[LEDE-DEV] [PATCH 1/3] mac80211: rt2x00: improve eeprom_file property handling

[Mathias Kresin]

Call the rt2x00lib_get_eeprom_file_name only once and from the function
where the EEPROM filename is required.

Error only out if an EEPROM file is mandatory. Use the
REQUIRE_EEPROM_FILE bit to determine if it is mandatory.

Do not set the REQUIRE_EEPROM_FILE bit while requesting an EEPROM file.
It should be (and is) set before requesting an EEPROM file.

Do not redirect users to upstream while using a function of a custom
patch.

Signed-off-by: Mathias Kresin 
---
 .../603-rt2x00-introduce-rt2x00eeprom.patch| 24 --
 ...-eeprom-on-SoC-from-a-mtd-device-defines-.patch |  4 ++--
 2 files changed, 11 insertions(+), 17 deletions(-)

diff --git 
a/package/kernel/mac80211/patches/603-rt2x00-introduce-rt2x00eeprom.patch 
b/package/kernel/mac80211/patches/603-rt2x00-introduce-rt2x00eeprom.patch
index 8ae5da3..0b191e6 100644
--- a/package/kernel/mac80211/patches/603-rt2x00-introduce-rt2x00eeprom.patch
+++ b/package/kernel/mac80211/patches/603-rt2x00-introduce-rt2x00eeprom.patch
@@ -152,7 +152,7 @@
  
 --- /dev/null
 +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c
-@@ -0,0 +1,111 @@
+@@ -0,0 +1,105 @@
 +/*
 +  Copyright (C) 2004 - 2009 Ivo van Doorn 
 +  Copyright (C) 2004 - 2009 Gertjan van Wingerde 
@@ -203,13 +203,14 @@
 +  int retval;
 +
 +  ee_name = rt2x00lib_get_eeprom_file_name(rt2x00dev);
-+  if (!ee_name) {
-+  rt2x00_err(rt2x00dev,
-+ "Invalid EEPROM filename.\n"
-+ "Please file bug report to %s.\n", DRV_PROJECT);
++  if (!ee_name && test_bit(REQUIRE_EEPROM_FILE, &rt2x00dev->cap_flags)) {
++  rt2x00_err(rt2x00dev, "Required EEPROM name is missing.");
 +  return -EINVAL;
 +  }
 +
++  if (!ee_name)
++  return 0;
++
 +  rt2x00_info(rt2x00dev, "Loading EEPROM data from '%s'.\n", ee_name);
 +
 +  retval = request_firmware(&ee, ee_name, rt2x00dev->dev);
@@ -245,16 +246,9 @@
 +{
 +  int retval;
 +
-+  if (!rt2x00lib_get_eeprom_file_name(rt2x00dev))
-+  return 0;
-+
-+  set_bit(REQUIRE_EEPROM_FILE, &rt2x00dev->cap_flags);
-+
-+  if (!rt2x00dev->eeprom_file) {
-+  retval = rt2x00lib_request_eeprom_file(rt2x00dev);
-+  if (retval)
-+  return retval;
-+  }
++  retval = rt2x00lib_request_eeprom_file(rt2x00dev);
++  if (retval)
++  return retval;
 +
 +  return 0;
 +}
diff --git 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
index 5dceff8..5097304 100644
--- 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
+++ 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
@@ -97,5 +97,5 @@ Signed-off-by: John Crispin 
 +  return 0;
 +
ee_name = rt2x00lib_get_eeprom_file_name(rt2x00dev);
-   if (!ee_name) {
-   rt2x00_err(rt2x00dev,
+   if (!ee_name && test_bit(REQUIRE_EEPROM_FILE, &rt2x00dev->cap_flags)) {
+   rt2x00_err(rt2x00dev, "Required EEPROM name is missing.");
-- 
2.7.4


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH 3/3] mac80211: rt2x00: add mtd-eeprom swab function

[Mathias Kresin]

Most of the lantiq devices with ralink wifi have the EEPROM stored
in big endian byte order in flash, but the driver expects the EEPROM to
be in little endian.

Signed-off-by: Mathias Kresin 
---
 ...x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch | 14 +-
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
index 6c264ef..a98b49c 100644
--- 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
+++ 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
@@ -7,8 +7,8 @@ Subject: [PATCH] rt2x00: load eeprom on SoC from a mtd device 
defines inside
 Signed-off-by: John Crispin 
 ---
  drivers/net/wireless/ralink/rt2x00/Kconfig|  1 +
- drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c | 61 +++
- 2 files changed, 62 insertions(+)
+ drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c | 65 +++
+ 2 files changed, 66 insertions(+)
 
 --- a/drivers/net/wireless/ralink/rt2x00/Kconfig
 +++ b/drivers/net/wireless/ralink/rt2x00/Kconfig
@@ -22,7 +22,7 @@ Signed-off-by: John Crispin 
  Supported chips: RT2880, RT3050, RT3052, RT3350, RT3352.
 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c
 +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c
-@@ -26,11 +26,69 @@
+@@ -26,11 +26,73 @@
  
  #include 
  #include 
@@ -40,7 +40,7 @@ Signed-off-by: John Crispin 
 +  static struct firmware mtd_fw;
 +  struct device_node *np = rt2x00dev->dev->of_node, *mtd_np = NULL;
 +  size_t retlen, len = rt2x00dev->ops->eeprom_size;
-+  int size, offset = 0;
++  int i, size, offset = 0;
 +  struct mtd_info *mtd;
 +  const char *part;
 +  const __be32 *list;
@@ -79,6 +79,10 @@ Signed-off-by: John Crispin 
 +  return ret;
 +  }
 +
++  if (of_find_property(np, "ralink,mtd-eeprom-swap", NULL))
++  for (i = 0; i < len/sizeof(u16); i++)
++  rt2x00dev->eeprom[i] = swab16(rt2x00dev->eeprom[i]);
++
 +  rt2x00dev->eeprom_file = &mtd_fw;
 +  mtd_fw.size = len;
 +  mtd_fw.data = (const u8 *) rt2x00dev->eeprom;
@@ -92,7 +96,7 @@ Signed-off-by: John Crispin 
  static const char *
  rt2x00lib_get_eeprom_file_name(struct rt2x00_dev *rt2x00dev)
  {
-@@ -58,6 +116,9 @@ static int rt2x00lib_request_eeprom_file
+@@ -58,6 +120,9 @@ static int rt2x00lib_request_eeprom_file
const char *ee_name;
int retval;
  
-- 
2.7.4


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH 2/3] mac80211: rt2x00: remove eeprom filename dependency from mtd-eeprom

[Mathias Kresin]

The ralink,mtd-eeprom device tree property doesn't require the
ralink,eeprom property to work.

