- 13 Apr, 2022 6 commits
-
-
Alvin Šipraga authored
This switch is not even supported, but if someone were to actually put this compatible string "realtek,rtl8366s" in their device tree, they would be greeted with a kernel panic because the probe function would dereference NULL. So let's just remove it. Link: https://lore.kernel.org/all/CACRpkdYdKZs0WExXc3=0yPNOwP+oOV60HRz7SRoGjZvYHaT=1g@mail.gmail.com/Signed-off-by:
Alvin Šipraga <alsi@bang-olufsen.dk> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Alvin Šipraga authored
The kernel test robot reported a build failure: or1k-linux-ld: drivers/net/dsa/realtek/realtek-smi.o:(.rodata+0x16c): undefined reference to `rtl8366rb_variant' ... with the following build configuration: CONFIG_NET_DSA_REALTEK=y CONFIG_NET_DSA_REALTEK_SMI=y CONFIG_NET_DSA_REALTEK_RTL8365MB=y CONFIG_NET_DSA_REALTEK_RTL8366RB=m The problem here is that the realtek-smi interface driver gets built-in, while the rtl8366rb switch subdriver gets built as a module, hence the symbol rtl8366rb_variant is not reachable when defining the OF device table in the interface driver. The Kconfig dependencies don't help in this scenario because they just say that the subdriver(s) depend on at least one interface driver. In fact, the subdrivers don't depend on the interface drivers at all, and can even be built even in their absence. Somewhat strangely, the interface drivers can also be built in the absence of any subdriver, BUT, if a subdriver IS enabled, then it must be reachable according to the linkage of the interface driver: effectively what the IS_REACHABLE() macro achieves. If it is not reachable, the above kind of linker error will be observed. Rather than papering over the above build error by simply using IS_REACHABLE(), we can do a little better and admit that it is actually the interface drivers that have a dependency on the subdrivers. So this patch does exactly that. Specifically, we ensure that: 1. The interface drivers' Kconfig symbols must have a value no greater than the value of any subdriver Kconfig symbols. 2. The subdrivers should by default enable both interface drivers, since most users probably want at least one of them; those interface drivers can be explicitly disabled however. What this doesn't do is prevent a user from building only a subdriver, without any interface driver. To that end, add an additional line of help in the menu to guide users in the right direction. Link: https://lore.kernel.org/all/202204110757.XIafvVnj-lkp@intel.com/Reported-by:
kernel test robot <lkp@intel.com> Fixes: aac94001 ("net: dsa: realtek: add new mdio interface for drivers") Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wirelessDavid S. Miller authored
Kalle Valo says: ==================== wireless fixes for v5.18 First set of fixes for v5.18. Maintainers file updates, two compilation warning fixes, one revert for ath11k and smaller fixes to drivers and stack. All the usual stuff. ==================== Signed-off-by: -
Dylan Hung authored
AST2600 MAC register 0x58 is writable only when the MAC clock is enabled. Usually, the MAC clock is enabled by the bootloader so register 0x58 is set normally when the bootloader is involved. To make ast2600 ftgmac100 work without the bootloader, postpone the register write until the clock is ready. Fixes: 137d23ce ("net: ftgmac100: Fix Aspeed ast2600 TX hang issue") Signed-off-by:
Dylan Hung <dylan_hung@aspeedtech.com> Signed-off-by:
-
Vladimir Oltean authored
This reverts commit 11fd667d. dsa_slave_change_mtu() updates the MTU of the DSA master and of the associated CPU port, but only if it detects a change to the master MTU. The blamed commit in the Fixes: tag below addressed a regression where dsa_slave_change_mtu() would return early and not do anything due to ds->ops->port_change_mtu() not being implemented. However, that commit also had the effect that the master MTU got set up to the correct value by dsa_master_setup(), but the associated CPU port's MTU did not get updated. This causes breakage for drivers that rely on the ->port_change_mtu() DSA call to account for the tagging overhead on the CPU port, and don't set up the initial MTU during the setup phase. Things actually worked before because they were in a fragile equilibrium where dsa_slave_change_mtu() was called before dsa_master_setup() was. So dsa_slave_change_mtu() could actually detect a change and update the CPU port MTU too. Restore the code to the way things used to work by reverting the reorder of dsa_tree_setup_master() and dsa_tree_setup_ports(). That change did not have a concrete motivation going for it anyway, it just looked better. Fixes: 066dfc42 ("Revert "net: dsa: stop updating master MTU from master.c"") Signed-off-by:
