delegated inode numbers from an MDS create-with-delegation reply. For each set it reads a 64-bit start and a 64-bit len with ceph_decode_64_safe(), which only validates that the eight bytes are present in the message, not the value, and then loops over len while inserting entries into s_delegated_inos. len is fully attacker controlled. A malicious or compromised MDS can send one huge interval, many intervals in one reply, duplicate intervals, or repeated replies that accumulate delegated inodes on the same session. The original code bounded none of these and could spin the insert loop or grow the xarray without limit. Bound both dimensions with a single enforcement point. Track the number of delegated inodes held by each MDS session in an atomic counter and grow it only in ceph_insert_deleg_ino(), which uses atomic_add_unless() to refuse to push the count past CEPH_MAX_DELEG_INOS. Because that helper is the only place the counter grows, the per-session population can never exceed the cap, so no separate per-session pre-check is needed. The counter is decremented when async create consumes a delegated inode or when an insert fails, incremented when a delegated inode is restored, initialized with the session xarray, and reset when reconnect destroys the xarray. A per-session cap alone still lets one reply spin the insert loop on duplicate ranges without growing the counter, so also cap the aggregate interval length accepted from a single reply. Together these bound both the loop trip count per reply and the xarray population across replies. The cap is a fixed, client-chosen constant rather than a value derived from the MDS. mds_client_prealloc_inos is a userspace MDS configuration option; it is never sent to the kernel client on the wire, and a server-supplied bound could not be trusted for a defensive limit in any case. The constant is set well above that option's documented default of 1000 (a generous multiple), so legitimate refill behavior is unaffected while the CPU and xarray memory a malformed delegation stream can consume stays bounded. Impact: a malicious or compromised Ceph MDS can no longer make a client spin through an unbounded delegated-inode interval or grow one session's delegated-inode xarray without limit. Cc: stable(a)vger.kernel.org Fixes: d48464878708 ("ceph: decode interval_sets for delegated inos") Suggested-by: Viacheslav Dubeyko <Slava.Dubeyko(a)ibm.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito <michael.bommarito(a)gmail.com> Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko(a)ibm.com> Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Conflicts: fs/ceph/mds_client.c [backport to OLK-6.6: context adjusted for this tree: pr_warn_ratelimited_client()/doutc() from 38d46409c4639 are not used here, so the new warnings use plain pr_warn_ratelimited() and the struct ceph_client *cl local is not added. CEPH_MAX_DELEG_INOS is inserted right after CEPH_INO_SYSTEM_BASE. s_num_deleg_inos is reset next to the existing xa_destroy() in send_mds_reconnect(), which still runs before mutex_lock(&session->s_mutex) in this tree (39fe30315893 not applied); this matches the linux-6.18.y stable backport c040e139f1a62.] Signed-off-by: Pan Taixi <pantaixi1(a)huawei.com> --- fs/ceph/mds_client.c | 56 +++++++++++++++++++++++++++++++++++++++----- fs/ceph/mds_client.h | 1 + fs/ceph/super.h | 9 +++++++ 3 files changed, 60 insertions(+), 6 deletions(-) diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 00b9f9f3f4d7..a3fa2fd0b974 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -566,13 +566,37 @@ static int parse_reply_info_filelock(void **p, void *end, #if BITS_PER_LONG == 64 #define DELEGATED_INO_AVAILABLE xa_mk_value(1) +static int ceph_insert_deleg_ino(struct ceph_mds_session *s, u64 ino) +{ + int err; + + /* + * Cap how many delegated inodes a single session may hold. This is + * the only place that grows the count, so atomic_add_unless() bounds + * it at exactly CEPH_MAX_DELEG_INOS; s_num_deleg_inos can never exceed + * that. + */ + if (!atomic_add_unless(&s->s_num_deleg_inos, 1, CEPH_MAX_DELEG_INOS)) { + pr_warn_ratelimited("MDS session already holds %d delegated inodes\n", + CEPH_MAX_DELEG_INOS); + return -EOVERFLOW; + } + + err = xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE, + GFP_KERNEL); + if (err) + atomic_dec(&s->s_num_deleg_inos); + return