aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* tar: fix reproducibility issueFelix Fietkau2016-12-191-0/+22
| | | | | | | Force root/root as names for uid0/gid0 instead of using the system names. This helps make packed download tarballs more reproducible Signed-off-by: Felix Fietkau <nbd@nbd.name>
* usbutils: fix download filenameFelix Fietkau2016-12-191-7/+6
| | | | | | | The name needs to include the revision, otherwise it is not automatically re-downloaded on updates. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* Revert "ath9k: Add airtime fairness scheduler"Felix Fietkau2016-12-196-104/+7
| | | | | | Accidentally pushed to the wrong branch This reverts commit 47bc081e76f6b04c0e9db687c8abb78eb437021f.
* ath9k: Add airtime fairness schedulerToke Høiland-Jørgensen2016-12-196-7/+104
| | | | | | | | | This adds a patch that introduces airtime fairness scheduling to ath9k, which can significantly improve network efficiency in mixed-rate environments. Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk> Signed-off-by: Felix Fietkau <nbd@nbd.name>
* mac80211: backport "cfg80211: limit scan results cache size"Christian Lamparter2016-12-191-0/+150
| | | | | | | | | | | | | | | | | | | | | | | The patch commit states: "It's possible to make scanning consume almost arbitrary amounts of memory, e.g. by sending beacon frames with random BSSIDs at high rates while somebody is scanning. Limit the number of BSS table entries we're willing to cache to 1000, limiting maximum memory usage to maybe 4-5MB, but lower in practice - that would be the case for having both full-sized beacon and probe response frames for each entry; this seems not possible in practice, so a limit of 1000 entries will likely be closer to 0.5 MB." Signed-off-by: Johannes Berg <johannes.berg@intel.com>" This patch was added in 4.4.36. But because LEDE backports cfg80211, mac80211 and the wifi drivers separately, it needs to be added manually for now. It can be dropped later as it will be part of the next mac80211 refresh. Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
* mac80211: merge a number of minstrel/minstrel_ht performance and memory ↵Felix Fietkau2016-12-199-0/+822
| | | | | | usage improvements Signed-off-by: Felix Fietkau <nbd@nbd.name>
* odhcpd: Bump to git HEAD version (various fixes)Hans Dedecker2016-12-171-3/+3
| | | | | | | | | e055530 Don't print non bound assignments in the state file 3af23ad config: Fix RA interface config being overwritten 41b5268 dhcpv6-ia : Fix static DHCPv6 assignments becoming non static be6c515 dhcpv6-ia: Fix assignment of static DHCPv6 leases 374dc3f cmake: Find libubox/uloop.h 01c919c odhcpd: Display infinite valid lifetime as -1
* scripts/update-package-md5sum: remove file, it is obsoleted by make check ↵Felix Fietkau2016-12-171-38/+0
| | | | | | FIXUP=1 Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: add CHECK_ALL variable to allow make download/check to include not ↵Felix Fietkau2016-12-172-0/+9
| | | | | | selected packages Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: add FIXUP option for make checkFelix Fietkau2016-12-172-1/+153
| | | | | | | | This will attempt to automatically fix common mistakes like using MD5 instead of SHA256, using the MD5SUM variable instead of HASH, or even a missing mirror file hash. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: implement make check and make package/X/checkFelix Fietkau2016-12-1711-22/+103
| | | | | | | | | | | | | | | | This is intended to be used for a wide array of package sanity checks. The first check that is implemented is for the hash of downloaded files. It checks: - Missing hash - Use of SHA256 instead of MD5 - dl/<file> hash not matching hash in makefile - deprecated MD5SUM variable The deprecated MD5SUM variable check is skipped for feeds/ until OpenWrt is updated as well Signed-off-by: Felix Fietkau <nbd@nbd.name>
* treewide: clean up download hashesFelix Fietkau2016-12-16233-283/+300
| | | | | | Replace *MD5SUM with *HASH, replace MD5 hashes with SHA256 Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath10k-ct: Update to latest CT ath10k driver.Ben Greear2016-12-161-2/+2
| | | | | | | | This works around deadlock and/or memory corruption during firmware crash and improves ability to configure number of tids in firmware. Signed-off-by: Ben Greear <greearb@candelatech.com>
* odhcp6c: Pass parameters to user dhcpv6 scriptHans Dedecker2016-12-162-3/+3
| | | | | | Pass all the parameters like device, dhcpv6 state to user script Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
* map: Have cmake find libubus.hFlorian Fainelli2016-12-161-0/+3
| | | | | | Update CMakeList.txt to look for libubus.h since we depend on it. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
* include/package.mk: sync default value for hash fallback with mirror hashFelix Fietkau2016-12-162-2/+1
| | | | | | Simplifies dealing with PKG_HASH vs PKG_MD5SUM Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: remove duplicate Download/default definition from include/host-build.mkFelix Fietkau2016-12-163-23/+13
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* download.pl: check for existing file before the first download attemptFelix Fietkau2016-12-161-2/+1
