OpenWrt Forum Archive

Topic: Build for WNDR3700/WNDR3800

The content of this topic has been archived between 9 Jul 2013 and 6 May 2018. Unfortunately there are posts – most likely complete pages – missing.

hnyman wrote:
mmhorda wrote:

Could you please compile a new build with kmod-usb-serial-option  I will try if that works.
Thanks in advance.

That is only for "Option" modems, so most likely you do not need it.
https://dev.openwrt.org/browser/trunk/package/kernel/linux/modules/usb.mk#L583

(I think that wiki has wrong info, and option is not needed, but it is used for selecting wwan (on which it depends))

Hi.
I had to make some other tricks to make it work without this package.

I tried to force install since the kernel version is the same. and it worked.
modswitch is working as it suppose to work now.


i have huawei k3765

did you force install this "option" package? or something else?

was going to make an overclocked version but i can't seem to compile

make[3]: Entering directory `/Openwrt/trunk/toolchain/uClibc/utils'
make[2]: Entering directory `/Openwrt/trunk'
make[3]: Entering directory `/Openwrt/trunk/target/linux'
make[4]: Entering directory `/Openwrt/trunk/target/linux/ar71xx'
Applying patch generic/003-11-004-hso-Earlier-catch-of-error-condition.patch
make[4]: *** [/Openwrt/trunk/build_dir/target-mips_34kc_uClibc-0.9.33.2/linux-ar71xx_generic/linux-3.10.18/.quilt_checked] Error 1
make[3]: *** [compile] Error 2
make[2]: *** [target/linux/compile] Error 2
make[1]: *** [/Openwrt/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/stamp/.target_compile] Error 2
make: *** [world] Error 2
u64g@u64g:/Openwrt/trunk$ 

First guess is that your new patch is malformed. Does it apply cleanly?
And did you "make clean" first?
Use the "singlecompile.sh" and then check the end of the build.log to see what is the reason for your error.

(Last edited by hnyman on 16 Nov 2013, 08:56)

hnyman wrote:

did you force install this "option" package? or something else?

Yes. I force install "kmod-usb-serial-option_3.3.8-1_ar71xx.ipk  " taken from here _http://downloads.openwrt.org/attitude_adjustment/12.09/ar71xx/generic/packages/

here is the dmesg:

[   34.600000] option 1-1:1.0: GSM modem (1-port) converter detected
[   34.600000] usb 1-1: GSM modem (1-port) converter now attached to ttyUSB0
[   34.610000] option 1-1:1.3: GSM modem (1-port) converter detected
[   34.620000] usb 1-1: GSM modem (1-port) converter now attached to ttyUSB1
[   34.620000] option 1-1:1.4: GSM modem (1-port) converter detected
[   34.630000] usb 1-1: GSM modem (1-port) converter now attached to ttyUSB2
[   34.640000] usbcore: registered new interface driver option
[   34.640000] option: v0.7.2:USB Driver for GSM modems


without this package i had to play with usbserial to make the modem work ) and even after some success I had NO CARRIER all the time.
Installing this package made the modem work just from the first plugin to the router.

(Last edited by mmhorda on 16 Nov 2013, 09:32)

does this build contain unaligned access patches for higher encryption/vpn performance?

burning daylight wrote:

does this build contain unaligned access patches for higher encryption/vpn performance?

My build has the default patches, so yes for the first patch that is incorporated in trunk, but no for the second "not yet implemented" patch.

(Last edited by hnyman on 16 Nov 2013, 21:11)

Hi Hnyman,
I can not see the graphics of the statistics page. This lasted for some time. When I do a clean installation of firmware, the graphics are just visible for a few seconds.

Erreur d'analyse XML : mal formé Emplacement : http://192.168.1.1/cgi-bin/luci/;stok=73a3c6d46cfb8b11482f2d5f67fc6f1b/admin/statistics/graph/uptime/? Numéro de ligne 173, Colonne 131 :    <img src="/cgi-bin/luci/;stok=73a3c6d46cfb8b11482f2d5f67fc6f1b/admin/statistics/graph/uptime?img=WNDR3700/uptime/1.3600.png&host=WNDR3700" /> ------------------------------------------------------------------------------------------------------------------------------------------------^
Manani wrote:

Hi Hnyman,
I can not see the graphics of the statistics page. This lasted for some time. When I do a clean installation of firmware, the graphics are just visible for a few seconds.