Rework the error handling and user notification as well. Do not log an
error if the mtd-eeprom parameter isn't used. It could be intentional
and should not scare the user.

Check if the number of bytes read from the mtd devices matches the
requested number of bytes.

In case of an mtd read error, give a hint to the user which partition
was tried to read from.

In case everything is fine, notify the user as well.

Signed-off-by: Mathias Kresin 
---
 ...-eeprom-on-SoC-from-a-mtd-device-defines-.patch | 27 --
 1 file changed, 15 insertions(+), 12 deletions(-)

diff --git 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
index 5097304..6c264ef 100644
--- 
a/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
+++ 
b/package/kernel/mac80211/patches/605-rt2x00-load-eeprom-on-SoC-from-a-mtd-device-defines-.patch
@@ -6,9 +6,9 @@ Subject: [PATCH] rt2x00: load eeprom on SoC from a mtd device 
defines inside
 
 Signed-off-by: John Crispin 
 ---
- drivers/net/wireless/ralink/rt2x00/Kconfig |1 +
- drivers/net/wireless/ralink/rt2x00/rt2800pci.c |   44 
++-
- 2 files changed, 39 insertions(+), 6 deletions(-)
+ drivers/net/wireless/ralink/rt2x00/Kconfig|  1 +
+ drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c | 61 +++
+ 2 files changed, 62 insertions(+)
 
 --- a/drivers/net/wireless/ralink/rt2x00/Kconfig
 +++ b/drivers/net/wireless/ralink/rt2x00/Kconfig
@@ -22,7 +22,7 @@ Signed-off-by: John Crispin 
  Supported chips: RT2880, RT3050, RT3052, RT3350, RT3352.
 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c
 +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00eeprom.c
-@@ -26,11 +26,66 @@
+@@ -26,11 +26,69 @@
  
  #include 
  #include 
@@ -47,10 +47,8 @@ Signed-off-by: John Crispin 
 +  phandle phandle;
 +
 +  list = of_get_property(np, "ralink,mtd-eeprom", &size);
-+  if (!list) {
-+  dev_err(rt2x00dev->dev, "failed to load eeprom property\n");
++  if (!list)
 +  return -ENOENT;
-+  }
 +
 +  phandle = be32_to_cpup(list++);
 +  if (phandle)
@@ -76,11 +74,16 @@ Signed-off-by: John Crispin 
 +  ret = mtd_read(mtd, offset, len, &retlen, (u_char *) rt2x00dev->eeprom);
 +  put_mtd_device(mtd);
 +
-+  if (!ret) {
-+  rt2x00dev->eeprom_file = &mtd_fw;
-+  mtd_fw.size = len;
-+  mtd_fw.data = (const u8 *) rt2x00dev->eeprom;
++  if ((retlen != rt2x00dev->ops->eeprom_size) || ret) {
++  dev_err(rt2x00dev->dev, "failed to load eeprom from device 
\"%s\"\n", part);
++  return ret;
 +  }
++
++  rt2x00dev->eeprom_file = &mtd_fw;
++  mtd_fw.size = len;
++  mtd_fw.data = (const u8 *) rt2x00dev->eeprom;
++
++  dev_info(rt2x00dev->dev, "loaded eeprom from mtd device \"%s\"\n", 
part);
 +#endif
 +
 +  return ret;
@@ -89,7 +92,7 @@ Signed-off-by: John Crispin 
  static const char *
  rt2x00lib_get_eeprom_file_name(struct rt2x00_dev *rt2x00dev)
  {
-@@ -58,6 +113,9 @@ static int rt2x00lib_request_eeprom_file
+@@ -58,6 +116,9 @@ static int rt2x00lib_request_eeprom_file
const char *ee_name;
int retval;
  
-- 
2.7.4


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH 0/3] mac80211: rt2x00: improve device tree eeprom handling

[Mathias Kresin]

The series aims to make the out of tree rt2x00 eeprom handling
functions better usable with  ralink pci chips. To my knowledge the
eeprom handling functions are used only on ramips so far.

The overall purpose is to get rid of the lantiq specific ralink eeprom
handling functions in favour to the ones already patched into the
driver.

A full working example can be found in the ralink branch of my staging
repository[0].

As long as no one raises concerns, I will merge this series during the
next days. An ACK of corse would speed up things.

[0] 
https://git.lede-project.org/?p=lede/mkresin/staging.git;a=shortlog;h=refs/heads/rt2x00

Mathias Kresin (3):
  mac80211: rt2x00: improve eeprom_file property handling
  mac80211: rt2x00: remove eeprom filename dependency from mtd-eeprom
  mac80211: rt2x00: add mtd-eeprom swab function

 .../603-rt2x00-introduce-rt2x00eeprom.patch| 24 ++
 ...-eeprom-on-SoC-from-a-mtd-device-defines-.patch | 37 +-
 2 files changed, 31 insertions(+), 30 deletions(-)

-- 
2.7.4


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] New CT ath10k firmware is available

[Ben Greear]

I don't really have time to work specifically on LEDE much.

But, if someone else takes care of the rest, then my firmware
can hopefully work well on its radio.

Thanks,
Ben


On 11/03/2016 07:48 PM, liudengf...@kunteng.org wrote:

thx for your kindly reply.
by the way, do u plan to support R7800, if plan, I think its a good idea.



