00:03
Guest8515 has quit [Ping timeout: 480 seconds]
00:05
Guest8516 has quit [Ping timeout: 480 seconds]
00:17
tidalf has quit [Remote host closed the connection]
00:20
gladiac is now known as Guest8519
00:20
gladiac has joined #openwrt-devel
00:24
gladiac is now known as Guest8520
00:24
gladiac has joined #openwrt-devel
00:25
Guest8519 has quit [Ping timeout: 480 seconds]
00:30
Guest8520 has quit [Ping timeout: 480 seconds]
00:34
gladiac has quit [Ping timeout: 480 seconds]
00:42
gladiac has joined #openwrt-devel
01:01
thejoker8814 is now known as Guest8522
01:01
thejoker8814 has joined #openwrt-devel
01:06
cation has quit [Read error: Connection reset by peer]
01:08
Guest8522 has quit [Ping timeout: 480 seconds]
01:14
cation has joined #openwrt-devel
01:54
minimal has quit [Quit: Leaving]
02:00
tSYS has quit [Quit: *squeak*]
02:00
tSYS has joined #openwrt-devel
02:01
madwoota has quit [Ping timeout: 480 seconds]
02:01
madwoota has joined #openwrt-devel
02:05
xback has quit [Remote host closed the connection]
02:06
xback has joined #openwrt-devel
03:09
svanheule has quit [Ping timeout: 480 seconds]
03:13
svanheule has joined #openwrt-devel
03:55
cation has quit [Read error: Connection reset by peer]
03:55
cation has joined #openwrt-devel
04:28
rua has quit [Quit: Leaving.]
04:53
rua has joined #openwrt-devel
05:57
goliath has joined #openwrt-devel
07:11
noahm has quit [Ping timeout: 480 seconds]
07:49
bwitte has quit [Remote host closed the connection]
08:10
gch981213 has joined #openwrt-devel
08:25
robimarko has joined #openwrt-devel
09:26
xback has quit [Quit: Konversation terminated!]
09:39
rua has quit [Quit: Leaving.]
09:42
rua has joined #openwrt-devel
10:43
rua has quit [Quit: Leaving.]
10:52
tidalf has joined #openwrt-devel
11:31
goliath has quit [Quit: SIGSEGV]
12:03
rua has joined #openwrt-devel
12:15
goliath has joined #openwrt-devel
13:53
mrnuke has quit [Ping timeout: 480 seconds]
14:29
minimal has joined #openwrt-devel
14:30
cmonroe has quit [Ping timeout: 480 seconds]
14:35
cmonroe has joined #openwrt-devel
15:46
noahm has joined #openwrt-devel
15:59
goliath has quit [Quit: SIGSEGV]
17:27
schmars[m] has joined #openwrt-devel
18:20
robimarko has joined #openwrt-devel
18:28
<
Ansuel >
mangix my wdr3600 arrived and testing the qca8k... any hint to repro the problem ?
18:28
<
Ansuel >
currently running ping command on loop
18:31
<
robimarko >
Oh, the good old undocumented mixed endianess :)
18:42
<
Ansuel >
robimarko my theory is that these early switch are just broken and miss command maybe?
18:43
<
robimarko >
That wouldnt be surprising as well
18:43
<
robimarko >
Considering its really not used at all by QCA
19:48
<
Ansuel >
ok yes the thing die for real o.o
19:49
<
Ansuel >
doesn't handle any more packet for mdio wth???
19:51
<
stintel >
sounds like usual QCA business :P
20:43
<
Mangix >
Ansuel: glad you can repro
20:43
<
Mangix >
all it takes is just idling
20:44
<
Mangix >
passing massive traffic like flent benchmarks make it fail faster but w/e
20:46
<
Mangix >
kabel said he used an oscilloscope to debug some issue similar to this. Alas I do not have one.
