full branch 5.15 and run:
./scripts/feeds update -a && ./scripts/feeds install -a && cp diffconfig .config && make defconfig && ./scripts/getver.sh
make -j5
Added the missing patch on my repo.
Built an image for just the r7800 with the 5.15 kernel and qsdk11 if anyone wants to try it. I’m testing it and found my main router (wifi turned off) seems to work just fine. My two dumb AP r7800s did not like it and would reboot after just a couple minutes. Haven’t trouble shot it yet.
organized all the commits on top of master for simplicity (master from about 4 days ago):
R7800 test images here. Seem to still have a bug (see above) but they post and work well on one of my devices:
sorry, I mean the full 5.15 branch from AC repo. What diff is your repo? Thanks.
I am currently on the 5.15-qsdk11 branch of qosmio with no changes. I've no repo.
The account i made on github was to test things but it is useless at present. You should not take it into account if you are referring to that repo.
In this post
I was referring to the 5.15 branch of ACwifidude before he updated it yesterday.
Thanks. re-make with AC's new updates today, All ok.
Compiled from your 5.15 branch yesterday. I compiled with ath10k (non -ct), rebased from master and added my usual packages (mainly I'm using unbound with odhcpd instead of Dnsmasq) and so far I have 16 hours uptime.
Edit: oh and for what it's worth I have this in my /etc/rc.local:
# move nss cores to cpu1 and cpu2
i=1
awk '$7=="nss"{gsub(":","");print $1,$7}' /proc/interrupts| while read num irq; do
echo $i > /proc/irq/$num/smp_affinity
i=$((i+1))
done
# move rpm, and usb to cpu2
awk '$7~/qcom_rpm_ack|xhci-hcd/{gsub(":","");print $1,$7}' /proc/interrupts| while read num irq; do
echo 2 > /proc/irq/$num/smp_affinity
done
# Set minimum CPU Frequency to 800Mhz
echo 800000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
echo 800000 > /sys/devices/system/cpu/cpu1/cpufreq/scaling_min_freq
Started testing 5.15 on NetGear 7800 and Zyxel NBG. ath10k non-ct. It works (at least at the moment).
it seems that some entries in /sys/kernel/debug/qcom_socinfo are not supported ipq806x.
it causes kernel panic.
BusyBox v1.35.0 (2022-10-30 23:33:43 UTC) built-in shell (ash)
_______ ________ __
| |.-----.-----.-----.| | | |.----.| |_
| - || _ | -__| || | | || _|| _|
|_______|| __|_____|__|__||________||__| |____|
|__| W I R E L E S S F R E E D O M
-----------------------------------------------------
OpenWrt SNAPSHOT, r21321-23a9f0c5c5
-----------------------------------------------------
root@OpenWrt:/#
root@OpenWrt:/#
root@OpenWrt:/#
root@OpenWrt:/# cat /sys/kernel/debug/qcom_socinfo/rpm/name
[ 83.347488] 8<--- cut here ---
[ 83.347526] Unable to handle kernel NULL pointer dereference at virtual address 0000017a
[ 83.349473] pgd = 0bfff502
[ 83.357835] [0000017a] *pgd=00000000
[ 83.360209] Internal error: Oops: 5 [#1] SMP ARM
[ 83.363942] Modules linked in: ecm xt_connlimit nf_conncount ath10k_pci ath10k_core ath xt_state xt_helper xt_conntrack xt_connmark xt_connbytes xt_CT nft_redir nft_nat nft_masq nft_flow_offload nft_fib_inet nft_ct nft_chain_nat nf_nat nf_flow_table_ipv6 nf_flow_table_ipv4 nf_flow_table_inet nf_flow_table nf_conntrack_netlink nf_conntrack mac80211 ipt_REJECT cfg80211 xt_time xt_tcpudp xt_recent xt_policy xt_multiport xt_mark xt_mac xt_limit xt_esp xt_comment xt_TCPMSS xt_LOG xfrm_interface pptp ppp_async nft_reject_ipv6 nft_reject_ipv4 nft_reject_inet nft_reject nft_quota nft_objref nft_numgen nft_log nft_limit nft_hash nft_fib_ipv6 nft_fib_ipv4 nft_fib nft_counter nf_tables nf_reject_ipv6 nf_reject_ipv4 nf_log_syslog nf_defrag_ipv6 nf_defrag_ipv4 libcrc32c l2tp_ppp iptable_raw iptable_mangle iptable_filter ipt_ah ip_tables crc_ccitt compat sch_tbf sch_ingress sch_htb sch_hfsc em_u32 cls_u32 cls_tcindex cls_route cls_matchall cls_fw cls_flow cls_basic act_skbedit act_mirred act_gact
