summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* ar71xx: add AR933x specific frequency initialization codeGabor Juhos2011-05-312-0/+76
| | | | SVN-Revision: 27056
* ar71xx: add AR933x specific early_printk supportGabor Juhos2011-05-312-5/+62
| | | | SVN-Revision: 27055
* ar71xx: add preliminary support for the Atheros AR933x SoCsGabor Juhos2011-05-315-4/+22
| | | | SVN-Revision: 27054
* ar71xx: stop producing -NA factory images and tag factory images with hd_id ↵Gabor Juhos2011-05-311-20/+19
| | | | | | | | | | | | | | | | | | | | | | for WNDR3700v2 NETGEAR doesn’t produce a distinct North American image for WNDR3700v2, they use the same image worldwide. This is a change from earlier models such as WNDR3700 (v1). NETGEAR's v2 images now contain an "hd_id" parameter, as well. All observed WNDR3700v2, WNDR3800, and WNDRMAC images use 29763654+16+64 as their hd_id value. This patch changes the OpenWrt WNDR3700v2 "factory" image generation to stop producing the extraneous -NA version and to tag the image with the same hd_id used in NETGEAR's images. There is no change to WNDR3700 (v1) image generation, as NETGEAR continues to produce distinct -NA and worldwide images, neither of which are tagged with hd_id. Signed-off-by: Mark Mentovai <mark@moxienet.com> SVN-Revision: 27053
* tools/firmware-utils: add support for hd_id tag into mkdniimgGabor Juhos2011-05-311-4/+13
| | | | | | Signed-off-by: Mark Mentovai <mark@moxienet.com> SVN-Revision: 27052
* ar71xx: create factory images for the WZR-HP-G30{0,1}NH boardsGabor Juhos2011-05-311-2/+9
| | | | SVN-Revision: 27051
* tools/firmware-utils: add buffalo specific fw toolsGabor Juhos2011-05-315-0/+1177
| | | | SVN-Revision: 27050
* ar71xx: Fix header offset for newer WRT160NL modelsGabor Juhos2011-05-311-9/+16
| | | | | | | | | | | | Newer WRT160NLs have a flash chip with 4K erase blocks instead of 64K, resulting in miscalculated partition sizes. Since the actual sizes did not change, hardcode them to their current sizes, and make sure they are at least one erase block big (in case Cisco decides to start to use chips with 128K erase blocks). Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com> SVN-Revision: 27049
* ar71xx: add SOC_AR{71XX,724X,913X,934X} config symbolsGabor Juhos2011-05-313-1/+63
| | | | SVN-Revision: 27048
* ar71xx: reduce the size of the squashfs-4k binariesGabor Juhos2011-05-311-2/+1
| | | | SVN-Revision: 27047
* mac80211: make ath9k available even if PCI support is disabledGabor Juhos2011-05-311-5/+6
| | | | SVN-Revision: 27046
* ar71xx: initialize is_clk_25mhz field for AR9340 WMACGabor Juhos2011-05-311-0/+4
| | | | SVN-Revision: 27045
* ar71xx: sync ath9k_platform_data.h with compat-wirelessGabor Juhos2011-05-311-0/+2
| | | | SVN-Revision: 27044
* ar71xx: use the diag LED for diagnostic on the WZR-HP-G30XNH boardsGabor Juhos2011-05-311-1/+1
| | | | SVN-Revision: 27043
* ar71xx: fix uci-defaults script for the WZR-HP-G30XNH boardsGabor Juhos2011-05-311-2/+2
| | | | SVN-Revision: 27042
* ar71xx: ag71xx: make ring sizes configurable via ethtoolGabor Juhos2011-05-312-0/+56
| | | | SVN-Revision: 27041
* ar71xx: ag71xx: prepare to make ring sizes configurableGabor Juhos2011-05-313-35/+32
| | | | SVN-Revision: 27040
* ar71xx: ag71xx: nuke unused AG71XX_TX_FIFO_LEN defineGabor Juhos2011-05-311-1/+0
| | | | | | Reported-by: Dave Täht <dave.taht@gmail.com> SVN-Revision: 27039