I see all included statistics graphs ok, both with Firefox and with IE.

You might try with another browser, and also try to purge your browser cache in case there is something malformed being stored.

hnyman wrote:

First guess is that your new patch is malformed. Does it apply cleanly?
And did you "make clean" first?
Use the "singlecompile.sh" and then check the end of the build.log to see what is the reason for your error.

I successfully compiled your build with the default settings. I then just deleted the directory and started all over.

Here's the end of the file, though

Applying patch platform/900-overclock.patch
patching file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Hunk #1 FAILED at 23.
1 out of 1 hunk FAILED -- rejects in file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Patch platform/900-overclock.patch does not apply (enforce with -f)
make[4]: *** [/Openwrt/trunk/build_dir/target-mips_34kc_uClibc-0.9.33.2/linux-ar71xx_generic/linux-3.10.18/.quilt_checked] Error 1
make[4]: Leaving directory `/Openwrt/trunk/target/linux/ar71xx'
make[3]: *** [compile] Error 2
make[3]: Leaving directory `/Openwrt/trunk/target/linux'
make[2]: *** [target/linux/compile] Error 2
make[2]: Leaving directory `/Openwrt/trunk'
make[1]: *** [/Openwrt/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/stamp/.target_compile] Error 2
make[1]: Leaving directory `/Openwrt/trunk'
make: *** [world] Error 2

inside is just

--- a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h.orig
+++ b/arch/mips/include/asm/mach-ath79/kernel-entry-init.h
@@ -23,6 +23,8 @@
     and    t0, t1
     ori    t0, CONF_CM_CACHABLE_NONCOHERENT
     mtc0    t0, CP0_CONFIG
+    li t2, 0xc0140198
+    sw t2, 0x18050000
     nop
     .endm


The file you linked,

/*
 *  Atheros AR71XX/AR724X/AR913X specific kernel entry setup
 *
 *  Copyright (C) 2009 Gabor Juhos <juhosg@openwrt.org>
 *
 *  This program is free software; you can redistribute it and/or modify it
 *  under the terms of the GNU General Public License version 2 as published
 *  by the Free Software Foundation.
 *
 */
#ifndef __ASM_MACH_ATH79_KERNEL_ENTRY_H
#define __ASM_MACH_ATH79_KERNEL_ENTRY_H

    /*
     * Some bootloaders set the 'Kseg0 coherency algorithm' to
     * 'Cacheable, noncoherent, write-through, no write allocate'
     * and this cause performance issues. Let's go and change it to
     * 'Cacheable, noncoherent, write-back, write allocate'
     */
    .macro    kernel_entry_setup
    mfc0    t0, CP0_CONFIG
    li    t1, ~CONF_CM_CMASK
    and    t0, t1
    ori    t0, CONF_CM_CACHABLE_NONCOHERENT
    mtc0    t0, CP0_CONFIG
    nop
    .endm

    .macro    smp_slave_setup
    .endm

#endif /* __ASM_MACH_ATH79_KERNEL_ENTRY_H */

I realize I could just edit that to be

/*
 *  Atheros AR71XX/AR724X/AR913X specific kernel entry setup
 *
 *  Copyright (C) 2009 Gabor Juhos <juhosg@openwrt.org>
 *
 *  This program is free software; you can redistribute it and/or modify it
 *  under the terms of the GNU General Public License version 2 as published
 *  by the Free Software Foundation.
 *
 */
#ifndef __ASM_MACH_ATH79_KERNEL_ENTRY_H
#define __ASM_MACH_ATH79_KERNEL_ENTRY_H

    /*
     * Some bootloaders set the 'Kseg0 coherency algorithm' to
     * 'Cacheable, noncoherent, write-through, no write allocate'
     * and this cause performance issues. Let's go and change it to
     * 'Cacheable, noncoherent, write-back, write allocate'
     */
    .macro    kernel_entry_setup
    mfc0    t0, CP0_CONFIG
    li    t1, ~CONF_CM_CMASK
    and    t0, t1
    ori    t0, CONF_CM_CACHABLE_NONCOHERENT
    mtc0    t0, CP0_CONFIG
+    li t2, 0xc0140198
+    sw t2, 0x18050000
    nop
    .endm

    .macro    smp_slave_setup
    .endm

#endif /* __ASM_MACH_ATH79_KERNEL_ENTRY_H */

I added two lines, but where would I put these changes? I can't find any file called kernel-entry-init.h

or would I just call this file 900-overclock.patch and place it into /Openwrt/trunk/target/linux/ar71xx/patches-3.10

sorry for the noobness





Thanks for your help, really appreciated.

