- 22 Jan, 2015 29 commits
-
-
Arseny Solokha authored
commit 56abcab8 upstream. Commit 8dccddbc ("OHCI: final fix for NVIDIA problems (I hope)") introduced into 3.1.9 broke boot on e.g. Freescale P2020DS development board. The code path that was previously specific to NVIDIA controllers had then become taken for all chips. However, the M5237 installed on the board wedges solid when accessing its base+OHCI_FMINTERVAL register, making it impossible to boot any kernel newer than 3.1.8 on this particular and apparently other similar machines. Don't readl() and writel() base+OHCI_FMINTERVAL on PCI ID 10b9:5237. The patch is suitable for the -next tree as well as all maintained kernels up to 3.2 inclusive. Signed-off-by:
Arseny Solokha <asolokha@kb.kras.ru> Acked-by:
Alan Stern <stern@rowland.harvard.edu> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
Luis Henriques <luis.henriques@canonical.com>
-
Hans de Goede authored
commit 3ca8c717 upstream. Just like all previous UAS capable Seagate disk enclosures, these need the US_FL_NO_ATA_1X to not crash when udev probes them. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Signed-off-by:
-
Hans de Goede authored
commit c6fa3945 upstream. Like the JMicron JMS567 enclosures with the JMS566 choke on report-opcodes, so avoid it. Tested-and-reported-by:
Takeo Nakayama <javhera@gmx.com> Signed-off-by:
-
Hans de Goede authored
commit e5797a3d upstream. This is yet another Seagate device which needs the US_FL_NO_ATA_1X quirk Reported-by:
Marcin Zajączkowski <mszpak@wp.pl> Signed-off-by:
-
Hans de Goede authored
commit 7f5c4d63 upstream. Streams do not work reliabe on Fresco Logic FL1000G xhci controllers, trying to use them results in errors like this: 21:37:33 kernel: xhci_hcd 0000:04:00.0: ERROR Transfer event for disabled endpoint or incorrect stream ring 21:37:33 kernel: xhci_hcd 0000:04:00.0: @00000000368b3570 9067b000 00000000 05000000 01078001 21:37:33 kernel: xhci_hcd 0000:04:00.0: ERROR Transfer event for disabled endpoint or incorrect stream ring 21:37:33 kernel: xhci_hcd 0000:04:00.0: @00000000368b3580 9067b400 00000000 05000000 01038001 As always I've ordered a pci-e addon card with a Fresco Logic controller for myself to see if I can come up with a better fix then the big hammer, in the mean time this will make uas devices work again (in usb-storage mode) for FL1000G users. Reported-by:
-
Alan Stern authored
commit 6d89252a upstream. Commit c3ee9b76 (EHCI: improved logic for isochronous scheduling) introduced the idea of using ehci->last_iso_frame as the origin (or base) for the circular calculations involved in modifying the isochronous schedule. However, the new code it added used ehci->last_iso_frame before the value was properly initialized. This patch rectifies the mistake by moving the initialization lines earlier in iso_stream_schedule(). This fixes Bugzilla #72891. Signed-off-by:
Joe Bryant <tenminjoe@yahoo.com> Tested-by:
Martin Long <martin@longhome.co.uk> Signed-off-by:
-
Mathias Nyman authored
commit f161ead7 upstream. Solves xhci error cases with debug messages: xhci_hcd 0000:00:14.0: Setup ERROR: setup context command for slot 1. usb 1-6: hub failed to enable device, error -22 xhci will give a context state error if we try to set a slot in default state to the same default state with a special address device command. Turns out this happends in several cases: - retry reading the device rescriptor in hub_port_init() - usb_reset_device() is called for a slot in default state - in resume path, usb_port_resume() calls hub_port_init() The default state is usually reached from most states with a reset device command without any context state errors, but using the address device command with BSA bit set (block set address) only works from the enabled state and will otherwise cause context error. solve this by checking if we are already in the default state before issuing a address device BSA=1 command. Fixes: 48fc7dbd ("usb: xhci: change enumeration scheme to 'new scheme'") Signed-off-by:
Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by:
-
Jisheng Zhang authored
commit 5138d5c5 upstream. The gpio4 and gpio5 are in 0xf7fc0000 apb which is located in the SM domain. This patch moves gpio4 and gpio5 to the correct location. This patch also renames them as the following to match the names we internally used in marvell: gpio4 -> sm_gpio1 gpio5 -> sm_gpio0 porte -> portf portf -> porte This also matches what we did for BG2 and BG2CD's SM GPIO. Fixes: cedf57fc ("ARM: dts: berlin: add the BG2Q GPIO nodes") Signed-off-by:
Jisheng Zhang <jszhang@marvell.com> Signed-off-by:
Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com> [ luis: backported to 3.16: adjusted context ] Signed-off-by:
-
Jisheng Zhang authored
commit 5adba7c2 upstream. There's no card detection for the eMMC, so this patch adds the missing broken-cd property. This patch also sets bus width as 8 to add MMC_CAP_8_BIT_DATA in the Host capabilities. Fixes: 3047086d ("ARM: dts: berlin: enable SD card reader and eMMC for the BG2Q DMP") Signed-off-by:
-
Jisheng Zhang authored
commit 96ed6046 upstream. On BG2Q, the sdhci2 host uses nfcecc for "io" clk and nfc for "core" clk. The shdci2 can't work without this patch due to the "core" clk is gated. Fixes: 0d859a6a ("ARM: dts: berlin: add the SDHCI nodes for the BG2Q") Signed-off-by:
-
Reinhard Speyerer authored
commit d80c0d14 upstream. As has been discussed in the thread starting with https://lkml.kernel.org/g/549748e9.d+SiJzqu50f1r4lSAL043YSc@arcor.de Sierra Wireless MC73xx devices with USB VID/PID 0x1199:0x68c0 require the option_send_setup() code to be used on the USB interface for the AT port to make unsolicited response codes work correctly. Move these devices from the qcserial driver where they have been added by commit 70a3615f ("usb: qcserial: add Sierra Wireless MC73xx") to the option driver and add a MC73xx-specific blacklist to ensure that 1. the sendsetup code is not used for the DIAG/DM and NMEA interfaces 2. the option driver does not attach to the QMI/network interfaces Signed-off-by:
Reinhard Speyerer <rspmn@arcor.de> Signed-off-by:
Johan Hovold <johan@kernel.org> Signed-off-by:
-
Nicholas Bellinger authored
commit 46243860 upstream. While looking at hch's recent conversion to drop the MSG_*_TAG definitions, I noticed a long standing bug in vhost-scsi where the VIRTIO_SCSI_S_* attribute definitions where incorrectly being passed directly into target_submit_cmd_map_sgls(). This patch adds the missing virtio-scsi to TCM/SAM task attribute conversion. Cc: Christoph Hellwig <hch@lst.de> Cc: Michael S. Tsirkin <mst@redhat.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by:
Nicholas Bellinger <nab@linux-iscsi.org> [ luis: backported to 3.16: - replaced TCM_*_TAG by MSG_*_TAG ] Signed-off-by:
-
David Peterson authored
commit 1ae78a48 upstream. Added virtual com port VID/PID entries for CEL USB sticks and MeshWorks devices. Signed-off-by:
David Peterson <david.peterson@cel.com> Signed-off-by:
-
Fabio Estevam authored
commit 7a9f0604 upstream. GPIO2_5 is the reset GPIO for the USB3317 ULPI PHY. Instead of modelling it as a regulator, the correct approach is to use the 'reset_gpios' property of the "usb-nop-xceiv" node. GPIO1_7 is the reset GPIO for the USB2517 USB hub. As we currently don't have dt bindings to describe a HUB reset, let's keep using the regulator approach. Rename the regulator to 'reg_hub_reset' to better describe its function and bind it with the USB host1 port instead. USB host support has been introduced by commit 9bf206a9 ("ARM: dts: imx51-babbage: Add USB Host1 support"), which landed in 3.16 and it seems that USB has only been functional due to previous bootloader initialization. With this patch applied we can get USB host to work without relying on the bootloader. Signed-off-by:
Fabio Estevam <fabio.estevam@freescale.com> Signed-off-by:
Shawn Guo <shawn.guo@linaro.org> Signed-off-by:
-
Trond Myklebust authored