--

liudengf...@kunteng.org

*From:* Ben Greear 
*Date:* 2016-11-04 10:39
*To:* liudengf...@kunteng.org ; LEDE Development 
List 
*Subject:* Re: [LEDE-DEV] New CT ath10k firmware is available
The 10.4 firmware supports 9984.  It is one of the chips I test against.
But, I am using it as a NIC in an x86-64 platform, not R7800 specifically.
Thanks,
Ben
On 11/03/2016 07:24 PM, liudengf...@kunteng.org wrote:
 > Does this version support QCA9984? I have tested ct ath10k in R7800 
before, but it can not run in LEDE.
 >
 >





--
 > liudengf...@kunteng.org
 >
 > *From:* Ben Greear 
 > *Date:* 2016-11-04 09:12
 > *To:* LEDE Development List 
 > *Subject:* [LEDE-DEV] New CT ath10k firmware is available
 > For those of you using my 10.1 or 10.4 CT ath10k firmware in LEDE, I 
have pushed new beta
 > builds for all of the various images I build.
 > For those with capability and time, it would help me out if you 
could do a bit of
 > testing of these firmware on LEDE.  If I can get some successful 
reports, I will
 > post a LEDE patch to update to these latest firmwares.
 > http://www.candelatech.com/ath10k.php
 > In particular, I have no 9887 hardware, so if someone tests that 
image I
 > am interested to know how it goes.
 > Thanks,
 > Ben
 > --
 > Ben Greear 
 > Candela Technologies Inc  http://www.candelatech.com
 > ___
 > Lede-dev mailing list
 > Lede-dev@lists.infradead.org
 > http://lists.infradead.org/mailman/listinfo/lede-dev
 >
--
Ben Greear 
Candela Technologies Inc  http://www.candelatech.com



--
Ben Greear 
Candela Technologies Inc  http://www.candelatech.com

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] New CT ath10k firmware is available

[Ben Greear]

The 10.4 firmware supports 9984.  It is one of the chips I test against.

But, I am using it as a NIC in an x86-64 platform, not R7800 specifically.

Thanks,
Ben

On 11/03/2016 07:24 PM, liudengf...@kunteng.org wrote:

Does this version support QCA9984? I have tested ct ath10k in R7800 before, but 
it can not run in LEDE.



--

liudengf...@kunteng.org

*From:* Ben Greear 
*Date:* 2016-11-04 09:12
*To:* LEDE Development List 
*Subject:* [LEDE-DEV] New CT ath10k firmware is available
For those of you using my 10.1 or 10.4 CT ath10k firmware in LEDE, I have 
pushed new beta
builds for all of the various images I build.
For those with capability and time, it would help me out if you could do a 
bit of
testing of these firmware on LEDE.  If I can get some successful reports, I 
will
post a LEDE patch to update to these latest firmwares.
http://www.candelatech.com/ath10k.php
In particular, I have no 9887 hardware, so if someone tests that image I
am interested to know how it goes.
Thanks,
Ben
--
Ben Greear 
Candela Technologies Inc  http://www.candelatech.com
___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev



--
Ben Greear 
Candela Technologies Inc  http://www.candelatech.com

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] New CT ath10k firmware is available

[Ben Greear]

For those of you using my 10.1 or 10.4 CT ath10k firmware in LEDE, I have 
pushed new beta
builds for all of the various images I build.

For those with capability and time, it would help me out if you could do a bit 
of
testing of these firmware on LEDE.  If I can get some successful reports, I will
post a LEDE patch to update to these latest firmwares.

http://www.candelatech.com/ath10k.php

In particular, I have no 9887 hardware, so if someone tests that image I
am interested to know how it goes.

Thanks,
Ben

--
Ben Greear 
Candela Technologies Inc  http://www.candelatech.com


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] How LEDE knows when to "reinstall" base-files (after revision change)?

[Rafał Miłecki]

On 11/03/2016 06:44 PM, Jo-Philipp Wich wrote:

I think its not actually noticing all git commits but rather most of them.

The base-files package registers some explicit file depends using:
  PKG_FILE_DEPENDS:=$(PLATFORM_DIR)/ $(GENERIC_PLATFORM_DIR)/base-files/

which means that its stamp files are invalided whenever any file in the
directories above is modified.


Please take a look at my test below. I added file to the top dir which should
be insignificant but still, base-files noticed it and updated openwrt_version.

$ cat 
build_dir/target-*/linux-*/base-files/ipkg-arm_cortex-a9/base-files/etc/openwrt_version
r2094+2
$ echo foo > bar.txt
$ git add bar.txt
$ git commit -m FooBar
$ make package/base-files/install
 make[1] package/base-files/install
 make[2] -C package/base-files install
$ cat 
build_dir/target-*/linux-*/base-files/ipkg-arm_cortex-a9/base-files/etc/openwrt_version
r2094+3

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH] apm821xx: add back end-of-UBI marker for the WNDR4700 and MR24

[Christian Lamparter]

The patch "build: drop UBI EOF marker from images by default"
(commit d27bce8d28eb129af0abd9c80a7756301b7d588a) removed the
end-of-UBI marker. Without this marker, the boards will panic
during boot:

[1.683458] ubi0: attaching mtd4
[1.696181] ubi0 warning: scan_peb: valid VID header but corrupted EC header 
at PEB 31
[1.704367] ubi0 error: scan_peb: bad image sequence number 549886691 in PEB 
32, expected 184585623
[1.713377] Erase counter header dump:
[1.717110]  magic  0x55424923
[1.720843]  version1
[1.723797]  ec 0
[1.726752]  vid_hdr_offset 512
[1.729880]  data_offset2048
[1.733094]  image_seq  549886691
[1.736740]  hdr_crc0x92ba8130
[1.740472] erase counter header hexdump:
[1.744493] ubi0 error: ubi_attach_mtd_dev: failed to attach mtd4, error -22
[1.751528] UBI error: cannot attach mtd4
[1.755373] hctosys: unable to open rtc device (rtc0)
[1.761130] VFS: Cannot open root device "(null)" or unknown-block(0,0): 
error -6
[1.768604] Please append a correct "root=" boot option; here are the 
available partitions:
[...]
[1.856992] Kernel panic - not syncing: VFS: Unable to mount root fs on 
unknown-block(0,0)
[1.866519] Rebooting in 1 seconds..Auto calibration ---