* x86: add additional companion chip drivers to alix2 and geosFelix Fietkau2011-05-312-5/+9
| | | | | | | | | | | | Add the MFGPT high precision timers, and integration into the schedule (clock events). Use OHCI, not UHCI usb module. Remove VIA SATA driver from Alix2... that was accidentally cloned from net5501 platform. Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com> SVN-Revision: 27038
* x86: remove pata-amd driver from generic x86 images, add addition cs553x modulesFelix Fietkau2011-05-316-2/+103
| | | | | | | | | | The pata-cs5535/pata-cs5536 driver is actually fairly rare: it only gets used on embedded platforms. Therefore, we remove it from the generic x86 configuration. Also, add definitions for other useful geode modules (high-res timers, MFGPT timers, etc). Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com> SVN-Revision: 27037
* purge out CONFIG_CRYPTO* symbols out of target specific kernel configs since ↵Mirko Vogt2011-05-316-10/+0
| | | | | | OpenWrt is taking care of them SVN-Revision: 27036
* ar71xx: only enable the rx engine after the link is up, fixes a race ↵Felix Fietkau2011-05-301-67/+68
| | | | | | condition that got rx stuck when the interface is brought up during lots of inbound traffic (thx, matteo) SVN-Revision: 27035
* ar71xx: disable flow control for ar724x, it can get stuck in a loop of ↵Felix Fietkau2011-05-301-5/+1
| | | | | | continously sending MAC pause frames SVN-Revision: 27034
* remove the now unused Target-Kernel field from the target metadataFelix Fietkau2011-05-302-2/+0
| | | | SVN-Revision: 27033
* remove kernel major version specific kernelpackage overridesFelix Fietkau2011-05-301-2/+0
| | | | SVN-Revision: 27032
* upgrade to 2.6.39Imre Kaloz2011-05-3015-95/+27126
| | | | SVN-Revision: 27030
* ath9k: reduce the number of rx buffers, 512 is excessively high, 128 should ↵Felix Fietkau2011-05-301-0/+11
| | | | | | work just as well, and will use a lot less memory SVN-Revision: 27029
* * backport 2.6.8 patches to .39 / .32.33John Crispin2011-05-29165-18458/+54116
| | | | | | | | | * remove lqtapi * bump tapi/dsl to .39 * migrate to new ltq_ style api * add amazon_se support SVN-Revision: 27026
* ath9k: merge a few bugfixesFelix Fietkau2011-05-283-4/+97
| | | | SVN-Revision: 27025
* ath5k: disable fast channel changes, some people have reported that they ↵Felix Fietkau2011-05-281-0/+18
| | | | | | introduce connection stability issues SVN-Revision: 27024
* remove legacy package/foo-compile wrapper targets, people should be used to ↵Felix Fietkau2011-05-281-2/+0
| | | | | | the correct target names by now SVN-Revision: 27023
* hostapd: fix a race condition related to assoc response tx status processingFelix Fietkau2011-05-281-0/+48
| | | | SVN-Revision: 27022
* hostapd: update to 20110527, includes a client mode wpa auth fixFelix Fietkau2011-05-2813-53/+53
| | | | SVN-Revision: 27021
* bcm63xx: Add support for 2.6.39Florian Fainelli2011-05-2738-0/+4717
| | | | | | | | Add support for Kernel 2.6.39. Boots fine on a BCM6358. Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com> SVN-Revision: 27019
* bcm63xx: Add support for 2.6.38Florian Fainelli2011-05-2738-0/+4725
| | | | | | | | Add support for Kernel 2.6.38, based on 2.6.38.6. Boots fine on a BCM6358. Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com> SVN-Revision: 27018
