All leds on these boards are green. v1 has RFKILL GPIO 23 for production
units (it had GPIO 13 only for test phase units, and these are rather
very rare to find). As for the previous attempt to fix this and revert
due to WDR boards have blue leds, it was wrong: WDR board does not use
common setup (false).
Petr Štetiar [Wed, 24 Jul 2019 14:00:39 +0000 (14:00 +0000)]
tfa-layerscape: fix fiptool host build
fiptool is a host tool, used in a firmware generation pipeline, but it's
not treated as such, leading to the build breakage on the hosts which
don't have {Open,Libre}SSL dev package installed:
In file included from fiptool.h:16:0,
from fiptool.c:19:
fiptool_platform.h:18:27: fatal error: openssl/sha.h:
No such file or directory
# include <openssl/sha.h>
So this patch promotes fiptool into the host tool with proper host
include and library paths under STAGING_DIR.
Ref: https://github.com/openwrt/openwrt/pull/2267 Signed-off-by: Petr Štetiar <ynezz@true.cz>
Petr Štetiar [Mon, 7 Jan 2019 00:57:01 +0000 (01:57 +0100)]
build: allow simple build system customization with local.mk
I tend to automate most of the boring and repetitive tasks like firmware
flashing, build config reconfiguration etc., so I always end up adding
`include local.mk` as a last line in my Makefile, where local.mk is
usually symlink to some other place, just to not accidentally delete it
during `git clean`.
Carrying this single uncommited modification along in the development
process is quite PITA, because it's causing problems during Git
workflow, while rebasing etc.
I hope, that I'm not alone using similar workflow, so I believe, that
this modification might be useful for others as well.
Change capitalization from spaces to tabs in order to match
the style of the file.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
[fix capitalization in commit title, add short description] Signed-off-by: David Bauer <mail@david-bauer.net>
1. Upgrade the factory.bin through the factory web interface or
the u-boot failsafe interface.
The firmware will boot up correctly for the first time.
Do not power off the device after OpenWrt has booted.
Otherwise the u-boot will enter failsafe mode as the checksum
of the firmware has been changed.
2. Upgrade the sysupgrade.bin in OpenWrt.
After upgrading completes the u-boot won't complain about the
firmware checksum and it's OK to use now.
3. If you powered off the device before upgrading the sysupgrade.bin,
just upgrade the factory.bin through the u-boot failsafe interface
and then goto step 2.
Signed-off-by: Jackson Lim <jackcolentern@gmail.com>
[Reword reset-hog comment, fix formatting] Signed-off-by: David Bauer <mail@david-bauer.net>
"
LibreSSL 2.9.1 now has a test that requires libtls.a, however, when building a
shared library only build, the --disable-static flag is passed to libressl,
which prevents the building of libtls.a.
With libtls.a not being built, the following error occurs:
libressl-2.9.1/tls/.libs/libtls.a', needed by 'handshake_table'. Stop.
There are three options to fix this:
1) Stick with autotools, and provide a patch that removes building anything in
the tests folder.
2) Pass --enable-static to LIBRESSL_CONF_OPTS
3) Change the package type to cmake, as a cmake build does not have this issue."
It appears we cannot change to cmake because cmake has a dependency on
an ssl library.
Take option 1 and do not build the tests.
Also take the opportunity to remove man page building as well.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
treewide: sysupgrade: get rid of platform_nand_pre_upgrade()
1) nand_do_upgrade() is always called by a target code
2) nand_do_upgrade() starts with calling platform_nand_pre_upgrade()
It means there is no need for the platform_nand_pre_upgrade() callback
at all. All code that was present there could bo moved & simplly called
by a target right before the nand_do_upgrade().
layerscape: sysupgrade: get rid of platform_pre_upgrade()
1) RAMFS_COPY_BIN and RAMFS_COPY_DATA can be defined at top of the file
like it's done for all other targets.
2) fw_printenv.lock can be created one step later in the
platform_do_upgrade(). It seems to be working well on many other
targets.