Signed-off-by: Christian Lamparter 
---
 target/linux/apm821xx/image/Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/target/linux/apm821xx/image/Makefile 
b/target/linux/apm821xx/image/Makefile
index f8aecdd..c032c54 100644
--- a/target/linux/apm821xx/image/Makefile
+++ b/target/linux/apm821xx/image/Makefile
@@ -80,6 +80,7 @@ define Device/mr24
  MerakiAdd-dtb | pad-to 2047k | MerakiAdd-initramfs | \
  MerakiNAND
   IMAGE/sysupgrade.tar := sysupgrade-tar
+  UBINIZE_OPTS := -E 5
 endef
 TARGET_DEVICES += mr24
 
@@ -162,6 +163,7 @@ define Device/WNDR4700
   IMAGE/sysupgrade.tar := sysupgrade-tar
   NETGEAR_BOARD_ID := WNDR4700
   NETGEAR_HW_ID := 29763875+128+256
+  UBINIZE_OPTS := -E 5
 endef
 TARGET_DEVICES += WNDR4700
 
-- 
2.10.2


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [PATCH v2] include: Cortex-A53 is also an AArch64 CPU

[Álvaro Fernández Rojas]

ACK.

El 3/11/16 a las 19:16, Florian Fainelli escribió:
> Specifying a mtune option with cortex-a53 is also valid for an aarch64
> toolchain
> 
> Fixes: SVN 48964
> Signed-off-by: Florian Fainelli 
> ---
> Changes in v2:
> - add the option to the aarch64 architecture conditional
> 
>  include/target.mk | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/include/target.mk b/include/target.mk
> index c9c02fa03632..a8d3b2b24929 100644
> --- a/include/target.mk
> +++ b/include/target.mk
> @@ -212,6 +212,7 @@ ifeq ($(DUMP),1)
>ifeq ($(ARCH),aarch64)
>  CPU_TYPE ?= armv8-a
>  CPU_CFLAGS_armv8-a = -mcpu=armv8-a
> +CPU_CFLAGS_cortex-a53 = -mcpu=cortex-a53
>endif
>ifeq ($(ARCH),arc)
>  CPU_TYPE ?= arc700
> 

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH v2] include: Cortex-A53 is also an AArch64 CPU

[Florian Fainelli]

Specifying a mtune option with cortex-a53 is also valid for an aarch64
toolchain

Fixes: SVN 48964
Signed-off-by: Florian Fainelli 
---
Changes in v2:
- add the option to the aarch64 architecture conditional

 include/target.mk | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/target.mk b/include/target.mk
index c9c02fa03632..a8d3b2b24929 100644
--- a/include/target.mk
+++ b/include/target.mk
@@ -212,6 +212,7 @@ ifeq ($(DUMP),1)
   ifeq ($(ARCH),aarch64)
 CPU_TYPE ?= armv8-a
 CPU_CFLAGS_armv8-a = -mcpu=armv8-a
+CPU_CFLAGS_cortex-a53 = -mcpu=cortex-a53
   endif
   ifeq ($(ARCH),arc)
 CPU_TYPE ?= arc700
-- 
2.10.1


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] How LEDE knows when to "reinstall" base-files (after revision change)?

[Jo-Philipp Wich]

Hi Rafal,

I think its not actually noticing all git commits but rather most of them.

The base-files package registers some explicit file depends using:
  PKG_FILE_DEPENDS:=$(PLATFORM_DIR)/ $(GENERIC_PLATFORM_DIR)/base-files/

which means that its stamp files are invalided whenever any file in the
directories above is modified.

~ Jo



signature.asc
Description: OpenPGP digital signature
___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] How LEDE knows when to "reinstall" base-files (after revision change)?

[Rafał Miłecki]

From: Rafał Miłecki 

I'm trying to understand how REVISION variable and getver.sh work. There is this
nice feature: whenever a new git commit appears, getver.sh is used to calculate
new revision and it's used for /etc/openwrt_version. LEDE never misses an update
of git commit.

Also as long as git commit stays the same, calling
make package/base-files/install V=s
has no effect.

I was trying to do the same trick for adding info on time of compilation to the
rootfs content. Please take a look at patch below. Obviously my TIME_TEST
variable changes on every "make" call, but content of
build_dir/target-*/linux-*/base-files/ipkg-arm_cortex-a9/base-files/etc/time_test
isn't updated as I expected.

Can someone explain me, how does it work for REVISION and getver.sh, please?
---

--- a/include/toplevel.mk
+++ b/include/toplevel.mk
@@ -25,6 +25,9 @@ export SOURCE_DATE_EPOCH
 export GIT_CONFIG_PARAMETERS='core.autocrlf=false'
 export MAKE_JOBSERVER=$(filter --jobserver%,$(MAKEFLAGS))
 
+TIME_TEST:=$(shell date +"%T")
+export TIME_TEST
+
 # prevent perforce from messing with the patch utility
 unexport P4PORT P4USER P4CONFIG P4CLIENT
 
--- a/package/base-files/Makefile
+++ b/package/base-files/Makefile
@@ -134,6 +134,8 @@ define Package/base-files/install
$(1)/etc/preinit \
$(1)/etc/profile
 
+   echo $(TIME_TEST) > $(1)/etc/time_test
+
mkdir -p $(1)/CONTROL
mkdir -p $(1)/dev
mkdir -p $(1)/etc/crontabs

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] LEDE Forum now open for business

[Ted Hess]

Hi all -

Here it is: https://forum.lede-project.org

You can now self-register without admin approval.

With a lot of help and comments from a bunch of the current users, I think we
now have a very usable site. We may still tweak the style, layouts and options
as we mature the site so don't be upset if the site changes.

So, visit the site, register and leave us a comment in the Site Feedback
category (https://forum.lede-project.org/c/site-feedback).

/ted


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[Dave Taht]

awesome detective work, everybody. thx.