commit 1fc0703a upstream. Currently, our trunking code will check for session trunking, but will fail to detect client id trunking. This is a problem, because it means that the client will fail to recognise that the two connections represent shared state, even if they do not permit a shared session. By removing the check for the server minor id, and only checking the major id, we will end up doing the right thing in both cases: we close down the new nfs_client and fall back to using the existing one. Fixes: 05f4c350 ("NFS: Discover NFSv4 server trunking when mounting") Cc: Chuck Lever <chuck.lever@oracle.com> Tested-by:
Chuck Lever <chuck.lever@oracle.com> Signed-off-by:
Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by:
-
Trond Myklebust authored
commit 06bed7d1 upstream. This commit fixes a race whereby nlmclnt_init() first starts the lockd daemon, and then calls nlm_bind_host() with the expectation that nlmsvc_timeout has already been initialised. Unfortunately, there is no no synchronisation between lockd() and lockd_up() to guarantee that this is the case. Fix is to move the initialisation of nlmsvc_timeout into lockd_create_svc Fixes: 9a1b6bf8 ("LOCKD: Don't call utsname()->nodename...") Cc: Bruce Fields <bfields@fieldses.org> Signed-off-by:
-
Lennart Sorensen authored
commit 999f934d upstream. If the boot loader enables HYP mode on the boot CPU, the secondary CPU also needs to call into the ROM to switch to HYP mode before booting. The firmwares on the omap5 and dra7xx unfortunately do not take care of this, so it has to be handled by the kernel. This patch is based on "[PATCH 2/2] ARM: OMAP5: Add HYP mode entry support for secondary CPUs" by Santosh Shilimkar <santosh.shilimkar@ti.com>, except this version does not require a compile time CONFIG to control if it should enable HYP mode or not, it simply does it based on the mode of the boot CPU, so it works whether the CPU boots in SVC or HYP mode, and should even work as a guest kernel inside kvm if qemu decides to support emulating the omap5 or dra7xx. Signed-off-by:
Len Sorensen <lsorense@csclub.uwaterloo.ca> Signed-off-by:
Tony Lindgren <tony@atomide.com> Signed-off-by:
-
Lennart Sorensen authored
commit 572b24e6 upstream. The switch statement of the possible list of SYSCLK1 frequencies is missing a 0 in 4 out of the 7 frequencies. Fixes: fa6d79d2 ("ARM: OMAP: Add initialisation for the real-time counter") Signed-off-by:
Lokesh Vutla <lokeshvutla@ti.com> Acked-by:
Nishanth Menon <nm@ti.com> Signed-off-by:
-
Fabio Estevam authored
commit 7c168ed8 upstream. The default qspi2_clk_sel field of register CCM_CS2CDR contains '110' which is marked as 'reserved', so we can't rely on the default value. Provide a proper parent for QSPI clocks to avoid a kernel oops: [ 1.037920] Division by zero in kernel. [ 1.041807] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 3.18.0-rc7-next-20141204-00002-g5aa23e1 #2143 [ 1.050967] Hardware name: Freescale i.MX6 SoloX (Device Tree) [ 1.056853] Backtrace: [ 1.059360] [<80011ea0>] (dump_backtrace) from [<8001203c>] (show_stack+0x18/0x1c) [ 1.066982] r6:00000000 r5:00000000 r4:00000000 r3:00000000 [ 1.072754] [<80012024>] (show_stack) from [<806b7100>] (dump_stack+0x88/0xa4) [ 1.080038] [<806b7078>] (dump_stack) from [<80011d20>] (__div0+0x18/0x20) [ 1.086958] r5:be018500 r4:be017c00 [ 1.090600] [<80011d08>] (__div0) from [<802aa418>] (Ldiv0+0x8/0x10) [ 1.097012] [<80504fbc>] (clk_divider_set_rate) from [<80503ddc>] (clk_change_rate+0x14c/0x17c) [ 1.105759] r7:00000000 r6:00000000 r5:be018500 r4:00000000 [ 1.111516] [<80503c90>] (clk_change_rate) from [<80503ea0>] (clk_set_rate+0x94/0x98) [ 1.119391] r8:be7e0368 r7:00000000 r6:be11a000 r5:be018500 r4:00000000 r3:00000000 [ 1.127290] [<80503e0c>] (clk_set_rate) from [<80410558>] (fsl_qspi_probe+0x23c/0x75c) [ 1.135260] r5:be11a010 r4:be350010 [ 1.138900] [<8041031c>] (fsl_qspi_probe) from [<80385a18>] (platform_drv_probe+0x50/0xac) Signed-off-by:
-
Eyal Shapira authored