| | | | | | | | | | | | In the build system, flock will prevent multiple concurrent downloads for the same file. However, if one download request for the same file is waiting for another one to finish, it will result in downloading the same file twice consecutively. Prevent this issue by exiting immediately if the file has already been downloaded Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath10k-firmware: fix typoFelix Fietkau2016-12-161-1/+1
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* package/kernel/linux: only access kernel config if DUMP is unsetFelix Fietkau2016-12-161-2/+1
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath10k-firmware: untangle CT firmware filenames, fix conflictsFelix Fietkau2016-12-161-13/+17
| | | | | | | | | CT firmware files use the same filename with a different URL for different chips. Since all files end up in dl/, filenames need to be unique as well. Add a chip prefix to the output filename to fix this issue. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: replace MD5SUM variables with HASHFelix Fietkau2016-12-162-3/+7
| | | | | | | Since we've switched to preferring SHA256 over MD5, the old variable name is misleading. Packages using the old name remain compatible. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ar71xx: add support for TP-LINK TL-WR940N v4David Lutz2016-12-1614-1/+186
| | | | Signed-off-by: David Lutz <kpanic@hirnduenger.de>
* dnsmasq: fix service reloadJohn Crispin2016-12-161-2/+1
| | | | | | | | | | | The SIGHUP also got sent to the reload script making it bail out with an error Revert "dnsmasq: reload config if host name is modified" This reverts commit 854459a2f923376e0e509ebc0fb8ff90e9f13c02. Reported-by: Hans Dedecker <dedeckeh@gmail.com> Signed-off-by: John Crispin <john@phrozen.org>
* realview: drop the targetYousong Zhou2016-12-168-422/+0
| | | | Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
* armvirt: new targetYousong Zhou2016-12-166-0/+363
| | | | | | | | | | | | | | | | | | | | An ARM Cortex-A15 machine provided by QEMU. Kernel drivers enabled: - pl011, uart - pl031, rtc - pl061, gpio - pci-host-generic - virtio_{mmio,pci,net,blk,scsi,9p,console,balloon} - smp with NR_CPUS=4 - cpu-hotplug - ext4 - neon, vfp extensions support (otherwise userland will fail with illegal instruction signal (code 0x00000004)) - DEBUG_BUGVERBOSE for debug purposes Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
* mt76: update to the latest version, fixes a build error on some platformsFelix Fietkau2016-12-151-2/+2
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* x86: revert default root size back to 256 MBFelix Fietkau2016-12-151-2/+1
| | | | | | | | | | | 2 GB is overkill and was only added to allow unlimited ext4 resizing, which is a pretty rare use case. 256 MB allows resizing up to 256 GB, which should be good enough for almost all users. A lot of this is mostly irrelevant anyway, since you can just use squashfs + ext4 overlay. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* odhcpd: update sha256sumHauke Mehrtens2016-12-141-1/+1
| | | | | | | The sha256sum was not updated in the last commit. Fixes: a7c231027 [odhcpd: Fix dnsmasq re-reading hostfile] Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
* base-files: fix CONFIG_VERSION_DIST defaultJo-Philipp Wich2016-12-141-1/+1
| | | | | | Fix the default value of CONFIG_VERSION_DIST to use uppercase "LEDE". Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* build: use SNAPSHOT instead of CURRENT to designate untagged branch buildsJo-Philipp Wich2016-12-142-5/+5
| | | | | | | Calling a build ##.##-CURRENT might mislead users into thinking that this build is the most current release of a branch. Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* ar71xx: fix TL-WR842N v2 switch port orderJo-Philipp Wich2016-12-141-1/+5
| | | | | | | | | | | | | | The TL-WR842N v2 has the following switch port layout with swconfig port 0 being connected to eth1: Power LAN4 LAN3 LAN2 LAN1 WAN USB (*) | [...] | [...] | [...] | [...] | [...] | [.] swconfig port: 1 4 3 2 n/a netdev: eth1 eth1 eth1 eth1 eth0 Adjust the board defaults to match this model specific layout. Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* ar71xx: Added missing support for Linksys E2100LGad Krumholz2016-12-146-0/+143
| | | | | | | | | | | | | It's based on the WRT160NL according to https://wiki.openwrt.org/toh/linksys/e2100l Based on research done here: https://forum.openwrt.org/viewtopic.php?id=24244 and here: https://forum.openwrt.org/viewtopic.php?pid=120791#p120791 this patch was conceived. Signed-off-by: Gad Krumholz <gad.krumholz@gmail.com> [Jo-Philipp Wich: fix ordering, move addpattern change into separate commit] Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* firmware-utils: add E2100L support to addpattern.cJo-Philipp Wich2016-12-141-0/+7
| | | | | | Based on OpenWrt PR#250 Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* linux: do not autoload sdhci.ko as sdhci-pltfm.ko already depends on itJo-Philipp Wich2016-12-141-1/+1
