Add the label name 'reserved_memory' to the reserved-memory node in all
K3 AM6* board level dts files. This is done so that the node can be
referenced and extended to add more carveout entries as needed in future
refactoring patches.
Signed-off-by: Beleswar Padhi <b-padhi@ti.com>
Acked-by: Andrew Davis <afd@ti.com>
Link: https://patch.msgid.link/20250908142826.1828676-13-b-padhi@ti.com
Signed-off-by: Nishanth Menon <nm@ti.com>
[ upstream commit: 4f1aee4723a796a92f17b23699dc861b582ddfd2 ]
(cherry picked from commit 58c447fe500d78f5adc373b4945d8317e11df072)
Simon Glass <sjg@chromium.org> says:
At present global bootmeths always run first, before all other
bootmeths. Optimisations in the code take advantage of this, putting
them at the end, so they can be used once and then forgotten.
In some cases it is useful to run global bootmeths later in the boot.
For example, the EFI-bootmgr bootmeth may itself scan devices and the
network, so running it first can hold up the boot significantly for
boards not actually relying on EFI-bootmgr to boot.
This series introduces a new field in global bootmeths which indicates
the priority, using the same scheme as is used with bootdev hunters.
Thus it is possible to insert the EFI-bootmgr bootmeth just before the
hunter for network bootdevs is invoked.
Despite the simplicity of the concept and the relatively small series,
this is a fairly significant enhancement. It is also quite tricky to
implement, largely due to the way the original code was written, with
global bootmeths being a small, size-optimised add-on to the original
bootstd implementation.
For now we only allow each global bootmeth to run at most once, but this
implementation is written in a way that we could relax that if needed.
Then the bootmeth itself could decide whether to run at any particular
point in the bootflow iteration.
Link: https://lore.kernel.org/r/20251015154423.908468-1-sjg@chromium.org
At present the EFI bootmgr scans all devices in the system before
deciding which one to boot. Ideally it would use the bootstd iterator
for this, but in the meantime, give it a lower priority, so it runs
just before the network devices.
Note that if there are no hunted network devices hunted, then it will
run at the end, after all bootdevs are exhausted. In other words, it
will always run.
Signed-off-by: Simon Glass <sjg@chromium.org>
When there are no more bootdevs we should still go through the global
bootmeths, since some may not have yet been used, if their priority has
not yet come up.
Add a final check for this at the end of the iterator.
Update the documentation to match the new behaviour of global bootmeths.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present before scanning global bootmeths, the iterator sets the
method count to the index of the first global bootmeth. Now that we
support scanning the global bootmeths multiple times, we must leave this
count alone.
Check against have_global and first_glob_method instead.
Signed-off-by: Simon Glass <sjg@chromium.org>
Allow bootmeths to select when they want to run, using the bootdev
priority. Provide a new bootmeth_glob_allowed() function which checks if
a bootmeth is ready to use.
Fix a comment in bootflow_system() which is a test for global bootmeths.
Signed-off-by: Simon Glass <sjg@chromium.org>
Use the methods_done flags to make sure that each global bootmeth is
only used once. For now this has no effect, since they are all processed
at the start.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add the logic to scan through the global bootmeths for every new
bootdev, in preparation for allowing global bootmeths to select where in
the hunter ordering they go.
Use a new bootmeth_glob_allowed() function to check if a bootmeth is
allowed, ensuring that each can run at most once.
For now this has no actual effect, since the global bootmeths are
unconditionally processed at the start, with iter->methods_done being
updated to include all of them. Therefore when scanning again, no
unprocessed global bootmeths will be found.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a bitfield which tracks when bootmeths have been used. This will be
needed when global bootmeths can be used later in the iteration.
Fix a missing bootflow_free() while here.
Signed-off-by: Simon Glass <sjg@chromium.org>
The current 'doing_global' refers to being in the state of processing
global bootmeths. Since global bootmeths are currently used once at the
start, it becomes false once the last global bootmeth has been used.
In preparation for allowing bootmeths to run at other points in the
bootstd interation, add a new 'have_global' flag which tracks whether
there are any global bootmeths in the method_order[] list. It is set up
when iteration starts. Unlike doing_global which resets back to false
after the global bootmeths have been handled, once have_global is set to
true, it remains true for the entire iteration process. This provides a
quick check as to whether global-bootmeth processing is needed.
