OpenWrt Forum Archive

Topic: Update on Linksys WRT1900AC support

The content of this topic has been archived between 16 Sep 2014 and 7 May 2018. Unfortunately there are posts – most likely complete pages – missing.

Post #3976
drewgarth2
30 Mar 2015, 20:09
LogicoZone wrote:
wrtpat wrote:

EDIT: @LogicoZone, I noticed you opened a ticket for the same (or similar) against Kaloz's Git repo, Here.  This may be helpful, or at least get us closer to a fix.

Marvell just checked in fix for the performance issue over wifi. I haven't had time to make a build. Maybe someone can try to change the mwlwifi Makefile to pull the revision and compile to test.

Great news, looking forward to hearing how this tests out.

Post #3977
gufus
30 Mar 2015, 20:32
drewgarth2 wrote:
LogicoZone wrote:
wrtpat wrote:

EDIT: @LogicoZone, I noticed you opened a ticket for the same (or similar) against Kaloz's Git repo, Here.  This may be helpful, or at least get us closer to a fix.

Marvell just checked in fix for the performance issue over wifi. I haven't had time to make a build. Maybe someone can try to change the mwlwifi Makefile to pull the revision and compile to test.

Great news, looking forward to hearing how this tests out.

FYI
https://github.com/kaloz/mwlwifi/commits/master

Post #3978
jklap
30 Mar 2015, 22:03
DavidMcWRT wrote:
jklap wrote:
DavidMcWRT wrote:

So I gave the latest trunk (r45092) a try.

Same problem as with the Kaloz Mar 23rd snapshot I came from: after sysupgrading (and keeping settings) the router was unresponsive.

Used the reset button and it eventually came up with default settings.

Imported my previous settings then proceeded to set up as normal.

Curious-- is this consistent? What happens if you sysupdate without keeping settings?

Haven't risked trying to reboot (which actually might be the root cause) after getting everything set up.

Haven't tried without keeping settings and likely won't have time to do so until next weekend.

However, my settings haven't changed for many, many builds, and re-importing the same settings after a factory reset results in stable operation (at least if I don't reboot).

Right now I need to use my connection rather than risk it smile, but I just wanted others to be aware that something wasn't right.

Anyone out there experiencing this that has a serial connection to their router such that they can tell us what is going on?

(Last edited by jklap on 30 Mar 2015, 22:13)

Post #3979
grimley
30 Mar 2015, 22:29

Debricking assistance required.

Console produces following:
NAND read: device 0 offset 0xa00000, size 0x1e00000
4194304 bytes read: OK
Wrong Image Format for bootm command
ERROR: Can't get kernel image!

Any idea where to go from here?
Thanks.

Post #3980
LogicoZone
30 Mar 2015, 22:31

Not sure if someone already asked. Is "WPA2-Enterprise" supported for this router? I don't see the UI option but wondering if I can just use UCI command to configure.

Has anyone tried this configuration?

Post #3981
gufus
30 Mar 2015, 22:46
grimley wrote:

Debricking assistance required.

Console produces following:
NAND read: device 0 offset 0xa00000, size 0x1e00000
4194304 bytes read: OK
Wrong Image Format for bootm command
ERROR: Can't get kernel image!

Any idea where to go from here?
Thanks.

run update_both_images

Post #3982
grimley
30 Mar 2015, 23:05
gufus wrote:
grimley wrote:

Debricking assistance required.

Console produces following:
NAND read: device 0 offset 0xa00000, size 0x1e00000
4194304 bytes read: OK
Wrong Image Format for bootm command
ERROR: Can't get kernel image!

Any idea where to go from here?
Thanks.

run update_both_images

Thanks.
TFTP was not communicating properly.
Found out that T T T T T T T means "T T T Time out...."

Post #3983
wrtpat
30 Mar 2015, 23:55
grimley wrote:

OpenWrt Chaos Calmer r45144 / LuCI (git-15.079.29361-3e37216)
Kernel Version    4.0.0-rc6
Local Time    Mon Mar 30 18:31:36 2015

Power cycled OK.
Fan has not stopped running (1 hour +)
Any additional command / install required?
Thank you.

