- Aug 12, 2021
-
-
Yu Kuai authored
[ Upstream commit 8d75d0ef ] If queue is dying while iolatency_set_limit() is in progress, blk_get_queue() won't increment the refcount of the queue. However, blk_put_queue() will still decrement the refcount later, which will cause the refcout to be unbalanced. Thus error out in such case to fix the problem. Fixes: 8c772a9b ("blk-iolatency: fix IO hang due to negative inflight counter") Signed-off-by: Yu Kuai <yukuai3@huawei.com> Acked-by: Tejun Heo <tj@kernel.org> Link: https://lore.kernel.org/r/20210805124645.543797-1-yukuai3@huawei.com Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Pavel Skripkin authored
[ Upstream commit 942e560a ] Smatch says: drivers/net/ethernet/neterion/vxge/vxge-main.c:3518 vxge_device_unregister() error: Using vdev after free_{netdev,candev}(dev); drivers/net/ethernet/neterion/vxge/vxge-main.c:3518 vxge_device_unregister() error: Using vdev after free_{netdev,candev}(dev); drivers/net/ethernet/neterion/vxge/vxge-main.c:3520 vxge_device_unregister() error: Using vdev after free_{netdev,candev}(dev); drivers/net/ethernet/neterion/vxge/vxge-main.c:3520 vxge_device_unregister() error: Using vdev after free_{netdev,candev}(dev); Since vdev pointer is netdev private data accessing it after free_netdev() call can cause use-after-free bug. Fix it by moving free_netdev() call at the end of the function Fixes: 6cca2003 ("vxge: cleanup probe error paths") Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Pavel Skripkin authored
[ Upstream commit 44712965 ] Smatch says: drivers/net/ethernet/freescale/fec_main.c:3994 fec_drv_remove() error: Using fep after free_{netdev,candev}(ndev); drivers/net/ethernet/freescale/fec_main.c:3995 fec_drv_remove() error: Using fep after free_{netdev,candev}(ndev); Since fep pointer is netdev private data, accessing it after free_netdev() call can cause use-after-free bug. Fix it by moving free_netdev() call at the end of the function Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: a31eda65 ("net: fec: fix clock count mis-match") Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Reviewed-by: Joakim Zhang <qiangqing.zhang@nxp.com> Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Pavel Skripkin authored
[ Upstream commit af35fc37 ] Syzbot reported uninit value pegasus_probe(). The problem was in missing error handling. get_interrupt_interval() internally calls read_eprom_word() which can fail in some cases. For example: failed to receive usb control message. These cases should be handled to prevent uninit value bug, since read_eprom_word() will not initialize passed stack variable in case of internal failure. Fail log: BUG: KMSAN: uninit-value in get_interrupt_interval drivers/net/usb/pegasus.c:746 [inline] BUG: KMSAN: uninit-value in pegasus_probe+0x10e7/0x4080 drivers/net/usb/pegasus.c:1152 CPU: 1 PID: 825 Comm: kworker/1:1 Not tainted 5.12.0-rc6-syzkaller #0 ... Workqueue: usb_hub_wq hub_event Call Trace: __dump_stack lib/dump_stack.c:79 [inline] dump_stack+0x24c/0x2e0 lib/dump_stack.c:120 kmsan_report+0xfb/0x1e0 mm/kmsan/kmsan_report.c:118 __msan_warning+0x5c/0xa0 mm/kmsan/kmsan_instr.c:197 get_interrupt_interval drivers/net/usb/pegasus.c:746 [inline] pegasus_probe+0x10e7/0x4080 drivers/net/usb/pegasus.c:1152 .... Local variable ----data.i@pegasus_probe created at: get_interrupt_interval drivers/net/usb/pegasus.c:1151 [inline] pegasus_probe+0xe57/0x4080 drivers/net/usb/pegasus.c:1152 get_interrupt_interval drivers/net/usb/pegasus.c:1151 [inline] pegasus_probe+0xe57/0x4080 drivers/net/usb/pegasus.c:1152 Reported-and-tested-by: <syzbot+02c9f70f3afae308464a@syzkaller.appspotmail.com> Fixes: 1da177e4 ("Linux-2.6.12-rc2") Signed-off-by: Pavel Skripkin <paskripkin@gmail.com> Link: https://lore.kernel.org/r/20210804143005.439-1-paskripkin@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Dan Carpenter authored
[ Upstream commit fb653827 ] Set the error code if bnx2x_alloc_fw_stats_mem() fails. The current code returns success. Fixes: ad5afc89 ("bnx2x: Separate VF and PF logic") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
H. Nikolaus Schaller authored
[ Upstream commit 28bbbb98 ] When cross compiling a MIPS kernel on a BSD based HOSTCC leads to errors like SYNC include/config/auto.conf.cmd - due to: .config egrep: empty (sub)expression UPD include/config/kernel.release HOSTCC scripts/dtc/dtc.o - due to target missing It turns out that egrep uses this egrep pattern: (|MINOR_|PATCHLEVEL_) This is not valid syntax or gives undefined results according to POSIX 9.5.3 ERE Grammar https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html It seems to be silently accepted by the Linux egrep implementation while a BSD host complains. Such patterns can be replaced by a transformation like "(|p1|p2)" -> "(p1|p2)?" Fixes: 48c35b2d ("[MIPS] There is no __GNUC_MAJOR__") Signed-off-by: H. Nikolaus Schaller <hns@goldelico.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Huang Pei authored