(I'm at linux plumbers, puzzling over why my wifi demo broke)

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[xinglp]

2016-11-03 22:28 GMT+08:00 Yousong Zhou :
> On 3 November 2016 at 22:20, xinglp  wrote:
>> 2016-11-03 18:34 GMT+08:00 Felix Fietkau :
>>> On 2016-11-03 04:18, xinglp wrote:
 I have tried to use several phone and app to play video via samba, I
 got broken picture or not fluent after this commit.
 Hardware is ar9331.
>>> Should be fixed in the latest version, please test.
>>
>> 'wifi detect' output nothing with latest version, but
>> /sys/class/net/wlan0 exist.
>>
>
> wifi detect has just been changed into a nop with
> 5f8f8a366136a07df661e31decce2458357c167a   The new method is "wifi
> config"
The wifi works fine when I streaming video now, thanks
>
> yousong
>
>> sh -x /sbin/wifi detect
>> + . /lib/functions.sh
>> + N=
>>
>> + _C=0
>> + NO_EXPORT=1
>> + LOAD_STATE=1
>> + LIST_SEP=
>> + reset_cb
>> + [ -z  -a -f /lib/config/uci.sh ]
>> + . /lib/config/uci.sh
>> + CONFIG_APPEND=
>> + . /usr/share/libubox/jshn.sh
>> + DEVICES=
>> + DRIVERS=
>> + include /lib/wifi
>> + local file
>> + ls /lib/wifi/mac80211.sh
>> + . /lib/wifi/mac80211.sh
>> + append DRIVERS mac80211
>> + local var=DRIVERS
>> + local value=mac80211
>> + local sep=
>> + eval export -n -- "DRIVERS=${DRIVERS:+${DRIVERS}${value:+$sep}}$value"
>> + export -n -- DRIVERS=mac80211
>> + scan_wifi
>> + local cfgfile=
>> + DEVICES=
>> + config_load wireless
>> + [ -n  ]
>> + uci_load wireless
>> + local PACKAGE=wireless
>> + local DATA
>> + local RET
>> + local VAR
>> + _C=0
>> + [ -z  ]
>> + export -n CONFIG_LIST_STATE=
>> + export -n CONFIG_SECTIONS=
>> + export -n CONFIG_NUM_SECTIONS=0
>> + export -n CONFIG_SECTION=
>> + /sbin/uci -P /var/state -S -n export wireless
>> + DATA=
>> + RET=1
>> + [ 1 != 0 -o -z  ]
>> + unset DATA
>> +
>> + return 1
>>
>>>
>>> - Felix
>>>
>>
>> ___
>> Lede-dev mailing list
>> Lede-dev@lists.infradead.org
>> http://lists.infradead.org/mailman/listinfo/lede-dev

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[Yousong Zhou]

On 3 November 2016 at 22:20, xinglp  wrote:
> 2016-11-03 18:34 GMT+08:00 Felix Fietkau :
>> On 2016-11-03 04:18, xinglp wrote:
>>> I have tried to use several phone and app to play video via samba, I
>>> got broken picture or not fluent after this commit.
>>> Hardware is ar9331.
>> Should be fixed in the latest version, please test.
>
> 'wifi detect' output nothing with latest version, but
> /sys/class/net/wlan0 exist.
>

wifi detect has just been changed into a nop with
5f8f8a366136a07df661e31decce2458357c167a   The new method is "wifi
config"

yousong

> sh -x /sbin/wifi detect
> + . /lib/functions.sh
> + N=
>
> + _C=0
> + NO_EXPORT=1
> + LOAD_STATE=1
> + LIST_SEP=
> + reset_cb
> + [ -z  -a -f /lib/config/uci.sh ]
> + . /lib/config/uci.sh
> + CONFIG_APPEND=
> + . /usr/share/libubox/jshn.sh
> + DEVICES=
> + DRIVERS=
> + include /lib/wifi
> + local file
> + ls /lib/wifi/mac80211.sh
> + . /lib/wifi/mac80211.sh
> + append DRIVERS mac80211
> + local var=DRIVERS
> + local value=mac80211
> + local sep=
> + eval export -n -- "DRIVERS=${DRIVERS:+${DRIVERS}${value:+$sep}}$value"
> + export -n -- DRIVERS=mac80211
> + scan_wifi
> + local cfgfile=
> + DEVICES=
> + config_load wireless
> + [ -n  ]
> + uci_load wireless
> + local PACKAGE=wireless
> + local DATA
> + local RET
> + local VAR
> + _C=0
> + [ -z  ]
> + export -n CONFIG_LIST_STATE=
> + export -n CONFIG_SECTIONS=
> + export -n CONFIG_NUM_SECTIONS=0
> + export -n CONFIG_SECTION=
> + /sbin/uci -P /var/state -S -n export wireless
> + DATA=
> + RET=1
> + [ 1 != 0 -o -z  ]
> + unset DATA
> +
> + return 1
>
>>
>> - Felix
>>
>
> ___
> Lede-dev mailing list
> Lede-dev@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/lede-dev

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[Arjen de Korte]

Citeren xinglp :


2016-11-03 18:34 GMT+08:00 Felix Fietkau :

On 2016-11-03 04:18, xinglp wrote:

I have tried to use several phone and app to play video via samba, I
got broken picture or not fluent after this commit.
Hardware is ar9331.

Should be fixed in the latest version, please test.


'wifi detect' output nothing with latest version, but
/sys/class/net/wlan0 exist.


This is expected. See  
https://git.lede-project.org/08de183485c8412ceb77e0260d131d0b9af580ac



sh -x /sbin/wifi detect
+ . /lib/functions.sh
+ N=

+ _C=0
+ NO_EXPORT=1
+ LOAD_STATE=1
+ LIST_SEP=
+ reset_cb
+ [ -z  -a -f /lib/config/uci.sh ]
+ . /lib/config/uci.sh
+ CONFIG_APPEND=
+ . /usr/share/libubox/jshn.sh
+ DEVICES=
+ DRIVERS=
+ include /lib/wifi
+ local file
+ ls /lib/wifi/mac80211.sh
+ . /lib/wifi/mac80211.sh
+ append DRIVERS mac80211
+ local var=DRIVERS
+ local value=mac80211
+ local sep=
+ eval export -n -- "DRIVERS=${DRIVERS:+${DRIVERS}${value:+$sep}}$value"
+ export -n -- DRIVERS=mac80211
+ scan_wifi
+ local cfgfile=
+ DEVICES=
+ config_load wireless
+ [ -n  ]
+ uci_load wireless
+ local PACKAGE=wireless
+ local DATA
+ local RET
+ local VAR
+ _C=0
+ [ -z  ]
+ export -n CONFIG_LIST_STATE=
+ export -n CONFIG_SECTIONS=
+ export -n CONFIG_NUM_SECTIONS=0
+ export -n CONFIG_SECTION=
+ /sbin/uci -P /var/state -S -n export wireless
+ DATA=
+ RET=1
+ [ 1 != 0 -o -z  ]
+ unset DATA
+
+ return 1



- Felix



___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev





___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[xinglp]

2016-11-03 18:34 GMT+08:00 Felix Fietkau :
> On 2016-11-03 04:18, xinglp wrote:
>> I have tried to use several phone and app to play video via samba, I
>> got broken picture or not fluent after this commit.
>> Hardware is ar9331.
> Should be fixed in the latest version, please test.

