On C2600 and EA8500 WiFi cal data have wrong addresses. This routers using hotplug script to change MAC address:
https://git.lede-project.org/?p=source. … ac;hb=HEAD
Topic: Support for TP-Link Archer C2600
The content of this topic has been archived between 29 Mar 2018 and 6 May 2018. Unfortunately there are posts – most likely complete pages – missing.
Page 32 of 49
Before I report a bug, can some intrepid soul try a fresh sysupgrade install and see if everything goes smoothly? I keep getting corrupt installs and having to force reinstall packages to work correctly. Often LUCI will either not work correctly (Bad Gateway), or the themes will be broken (just look like unstyled HTML), or libiwinfo-lua won't be working correctly and you can't see anything about the wireless. Some other times it brings up errors with opkg and the .list files for the package. Sometimes directories for the package config files aren't created.
Here are the packages I install:
opkg install luci luci-theme-material
opkg install luci-app-uhttpd luci-app-upnp
opkg install ddns-scripts ddns-scripts_no-ip_com luci-app-ddns
opkg install sqm-scripts luci-app-sqm
opkg install openssh-sftp-server nanoI've tried installing one by one, and all as a big list together, and all of them are giving similar random failure results. Always seems to end up in one of the failure above -- not sure what else isn't working right then. I haven't had this kind of install problem before.
I tried going back to either of the two working builds I had before, and they would work, but sqm would fail since it requires a kmod that no longer matches the kernel version.
Between this and the wireless issue, it seems like stock is running better at the moment. 
Edit: I should also note that once something seems to have gone wrong, the SSH shell named turns from
root@lede:~#
to
root@lede:(unreachable)/root#
(Last edited by TeutonJon78 on 2 Sep 2016, 09:36)
Heinz wrote:
On C2600 and EA8500 WiFi cal data have wrong addresses. This routers using hotplug script to change MAC address:
https://git.lede-project.org/?p=source. … ac;hb=HEAD
Is that hotplug script actually working for you? Have you verified?
It's not working for me. Also ianchi's post says the same thing. So, I figured it was still a problem for the your device.
The sysfs device node gets set properly, and then the driver just ignores it.
(Last edited by jmomo on 2 Sep 2016, 09:30)
jmomo wrote:
Heinz wrote:On C2600 and EA8500 WiFi cal data have wrong addresses. This routers using hotplug script to change MAC address:
https://git.lede-project.org/?p=source. … ac;hb=HEADIs that hotplug script actually working for you? Have you verified?
It's not working for me. Also ianchi's post says the same thing. So, I figured it was still a problem for the your device.
The sysfs device node gets set properly, and then the driver just ignores it.
All of my MAC addresses match what stock would setup for each of the network interfaces, so it seems to be working for the C2600.
Interesting. Thanks Heinz. Thanks TeutonJon.
TeutonJon78 I confirm random luci package installation failure bug, the only sure way for me to have luci is to embed it into the firmware.
jmomo same problem with wifi mac on Netgear r7800, those scripts are not applicable either.
Heinz wrote:
New build: r1346
update 4.4.15
fix-30dBm
luci
wget-full
WiFi default ON
fix spi-qup unexpected irq
Added usb-dwc3-of-simple for supporting DWC3 USB
Hi!
I downloaded the factory image from this link and flashed it. I was on the stock firmware. Everything seems te be fine, it booted and I can reach the Luci login screen. However, it requires a login/pass and I did not see the firstboot screen. I've tried all passwords (on stock i had admin/admin, root, admin, <blank>, changeme) but none of them work. Is there a default password in this firmware? If there isn't, anyone knows how to reset/recover the current password? Obviously I can't reset at the moment because I'm locked out
Many thanks in advance!
Michel.
@MichelLoyen My firmware not need password. On Luci just click login, On SSH enter login root and password not needed.
Du You try hold reset button ~11sec?
Du You try failsafe mode?
@jmomo here is PR to fix MAC: https://github.com/lede-project/source/pull/142 Without this mac is random on wireless interfaces
(Last edited by Heinz on 2 Sep 2016, 10:47)
Thanks again Heinz.
It sounds like from dissent that the Netgear R7800 has the same problem I do. The hotplug script isn't working for us.
I get the impression that the EA8500 might have similar problems, but I'm not sure.
This might only be working for the TP-Link.
I will come back to the issue later and I'll post something if I figure out a fix.
(Last edited by jmomo on 2 Sep 2016, 11:17)
Btw any idea if http://patchwork.ozlabs.org/patch/664931/ will work on ipq806x?
Heinz wrote:
@MichelLoyen My firmware not need password. On Luci just click login, On SSH enter login root and password not needed.
Du You try hold reset button ~11sec?
Du You try failsafe mode?
Hi Heinz,
Thanks for the quick reply! I tried a reset and failsafe. Reset didn't change anything (it still cannot login: invalid username/password). In failsafe I'm able to change the password. However, when I reboot I still cannot login because of invalid username/password! I'm baffled!