Vladimir Oltean <vladimir.oltean@nxp.com> Signed-off-by:
-
Martin Willi authored
The MACVLAN receive handler clones skbs to all matching source MACVLAN interfaces, before it passes the packet along to match on destination based MACVLANs. When using the MACVLAN nodst mode, passing the packet to destination based MACVLANs is omitted and the handler returns with RX_HANDLER_CONSUMED. However, the passed skb is not freed, leaking for any packet processed with the nodst option. Properly free the skb when consuming packets to fix that leak. Fixes: 427f0c8c ("macvlan: Add nodst option to macvlan type source") Signed-off-by:
Martin Willi <martin@strongswan.org> Signed-off-by:
-
- 12 Apr, 2022 14 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nfJakub Kicinski authored
Pablo Neira Ayuso says: ==================== Netfilter fixes for net 1) Fix cgroupv2 from the input path, from Florian Westphal. 2) Fix incorrect return value of nft_parse_register(), from Antoine Tenart. * git://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf: netfilter: nf_tables: nft_parse_register can return a negative value netfilter: nft_socket: make cgroup match work in input too ==================== Link: https://lore.kernel.org/r/20220412094246.448055-1-pablo@netfilter.orgSigned-off-by:
Jakub Kicinski <kuba@kernel.org>
-
Antoine Tenart authored
Since commit 6e1acfa3 ("netfilter: nf_tables: validate registers coming from userspace.") nft_parse_register can return a negative value, but the function prototype is still returning an unsigned int. Fixes: 6e1acfa3 ("netfilter: nf_tables: validate registers coming from userspace.") Signed-off-by:
Antoine Tenart <atenart@kernel.org> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org>
-
Jakub Kicinski authored
Horatiu Vultur says: ==================== net: lan966x: lan966x fixes This contains different fixes for lan966x in different areas like PTP, MAC, Switchdev and IGMP processing. ==================== Link: https://lore.kernel.org/r/20220409184143.1204786-1-horatiu.vultur@microchip.comSigned-off-by:
-
Horatiu Vultur authored
Currently when getting a new MAC is learn, the HW generates an interrupt. So then the SW will check the new entry and checks if it arrived on a correct port. If it didn't just generate a warning. But this could still crash the system. Therefore stop processing that entry when an issue is seen. Fixes: 5ccd66e0 ("net: lan966x: add support for interrupts from analyzer") Signed-off-by:
Horatiu Vultur <horatiu.vultur@microchip.com> Signed-off-by:
-
Horatiu Vultur authored
On lan966x it is not allowed to have foreign interfaces under a bridge which already contains lan966x ports. So when a port leaves the bridge it would call switchdev_bridge_port_unoffload which eventually will notify the other ports that bridge left the vlan group but that is not true because the bridge is still part of the vlan group. Therefore when a port leaves the bridge, stop generating replays because already the HW cleared after itself and the other ports don't need to do anything else. Fixes: cf2f6089 ("net: lan966x: Add support to offload the forwarding.") Signed-off-by:
-
Horatiu Vultur authored
In case an IGMP frame has a vlan tag, then the function lan966x_hw_offload couldn't figure out that is a IGMP frame. Therefore the SW thinks that the frame was already forward by the HW which is not true. Extend lan966x_hw_offload to pop the vlan tag if are any and then check for IGMP frames. Fixes: 47aeea0d ("net: lan966x: Implement the callback SWITCHDEV_ATTR_ID_BRIDGE_MC_DISABLED ") Signed-off-by:
-
Horatiu Vultur authored