Petr Štetiar [Fri, 19 Jul 2019 21:04:23 +0000 (23:04 +0200)]
firmware-utils: mkfwimage: fix build failure on macOS with gcc 9
Fixes following errors:
src/mkfwimage.c:279:8: error: format specifies type 'long' but the argument has type 'off_t' (aka 'long long') [-Werror,-Wformat]
d->stats.st_size,
^~~~~~~~~~~~~~~~
src/mkfwimage.c:280:8: error: format specifies type 'long' but the argument has type 'long long' [-Werror,-Wformat]
d->partition_length - d->stats.st_size);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
src/mkfwimage.c:378:6: error: format specifies type 'unsigned long' but the argument has type 'long long' [-Werror,-Wformat]
d->stats.st_size - d->partition_length);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Reported-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk> Signed-off-by: Petr Štetiar <ynezz@true.cz>
Petr Štetiar [Fri, 19 Jul 2019 19:34:16 +0000 (21:34 +0200)]
firmware-utils: mkfwimage: fix more errors reported by gcc-6/7/9
src/mkfwimage.c:276:8: error: format '%lld' expects argument of type 'long long int', but argument 4 has type '__off_t' {aka 'const long int'} [-Werror=format=]
src/fw.h:71:36: error: format '%llu' expects argument of type 'long long unsigned int', but argument 6 has type '__off_t' {aka 'long int'} [-Werror=format=]
inlined from 'main' at src/mkfwimage.c:543:12:
/string_fortified.h:106:10: error: '__builtin_strncpy' output truncated before terminating nul copying 4 bytes from a string of the same length [-Werror=stringop-truncation]
inlined from 'write_part' at src/mkfwimage.c:235:2,
string_fortified.h:106:10: error: '__builtin_strncpy' specified bound 16 equals destination size [-Werror=stringop-truncation]
inlined from 'main' at src/mkfwimage.c:477:5:
string_fortified.h:106:10: error: '__builtin_strncpy' specified bound 256 equals destination size [-Werror=stringop-truncation]
inlined from 'main' at src/mkfwimage.c:496:5:
string_fortified.h:106:10: error: '__builtin_strncpy' specified bound 4096 equals destination size [-Werror=stringop-truncation]
inlined from 'main' at src/mkfwimage.c:481:5:
string_fortified.h:106:10: error: '__builtin_strncpy' specified bound 4096 equals destination size [-Werror=stringop-truncation]
inlined from 'main' at src/mkfwimage.c:485:5:
string_fortified.h:106:10: error: '__builtin_strncpy' specified bound 16 equals destination size [-Werror=stringop-truncation]
the ath10k-ct package ships multiple versions of the ath10k-ct driver,
OpenWrt currently only uses the version 4.19, but we still ship some
patches for older versions. Remove all patches only touching older
versions and also remove the patch for older versions from patches which
do the same changes to multiple versions of ath10k-ct.
This removes some unneeded patches, the end binary should stay the same.
This updates mac80211 to version 5.2-rc7, this contains all the changes
to the wireless subsystem up to Linux 5.2-rc7.
* The removed patches are applied upstream
* b43 now uses kmod-lib-cordic
* Update the nl80211.h file in iw to match backports version.
* Remove the two backports from kernel 4.9, they were needed for mt76,
but that can use the version from backports now, otherwise they
collide and cause compile errors.
INAGAKI Hiroshi [Sun, 16 Jun 2019 16:23:32 +0000 (01:23 +0900)]
ath79: modify device name of I-O DATA WN-AC1600DGR2
The hardware of I-O DATA WN-AC1600DGR3 has completely compatibility
with WN-AC1600DGR2, so modify the device name of WN-AC1600DGR2 to show
the images for DGR2 also support DGR3.
Flash instruction (WebUI):
Download *-factory.bin image and upload it via the firmwary upgrade
function of the stock firmware WebUI.
Flash instruction (TFTP):
1. Set PC to fixed IP address 192.168.0.66
2. Download *-factory.bin image and rename it to tp_recovery.bin
3. Start a tftp server with the file tp_recovery.bin in its root
directory
4. Turn off the router
5. Press and hold reset button
6. Turn on router with the reset button pressed and wait ~15 seconds
7. Release the reset button and after a short time the firmware should
be transferred from the tftp server
8. Wait ~30 second to complete recovery
Flash instruction (WebUI):
Download *-factory.bin image and upload it via the firmwary upgrade
function of the stock firmware WebUI.
