Set the defaults on probe for the packet buffer size registers
for the i210.
The TX/RX PBSIZE register of the i210 resets to its default value
only at power-on - see Intel Ethernet Controller I210 Datasheet rev 3.5
chapter 8.3 'Internal Packet Buffer Size Registers'.
If something (another driver, another OS, etc.) modifies this register
from its default value, the e1000 driver doesn't function correctly. It
detects a hang of the transmitter and continuously resets the adapter.
Here we set this value to its default when resetting the i210 to
resolve this issue.
Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
checkpatch.pl: Make CONFIG_IS_ENABLED(CONFIG_*) an error
CONFIG_IS_ENABLED() takes the kconfig name without the CONFIG_ prefix,
e.g. CONFIG_IS_ENABLED(CLK) for CONFIG_CLK. Make including the prefix
an error in checkpatch.pl so calls in the wrong format aren't
accidentally reintroduced.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
CONFIG_IS_ENABLED() takes the kconfig name without the CONFIG_ prefix,
e.g. CONFIG_IS_ENABLED(CLK) for CONFIG_CLK. Some of these were being
fixed every now and then, see:
Sean Anderson [Wed, 9 Sep 2020 20:24:56 +0000 (16:24 -0400)]
time: Fix get_ticks being non-monotonic
get_ticks does not always succeed. Sometimes it can be called before the
timer has been initialized. If it does, it returns a negative errno.
This causes the timer to appear non-monotonic, because the value will
become much smaller after the timer is initialized.
No users of get_ticks which I checked handle errors of this kind. Further,
functions like tick_to_time mangle the result of get_ticks, making it very
unlikely that one could check for an error without suggesting a patch such
as this one.
This patch panics if we ever get an error. There are two cases in which
this can occur. The first is if we couldn't find/probe the timer for some
reason. One reason for this is if the timer is not available so early. This
likely indicates misconfiguration. Another reason is that the timer has an
invalid/missing device tree binding. In this case, panicing is also
correct. The second case covers errors calling get_count. This can only
occur if the timer is missing a get_count function (or on RISC-V, but that
should be fixed soon).
Fixes: c8a7ba9e6a5 Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
David Woodhouse [Tue, 4 Aug 2020 09:05:47 +0000 (10:05 +0100)]
env/ext4.c: allow loading from an EXT4 partition on the MMC boot device
This parallels what I added for FAT in commit 6731bef6966, allowing the
environment to be found in a specific partition on the device that the
board's mmc_get_env_dev() returns. On the Banana Pi R2 that means the
device that U-Boot was loaded from; either the internal eMMC or an SD
card.
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Frédéric Danis [Fri, 20 Mar 2020 09:59:24 +0000 (10:59 +0100)]
cmd: Fixup DT to pass PStore Ramoops parameters
To simplify configuration and keep synchronized the PStore/Ramoops between
U-Boot and the Linux kernel, this commit dynamically adds the Ramoops
parameters defined in the U-Boot session to the Device Tree.
Signed-off-by: Frédéric Danis <frederic.danis@collabora.com> Cc: Tom Rini <trini@konsulko.com> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
Frédéric Danis [Fri, 20 Mar 2020 09:59:23 +0000 (10:59 +0100)]
test: Add PStore command tests
Add PStore command to sandbox and sandbox64 defconfigs.
Add test checking:
- 'pstore display' of all records
- 'pstore display' only the 2nd dump record
- 'pstore save' of all records
Signed-off-by: Frédéric Danis <frederic.danis@collabora.com>
[trini: Adjust to always load files from source directory] Signed-off-by: Tom Rini <trini@konsulko.com> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
Frédéric Danis [Fri, 20 Mar 2020 09:59:22 +0000 (10:59 +0100)]
cmd: Add command to display or save Linux PStore dumps
This patch adds a new pstore command allowing to display or save ramoops
logs (oops, panic, console, ftrace and user) generated by a previous
kernel crash.
PStore parameters can be set in U-Boot configuration file, or at run-time
using "pstore set" command. Records size should be the same as the ones
used by kernel, and should be a power of 2.