The clk_per_cfg register represents the value added to the system clock for each clock cycle. The issue is that the default value is wrong, meaning that in case the DUT was a grandmaster then everone in the network was too slow. In case there was a grandmaster, then there is no issue because the DUT will configure clk_per_cfg register based on the master frequency. Fixes: d0964594 ("net: lan966x: Add support for ptp clocks") Signed-off-by:
-
Petr Malat authored
Function sctp_do_peeloff() wrongly initializes daddr of the original socket instead of the peeled off socket, which makes getpeername() return zeroes instead of the primary address. Initialize the new socket instead. Fixes: d570ee49 ("[SCTP]: Correctly set daddr for IPv6 sockets during peeloff") Signed-off-by:
Petr Malat <oss@malat.biz> Acked-by:
Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Link: https://lore.kernel.org/r/20220409063611.673193-1-oss@malat.bizSigned-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queueJakub Kicinski authored
Tony Nguyen says: ==================== Intel Wired LAN Driver Updates 2022-04-08 Alexander fixes a use after free issue with aRFS for ice driver. Mateusz reverts a commit that introduced issues related to device resets for iavf driver. * '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue: Revert "iavf: Fix deadlock occurrence during resetting VF interface" ice: arfs: fix use-after-free when freeing @rx_cpu_rmap ==================== Link: https://lore.kernel.org/r/20220408163411.2415552-1-anthony.l.nguyen@intel.comSigned-off-by:
-
Jakub Kicinski authored
Karsten Graul says: ==================== net/smc: fixes 2022-04-08 Patch 1 fixes two usages of snprintf() with non null-terminated string which results into an out-of-bounds read. Pach 2 fixes a syzbot finding where a pointer check was missed before the call to dev_name(). Patch 3 fixes a crash when already released memory is used as a function pointer. ==================== Link: https://lore.kernel.org/r/20220408151035.1044701-1-kgraul@linux.ibm.comSigned-off-by:
-
Karsten Graul authored
Child sockets may inherit the af_ops from the parent listen socket. When the listen socket is released then the af_ops of the child socket points to released memory. Solve that by restoring the original af_ops for child sockets which inherited the parent af_ops. And clear any inherited user_data of the parent socket. Fixes: 8270d9c2 ("net/smc: Limit backlog connections") Reviewed-by:
Wenjia Zhang <wenjia@linux.ibm.com> Signed-off-by:
Karsten Graul <kgraul@linux.ibm.com> Reviewed-by:
D. Wythe <alibuda@linux.alibaba.com> Signed-off-by:
-
Karsten Graul authored
dev_name() was called with dev.parent as argument but without to NULL-check it before. Solve this by checking the pointer before the call to dev_name(). Fixes: af5f60c7 ("net/smc: allow PCI IDs as ib device names in the pnet table") Reported-by: syzbot+03e3e228510223dabd34@syzkaller.appspotmail.com Signed-off-by:
-
Karsten Graul authored
Using snprintf() to convert not null-terminated strings to null terminated strings may cause out of bounds read in the source string. Therefore use memcpy() and terminate the target string with a null afterwards. Fixes: fa086662 ("net/smc: add support for user defined EIDs") Fixes: 3c572145 ("net/smc: add generic netlink support for system EID") Signed-off-by:
-
Tomas Melin authored
commit 42983885 ("net: macb: restart tx after tx used bit read") added support for restarting transmission. Restarting tx does not work in case controller asserts TXUBR interrupt and TQBP is already at the end of the tx queue. In that situation, restarting tx will immediately cause assertion of another TXUBR interrupt. The driver will end up in an infinite interrupt loop which it cannot break out of. For cases where TQBP is at the end of the tx queue, instead only clear TX_USED interrupt. As more data gets pushed to the queue, transmission will resume. This issue was observed on a Xilinx Zynq-7000 based board. During stress test of the network interface, driver would get stuck on interrupt loop within seconds or minutes causing CPU to stall. Signed-off-by:
Tomas Melin <tomas.melin@vaisala.com> Tested-by:
Claudiu Beznea <claudiu.beznea@microchip.com> Reviewed-by:
-
- 11 Apr, 2022 9 commits
-
-
Nicolas Dichtel authored