[ Upstream commit 6aa32467 ] +. According to Documentation/vm/split_page_table_lock, handle failure of pgtable_pmd_page_ctor +. Use GFP_KERNEL_ACCOUNT instead of GFP_KERNEL|__GFP_ACCOUNT +. Adjust coding style Fixes: ed914d48 ("MIPS: add PMD table accounting into MIPS') Reported-by: Joshua Kinard <kumba@gentoo.org> Signed-off-by: Huang Pei <huangpei@loongson.cn> Reviewed-by: Joshua Kinard <kumba@gentoo.org> Signed-off-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Yunsheng Lin authored
[ Upstream commit 06f5553e ] According to comment in qdisc_alloc(), sch->seqlock's lockdep class key should be set to qdisc_tx_busylock, due to possible type error, sch->busylock's lockdep class key is set to qdisc_tx_busylock, which is duplicated because sch->busylock's lockdep class key is already set in qdisc_alloc(). So fix it by replacing sch->busylock with sch->seqlock. Fixes: 96009c7d ("sched: replace __QDISC_STATE_RUNNING bit with a spin lock") Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Oleksij Rempel authored
[ Upstream commit d1a58c01 ] In case of this switch we work with 32bit registers on top of 16bit bus. Some registers (for example access to forwarding database) have trigger bit on the first 16bit half of request and the result + configuration of request in the second half. Without this patch, we would trigger database operation and overwrite result in one run. To make it work properly, we should do the second part of transfer before the first one is done. So far, this rule seems to work for all registers on this switch. Fixes: ec6698c2 ("net: dsa: add support for Atheros AR9331 built-in switch") Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Vladimir Oltean <olteanv@gmail.com> Link: https://lore.kernel.org/r/20210803063746.3600-1-o.rempel@pengutronix.de Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Antoine Tenart authored
[ Upstream commit 40391467 ] The patch fixing the returned value of ip6_skb_dst_mtu (int -> unsigned int) was rebased between its initial review and the version applied. In the meantime fade5641 was applied, which added a new variable (int) used as the returned value. This lead to a mismatch between the function prototype and the variable used as the return value. Fixes: 40fc3054 ("net: ipv6: fix return value of ip6_skb_dst_mtu") Cc: Vadim Fedorenko <vfedorenko@novek.ru> Signed-off-by: Antoine Tenart <atenart@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Fei Qin authored
[ Upstream commit 9fdc5d85 ] Pauseframe control is set to symmetric mode by default on the NFP. Pause frames can not be configured through ethtool now, but ethtool can report the supported mode. Fixes: 265aeb51 ("nfp: add support for .get_link_ksettings()") Signed-off-by: Fei Qin <fei.qin@corigine.com> Signed-off-by: Louis Peens <louis.peens@corigine.com> Signed-off-by: Simon Horman <simon.horman@corigine.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Xin Long authored
[ Upstream commit ae954bbc ] In commit 58acd100 ("sctp: update active_key for asoc when old key is being replaced"), sctp_auth_asoc_init_active_key() is called to update the active_key right after the old key is deleted and before the new key is added, and it caused that the active_key could be found with the key_id. In Ying Xu's testing, the BUG_ON in sctp_auth_asoc_init_active_key() was triggered: [ ] kernel BUG at net/sctp/auth.c:416! [ ] RIP: 0010:sctp_auth_asoc_init_active_key.part.8+0xe7/0xf0 [sctp] [ ] Call Trace: [ ] sctp_auth_set_key+0x16d/0x1b0 [sctp] [ ] sctp_setsockopt.part.33+0x1ba9/0x2bd0 [sctp] [ ] __sys_setsockopt+0xd6/0x1d0 [ ] __x64_sys_setsockopt+0x20/0x30 [ ] do_syscall_64+0x5b/0x1a0 So fix it by moving the active_key update after sh_keys is added. Fixes: 58acd100 ("sctp: update active_key for asoc when old key is being replaced") Reported-by: Ying Xu <yinxu@redhat.com> Signed-off-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Aharon Landau authored
[ Upstream commit d6793ca9 ] Fixing a typo that causes a cache entry to shrink immediately after adding to it new MRs if the entry size exceeds the high limit. In doing so, the cache misses its purpose to prevent the creation of new mkeys on the runtime by using the cached ones. Fixes: b9358bdb ("RDMA/mlx5: Fix locking in MR cache work queue") Link: https://lore.kernel.org/r/fcb546986be346684a016f5ca23a0567399145fa.1627370131.git.leonro@nvidia.com Signed-off-by: Aharon Landau <aharonl@nvidia.com> Reviewed-by: Maor Gottlieb <maorg@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Matthias Schiffer authored