err; +} + static int ceph_parse_deleg_inos(void **p, void *end, struct ceph_mds_session *s) { + u64 msg_deleg_inos = 0; u32 sets; ceph_decode_32_safe(p, end, sets, bad); dout("got %u sets of delegated inodes\n", sets); while (sets--) { @@ -585,20 +609,37 @@ static int ceph_parse_deleg_inos(void **p, void *end, if (start < CEPH_INO_SYSTEM_BASE) { pr_warn_ratelimited("ceph: ignoring reserved inode range delegation (start=0x%llx len=0x%llx)\n", start, len); continue; } + + /* + * Bound the number of inodes one reply may delegate. + * ceph_insert_deleg_ino() separately caps the per-session + * population, so this only has to stop one reply from spinning + * the insert loop under an attacker-controlled len. + */ + if (len > (u64)CEPH_MAX_DELEG_INOS || + msg_deleg_inos > (u64)CEPH_MAX_DELEG_INOS - len) { + pr_warn_ratelimited("MDS reply delegates too many inodes (have %llu, +%llu, max %d)\n", + msg_deleg_inos, len, CEPH_MAX_DELEG_INOS); + return -EIO; + } + msg_deleg_inos += len; + while (len--) { - int err = xa_insert(&s->s_delegated_inos, start++, - DELEGATED_INO_AVAILABLE, - GFP_KERNEL); + int err = ceph_insert_deleg_ino(s, start++); + if (!err) { dout("added delegated inode 0x%llx\n", start - 1); } else if (err == -EBUSY) { pr_warn("MDS delegated inode 0x%llx more than once.\n", start - 1); + } else if (err == -EOVERFLOW) { + /* ceph_insert_deleg_ino() already warned. */ + return -EIO; } else { return err; } } } @@ -612,20 +653,21 @@ u64 ceph_get_deleg_ino(struct ceph_mds_session *s) unsigned long ino; void *val; xa_for_each(&s->s_delegated_inos, ino, val) { val = xa_erase(&s->s_delegated_inos, ino); - if (val == DELEGATED_INO_AVAILABLE) + if (val == DELEGATED_INO_AVAILABLE) { + atomic_dec(&s->s_num_deleg_inos); return ino; + } } return 0; } int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino) { - return xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE, - GFP_KERNEL); + return ceph_insert_deleg_ino(s, ino); } #else /* BITS_PER_LONG == 64 */ /* * FIXME: xarrays can't handle 64-bit indexes on a 32-bit arch. For now, just * ignore delegated_inos on 32 bit arch. Maybe eventually add xarrays for top @@ -1002,10 +1044,11 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, INIT_LIST_HEAD(&s->s_caps); refcount_set(&s->s_ref, 1); INIT_LIST_HEAD(&s->s_waiting); INIT_LIST_HEAD(&s->s_unsafe); xa_init(&s->s_delegated_inos); + atomic_set(&s->s_num_deleg_inos, 0); INIT_LIST_HEAD(&s->s_cap_releases); INIT_WORK(&s->s_cap_release_work, ceph_cap_release_work); INIT_LIST_HEAD(&s->s_cap_dirty); INIT_LIST_HEAD(&s->s_cap_flushing); @@ -4619,10 +4662,11 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc, reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false); if (!reply) goto fail_nomsg; xa_destroy(&session->s_delegated_inos); + atomic_set(&session->s_num_deleg_inos, 0); mutex_lock(&session->s_mutex); session->s_state = CEPH_MDS_SESSION_RECONNECTING; session->s_seq = 0; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 22398ccb21a4..9916acd901ab 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -228,10 +228,11 @@ struct ceph_mds_session { u64 s_renew_seq; struct list_head s_waiting; /* waiting requests */ struct list_head s_unsafe; /* unsafe requests */ struct xarray s_delegated_inos; + atomic_t s_num_deleg_inos; }; /* * modes of choosing which MDS to send a request to */ diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 5903e3fb6d75..09bf04f4452e 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -588,10 +588,19 @@ static inline int ceph_ino_compare(struct inode *inode, void *data) #define CEPH_NUM_STRAY 10 #define CEPH_MDS_INO_MDSDIR_OFFSET (1 * CEPH_MAX_MDS) #define CEPH_MDS_INO_LOG_OFFSET (2 * CEPH_MAX_MDS) #define CEPH_INO_SYSTEM_BASE ((6*CEPH_MAX_MDS) + (CEPH_MAX_MDS * CEPH_NUM_STRAY)) +/* + * Upper bound on the number of delegated inodes a single MDS session may + * hold. The MDS normally hands out a small preallocation window (the + * userspace mds_client_prealloc_inos option defaults to 1000) and refills + * it as the client consumes entries. This leaves generous headroom while + * bounding the CPU and memory a malformed delegation interval can consume. + */ +#define CEPH_MAX_DELEG_INOS 8192 + static inline bool ceph_vino_is_reserved(const struct ceph_vino vino) { if (vino.ino >= CEPH_INO_SYSTEM_BASE || vino.ino < CEPH_MDS_INO_MDSDIR_OFFSET) return false; -- 2.34.1