kongweibin reported a kernel panic in ip6_forward() when input interface has no in6 dev associated. The following tc commands were used to reproduce this panic: tc qdisc del dev vxlan100 root tc qdisc add dev vxlan100 root netem corrupt 5% CC: stable@vger.kernel.org Fixes: ccd27f05 ("ipv6: fix 'disable_policy' for fwd packets") Reported-by:
kongweibin <kongweibin2@huawei.com> Signed-off-by:
Nicolas Dichtel <nicolas.dichtel@6wind.com> Reviewed-by:
David Ahern <dsahern@kernel.org> Signed-off-by:
-
Lv Ruyi authored
Both of of_get_parent() and of_parse_phandle() return node pointer with refcount incremented, use of_node_put() on it to decrease refcount when done. Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Lv Ruyi <lv.ruyi@zte.com.cn> Signed-off-by:
-
Florian Westphal authored
cgroupv2 helper function ignores the already-looked up sk and uses skb->sk instead. Just pass sk from the calling function instead; this will make cgroup matching work for udp and tcp in input even when edemux did not set skb->sk already. Fixes: e0bb96db ("netfilter: nft_socket: add support for cgroupsv2") Signed-off-by:
Florian Westphal <fw@strlen.de> Tested-by:
Topi Miettinen <toiwoton@gmail.com> Signed-off-by:
-
Ben Greear authored
Don't use sizeof(pointer) when calculating scnprintf offset. Fixes: 01f84f0e ("mac80211: reduce stack usage in debugfs") Signed-off-by:
Ben Greear <greearb@candelatech.com> Link: https://lore.kernel.org/r/20220406175659.20611-1-greearb@candelatech.com [correct the Fixes tag] Signed-off-by:
Johannes Berg <johannes.berg@intel.com>
-
Rameshkumar Sundaram authored
Synchronize additions to nontrans_list of transmitting BSS with bss_lock to avoid races. Also when cfg80211_add_nontrans_list() fails __cfg80211_unlink_bss() needs bss_lock to be held (has lockdep assert on bss_lock). So protect the whole block with bss_lock to avoid races and warnings. Found during code review. Fixes: 0b8fb823 ("cfg80211: Parsing of Multiple BSSID information in scanning") Signed-off-by:
Rameshkumar Sundaram <quic_ramess@quicinc.com> Link: https://lore.kernel.org/r/1649668071-9370-1-git-send-email-quic_ramess@quicinc.comSigned-off-by:
-
Johannes Berg authored
We need this to be at least two bytes, so we can access alpha2[0] and alpha2[1]. It may be three in case some userspace used NUL-termination since it was NLA_STRING (and we also push it out with NUL-termination). Cc: stable@vger.kernel.org Reported-by:
Lee Jones <lee.jones@linaro.org> Link: https://lore.kernel.org/r/20220411114201.fd4a31f06541.Ie7ff4be2cf348d8cc28ed0d626fc54becf7ea799@changeidSigned-off-by:
-
Benedikt Spranger authored
A user may set the SO_TXTIME socket option to ensure a packet is send at a given time. The taprio scheduler has to confirm, that it is allowed to send a packet at that given time, by a check against the packet time schedule. The scheduler drop the packet, if the gates are closed at the given send time. The check, if SO_TXTIME is set, may fail since sk_flags are part of an union and the union is used otherwise. This happen, if a socket is not a full socket, like a request socket for example. Add a check to verify, if the union is used for sk_flags. Fixes: 4cfd5779 ("taprio: Add support for txtime-assist mode") Signed-off-by:
Benedikt Spranger <b.spranger@linutronix.de> Reviewed-by:
Kurt Kanzenbach <kurt@linutronix.de> Acked-by:
Vinicius Costa Gomes <vinicius.gomes@intel.com> Signed-off-by:
-
Dinh Nguyen authored
When using a fixed-link, the altr_tse_pcs driver crashes due to null-pointer dereference as no phy_device is provided to tse_pcs_fix_mac_speed function. Fix this by adding a check for phy_dev before calling the tse_pcs_fix_mac_speed() function. Also clean up the tse_pcs_fix_mac_speed function a bit. There is no need to check for splitter_base and sgmii_adapter_base because the driver will fail if these 2 variables are not derived from the device tree. Fixes: fb3bbdb8 ("net: ethernet: Add TSE PCS support to dwmac-socfpga") Signed-off-by:
Dinh Nguyen <dinguyen@kernel.org> Signed-off-by:
-
Arun Ramadoss authored