[ Upstream commit 9b87f435 ] The tqmx86 MFD driver was passing IRQ 0 for "no IRQ" in the past. This causes warnings with newer kernels. Prepare the gpio-tqmx86 driver for the fixed MFD driver by handling a missing IRQ properly. Fixes: b868db94 ("gpio: tqmx86: Add GPIO from for this IO controller") Signed-off-by: Matthias Schiffer <matthias.schiffer@ew.tq-group.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Acked-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Wang Hai authored
[ Upstream commit 7fe74dfd ] Replace pci_enable_device() with pcim_enable_device(), pci_disable_device() and pci_release_regions() will be called in release automatically. Fixes: 1da177e4 ("Linux-2.6.12-rc2") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wang Hai <wanghai38@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Steve Bennett authored
[ Upstream commit a5e63c7d ] The logic for discerning between KSZ8051 and KSZ87XX PHYs is incorrect such that the that KSZ87XX switch is not identified correctly. ksz8051_ksz8795_match_phy_device() uses the parameter ksz_phy_id to discriminate whether it was called from ksz8051_match_phy_device() or from ksz8795_match_phy_device() but since PHY_ID_KSZ87XX is the same value as PHY_ID_KSZ8051, this doesn't work. Instead use a bool to discriminate the caller. Without this patch, the KSZ8795 switch port identifies as: ksz8795-switch spi3.1 ade1 (uninitialized): PHY [dsa-0.1:03] driver [Generic PHY] With the patch, it identifies correctly: ksz8795-switch spi3.1 ade1 (uninitialized): PHY [dsa-0.1:03] driver [Micrel KSZ87XX Switch] Fixes: 8b95599c ("net: phy: micrel: Discern KSZ8051 and KSZ8795 PHYs") Signed-off-by: Steve Bennett <steveb@workware.net.au> Reviewed-by: Marek Vasut <marex@denx.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Vladimir Oltean authored
[ Upstream commit 47c2c0c2 ] On SJA1105P/Q/R/S and SJA1110, the L2 Lookup Table entries contain a maskable "inner/outer tag" bit which means: - when set to 1: match single-outer and double tagged frames - when set to 0: match untagged and single-inner tagged frames - when masked off: match all frames regardless of the type of tag This driver does not make any meaningful distinction between inner tags (matches on TPID) and outer tags (matches on TPID2). In fact, all VLAN table entries are installed as SJA1110_VLAN_D_TAG, which means that they match on both inner and outer tags. So it does not make sense that we install FDB entries with the IOTAG bit set to 1. In VLAN-unaware mode, we set both TPID and TPID2 to 0xdadb, so the switch will see frames as outer-tagged or double-tagged (never inner). So the FDB entries will match if IOTAG is set to 1. In VLAN-aware mode, we set TPID to 0x8100 and TPID2 to 0x88a8. So the switch will see untagged and 802.1Q-tagged packets as inner-tagged, and 802.1ad-tagged packets as outer-tagged. So untagged and 802.1Q-tagged packets will not match FDB entries if IOTAG is set to 1, but 802.1ad tagged packets will. Strange. To fix this, simply mask off the IOTAG bit from FDB entries, and make them match regardless of whether the VLAN tag is inner or outer. Fixes: 1da73821 ("net: dsa: sja1105: Add FDB operations for P/Q/R/S series") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Vladimir Oltean authored
[ Upstream commit 589918df ] Similar but not quite the same with what was done in commit b11f0a4c ("net: dsa: sja1105: be stateless when installing FDB entries") for SJA1105E/T, it is desirable to drop the priv->vlan_aware check and simply go ahead and install FDB entries in the VLAN that was given by the bridge. As opposed to SJA1105E/T, in SJA1105P/Q/R/S and SJA1110, the FDB is a maskable TCAM, and we are installing VLAN-unaware FDB entries with the VLAN ID masked off. However, such FDB entries might completely obscure VLAN-aware entries where the VLAN ID is included in the search mask, because the switch looks up the FDB from left to right and picks the first entry which results in a masked match. So it depends on whether the bridge installs first the VLAN-unaware or the VLAN-aware FDB entries. Anyway, if we had a VLAN-unaware FDB entry towards one set of DESTPORTS and a VLAN-aware one towards other set of DESTPORTS, the result is that the packets in VLAN-aware mode will be forwarded towards the DESTPORTS specified by the VLAN-unaware entry. To solve this, simply do not use the masked matching ability of the FDB for VLAN ID, and always match precisely on it. In VLAN-unaware mode, we configure the switch for shared VLAN learning, so the VLAN ID will be ignored anyway during lookup, so it is redundant to mask it off in the TCAM. This patch conflicts with net-next commit 0fac6aa0 ("net: dsa: sja1105: delete the best_effort_vlan_filtering mode") which changed this line: if (priv->vlan_state != SJA1105_VLAN_UNAWARE) { into: if (priv->vlan_aware) { When merging with net-next, the lines added by this patch should take precedence in the conflict resolution (i.e. the "if" condition should be deleted in both cases). Fixes: 1da73821 ("net: dsa: sja1105: Add FDB operations for P/Q/R/S series") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Vladimir Oltean authored