* fold 446-board-D4PW_correction.patch in 440-board-D4PW.patchFlorian Fainelli2011-05-267-98/+34
| | | | SVN-Revision: 27014
* renumber patches accordinglyFlorian Fainelli2011-05-2625-64/+60
| | | | SVN-Revision: 27013
* renumber board patchesFlorian Fainelli2011-05-2612-20/+20
| | | | SVN-Revision: 27012
* kernel: exportfs: reallow building as a moduleHauke Mehrtens2011-05-251-0/+11
| | | | | | | | | | | | | | | It was originally tristate, but was changed to bool to prevent a build breakage in an earlier patch revision (for a new syscall). The actual commit[1] changed it in a way this isn't needed anymore, but accidentially left the EXPORTFS change in. There's no reason for it not being buildable as a module, so changing it back to tristate is fine. see: https://patchwork.kernel.org/patch/812832/ Thank you Jonas Gorski for the patch. SVN-Revision: 27011
* kernel: generic: Update 2.6.39 to finalHauke Mehrtens2011-05-255-12/+43
| | | | | | | | | | | * Add missing config symbols * Update package/kernel for 2.6.39 CONFIG_BT_L2CAP=y and CONFIG_BT_SCO=y are set to build them into the bluetooth module. Thank you Jonas Gorski for the patch SVN-Revision: 27010
* kernel: some updates for kernel 2.6.39Hauke Mehrtens2011-05-2527-79/+215
| | | | | | | | | | * add md5sum for kernel * add some missing config symbols * add some ssb patches to make compat-wireless compile Thank you Peter Wagner for the patch SVN-Revision: 27009
* brcm47xx: initial 2.6.39 support for brcm47xxHauke Mehrtens2011-05-2537-0/+3451
| | | | | | Thank you Peter Wagner for the patch SVN-Revision: 27008
* fix mtd warningsFlorian Fainelli2011-05-255-10/+30
| | | | | | | | | | | | | | | Fixes for the following warnings: drivers/mtd/devices/block2mtd.c: In function '_open_bdev': drivers/mtd/devices/block2mtd.c:272:3: warning: ISO C90 forbids mixed declarations and code drivers/mtd/devices/block2mtd.c: In function 'add_device': drivers/mtd/devices/block2mtd.c:399:13: warning: assignment discards qualifiers from pointer target type In the case of 442-block2mtd_probe.patch, it's not until the insertion of a executable statement before the definition that causes a warning, so I've chosen to resolve the issue here by splitting the definition and the initialization into a definition and a statement (and assignment) instead. Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com> SVN-Revision: 27007
* brcm47xx: remove old kernelHauke Mehrtens2011-05-2537-3465/+0
| | | | SVN-Revision: 27006
* brcm47xx: fix partition layout on SimpleShare devices.Hauke Mehrtens2011-05-251-21/+66
| | | | | | | | | | The flash driver code should be cleaned up and the broad detection code should be placed into arch code and used here. This fixes #9323 Thank you Will Holmes for the patch. SVN-Revision: 27005
* use 'fputws_unlocked(S,F)' instead of 'fputws(S,F)'Mirko Vogt2011-05-241-0/+21
| | | | | | | this eliminates a source of reproduceable freezes http://lists.uclibc.org/pipermail/uclibc/2011-May/045367.html SVN-Revision: 27001
* libpcap: properly fix fPIC handling (reverts $(FPIC) added in r26999)Felix Fietkau2011-05-244-19/+42
| | | | SVN-Revision: 27000
* libpcap: add -fpic, refresh patchesJo-Philipp Wich2011-05-244-8/+9
| | | | SVN-Revision: 26999
* Clean up patch not applyingHamish Guthrie2011-05-241-1/+1
| | | | SVN-Revision: 26997
* libpcap: remove empty patchfilesJo-Philipp Wich2011-05-242-0/+0
| | | | SVN-Revision: 26996