2 1

[PATCH OLK-5.10] ceph: cap delegated inode count in ceph_parse_deleg_inos()
by Pan Taixi 20 Sep '26

20 Sep '26

From: Michael Bommarito <michael.bommarito(a)gmail.com> mainline inclusion from mainline-v7.3-rc1 commit 4bd3158bd62466d57ed72a3f7bc5f205fedd6919 category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/18927 CVE: CVE-2026-89648 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- ceph_parse_deleg_inos() decodes interval sets of delegated inode numbers from an MDS create-with-delegation reply. For each set it reads a 64-bit start and a 64-bit len with ceph_decode_64_safe(), which only validates that the eight bytes are present in the message, not the value, and then loops over len while inserting entries into s_delegated_inos. len is fully attacker controlled. A malicious or compromised MDS can send one huge interval, many intervals in one reply, duplicate intervals, or repeated replies that accumulate delegated inodes on the same session. The original code bounded none of these and could spin the insert loop or grow the xarray without limit. Bound both dimensions with a single enforcement point. Track the number of delegated inodes held by each MDS session in an atomic counter and grow it only in ceph_insert_deleg_ino(), which uses atomic_add_unless() to refuse to push the count past CEPH_MAX_DELEG_INOS. Because that helper is the only place the counter grows, the per-session population can never exceed the cap, so no separate per-session pre-check is needed. The counter is decremented when async create consumes a delegated inode or when an insert fails, incremented when a delegated inode is restored, initialized with the session xarray, and reset when reconnect destroys the xarray. A per-session cap alone still lets one reply spin the insert loop on duplicate ranges without growing the counter, so also cap the aggregate interval length accepted from a single reply. Together these bound both the loop trip count per reply and the xarray population across replies. The cap is a fixed, client-chosen constant rather than a value derived from the MDS. mds_client_prealloc_inos is a userspace MDS configuration option; it is never sent to the kernel client on the wire, and a server-supplied bound could not be trusted for a defensive limit in any case. The constant is set well above that option's documented default of 1000 (a generous multiple), so legitimate refill behavior is unaffected while the CPU and xarray memory a malformed delegation stream can consume stays bounded. Impact: a malicious or compromised Ceph MDS can no longer make a client spin through an unbounded delegated-inode interval or grow one session's delegated-inode xarray without limit. Cc: stable(a)vger.kernel.org Fixes: d48464878708 ("ceph: decode interval_sets for delegated inos") Suggested-by: Viacheslav Dubeyko <Slava.Dubeyko(a)ibm.com> Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito <michael.bommarito(a)gmail.com> Reviewed-by: Viacheslav Dubeyko <Slava.Dubeyko(a)ibm.com> Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Conflicts: fs/ceph/mds_client.c [backport to OLK-5.10: context adjusted for this tree: pr_warn_ratelimited_client()/doutc() from 38d46409c4639 are not used here, so the new warnings use plain pr_warn_ratelimited() and the struct ceph_client *cl local is not added. The redundant 'ino' local (2ecd0edd13a8b not applied here) is dropped together with its only user, the open-coded xa_insert() in ceph_parse_deleg_inos(). CEPH_MAX_DELEG_INOS is inserted right after CEPH_INO_SYSTEM_BASE. s_num_deleg_inos is reset next to the existing xa_destroy() in send_mds_reconnect(), which still runs before mutex_lock(&session->s_mutex) in this tree (39fe30315893 not applied); this matches the linux-6.18.y stable backport c040e139f1a62.] Signed-off-by: Pan Taixi <pantaixi1(a)huawei.com> --- fs/ceph/mds_client.c | 58 ++++++++++++++++++++++++++++++++++++++------ fs/ceph/mds_client.h | 1 + fs/ceph/super.h | 9 +++++++ 3 files changed, 61 insertions(+), 7 deletions(-) diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index aea1b42b9b18..ef39852280c8 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -419,39 +419,80 @@ static int parse_reply_info_filelock(void **p, void *end, #if BITS_PER_LONG == 64 #define DELEGATED_INO_AVAILABLE xa_mk_value(1) +static int ceph_insert_deleg_ino(struct ceph_mds_session *s, u64 ino) +{ + int err; + + /* + * Cap how many delegated inodes a single session may hold. This is + * the only place that grows the count, so atomic_add_unless() bounds + * it at exactly CEPH_MAX_DELEG_INOS; s_num_deleg_inos can never exceed + * that. + */ + if (!atomic_add_unless(&s->s_num_deleg_inos, 1, CEPH_MAX_DELEG_INOS)) { + pr_warn_ratelimited("MDS session already