When trying to boot an OS installer or a live image via EFI HTTP the
following happens
- U-Boot downloads the image and mounts it in memory
- The EFI subsystem is invoked and the image is started
- The OS calls ExitBootServices and the memory that holds the mounted
image might get overwritten
This results in installers complaining that they can't find installer
medium or live images complaining they can't find the root filesystem.
ACPI already deals with it by having NFIT and NVDIMM to provide ramdisks
that need to be preserved by the OS. Linux and device trees have support
for persistent memory devices (pmem).
We can use them and inject a pmem node in the DT to preserve memory across the
entire boot sequence. Linux will just create a block device over the reserved
memory and installers/images can re-discover it.
This is what it looks like from the OS perspective:
nd_pmem namespace0.0: unable to guarantee persistence of writes
pmem0: p1 p2 p3
EXT4-fs (pmem0p3): mounted filesystem f40f64a4-5b41-4828-856e-caaae2c1c2a0 r/w with ordered data mode. Quota mode: disabled.
EXT4-fs (pmem0p3): re-mounted f40f64a4-5b41-4828-856e-caaae2c1c2a0 r/w. Quota mode: disabled.
Adding 45052k swap on /dev/pmem0p2. Priority:-2 extents:1 across:45052k SS
root@genericarm64:~# mount | grep pmem
/dev/pmem0p3 on / type ext4 (rw,relatime)
/dev/pmem0p1 on /boot type vfat (rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=iso8859-1,shortname=mixed,errors=remount-ro)
It's worth noting that Linux behaves differently with reserved memory
(at least on arm64) and that depends on kernel config options.
CONFIG_ZONE_DEVICES and CONFIG_ARM64_PMEM are such options. It boils down to
how the kernel tries to map pages. If devm_memremap_pages() gets called instead
of devm_memremap() mapping the memory fails.
The only safe way is to remove the memory from the EFI memory map,
rather than defining it as /reserved no-map;/ in the DT.
Commit 56f186a68b ("lmb: check if a region can be reserved by
lmb_reserve()") fixed the lmb_reserve() and lmb_alloc_addr() API's for
some corner case scenarios, and also added corresonding test cases for
these corner cases. These tests were checking, among other things, the
lmb_alloc_addr() API. The above commit was applied to the next branch.
Subsequently, there was commit 67be24906f
("lmb: change the return code on lmb_alloc_addr()") which was first
applied on the master branch, and subsequently got merged to next as
part of the rebase. The second commit changes the return value of the
lmb_alloc_addr() API, which now results in some of the tests added as
part of the first commit to fail. Fix those test cases.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
The EFI HTTP boot puts the ISO installer image at some location in
memory. Information about this image has to be passed on to the OS
kernel, which is done by adding a persistent memory(pmem) node to the
devicetree(DT) that is passed to the OS. The OS kernel then gets
information about the presence of this ISO image and proceeds with the
installation.
In U-Boot, this ISO image gets mounted as a memory mapped blkmap
device slice, with the 'preserve' attribute. Add a helper function
which iterates through all such slices, and invokes a callback. The
callback adds the pmem node to the DT and removes the corresponding
memory region from the EFI memory map. Invoke this helper function as
part of the DT fixup which happens before booting the OS.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Tobias Waldekranz <tobias@waldekranz.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Some blkmap memory mapped devices might have to be relevant even
after U-Boot passes control to the next image as part of the platform
boot. An example of such a mapping would be an OS installer ISO image,
information for which has to be provided to the OS kernel. Use the
'preserve' attribute for such mappings. The code for adding a pmem
node to the device-tree then checks if this attribute is set, and adds
a node only for mappings which have this attribute.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Tobias Waldekranz <tobias@waldekranz.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Add information about the type of blkmap slices as an attribute in the
corresponding slice structure. Put information in the blkmap slice
structure to identify if it is associated with a memory or linear
mapped device. Which can then be used to take specific action based on
the type of the blkmap slice.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Tobias Waldekranz <tobias@waldekranz.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
ACPI has NFIT and NVDIMM support to provide ramdisks to the OS. Linux
and device trees have support for persistent memory(pmem) devices. The
firmware can then add a pmem node for the region of memory occupied by
the ramdisk when passing the device-tree to the OS.