(Last edited by jigglywiggly on 17 Nov 2013, 10:42)

jigglywiggly wrote:
hnyman wrote:

First guess is that your new patch is malformed. Does it apply cleanly?

Applying patch platform/900-overclock.patch
patching file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Hunk #1 FAILED at 23.
1 out of 1 hunk FAILED -- rejects in file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Patch platform/900-overclock.patch does not apply (enforce with -f)

So, the patch is malformed.
Check the whitespace inside that patch. Most likely you have spaces instead of tabs on the lines. Or something like that.

EDIT: Or did you forget to correct the filename on the first a/ line?
not a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h.orig
but a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h

(Although your error message refers to the correct file?)

EDIT2:
Link to patch that applies cleanly for me: https://www.dropbox.com/sh/t52c02rm20y8x9p/lv-IYMq3ra/900-overclock.patch

(Last edited by hnyman on 17 Nov 2013, 10:56)

hnyman wrote:
Manani wrote:

Hi Hnyman,
I can not see the graphics of the statistics page. This lasted for some time. When I do a clean installation of firmware, the graphics are just visible for a few seconds.

I see all included statistics graphs ok, both with Firefox and with IE.

You might try with another browser, and also try to purge your browser cache in case there is something malformed being stored.

I tried unsuccessfully with Firefox, IE and Google Chrome.

Google Chrome:

This page contains the following errors:
error on line 173 at column 130: EntityRef: expecting ';'
Below is a rendering of the page up to the first error.

Maybe a problem with my router. I still have problems routing IPv6.

Manani wrote:

I still have problems routing IPv6.

There will be a 6rd patch in the next build. Hopefully it helps you.
https://dev.openwrt.org/changeset/38818

(Last edited by hnyman on 17 Nov 2013, 11:26)

hnyman wrote:
Manani wrote:

I still have problems routing IPv6.

There will be a 6rd patch in the next build. Hopefully it helps you.
https://dev.openwrt.org/changeset/38818

I hope so too.
Thank you!

hnyman wrote:
jigglywiggly wrote:
hnyman wrote:

First guess is that your new patch is malformed. Does it apply cleanly?

Applying patch platform/900-overclock.patch
patching file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Hunk #1 FAILED at 23.
1 out of 1 hunk FAILED -- rejects in file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Patch platform/900-overclock.patch does not apply (enforce with -f)

So, the patch is malformed.
Check the whitespace inside that patch. Most likely you have spaces instead of tabs on the lines. Or something like that.

EDIT: Or did you forget to correct the filename on the first a/ line?
not a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h.orig
but a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h

(Although your error message refers to the correct file?)

EDIT2:
Link to patch that applies cleanly for me: https://www.dropbox.com/sh/t52c02rm20y8x9p/lv-IYMq3ra/900-overclock.patch

Weird, same issue

Applying patch platform/900-overclock.patch
patching file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Hunk #1 FAILED at 23.
1 out of 1 hunk FAILED -- rejects in file arch/mips/include/asm/mach-ath79/kernel-entry-init.h
Patch platform/900-overclock.patch does not apply (enforce with -f)
make[4]: *** [/Openwrt/trunk/build_dir/target-mips_34kc_uClibc-0.9.33.2/linux-ar71xx_generic/linux-3.10.18/.quilt_checked] Error 1
make[4]: Leaving directory `/Openwrt/trunk/target/linux/ar71xx'
make[3]: *** [compile] Error 2
make[3]: Leaving directory `/Openwrt/trunk/target/linux'
make[2]: *** [target/linux/compile] Error 2
make[2]: Leaving directory `/Openwrt/trunk'
make[1]: *** [/Openwrt/trunk/staging_dir/target-mips_34kc_uClibc-0.9.33.2/stamp/.target_compile] Error 2
make[1]: Leaving directory `/Openwrt/trunk'
make: *** [world] Error 2

When you run

./updateNmake.sh

I am immediately hit exit, and choose to keep the default configuration, is that how you did it?