Flash instruction (TFTP):
1. Set PC to fixed ip address 192.168.0.66
2. Download *-factory.bin image and rename it to tp_recovery.bin
3. Start a tftp server with the file tp_recovery.bin in its root directory
4. Turn off the router
5. Press and hold Reset button
6. Turn on router with the reset button pressed and wait ~15 seconds
7. Release the reset button and after a short time
the firmware should be transferred from the tftp server
8. Wait ~30 second to complete recovery.
Andrew Cameron [Wed, 26 Jun 2019 21:40:54 +0000 (16:40 -0500)]
ath79: add support for TP-Link CPE510-v2/v3
TP-Link CPE510-v2/v3 is an outdoor wireless CPE for 5 GHz with
one Ethernet port based on Atheros AR9344
Specifications:
- Based on the same underlying hardware as the TP-Link CPE510
- Power, LAN, and 4 green LEDs
- 1 10/100Mbps Shielded Ethernet Port (Passive PoE in)
- Built-in 13dBi 2x2 dual-polarized directional MIMO antenna
- Adjustable transmission power from 0 to 23dBm/200mw
Flashing instructions:
Flash factory image through stock firmware WEB UI
or through TFTP
To get to TFTP recovery just hold reset button while powering on for
around 4-5 seconds and release.
Rename factory image to recovery.bin
Stock TFTP server IP:192.168.0.100
Stock device TFTP adress:192.168.0.254
Signed-off-by: Andrew Cameron <apcameron@softhome.net>
[whitespace fixes] Signed-off-by: Petr Štetiar <ynezz@true.cz>
Georgi Vlaev [Wed, 17 Apr 2019 18:17:45 +0000 (21:17 +0300)]
ramips: add support for TP-Link RE650 v1
TP-Link RE650 v1 is a dual-band AC2600 range extender,
based on MediaTek MT7621A and MT7615E. According to the
wikidevi entry for RE650 this device is identical with
TP-Link RE500 as hardware. This patch supports only RE650.
Upload
openwrt-ramips-mt7621-tplink_re650-v1-squashfs-factory.bin
from the RE650 web interface.
TFTP recovery to stock firmware:
Unfortunately, I can't find an easy way to recover the RE
without opening the device and using modified binaries. The
TFTP upload will only work if selected from u-boot, which
means you have to open the device and attach to the serial
console. The TFTP update procedure does *not* accept the
published vendor firmware binaries. However, it allows to
flash kernel + rootfs binaries, and this works if you have
a backup of the original contents of the flash. It's probably
possible to create special image out of the vendor binaries
and use that as recovery image.
Signed-off-by: Georgi Vlaev <georgi.vlaev@gmail.com>
[re-added variables for kernel header] Signed-off-by: David Bauer <mail@david-bauer.net>
Petr Štetiar [Wed, 17 Jul 2019 21:04:22 +0000 (23:04 +0200)]
mtd: cleanup unused code and variables in fis.c
While compile checking mtd changes in PR#1359 I've noticed following
compiler warnings and cleaned them up:
fis.c: In function 'fis_remap':
fis.c:143:25: warning: variable 'redboot' set but not used [-Wunused-but-set-variable]
struct fis_image_desc *redboot = NULL;
^~~~~~~
fis.c:142:25: warning: variable 'fisdir' set but not used [-Wunused-but-set-variable]
struct fis_image_desc *fisdir = NULL;
^~~~~~
This board was used in dual-band 802.11n enterprise access points, models
BSAP-1800v2 and BSAP-1840, introduced in 2010 by Bluesocket, which was
acquired by Adtran in 2011, who has now EOL'ed them. They differed only in
that the BSAP-1840's antennae were detachable, while the BSAP-1800v2's were
inside the case. They have an external RJ-45 console port, which works with
standard Cisco 72-3383-01 console cables.