It's worth noting that for linux to instantiate the /dev/pmemX device,
the memory described in the pmem node has to be omitted from the EFI
memory map we hand over to the OS if ZONE_DEVICES and SPARSEMEM is
enabled. With those enabled the pmem driver ends up calling
devm_memremap_pages() instead of devm_memremap(). The latter works
whether the memory is omitted or marked as reserved, but mapping pages
only works if the memory is omitted.
On top of that, depending on how the kernel is configured, that memory
area must be page aligned or 2MiB aligned. PowerPC is an exception here
and requires 16MiB alignment, but since we don't have EFI support for
it, limit the alignment to 2MiB.
Ensure that the ISO image is 2MiB aligned and remove the region
occupied by the image from the EFI memory map.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
With upcoming changes supporting pmem nodes, we need to remove the
pmem area from the EFI memory map. Rename efi_add_memory_map_pg() to
efi_update_memory_map(), and allow removing memory from the EFI memory
map.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
One of the problems an OS may face, when running in EFI, is that
a mounted ISO, after calling ExitBootServices goes away, if that ISO
is resident in RAM memory as a ramdisk.
ACPI has NFIT and NVDIMM support to provide ramdisks to the OS, but we
don't have anything in place for DTs. Linux and device trees have support
for persistent memory devices. So add a function that can inject a pmem
node in a DT, so we can pass information on the ramdisk the OS.
Signed-off-by: Masahisa Kojima <kojima.masahisa@socionext.com>
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
The efi_install_fdt() function is called before booting an EFI binary,
either directly, or through a bootmanager. This function installs a
copy of the device-tree(DT) on the EFI configuration table, which is
passed on to the OS.
The current logic in this function does not install a DT if a
device-tree is already installed as an EFI configuration
table. However, this existing copy of the DT might not be up-to-date,
or it could be a wrong DT for the image that is being booted. Always
install a DT afresh to the configuration table before booting the EFI
binary.
Installing a new DT also involves some additional checks that are
needed to clean up memory associated with the existing DT copy. Check
for an existing copy, and free up that memory.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
There is logic in the copy_fdt() function which is iterating over the
platform's DRAM banks and setting the fdt_ram_start variable. However,
this variable is not used subsequently in the function. Remove this
superfluous code.
Signed-off-by: Sughosh Ganu <sughosh.ganu@linaro.org>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
The imx8_cpu driver is a CPU Driver that supports the i.MX9
family to display the CPU type, temperature grade and
current operating temperature. The older file,
arch/arm/mach-imx/cpu.c, does not support i.MX9, so this config
is enabled in various IMX9 boards. Instead of having this option
enabled in every IMX9, select this driver by default for the
platform.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The imx8_cpu driver is a CPU Driver that supports the i.MX8Q
family. When it is enabled, it acts as an alternative to
arch/arm/mach-imx/cpu.c, but the imx8_cpu supports the driver
model where cpu.c does not. Imply this newer driver by default.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The imx8_cpu driver is a CPU Driver that supports the i.MX8M
family, and when it is enabled, it acts as an alternative to
arch/arm/mach-imx/cpu.c, but the imx8_cpu supports the driver
model where cpu.c does not. Imply this newer driver by default.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Much of the data that is display by imx8_cpu.c is also displayed from
arch/arm/mach-imx/cpu.c, except the temperature grade and active
temperature are only displayed when SoC is an i.MX9. Since IMX9 now
implies IMX_TMU, change this to check for IMX_TMU in the same way
it's done in mach-imx/cpu.c to enable displaying this information
for any SoC with either of this config enabled.
Since additional text may appear due to this commit, remove
the extra space in the message displaying the temperature
grade.