This command allows:
- to display uncompressed logs
- to save compressed or uncompressed logs, compressed logs are saved as a
compressed stream, it may need some work to be able to decompress it,
e.g. adding a fake header:
"printf "\x1f\x8b\x08\x00\x00\x00\x00\x00\x00\x00" |
cat - dmesg-ramoops-0.enc.z | gzip -dc"
- ECC part is not used to check memory corruption
- only 1st FTrace log is displayed or saved
Signed-off-by: Frédéric Danis <frederic.danis@collabora.com>
[trini: Minor updates for current design, correct spacing in rST] Signed-off-by: Tom Rini <trini@konsulko.com> Cc: Heinrich Schuchardt <xypron.glpk@gmx.de> Cc: Wolfgang Denk <wd@denx.de> Cc: Heiko Schocher <hs@denx.de>
- Fix verified boot on BE targets
- Add support for multiple required keys in verified boots
- Add support for Initialization Vectors in AES keys in FIT images
- Assorted fixes in the RSA code
rsa: fix retrieving public exponent on big-endian systems
Commit fdf0819afb (rsa: fix alignment issue when getting public
exponent) changed the logic to avoid doing an 8-byte access to a
possibly-not-8-byte-aligned address.
However, using rsa_convert_big_endian is wrong: That function converts
an array of big-endian (32-bit) words with the most significant word
first (aka a BE byte array) to an array of cpu-endian words with the
least significant word first. While the exponent is indeed _stored_ as
a big-endian 64-bit word (two BE words with MSW first), we want to
extract it as a cpu-endian 64 bit word. On a little-endian host,
swapping the words and byte-swapping each 32-bit word works, because
that's the same as byte-swapping the whole 64 bit word. But on a
big-endian host, the fdt32_to_cpu are no-ops, but
rsa_convert_big_endian() still does the word-swapping, breaking
verified boot.
To fix that, while still ensuring we don't do unaligned accesses, add
a little helper that first memcpy's the bytes to a local fdt64_t, then
applies fdt64_to_cpu(). [The name is chosen based on the
[bl]eXX_to_cpup in linux/byteorder/generic.h].
Fixes: fdf0819afb ("rsa: fix alignment issue when getting public exponent") Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Simon Glass <sjg@chromium.org>
lib: rsa: check algo match in rsa_verify_with_keynode
The algo name should match between the FIT's signature node and the
U-Boot's control FDT.
If we do not check it, U-Boot's control FDT can expect sha512 hash but
nothing will prevent to accept image with sha1 hash if the signature is correct.
Philippe Reynes [Thu, 17 Sep 2020 13:01:46 +0000 (15:01 +0200)]
fit: cipher: aes: allow to store the IV in the FIT image
Binaries may be encrypted in a FIT image with AES. This
algo needs a key and an IV (Initialization Vector). The
IV is provided in a file (pointer by iv-name-hint in the
ITS file) when building the ITB file.
This commits adds provide an alternative way to manage
the IV. If the property iv-name-hint is not provided in
the ITS file, the tool mkimage will generate an random
IV and store it in the FIT image.
Signed-off-by: Philippe Reynes <philippe.reynes@softathome.com>
This patch adds vboot tests to verify the support for multiple
required keys using new required-mode DTB policy.
This patch also fixes existing test where dev
key is assumed to be marked as not required, although
it is marked as required.
Note that this patch re-added sign_fit_norequire().
sign_fit_norequire() was removed as part of the following:
commit b008677daf2a ("test: vboot: Fix pylint errors").
This patch leverages sign_fit_norequire() to fix the
existing bug.
Signed-off-by: Thirupathaiah Annapureddy <thiruan@linux.microsoft.com> Reviewed-by: Simon Glass <sjg@chromium.org>
vboot: add DTB policy for supporting multiple required conf keys
Currently FIT image must be signed by all required conf keys. This means
Verified Boot fails if there is a signature verification failure
using any required key in U-Boot DTB.
This patch introduces a new policy in DTB that can be set to any required
conf key. This means if verified boot passes with one of the required
keys, U-Boot will continue the OS hand off.
There were prior attempts to address this:
https://lists.denx.de/pipermail/u-boot/2019-April/366047.html
The above patch was failing "make tests".
https://lists.denx.de/pipermail/u-boot/2020-January/396629.html
Signed-off-by: Thirupathaiah Annapureddy <thiruan@linux.microsoft.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Mon, 12 Oct 2020 11:26:57 +0000 (07:26 -0400)]
Merge tag 'ti-v2021.01-rc1' of https://gitlab.denx.de/u-boot/custodians/u-boot-ti
- Minor cleanup on K3 env variables
- Fix OSPI compatible for J721e
- Drop unused property in omap-usb2-phy
- Update Maintainer for am335x-guardian board.
phy: omap-usb2-phy: Drop usage of "ti, dis-chg-det-quirk" DT property
"ti,dis-chg-det-quirk" property is not part of Linux kernel DT binding
documentation. Therefore drop this and instead use soc_device_match()
to distinguish b/w AM654 SR1.0 and SR2.0 devices similar to Linux kernel
driver.