[ Upstream commit 6c5fc159 ] The procedure to add a static FDB entry in sja1105 is concurrent with dynamic learning performed on all bridge ports and the CPU port. The switch looks up the FDB from left to right, and also learns dynamically from left to right, so it is possible that between the moment when we pick up a free slot to install an FDB entry, another slot to the left of that one becomes free due to an address ageing out, and that other slot is then immediately used by the switch to learn dynamically the same address as we're trying to add statically. The result is that we succeeded to add our static FDB entry, but it is being shadowed by a dynamic FDB entry to its left, and the switch will behave as if our static FDB entry did not exist. We cannot really prevent this from happening unless we make the entire process to add a static FDB entry a huge critical section where address learning is temporarily disabled on _all_ ports, and then re-enabled according to the configuration done by sja1105_port_set_learning. However, that is kind of disruptive for the operation of the network. What we can do alternatively is to simply read back the FDB for dynamic entries located before our newly added static one, and delete them. This will guarantee that our static FDB entry is now operational. It will still not guarantee that there aren't dynamic FDB entries to the _right_ of that static FDB entry, but at least those entries will age out by themselves since they aren't hit, and won't bother anyone. Fixes: 291d1e72 ("net: dsa: sja1105: Add support for FDB and MDB management") Fixes: 1da73821 ("net: dsa: sja1105: Add FDB operations for P/Q/R/S series") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Vladimir Oltean authored
[ Upstream commit e11e865b ] The SJA1105 switch family leaves it up to software to decide where within the FDB to install a static entry, and to concatenate destination ports for already existing entries (the FDB is also used for multicast entries), it is not as simple as just saying "please add this entry". This means we first need to search for an existing FDB entry before adding a new one. The driver currently manages to fool itself into thinking that if an FDB entry already exists, there is nothing to be done. But that FDB entry might be dynamically learned, case in which it should be replaced with a static entry, but instead it is left alone. This patch checks the LOCKEDS ("locked/static") bit from found FDB entries, and lets the code "goto skip_finding_an_index;" if the FDB entry was not static. So we also need to move the place where we set LOCKEDS = true, to cover the new case where a dynamic FDB entry existed but was dynamic. Fixes: 291d1e72 ("net: dsa: sja1105: Add support for FDB and MDB management") Fixes: 1da73821 ("net: dsa: sja1105: Add FDB operations for P/Q/R/S series") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Jakub Sitnicki authored
[ Upstream commit d51c5907 ] GSO expects inner transport header offset to be valid when skb->encapsulation flag is set. GSO uses this value to calculate the length of an individual segment of a GSO packet in skb_gso_transport_seglen(). However, tcp/udp gro_complete callbacks don't update the skb->inner_transport_header when processing an encapsulated TCP/UDP segment. As a result a GRO skb has ->inner_transport_header set to a value carried over from earlier skb processing. This can have mild to tragic consequences. From miscalculating the GSO segment length to triggering a page fault [1], when trying to read TCP/UDP header at an address past the skb->data page. The latter scenario leads to an oops report like so: BUG: unable to handle page fault for address: ffff9fa7ec00d008 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 123f201067 P4D 123f201067 PUD 123f209067 PMD 0 Oops: 0000 [#1] SMP NOPTI CPU: 44 PID: 0 Comm: swapper/44 Not tainted 5.4.53-cloudflare-2020.7.21 #1 Hardware name: HYVE EDGE-METAL-GEN10/HS-1811DLite1, BIOS V2.15 02/21/2020 RIP: 