@grimley
The fan isn't shutting off cuz path names have changed slightly in linux-4.0
As a consequence, fan_ctrl.sh (in /sbin) needs to be fixed-up

This is what worked for me....
Edit the fan_ctrl.sh script, and change all occurrences of;
    /sys/devices/pwm_fan/hwmon/hwmon0/pwm1
to;
    /sys/devices/platform/pwm_fan/hwmon/hwmon0/pwm1

Post #3984
grimley
31 Mar 2015, 00:30
wrtpat wrote:
grimley wrote:

OpenWrt Chaos Calmer r45144 / LuCI (git-15.079.29361-3e37216)
Kernel Version    4.0.0-rc6
Local Time    Mon Mar 30 18:31:36 2015

Power cycled OK.
Fan has not stopped running (1 hour +)
Any additional command / install required?
Thank you.

@grimley
The fan isn't shutting off cuz path names have changed slightly in linux-4.0
As a consequence, fan_ctrl.sh (in /sbin) needs to be fixed-up

This is what worked for me....
Edit the fan_ctrl.sh script, and change all occurrences of;
    /sys/devices/pwm_fan/hwmon/hwmon0/pwm1
to;
    /sys/devices/platform/pwm_fan/hwmon/hwmon0/pwm1

Thank you.
Added to my notes and lessons for the day.
Appreciated.

Post #3985
wrtpat
31 Mar 2015, 01:12
gufus wrote:
drewgarth2 wrote:
LogicoZone wrote:

Marvell just checked in fix for the performance issue over wifi. I haven't had time to make a build. Maybe someone can try to change the mwlwifi Makefile to pull the revision and compile to test.

Great news, looking forward to hearing how this tests out.

FYI
https://github.com/kaloz/mwlwifi/commits/master

I tried the changes today.
Unfortunately, the changes don't appear to fix the performance issue.  Performance is no better (perhaps worse) than it was prior to  today's commits.

Post #3986
LogicoZone
31 Mar 2015, 01:22
wrtpat wrote:
gufus wrote:
drewgarth2 wrote:

Great news, looking forward to hearing how this tests out.

FYI
https://github.com/kaloz/mwlwifi/commits/master

I tried the changes today.
Unfortunately, the changes don't appear to fix the performance issue.  Performance is no better (perhaps worse) than it was prior to  today's commits.

I'm trying to find a time to make a build. Anyway, maybe you can share your result to the ticket on github.

Post #3987
alirz
31 Mar 2015, 01:23
alirz wrote:

@nitroshift

I did actually reply to you yesterday but no worries at all. I'm thankful that you are helping me with this.

I would like to keep the same kernel that is in Kaloz's latest build.
i.e 3.18.9 . I want to make sure package compatibility is not broken with the newer kernel. Unless the newer kernel has better performance?
Will the image also have all the packages kalzo had in his snapshot?
In Any case heres a txt file with all the packages that came in Kalzo's snapshot and also including the few that installed myself. I would like to have all those in the image. I honestly dont know how much effort it is on your part to include packages in a custom build. Feel free to let me know if it takes too much time on your part and if you wouldn't want to do a custom build for me. Thanks

Full package list: https://dl.dropboxusercontent.com/u/488 … ckages.txt
Kaloz's build: https://downloads.openwrt.org/people/kaloz/

My custom installed packages:

luci-theme-openwrt
block-mount
python-setuptools
python-pip
ip  (ip_3.19.0-1_mvebu.ipk)
openssh-sftp-server
postfix
tcpdump


Any reason why so much space is reserved for /tmp and only 25MB for / ?


@nitroshift

Thank you for the build. It installed fine. However some of the important packages from my original request are missing in the build

python-pip
postfix
tcpdump

Post #3988
alirz
31 Mar 2015, 01:34

@nitroshift

i think perhaps its better to just stop and not worry about a custom build. Thats going to open a whole new can of worms most likely.

Why i say this?

After i installed your customer built image for me. I tried to install "python-pip" as i normally do, but this time i got this error message:

root@OpenWrt:~# opkg install python-pip
Installing python-pip (1.5.6-1) to root...
Downloading http://downloads.openwrt.org/snapshots/ … mvebu.ipk.
sh: /tmp/opkg-tugV6q/python-pip-iDgPko/preinst: Permission denied
Collected errors:
* pkg_run_script: package "python-pip" preinst script returned status 126.
* preinst_configure: Aborting installation of python-pip.
* opkg_install_cmd: Cannot install package python-pip.