holds %d delegated inodes\n", + CEPH_MAX_DELEG_INOS); + return -EOVERFLOW; + } + + err = xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE, + GFP_KERNEL); + if (err) + atomic_dec(&s->s_num_deleg_inos); + return err; +} + static int ceph_parse_deleg_inos(void **p, void *end, struct ceph_mds_session *s) { + u64 msg_deleg_inos = 0; u32 sets; ceph_decode_32_safe(p, end, sets, bad); dout("got %u sets of delegated inodes\n", sets); while (sets--) { - u64 start, len, ino; + u64 start, len; ceph_decode_64_safe(p, end, start, bad); ceph_decode_64_safe(p, end, len, bad); /* Don't accept a delegation of system inodes */ if (start < CEPH_INO_SYSTEM_BASE) { pr_warn_ratelimited("ceph: ignoring reserved inode range delegation (start=0x%llx len=0x%llx)\n", start, len); continue; } + + /* + * Bound the number of inodes one reply may delegate. + * ceph_insert_deleg_ino() separately caps the per-session + * population, so this only has to stop one reply from spinning + * the insert loop under an attacker-controlled len. + */ + if (len > (u64)CEPH_MAX_DELEG_INOS || + msg_deleg_inos > (u64)CEPH_MAX_DELEG_INOS - len) { + pr_warn_ratelimited("MDS reply delegates too many inodes (have %llu, +%llu, max %d)\n", + msg_deleg_inos, len, CEPH_MAX_DELEG_INOS); + return -EIO; + } + msg_deleg_inos += len; + while (len--) { - int err = xa_insert(&s->s_delegated_inos, ino = start++, - DELEGATED_INO_AVAILABLE, - GFP_KERNEL); + int err = ceph_insert_deleg_ino(s, start++); + if (!err) { dout("added delegated inode 0x%llx\n", start - 1); } else if (err == -EBUSY) { pr_warn("ceph: MDS delegated inode 0x%llx more than once.\n", start - 1); + } else if (err == -EOVERFLOW) { + /* ceph_insert_deleg_ino() already warned. */ + return -EIO; } else { return err; } } } @@ -465,20 +506,21 @@ u64 ceph_get_deleg_ino(struct ceph_mds_session *s) unsigned long ino; void *val; xa_for_each(&s->s_delegated_inos, ino, val) { val = xa_erase(&s->s_delegated_inos, ino); - if (val == DELEGATED_INO_AVAILABLE) + if (val == DELEGATED_INO_AVAILABLE) { + atomic_dec(&s->s_num_deleg_inos); return ino; + } } return 0; } int ceph_restore_deleg_ino(struct ceph_mds_session *s, u64 ino) { - return xa_insert(&s->s_delegated_inos, ino, DELEGATED_INO_AVAILABLE, - GFP_KERNEL); + return ceph_insert_deleg_ino(s, ino); } #else /* BITS_PER_LONG == 64 */ /* * FIXME: xarrays can't handle 64-bit indexes on a 32-bit arch. For now, just * ignore delegated_inos on 32 bit arch. Maybe eventually add xarrays for top @@ -760,10 +802,11 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, s->s_nr_caps = 0; refcount_set(&s->s_ref, 1); INIT_LIST_HEAD(&s->s_waiting); INIT_LIST_HEAD(&s->s_unsafe); xa_init(&s->s_delegated_inos); + atomic_set(&s->s_num_deleg_inos, 0); s->s_num_cap_releases = 0; s->s_cap_reconnect = 0; s->s_cap_iterator = NULL; INIT_LIST_HEAD(&s->s_cap_releases); INIT_WORK(&s->s_cap_release_work, ceph_cap_release_work); @@ -3908,10 +3951,11 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc, reply = ceph_msg_new2(CEPH_MSG_CLIENT_RECONNECT, 0, 1, GFP_NOFS, false); if (!reply) goto fail_nomsg; xa_destroy(&session->s_delegated_inos); + atomic_set(&session->s_num_deleg_inos, 0); mutex_lock(&session->s_mutex); session->s_state = CEPH_MDS_SESSION_RECONNECTING; session->s_seq = 0; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index b1939f916228..847f04424be3 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -208,10 +208,11 @@ struct ceph_mds_session { u64 s_renew_seq; struct list_head s_waiting; /* waiting requests */ struct list_head s_unsafe; /* unsafe requests */ struct xarray s_delegated_inos; + atomic_t s_num_deleg_inos; }; /* * modes of choosing which MDS to send a request to */ diff --git a/fs/ceph/super.h b/fs/ceph/super.h index c0c5ed3e2718..6b2e0cfb9ae3 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -538,10 +538,19 @@ static inline int ceph_ino_compare(struct inode *inode, void *data) #define CEPH_MAX_MDS 0x100 #define CEPH_NUM_STRAY 10 #define CEPH_MDS_INO_MDSDIR_OFFSET (1 * CEPH_MAX_MDS) #define CEPH_INO_SYSTEM_BASE ((6*CEPH_MAX_MDS) + (CEPH_MAX_MDS * CEPH_NUM_STRAY)) +/* + * Upper bound on the number of delegated inodes a single MDS session may + * hold. The MDS normally hands out a small preallocation window (the + * userspace mds_client_prealloc_inos option defaults to 1000) and refills + * it as the client consumes entries. This leaves generous headroom while + * bounding the CPU and memory a malformed delegation interval can consume. + */ +#define CEPH_MAX_DELEG_INOS 8192 + static inline bool ceph_vino_is_reserved(const struct ceph_vino vino) { if (vino.ino < CEPH_INO_SYSTEM_BASE && vino.ino >= CEPH_MDS_INO_MDSDIR_OFFSET) { WARN_RATELIMIT(1, "Attempt to access reserved inode number 0x%llx", vino.ino); -- 2.34.1