0010:skb_gso_transport_seglen+0x44/0xa0 Code: c0 41 83 e0 11 f6 87 81 00 00 00 20 74 30 0f b7 87 aa 00 00 00 0f [...] RSP: 0018:ffffad8640bacbb8 EFLAGS: 00010202 RAX: 000000000000feda RBX: ffff9fcc8d31bc00 RCX: ffff9fa7ec00cffc RDX: ffff9fa7ebffdec0 RSI: 000000000000feda RDI: 0000000000000122 RBP: 00000000000005c4 R08: 0000000000000001 R09: 0000000000000000 R10: ffff9fe588ae3800 R11: ffff9fe011fc92f0 R12: ffff9fcc8d31bc00 R13: ffff9fe0119d4300 R14: 00000000000005c4 R15: ffff9fba57d70900 FS: 0000000000000000(0000) GS:ffff9fe68df00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: ffff9fa7ec00d008 CR3: 0000003e99b1c000 CR4: 0000000000340ee0 Call Trace: <IRQ> skb_gso_validate_network_len+0x11/0x70 __ip_finish_output+0x109/0x1c0 ip_sublist_rcv_finish+0x57/0x70 ip_sublist_rcv+0x2aa/0x2d0 ? ip_rcv_finish_core.constprop.0+0x390/0x390 ip_list_rcv+0x12b/0x14f __netif_receive_skb_list_core+0x2a9/0x2d0 netif_receive_skb_list_internal+0x1b5/0x2e0 napi_complete_done+0x93/0x140 veth_poll+0xc0/0x19f [veth] ? mlx5e_napi_poll+0x221/0x610 [mlx5_core] net_rx_action+0x1f8/0x790 __do_softirq+0xe1/0x2bf irq_exit+0x8e/0xc0 do_IRQ+0x58/0xe0 common_interrupt+0xf/0xf </IRQ> The bug can be observed in a simple setup where we send IP/GRE/IP/TCP packets into a netns over a veth pair. Inside the netns, packets are forwarded to dummy device: trafgen -> [veth A]--[veth B] -forward-> [dummy] For veth B to GRO aggregate packets on receive, it needs to have an XDP program attached (for example, a trivial XDP_PASS). Additionally, for UDP, we need to enable GSO_UDP_L4 feature on the device: ip netns exec A ethtool -K AB rx-udp-gro-forwarding on The last component is an artificial delay to increase the chances of GRO batching happening: ip netns exec A tc qdisc add dev AB root \ netem delay 200us slot 5ms 10ms packets 2 bytes 64k With such a setup in place, the bug can be observed by tracing the skb outer and inner offsets when GSO skb is transmitted from the dummy device: tcp: FUNC DEV SKB_LEN NH TH ENC INH ITH GSO_SIZE GSO_TYPE ip_finish_output dumB 2830 270 290 1 294 254 1383 (tcpv4,gre,) ^^^ udp: FUNC DEV SKB_LEN NH TH ENC INH ITH GSO_SIZE GSO_TYPE ip_finish_output dumB 2818 270 290 1 294 254 1383 (gre,udp_l4,) ^^^ Fix it by updating the inner transport header offset in tcp/udp gro_complete callbacks, similar to how {inet,ipv6}_gro_complete callbacks update the inner network header offset, when skb->encapsulation flag is set. [1] https://lore.kernel.org/netdev/CAKxSbF01cLpZem2GFaUaifh0S-5WYViZemTicAg7FCHOnh6kug@mail.gmail.com/ Fixes: bf296b12 ("tcp: Add GRO support") Fixes: f993bc25 ("net: core: handle encapsulation offloads when computing segment lengths") Fixes: e20cf8d3 ("udp: implement GRO for plain UDP sockets.") Reported-by: Alex Forster <aforster@cloudflare.com> Signed-off-by: Jakub Sitnicki <jakub@cloudflare.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Juergen Borleis authored
[ Upstream commit 7199dded ] Commit dea7a9fb ("dmaengine: imx-dma: remove dma_slave_config direction usage") changes the method from a "configuration when called" to an "configuration when used". Due to this, only the cyclic DMA type gets configured correctly, while the generic DMA type is left non-configured. Without this additional call, the struct imxdma_channel::word_size member is stuck at DMA_SLAVE_BUSWIDTH_UNDEFINED and imxdma_prep_slave_sg() always returns NULL. Signed-off-by: Juergen Borleis <jbe@pengutronix.de> Fixes: dea7a9fb ("dmaengine: imx-dma: remove dma_slave_config direction usage") Link: https://lore.kernel.org/r/20210729071821.9857-1-jbe@pengutronix.de Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Marek Vasut authored
[ Upstream commit 15f68f02 ] While 7e5f3155 ("ARM: dts: stm32: Fix LED5 on STM32MP1 DHCOM PDK2") fixed the LED0 assignment on the PDK2 board, the same commit did not update the touchscreen IRQ line assignment, which is the same GPIO line, shared between the LED0 output and touchscreen IRQ input. To make this more convoluted, the same EXTI input (not the same GPIO line) is shared between Button B which is Active-Low IRQ, and touchscreen IRQ which is Edge-Falling IRQ, which cannot be used at the same time. In case the LCD board with touchscreen is in use, which is the case here, LED0 must be disabled, Button B must be polled, so the touchscreen interrupt works as it should. Update the touchscreen IRQ line assignment, disable LED0 and use polled GPIO button driver for Button B, since the DT here describes baseboard with LCD board. Fixes: 7e5f3155 ("ARM: dts: stm32: Fix LED5 on STM32MP1 DHCOM PDK2") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Patrice Chotard <patrice.chotard@foss.st.com> Cc: Patrick Delaunay <patrick.delaunay@foss.st.com> Cc: linux-stm32@st-md-mailman.stormreply.com To: linux-arm-kernel@lists.infradead.org Signed-off-by: Alexandre Torgue <alexandre.torgue@foss.st.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Marek Vasut authored