Signed-off-by: Simon Glass <sjg@chromium.org>
For now we only support dropping non-global bootmeths from the
iteration. Update first_glob_method in that case and add a few checks
that things are correct.
Signed-off-by: Simon Glass <sjg@chromium.org>
These have different behaviour from normal bootmeths and we are about to
enhance it. So add a test and also an extra check in bootflow_iter()
Signed-off-by: Simon Glass <sjg@chromium.org>
At present, normally when one bootmeth fails on a partition, we move on
and try the next bootmeth. However, this was not the case for the final
partition due to a bug. Rework the logic so that all partitions are
treated the same.
Signed-off-by: Simon Glass <sjg@chromium.org>
Anshul Dalal <anshuld@ti.com> says:
In U-Boot, TI only provides a single memory map for all k3 platforms, this
does not scale for devices where atf and optee lie outside the range 0x80000000
- 0x80080000 and 0x9e780000 - 0xa0000000 respectively.
There are also issues for devices with < 2GiB of memory (eg am62SiP with 512MiB
of RAM) as the maximum size for the first DRAM bank is hardcoded to 2GiB in the
current memory map. Furthermore the second DRAM bank is mapped even for devices
that only have a single bank.
Therefore this patch set adds the required functionality to create the MMU table
at runtime based on the device-tree.
The patch set has been build tested on all effected platforms but boot-tested
only on TI's K3 EVMs, the beagleplay and phytec's phycore-am6* platforms.
The following effected boards have not been boot tested:
- verdin-am62
- iot2050
Link: https://lore.kernel.org/r/20251017131540.3636067-1-anshuld@ti.com
With the memory map configuration being done dynamically, reserve extra
space during U-Boot relocation to ensure we have enough for the fixups.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
K3 platforms have reserved memory regions for TFA and OPTEE which should
be unmapped for U-Boot. While other "no-map" memory regions like the
memory pools for remote cores should not be unmapped to allow U-Boot to
load firmware during remoteproc.
Therefore this patch adds the necessary fdt fixups to properly set the
load address for TFA/OPTEE and unmaps both by mmu_unmap_reserved_mem.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
The boot flow in K3 platforms requires the subsequent boot stages to be
aware of the memory reserved for previous boot binaries. In the regular
boot sequence of:
R5 SPL -> ATF -> OP-TEE -> A-core SPL -> U-Boot -> Kernel,
Both A-core SPL and U-Boot should be made aware of the memory reserved
for ATF and OP-TEE from their device-tree. Currently this information is
absent.
Therefore this patch adds the reserved-memory regions as part of
spl_perform_arch_fixups for the next stage's DT. This is called during
both R5 and A-core SPL which fixes up the DT for their respective next
stages: A-core SPL and U-Boot proper.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
The current spl_board_fixups API allows for modification of spl_image
before the SPL jumps to it. This can be used to modify the DT for the
next boot stage, however the current API only allows either the machine
arch or the board to use it.
This limits the utility of the API as there might be certain fixups that
should be applied to all boards sharing the same machine architecture
with others being board specific.
For TI's K3 specifically, this prevents us from performing architecture
level fixups since a lot of TI boards are already making use of the
spl_board_fixups API.
Therefore this patch splits the API into two to allow both board and the
architecture specific fixups. The order is kept as arch then board to
give board specific fixups the precedence.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
For armv8, U-Boot uses a static map defined as 'mem_map' for configuring
the MMU's page tables, done by mmu_setup.
Though this works well for simpler platforms, it makes creating runtime
carveouts by modifying the static array at runtime exceedingly complex
like in mach-snapdragon/board.c.
Creation of such carveouts are much better handled by APIs such as
mmu_change_region_attr once the page tables are configured. Usually such
carveouts are configured via the device-tree's reserved-memory node
which provides the address and size for the carveout.
Therefore this patch adds mmu_unmap_reserved_mem which acts as a wrapper
over mmu_change_region_attr, helping unmap a reserved-memory region.
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
The static memory map for K3 (k3_mem_map) only maps the first DRAM bank
and therefore doesn't scale for platforms with multiple memory banks.
This patch modifies enable_caches to add mem_map_from_dram_banks which
appends all the memory banks to k3_mem_map before calling mmu_setup.