2 1

[PATCH OLK-6.6] ceph: do not repeat ceph_trim_dentries() if no progress possible
by Pan Taixi 20 Sep '26

20 Sep '26

From: Max Kellermann <max.kellermann(a)ionos.com> mainline inclusion from mainline-v7.3-rc1 commit e7d7aa7b730178278109c41fa1b17b06873065d5 category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/19026 CVE: CVE-2026-89647 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- ceph_cap_reclaim_work() re-queues itself for as long as ceph_trim_dentries() returns -EAGAIN, which happens whenever a lease walk exhausts its `nr_to_scan` budget. This creates a busy loop that consumes CPU without making any progress when there is nothing to reclaim: with no cap pressure (`count==0`) and every scanned lease still valid, each pass runs the full scan budget down to zero and returns `-EAGAIN`, only to be queued again immediately. The dir-lease walk made this worse. When `expire_dir_lease` is `false` (i.e. we have no intention of reclaiming dir leases), __dir_lease_check() returned `TOUCH` for every valid lease. `TOUCH` moves the dentry to the tail of the list and resets `di->time` via __dentry_dir_lease_touch(), so a walk over N valid leases pointlessly rewrote the list, refreshed the timestamps (preventing them from ever aging out) and always drained `nr_to_scan`, guaranteeing the `-EAGAIN` requeue. Fix this in three steps: - Return `KEEP` instead of `TOUCH` when `expire_dir_lease` is `false`. If we are not going to reclaim the lease, leave it in place instead of churning the list and resetting its timestamp; the walk then terminates naturally (or via `STOP` at the first fresh lease). - Only return `-EAGAIN` from the first (dentry-lease) walk when something was actually freed. A full batch that frees nothing means retrying the same list immediately is futile; fall through to the dir-lease walk instead. - After both walks, bail out with success (0) when nothing was freed and there is no cap pressure (`count==0`). There is no reason to keep retrying when we are not over the cap limit and made no progress. Under real cap pressure (`count>0`) the reclaim path is unchanged and still retries via `-EAGAIN`. Without this patch, I saw 500 ceph_trim_dentries() calls per second on our web servers. This is very visible in `/proc/lock_stat` (5 minute capture): class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 126180 128218 0.04 8063.44 15986965.20 124.69 1573354 5296812 0.04 8291.28 74164526.48 14.00 ----------------------- &mdsc->dentry_list_lock 111736 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2631 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 3878 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 9973 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 ----------------------- &mdsc->dentry_list_lock 123621 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 1822 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2720 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 55 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 With this patch: class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 1203 1215 0.16 408.88 33082.88 27.23 4320501 7357389 0.04 500.64 1961578.00 0.27 ----------------------- &mdsc->dentry_list_lock 1029 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 169 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 16 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 1 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 ----------------------- &mdsc->dentry_list_lock 158 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 858 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 182 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 17 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 __dentry_leases_walk() is almost gone. The total wait time is reduced by a factor of 483. That will give some latency gains to ceph_readdir(). Cc: stable(a)vger.kernel.org Fixes: 37c4efc1ddf9 ("ceph: periodically trim stale dentries") Signed-off-by: Max Kellermann <max.kellermann(a)ionos.com> Reviewed-by: Alex Markuze <amarkuze(a)redhat.com> Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Conflicts: fs/ceph/dir.c [backport to OLK-6.6: the three upstream semantic changes are taken verbatim - __dir_lease_check() now returns KEEP instead of TOUCH when expire_dir_lease is false (moved above the round-robin comment), the first -EAGAIN is gated on freed > 0, and ceph_trim_dentries() bails out with 0 when freed == 0 && count == 0. Only the two __dentry_leases_walk() call sites had to be adapted: like linux-5.10.y, this tree still passes the check callback as a third argument, because 2a965d1b15d28 ("ceph: get rid of passing callbacks in __dentry_leases_walk()", v6.8-rc1, which is what the linux-6.12.y stable backport is based on) is a pure cleanup that is neither present in OLK-6.6 nor part of this fix's dependency chain. linux-6.6.y LTS has not taken this fix, so mainline e7d7aa7b73017 is used as the baseline.] Signed-off-by: Pan Taixi <pantaixi1(a)huawei.com> --- fs/ceph/dir.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 949c2a450cfb..5f82537ba37c 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1674,15 +1674,15 @@ static int __dir_lease_check(struct dentry *dentry, void *arg) if (ret == -EBUSY) return KEEP; if (ret > 0) { if (time_before(jiffies, di->time + lwc->dir_lease_ttl)) return STOP; + if (!lwc->expire_dir_lease) + return KEEP; /* Move dentry to tail of dir lease list if we don't want * to delete it. So dentries in the list are checked in a * round robin manner */ - if (!lwc->expire_dir_lease) - return TOUCH; if (dentry->d_lockref.count > 0 || (di->flags & CEPH_DENTRY_REFERENCED)) return TOUCH; /* invalidate dir lease */ di->lease_shared_gen = 0; @@ -1705,20 +1705,24 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc) spin_unlock(&mdsc->caps_list_lock); lwc.dir_lease = false; lwc.nr_to_scan = CEPH_CAPS_PER_RELEASE * 2; freed = __dentry_leases_walk(mdsc, &lwc, __dentry_lease_check); - if (!lwc.nr_to_scan) /* more invalid leases */ + if (freed > 0 && !lwc.nr_to_scan) /* more invalid leases */ return -EAGAIN; if (lwc.nr_to_scan < CEPH_CAPS_PER_RELEASE) lwc.nr_to_scan = CEPH_CAPS_PER_RELEASE; lwc.dir_lease = true; lwc.expire_dir_lease = freed < count; lwc.dir_lease_ttl = mdsc->fsc->mount_options->caps_wanted_delay_max * HZ; freed +=__dentry_leases_walk(mdsc, &lwc, __dir_lease_check); + if (freed == 0 && count == 0) + /* no progress possible currently, retry futile */ + return 0; + if (!lwc.nr_to_scan) /* more to check */ return -EAGAIN; return freed > 0 ? 1 : 0; } -- 2.34.1

2 1

[PATCH OLK-5.10] ceph: do not repeat ceph_trim_dentries() if no progress possible
by Pan Taixi 20 Sep '26