[ Upstream commit 36862c1e ] The LAN8710 Energy Detect Power Down (EDPD) functionality might cause unreliable cable detection. There are multiple accounts of this in the SMSC PHY driver patches which attempted to make EDPD reliable, however it seems there is always some sort of corner case left. Unfortunatelly, there is no errata documented which would confirm this to be a silicon bug on the LAN87xx series of PHYs (LAN8700, LAN8710, LAN8720 at least). Disable EDPD on the DHCOM SoM, just like multiple other boards already do as well, to make the cable detection reliable. Fixes: 34e0c784 ("ARM: dts: stm32: Add DH Electronics DHCOM STM32MP1 SoM and PDK2 board") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Patrice Chotard <patrice.chotard@foss.st.com> Cc: Patrick Delaunay <patrick.delaunay@foss.st.com> Cc: linux-stm32@st-md-mailman.stormreply.com To: linux-arm-kernel@lists.infradead.org Signed-off-by: Alexandre Torgue <alexandre.torgue@foss.st.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Hans Verkuil authored
[ Upstream commit c592b469 ] If a vb2_queue sets q->min_buffers_needed then when the number of queued buffers reaches q->min_buffers_needed, vb2_core_qbuf() will call the start_streaming() callback. If start_streaming() returns an error, then that error was just returned by vb2_core_qbuf(), but the buffer was still queued. However, userspace expects that if VIDIOC_QBUF fails, the buffer is returned dequeued. So if start_streaming() fails, then remove the buffer from the queue, thus avoiding this unwanted side-effect. Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Tested-by: Kieran Bingham <kieran.bingham@ideasonboard.com> Fixes: b3379c62 ("[media] vb2: only call start_streaming if sufficient buffers are queued") Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Li Manyi authored
[ Upstream commit 5c04243a ] Media event code 3 is defined in the MMC-6 spec as follows: "MediaRemoval: The media has been removed from the specified slot, and the Drive is unable to access the media without user intervention. This applies to media changers only." This indicated that treating the condition as an EJECT_REQUEST was appropriate. However, doing so had the unfortunate side-effect of causing the drive tray to be physically ejected on resume. Instead treat the event as a MEDIA_CHANGE request. Fixes: 7dd753ca ("scsi: sr: Return appropriate error code when disk is ejected") Link: https://bugzilla.kernel.org/show_bug.cgi?id=213759 Link: https://lore.kernel.org/r/20210726114913.6760-1-limanyi@uniontech.com Signed-off-by: Li Manyi <limanyi@uniontech.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Marek Vasut authored
[ Upstream commit 53ca18ac ] The spi_imx->spi_bus_clk may be uninitialized and thus also zero in mx51_ecspi_prepare_message(), which would lead to division by zero in kernel. Since bitbang .setup_transfer callback which initializes the spi_imx->spi_bus_clk is called after bitbang prepare_message callback, iterate over all the transfers in spi_message, find the one with lowest bus frequency, and use that bus frequency for the delay calculation. Note that it is not possible to move this CONFIGREG delay back into the .setup_transfer callback, because that is invoked too late, after the GPIO chipselects were already configured. Fixes: 135cbd37 ("spi: imx: mx51-ecspi: Reinstate low-speed CONFIGREG delay") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Cc: Mark Brown <broonie@kernel.org> Link: https://lore.kernel.org/r/20210726100102.5188-1-marex@denx.de Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Marek Vasut authored
[ Upstream commit 135cbd37 ] Since 00b80ac9 ("spi: imx: mx51-ecspi: Move some initialisation to prepare_message hook."), the MX51_ECSPI_CONFIG write no longer happens in prepare_transfer hook, but rather in prepare_message hook, however the MX51_ECSPI_CONFIG delay is still left in prepare_transfer hook and thus has no effect. This leads to low bus frequency operation problems described in 6fd8b850 ("spi: spi-imx: Fix out-of-order CS/SCLK operation at low speeds") again. Move the MX51_ECSPI_CONFIG write delay into the prepare_message hook as well, thus reinstating the low bus frequency fix. Fixes: 00b80ac9 ("spi: imx: mx51-ecspi: Move some initialisation to prepare_message hook.") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Cc: Mark Brown <broonie@kernel.org> Link: https://lore.kernel.org/r/20210703022300.296114-1-marex@denx.de Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Zhang Qilong authored