20:47
<
robimarko >
Ansuel: I would just make this an KConfig option and disable it
20:49
<
Ansuel >
[ 65.024937] br-lan: port 3(lan3) entered blocking state
20:49
<
Ansuel >
[ 65.030291] br-lan: port 3(lan3) entered forwarding state
20:49
<
Ansuel >
[ 1682.207465] br-lan: port 3(lan3) entered disabled state
20:49
<
Ansuel >
[ 1682.215430] qca8k mdio.0:10 lan3: Link is Down
20:49
* Mangix
contemplates replacing his RT1800 with RT3200
20:49
<
Ansuel >
robimarko yes that is the easy way but i'm curious if I can get the problem
20:50
<
Mangix >
was the benefit lower CPU usage?
20:51
<
robimarko >
CPU and speed, since MDIO is slot and polling is frequent
20:52
<
Ansuel >
(still can't understand why oem loves to use polling instead of attaching a damn interrupt....)
20:53
<
Ansuel >
it would be so much better...
20:53
dermoth has quit [Remote host closed the connection]
20:54
dermoth has joined #openwrt-devel
21:01
<
Ansuel >
Mangix we probably forgot on that and Daniel never sent a v2 with the of ref fixed
21:04
<
Ansuel >
i'm trying to find a way to repro the thing quickly tho...
21:06
<
Mangix >
maybe ask him
21:08
<
Mangix >
interesting. he gets even lan1 to die...
21:08
<
Ansuel >
those are way too much generic report and I can't see any pattern there...
21:08
<
Ansuel >
also still not clear to me if the thing works correctly with the mgmt feature turned off
21:10
<
Mangix >
it at least stays alive. I've had it run multiple days maybe back in kernel 5.10 days
21:11
<
Ansuel >
imagine if it's something strange like the mdio bus going in some kind of sleep mode or autoturn off after a while due to us using ethernet packet and keeping everything internally...
21:13
tidalf has quit [Remote host closed the connection]
21:14
tidalf has joined #openwrt-devel
21:16
<
Ansuel >
[ 64.944942] br-lan: port 3(lan3) entered blocking state
21:16
<
Ansuel >
[ 64.950265] br-lan: port 3(lan3) entered forwarding state
21:16
<
Ansuel >
[ 910.454297] qca8k mdio.0:10 lan3: Link is Down
21:16
<
Ansuel >
[ 910.446972] br-lan: port 3(lan3) entered disabled state
21:16
<
Ansuel >
mh fastwer
21:17
<
Mangix >
welcome to QCA MIPS
21:21
<
Ansuel >
btw this makes me owner of 3 wdr3600....
21:28
<
Mangix >
hmm wasn't there a feature to ignore package failures in menuconfig?
21:28
<
Ansuel >
mhhh no but there was the thing IGNORE_ERRORS=1
21:52
<
Mangix >
ugh. tools/sparse and tools/mold fail to build
21:52
<
slh >
it's been a while, but I did run the old (probably around v5.10, not quite sure now which version exactly) qca8k on tl-wdr3600 and tl-wdr4300 for months at a time
21:54
<
slh >
Mangix: at this point, I'd skip the rt3200 for filogic 820/ 830, e.g. xiaomi ax3000t is quite tempting (regularly for under 30 EUR delivered, no I haven't bought one yet - xiaomi ax3600/ ipq8071a works well enough)
21:59
<
Mangix >
slh: but I have the rt3200 just sitting. unfortunately, it seems to be plagued by NAND/bootloader bugs
22:04
<
schmars[m] >
the router that got its own poem eh
22:04
<
schmars[m] >
definitely skip the rt3200
22:05
<
schmars[m] >
it bricks itself on powercycle / reboot
22:06
<
schmars[m] >
we've had three dead rt3200 in our community network the past few months, and now we're scrambling to replace them
22:06
<
schmars[m] >
s/them/the other dozen/
22:06
<
Mangix >
is this a new problem?
22:07
<
schmars[m] >
i don't know tbh, but it's related to something something ubifs
22:09
<
Mangix >
tools/mold error: /usr/bin/ld: cannot find -lstdc++: No such file or directory
22:09
<
Mangix >
is this a joke?
22:09
<
slh >
yeah, the OKD thing is another concern (obviously if you have a (working) rt3200, no reason to replace, just as with the ax3600 here)
22:10
<
schmars[m] >
we thought our rt3200 were working :-)
22:10
<
schmars[m] >
just dont reboot or powercycle, and cross fingers
22:41
tersono has joined #openwrt-devel