Nishanth Menon [Thu, 8 Oct 2020 02:48:01 +0000 (21:48 -0500)]
configs: am65x_evm: Use DEFAULT_LINUX_BOOT_ENV and remove overlayaddr
Use DEFAULT_LINUX_BOOT_ENV to define the standard addresses used in rest
of TI platforms as defined in ti_armv7_common.h
This avoids the standard pitfalls we've had with kernel images and fdt
addresses stomping on each other.
As part of this process, redefine overlayaddr to be dtboaddr (defined
in ti_armv7_common.h for this very purpose) and get rid of the
definition of overlayaddr..
Nishanth Menon [Thu, 8 Oct 2020 02:47:58 +0000 (21:47 -0500)]
configs: j721e_evm: Use DEFAULT_LINUX_BOOT_ENV
Use DEFAULT_LINUX_BOOT_ENV to define the standard addresses used in rest
of TI platforms as defined in ti_armv7_common.h
This avoids the standard pitfalls we've had with kernel images and fdt
addresses stomping on each other.
As part of this process, redefine overlayaddr to be dtboaddr (defined
in ti_armv7_common.h for this very purpose).. we will get rid of
overlayaddr later in the series.
Sean Anderson [Sat, 12 Sep 2020 21:45:43 +0000 (17:45 -0400)]
net: Expose some errors generated in net_init
net_init does not always succeed, and there is no existing mechanism to
discover errors. This patch allows callers of net_init (such as net_init)
to handle errors. The root issue is that eth_get_dev can fail, but
net_init_loop doesn't expose that. The ideal way to fix eth_get_dev would
be to return an error with ERR_PTR, but there are a lot of callers, and all
of them just check if it's NULL. Another approach would be to change the
signature to something like
int eth_get_dev(struct udevice **pdev)
but that would require rewriting all of the many callers.
Signed-off-by: Sean Anderson <seanga2@gmail.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Simon Glass [Sat, 12 Sep 2020 18:28:50 +0000 (12:28 -0600)]
log: Disable the syslog driver by default
This driver interferes with other sandbox tests since it causes log output
to be interspersed with "No ethernet found." messages. Disable this driver
by default.
Enable it for the syslog tests so that they still pass.
Simon Glass [Sat, 12 Sep 2020 18:28:47 +0000 (12:28 -0600)]
log: Add a flag to enable log drivers
At present there is no way to disable a log driver. But the syslog driver
causes (attempted) network traffic in sandbox every time a log message
is printed, which is often.
Add a flag to enable a log driver. Adjust struct log_device to use a short
for next_filter_num so that no more memory is used for devices. Also fix
a missing line in the struct log_driver comment while here.
To maintain compatibility, enable it for all drivers for now.
Simon Glass [Sat, 12 Sep 2020 17:13:34 +0000 (11:13 -0600)]
log: Allow LOG_DEBUG to always enable log output
At present if CONFIG_LOG enabled, putting LOG_DEBUG at the top of a file
(before log.h inclusion) causes _log() to be executed for every log()
call, regardless of the build- or run-time logging level.
However there is no guarantee that the log record will actually be
displayed. If the current log level is lower than LOGL_DEBUG then it will
not be.
Add a way to signal that the log record should always be displayed and
update log_passes_filters() to handle this.
With the new behaviour, log_debug() will always log if LOG_DEBUG is
enabled.
Move log_test_syslog_nodebug() into its own file since it cannot be made
to work where it is, with LOG_DEBUG defined.
Simon Glass [Fri, 11 Sep 2020 02:21:27 +0000 (20:21 -0600)]
Kconfig: Create a new tools menu
At present MKIMAGE_DTC_PATH is in the devicetree menu but not within
'devicetree control' since it does not relate to that. As a result it
shows up in the top menu.
It actually relates to the mkimage tool, so create a new tools menu for it
and move it there.
Simon Glass [Fri, 11 Sep 2020 02:21:16 +0000 (20:21 -0600)]
Kconfig: Move autoboot options under boot options
At present the autoboot options are in cmd/Kconfig but they don't really
relate to commands. They relate to booting, so move this menu under the
boot menu.
Tom Rini [Fri, 9 Oct 2020 12:58:56 +0000 (08:58 -0400)]
Merge branch '2020-10-08-misc-board-improvements'
- Move ASPEED ram driver, update.
- Exhance pinctrl/gpio support, update Kendryte K210 support
- Enhance qemu_arm64 support for a single binary to work with and
without TF-A
Chee Hong Ang [Tue, 11 Aug 2020 13:52:30 +0000 (21:52 +0800)]
Makefile: socfpga: Generate sfp file with 4 SPL images
Generate 'u-boot-splx4.sfp' which consist of 4 SPL images required
for booting up Cyclone5/Arria10.