[ Upstream commit baa16371 ] pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. We fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced. Fixes: 4f3ceca2 ("dmaengine: stm32-dmamux: Add PM Runtime support") Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> Link: https://lore.kernel.org/r/20210607064640.121394-3-zhangqilong3@huawei.com Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Zhang Qilong authored
[ Upstream commit d54db74a ] pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. We fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced. Fixes: 48bc73ba ("dmaengine: stm32-dma: Add PM Runtime support") Fixes: 05f8740a ("dmaengine: stm32-dma: add suspend/resume power management support") Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com> Link: https://lore.kernel.org/r/20210607064640.121394-2-zhangqilong3@huawei.com Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Dmitry Osipenko authored
[ Upstream commit 2bcc025a ] Implement disable_unused() callback of tegra_clk_sdmmc_mux_ops to fix imbalanced disabling of the unused MMC clock on Tegra210 Jetson Nano. Fixes: c592c8a2 ("clk: tegra: Fix refcounting of gate clocks") Reported-by: Jon Hunter <jonathanh@nvidia.com> # T210 Nano Tested-by: Jon Hunter <jonathanh@nvidia.com> # T210 Nano Acked-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Dmitry Osipenko <digetx@gmail.com> Link: https://lore.kernel.org/r/20210717112742.7196-1-digetx@gmail.com Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Kunihiko Hayashi authored
[ Upstream commit 55f24c27 ] The function uniphier_xdmac_chan_stop() is only called in atomic state. Should use readl_poll_timeout_atomic() there instead of readl_poll_timeout(). Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Fixes: 667b9251 ("dmaengine: uniphier-xdmac: Add UniPhier external DMA controller driver") Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com> Link: https://lore.kernel.org/r/1627364852-28432-1-git-send-email-hayashi.kunihiko@socionext.com Signed-off-by: Vinod Koul <vkoul@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
H. Nikolaus Schaller authored
[ Upstream commit c68ef4ad ] This device tree include file describes a fixed-regulator connecting smps7_reg output (1.8V) to some 1.8V rail and consumers (vdds_1v8_main). This regulator does not physically exist. I assume it was introduced as a wrapper around smps7_reg to provide a speaking signal name "vdds_1v8_main" as label. This fixed-regulator without real function was not an issue in driver code until Commit 98e48cd9 ("regulator: core: resolve supply for boot-on/always-on regulators") introduced a new check for regulator initialization which makes Palmas regulator registration fail: [ 5.407712] ldo1: supplied by vsys_cobra [ 5.412748] ldo2: supplied by vsys_cobra [ 5.417603] palmas-pmic 48070000.i2c:palmas@48:palmas_pmic: failed to register 48070000.i2c:palmas@48:palmas_pmic regulator The reason is that the supply-chain of regulators is too long and goes from ldo3 through the virtual vdds_1v8_main regulator and then back to smps7. This adds a cross-dependency of probing Palmas regulators and the fixed-regulator which leads to probe deferral by the new check and is no longer resolved. Since we do not control what device tree files including this one reference (either &vdds_1v8_main or &smps7_reg or both) we keep both labels for smps7 for compatibility. Fixes: 98e48cd9 ("regulator: core: resolve supply for boot-on/always-on regulators") Signed-off-by: H. Nikolaus Schaller <hns@goldelico.com> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Dario Binacchi authored
[ Upstream commit 0162a996 ] Replace clock-name with clock-names. Fixes: 2a4117df ("ARM: dts: Fix dcan driver probe failed on am437x platform") Signed-off-by: Dario Binacchi <dariobin@libero.it> Signed-off-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Dario Binacchi authored