--- a/arch/mips/include/asm/mach-ath79/kernel-entry-init.h
+++ b/arch/mips/include/asm/mach-ath79/kernel-entry-init.h
@@ -23,6 +23,8 @@
    and t0, t1
    ori t0, CONF_CM_CACHABLE_NONCOHERENT
    mtc0    t0, CP0_CONFIG
+   li t2, 0xc0140198
+   sw t2, 0x18050000
    nop
    .endm

is what my 900-overclock.patch looks like

hnyman wrote:
Manani wrote:

Hi Hnyman,
I can not see the graphics of the statistics page. This lasted for some time. When I do a clean installation of firmware, the graphics are just visible for a few seconds.

I see all included statistics graphs ok, both with Firefox and with IE.

You might try with another browser, and also try to purge your browser cache in case there is something malformed being stored.

I understand the problem. This is the theme that I change after each update:

luci-theme-openwrt
Manani wrote:

I can not see the graphics of the statistics page. This lasted for some time.
...
I understand the problem. This is the theme that I change after each update:

luci-theme-openwrt

When has this problem started for you? A few days ago? A few months ago? (The default theme was changed to "bootstrap" in April.)

I added one line in early November to Luci config making the default theme selection more robust, but I am wondering if that has caused this problem for you (as you are changing the theme). Have you had the problem already in October or earlier?

I am not quite sure if that old "openwrt" theme is still compatible. It has not been maintained since Nov 2012...

(Last edited by hnyman on 18 Nov 2013, 13:47)

Oh wait I got it. I started over again, and this time after I ran

./updateNmake.sh

I hit, exit and hit yes to save. It successfully compiled, will test now.

EDIT: it works!



if anyone wants to try it https://www.dropbox.com/sh/u0bspioeoq4hyig/Zdh67nRIQA

root@OpenWrt:~# cat proc/cpuinfo
cat: can't open 'proc/cpuinfo': No such file or directory
root@OpenWrt:~# cat /proc/cpuinfo
system type             : Atheros AR7161 rev 2
machine                 : NETGEAR WNDR3700/WNDR3800/WNDRMAC
processor               : 0
cpu model               : MIPS 24Kc V7.4
BogoMIPS                : 531.66
wait instruction        : yes
microsecond timers      : yes
tlb_entries             : 16
extra interrupt vector  : yes
hardware watchpoint     : yes, count: 4, address/irw mask: [0x0ffc, 0x0ffc, 0x0ffb, 0x0ffb]
isa                     : mips1 mips2 mips32r1 mips32r2
ASEs implemented        : mips16
shadow register sets    : 1
kscratch registers      : 0
core                    : 0
VCED exceptions         : not available
VCEI exceptions         : not available

Thanks a lot man for your help.

(Last edited by jigglywiggly on 18 Nov 2013, 14:57)

jigglywiggly wrote:

EDIT: it works!

if anyone wants to try it https://www.dropbox.com/sh/u0bspioeoq4hyig/Zdh67nRIQA

Thanks a lot man for your help.

You are welcome.
One hint:
Right now the patch is shown as an unknown file in the status report and is not shown in the generated patch files.
  ?       trunk/target/linux/ar71xx/patches-3.10/900-overclock.patch
You should "svn add" needed patch for it to be properly included in the generated patch files.
  svn add target/linux/ar71xx/patches-3.10/900-overclock.patch

(Last edited by hnyman on 18 Nov 2013, 15:18)

Got it, it was changed to an 'A' now, instead of a '?'

https://www.dropbox.com/sh/nkh6sqvgh0ku4rr/_uAOR20NYn

I haven't actually tested this, but nothing else was changed other than the svn addition, so it should definitely work for anyone who wants to try it.

(Last edited by jigglywiggly on 18 Nov 2013, 16:05)

jigglywiggly wrote:

Got it, it was changed to an 'A' now, instead of a '?'