20 Sep '26

From: Max Kellermann <max.kellermann(a)ionos.com> mainline inclusion from mainline-v7.3-rc1 commit e7d7aa7b730178278109c41fa1b17b06873065d5 category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/19026 CVE: CVE-2026-89647 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- ceph_cap_reclaim_work() re-queues itself for as long as ceph_trim_dentries() returns -EAGAIN, which happens whenever a lease walk exhausts its `nr_to_scan` budget. This creates a busy loop that consumes CPU without making any progress when there is nothing to reclaim: with no cap pressure (`count==0`) and every scanned lease still valid, each pass runs the full scan budget down to zero and returns `-EAGAIN`, only to be queued again immediately. The dir-lease walk made this worse. When `expire_dir_lease` is `false` (i.e. we have no intention of reclaiming dir leases), __dir_lease_check() returned `TOUCH` for every valid lease. `TOUCH` moves the dentry to the tail of the list and resets `di->time` via __dentry_dir_lease_touch(), so a walk over N valid leases pointlessly rewrote the list, refreshed the timestamps (preventing them from ever aging out) and always drained `nr_to_scan`, guaranteeing the `-EAGAIN` requeue. Fix this in three steps: - Return `KEEP` instead of `TOUCH` when `expire_dir_lease` is `false`. If we are not going to reclaim the lease, leave it in place instead of churning the list and resetting its timestamp; the walk then terminates naturally (or via `STOP` at the first fresh lease). - Only return `-EAGAIN` from the first (dentry-lease) walk when something was actually freed. A full batch that frees nothing means retrying the same list immediately is futile; fall through to the dir-lease walk instead. - After both walks, bail out with success (0) when nothing was freed and there is no cap pressure (`count==0`). There is no reason to keep retrying when we are not over the cap limit and made no progress. Under real cap pressure (`count>0`) the reclaim path is unchanged and still retries via `-EAGAIN`. Without this patch, I saw 500 ceph_trim_dentries() calls per second on our web servers. This is very visible in `/proc/lock_stat` (5 minute capture): class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 126180 128218 0.04 8063.44 15986965.20 124.69 1573354 5296812 0.04 8291.28 74164526.48 14.00 ----------------------- &mdsc->dentry_list_lock 111736 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2631 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 3878 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 9973 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 ----------------------- &mdsc->dentry_list_lock 123621 [<0000000050597999>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 1822 [<000000007b11e319>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 2720 [<000000002f27cb6f>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 55 [<00000000c0022f62>] __ceph_dentry_lease_touch+0x5c/0xa8 With this patch: class name con-bounces contentions waittime-min waittime-max waittime-total waittime-avg acq-bounces acquisitions holdtime-min holdtime-max holdtime-total holdtime-avg &mdsc->dentry_list_lock: 1203 1215 0.16 408.88 33082.88 27.23 4320501 7357389 0.04 500.64 1961578.00 0.27 ----------------------- &mdsc->dentry_list_lock 1029 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 169 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 