When the T1 phy master/slave state is changed, at the end of config_aneg function genphy_softreset is called. After the reset all the registers configured during the config_init are restored to default value. To avoid this, removed the genphy_softreset call. v1->v2 ------ Added the author in cc Fixes: 8a1b415d ("net: phy: added ethtool master-slave configuration support") Signed-off-by:
Arun Ramadoss <arun.ramadoss@microchip.com> Signed-off-by:
-
- 10 Apr, 2022 3 commits
-
-
Toke Høiland-Jørgensen authored
The ieee80211_tx_info_clear_status() helper also clears the rate counts and the driver-private part of struct ieee80211_tx_info, so using it breaks quite a few other things. So back out of using it, and instead define a ath-internal helper that only clears the area between the status_driver_data and the rates info. Combined with moving the ath_frame_info struct to status_driver_data, this avoids clearing anything we shouldn't be, and so we can keep the existing code for handling the rate information. While fixing this I also noticed that the setting of tx_info->status.rates[tx_rateindex].count on hardware underrun errors was always immediately overridden by the normal setting of the same fields, so rearrange the code so that the underrun detection actually takes effect. The new helper could be generalised to a 'memset_between()' helper, but leave it as a driver-internal helper for now since this needs to go to stable. Cc: stable@vger.kernel.org Reported-by:
Peter Seiderer <ps.report@gmx.net> Fixes: 037250f0 ("ath9k: Properly clear TX status area before reporting to mac80211") Signed-off-by:
Toke Høiland-Jørgensen <toke@redhat.com> Reviewed-by:
Kalle Valo <kvalo@kernel.org> Link: https://lore.kernel.org/r/20220404204800.2681133-1-toke@toke.dk
-
Borislav Petkov authored
Fix: drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’: drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17): ^~~~ drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13): ^~~~ See https://lore.kernel.org/r/YkwQ6%2BtIH8GQpuct@zn.tnic for the gory details as to why it triggers with older gccs only. Signed-off-by:Borislav Petkov <bp@suse.de> Cc: Arend van Spriel <aspriel@gmail.com> Cc: Franky Lin <franky.lin@broadcom.com> Cc: Hante Meuleman <hante.meuleman@broadcom.com> Cc: Kalle Valo <kvalo@kernel.org> Cc: "David S. Miller" <davem@davemloft.net> Cc: Jakub Kicinski <kuba@kernel.org> Cc: brcm80211-dev-list.pdl@broadcom.com Cc: netdev@vger.kernel.org Acked-by:
Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by:
-
Borislav Petkov authored
Fix: drivers/net/wireless/mediatek/mt76/mt76x2/pci.c: In function ‘mt76x2e_probe’: ././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_946’ \ declared with attribute error: FIELD_PREP: mask is not constant _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__) See https://lore.kernel.org/r/YkwQ6%2BtIH8GQpuct@zn.tnic for the gory details as to why it triggers with older gccs only. Signed-off-by:
-
- 09 Apr, 2022 2 commits
-
-
Kai-Heng Feng authored
UBSAN warnings are observed on atlantic driver: [ 294.432996] UBSAN: array-index-out-of-bounds in /build/linux-Qow4fL/linux-5.15.0/drivers/net/ethernet/aquantia/atlantic/aq_nic.c:484:48 [ 294.433695] index 8 is out of range for type 'aq_vec_s *[8]' The ring is dereferenced right before breaking out the loop, to prevent that from happening, only use the index in the loop to fix the issue. BugLink: https://bugs.launchpad.net/bugs/1958770Tested-by:
Mario Limonciello <mario.limonciello@amd.com> Signed-off-by:
Kai-Heng Feng <kai.heng.feng@canonical.com> Reviewed-by:
Igor Russkikh <irusskikh@marvell.com> Link: https://lore.kernel.org/r/20220408022204.16815-1-kai.heng.feng@canonical.comSigned-off-by:
-
Michael Walle authored
The DSA master might not have been probed yet in which case the probe of the felix switch fails with -EPROBE_DEFER: [ 4.435305] mscc_felix 0000:00:00.5: Failed to register DSA switch: -517 It is not an error. Use dev_err_probe() to demote this particular error to a debug message. Fixes: 56051948 ("net: dsa: ocelot: add driver for Felix switch family") Signed-off-by:
Michael Walle <michael@walle.cc> Reviewed-by:
-
- 08 Apr, 2022 6 commits
-
-
Marcelo Ricardo Leitner authored