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
For armv8, U-Boot uses a static map defined as 'mem_map' for configuring
the MMU as part of mmu_setup.
But since the exact configuration of memory banks might not be known at
build time, many platforms such as imx9, versal2 etc. utilize
gd->bd->bi_dram to configure the static map at runtime.
Therefore this patch adds a new API mem_map_from_dram_banks that
modifies the static map in a similar way. Allowing the caller to map all
dram banks by just passing the index to last entry in their mem_map and
it's length.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
In dcache_enable, currently the dcache entries are only invalidated when
the MMU is not enabled. This causes issues when dcache_enable is called
with the MMU already configured, in such cases the existing dcache
entries are not flushed which might result in un-expected behavior.
This patch invalidates the cache entries on every call of dcache_enable
before enabling dcache (by setting CR_C). This makes dcache_enable
behave similar to icache_enable as well.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
The mmu_setup function configures the page tables based on the board
supplied mem_map struct array. It is called implicitly as part of
dcache_enable but this limits us to only be able to use APIs such as
mmu_change_region_attr only after caches are enabled.
This might lead to speculative accesses before we can unmap a region
that is marked as cacheable in the static memory map.
Therefore this patch exports the mmu_setup function in mmu.h allowing
users to have more control over when the mmu is configured.
For K3 specifically this allows for the following configuration sequence
as part of enable_caches:
static mem_map fixups (TODO) -> mmu_setup -> carveouts using
mmu_change_region_attr (TODO) -> icache/dcache enable
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Reviewed-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
U-Boot's provided enable_caches enforces the following sequence:
icache_enable -> mmu_setup (as part of dcache_enable) -> dcache_enable
Whereas for K3 devices, we would like to add entries to the provided
static array (k3_mem_map) as per gd->bd->bi_dram and then call mmu_setup
to configure the MMU but also create carveouts for TFA/TEE before we
enable caches to prevent speculative accesses to the region. Thus the
following desired sequence:
add dram banks -> mmu_setup -> carveout TFA/TEE -> icache/dcache enable
Therefore this patch adds K3's own implementation of enable_cache for
ARM64, allowing for greater control over the cache enablement sequence.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
The K3 family of SoCs encompasses a wide variety of devices with varying
DDR configurations and memory carveout requirements, the current static
memory map provides basic support for TI EVMs but does not scale well
for newer platforms (such as AM62SiP with 512MiB of RAM).
Therefore this patch replaces the existing memory map with a minimal
one, that could be more easily modified at runtime.
Reviewed-by: Dhruva Gole <d-gole@ti.com>
Signed-off-by: Anshul Dalal <anshuld@ti.com>
Tested-by: Wadim Egorov <w.egorov@phytec.de>
Pull request net-20251022
net:
- airoha: improvements
- Tighten a few more driver dependencies
- designware: fix bitbang init error
- phy: Make driver overloading get_phy_id depend on !COMPILE_TEST
- phy: add paged PHY register accessors
- make dhcp_run() common for NET and NET_LWIP
- dwc_eth_ops: Correct check for FDT_64BIT
- mediatek: mt7988: various fixup + MDIO detach
- phy: aquantia: switch to use phy_get_ofnode(), fix bindings typo
net-legacy:
- bootp: Prevent buffer overflow to avoid leaking the RAM content
- tftp: make TFTP ports unconditionally configurable
misc:
- uthreads: Make use of CONFIG_IS_ENABLED consistently
First set of u-boot-at91 features for the 2026.01 cycle:
This small fixes set includes a change on clocks register value and a
fix for the sam9x60ek default timer.
The Synchronous Abort and reset errors occurred due
to incorrect parameter passing during initialization.
Signed-off-by: Jim Liu <JJLIU0@nuvoton.com>
[jf: add missing #if IS_ENABLED(CONFIG_BITBANGMII)]
Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
With commit 597fe041a8 ("net/phy: enable get_phy_id redefinable") we
made get_phy_id a public but weak function, so that PHY drivers that had
required non-standard ways of getting the PHY ID could be supported.
However, overloading a weak function multiple times is (rightly) a link
error. At this point, we have two PHYs which make use of this feature,
so make both of them only available when COMPILE_TEST is unset, as part
of being able to support "allyesconfig" in the future.