By default, this 'u-boot-splx4.sfp' is generated without extra
padding after each SPL image.
For Cyclone5, 'u-boot-splx4.sfp' contains:
4 x SPL(64KB) = 256KB
For Arria10, 'u-boot-splx4.sfp' contains:
4 x SPL(256KB) = 1024KB
For Cyclone5 using NAND flash image layout for 128 KB memory blocks,
user can 'make' the following target to generate 4 SPL images with
padding:
make u-boot-spl-padx4.sfp
'u-boot-spl-padx4.sfp' contains four 128KB SPL images (each 64KB SPL is
followed by 64KB of zero-padding).
4 x (SPL(64KB) + zero-padding(64KB)) = 512KB
Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com> Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chee Hong Ang [Thu, 1 Oct 2020 09:15:59 +0000 (02:15 -0700)]
arm: socfpga: soc64: Remove CONFIG_OF_EMBED
CONFIG_OF_EMBED was primarily enabled to support the S10/Agilex
spl hex file requirements. Since this option now produces a
warning during build, and the spl hex can be created using
alternate methods, CONFIG_OF_EMBED is no longer needed.
Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com>
arm: dts: socfpga: arria10: Move to use generic handoff dtsi
Move to use generic handoff dtsi (socfpga_arria10-handoff.dtsi) and include
the specify generated _handoff.h header file from qts-filter-a10.sh script.
Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
arm: socfpga: arria10: Add qts-filter for Arria10 socfpga
Add a script to process HPS handoff data and generate a header
for inclusion in u-boot specific devicetree addons. The header
should be included in the top level of u-boot.dtsi.
Signed-off-by: Dalon Westergreen <dalon.westergreen@intel.com> Signed-off-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chin Liang See [Mon, 10 Aug 2020 02:55:56 +0000 (10:55 +0800)]
arm: socfpga: soc64: Document down boot_scratch_cold register usage
Document down the usage of boot_scratch_cold register to avoid
overlapping of usage in the code for S10 & Agilex.
The boot_scratch_cold register is generally used for passing
critical system info between SPL, U-Boot and Linux.
Signed-off-by: Chin Liang See <chin.liang.see@intel.com> Signed-off-by: Siew Chin Lim <elly.siew.chin.lim@intel.com> Reviewed-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chee Hong Ang [Fri, 7 Aug 2020 03:50:03 +0000 (11:50 +0800)]
fpga: altera: Rename Stratix10 FPGA to Intel FPGA SDM Mailbox
Rename Stratix10 FPGA driver to Intel FPGA SDM Mailbox driver
because it is using generic SDM (Secure Device Manager) Mailbox
interface shared by other platform (e.g. Agilex) as well.
Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com> Reviewed-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chee Hong Ang [Wed, 5 Aug 2020 13:15:56 +0000 (21:15 +0800)]
arm: socfpga: soc64: Add SDM triggered warm reset bit mask
Include SDM triggered warm reset bit (BIT1) in Reset Manager's stat
register when checking for HPS warm reset status.
Refactor the warm reset mask macro for clarity purpose.
Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com> Reviewed-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chin Liang See [Wed, 5 Aug 2020 10:34:33 +0000 (18:34 +0800)]
configs: socfpga: soc64: Avoid SPL enter infinite loop during exception
In current implementation, any exception would trigger a CPU reset.
But a bad written SPL would cause infinite loop where the system
will reload the same SPL instead of loading factory safe image.
Hence this patch is to ensure any exception will cause a hang. At this
moment, watchdog shall be triggered and Remote System Update mechanism
shall load the next production image or factory safe image.
Signed-off-by: Chin Liang See <chin.liang.see@intel.com> Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com> Reviewed-by: Ley Foon Tan <ley.foon.tan@intel.com>
Chee Hong Ang [Fri, 10 Jul 2020 12:55:22 +0000 (20:55 +0800)]
clk: agilex: Handle clock configuration differently in SPL and U-Boot proper
Since warm reset may optionally set the CLock Manager to'boot mode',
the clock driver should always force the Agilex's Clock Manager to
'boot mode' before the clock driver start configuring the Clock Manager
in SPL.
In SSBL, clock driver will skip the Clock Manager configuration
if it's already being setup by SPL (Clock Manager NOT in 'boot
mode') to prevent any inaccurate clocking issues happened on HPS
peripherals such as UART, MAC and etc.
Signed-off-by: Chee Hong Ang <chee.hong.ang@intel.com>
projects / u-boot.git / log