| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
|
| |
Th v2022.08.0 release added Raspberry Pi i2c controller support as
well as Raspberry Pi 4 support. We enable both here and will put
the Raspberry Pi 4 support into use in a follow-up commit.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220930162159.814389-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
| |
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
- enabling drivers for dwc2 gadget, dwc2 host, stm32_usbphyc and stmp32_pwr
Signed-off-by: Michael Grzeschik <m.grzeschik@pengutronix.de>
Link: https://lore.pengutronix.de/20220907195035.2643740-1-m.grzeschik@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220907132343.3321447-2-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220907132343.3321447-1-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
It was enabled for mx6&rpi&stm32mp, but not for am335x&at91&vexpress.
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220906103328.3303545-4-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
It was enabled for mx6&rpi&stm32mp, but not for am335x&at91&vexpress.
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220906103328.3303545-3-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
It was enabled for mx6&rpi&stm32mp&vexpress, but not for am335x&at91.
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220906103328.3303545-2-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
It was enabled for mx6&rpi&stm32mp&vexpress, but not for am335x&at91.
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220906103328.3303545-1-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
Package name and symbol name should match.
Signed-off-by: Michael Olbrich <m.olbrich@pengutronix.de>
Link: https://lore.pengutronix.de/20220902104854.2392497-1-m.olbrich@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
Use the same name as the other barebox rules.
Signed-off-by: Jan Luebbe <jlu@pengutronix.de>
Link: https://lore.pengutronix.de/20220803130344.1895310-1-jlu@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In remote labs co-located with other hardware, we've observed card
detect levers of different boards to sporadically fail to detect
the card, e.g. because the cable on the usbsdmux was yanked around
by accident. When this happens, barebox usually boots up normally as
the card detect is ignored and then Linux waits indefinitely for
the card-detect to turn active. Add a new config option that can be
enabled to avoid these issues altogether.
We enable this for all configurations, except for barebox-am335x-mlo,
which is size-constrained as it needs to fit into on-chip SRAM
and those 100 extra bytes are deemed too expensive.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-8-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
| |
There has been two barebox releases in the mean time, so give the v7a
barebox-common an update to get rid of the patches we carry.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-7-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
| |
We have everything in place now to switch from stm32image to FIP.
This gives us upstream support for factory bootstrap via DFU and
avoids the breakage on v2.8, which will be the first release removing
support of the deprecated stm32image format.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-6-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For building FIP images, we need a firmware config describing e.g.
where to place components, a device tree for the bl32, a secure
monitor (SP_min, bl32.bin), as well as non-trusted firmware and
hardware configuration. The latter two will be provided by barebox,
so install here the ones provided by TF-A in addition to the TF-A
binaries itself (which will load the FIP).
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-5-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
| |
FIP images are passed along a device tree via r2 (called hardware
config). As we want barebox to be booted with the barebox DT to
decouple kernel and bootloader updates, install the relevant barebox
DTs, but with a barebox- prefix as they would conflict with kernel
device tree blobs otherwise.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-4-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
| |
We need at least v2.6 for FIP support, but there's a freshly released
v2.7, so lets use that. This unfortunately means we need to carry a
patch for the MC-1 because upstream broke it. To ensure the legacy
image works for now, we need to set a new build option, but that will
be removed soon enough with the switch to FIP.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
| |
We only generate SD/eMMC images for STM32MP1, so having NAND/NOR
drivers in the FSBL just increases size pressure On-Chip RAM,
so drop them.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220725120445.1295669-1-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
A new ptxdist version is out, update.
Signed-off-by: Johannes Zink <j.zink@pengutronix.de>
Link: https://lore.pengutronix.de/20220705084150.3272186-1-j.zink@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
A new ptxdist version is out, update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
barebox has had the ability to generate an extra image that's bootable
like a kernel for a while now. With small adjustment, we can let the
videocore use this image and pass it the appropriate device tree
depending on the board used. Moving this decision into the VideoCore
allows us to use the same image for both Raspberry Pi 2, Rasperry Pi 3
and Raspberry Pi CM3.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220426083821.1829442-3-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Config changes are those of olddefconfig with the exception of
CONFIG_FS_UBIFS_COMPRESSION_ZSTD for the i.MX6 barebox: We already have
UBIFS support enabled, so it makes sense to allow mounting UBIFS with
zstd as well.
This update imports two barebox patches to address regressions due to
recent Raspberry Pi upstream rework.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220426083821.1829442-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
A new ptxdist version is out, update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
| |
Sync barebox-at91 make file with other make variants and enable build
out of tree, to avoid build conflicts if local_src is used.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://lore.pengutronix.de/20220407105352.1256734-3-o.rempel@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
| |
Sync barebox-stm32mp make file with other make variants and enable build
out of tree, to avoid build conflicts if local_src is used.