Installation:
1. Connect to the serial console using a terminal that supports YMODEM at
115200 bps, 8 data bits, no parity, 1 stop bit
2. Interrupt the bootloader using its password, which is: r00t
3. Issue the "fis init" command, confirming if prompted
4. Look at the length of the openwrt-ath79-generic-*-squashfs-kernel.bin
file, and substitute it below, instead of where I have "LeNgTh"
5. Issue the following command, and upload this file using YMODEM protocol
load -r -v -b 0x80060000 -m ymodem
6. Issue the following commands, substituting as mentioned above:
fis create -b 0x80060000 -l LeNgTh vmlinux_2
load -r -v -b 0x80100000 -m ymodem
7. Using YMODEM, upload openwrt-ath79-generic-*-squashfs-rootfs.bin
8. Issue the "fis free" command, and for the first range in its response,
use a hexadecimal calculator to subtract the start from the end in order
to substitute it below, with the leading "0x" to specify it in
hexadecimal, instead of where I have "LeNgTh"
9. Issue the following commands, substituting as mentioned above:
fis create -b 0x80100000 -l LeNgTh -e 0 -r 0 rootfs
reset
10.Wait for the status LED to go solid green
Tested-by: Brian Gonyer <bgonyer@gmail.com> Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
[fixed obsolete $ARGV in platform_do_upgrade] Signed-off-by: Petr Štetiar <ynezz@true.cz>
The Bluesocket BSAP1880 board has a Vitesse PHY, for which the driver was
not being included, and its RedBoot directory block is earlier in the
flash than the search was allowing. This commit prepares for supporting it.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
The code for calculating the CRC32 signatures for RedBoot FIS partitions
was already included, but for unknown reasons, it was never invoked. Some
bootloaders enforce checking these for loaded kernels, so they should be
written. This patch does so.
Tested-by: Brian Gonyer <bgonyer@gmail.com> Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
brcm47xx: fix 831-old_gpio_wdt.patch for linux 4.19
Fix the following compilation error when using linux 4.19 on brcm47xx:
drivers/watchdog/old_gpio_wdt.c:236:2: error: implicit declaration of function 'setup_timer'
setup_timer was removed and replaced by timer_setup in linux >=4.15:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=513ae785
Petr Štetiar [Tue, 16 Jul 2019 19:46:49 +0000 (21:46 +0200)]
kernel: generic: fix perf build breakage on 4.19
Drop 211-host_tools_portability.patch which is breaking perf build on
4.19 kernels by removing the include directory from the host's CFLAGS
leading to the following build breakage:
pmu-events/jevents.c:48:10: fatal error: linux/list.h: No such file or directory
#include <linux/list.h>
^~~~~~~~~~~~~~
Petr Štetiar [Tue, 16 Jul 2019 08:23:18 +0000 (10:23 +0200)]
imx6: apalis: fix sysupgrade paths
We can't use $board_name in sysupgrade paths for kernel and rootfs as we
currently generate one image which could be used on two different base
boards.
Alberto Bursi [Sun, 14 Jul 2019 01:11:00 +0000 (03:11 +0200)]
x86: add modern network modules to Generic target
Many Atom-based embedded/industrial x86 boards can't run 64bit operating
systems due to either processor or board firmware limitations, but they
have modern interfaces (PCIe) or have modern Intel gigabit controllers
onboard. With the current default package selection for x86 Generic
target their network won't work.
Add the modern gigabit network modules needed or most likely going to be
used as add-in cards, similar to what is the list on x86_64 target.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
[fixed whitespace issue] Signed-off-by: Petr Štetiar <ynezz@true.cz>
Felix Fietkau [Wed, 19 Jun 2019 10:32:20 +0000 (12:32 +0200)]
mac80211: add config tweak for tx bursting when using VHT
By default, set BE tx queue TXOP limit to 1.0 in the hostapd config
Many vendor drivers are doing similar things to boost throughput.
On MT7612 under ideal conditions, it improves tx throughput from 470 Mbit/s
to about 570 Mbit/s.
The only step between platform_pre_upgrade() and platform_do_upgrade()
is switching to ramdisk. It should be fine to "mtd erase firmware" from
the later callback and get rid of the first one.
This change wasn't tested on affected target but identical code logic
was verified to work as expected on brcm47xx with initramfs firmware.
The only step between platform_pre_upgrade() and platform_do_upgrade()
is switching to ramdisk. It should be fine to "mtd erase firmware" from
the later callback and get rid of the first one.
This change wasn't tested on affected target but identical code logic
was verified to work as expected on brcm47xx with initramfs firmware.
fixes intermittent loss of connectivity on 1Gbit port, with log message:
> 803x_aneg_done: SGMII link is not ok
Thanks to David Bauer for pointing me in the right direction.