| | | | | | | | | | | Currently, installing kmod-sdhci fails with "sdhci is already loaded" since "sdhci.ko" is inserted explicitely first, the implicitely loaded again when "sdhci-pltfm.ko" is inserted as the latter depends on the former. Remove the explicit autoprobe of "sdhci.ko" to fix the postinstall script. Acked-by: John Crispin <john@phrozen.org> Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* brcm63xx: image: trim revision code used for --rsa-signatureJo-Philipp Wich2016-12-141-3/+3
| | | | | | | | | | | | | | | | | The imagetag RSA signature field may not exceed 20 characters, so trim the Git hash portion from the REVISION variable in order to ensure that the resulting string fits within this field. The current revision values may be longer than 20 chars, especially when building within a dirty tree. In this case, the revision will look like "r2435+75-b4aa3c8" which, combined with the "LEDE-" prefix, is 21 chars long. Fixes the following error spotted by the buildbots: Error: RSA Signature (rsa_signature,r) too long. make[4]: *** [.../lede-brcm63xx-generic-NEUFBOX6-squashfs-cfe.bin] Error 1 Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* ipkg-make-index.sh: drop a few non-essential fieldsFelix Fietkau2016-12-141-1/+1
| | | | | | Saves some more space on the package index Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ipkg-make-index.sh: drop md5sum from package indexFelix Fietkau2016-12-141-3/+0
| | | | | | | We have switched opkg to sha256 a long time ago, and shrinking package lists is useful for systems that are running low on RAM Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: netfilter: split out iptable_raw into a separate packageFelix Fietkau2016-12-142-2/+22
| | | | | | | | This will avoid loading it in the default configuration, which reduces image size a bit, and (more importantly) improves performance by avoiding some unnecessary netfilter hooks Signed-off-by: Felix Fietkau <nbd@nbd.name>
* gcc: rip out transactional memory related bloat from crtbeginFelix Fietkau2016-12-142-0/+22
| | | | | | | Slightly improves compression for each executable, saving about 4k from the default ar71xx rootfs Signed-off-by: Felix Fietkau <nbd@nbd.name>
* swconfig: replace the shared library with a static oneFelix Fietkau2016-12-142-5/+5
| | | | | | Reduces binary size Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ppp: use --gc-sections to save a tiny bit of spaceFelix Fietkau2016-12-141-0/+3
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* libs/libpcap: update to 1.8.1p-wassi2016-12-148-202/+81
| | | | | | | | | | Update libpcap to upstream release 1.8.1 Change the name from libpcap.so.1.3 to libpcap.so.1 Remove parts of patch 201 which moved code among src files. Import patch 204 from Debian to update the USB path. Signed-off-by: Paul Wassi <p.wassi@gmx.at> Signed-off-by: Felix Fietkau <nbd@nbd.name> [fix parallel build bug]
* ath10k: fix a soft-lockup on firmware restartFelix Fietkau2016-12-142-2/+51
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath10k: fix a bug on sending null-func framesFelix Fietkau2016-12-141-0/+42
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath10k-firmware: removed broken submenuAlberto Bursi2016-12-141-1/+0
| | | | | | | this package references an undefined variable for its submenu. Remove this NOP variable assignment. Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
* kernel: add a missing submenuAlberto Bursi2016-12-141-0/+1
| | | | | | | this kernel module currently does not set submenu. Fix this by adding it to the "Others" submenu Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
* comgt-ncm: Add support for specifying profile indexMatti Laakso2016-12-144-56/+51
| | | | | | | | | | | | | | | Add support for specifying a call profile index instead of APN. A specific index different from 1 must be used for some service provider and modem combinations. In addition, change the manufacturer detection to use the standard AT+CGMI command, which produces more predictable output than ATI, remove the redundant ipv6 option, since it is less ambiguous to directly specify the PDP context type with mobile connections, and fix missing device during teardown when using ncm through the wwan proto. Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
* uqmi: Add support for specifying profile indexMatti Laakso2016-12-142-82/+93
| | | | | | | | | | | | | | Update uqmi to latest version, which brings about support for specifying a call profile index instead of APN. A specific index different from 1 must be used for some service provider and modem combinations. Also change option dhcp to dhcpv6, since IPv4 now always uses DHCP, replace option ipv6 with pdptype, which is less ambiguous, and make autoconnect optional and default it to off for IPv6 due to it not working with statically configured IPv6. Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
* dropbear: clean up default PATH handling in makefileDario Ernst2016-12-142-14/+2
| | | | | | | | Harmonise handling of DEFAULT_PATH by removing the patch introducing #ifndef guards around the path, and only using one means to set the path in the makefile. Signed-off-by: Dario Ernst <Dario.Ernst@riverbed.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-19 01:44:16 +0000