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Link: https://lore.pengutronix.de/20220407105352.1256734-1-o.rempel@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Since 5.16, there is a bug that results in this backtrace:
[ 0.000000] L2C: platform modifies aux control register: 0x02020000 -> 0x02420000
[ 0.000000] L2C: DT/platform modifies aux control register: 0x02020000 -> 0x02420000
[ 0.009001] Failed to initialize '/bus@40000000/motherboard-bus@40000000/iofpga@7,00000000/timer@12000': -22
[ 0.011756] smp_twd: clock not found -2
[ 0.922280] irq: type mismatch, failed to map hwirq-75 for interrupt-controller@1e001000!
[ 1.493972] physmap-flash 48000000.psram: map_probe failed
[ 1.668170] Internal error: Oops - undefined instruction: 0 [#1] PREEMPT SMP ARM
[ 1.668867] Modules linked in:
[ 1.669245] CPU: 1 PID: 1 Comm: swapper/0 Not tainted 5.16.0 #1
[ 1.669593] Hardware name: ARM-Versatile Express
[ 1.669953] PC is at crypto_unregister_alg+0xf8/0x100
[ 1.670857] LR is at 0x0
[ 1.671324] pc : [<c04bcc84>] lr : [<00000000>] psr: 20000013
[ 1.671595] sp : c18c7ea0 ip : 00000000 fp : c0d7e100
[ 1.671826] r10: c0e55858 r9 : c100c000 r8 : 00000000
[ 1.672067] r7 : c100c41c r6 : 00000001 r5 : c198ad00 r4 : c1c23080
[ 1.672354] r3 : 00000002 r2 : ffffffff r1 : 00000001 r0 : c0f9e0d0
[ 1.672702] Flags: nzCv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
[ 1.673012] Control: 10c5387d Table: 6000406a DAC: 00000051
[ 1.673309] Register r0 information: non-slab/vmalloc memory
[ 1.674112] Register r1 information: non-paged memory
[ 1.674405] Register r2 information: non-paged memory
[ 1.674629] Register r3 information: non-paged memory
[ 1.674853] Register r4 information: slab kmalloc-512 start c1c23000 pointer offset 128 size 512
[ 1.675788] Register r5 information: slab task_struct start c198ad00 pointer offset 0
[ 1.676170] Register r6 information: non-paged memory
[ 1.676474] Register r7 information: non-slab/vmalloc memory
[ 1.676727] Register r8 information: NULL pointer
[ 1.676938] Register r9 information: non-slab/vmalloc memory
[ 1.677182] Register r10 information: non-slab/vmalloc memory
[ 1.677429] Register r11 information: non-slab/vmalloc memory
[ 1.677686] Register r12 information: NULL pointer
[ 1.677923] Process swapper/0 (pid: 1, stack limit = 0x(ptrval))
We work around this by switching on CONFIG_CRYPTO_CBC=y.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
A new ptxdist version is out, update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
A new ptxdist version is out, update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
| |
Freshest barebox release incorporates the OMAP patches we have in the
BSP. Bump up the release and drop these two patches in return.
The release also introduced CONFIG_STM32_BSEC_WRITE, which enables
blowing e-fuses on STM32MP1 SoCs, e.g. to persist an Ethernet address.
We support a couple of STM32MP1 boards, so it makes sense to enable
this option. All other Kconfig changes are those of olddefconfig.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
|
|
|
|
|
|
| |
Update to the latest ptxdist release.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The upstream device tree change addressed by the previous commit
also made it into barebox and broke Beaglebone networking there.
This is fixed in master, so import the patch.
It was also reported that bumping bumping up core clock frequency for
faster boot times broke reboot from within Linux.
This might be brown out resulting from Linux scaling down the
voltage and Linux system reset not affecting the PMIC. Until that's
resolved, barebox master reverts the overclocking, so import that patch
as well. Both patches are scheduled to appear in v2022.01.0.
Fixes: 949c5ef2 ("v7a: barebox: version bump v2021.08.0 → v2021.11.0") # CPSW
Fixes: afdae878 ("v7a: barebox: version bump v2021.05.0 → v2021.07.0") # reset
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220108075723.432333-3-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The kernel now has two CPSW drivers and the device tree has two device
tree nodes, where one or the other may be enabled. The upstream device
tree was switched to new switchdev-based CPSW in v5.15-rc1.
Update our kernelconfig to enable the appropriate driver to get
networking working again on the Beaglebone Black.