I just had to figure out the right bus_id, which you find in this log:
> ag71xx ag71xx.1: connected to PHY at gpio-1:00 [uid=004dd074,
driver=Atheros 8031 ethernet]
ramips: Remove left-over mistyped device string from 02_network
In 2011 (!), network setup for nexaira,bc2 was moved to network
defaults script with a typo so it became b2c: 295e04084c ("ramips: setup bc2 mac addresses from the generic network script")
This patch just removes the useless entry without replacement,
since it seems to have worked for 8 years anyway.
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
ramips: mtk-mmc: mt76x8: check ESD_MODE before applying AGPIO_CFG
Since mt76x8an ver1 eco2, SDXC pins can be switched to the
following pinmap:
sd_d1 -> PAD_I2S_SDI
sd_d0 -> PAD_I2S_WS
sd_cmd -> PAD_I2S_CLK
sd_d3 -> PAD_I2C_SCLK
sd_d2 -> PAD_I2C_SD
sd_clk -> PAD_GPIO0
sd_wp -> PAD_TXD1
sd_cd -> PAD_RXD1
To use this pinmap, one would need to set ESD_MODE bit (bit 15)
to 1 in GPIO1_MODE and switch other used pads into GPIO mode.
In this mode, we don't need to switch ethernet pins to digital
pad.
Check ESD_MODE bit before applying AGPIO_CFG and use rt_sysc_m32
to set it.
Petr Štetiar [Sun, 14 Jul 2019 20:10:37 +0000 (22:10 +0200)]
fstools: add direct dependencies on libblobmsg-json and libjson-c
The OpenWrt buildroot ABI version rebuild tracker does not handle
transient dependencies, therefore add all libraries linked by
block-mount and blockd as direct dependencies to the corresponding
binary package definition.
This ensures that block-mount and blockd is automatically rebuilt and
relinked if any of these libraries has its ABI_VERSION updated in the
future.
Fixes: FS#2373
[jow: similar fix for procd and 98.42% of commit message] Signed-off-by: Jo-Philip Wich <jow@mein.io> Signed-off-by: Petr Štetiar <ynezz@true.cz>
David Bauer [Tue, 11 Jun 2019 15:27:51 +0000 (17:27 +0200)]
ath79: add support for devolo WiFi pro 1750x
Hardware
--------
CPU: Qualcomm Atheros QCA9558
RAM: 128M DDR2
FLASH: 16MiB
ETH: 1x Atheros AR8035 (PoE in)
WiFi2: QCA9558 3T3R
WiFi5: QCA9880 3T3R
BTN: 1x Reset
LED: 1x LED Power (non-controllable)
1x LED Status (internal)
1x LED LAN (controlled by PHY)
1x LED WLAN
BEEP: 1x GPIO attached piezo beeper
UART: 3.3V GND TX RX (115200-N-8) (3.3V is square pad)
Header is located next to external-LED header.
Installation
------------
Make sure you set a password for the root user as prompted on first
setup!
1. Upload OpenWRT sysupgrade image via SSH to the device.
Use /tmp as the destination folder on the device.
User is root, password the one set in the web interface.
The command line enforced by the kernel is harmful, we want to
use the command line passed from the device tree. This is
especially important for the SQ201, Teltonika RUT and WBD111
which has a console set to 115200n8 and not 19200n8 like all
the others, so their consoles will be confusingly changing
baudrate between boot loader and kernel. The setting in the
device tree corresponds to what the boot loader is using.
David Bauer [Tue, 2 Jul 2019 21:57:18 +0000 (23:57 +0200)]
gpio-button-hotplug: unify polled and interrupt code
This patch unifies the polled and interrupt-driven gpio_keys code
paths as well implements consistent handling of the debounce
interval set for the GPIO buttons and switches.
Hotplug events will only be fired if
1. The input changes its state and remains stable for the duration
of the debounce interval (default is 5 ms).
2. In the initial stable (no state-change for duration of the
debounce interval) state once the driver module gets loaded.
Switch type inputs will always report their stable state.
Unpressed buttons will not trigger an event for the initial
stable state. Whereas pressed buttons will trigger an event.