[ 83.364694] qca_nss_tunipip6 qca_nss_tun6rd qca_nss_ipsecmgr qca_nss_cfi_cryptoapi qca_nss_qdisc qca_nss_crypto qca_nss_pppoe pppoe pppox qca_nss_l2tpv2 ppp_generic slhc ledtrig_usbport xt_set x_tables ip_set_list_set ip_set_hash_netportnet ip_set_hash_netport ip_set_hash_netnet ip_set_hash_netiface ip_set_hash_net ip_set_hash_mac ip_set_hash_ipportnet ip_set_hash_ipportip ip_set_hash_ipport ip_set_hash_ipmark ip_set_hash_ipmac ip_set_hash_ip ip_set_bitmap_port ip_set_bitmap_ipmac ip_set_bitmap_ip ip_set nfnetlink ip_gre gre ifb nat46 ip6_vti ip_vti sit ipcomp6 xfrm6_tunnel esp6 ah6 xfrm4_tunnel ipcomp esp4 ah4 ip6_tunnel tunnel6 tunnel4 xfrm_user xfrm_ipcomp af_key xfrm_algo sha1_generic seqiv md5 echainiv des_generic libdes cmac cbc authencesn authenc leds_gpio xhci_plat_hcd xhci_pci xhci_hcd dwc3 dwc3_qcom ohci_platform ohci_hcd phy_qcom_ipq806x_usb ahci fsl_mph_dr_of ehci_platform ehci_fsl sd_mod ahci_platform libahci_platform libahci libata scsi_mod scsi_common ehci_hcd
[ 83.433912] qca_nss_drv qca_nss_gmac pstore gpio_button_hotplug crc32c_generic
[ 83.541723] CPU: 0 PID: 6123 Comm: cat Not tainted 5.15.76 #0
[ 83.548924] Hardware name: Generic DT based system
[ 83.554825] PC is at show_image_name+0x4/0x3c
[ 83.559512] LR is at seq_read_iter+0x1e4/0x544
[ 83.563939] pc : [<c06b7160>] lr : [<c04c3498>] psr: 00000013
[ 83.568281] sp : c395be40 ip : 1ce49000 fp : 00400cc0
[ 83.574443] r10: 7ffff000 r9 : c20b60b8 r8 : c1811440
[ 83.579650] r7 : c395beb0 r6 : 00000001 r5 : 00000000 r4 : c20b60a0
[ 83.584861] r3 : c06b715c r2 : 0000017a r1 : 00000001 r0 : c20b60a0
[ 83.591t59] Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
[ 83.597972] Control: 10c5787d Table: 4941806a DAC: 00000051
[ 83.605173] Register r0 information: slab seq_file start c20b60a0 pointer offset 0
[ 83.610905] Register r1 information: non-paged memory
[ 83.618366] Register r2 information: non-paged memory
[ 83.623488] Register r3 information: non-slab/vmalloc memory
[ 83.628523] Register r4 information: slab seq_file start c20b60a0 pointer offset 0
[ 83.634256] Register r5 information: NULL pointer
[ 83.641631] Register r6 information: non-paged memory
[ 83.646404] Register r7 information: non-slab/vmalloc memory
[ 83.651440] Register r8 information: slab dentry start c1811440 pointer offset 0 size 36
[ 83.657174] Register r9 information: slab seq_file start c20b60a0 pointer offset 24
[ 83.665246] Register r10 information: non-paged memory
[ 83.672620] Register r11 information: non-paged memory
[ 83.677828] Register r12 information: non-paged memory
[ 83.682950] Process cat (pid: 6123, stack limit = 0x58e197e8)
[ 83.688074] Stack: (0xc395be40 to 0xc395c000)
[ 83.693895] be40: c577a060 00000000 c395be98 c20b60c8 c577af10 01603000 c5134400 c184a4c8
[ 83.698239] be60: c3969d90 00000000 00000000 c395bf88 00000000 c1811440 c395bf88 00000003
[ 83.706399] be80: 00000003 c04c38e0 00001000 c049fcf8 bed5fde0 00001000 00000000 00000000
[ 83.714559] bea0: 00001000 c395be90 00000001 00000000 c576be40 00000000 00000000 00000000
[ 83.722718] bec0: 00000000 00000000 00000000 00000000 00000000 00000000 c576be40 c04c37f8
[ 83.730878] bee0: c576be40 bed5fde0 00001000 c058b6c0 c576be40 c395bf88 00000001 c058b65c
[ 83.739037] bf00: c03002a4 c395a000 00000003 c0494b70 00000000 00000000 00000000 c576be40
[ 83.747197] bf20: 01000000 c0493ee4 01000000 00000000 c513443c 00000000 c395bf7c c141ca80
[ 83.755356] bf40: 00000000 00000000 00000000 00000000 c5134440 00000000 00000000 00000000
[ 83.763517] bf60: 00000000 00000000 bed5fde0 c576be40 c576be40 00000000 00000000 c03002a4
[ 83.771675] bf80: c395a000 c0495224 00000000 00000000 00000003 00000000 00000000 00000000
[ 83.779836] bfa0: 00000003 c0300060 00000000 00000000 00000003 bed5fde0 00001000 00000000