Signed-off-by: Tom Rini <trini@konsulko.com>
CVE-2024-42040 describes a possible buffer overflow when calling
bootp_process_vendor() in bootp_handler() since the total length
of the packet is passed to bootp_process_vendor() without being
reduced to len-(offsetof(struct bootp_hdr,bp_vend)+4).
The packet length is also checked against its minimum size to avoid
reading data from struct bootp_hdr outside of the packet length.
Signed-off-by: Paul HENRYS <paul.henrys_ext@softathome.com>
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
Synchronize paged PHY helpers with Linux v6.17.
Add support for PHY devices that use paged register access by
implementing the following functions:
- phy_save_page(): Save current page number
- phy_select_page(): Switch to a specific page and return previous page
- phy_restore_page(): Restore previously saved page
Also adds read_page and write_page callbacks to the phy_driver
structure to enable driver-specific page handling.
These helpers allow safe access to paged PHY registers by ensuring
proper page selection and restoration,
even in error conditions, which will be used by the Airoha PHY driver.
Signed-off-by: Lucien.Jheng <lucienzx159@gmail.com>
According to commit 997786bbf4 ("drivers/net/airoha_eth: fix stalling
in package receiving") the minimal possible value of SYS_RX_ETH_BUFFER
is 4. Unfortunately it's too small for reliable ping.
Observations shows that SYS_RX_ETH_BUFFER must be at least 8.
Signed-off-by: Mikhail Kshevetskiy <mikhail.kshevetskiy@iopsys.eu>
The commit 997786bbf4 ("drivers/net/airoha_eth: fix stalling in package
receiving") can be improved. Instead of returning previous descriptor
it's possible:
* do nothing in even descriptor case
* return 2 descriptor to the queue (current and previous) in the odd
descriptor case.
This patch:
* implements above approach
* remove logic not required within new approach
* adds note that PKTBUFSRX must be even and larger than 7
for reliable driver operations
Signed-off-by: Mikhail Kshevetskiy <mikhail.kshevetskiy@iopsys.eu>
There are currently two implementations of dhcp_run(): one in cmd/net.c
for NET and one in net/lwip/dhcp.c for NET_LWIP. There is no
justification for that. Therefore, move the NET version into
net/net-common.c to be used by both stacks, and drop the NET_LWIP
version which by the way does not look totally correct.
Signed-off-by: Jerome Forissier <jerome.forissier@linaro.org>
Suggested-by: Tom Rini <trini@konsulko.com>
Acked-by: Benjamin Hahn <B.Hahn@phytec.de>
In preparation for support of MDIO on AN7581, move the MT7531 MMIO logic
to a dedicated driver and permit usage of the mdio read/write function
to the mtk_eth driver.
This only affect MT7988 that can use MMIO operation to access the Switch
register. The MT7988 code is updated to make use of the external driver.
This permits Airoha driver to make use of DM_MDIO to bind for the MT7531
driver that have the same exact register.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
Correctly free the MDIO Bus on calling cleanup function. While at it
also fix a copy-paste error and rename the cleanup function name to the
more specific name.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
On exiting the phy_setting function for MT7988, the PHY page is never
restored to Page 0. This can cause all kind of problem with reading the
status of the PHY at runtime.
Correctly restore PHY page on exiting the PHY setting function.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
The PHY for MT7531/MT7988 are never actully turned ON/OFF for the
affected PHY as we are read/writing to the wrong PHY address.
This is caused by the fact that we use the MT753X_PHY_ADDR macro 2
times offsetting the address multiple times.
One in the _setup() function and one in the mt7531_mii_read/write.
Drop the additional usage of MT753X_PHY_ADDR in setup() to correctly
set the PHY.
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
In this case, the mediatek network drivers cannot build outside of
ARCH_MEDIATEK or ARCH_MTMIPS, and so express this requirement in Kconfig
as well. In the case of DWC_ETH_XGMAC / DWC_ETH_XGMAC_SOCFPGA, the file
controlled by the DWC_ETH_XGMAC option references a socfpga-specific
array defined in the file controlled by DWC_ETH_XGMAC_SOCFPGA. Rework
these options in Kconfig to handle this dependency.
Signed-off-by: Tom Rini <trini@konsulko.com>
It has been over a decade since we had a platform that implemented the
bootp vendor extension support hook. Remove this option due to lack of
use.