16 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 &mdsc->dentry_list_lock 1 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 ----------------------- &mdsc->dentry_list_lock 158 [<000000002038c577>] __dentry_lease_unlist+0x50/0xa0 &mdsc->dentry_list_lock 858 [<000000003c9aea8a>] __ceph_dentry_dir_lease_touch+0x7c/0xa8 &mdsc->dentry_list_lock 182 [<00000000612fe15f>] __dentry_leases_walk+0x64/0x2c8 &mdsc->dentry_list_lock 17 [<00000000c991106d>] __ceph_dentry_lease_touch+0x5c/0xa8 __dentry_leases_walk() is almost gone. The total wait time is reduced by a factor of 483. That will give some latency gains to ceph_readdir(). Cc: stable(a)vger.kernel.org Fixes: 37c4efc1ddf9 ("ceph: periodically trim stale dentries") Signed-off-by: Max Kellermann <max.kellermann(a)ionos.com> Reviewed-by: Alex Markuze <amarkuze(a)redhat.com> Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com> Conflicts: fs/ceph/dir.c [backport to OLK-5.10: the three upstream semantic changes are taken verbatim - __dir_lease_check() now returns KEEP instead of TOUCH when expire_dir_lease is false (moved above the round-robin comment), the first -EAGAIN is gated on freed > 0, and ceph_trim_dentries() bails out with 0 when freed == 0 && count == 0. Only the two __dentry_leases_walk() call sites had to be adapted: this tree still passes the check callback as a third argument, because 2a965d1b15d28 ("ceph: get rid of passing callbacks in __dentry_leases_walk()", v6.8-rc1) is a pure cleanup that is neither present in OLK-5.10 nor part of this fix's dependency chain. linux-5.10.y LTS has not taken this fix, so mainline e7d7aa7b73017 is used as the baseline.] Signed-off-by: Pan Taixi <pantaixi1(a)huawei.com> --- fs/ceph/dir.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 504693f6b72d..fafff54e30d5 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1468,15 +1468,15 @@ static int __dir_lease_check(struct dentry *dentry, void *arg) if (ret == -EBUSY) return KEEP; if (ret > 0) { if (time_before(jiffies, di->time + lwc->dir_lease_ttl)) return STOP; + if (!lwc->expire_dir_lease) + return KEEP; /* Move dentry to tail of dir lease list if we don't want * to delete it. So dentries in the list are checked in a * round robin manner */ - if (!lwc->expire_dir_lease) - return TOUCH; if (dentry->d_lockref.count > 0 || (di->flags & CEPH_DENTRY_REFERENCED)) return TOUCH; /* invalidate dir lease */ di->lease_shared_gen = 0; @@ -1499,20 +1499,24 @@ int ceph_trim_dentries(struct ceph_mds_client *mdsc) spin_unlock(&mdsc->caps_list_lock); lwc.dir_lease = false; lwc.nr_to_scan = CEPH_CAPS_PER_RELEASE * 2; freed = __dentry_leases_walk(mdsc, &lwc, __dentry_lease_check); - if (!lwc.nr_to_scan) /* more invalid leases */ + if (freed > 0 && !lwc.nr_to_scan) /* more invalid leases */ return -EAGAIN; if (lwc.nr_to_scan < CEPH_CAPS_PER_RELEASE) lwc.nr_to_scan = CEPH_CAPS_PER_RELEASE; lwc.dir_lease = true; lwc.expire_dir_lease = freed < count; lwc.dir_lease_ttl = mdsc->fsc->mount_options->caps_wanted_delay_max * HZ; freed +=__dentry_leases_walk(mdsc, &lwc, __dir_lease_check); + if (freed == 0 && count == 0) + /* no progress possible currently, retry futile */ + return 0; + if (!lwc.nr_to_scan) /* more to check */ return -EAGAIN; return freed > 0 ? 1 : 0; } -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS V1] media: saa7164: fix cleanup on resource allocation failure
by Yao Yiqi 20 Sep '26