Perhaps, you had tried to include it into the image, but possibly it error ed out and got excluded due to some errors behind the seen?
Anyways, i do appreciate your help but i wont waste your time anymore. ill stick to manually installing the packages i need.


EDIT: i reverted back to the original KALZOs image was running. im getting the python pip errors on that also. Looks like  something change in the repo?

(Last edited by alirz on 31 Mar 2015, 02:15)

Post #3989
raven-au
31 Mar 2015, 01:46
doITright wrote:
raven-au wrote:
doITright wrote:

Question for all...

Has anyone managed to get bridged mode working with the new driver?

My second unit is still running stock and I would love to switch it to trunk....

What are you asking?

I set mine up as a pseudo bridge, http://wiki.openwrt.org/doc/recipes/relayclient, which is probably about all that can be done.
And I posted saying that not long after the new driver was available.

If your asking if the router can do transparent bridging then, no it won't, since it isn't a broadcom based router using broadcom drivers.

I was under the impression that it was the driver that was preventing the use of client/station modes, and was hoping that the new driver would give us this functionality already....

At the moment I am running my second unit on stock firmware 1.1.8.164461 (2014-11-20 ) in wireless bridge mode, but I want to do this with openwrt....

Since you got the pseudo bridge working, I have high hopes that an AP-to-STA WDS or similar is finally possible smile

I will give it a try...

The new release of the driver did add wireless station support.

Part of the pseudobridge setup is to set the wireless interface to station mode and that worked fine for me.
I don't know anything about the requirements for mesh network setup.

We need to take a little care talking about this because the stock firmware has, what I thought was, unusual names for this functionality.
The stock firmware supports several types of wireless configuration including media bridge (seems to be what most implementations are calling it) and that doesn't behave like a transparent bridge either.

Post #3990
raven-au
31 Mar 2015, 02:00
wrtpat wrote:
gufus wrote:
drewgarth2 wrote:

Great news, looking forward to hearing how this tests out.

FYI
https://github.com/kaloz/mwlwifi/commits/master

I tried the changes today.
Unfortunately, the changes don't appear to fix the performance issue.  Performance is no better (perhaps worse) than it was prior to  today's commits.

If the failure to ack data frames really is the problem perhaps the:
"if (skb_queue_len(&priv->delay_freeq) > SYSADPT_DELAY_FREE_Q_LIMIT)"
is false which would still prevent the data frame from being acked.

I don't understand the wireless infrastructure much at all which is something that is essential to understanding and fixing this.

Post #3991
LogicoZone
31 Mar 2015, 02:04

I tried to make a build today with no additional packages installed other than add ext4 support. I got following compilation error. Anyone can share some leads?

/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/ustream-ssl-cyassl/ustream-ssl-2014-03-30/ustream-openssl.h: In function '__ustream_ssl_set_server_name':
/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/ustream-ssl-cyassl/ustream-ssl-2014-03-30/ustream-openssl.h:44:2: error: implicit declaration of function 'SSL_set_tlsext_host_name' [-Werror=implicit-function-declaration]
  SSL_set_tlsext_host_name(us->ssl, us->server_name);
  ^
cc1: all warnings being treated as errors
make[6]: *** [CMakeFiles/ustream-ssl.dir/ustream-ssl.c.o] Error 1

P.S. I'm compiling on Ubuntu 14.04.

(Last edited by LogicoZone on 31 Mar 2015, 02:05)

Post #3992
alirz
31 Mar 2015, 03:11

Could someone please try to intsll "python-pip" and let me know they get an error like:
Collected errors:
* pkg_run_script: package "python-pip" preinst script returned status 126.


I had installed it with no issues three days ago on Kalzos snapshot build. I just re flashed my router back to Kalzo's and i cant seem to install it now. I even tried on the RC6 trunk build, got the same error.

Please note that it takes about 10-13MB to install. So someone who is about to re flash their router would be the best candidate to test it out.
i notice the date on the repo pacakges is now MAR 29th. Did something change perhaps?