Yep, and the patch itself is now at the end of ...openwrt.patch

(Last edited by hnyman on 18 Nov 2013, 19:04)

hnyman wrote:
Manani wrote:

I can not see the graphics of the statistics page. This lasted for some time.
...
I understand the problem. This is the theme that I change after each update:

luci-theme-openwrt

When has this problem started for you? A few days ago? A few months ago? (The default theme was changed to "bootstrap" in April.)

I am not quite sure if that old "openwrt" theme is still compatible. It has not been maintained since Nov 2012...

If I change theme to the old Openwrt, I get the same error for the statistics graphs.
  "XML Parsing Error: not well-formed",
And the error pointer points to the only = char there.
  1.3600.png&host=WNDR3700"
Possibly that = char should be escaped in proper XML, or actually the & starts an escapte sequence (of style &jotakin; ) and then = suddenly stops it.

When I look at the generated page source code, the file from Openwrt theme claims to be XML, while the bootstrap page does not. And public XML validators choke on the page...

(Last edited by hnyman on 18 Nov 2013, 19:06)

hnyman wrote:
hnyman wrote:

When has this problem started for you? A few days ago? A few months ago? (The default theme was changed to "bootstrap" in April.)

I am not quite sure if that old "openwrt" theme is still compatible. It has not been maintained since Nov 2012...

If I change theme to the old Openwrt, I get the same error for the statistics graphs.
  "XML Parsing Error: not well-formed",
And the error pointer points to the only = char there.
  1.3600.png&host=WNDR3700"
Possibly that = char should be escaped in proper XML, or actually the & starts an escapte sequence (of style &jotakin; ) and then = suddenly stops it.

When I look at the generated page source code, the file from Openwrt theme claims to be XML, while the bootstrap page does not. And public XML validators choke on the page...

This problem started there about a month. I'll have to abandon the old theme.

I found another problem, customizing the data directory (/mnt/sda1/storage/rrd) results in a display bug even with the new theme.

/usr/lib/lua/luci/dispatcher.lua:448: Failed to execute call dispatcher target for entry '/admin/statistics/graph/memory'.
The called action terminated with an exception:
/usr/lib/lua/luci/util.lua:209: attempt to get length of local 'str' (a nil value)
stack traceback:
    [C]: in function 'assert'
    /usr/lib/lua/luci/dispatcher.lua:448: in function 'dispatch'
    /usr/lib/lua/luci/dispatcher.lua:195: in function </usr/lib/lua/luci/dispatcher.lua:194>

Edit: Reverting to the default directory (/tmp/rrd) does not correct the display abnormality.

(Last edited by Manani on 18 Nov 2013, 22:23)

Hello hnyman,

I just tried your very new build r38896 but if fails:Installing package 'batman-adv'
Installing package 'luci-app-bmx6'
svn: warning: W155010: »/Openwrt/attitude/files« nicht gefunden
svn: E200009: Could not add all targets because some targets don't exist
svn: E200009: Ungültiges Ziel für die angeforderte Operation
svn: warning: W155010: »/Openwrt/attitude/.config.init« nicht gefunden
svn: E200009: Could not add all targets because some targets don't exist
svn: E200009: Ungültiges Ziel für die angeforderte Operation
chmod: Zugriff auf „files/etc/*.sh“ nicht möglich: Datei oder Verzeichnis nicht gefunden
chmod: Zugriff auf „files/etc/rc.button/*“ nicht möglich: Datei oder Verzeichnis nicht gefunden
cp: der Aufruf von stat für „.config.init“ ist nicht möglich: Datei oder Verzeichnis nicht gefunden

Have I missed something? Well, I am missing <target>/files. Where can I find it?


Thanx,
Michael

I am not quite sure what you are trying to do.
You talk about 38896 build, which is trunk, but the texts do include Openwrt/attitude/

Are you trying to create a build environment?
If yes, I guess that you are trying to create an Attitude 12.09 build environment from the trunk build without modifying the 3 patch names. (or you have not downloaded the patch contents)

Have you read the message #2 in this thread?

If you want an attitude build environment, easiest would be to use the latest attitude build for that. (there will probably be /etc/rc.button error for attitude, but I guess no other errors should be visible)

(Last edited by hnyman on 23 Nov 2013, 16:47)