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Jerome Forissier <jerome.forissier@linaro.org>
The support found under SYS_FAULT_ECHO_LINK_DOWN requires that the
SYS_FAULT_MII_ADDR symbol also be set. This wasn't previously found in
Kconfig, so add it now.
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Jerome Forissier <jerome.forissier@linaro.org>
Some drivers which depend on SoCFPGA specific headers had not had
appropriate dependencies list in Kconfig. Add ARCH_SOCFPGA or
TARGET_SOCFPGA_SOC64 where appropriate.
Signed-off-by: Tom Rini <trini@konsulko.com>
We do not yet support UTHREADS in xPL phases. However, we have the need
to dummy out certain functions so that xPL can build when full U-Boot
has UTHREADS enabled. Update the few places that need to use
CONFIG_IS_ENABLED so that we have the correct dummy in xPL.
Signed-off-by: Tom Rini <trini@konsulko.com>
Acked-by: Jerome Forissier <jerome.forissier@linaro.org>
A few lines of code being guarded by the CONFIG_TFTP_PORT option seems
an unnecessary restriction on the TFTP support provided by a vanilla
U-Boot image. In cases where the TFTP server cannot run as superuser -
and hence cannot run on the well-known port 69 - this quirk incurs a
full reconfiguration and rebuild of the bootloader only in order to
select the appropriate destination port.
Remove the CONFIG_TFTP_PORT option entirely and make the tftpdstp and
tftpsrcp variables always have an effect. Their being unset will mean
that U-Boot behaves the same as if CONFIG_TFTP_PORT was unset. Update
the documentation accordingly. And fix up the single board which was
originally enabling this option.
Signed-off-by: Alvin Šipraga <alvin@pqrs.dk>
Reviewed-by: Quentin Schulz <quentin.schulz@cherry.de>
The email forwarder I'm using has run into severe problems with Gmail
lately. Switch over to my kernel.org address.
Signed-off-by: Chen-Yu Tsai <wens@kernel.org>
Acked-by: Jernej Skrabec <jernej.skrabec@gmail.com>
The A/B update strategy of the env's has a gap in the first 2 calls of saveenv.
The env's are stored twice on the first memory area if:
gd->env_valid == ENV_INVALID.
u-boot=> saveenv
Saving Environment to MMC... Writing to MMC(1)... OK
u-boot=> saveenv
Saving Environment to MMC... Writing to MMC(1)... OK <-- !!!
u-boot=> saveenv
Saving Environment to MMC... Writing to redundant MMC(1)... OK
u-boot=> saveenv
Saving Environment to MMC... Writing to MMC(1)... OK
This is the same issue as resolved in commit e589d5822c ("env: spi:
Fix gd->env_valid for the first write")
Signed-off-by: Michael Glembotzki <Michael.Glembotzki@iris-sensing.com>
Signed-off-by: Jasper Orschulko <jasper@fancydomain.eu>
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
[trini: Amend the commit message]
---
Changes in v2:
- Rebase to current master
- Amend the commit message to reference the SPI version of this fix,
which had significant follow-up discussion.
When falcon mode is enabled and SPL_OS_BOOT_SECURE is not enabled,
restore the previous default behavior of having an args file be
expected. Platforms which are using a FIT image here and do not need
this can update at their convenience to disable this option now.
Fixes: b1a3ed0688 ("spl: make args file optional in falcon mode")
Signed-off-by: Tom Rini <trini@konsulko.com>
Anshul Dalal <anshuld@ti.com> says:
Continuing from the last series[1], this patch series addresses the requirement
to disable the args file in falcon mode.
The args file is used in falcon mode for loading the device-tree for the kernel.
However in secure falcon mode, the expected payload is a FIT containing a signed
device-tree and kernel image. Thus removing the need to load the extra args
file in the first place. Also, loading the extra file without any authentication
mechanism exposes an attack vector and should therefore be disabled to keep the
boot secure.
This patch set builds on the last few to first optionally allow for loading the
args file in non-secure falcon boot flow [1/3] and then disable them altogether
in the next patch [2/3] for secure falcon mode.
[1]: https://lore.kernel.org/u-boot/20251006101057.4172248-1-anshuld@ti.com/
Link: https://lore.kernel.org/r/20251009115846.897186-1-anshuld@ti.com
