Re: [LEDE-DEV] [PATCH v5] base-files: seed /dev/urandom

2016-07-19 7:40 GMT+02:00 Etienne Champetier : > 2016-07-19 7:09 GMT+02:00 Jo-Philipp Wich : >> Hi Daniel, >> >> interesting failure mode ;( >> >> To me it seems the easiest fix is changing the getrandom syscall wrapper >> utility to use the GRND_NONBLOCK flag in order to be interruptible by >> sig

Re: [LEDE-DEV] [PATCH v5] base-files: seed /dev/urandom

2016-07-19 7:09 GMT+02:00 Jo-Philipp Wich : > Hi Daniel, > > interesting failure mode ;( > > To me it seems the easiest fix is changing the getrandom syscall wrapper > utility to use the GRND_NONBLOCK flag in order to be interruptible by > signals. > > Thoughts? According to the man page it's alre

Re: [LEDE-DEV] [PATCH v5] base-files: seed /dev/urandom

Hi Daniel, interesting failure mode ;( To me it seems the easiest fix is changing the getrandom syscall wrapper utility to use the GRND_NONBLOCK flag in order to be interruptible by signals. Thoughts? ~ Jo ___ Lede-dev mailing list Lede-dev@lists.inf

Re: [LEDE-DEV] wrong switch configuration

Hi Hartmut, can you please provide the output of "./scripts/diffconfig.sh" in your buildroot and the contents of "/etc/board.json" on your device? Can you also see if the configuration looks saner after an "rm -f /etc/config/network /etc/config/system; /bin/config_generate" ? ~ Jo _

[LEDE-DEV] wrong switch configuration

Hi, I'm using a modified TP-Link wdr3600 router. Flash is replaced by by a 16MB type. I update the build system for the larger flash. I'm using this configuration since 2014 without any problems (previously with OpenWrt, now with Lede). Currently I've problems with the generated images. If I'm

Re: [LEDE-DEV] [PATCH v2 8/8] apm821xx: add support for the Netgear Centria N900 WNDR4700/WNDR4720 tested-by

On 07/13/2016 02:23 PM, Christian Lamparter wrote: > From: Christian Lamparter > > This patch adds support for Netgear Centria N900 WNDR4700/WNDR4720 > > hardware highlights: > > CPU: AMCC PowerPC APM82181 Rev. E at 1000 MHz (PLB=166, OPB=83, EBC=83 MHz) > Security support, Boot ROM Loca

Re: [LEDE-DEV] [PATCH v5] base-files: seed /dev/urandom

Hi Etienne, Hi Arjen, Hi John, Hi Jo, while wondering which process is blocking my rootfs from being unmounted for sysupgrade I discovered that getrandom story going on and while waiting for blocking I/O read from /dev/urandom, it even ignored any TERM or KILL signals. This then prevented the root

[LEDE-DEV] WDS and WPA Enterprise: Can it be done

Hi, I try to setup a station that connects to an ap via WDS and WPA Enterprise. The idea is that the wds connection at the ap is put in the right vlan, cf packetfence. This works will for stations connecting via WPA enterprise *without* WDS, using the the dynamic vlan feature of hostapd However,

Re: [LEDE-DEV] How to compile hostapd such that there is a full WPA2 Enterprise auhenticator/supplicant?

Hi, I never submitted a patch, and the LEDE site does not really give clue how to do so. The diff below is for the file: build_dir/target-mips_34kc_musl-1.1.15/hostapd-wpad-full/hostapd-2016-06-15/ hostapd/config_file.c It addresses the issue that the radius_das_client line in the config file is

[LEDE-DEV] [PATCH] x86/generic: enable CONFIG_SATA_VIA

This enables support for VIA SATA. --- target/linux/x86/generic/config-default | 1 + 1 file changed, 1 insertion(+) diff --git a/target/linux/x86/generic/config-default b/target/linux/x86/generic/config-default index 6d6f94a..6d19b59 100644 --- a/target/linux/x86/generic/config-default +++ b/ta

[LEDE-DEV] [PATCH] ledtrig-usbdev: use upstream function for iterating USB devices

This will allow us to drop LEDE patch adding usb_find_device_by_name. Signed-off-by: Rafał Miłecki --- .../generic/files/drivers/leds/ledtrig-usbdev.c| 32 ++ 1 file changed, 27 insertions(+), 5 deletions(-) diff --git a/target/linux/generic/files/drivers/leds/ledtrig-us

Re: [LEDE-DEV] How to compile hostapd such that there is a full WPA2 Enterprise auhenticator/supplicant?

Hi "frm", can you send a proper diff? WIthout any context it is totally unclear what you intend to tell by pasting the code below. Also how does it relate to dynamic vlan support? ~ Jo ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists

Re: [LEDE-DEV] How to compile hostapd such that there is a full WPA2 Enterprise auhenticator/supplicant?

Hi Felix, I guess you missed the big in hostapd/config_file.c: /* trunk val, such val contains only the ip address and not the secret */ *secret= '\0'; -- static int hostapd_parse_das_client(struct hostapd_bss_config *bss, const char *

Re: [LEDE-DEV] How to compile hostapd such that there is a full WPA2 Enterprise auhenticator/supplicant?

On 2016-07-18 13:22, frm frm wrote: > In package/network/services/hostapd/Makefile > Disable CONFIG_IBSS_RSN (enables the dynamic vlan code) Fixed in the 75329fc1618c (r1010) - Felix ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.in

Re: [LEDE-DEV] How to compile hostapd such that there is a full WPA2 Enterprise auhenticator/supplicant?

In package/network/services/hostapd/Makefile Disable CONFIG_IBSS_RSN (enables the dynamic vlan code) In build_dir/target-mips_34kc_musl-1.1.15/hostapd-wpad-full/hostapd-2016-06-15/ hostapd/config_file.c: static int hostapd_parse_das_client(struct hostapd_bss_config *bss, const char *val) { char *s