Hello, I applied these along with the reverting : xhci: Set scatter-gather limit to avoid failed block writes. usb: xhci: Link TRB must not occur within a USB payload burst to Linus' tree. The excellent news is that my system no longer hard freezes with an ax88179_178a connected via a VIA VL800 host when doing certain types of network traffic. The bad news is the following: [ 3337.877698] ------------[ cut here ]------------ [ 3337.877721] WARNING: CPU: 5 PID: 0 at net/sched/sch_generic.c:264 dev_watchdog+0x256/0x260() [ 3337.877727] NETDEV WATCHDOG: enp2s0u1 (ax88179_178a): transmit queue 0 timed out [ 3337.877731] Modules linked in: nls_iso8859_1 vfat fat ax88179_178a usbnet mii usb_storage xhci_hcd fuse ipt_MASQUERADE iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack ipt_REJECT xt_CHECKSUM iptable_mangle xt_tcpudp ip6table_filter ip6_tables iptable_filter ip_tables ebtable_nat ebtables x_tables af_packet bridge stp llc vhost_net vhost macvtap macvlan vfio_pci vfio_iommu_type1 vfio it87 hwmon_vid hid_generic usbhid hid btrfs raid6_pq libcrc32c xor mxm_wmi ehci_pci ohci_pci ohci_hcd kvm_amd ehci_hcd snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic kvm crct10dif_pclmul radeon fbcon ttm bitblit snd_hda_intel softcursor crc32_pclmul snd_hda_codec font tileblit ghash_clmulni_intel drm_kms_helper aesni_intel aes_x86_64 lrw gf128mul usbcore snd_hwdep [ 3337.877807] drm snd_pcm snd_timer snd agpgart cfbfillrect cfbimgblt cfbcopyarea i2c_algo_bit fb psmouse glue_helper fam15h_power k10temp usb_common serio_raw wmi ablk_helper pcspkr soundcore fbdev sp5100_tco i2c_piix4 cryptd edac_core tpm_infineon edac_mce_amd tpm_tis mac_hid microcode ipv6 autofs4 unix [ 3337.877843] CPU: 5 PID: 0 Comm: swapper/5 Not tainted 3.13.0+ #4 [ 3337.877849] Hardware name: To be filled by O.E.M. To be filled by O.E.M./M5A99FX PRO R2.0, BIOS 2201 11/22/2013 [ 3337.877853] 0000000000000009 ffff88043ed43de0 ffffffffb04f07d6 ffff88043ed43e28 [ 3337.877860] ffff88043ed43e18 ffffffffb0041883 0000000000000000 ffff8804234f0000 [ 3337.877866] 0000000000000001 0000000000000005 ffff880428560000 ffff88043ed43e78 [ 3337.877873] Call Trace: [ 3337.877877] <IRQ> [<ffffffffb04f07d6>] dump_stack+0x45/0x56 [ 3337.877893] [<ffffffffb0041883>] warn_slowpath_common+0x73/0x90 [ 3337.877901] [<ffffffffb00418e7>] warn_slowpath_fmt+0x47/0x50 [ 3337.877909] [<ffffffffb04677d6>] dev_watchdog+0x256/0x260 [ 3337.877916] [<ffffffffb0467580>] ? dev_graft_qdisc+0x80/0x80 [ 3337.877924] [<ffffffffb004b538>] call_timer_fn.isra.36+0x18/0x80 [ 3337.877931] [<ffffffffb004b70a>] run_timer_softirq+0x16a/0x200 [ 3337.877939] [<ffffffffb00457fc>] __do_softirq+0xdc/0x1f0 [ 3337.877947] [<ffffffffb0045b1d>] irq_exit+0x9d/0xb0 [ 3337.877956] [<ffffffffb002c21f>] smp_apic_timer_interrupt+0x3f/0x50 [ 3337.877963] [<ffffffffb04fb68a>] apic_timer_interrupt+0x6a/0x70 [ 3337.877968] <EOI> [<ffffffffb03ff217>] ? cpuidle_enter_state+0x47/0xc0 [ 3337.877979] [<ffffffffb03ff326>] cpuidle_idle_call+0x96/0x130 [ 3337.877987] [<ffffffffb000b989>] arch_cpu_idle+0x9/0x20 [ 3337.877995] [<ffffffffb007edfa>] cpu_startup_entry+0xda/0x1d0 [ 3337.878002] [<ffffffffb002a861>] start_secondary+0x1e1/0x240 [ 3337.878008] ---[ end trace 95a68de5267295c3 ]--- [ 3337.878915] xhci_hcd 0000:02:00.0: HC gave bad length of 124978 bytes left I honestly think it's worth pursuing this as my system is crashable even with sg disabled for the ax88179 without these patches. With this error i'm still able to just rmmod xhci_hcd and reload it and the whole system hasn't been brought down. Regards, Will Trives -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