[ Upstream commit 24b5b197 ] Enabling the framebuffer leads to a system hang. Running, as a debug hack, the store_pan() function in drivers/video/fbdev/core/fbsysfs.c without taking the console_lock, allows to see the crash backtrace on the serial line. ~ # echo 0 0 > /sys/class/graphics/fb0/pan [ 9.719414] Unhandled exception: IPSR = 00000005 LR = fffffff1 [ 9.726937] CPU: 0 PID: 49 Comm: sh Not tainted 5.13.0-rc5 #9 [ 9.733008] Hardware name: STM32 (Device Tree Support) [ 9.738296] PC is at clk_gate_is_enabled+0x0/0x28 [ 9.743426] LR is at stm32f4_pll_div_set_rate+0xf/0x38 [ 9.748857] pc : [<0011e4be>] lr : [<0011f9e3>] psr: 0100000b [ 9.755373] sp : 00bc7be0 ip : 00000000 fp : 001f3ac4 [ 9.760812] r10: 002610d0 r9 : 01efe920 r8 : 00540560 [ 9.766269] r7 : 02e7ddb0 r6 : 0173eed8 r5 : 00000000 r4 : 004027c0 [ 9.773081] r3 : 0011e4bf r2 : 02e7ddb0 r1 : 0173eed8 r0 : 1d3267b8 [ 9.779911] xPSR: 0100000b [ 9.782719] CPU: 0 PID: 49 Comm: sh Not tainted 5.13.0-rc5 #9 [ 9.788791] Hardware name: STM32 (Device Tree Support) [ 9.794120] [<0000afa1>] (unwind_backtrace) from [<0000a33f>] (show_stack+0xb/0xc) [ 9.802421] [<0000a33f>] (show_stack) from [<0000a8df>] (__invalid_entry+0x4b/0x4c) The `pll_num' field in the post_div_data configuration contained a wrong value which also referenced an uninitialized hardware clock when clk_register_pll_div() was called. Fixes: 517633ef ("clk: stm32f4: Add post divisor for I2S & SAI PLLs") Signed-off-by: Dario Binacchi <dariobin@libero.it> Reviewed-by: Gabriel Fernandez <gabriel.fernandez@st.com> Link: https://lore.kernel.org/r/20210725160725.10788-1-dariobin@libero.it Signed-off-by: Stephen Boyd <sboyd@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
chihhao.chen authored
[ Upstream commit 4511781f ] The following scenario describes an echo test for Samsung USBC Headset (AKG) with VID/PID (0x04e8/0xa051). We first start a capture stream(USB IN transfer) in 96Khz/24bit/1ch mode. In clock find source function, we get value 0x2 for clock selector and 0x1 for clock source. Kernel-4.14 behavior Since clock source is valid so clock selector was not set again. We pass through this function and start a playback stream(USB OUT transfer) in 48Khz/32bit/2ch mode. This time we get value 0x1 for clock selector and 0x1 for clock source. Finally clock id with this setting is 0x9. Kernel-5.10 behavior Clock selector was always set one more time even it is valid. When we start a playback stream, we will get 0x2 for clock selector and 0x1 for clock source. In this case clock id becomes 0xA. This is an incorrect clock source setting and results in severe noises. We see wrong data rate in USB IN transfer. (From 288 bytes/ms becomes 144 bytes/ms) It should keep in 288 bytes/ms. This earphone works fine on older kernel version load because this is a newly-added behavior. Fixes: d2e8f641 ("ALSA: usb-audio: Explicitly set up the clock selector") Signed-off-by: chihhao.chen <chihhao.chen@mediatek.com> Link: https://lore.kernel.org/r/1627100621-19225-1-git-send-email-chihhao.chen@mediatek.com Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Pali Rohár authored
[ Upstream commit ee7ab3f2 ] Some SFP modules are not detected when i2c-fast-mode is enabled even when clock-frequency is already set to 100000. The I2C bus violates the timing specifications when run in fast mode. So disable fast mode on Turris Mox. Same change was already applied for uDPU (also Armada 3720 board with SFP) in commit fe3ec631 ("arm64: dts: uDPU: remove i2c-fast-mode"). Fixes: 7109d817 ("arm64: dts: marvell: add DTS for Turris Mox") Signed-off-by: Pali Rohár <pali@kernel.org> Reviewed-by: Marek Behún <kabel@kernel.org> Acked-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk> Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Vladimir Oltean authored
[ Upstream commit 923f9892 ] Since drivers/mmc/host/sdhci-xenon.c declares the PROBE_PREFER_ASYNCHRONOUS probe type, it is not guaranteed whether /dev/mmcblk0 will belong to sdhci0 or sdhci1. In turn, this will break booting by: root=/dev/mmcblk0p1 Fix the issue by adding aliases so that the old MMC controller indices are preserved. Fixes: 7320915c ("mmc: Set PROBE_PREFER_ASYNCHRONOUS for drivers that existed in v4.14") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Marek Vasut authored
[ Upstream commit 3d9e30a5 ] The pinctrl_power_button/pinctrl_power_out each define single GPIO pinmux, except it is exactly the other one than the matching gpio-keys and gpio-poweroff DT nodes use for that functionality. Swap the two GPIOs to correct this error. Fixes: 50d29fdb ("ARM: dts: imx53: Add power GPIOs on M53Menlo") Signed-off-by: Marek Vasut <marex@denx.de> Cc: Shawn Guo <shawnguo@kernel.org> Cc: Fabio Estevam <festevam@gmail.com> Cc: NXP Linux Team <linux-imx@nxp.com> Reviewed-by: Fabio Estevam <festevam@gmail.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-
Colin Ian King authored
[ Upstream commit 20fb7391 ] The function imx_mmdc_perf_init recently had a 3rd argument added to it but the equivalent macro was not updated and is still the older 2 argument version. Fix this by adding in the missing 3rd argumement mmdc_ipg_clk. Fixes: f07ec853 ("ARM: imx: add missing clk_disable_unprepare()") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Shawn Guo <shawnguo@kernel.org> Signed-off-by: Sasha Levin <sashal@kernel.org>
-