If that options is disabled in uClibc then net-snmp doesn't
detect sighold support on its configuration stage and so
defines HAVE_SIGHOLD as 0. That in its turn causes compilation of
the following branch of timerPause() in apps/snmpnetstat:
------------------>8----------------
int oldmask;
oldmask = sigblock(sigmask(SIGALRM));
if (!signalled) {
sigpause(0);
}
sigsetmask(oldmask);
------------------>8----------------
Now in uClibc all 3 sigblock(), sigmask() and sigsetmask()
were removed back in 2005, see:
https://git.busybox.net/uClibc/commit/?id=5aa7aa7fa7ec2a0fe567ac0b2595b46add6f3594
And all that causes net-snmp linkage to fail this way:
------------------>8----------------
.libs/if.o: In function `intpr':
if.c:(.text+0x908): undefined reference to `sigmask'
if.c:(.text+0x90e): undefined reference to `sigblock'
if.c:(.text+0x924): undefined reference to `sigsetmask'
------------------>8----------------
If we enable UCLIBC_HAS_OBSOLETE_BSD_SIGNAL in uClibc then
branch with sighold is used in timerPause() and everything builds as
expected.
scripts/getver.sh: Use 'git-rev-parse' to detect if tree lies in Git repository
Path to the Git repository directory can be overriden by using the '$GIT_DIR'
environment variable. This patch improves detection of Git repository by using
'git-rev-parse', which respects '$GIT_DIR' environment variable, instead of just
checking the existence of '.git' directory.
John Crispin [Sat, 9 Apr 2016 10:26:37 +0000 (10:26 +0000)]
ar71xx: fix nondeterministic hangs during bootconsole/console handover
Reconfiguring the UART when the FIFO is not empty may cause the boot to
hang. This hang is extremely suspectible to timing differences; recompiling
the kernel with the same configuration, but different UTS_VERSION
timestamps can yield images that hang more or less often. Sometimes images
are produced that hang reproducibly.
This patch should also make it into one of the next linux-stable releases,
but it's better to get this fixed as soon as possible.
John Crispin [Sat, 9 Apr 2016 10:25:42 +0000 (10:25 +0000)]
scripts/om-fwupgradecfg-gen.sh: Generate sha256sum for uboot verification
Future Open Mesh u-boot versions are changing the check of the image files
(vmlinux, rootfs) from md5 to sha256. Having both in them should be enough
to ensure backward and forward compatibility.
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
SVN-Revision: 49140
John Crispin [Sat, 9 Apr 2016 10:25:16 +0000 (10:25 +0000)]
arc770: enable unaligned access handling simulation in software
This enables misaligned access handling by software in Linux kernel.
With some wireless drivers (ath9k-htc and mt7601u for example) we see
misaligned accesses here and there and to cope with that without
fixing stuff in the drivers we're just gracefully handling it on ARC.
John Crispin [Sat, 9 Apr 2016 10:25:03 +0000 (10:25 +0000)]
ramips: Disable the SD driver by default on MT7621
The SD driver has been a long and dragging issue on the MT7621 platforms as it hasn't worked for
months resulting in unbootable devices. Until it's fixed, disable it in the default profiles.
John Crispin [Sat, 9 Apr 2016 10:24:58 +0000 (10:24 +0000)]
ramips: Fixes for LinkIt 7688
Misc fixes for LinkIt 7688 board:
- Copy the right wireless firmware for the mt7688
- Add back '0065-mt7688-fixes.patch', left out after the move to Linux 4.4.
- Remove SPI_DEV from linux config which otherwise causes a massive warning
- Add wmac to LINKIT7688.dts so wireless works
Signed-off-by: Adam Kent <adam@semicircular.net>
SVN-Revision: 49130
4.3 isn't oficially used anymore but it may be useful for debugging PCI
host driver regression. SPROM driver doesn't seem to work correctly with
4.3:
[ 2.336686] bcma: bus0: Using fallback SPROM failed (err -2)
[ 2.342366] bcma: bus0: No SPROM available
but it at least compiles now.
John Crispin [Fri, 1 Apr 2016 07:11:56 +0000 (07:11 +0000)]
ar71xx: avoid AR71XX_MODEL ending with a space on some TP-LINK-like devices
Instead of adding the space when combining $model and $hwver, add the space
to the beginning of $hwver, so the resulting string won't end with a space
when $hwver is set to the empty string.
John Crispin [Fri, 1 Apr 2016 07:11:48 +0000 (07:11 +0000)]
ramips: move different subtarget image generation to separate makefiles to improve maintainability
- all subtarget specific defines are in their own files
- common defines left in main Makefile
- each subtarget makefile idefed with SUBTARGET
- all subtargets compile tested
- few seems to be broken/unneeded things marked with FIXME
Signed-off-by: Roman Yeryomin <roman@advem.lv>
SVN-Revision: 49104
John Crispin [Fri, 1 Apr 2016 07:11:27 +0000 (07:11 +0000)]
ar71xx: WNR2200: fix for random WLAN MAC
Fix for invalid/random/duplicate WLAN MAC address in WNR2200.