Any ideas? :S
Many thanks.
dissent wrote:
Btw any idea if http://patchwork.ozlabs.org/patch/664931/ will work on ipq806x?
Now i compile firmware with this option and flash it to router.
How to test this crypto?
@MichelLoyen try reflash LEDE firmware using Recovery mode.
MichelLoyen wrote:
Heinz wrote:@MichelLoyen My firmware not need password. On Luci just click login, On SSH enter login root and password not needed.
Du You try hold reset button ~11sec?
Du You try failsafe mode?Hi Heinz,
Thanks for the quick reply! I tried a reset and failsafe. Reset didn't change anything (it still cannot login: invalid username/password). In failsafe I'm able to change the password. However, when I reboot I still cannot login because of invalid username/password! I'm baffled!
Any ideas? :S
Many thanks.
OMG! Nevermind! I'm such a n00b! There is another OpenWRT router on my network. Due to network config I did not realize I went to the webinterface of the other router! I just connected to the open Wifi LEDE network and it works just fine...
Very sorry to have spammed this thread with such a basic error on my part. Many many thanks for the help and quick response though 
Could it be your NOR flash is going kaput due to daily flashing?
Heinz wrote:
dissent wrote:Btw any idea if http://patchwork.ozlabs.org/patch/664931/ will work on ipq806x?
Now i compile firmware with this option and flash it to router.
How to test this crypto?
@MichelLoyen try reflash LEDE firmware using Recovery mode.
Might work out of the box... can't test it, not near the router.
The config description http://lxr.free-electrons.com/source/ar … to/Kconfig
Try that OpenSSL test again
(Last edited by dissent on 2 Sep 2016, 13:13)
https://chromium-review.googlesource.co … mon.config
Guess this can also be tried
Heinz wrote:
dissent wrote:Btw any idea if http://patchwork.ozlabs.org/patch/664931/ will work on ipq806x?
Now i compile firmware with this option and flash it to router.
How to test this crypto?
Pls run:
openssl speed md5 sha1 sha256 sha512 des des-ede3 aes-128-cbc aes-192-cbc aes-256-cbc rsa2048 dsa2048 | tee /tmp/sslspeed
echo "|" `awk 'match($0,/r[0-9]+/) {print substr($0,RSTART,RLENGTH)}' /etc/banner` `awk -v FS=": " -v ORS="" '/(Processor|BogoMIPS|Hardware|machine|cpu model|system type)/ { print "| " $2 " " } END { print "" }' /proc/cpuinfo` `awk -v ORS="" '$1 ~ /OpenSSL/ {print "| " $2 " |"} $1 ~ /(md5|sha)/ {print " " $5 " |"} $1 ~ /(des|aes)/ {b = b " " $6 " |"} $1 ~ /(rsa|dsa)/ {print b " " $6 " | " $7 " ";b=""} END { print "|" }' /tmp/sslspeed | sed 's/\.\(..\)k/\10/g'`
echoThen find it in tmp and paste it here pls
Result can be compared to other devices here: https://wiki.openwrt.org/doc/howto/benchmark.openssl
Another thing, when you guys post bench results or report issues pls state your HW rev, we dont know if the flash chip is the only thing they changed between v1.0 and 1.1..
Given 1.1 doesnt reboot that suggests there are other changes.
dissent wrote:
TeutonJon78 I confirm random luci package installation failure bug, the only sure way for me to have luci is to embed it into the firmware.
OK, thanks...glad I'm not going crazy.
james04 wrote:
Could it be your NOR flash is going kaput due to daily flashing?
Stock flashed fine with no problem. Plus, these NOR chips are rated for 100k+ writes. (And I only update when when there is something new in the build for us, so only like 4 times now. (But that's my fear as well.)