Would really appreciate if someone can let me know if they get the same error so that rule out its something on my side.
thanks

Post #3993
LogicoZone
31 Mar 2015, 04:35

I got roadblock on compile a build at e2fsprogs-1.42.12. Not sure why I have this issue since this package is fairly old. Can anyone share experience? 

make[6]: Entering directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
    SUBST ../lib/dirpaths.h
    CC mke2fs.c
    CC util.c
    CC ../e2fsck/profile.c
    CC mk_hugefiles.c
    LD mke2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC badblocks.c
    LD badblocks
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC tune2fs.c
    LD tune2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC dumpe2fs.c
    LD dumpe2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC blkid.c
    LD blkid
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC logsave.c
    LD logsave
    LD e2image
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC fsck.c
    CC base_device.c
    CC ismounted.c
    LD fsck
    CC e2undo.c
    LD e2undo
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC chattr.c
    LD chattr
    CC lsattr.c
    LD lsattr
    CC mklost+found.c
    LD mklost+found
    CC filefrag.c
    LD filefrag
    CC e2freefrag.c
    LD e2freefrag
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC e4defrag.c
e4defrag.c:197:2: error: #error sync_file_range not available!
 #error sync_file_range not available!
  ^
make[6]: *** [e4defrag.o] Error 1
make[6]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
make[5]: *** [all-progs-recursive] Error 1
make[5]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12'
make[4]: *** [all] Error 2
Post #3994
wrtpat
31 Mar 2015, 04:59
LogicoZone wrote:

I got roadblock on compile a build at e2fsprogs-1.42.12. Not sure why I have this issue since this package is fairly old. Can anyone share experience? 

make[6]: Entering directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
    SUBST ../lib/dirpaths.h
    CC mke2fs.c
    CC util.c
    CC ../e2fsck/profile.c
    CC mk_hugefiles.c
    LD mke2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC badblocks.c
    LD badblocks
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC tune2fs.c
    LD tune2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC dumpe2fs.c
    LD dumpe2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC blkid.c
    LD blkid
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC logsave.c
    LD logsave
    LD e2image
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC fsck.c
    CC base_device.c
    CC ismounted.c
    LD fsck
    CC e2undo.c
    LD e2undo
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC chattr.c
    LD chattr
    CC lsattr.c
    LD lsattr
    CC mklost+found.c
    LD mklost+found
    CC filefrag.c
    LD filefrag
    CC e2freefrag.c
    LD e2freefrag
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC e4defrag.c
e4defrag.c:197:2: error: #error sync_file_range not available!
 #error sync_file_range not available!
  ^
make[6]: *** [e4defrag.o] Error 1
make[6]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
make[5]: *** [all-progs-recursive] Error 1
make[5]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12'
make[4]: *** [all] Error 2

Saw something about this come across the mail list today.
There's something in the works... coming soon.
See here -> https://www.mail-archive.com/openwrt-de … 30391.html

(Last edited by wrtpat on 31 Mar 2015, 05:00)

Post #3995
Kaloz
31 Mar 2015, 09:30

New images with the latest changes (eg. from about 20 minutes ago) will be up soon.

Post #3996
sachingopal
31 Mar 2015, 10:23

@DavidMcWRT Thanks for the info, really appreciate your time. Also though the local vendor told me that the shipped version is v 1.1, when i rechecked it was v1. Sorry for the confusion.

Sachin

Post #3997
belliash
31 Mar 2015, 10:25
sachingopal wrote:

@DavidMcWRT Thanks for the info, really appreciate your time. Also though the local vendor told me that the shipped version is v 1.1, when i rechecked it was v1. Sorry for the confusion.

Sachin

Where do you chceck it?

Post #3998
LogicoZone
31 Mar 2015, 11:37

Just compiled with latest mwlwifi driver. I can confirm the performance issue is fixed. Now, with ext4 hard drive connected to USB3 port, I can download file from SAMBA share to my local laptop over 5Ghz with speed of 30 MB/s. In the past, I couldn't even download couple kbytes.

Awesome. This router is getting to perfection.

Here is the iperf3 test with latest driver.

Connecting to host 10.0.1.201, port 5201
[  4] local 10.0.1.53 port 52272 connected to 10.0.1.201 port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.00   sec  17.5 MBytes   146 Mbits/sec             
[  4]   1.00-2.00   sec  24.8 MBytes   208 Mbits/sec             
[  4]   2.00-3.00   sec  24.4 MBytes   205 Mbits/sec             
[  4]   3.00-4.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   4.00-5.00   sec  24.9 MBytes   209 Mbits/sec             
[  4]   5.00-6.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   6.00-7.00   sec  23.1 MBytes   194 Mbits/sec             
[  4]   7.00-8.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   8.00-9.01   sec  24.4 MBytes   204 Mbits/sec             
[  4]   9.01-10.00  sec  24.6 MBytes   207 Mbits/sec             
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec   239 MBytes   200 Mbits/sec              sender
[  4]   0.00-10.00  sec   239 MBytes   200 Mbits/sec              receiver

iperf Done.