commit c93edc63 upstream. commit 5c904224 "iwlwifi: mvm: don't allow diversity if BT Coex / TT forbid it" broke Rx with 2 chains for diversity. This had an impact on throughput where we're using only a single stream (11a/b/g APs, single stream APs, static SMPS). Fixes: 5c904224 ("iwlwifi: mvm: don't allow diversity if BT Coex / TT forbid it") Signed-off-by:
Eyal Shapira <eyalx.shapira@intel.com> Signed-off-by:
Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by:
-
Johan Hovold authored
commit b5122236 upstream. Fix null-pointer dereference during probe if the interface-status completion handler is called before the individual ports have been set up. Fixes: f79b2d0f ("USB: keyspan: fix NULL-pointer dereferences and memory leaks") Reported-by:
Richard <richjunk@pacbell.net> Tested-by:
-
Preston Fick authored
commit 90441b4d upstream. Fixing typo for MeshConnect IDs. The original PID (0x8875) is not in production and is not needed. Instead it has been changed to the official production PID (0x8857). Signed-off-by:
Preston Fick <pffick@gmail.com> Signed-off-by:
-
Fabio Estevam authored
commit 7a87e9cb upstream. From Documentation/devicetree/bindings/clock/imx25-clock.txt: cspi1_ipg 78 cspi2_ipg 79 cspi3_ipg 80 , so fix the SPI1 clocks accordingly to avoid a kernel hang when trying to access SPI1. Signed-off-by:
-
Gary Bisson authored
commit 81ef4479 upstream. The post dividers do not work on i.MX6Q rev T0 1.0 so they must be fixed to 1. As the table index was wrong, a divider a of 4 could still be requested which implied the clock not to be set properly. This is the root cause of the HDMI not working at high resolution on rev T0 1.0 of the SoC. Signed-off-by:
Gary Bisson <bisson.gary@gmail.com> Signed-off-by:
-
Dmitry Voytik authored
commit d2a10a17 upstream. Drop unnecessary semicolon after closing curly bracket. Signed-off-by:
Dmitry Voytik <voytikd@gmail.com> Signed-off-by:
-
Bo Shen authored
commit 6785a103 upstream. When receive data, the RXRDY in status register set by hardware after a new packet has been stored in the endpoint FIFO. When it is copied from FIFO, this bit is cleared which make the FIFO can be accessed again. In the receive_data() function, this bit RXRDY has been cleared. So, after the receive_data() function return, this bit should not be cleared again, or else it may cause the accessing FIFO corrupt, which will make the data loss. Fixes: 914a3f3b (USB: add atmel_usba_udc driver) Acked-by:
Nicolas Ferre <nicolas.ferre@atmel.com> Signed-off-by:
Bo Shen <voice.shen@atmel.com> Signed-off-by:
Felipe Balbi <balbi@ti.com> [ luis: backported to 3.16: - file rename: drivers/usb/gadget/udc/atmel_usba_udc.c -> drivers/usb/gadget/atmel_usba_udc.c ] Signed-off-by:
-
Bo Shen authored
commit f40afddd upstream. According to the datasheet, when transfer using DMA, the control setting for IN packet only need END_BUF_EN, END_BUF_IE, CH_EN, while for OUT packet, need more two bits END_TR_EN and END_TR_IE to be configured. Fixes: 914a3f3b (USB: add atmel_usba_udc driver) Acked-by:
-
Mario Schuknecht authored
commit b44be246 upstream. Commit 3b74c73f switched over to memdup_user() in ep_write() function and removed kfree (kbuf). memdup_user() function allocates memory which is never freed. Fixes: 3b74c73f (usb: gadget: inode: switch over to memdup_user()) Signed-off-by:
Mario Schuknecht <mario.schuknecht@dresearch-fe.de> Signed-off-by:
-
Sebastian Andrzej Siewior authored
commit 68693b8e upstream. since the split of host+gadget mode in commit 74c2e936 ("usb: musb: factor out hcd initalization") we leak the usb_hcd struct. We call now musb_host_cleanup() which does basically usb_remove_hcd() and also sets the hcd variable to NULL. Doing so makes the finall call to musb_host_free() basically a nop and the usb_hcd remains around for ever without anowner. This patch drops that NULL assignment for that reason. Fixes: 74c2e936 ("usb: musb: factor out hcd initalization") Cc: Daniel Mack <zonque@gmail.com> Signed-off-by:
Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by:
-
- 21 Jan, 2015 2 commits
-
-
David Vrabel authored
commit 11d3d2a1 upstream. Commit 97a6d1bb (xen-netfront: Fix handling packets on compound pages with skb_linearize) attempted to fix a problem where an skb that would have required too many slots would be dropped causing TCP connections to stall. However, it filled in the first slot using the original buffer and not the new one and would use the wrong offset and grant access to the wrong page. Netback would notice the malformed request and stop all traffic on the VIF, reporting: vif vif-3-0 vif3.0: txreq.offset: 85e, size: 4002, end: 6144 vif vif-3-0 vif3.0: fatal error; disabling device Reported-by:
Anthony Wright <anthony@overnetdata.com> Tested-by:
David Vrabel <david.vrabel@citrix.com> Signed-off-by:
David S. Miller <davem@davemloft.net> Cc: Stefan Bader <stefan.bader@canonical.com> Signed-off-by:
-
Zoltan Kiss authored
commit 97a6d1bb upstream. There is a long known problem with the netfront/netback interface: if the guest tries to send a packet which constitues more than MAX_SKB_FRAGS + 1 ring slots, it gets dropped. The reason is that netback maps these slots to a frag in the frags array, which is limited by size. Having so many slots can occur since compound pages were introduced, as the ring protocol slice them up into individual (non-compound) page aligned slots. The theoretical worst case scenario looks like this (note, skbs are limited to 64 Kb here): linear buffer: at most PAGE_SIZE - 17 * 2 bytes, overlapping page boundary, using 2 slots first 15 frags: 1 + PAGE_SIZE + 1 bytes long, first and last bytes are at the end and the beginning of a page, therefore they use 3 * 15 = 45 slots last 2 frags: 1 + 1 bytes, overlapping page boundary, 2 * 2 = 4 slots Although I don't think this 51 slots skb can really happen, we need a solution which can deal with every scenario. In real life there is only a few slots overdue, but usually it causes the TCP stream to be blocked, as the retry will most likely have the same buffer layout. This patch solves this problem by linearizing the packet. This is not the fastest way, and it can fail much easier as it tries to allocate a big linear area for the whole packet, but probably easier by an order of magnitude than anything else. Probably this code path is not touched very frequently anyway. Signed-off-by:
Zoltan Kiss <zoltan.kiss@citrix.com> Cc: Wei Liu <wei.liu2@citrix.com> Cc: Ian Campbell <Ian.Campbell@citrix.com> Cc: Paul Durrant <paul.durrant@citrix.com> Cc: netdev@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: xen-devel@lists.xenproject.org Signed-off-by:
-
- 20 Jan, 2015 1 commit
-
-
Florian Westphal authored
commit db29a950 upstream. Given following iptables ruleset: -P FORWARD DROP -A FORWARD -m sctp --dport 9 -j ACCEPT -A FORWARD -p tcp --dport 80 -j ACCEPT -A FORWARD -p tcp -m conntrack -m state ESTABLISHED,RELATED -j ACCEPT One would assume that this allows SCTP on port 9 and TCP on port 80. Unfortunately, if the SCTP conntrack module is not loaded, this allows *all* SCTP communication, to pass though, i.e. -p sctp -j ACCEPT, which we think is a security issue. This is because on the first SCTP packet on port 9, we create a dummy "generic l4" conntrack entry without any port information (since conntrack doesn't know how to extract this information). All subsequent packets that are unknown will then be in established state since they will fallback to proto_generic and will match the 'generic' entry. Our originally proposed version [1] completely disabled generic protocol tracking, but Jozsef suggests to not track protocols for which a more suitable helper is available, hence we now mitigate the issue for in tree known ct protocol helpers only, so that at least NAT and direction information will still be preserved for others. [1] http://www.spinics.net/lists/netfilter-devel/msg33430.html Joint work with Daniel Borkmann. Signed-off-by:
Florian Westphal <fw@strlen.de> Signed-off-by:
Daniel Borkmann <dborkman@redhat.com> Acked-by:
Jozsef Kadlecsik <kadlec@blackhole.kfki.hu> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by:
-
- 19 Jan, 2015 8 commits
-
-
Govindarajulu Varadarajan authored
commit 17e96834 upstream. Hardware always provides compliment of IP pseudo checksum. Stack expects whole packet checksum without pseudo checksum if CHECKSUM_COMPLETE is set. This causes checksum error in nf & ovs. kernel: qg-19546f09-f2: hw csum failure kernel: CPU: 9 PID: 0 Comm: swapper/9 Tainted: GF O-------------- 3.10.0-123.8.1.el7.x86_64 #1 kernel: Hardware name: Cisco Systems Inc UCSB-B200-M3/UCSB-B200-M3, BIOS B200M3.2.2.3.0.080820141339 08/08/2014 kernel: ffff881218f40000 df68243feb35e3a8 ffff881237a43ab8 ffffffff815e237b kernel: ffff881237a43ad0 ffffffff814cd4ca ffff8829ec71eb00 ffff881237a43af0 kernel: ffffffff814c6232 0000000000000286 ffff8829ec71eb00 ffff881237a43b00 kernel: Call Trace: kernel: <IRQ> [<ffffffff815e237b>] dump_stack+0x19/0x1b kernel: [<ffffffff814cd4ca>] netdev_rx_csum_fault+0x3a/0x40 kernel: [<ffffffff814c6232>] __skb_checksum_complete_head+0x62/0x70 kernel: [<ffffffff814c6251>] __skb_checksum_complete+0x11/0x20 kernel: [<ffffffff8155a20c>] nf_ip_checksum+0xcc/0x100 kernel: [<ffffffffa049edc7>] icmp_error+0x1f7/0x35c [nf_conntrack_ipv4] kernel: [<ffffffff814cf419>] ? netif_rx+0xb9/0x1d0 kernel: [<ffffffffa040eb7b>] ? internal_dev_recv+0xdb/0x130 [openvswitch] kernel: [<ffffffffa04c8330>] nf_conntrack_in+0xf0/0xa80 [nf_conntrack] kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40 kernel: [<ffffffffa049e302>] ipv4_conntrack_in+0x22/0x30 [nf_conntrack_ipv4] kernel: [<ffffffff815005ca>] nf_iterate+0xaa/0xc0 kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40 kernel: [<ffffffff81500664>] nf_hook_slow+0x84/0x140 kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40 kernel: [<ffffffff81509dd4>] ip_rcv+0x344/0x380 Hardware verifies IP & tcp/udp header checksum but does not provide payload checksum, use CHECKSUM_UNNECESSARY. Set it only if its valid IP tcp/udp packet. Cc: Jiri Benc <jbenc@redhat.com> Cc: Stefan Assmann <sassmann@redhat.com> Reported-by:
Sunil Choudhary <schoudha@redhat.com> Signed-off-by:
Govindarajulu Varadarajan <_govind@gmx.com> Reviewed-by:
Jiri Benc <jbenc@redhat.com> Signed-off-by:
-
Jiri Pirko authored
commit b0d11b42 upstream. This patch is fixing a race condition that may cause setting count_pending to -1, which results in unwanted big bulk of arp messages (in case of "notify peers"). Consider following scenario: count_pending == 2 CPU0 CPU1 team_notify_peers_work atomic_dec_and_test (dec count_pending to 1) schedule_delayed_work team_notify_peers atomic_add (adding 1 to count_pending) team_notify_peers_work atomic_dec_and_test (dec count_pending to 1) schedule_delayed_work team_notify_peers_work atomic_dec_and_test (dec count_pending to 0) schedule_delayed_work team_notify_peers_work atomic_dec_and_test (dec count_pending to -1) Fix this race by using atomic_dec_if_positive - that will prevent count_pending running under 0. Fixes: fc423ff0 ("team: add peer notification") Fixes: 492b200e ("team: add support for sending multicast rejoins") Signed-off-by:
Jiri Pirko <jiri@resnulli.us> Signed-off-by:
-
Eric Dumazet authored
commit 7a05dc64 upstream. Commit d75b1ade ("net: less interrupt masking in NAPI") uncovered wrong alx_poll() behavior. A NAPI poll() handler is supposed to return exactly the budget when/if napi_complete() has not been called. It is also supposed to return number of frames that were received, so that netdev_budget can have a meaning. Also, in case of TX pressure, we still have to dequeue received packets : alx_clean_rx_irq() has to be called even if alx_clean_tx_irq(alx) returns false, otherwise device is half duplex. Signed-off-by:
Eric Dumazet <edumazet@google.com> Fixes: d75b1ade ("net: less interrupt masking in NAPI") Reported-by:
Oded Gabbay <oded.gabbay@amd.com> Bisected-by:
-
Palik, Imre authored