Before:
CPU: NXP i.MX8MP Rev1.1 A53 at 1200 MHz
Model: Beacon EmbeddedWorks i.MX8MPlus Development kit
After:
CPU: NXP i.MX8MP Rev1.1 A53 at 1200 MHz
CPU: Industrial temperature grade (-40C to 105C) at 28C
Model: Beacon EmbeddedWorks i.MX8MPlus Development kit
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
If the CPU Information is displayed from imx8_cpu, it displays the
cpu temperature grade and operating temperature if CONFIG_IMX9 is
defined. This behavior is similar to what happens mach-imx/cpu.c,
except that the latter checks for IMX_THERMAL or IMX_TMU.
In preparation to make imx8_cpu act like the previous implementation
for any CPU, make IMX8M imply IMX_TMU so it will be always displayed
unless a user decides to disable it.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
The imx8_cpu is capable of running on IMX8, IMX8M, and IMX9
families, but the CPU list is limited on the 8M variants. Expand
this list to show more variants and their respective names.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
If the CPU Information is displayed from imx8_cpu, it displays
the cpu temperature grade and operating temperature if
CONFIG_IMX9 is defined. This behavior is similar to what
happens arch/arm/mach-imx/cpu.c except that the latter
checks for CONFIG_IMX_THERMAL or CONFIG_IMX_TMU.
In preparation to make imx8_cpu act like the previous
implementation for any CPU, make IMX9 imply IMX_TMU, so
it will be always displayed unless a user decides to
disable it.
Signed-off-by: Adam Ford <aford173@gmail.com>
Reviewed-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
There is a config option to run the PMIC at nominal voltages
which is not enabled on the i.MX8MP Beacon kit, so it the PMIC
runs at overdrive voltages. Unfortuately, the check for this
condition to set the GIC clock parent and rate is backwards from
what it should be, and accidentally sets the GIC clock to nominal
if the PMIC is in overdrive, and sets the GIC clock to overdrive
if the PMIC is in nominal. Fix this by inverting the logic on the
check.
Fixes: ab53bd43db ("arm64: imx: Add support for imx8mp-beacon-kit")
Signed-off-by: Adam Ford <aford173@gmail.com>
The configuration descriptions generated by binman contain the vendor
device-tree directory. Instead of adding it to all match strings just strip
it off.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
JH7110 boards are currently the only use case for multi DTB FIT images
on RISC-V.
Booting JH7110 systems with a VisionFive 2 device-tree used to kind of
work without causing harm to the hardware. But there is no guarantee
that this will hold true in future. So we should not rely on it.
Before the current patch series booting failed on unsupported boards due
to the lack of a device-tree in the binman generated default configuration
when reaching main U-Boot.
By not setting a default configuration booting will now fail on
unsupported boards already in SPL. This allows SPL to
continue with the next boot source for a possible recovery.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: E Shattow <e@freeshell.de>
Binman can automatically generate device-tree and configuration entries in
the FIT image based on CONFIG_MULTI_DTB_FIT if the binman node includes the
right sub-nodes.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: E Shattow <e@freeshell.de>
Remove CFG_SYS_SDRAM_BASE so that we can get DRAM base
from dt instead of compile time config.
Removing this config helps the u-boot more portable.
Signed-off-by: Jimmy Ho <jimmy.ho@sifive.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Introduced in commit 5a4e0625ac ("riscv: dts: sophgo: Add ethernet
node"), eth_{csrclk,ptpclk} were used as placeholders for ethernet
controller. As the real clock controller has been added, drop them to
clean the devicetree up.
Signed-off-by: Yao Zi <ziyao@disroot.org>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
The zkr ISA extension can be used to generate random numbers. Since RVA22
zkr is an optional ISA extension. It can be emulated by QEMU. Our RNG
driver detects if the extension is usable during driver binding. Let's
enable it by default on QEMU.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Without 'nvme scan' the ESP on the NVMe drive is not found early.
EFI variables cannot be persisted.
Hit any key to stop autoboot: 0
Cannot persist EFI variables without system partition
** Booting bootflow '<NULL>' with efi_mgr
Loading Boot0000 'mmc 0' failed
EFI boot manager: Cannot load any image
Boot failed (err=-14)
scanning bus for devices...