Permanent platform MAC is calculated and assigned during system startup.
WLAN MAC follows wired Ethernet interface addresses.
Signed-off-by: Michal Cieslakiewicz <michal.cieslakiewicz@wp.pl>
SVN-Revision: 49100
John Crispin [Fri, 1 Apr 2016 07:11:23 +0000 (07:11 +0000)]
ar71xx: Add TL-WR841N/ND v11 support
This patch adds support for the TP-Link TL-WR841N/ND to trunk. It is
similar to the already supported v10 - see [1]. I have added support
based on that topic; in the meantime it has been confirmed working
(see the forum thread, and also [2]).
Signed-off by Stijn Segers <francesco.borromini@inventati.org>
Zoltan Herpai [Tue, 29 Mar 2016 11:42:14 +0000 (11:42 +0000)]
package: uboot-sunxi: various changes - bump to 2016.03 - add bugfixes related to 2016.03 update - sync DTS files with mainline - add support for non-standard uEnv.txt - add initial support for Theobroma A31-yQ7 devboard
John Crispin [Wed, 23 Mar 2016 12:52:47 +0000 (12:52 +0000)]
ar71xx/image: allow builds of loader without cmdline injecting
On most platforms the cmdline is supplied by the loader to the kernel.
When using an elf loader with cmdline the size is bloated by 512k.
For loader-elf platforms like the cpe510 better use loader-elf but patch
the cmdline into the kernel.
wdr1043: Tested loader still works using sysupgade image.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
SVN-Revision: 49080
John Crispin [Wed, 23 Mar 2016 12:52:20 +0000 (12:52 +0000)]
Revert "ar71xx: Clear bits in ath79_setup_qca955x_eth_cfg"
Not all mach-* files set all boards correctly in ETH_CFG. They depend on
some preset values by u-boot which were not previously modified by
ath79_setup_qca955x_eth_cfg. Avoiding to modify them in this function keeps
it backward compatible for these boards.
John Crispin [Wed, 23 Mar 2016 12:52:17 +0000 (12:52 +0000)]
Revert "ar71xx: Allow to set the RXDV, RXD, TXD, TXE delays for QCA955x"
The default delays RXD 3. RDV 3, TXD 0, TXE 0 doesn't seem to work for some
boards. These boards depend on the preset values of u-boot which may
differ.
John Crispin [Wed, 23 Mar 2016 12:52:12 +0000 (12:52 +0000)]
ar71xx: Use private version of ath79_setup_qca955x_eth_cfg for MR1750
The MR1750 must unset some bits in ETH_CFG which were set by u-boot to work
correctly under OpenWrt. But the global function
ath79_setup_qca955x_eth_cfg will not unset all of them to increase the
backward compatiblity with older mach-* files. A private (simplified)
version for MR1750 can be used instead.
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
SVN-Revision: 49070
John Crispin [Wed, 23 Mar 2016 12:52:09 +0000 (12:52 +0000)]
ar71xx: Use private version of ath79_setup_qca955x_eth_cfg for MR900
The MR900 must unset some bits in ETH_CFG which were set by u-boot to work
correctly under OpenWrt. But the global function
ath79_setup_qca955x_eth_cfg will not unset all of them to increase the
backward compatiblity with older mach-* files. A private (simplified)
version for MR900 can be used instead.
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
SVN-Revision: 49069
John Crispin [Wed, 23 Mar 2016 12:52:05 +0000 (12:52 +0000)]
oxnas: set ubifs parameters for Default profile
Work-around obsolete mkfs/ubifs code.
Unlike squashfs, ubifs images should be generated per-device so
minimum I/O unit size, logical erase block size, maximum logical erase
block count can be defined for the specific flash-chips parameters.
Also, generating ubinized images is unrelated to rootfs being ubifs or
squashfs, but rather depends on the device supporting UBI or not.
In the meantime and in order to fix an error which currently causes the
image generation on buildbot to fail, work-around by defining
UBIFS_OPTS for the Default profile.
See also
http://buildbot.openwrt.org:8010/builders/oxnas/builds/246/steps/compile_8/logs/stdio
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 49068
Gabor Juhos [Mon, 21 Mar 2016 13:15:29 +0000 (13:15 +0000)]
ramips: lzma-loader: fix O32 ABI conformance
According to the calling convention of the o32 ABI the
caller function must reserve stack space for $a0-$a3
registers in case the callee needs to save its arguments.
The assembly startup code does not reserve stack space
for these registers thus when the main C function needs
to save its arguments, that will cause a stack overflow.
Fix the assembly code to reserve stack space for the
registers to avoid that.
Untested. It seems that the lzma-loader is not used at all?
projects / thirdparty / openwrt.git / log