commit 07ff890d upstream. Since e9ce7cb6 ("xen-netback: Factor queue-specific data into queue struct"), the transimt shaper timeout is always set to 0. The value the user sets via xenbus is never propagated to the transmit shaper. This patch fixes the issue. Cc: Anthony Liguori <aliguori@amazon.com> Signed-off-by:
Imre Palik <imrep@amazon.de> Acked-by:
Ian Campbell <ian.campbell@citrix.com> Signed-off-by:
-
Herbert Xu authored
commit 843925f3 upstream. Thomas Jarosch reported IPsec TCP stalls when a PMTU event occurs. In fact the problem was completely unrelated to IPsec. The bug is also reproducible if you just disable TSO/GSO. The problem is that when the MSS goes down, existing queued packet on the TX queue that have not been transmitted yet all look like TSO packets and get treated as such. This then triggers a bug where tcp_mss_split_point tells us to generate a zero-sized packet on the TX queue. Once that happens we're screwed because the zero-sized packet can never be removed by ACKs. Fixes: 1485348d ("tcp: Apply device TSO segment limit earlier") Reported-by:
Thomas Jarosch <thomas.jarosch@intra2net.com> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Cheers, Signed-off-by:
-
Jay Vosburgh authored
commit 2c26d34b upstream. When using VXLAN tunnels and a sky2 device, I have experienced checksum failures of the following type: [ 4297.761899] eth0: hw csum failure [...] [ 4297.765223] Call Trace: [ 4297.765224] <IRQ> [<ffffffff8172f026>] dump_stack+0x46/0x58 [ 4297.765235] [<ffffffff8162ba52>] netdev_rx_csum_fault+0x42/0x50 [ 4297.765238] [<ffffffff8161c1a0>] ? skb_push+0x40/0x40 [ 4297.765240] [<ffffffff8162325c>] __skb_checksum_complete+0xbc/0xd0 [ 4297.765243] [<ffffffff8168c602>] tcp_v4_rcv+0x2e2/0x950 [ 4297.765246] [<ffffffff81666ca0>] ? ip_rcv_finish+0x360/0x360 These are reliably reproduced in a network topology of: container:eth0 == host(OVS VXLAN on VLAN) == bond0 == eth0 (sky2) -> switch When VXLAN encapsulated traffic is received from a similarly configured peer, the above warning is generated in the receive processing of the encapsulated packet. Note that the warning is associated with the container eth0. The skbs from sky2 have ip_summed set to CHECKSUM_COMPLETE, and because the packet is an encapsulated Ethernet frame, the checksum generated by the hardware includes the inner protocol and Ethernet headers. The receive code is careful to update the skb->csum, except in __dev_forward_skb, as called by dev_forward_skb. __dev_forward_skb calls eth_type_trans, which in turn calls skb_pull_inline(skb, ETH_HLEN) to skip over the Ethernet header, but does not update skb->csum when doing so. This patch resolves the problem by adding a call to skb_postpull_rcsum to update the skb->csum after the call to eth_type_trans. Signed-off-by:
Jay Vosburgh <jay.vosburgh@canonical.com> Signed-off-by:
-
Thomas Graf authored
commit b8fb4e06 upstream. skb_scrub_packet() is called when a packet switches between a context such as between underlay and overlay, between namespaces, or between L3 subnets. While we already scrub the packet mark, connection tracking entry, and cached destination, the security mark/context is left intact. It seems wrong to inherit the security context of a packet when going from overlay to underlay or across forwarding paths. Signed-off-by:
Thomas Graf <tgraf@suug.ch> Acked-by:
Flavio Leitner <fbl@sysclose.org> Signed-off-by:
-
Toshiaki Makita authored
commit 796f2da8 upstream. When vlan tags are stacked, it is very likely that the outer tag is stored in skb->vlan_tci and skb->protocol shows the inner tag's vlan_proto. Currently netif_skb_features() first looks at skb->protocol even if there is the outer tag in vlan_tci, thus it incorrectly retrieves the protocol encapsulated by the inner vlan instead of the inner vlan protocol. This allows GSO packets to be passed to HW and they end up being corrupted. Fixes: 58e998c6 ("offloading: Force software GSO for multiple vlan tags.") Signed-off-by:
Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp> Signed-off-by:
-