This is consistent with upstream's gpio-key driver that uses
the input subsystem (and dont use autorepeat).
Prior to this patch, this was handled inconsistently for interrupt-based
an polled gpio-keys. Hence this patch unifies the shared logic into the
gpio_keys_handle_button() function and modify both implementations to
handle the initial state properly.
The changes described in 2. ) . can have an impact on the
failsafe trigger. Up until now, the script checked for button
state changes. On the down side, this allowed to trigger the
failsafe by releasing a held button at the right time. On the
plus side, the button's polarity setting didn't matter.
Now, the failsafe will only engage when a button was pressed
at the right moment (same as before), but now it can
theoretically also trigger when the button was pressed the
whole time the kernel booted and well into the fast-blinking
preinit phase. However, the chances that this can happen are
really small. This is because the gpio-button module is usually
up and ready even before the preinit state is entered. So, the
initial pressed button event gets lost and most devices behave
as before.
Bisectors: If this patch causes a device to permanently go into
failsafe or experience weird behavior due to inputs, please
check the following:
- the GPIO polarity setting for the button
- the software-debounce value
Run-tested for 'gpio-keys' and 'gpio-keys-polled' on
- devolo WiFi pro 1200e
- devolo WiFi pro 1750c
- devolo WiFi pro 1750x
- Netgear WNDR4700
- Meraki MR24
- RT-AC58U
Signed-off-by: David Bauer <mail@david-bauer.net> Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [further
cleanups, simplification and unification]
mac80211: Do not build b43legacy on BRCM47xx mips74 subtarget
b43legacy needs ssb support and we do not compile the mips74 subtarget
of the brcm47xx target with SSB support. This causes a build failure in
the mac80211 package and only some of the kernel modules are being
created.
I am not aware of any device with a BRCM47xx mips74 CPU which uses a
b43legacy compatible device.
David Bauer [Thu, 11 Jul 2019 15:15:48 +0000 (17:15 +0200)]
ramips: add support for ASUS RT-AC57U
SoC: MediaTek MT7621AT
RAM: 128M (Winbond W631GG6KB-15)
FLASH: 16MB (Spansion S25FL128SA)
WiFi: MediaTek MT7603EN bgn 2SS
WiFi: MediaTek MT7612EN nac 2SS
BTN: Reset - WPS
LED: - Power
- LAN {1-4}
- WAN
- WiFi 2.4 GHz
- WiFi 5 GHz
- USB
UART: UART is present next to the Power LED.
TX - RX - GND - 3V3 / 57600-8N1
3V3 is the nearest one to the Power LED.
Installation
------------
Via TFTP:
1. Set your computers IP-Address to 192.168.1.75.
2. Power up the Router with the Reset button pressed.
3. Release the Reset button after 5 seconds.
4. Upload OpenWRT sysupgrade image via TFTP:
> tftp -4 -v -m binary 192.168.1.1 -c put <IMAGE>
Via SSH:
Note: User/password for SSH is identical with the one used in the
Web-interface.
1. Complete the initial setup wizard.
2. Activate SSH under "Administration" -> "System".
3. Transfer the OpenWrt sysupgrade image via scp:
> scp owrt.bin admin@192.168.1.1:/tmp
4. Connect via SSH to the router.
> ssh admin@192.168.1.1
5. Write the OpenWrt image to flash.
> mtd-write -i /tmp/owrt.bin -d linux
6. Reboot the router
> reboot
Update the lua5.3 package to build a shared object just like the old lua
package. Ported / recreated the same patch number as the other lua
package. Built and tested library / interpreter on BCM5301X.
mt7621 has the following memory map:
0x0-0x1c000000: lower 448m memory
0x1c000000-0x2000000: peripheral registers
0x20000000-0x2400000: higher 64m memory
detect_memory_region in arch/mips/kernel/setup.c only add the first
memory region and isn't suitable for 512m memory detection because
it may accidentally read the memory area for peripheral registers.
This commit adds memory detection capability for mt7621:
1. add the highmem area when 512m is detected.
2. guard memcmp from accessing peripheral registers:
This only happens when some weird user decided to change
kernel load address to 256m or higher address. Since this
is a quite unusual case, we just skip 512m testing and return
256m as memory size.
projects / thirdparty / openwrt.git / log