@james04
root@lede:~# cat /tmp/sslspeed
OpenSSL 1.0.2h 3 May 2016
built on: reproducible build, date unspecified
options:bn(64,32) rc4(ptr,char) des(idx,cisc,2,long) aes(partial) blowfish(ptr)
compiler: arm-openwrt-linux-muslgnueabi-gcc -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -I/home/hyniu/2600/source/staging_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/usr/include -I/home/hyniu/2600/source/staging_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/include -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/usr/include -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/include/fortify -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/include -znow -zrelro -DOPENSSL_SMALL_FOOTPRINT -DOPENSSL_NO_ERR -DTERMIOS -Os -pipe -mcpu=cortex-a15 -mfpu=neon-vfpv4 -fno-caller-saves -fno-plt -fhonour-copts -Wno-error=unused-but-set-variable -Wno-error=unused-result -mfloat-abi=hard -iremap /home/hyniu/2600/source/build_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/openssl-1.0.2h:openssl-1.0.2h -Wformat -Werror=format-security -fstack-protector -D_FORTIFY_SOURCE=1 -Wl,-z,now -Wl,-z,relro -fpic -I/home/hyniu/2600/source/package/libs/openssl/include -ffunction-sections -fdata-sections -fomit-frame-pointer -Wall -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
md5 7391.40k 25775.64k 69470.21k 119189.85k 151868.76k
sha1 8272.85k 30054.78k 85402.71k 156865.88k 209461.25k
des cbc 19155.01k 19920.68k 20166.74k 20293.63k 20368.04k
des ede3 7108.97k 7212.91k 7233.02k 7244.12k 7252.65k
aes-128 cbc 44727.57k 50262.29k 52431.36k 52861.61k 52655.45k
aes-192 cbc 37797.34k 40951.40k 42373.38k 42650.62k 42625.89k
aes-256 cbc 34918.38k 37656.06k 38808.49k 38888.45k 38939.31k
sha256 15973.09k 41645.67k 80421.29k 105402.71k 116116.14k
sha512 6386.43k 25624.36k 38937.17k 54500.69k 61745.83k
sign verify sign/s verify/s
rsa 2048 bits 0.009294s 0.000203s 107.6 4928.2
sign verify sign/s verify/s
dsa 2048 bits 0.001749s 0.002041s 571.9 489.9| r1480 | ARMv7 Processor rev 0 (v7l) | 21.87 | ARMv7 Processor rev 0 (v7l) | 45.57 | Qualcomm (Flattened Device Tree) | 1.0.2h | 119189850 | 156865880 | 105402710 | 54500690 | 20293630 | 7244120 | 52861610 | 42650620 | 38888450 | 107.6 | 4928.2 571.9 | 489.9 |Archer C2600 v1.1
(Last edited by Heinz on 2 Sep 2016, 19:41)
Test on both cores:
OpenSSL 1.0.2h 3 May 2016
built on: reproducible build, date unspecified
options:bn(64,32) rc4(ptr,char) des(idx,cisc,2,long) aes(partial) blowfish(ptr)
compiler: arm-openwrt-linux-muslgnueabi-gcc -I. -I.. -I../include -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -I/home/hyniu/2600/source/staging_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/usr/include -I/home/hyniu/2600/source/staging_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/include -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/usr/include -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/include/fortify -I/home/hyniu/2600/source/staging_dir/toolchain-arm_cortex-a15+neon-vfpv4_gcc-5.4.0_musl-1.1.15_eabi/include -znow -zrelro -DOPENSSL_SMALL_FOOTPRINT -DOPENSSL_NO_ERR -DTERMIOS -Os -pipe -mcpu=cortex-a15 -mfpu=neon-vfpv4 -fno-caller-saves -fno-plt -fhonour-copts -Wno-error=unused-but-set-variable -Wno-error=unused-result -mfloat-abi=hard -iremap /home/hyniu/2600/source/build_dir/target-arm_cortex-a15+neon-vfpv4_musl-1.1.15_eabi/openssl-1.0.2h:openssl-1.0.2h -Wformat -Werror=format-security -fstack-protector -D_FORTIFY_SOURCE=1 -Wl,-z,now -Wl,-z,relro -fpic -I/home/hyniu/2600/source/package/libs/openssl/include -ffunction-sections -fdata-sections -fomit-frame-pointer -Wall -DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_GF2m -DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DAES_ASM -DBSAES_ASM -DGHASH_ASM
md5 14484.25k 50922.58k 137284.95k 234617.86k 300894.89k
sha1 16189.46k 59861.46k 168300.71k 311517.87k 415307.09k
des cbc 37891.65k 39127.89k 39983.79k 40216.58k 39987.88k
des ede3 14101.49k 14191.30k 14369.62k 14256.81k 14404.27k
aes-128 cbc 84738.28k 97499.75k 103551.49k 103688.19k 104385.19k
aes-192 cbc 71118.35k 80779.99k 83815.94k 83672.75k 84656.13k
aes-256 cbc 66764.96k 73996.44k 75832.66k 77181.95k 77028.37k
sha256 30591.23k 82121.62k 157753.09k 209011.03k 227942.40k
sha512 12669.88k 50494.25k 76444.93k 108032.68k 121206.10k
sign verify sign/s verify/s
rsa 2048 bits 0.004710s 0.000103s 212.3 9663.9
sign verify sign/s verify/s
dsa 2048 bits 0.000889s 0.001031s 1125.1 970.3| r1480 | ARMv7 Processor rev 0 (v7l) | 21.87 | ARMv7 Processor rev 0 (v7l) | 45.57 | Qualcomm (Flattened Device Tree) | 1.0.2h | 234617860 | 311517870 | 209011030 | 108032680 | 40216580 | 14256810 | 103688190 | 83672750 | 77181950 | 212.3 | 9663.9 1125.1 | 970.3 |@Heinz
Not bad!
Those numbers are with the arm crypto flags enabled?
james04 wrote:
@Heinz
Not bad!
Those numbers are with the arm crypto flags enabled?
It's with this commit https://git.lede-project.org/?p=source. … 27ad5d5b8f
Heinz
So nothing changed?
How do you test 1 or 2 cores?
@james Yes, but it nothing change.
@dissent Use command: openssl speed sha256 -multi 2
Wonder could the assembly commit be the reason of no effect
Sorry, posts 801 to 800 are missing from our archive.
Page 32 of 49