[ 83.787996] bfc0: 00000000 00000000 00000000 00000003 004ee5c0 00000000 01000000 00000003
[ 83.796155] bfe0: bed5fd78 bed5fd68 b6f2839c b6f2787c 60000010 00000003 00000000 00000000
[ 83.804313] [<c06b7160>] (show_image_name) from [<00000000>] (0x0)
[ 83.812475] Code: e28dd004 e3a00000 e49df004 e5902048 (e5d23000)
[ 83.818682] ---[ end trace 8e7aff5a1751e641 ]---
[ 83.824795] Kernel panic - not syncing: Fatal exception
[ 83.829397] CPU1: stopping
[ 83.834338] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G D 5.15.76 #0
[ 83.837120] Hardware name: Generic DT based system
[ 83.844672] [<c030e5b4>] (unwind_backtrace) from [<c030a130>] (show_stack+0x14/0x20)
[ 83.849277] [<c030a130>] (show_stack) from [<c0641f9c>] (dump_stack_lvl+0x40/0x4c)
[ 83.857174] [<c0641f9c>] (dump_stack_lvl) from [<c030cfdc>] (do_handle_IPI+0x144/0x184)
[ 83.864553] [<c030cfdc>] (do_handle_IPI) from [<c030d038>] (ipi_handler+0x1c/0x2c)
[ 83.872451] [<c030d038>] (ipi_handler) from [<c0379e9c>] (handle_percpu_devid_irq+0x84/0x16c)
[ 83.880093] [<c0379e9c>] (handle_percpu_devid_irq) from [<c0373760>] (handle_domain_irq+0x68/0x94)
[ 83.888686] [<c0373760>] (handle_domain_irq) from [<c065c08c>] (gic_handle_irq+0x8c/0xb4)
[ 83.897540] [<c065c08c>] (gic_handle_irq) from [<c0300b7c>] (__irq_svc+0x5c/0x78)
[ 83.905783] Exception stack(0xc146df58 to 0xc146dfa0)
[ 83.913249] df40: 0003695a 00000000
[ 83.918291] df60: 00000001 c0314800 c146c000 00000001 c0e04f10 c0e04f54 00000000 00000000
[ 83.926451] df80: 00000000 c0d466e8 c0ec0e98 c146dfa8 c03075ec c03075f0 60000013 ffffffff
[ 83.934606] [<c0300b7c>] (__irq_svc) from [<c03075f0>] (arch_cpu_idle+0x44/0x50)
[ 83.942764] [<c03075f0>] (arch_cpu_idle) from [<c035139c>] (do_idle+0x240/0x2a4)
[ 83.950230] [<c035139c>] (do_idle) from [<c0351724>] (cpu_startup_entry+0x1c/0x20)
[ 83.957607] [<c0351724>] (cpu_startup_entry) from [<42301530>] (0x42301530)
[ 84.965119] Rebooting in 3 seconds..
ath10k-ct still not good as non-ct version.
5G wifi connected to my ISP router as wwan, my pc wired to LAN port, ath10k-ct got ~100Mb internet speed.
But non-ct version could be >500Mb speed in the same conditions.
@ACwifidude Hi,
The kernel 5.15 branch code, bricked my Linksys-EA8500, wan port and lan ports not work.
Netgear-R7800 is ok on the same configs.
Seems like the ea8500 has a bug with OpenWrt master there is a commit that actually disables the ea8500 while a fix is being found. I plan on building for all the other ipq806x models I typically build for minus the ea8500 (will add it back in once the main devs find a fix).
Thanks, but your code branch 22.03 (knernel 5.10) is all ok for EA8500.
Yes 22.03 is unaffected. All 5.15 master builds are currently affected and the nss enabled builds here.
Crashed after 3 days just now.
Updated the build and added all the usual ipq806x devices. All three of my r7800s are running it. The performance governor seems more stable on 5.15.
Master is 22.03, right?
Any chance of updating 21.02 as well? I've been trying to build it myself, but so far no luck. And as I rely on some packages that have not been updated to work well on 22.03, I can't switch yet.
would be great to have some help about ea8500 would be curious to check if things works with a nss build.... currently very little time to produce more ideas about the problem so i would love some devs trying to figure what is wrong with ea8500
Master is not 22.03.
22.03 is the current stable release. Master is always "the bleeding edge" were continuous development occurs on features. The current stable release will get maintenance/security updates only, no new features basically.
On building 21.02 yourself, do you rebase to the latest 21.02 service release? I haven't built it myself but I can imagine that the package feeds for 21.02 need to have some alignment with the latest service release?