20 Sep '26

From: Guangshuo Li <lgs201920130244(a)gmail.com> stable inclusion from stable-v6.6.157 commit 77f216f8de6221efb6f6c471bc9e598f1879825b category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/19083 CVE: CVE-2026-89877 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ------------------------------ commit 28e84c6e2e6753ed238ea097b2842a32a6a6879b upstream. saa7164_dev_setup() adds the device to the global saa7164_devlist before requesting the PCI BAR memory regions. If get_resources() fails, saa7164_dev_setup() decrements the device count and returns an error, but leaves the device on saa7164_devlist. The probe error path then frees the device, leaving a dangling entry on the global list. Reuse the existing MMIO mapping error path to remove the device from saa7164_devlist and decrement the device count before returning. Also release BAR0 if it was successfully requested but the BAR2 request fails. Fixes: 443c1228d505 ("V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon") Cc: stable(a)vger.kernel.org Signed-off-by: Guangshuo Li <lgs201920130244(a)gmail.com> Signed-off-by: Hans Verkuil <hverkuil+cisco(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: drivers/media/pci/saa7164/saa7164-core.c [ context conflicts ] Signed-off-by: Yao Yiqi <yaoyiqi3(a)huawei.com> --- drivers/media/pci/saa7164/saa7164-core.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/media/pci/saa7164/saa7164-core.c b/drivers/media/pci/saa7164/saa7164-core.c index 5102519df108..c067550db0d3 100644 --- a/drivers/media/pci/saa7164/saa7164-core.c +++ b/drivers/media/pci/saa7164/saa7164-core.c @@ -892,6 +892,9 @@ static int get_resources(struct saa7164_dev *dev) if (request_mem_region(pci_resource_start(dev->pci, 2), pci_resource_len(dev->pci, 2), dev->name)) return 0; + + release_mem_region(pci_resource_start(dev->pci, 0), + pci_resource_len(dev->pci, 0)); } printk(KERN_ERR "%s: can't get MMIO memory @ 0x%llx or 0x%llx\n", @@ -1006,6 +1009,10 @@ static int saa7164_dev_setup(struct saa7164_dev *dev) dev->name, dev->pci->subsystem_vendor, dev->pci->subsystem_device); + mutex_lock(&devlist); + list_del(&dev->devlist); + mutex_unlock(&devlist); + saa7164_devcount--; return -ENODEV; } -- 2.34.1

2 1