** Booting bootflow 'nvme#1.blk#1.bootdev.part_1' with efi
Booting /\EFI\BOOT\BOOTRISCV64.EFI
error: no suitable video mode found.
GNU GRUB version 2.12
With 'nmve scan' booting works as expected.
Hit any key to stop autoboot: 0
** Booting bootflow '<NULL>' with efi_mgr
Loading Boot0000 'mmc 0' failed
Loading Boot0001 'nvme 0' failed
Booting: nvme 1
error: no suitable video mode found.
GNU GRUB version 2.12
Reported by Yuri Zaporozhets <yuriz@vodafonemail.de>
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
qcom-next-20230324:
* msm8916 gets proper sysreset and spin-table support
* The first new IPQ platform is added - the IPQ9574. The IPQ series are
used in routers. The flashing process is also documented
* mach-snapdragon gains the ability to boot with an internal FDT and
still parse memory from an externally provided one
* SC7280 gets a pinctrl driver and various clock driver improvements.
* Qualcom clock drivers will now actually return an error when
attempting
to enable a clock which isn't described.
* Qualcomm pinctrl drivers will now return an error when attempting to
configure an invalid function mux
Ben reports a failure to boot the kernel on hardware that starts its
physical memory from 0x0.
The reason is that lmb_alloc_addr(), which is supposed to reserve a
specific address, takes the address as the first argument, but then also
returns the address for success or failure and treats 0 as a failure.
Since we already know the address change the prototype to return an int.
Reported-by: Ben Schneider <ben@bens.haus>
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Tested-by: Ben Schneider <ben@bens.haus>
Reviewed-by: Sughosh Ganu <sughosh.ganu@linaro.org>
In case ENV_IS_IN_SPI_FLASH is disabled, returning ENVL_SPI_FLASH
leads to failure to find environment driver on start up. Fix this
by testing whether ENV_IS_IN_SPI_FLASH is enabled and if not, then
return ENVL_NOWHERE instead.
Signed-off-by: Marek Vasut <marex@denx.de>
Pass struct udevice * into imx_clk_fixed_factor*() functions, so the
clock core would have access to parent struct udevice *.
Signed-off-by: Marek Vasut <marex@denx.de>
Use clock-names property which is accessible via parent clock OF node
to look up the parent clock by name instead of depending on unreliable
global clock name to perform look up.
Signed-off-by: Marek Vasut <marex@denx.de>
Use U-Boot specific struct udevice instead of Linux compatibility
struct device in clk-fixed-factor registration.
Signed-off-by: Marek Vasut <marex@denx.de>
Use clock-names property which is accessible via parent clock OF node
to look up the parent clock by name instead of depending on unreliable
global clock name to perform look up.
Signed-off-by: Marek Vasut <marex@denx.de>
Pass struct udevice * into imx_clk_divider*() functions, so the
clock core would have access to parent struct udevice *.
Signed-off-by: Marek Vasut <marex@denx.de>
Use U-Boot specific struct udevice instead of Linux compatibility
struct device in clk-divider clock registration.
Signed-off-by: Marek Vasut <marex@denx.de>
Use clock-names property which is accessible via parent clock OF node
to look up the parent clock by name instead of depending on unreliable
global clock name to perform look up.
Signed-off-by: Marek Vasut <marex@denx.de>
Pass struct udevice * into imx_clk_pllv3*() functions, so the
clock core would have access to parent struct udevice *.
Signed-off-by: Marek Vasut <marex@denx.de>
Convert clock-osc-24m back to osc_24m and clock-osc-32k back to osc_32k.
These are the clock which match clock tables in Linux. This is now
possible because the clock drivers now resolve clock names based on
clock-names DT property in the CCM DT node.
Signed-off-by: Marek Vasut <marex@denx.de>
Pass struct udevice * into imx_clk_composite*() functions, so the
clock core would have access to parent struct udevice *.
Signed-off-by: Marek Vasut <marex@denx.de>
Use clock-names property which is accessible via parent clock OF node
to look up the parent clock by name instead of depending on unreliable
global clock name to perform look up.
Signed-off-by: Marek Vasut <marex@denx.de>