Reported-by: Sohaib Mohamed <sohaib.amhmd@gmail.com>
Fixes: 6ec9f5202c28 ("platform-v7a: kernel: version bump v5.14 -> v5.15")
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20220108075723.432333-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
| |
A new ptxdist version has been released, update.
ptxdist-2021.12.0 has genimage-15 which has a strikter check for
partitions that overlay: specify holes in the barebox partition that
allow to overlap with MBR.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
We have a new kernel, update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
Update to latest barebox release. Config changes are the default.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20211126064836.1306873-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
| |
The variable expansion braces should have been arithmetic evaluation
parentheses, which had the effect that the PPID was always used as a
port number (which incidentally works fine for PPIDs < 65535).
Fixes: d975eb8b04087f8606d8 (2021-04-26, "v7a: run: make the SSH forwarding port multiprocess-compatible")
Reported-by: Jürgen Borleis <jbe@pengutronix.de>
Signed-off-by: Roland Hieber <rhi@pengutronix.de>
|
|
|
|
|
|
|
| |
We missed some ptxdist updates, so it might be a good time to catch up
now.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
| |
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This way boards can keep time over reboots:
[ 2.661867] at91_rtc f80480b0.rtc: registered as rtc0
[ 2.666917] at91_rtc f80480b0.rtc: setting system clock to 2021-09-16T08:37:00 UTC (1631781420)
[ 2.675817] at91_rtc f80480b0.rtc: AT91 Real Time Clock driver.
Signed-off-by: Alexander Dahl <ada@thorsis.com>
Reviewed-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210916084101.26647-1-ada@thorsis.com
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
| |
Linus released a new kernel, so we update.
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
| |
Update to latest barebox release. Config changes are the default.
In return we can drop the one patch that went upstream in the mean time.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210824125248.22580-1-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
| |
The current rootfs has a size of about 55 MiB, which does not leave much
space until 64 MiB if additional kernel features or userspace packages
are enabled. Increase the partition size so that vexpress.hdimg fills
the next larger power of two, which qemu is very picky about.
Signed-off-by: Roland Hieber <rhi@pengutronix.de>
Link: https://lore.pengutronix.de/20210818205819.26285-1-rhi@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The STM32MP157C-EV1 Evaluation board is the full-feature demonstration
and development platform for STMicroelectronics Arm®-based dual
Cortex®-A7 32 bits and Cortex®-M4 32 bits MPUs in the STM32MP1 Series.
The small siblings DK1 and DK2 are already supported. Add now support
for the EV1 as well, so DistroKit can be used out-of-the-box with
all the official STM32MP1 development kits.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210722133520.17946-4-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
| |
The EV1 device tree describes a GPIO regulator controlling the I/O voltage
for the SD-Card. Enable the respective Linux driver.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210722133520.17946-3-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
| |
There's been two Linux releases since the last bump, so it's about time
we update. The config changes by olddefconfig look normal, so they are
taken as is. Only thing standing out is the removal of /dev/kmem.
It will likely not be missed.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210722133520.17946-2-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
barebox v2021.07.0 now comes with support for first stage boot on the
SAMA5D3. The SAMA5D3 has a v7a CPU but lacks NEON, so it doesn't make
sense to enable for the v7a platform. Unfortunately, barebox master
is upset if you don't build both sama5d2 and sama5d3 at the same time.
Instead of working around this in the defconfig and enabling a board
we don't need and may forget to disable, import a small patch.
We should be able to drop this with v2021.08.0.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210722133520.17946-1-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
| |
This is possible since ptxdist-2021.03.0 introduced in DistroKit with
ee0945e61587 ("ptxdist: migrate to 2021.03.0").
Signed-off-by: Alexander Dahl <ada@thorsis.com>
Reviewed-by: Roland Hieber <rhi@pengutronix.de>
Link: https://lore.pengutronix.de/20210713115218.16671-1-ada@thorsis.com
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
| |
The v2.5 release of the ARM trusted firmware now contains support
for the Linux Automation MC-1, allowing us to drop our board support
patches. Use it.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210706045740.20866-1-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
| |
Migrate all platforms with default options.
Signed-off-by: Roland Hieber <rhi@pengutronix.de>
Link: https://lore.pengutronix.de/20210620180725.5193-1-rhi@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Updating to v2021.05.0 lets us drop the single patch we have,
as it went upstream in revised form.
Kconfig changes are those produced by oldconfig. One notable
new option is CONFIG_DEFAULT_ENVIRONMENT_GENERIC_NEW_IKCONFIG,
which allows barebox to embed its configuration as /env/data/config
analogously to /proc/config.gz, which we already enable for Linux.
This is not enabled here, but might be something we want to use
eventually.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Link: https://lore.pengutronix.de/20210602111803.15832-1-a.fatoum@pengutronix.de
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de>
|