'wifi detect' output nothing with latest version, but
/sys/class/net/wlan0 exist.

sh -x /sbin/wifi detect
+ . /lib/functions.sh
+ N=

+ _C=0
+ NO_EXPORT=1
+ LOAD_STATE=1
+ LIST_SEP=
+ reset_cb
+ [ -z  -a -f /lib/config/uci.sh ]
+ . /lib/config/uci.sh
+ CONFIG_APPEND=
+ . /usr/share/libubox/jshn.sh
+ DEVICES=
+ DRIVERS=
+ include /lib/wifi
+ local file
+ ls /lib/wifi/mac80211.sh
+ . /lib/wifi/mac80211.sh
+ append DRIVERS mac80211
+ local var=DRIVERS
+ local value=mac80211
+ local sep=
+ eval export -n -- "DRIVERS=${DRIVERS:+${DRIVERS}${value:+$sep}}$value"
+ export -n -- DRIVERS=mac80211
+ scan_wifi
+ local cfgfile=
+ DEVICES=
+ config_load wireless
+ [ -n  ]
+ uci_load wireless
+ local PACKAGE=wireless
+ local DATA
+ local RET
+ local VAR
+ _C=0
+ [ -z  ]
+ export -n CONFIG_LIST_STATE=
+ export -n CONFIG_SECTIONS=
+ export -n CONFIG_NUM_SECTIONS=0
+ export -n CONFIG_SECTION=
+ /sbin/uci -P /var/state -S -n export wireless
+ DATA=
+ RET=1
+ [ 1 != 0 -o -z  ]
+ unset DATA
+
+ return 1

>
> - Felix
>

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] Smaller /overlay with latest snapshot on WDR-3600

[Damien Wyart]

Hi,

Thanks for your reply, I had not thought about the larger packages
went I sent my email, and I think this is the correct explanation.

I will either reduce my list or build a custom image.

Thanks

Damien


On Thu, Nov 3, 2016 at 10:25 AM, Stijn Segers
 wrote:
> Hi,
>
> 4,5 MiB sounds about right for a device with 8 MiB flash. If you do a
> factory reset, does the flash layout look identical to an older LEDE
> installation? It should, afaik, I haven't seen any changes in that sense in
> git lately. Most probably one of the packages you install afterwards grew in
> size (or maybe pulled in new dependencies).
>
> You might be better off using the image generator to include the packages
> you want in your image, prior to flashing it. That's more efficient
> space-wise.
>
> Hope this helps.
>
> Stijn

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH netifd] device: extend interface.macaddr option to refer to another interface

[Günther Kelleter]

e.g. option macaddr "@eth0" tells netifd to use the MAC of interface eth0.
Use case: set MAC address of a bridge interface to a specific interface's
MAC regardless of bridge/interface initialization order.

Signed-off-by: Günther Kelleter 
---
 device.c | 16 +++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/device.c b/device.c
index 82596e4..c63ddc7 100644
--- a/device.c
+++ b/device.c
@@ -256,7 +256,21 @@ device_init_settings(struct device *dev, struct blob_attr 
**tb)
}
 
if ((cur = tb[DEV_ATTR_MACADDR])) {
-   ea = ether_aton(blobmsg_data(cur));
+   const char *data = blobmsg_data(cur);
+   struct device_settings ds = {0};
+   if (data[0] == '@') {
+   struct device dev = {0};
+   strncpy(dev.ifname, data + 1, sizeof dev.ifname);
+   dev.ifname[sizeof dev.ifname - 1] = '\0';
+   system_if_get_settings(&dev, &ds);
+   if (ds.flags & DEV_OPT_MACADDR) {
+   ea = (struct ether_addr *)ds.macaddr;
+   } else {
+   ea = NULL;
+   }
+   } else {
+   ea = ether_aton(blobmsg_data(cur));
+   }
if (ea) {
memcpy(s->macaddr, ea, 6);
s->flags |= DEV_OPT_MACADDR;
-- 
2.10.0.100.gddd73ec


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [PATCH netifd] system-linux: check for open failure

[Günther Kelleter]

Signed-off-by: Günther Kelleter 
---
 system-linux.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/system-linux.c b/system-linux.c
index 6e4a194..a038987 100644
--- a/system-linux.c
+++ b/system-linux.c
@@ -2489,6 +2489,8 @@ int system_update_ipv6_mtu(struct device *dev, int mtu)
dev->ifname);
 
int fd = open(buf, O_RDWR);
+   if (fd < 0)
+   return ret;
 
if (!mtu) {
ssize_t len = read(fd, buf, sizeof(buf) - 1);
-- 
2.10.0.100.gddd73ec


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[xinglp]

2016-11-03 18:34 GMT+08:00 Felix Fietkau :
> On 2016-11-03 04:18, xinglp wrote:
>> I have tried to use several phone and app to play video via samba, I
>> got broken picture or not fluent after this commit.
>> Hardware is ar9331.
> Should be fixed in the latest version, please test.
Okay, I'm compiling it now.
Another question:
How to make login banner show something like "r1234"  other than
"Reboot (HEAD, unknown)"
>
> - Felix
>

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[Felix Fietkau]

On 2016-11-03 04:18, xinglp wrote:
> I have tried to use several phone and app to play video via samba, I
> got broken picture or not fluent after this commit.
> Hardware is ar9331.
Should be fixed in the latest version, please test.