(Last edited by LogicoZone on 31 Mar 2015, 11:51)

Post #3999
belliash
31 Mar 2015, 13:11
LogicoZone wrote:

Just compiled with latest mwlwifi driver. I can confirm the performance issue is fixed. Now, with ext4 hard drive connected to USB3 port, I can download file from SAMBA share to my local laptop over 5Ghz with speed of 30 MB/s. In the past, I couldn't even download couple kbytes.

Awesome. This router is getting to perfection.

Here is the iperf3 test with latest driver.

Connecting to host 10.0.1.201, port 5201
[  4] local 10.0.1.53 port 52272 connected to 10.0.1.201 port 5201
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-1.00   sec  17.5 MBytes   146 Mbits/sec             
[  4]   1.00-2.00   sec  24.8 MBytes   208 Mbits/sec             
[  4]   2.00-3.00   sec  24.4 MBytes   205 Mbits/sec             
[  4]   3.00-4.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   4.00-5.00   sec  24.9 MBytes   209 Mbits/sec             
[  4]   5.00-6.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   6.00-7.00   sec  23.1 MBytes   194 Mbits/sec             
[  4]   7.00-8.00   sec  25.1 MBytes   211 Mbits/sec             
[  4]   8.00-9.01   sec  24.4 MBytes   204 Mbits/sec             
[  4]   9.01-10.00  sec  24.6 MBytes   207 Mbits/sec             
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bandwidth
[  4]   0.00-10.00  sec   239 MBytes   200 Mbits/sec              sender
[  4]   0.00-10.00  sec   239 MBytes   200 Mbits/sec              receiver

iperf Done.

hmm ... iperf3 doesn't show 30MB/s ;-)
BTW: Did you compile trunk? Or you had to modify anything?

(Last edited by belliash on 31 Mar 2015, 13:17)

Post #4000
jklap
31 Mar 2015, 13:53
LogicoZone wrote:

I got roadblock on compile a build at e2fsprogs-1.42.12. Not sure why I have this issue since this package is fairly old. Can anyone share experience? 

make[6]: Entering directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
    SUBST ../lib/dirpaths.h
    CC mke2fs.c
    CC util.c
    CC ../e2fsck/profile.c
    CC mk_hugefiles.c
    LD mke2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC badblocks.c
    LD badblocks
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC tune2fs.c
    LD tune2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC dumpe2fs.c
    LD dumpe2fs
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC blkid.c
    LD blkid
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC logsave.c
    LD logsave
    LD e2image
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC fsck.c
    CC base_device.c
    CC ismounted.c
    LD fsck
    CC e2undo.c
    LD e2undo
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC chattr.c
    LD chattr
    CC lsattr.c
    LD lsattr
    CC mklost+found.c
    LD mklost+found
    CC filefrag.c
    LD filefrag
    CC e2freefrag.c
    LD e2freefrag
../lib/libext2fs.so: warning: the use of OBSOLESCENT `utime' is discouraged, use `utimes'
    CC e4defrag.c
e4defrag.c:197:2: error: #error sync_file_range not available!
 #error sync_file_range not available!
  ^
make[6]: *** [e4defrag.o] Error 1
make[6]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12/misc'
make[5]: *** [all-progs-recursive] Error 1
make[5]: Leaving directory `/home/parallels/OpenWRT/build_dir/target-arm_cortex-a9+vfpv3_uClibc-0.9.33.2_eabi/e2fsprogs-1.42.12'
make[4]: *** [all] Error 2

I'm running into the same issue after I updated trunk, https://dev.openwrt.org/ticket/19336.

Note the previous comment was linking to the fallocate64 issue which has been fixed already I believe.

Will be backing trunk down until the sync_file_range is fixed.

FYI, in case you want to update to the latest *before* the break, try r45074, or 7a92622eb3c15e2ee396cff35aadcb9b46783972 or maybe 45090, 88057fdc36d4bb170f3deae464ba1092c5fdbc7e... which looks to related.

Edit: I was able to build against r45074

(Last edited by jklap on 31 Mar 2015, 16:38)