Currently, when inserting a new filter that needs to sit at the head of chain 0, it will first update the heads pointer on all devices using the (shared) block, and only then complete the initialization of the new element so that it has a "next" element. This can lead to a situation that the chain 0 head is propagated to another CPU before the "next" initialization is done. When this race condition is triggered, packets being matched on that CPU will simply miss all other filters, and will flow through the stack as if there were no other filters installed. If the system is using OVS + TC, such packets will get handled by vswitchd via upcall, which results in much higher latency and reordering. For other applications it may result in packet drops. This is reproducible with a tc only setup, but it varies from system to system. It could be reproduced with a shared block amongst 10 veth tunnels, and an ingress filter mirroring packets to another veth. That's because using the last added veth tunnel to the shared block to do the actual traffic, it makes the race window bigger and easier to trigger. The fix is rather simple, to just initialize the next pointer of the new filter instance (tp) before propagating the head change. The fixes tag is pointing to the original code though this issue should only be observed when using it unlocked. Fixes: 2190d1d0 ("net: sched: introduce helpers to work with filter chains") Signed-off-by:
Vlad Buslov <vladbu@nvidia.com> Reviewed-by:
Davide Caratti <dcaratti@redhat.com> Link: https://lore.kernel.org/r/b97d5f4eaffeeb9d058155bcab63347527261abf.1649341369.git.marcelo.leitner@gmail.comSigned-off-by:
-
Xin Long authored
Yi Chen reported an unexpected sctp connection abort, and it occurred when COOKIE_ECHO is bundled with DATA Fragment by SCTP HW GSO. As the IP header is included in chunk->head_skb instead of chunk->skb, it failed to check IP header version in security_sctp_assoc_request(). According to Ondrej, SELinux only looks at IP header (address and IPsec options) and XFRM state data, and these are all included in head_skb for SCTP HW GSO packets. So fix it by using head_skb when calling security_sctp_assoc_request() in processing COOKIE_ECHO. v1->v2: - As Ondrej noticed, chunk->head_skb should also be used for security_sctp_assoc_established() in sctp_sf_do_5_1E_ca(). Fixes: e215dab1 ("security: call security_sctp_assoc_request in sctp_sf_do_5_1D_ce") Reported-by:Yi Chen <yiche@redhat.com> Signed-off-by:
Xin Long <lucien.xin@gmail.com> Reviewed-by:
Ondrej Mosnacek <omosnace@redhat.com> Acked-by:
-
Vadim Pasternak authored
Add mutex_destroy() call in driver initialization error flow. Fixes: 6882b0ae ("mlxsw: Introduce support for I2C bus") Signed-off-by:
Vadim Pasternak <vadimp@nvidia.com> Signed-off-by:
Ido Schimmel <idosch@nvidia.com> Link: https://lore.kernel.org/r/20220407070703.2421076-1-idosch@nvidia.comSigned-off-by:
-
Vladimir Oltean authored
When a driver for an interrupt controller is missing, of_irq_get() returns -EPROBE_DEFER ad infinitum, causing fwnode_mdiobus_phy_device_register(), and ultimately, the entire of_mdiobus_register() call, to fail. In turn, any phy_connect() call towards a PHY on this MDIO bus will also fail. This is not what is expected to happen, because the PHY library falls back to poll mode when of_irq_get() returns a hard error code, and the MDIO bus, PHY and attached Ethernet controller work fine, albeit suboptimally, when the PHY library polls for link status. However, -EPROBE_DEFER has special handling given the assumption that at some point probe deferral will stop, and the driver for the supplier will kick in and create the IRQ domain. Reasons for which the interrupt controller may be missing: - It is not yet written. This may happen if a more recent DT blob (with an interrupt-parent for the PHY) is used to boot an old kernel where the driver didn't exist, and that kernel worked with the vintage-correct DT blob using poll mode. - It is compiled out. Behavior is the same as above. - It is compiled as a module. The kernel will wait for a number of seconds specified in the "deferred_probe_timeout" boot parameter for user space to load the required module. The current default is 0, which times out at the end of initcalls. It is possible that this might cause regressions unless users adjust this boot parameter. The proposed solution is to use the driver_deferred_probe_check_state() helper function provided by the driver core, which gives up after some -EPROBE_DEFER attempts, taking "deferred_probe_timeout" into consideration. The return code is changed from -EPROBE_DEFER into -ENODEV or -ETIMEDOUT, depending on whether the kernel is compiled with support for modules or not. Fixes: 66bdede4 ("of_mdio: Fix broken PHY IRQ in case of probe deferral") Suggested-by:
Robin Murphy <robin.murphy@arm.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Link: https://lore.kernel.org/r/20220407165538.4084809-1-vladimir.oltean@nxp.comSigned-off-by:
-
Mateusz Palczewski authored
This change caused a regression with resetting while changing network namespaces. By clearing the IFF_UP flag, the kernel now thinks it has fully closed the device. This reverts commit 0cc318d2. Fixes: 0cc318d2 ("iavf: Fix deadlock occurrence during resetting VF interface") Signed-off-by:
Mateusz Palczewski <mateusz.palczewski@intel.com> Tested-by:
Konrad Jankowski <konrad0.jankowski@intel.com> Signed-off-by:
Tony Nguyen <anthony.l.nguyen@intel.com>
-
Alexander Lobakin authored
The CI testing bots triggered the following splat: [ 718.203054] BUG: KASAN: use-after-free in free_irq_cpu_rmap+0x53/0x80 [ 718.206349] Read of size 4 at addr ffff8881bd127e00 by task sh/20834 [ 718.212852] CPU: 28 PID: 20834 Comm: sh Kdump: loaded Tainted: G S W IOE 5.17.0-rc8_nextqueue-devqueue-02643-g23f3121aca93 #1 [ 718.219695] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0012.070720200218 07/07/2020 [ 718.223418] Call Trace: [ 718.227139] [ 718.230783] dump_stack_lvl+0x33/0x42 [ 718.234431] print_address_description.constprop.9+0x21/0x170 [ 718.238177] ? free_irq_cpu_rmap+0x53/0x80 [ 718.241885] ? free_irq_cpu_rmap+0x53/0x80 [ 718.245539] kasan_report.cold.18+0x7f/0x11b [ 718.249197] ? free_irq_cpu_rmap+0x53/0x80 [ 718.252852] free_irq_cpu_rmap+0x53/0x80 [ 718.256471] ice_free_cpu_rx_rmap.part.11+0x37/0x50 [ice] [ 718.260174] ice_remove_arfs+0x5f/0x70 [ice] [ 718.263810] ice_rebuild_arfs+0x3b/0x70 [ice] [ 718.267419] ice_rebuild+0x39c/0xb60 [ice] [ 718.270974] ? asm_sysvec_apic_timer_interrupt+0x12/0x20 [ 718.274472] ? ice_init_phy_user_cfg+0x360/0x360 [ice] [ 718.278033] ? delay_tsc+0x4a/0xb0 [ 718.281513] ? preempt_count_sub+0x14/0xc0 [ 718.284984] ? delay_tsc+0x8f/0xb0 [ 718.288463] ice_do_reset+0x92/0xf0 [ice] [ 718.292014] ice_pci_err_resume+0x91/0xf0 [ice] [ 718.295561] pci_reset_function+0x53/0x80 <...> [ 718.393035] Allocated by task 690: [ 718.433497] Freed by task 20834: [ 718.495688] Last potentially related work creation: [ 718.568966] The buggy address belongs to the object at ffff8881bd127e00 which belongs to the cache kmalloc-96 of size 96 [ 718.574085] The buggy address is located 0 bytes inside of 96-byte region [ffff8881bd127e00, ffff8881bd127e60) [ 718.579265] The buggy address belongs to the page: [ 718.598905] Memory state around the buggy address: [ 718.601809] ffff8881bd127d00: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc [ 718.604796] ffff8881bd127d80: 00 00 00 00 00 00 00 00 00 00 fc fc fc fc fc fc [ 718.607794] >ffff8881bd127e00: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc [ 718.610811] ^ [ 718.613819] ffff8881bd127e80: 00 00 00 00 00 00 00 00 00 00 00 00 fc fc fc fc [ 718.617107] ffff8881bd127f00: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc This is due to that free_irq_cpu_rmap() is always being called *after* (devm_)free_irq() and thus it tries to work with IRQ descs already freed. For example, on device reset the driver frees the rmap right before allocating a new one (the splat above). Make rmap creation and freeing function symmetrical with {request,free}_irq() calls i.e. do that on ifup/ifdown instead of device probe/remove/resume. These operations can be performed independently from the actual device aRFS configuration. Also, make sure ice_vsi_free_irq() clears IRQ affinity notifiers only when aRFS is disabled -- otherwise, CPU rmap sets and clears its own and they must not be touched manually. Fixes: 28bf2672 ("ice: Implement aRFS") Co-developed-by:Ivan Vecera <ivecera@redhat.com> Signed-off-by:
Alexander Lobakin <alexandr.lobakin@intel.com> Tested-by:
-