- Felix


___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[Tino Reichardt]

* John Crispin  wrote:
> Hi Tino,
> 
> supporting ASE is possible, however once you have the wifi driver and
> dsl stuff installed there is not much space left for even a webui on the
> system. there was once support for dgn1000 but it got dropped as it was
> never really useful. if you want to add support for the allnet router,
> then best look at the commit history and find the commit that removed
> ase support and use that as a base to create a new patch for adding ASE
> back.

I will search in the git history for this.

Also, the Allnet Modem has no wifi @all ;)

1x LAN, 1x WAN, 1x serial (intern), 1x reset button

Here is a picture:
http://www.allnet-shop.de/out/pictures/master/product/3/all0333cj_3.jpg


-- 
Best regards, TR

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[Alberto Bursi]

On 11/03/2016 10:13 AM, Tino Reichardt wrote:
> Most of these Modems have only a very small flash of 2MB or 4MB .. so
> there is no need for lot of stuff. My aim is to have nice DSL modems
> with recent kernel and dsl firmware. pppoe is not really needed, but
> will fit also into that flash, I think.
>

Default LEDE image (router firmware with switch/router/firewall/wifi) 
fits in 4MiB of flash (also with luci webinterface, if it is integrated 
at firmware generation time), there are plenty of other supported 
devices with 4MiB of flash.

2MiB is very low and might not be enough even if you cut down things, 
but for a modem you don't really need much.

> Where should I add my pictures and descriptions of that modem. Should I
> make an extra project page onto my homepage, or is the lede wiki okay
> for that? When okay, I need some sample device page of current lede
> wiki.
>

For now the lede wiki isn't ready yet for devices, you can open a thread 
in the LEDE forum in the development section, and the info will be moved 
to the wiki when it is ready.
https://forum.lede-project.org/

>
> Is it important, to have an own working uboot, or should I create some
> kernel/rootfs image first ? What is the correct way? When I brick the
> current vendor uboot, is there a way, to restore it via serial
> interface?
>

Most LEDE devices don't touch the vendor u-boot (or bootloaders) unless 
they really have to, the LEDE firmware you make should mimic the stock 
firmware's place on flash so the bootloader will boot it without 
complaining.

The reason is that if you do a mistake and corrupt/damage the uboot, the 
device will not boot anymore and recovery will require complex 
procedures with (expensive) JTAG tools or desoldering the flash chip and 
using another device to re-program it.

-Alberto

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[John Crispin]

Hi Tino,

supporting ASE is possible, however once you have the wifi driver and
dsl stuff installed there is not much space left for even a webui on the
system. there was once support for dgn1000 but it got dropped as it was
never really useful. if you want to add support for the allnet router,
then best look at the commit history and find the commit that removed
ase support and use that as a base to create a new patch for adding ASE
back.

John

On 03/11/2016 10:13, Tino Reichardt wrote:
> Hello,
> 
> I have about 10-15 Allnet DSL Modems in use and just realized, that
> they have an Linux SoC, which can be updated to more recent kernels...
> 
> Name: ALLNET ALL0333CJ Rev. C
> Short Product Info (german) [1]
> 
> This is what I have done so far:
> - made some pictures and checked the chips
> - bought an serial to usb adapter and connected sucessfully to uboot
>   and / or the system itself
> - I have also telnet root access and a lot infos via webinterface
> - download a lot code from vendors, which also use this SoC
> - created a github repo for Amazon SE DSL Firmware [2]
> 
> Hardware Infos:
> SoC, Amazon-SE PSB 50601: [3] [4]
> - is supported in upstream Linux as "Amazon SE"
> - they have three groups: 1) Amazon SE 2) XWAY 3) FALCON
> - LEDE Kernel has these Lantiq groups:
>   1) XRX200 2) XWAY 3) XWAY Legacy
>   - for what does xway / xway legacy stand for?
>   - the menuconfig help is not very descriptive there
>   - can I add "Amazon SE" as another group, or should it belong to some
> existing group?
> 
> Flash: 4MB (S29GL032N90TF104)
> - not supported by current lede uboot (also no patches)
> - I have to add this via patch for it (found in vendor code)
> 
> Downloaded Sourcecode for reference:
> - netgear gpl code of DGN1000B [4] [5]
> - the code of samsung modems [6]
> - the allnet code [7]
> 
> Most of these Modems have only a very small flash of 2MB or 4MB .. so
> there is no need for lot of stuff. My aim is to have nice DSL modems
> with recent kernel and dsl firmware. pppoe is not really needed, but
> will fit also into that flash, I think.
> 
> Where should I add my pictures and descriptions of that modem. Should I
> make an extra project page onto my homepage, or is the lede wiki okay
> for that? When okay, I need some sample device page of current lede
> wiki.
> 
> The openwrt wiki is not accessable today via chromium, cause the
> letsencrypt cert was valid until yesterday :/
> 
> Is it important, to have an own working uboot, or should I create some
> kernel/rootfs image first ? What is the correct way? When I brick the
> current vendor uboot, is there a way, to restore it via serial
> interface?
> 
> So, thats enough, for an first message to the list ;)
> Please comment some of my question, so the support for Lantiq Amazon-SE
> boards is possible.
> 
> 
> Links:
> [1] http://www.allnet.de/fileadmin/transfer/products/98791.pdf
> [2] https://github.com/mcmilk/Amazon-SE-Firmware
> [3] https://wikidevi.com/wiki/Lantiq#AMAZON-S.2FSE.2FME
> [3] http://ark.intel.com/products/92478/Intel-XWAY-AMAZONSE-Network-Processors
> [4] https://sourceforge.net/projects/officialnetgearfirmware/files/
> [5] 
> http://kb.netgear.de/app/answers/detail/a_id/2649/~/netgear-open-source-code-for-programmers-(gpl)
> [6] 
> http://www.samsung.com/global/business/telecommunication-systems/resource/opensource/iad.html
> [7] 
> ftp://212.18.29.48/ftp/pub/allnet/dsl-modems/all0333cj/sourcecode_ALL0333CJ.zip
> 

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[Alberto Bursi]

On 11/03/2016 10:39 AM, J Mo wrote:
>
>
> On 11/03/2016 02:13 AM, Tino Reichardt wrote:
>> Where should I add my pictures and descriptions of that modem. Should I
>> make an extra project page onto my homepage, or is the lede wiki okay
>> for that? When okay, I need some sample device page of current lede
>> wiki.
>
> This is what a forum would be great for,

Someone here has missed the announcement on the mailing list, twice, 
also the forum link has been added to the wiki too.

LEDE forum is here https://forum.lede-project.org/

-Alberto Bursi.

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[xinglp]

2016-11-03 16:48 GMT+08:00 Stijn Segers :
> There is a bug open for this:
>
> https://bugs.lede-project.org/index.php?do=details&task_id=241&order=id&sort=desc
>
> Stijn
Thanks, and I know I'm not alone.

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[J Mo]

On 11/03/2016 02:13 AM, Tino Reichardt wrote:

Where should I add my pictures and descriptions of that modem. Should I
make an extra project page onto my homepage, or is the lede wiki okay
for that? When okay, I need some sample device page of current lede
wiki.


This is what a forum would be great for, but since LEDE's actions 
indicate it doesn't seem to care about having a userbase...


Go use the OpenWRT forum and start a new thread. That's what I did 
recently when I started working on a new router. He's my tread as an 
example:


https://forum.openwrt.org/viewtopic.php?id=65956

Post pictures (imgur or whatever), document you progress as you go, and 
even if you give up, your work is there for others to continue on with 
if they want to do so.





Is it important, to have an own working uboot, or should I create some
kernel/rootfs image first ? What is the correct way? When I brick the
current vendor uboot, is there a way, to restore it via serial
interface?


Most OpenWRT/LEDE supported routers leave the OEM u-boot/bootloader in 
place and work around it where needed. Replacing it isn't impossible, 
but it's not easy.


You probably can't fix a broken u-boot without using a JTAG or some 
other means to write directly to the flash storage device. That's why 
most people don't mess with changing the u-boot.




This all being said, the first thing you should probably do is find out 
of those ADSL modems are supported under Linux already. If they are not, 
and you don't have the skills to write a new driver for it, just turn 
around, go home, and forget all about this because they will probably 
never get supported.


Good luck!



___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] Smaller /overlay with latest snapshot on WDR-3600

[Stijn Segers]

Hi,

4,5 MiB sounds about right for a device with 8 MiB flash. If you do a 
factory reset, does the flash layout look identical to an older LEDE 
installation? It should, afaik, I haven't seen any changes in that sense 
in git lately. Most probably one of the packages you install afterwards 
grew in size (or maybe pulled in new dependencies).


You might be better off using the image generator to include the 
packages you want in your image, prior to flashing it. That's more 
efficient space-wise.


Hope this helps.

Stijn

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

[LEDE-DEV] [RFC] ADSL Modem Allnet All0333CJ Rev.C / Amazon-SE SoC

[Tino Reichardt]

Hello,

I have about 10-15 Allnet DSL Modems in use and just realized, that
they have an Linux SoC, which can be updated to more recent kernels...

Name: ALLNET ALL0333CJ Rev. C
Short Product Info (german) [1]

This is what I have done so far:
- made some pictures and checked the chips
- bought an serial to usb adapter and connected sucessfully to uboot
  and / or the system itself
- I have also telnet root access and a lot infos via webinterface
- download a lot code from vendors, which also use this SoC
- created a github repo for Amazon SE DSL Firmware [2]

Hardware Infos:
SoC, Amazon-SE PSB 50601: [3] [4]
- is supported in upstream Linux as "Amazon SE"
- they have three groups: 1) Amazon SE 2) XWAY 3) FALCON
- LEDE Kernel has these Lantiq groups:
  1) XRX200 2) XWAY 3) XWAY Legacy
  - for what does xway / xway legacy stand for?
  - the menuconfig help is not very descriptive there
  - can I add "Amazon SE" as another group, or should it belong to some
existing group?

Flash: 4MB (S29GL032N90TF104)
- not supported by current lede uboot (also no patches)
- I have to add this via patch for it (found in vendor code)

Downloaded Sourcecode for reference:
- netgear gpl code of DGN1000B [4] [5]
- the code of samsung modems [6]
- the allnet code [7]

Most of these Modems have only a very small flash of 2MB or 4MB .. so
there is no need for lot of stuff. My aim is to have nice DSL modems
with recent kernel and dsl firmware. pppoe is not really needed, but
will fit also into that flash, I think.

Where should I add my pictures and descriptions of that modem. Should I
make an extra project page onto my homepage, or is the lede wiki okay
for that? When okay, I need some sample device page of current lede
wiki.

The openwrt wiki is not accessable today via chromium, cause the
letsencrypt cert was valid until yesterday :/

Is it important, to have an own working uboot, or should I create some
kernel/rootfs image first ? What is the correct way? When I brick the
current vendor uboot, is there a way, to restore it via serial
interface?

So, thats enough, for an first message to the list ;)
Please comment some of my question, so the support for Lantiq Amazon-SE
boards is possible.


Links:
[1] http://www.allnet.de/fileadmin/transfer/products/98791.pdf
[2] https://github.com/mcmilk/Amazon-SE-Firmware
[3] https://wikidevi.com/wiki/Lantiq#AMAZON-S.2FSE.2FME
[3] http://ark.intel.com/products/92478/Intel-XWAY-AMAZONSE-Network-Processors
[4] https://sourceforge.net/projects/officialnetgearfirmware/files/
[5] 
http://kb.netgear.de/app/answers/detail/a_id/2649/~/netgear-open-source-code-for-programmers-(gpl)
[6] 
http://www.samsung.com/global/business/telecommunication-systems/resource/opensource/iad.html
[7] 
ftp://212.18.29.48/ftp/pub/allnet/dsl-modems/all0333cj/sourcecode_ALL0333CJ.zip

-- 
Best regards, TR

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev

Re: [LEDE-DEV] wireless became unstable after "mac80211: update to wireless-testing 2016-10-08" commit ad51e09fd1301484820a466a49447a34d7504882

[Stijn Segers]

There is a bug open for this:

https://bugs.lede-project.org/index.php?do=details&task_id=241&order=id&sort=desc

Stijn

___
Lede-dev mailing list
Lede-dev@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/lede-dev