Tom Rini [Sat, 29 Oct 2022 00:27:10 +0000 (20:27 -0400)]
Convert CONFIG_SYS_NONCACHED_MEMORY to Kconfig
This converts the following to Kconfig:
CONFIG_SYS_NONCACHED_MEMORY
To do this we introduce CONFIG_SYS_HAS_NONCACHED_MEMORY as a bool to
gate if we are going to have noncached_... functions available and then
continue to use CONFIG_SYS_NONCACHED_MEMORY to store the size of said
cache. We make this new option depend on both the architectures which
implement support and the drivers which make use of it.
Cc: Tom Warren <twarren@nvidia.com> Cc: Mingming lee <mingming.lee@mediatek.com> Cc: "Ying-Chun Liu (PaulLiu)" <paul.liu@linaro.org> Cc: Alban Bedel <alban.bedel@avionic-design.de> Cc: Stephen Warren <swarren@nvidia.com> Cc: Marcel Ziswiler <marcel.ziswiler@toradex.com> Cc: Allen Martin <amartin@nvidia.com> Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sat, 29 Oct 2022 00:27:07 +0000 (20:27 -0400)]
Convert CONFIG_SYS_MONITOR_LEN to Kconfig
This converts the following to Kconfig:
CONFIG_SYS_MONITOR_LEN
To do this, we set a default of 0 for everyone because there are a
number of cases where we define CONFIG_SYS_MONITOR_LEN but the only
impact is that we set TOTAL_MALLOC_LEN to be CONFIG_SYS_MALLOC_LEN +
CONFIG_ENV_SIZE, so we must continue to allow all boards to set this
value. Update the SPL code to use 200 KB as the default raw U-Boot size
directly, if we don't have a real CONFIG_SYS_MONITOR_LEN value.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sat, 29 Oct 2022 00:27:02 +0000 (20:27 -0400)]
Convert CONFIG_SYS_LOADS_BAUD_CHANGE et al to Kconfig
This converts the following to Kconfig:
CONFIG_SYS_LOADS_BAUD_CHANGE
CONFIG_LOADS_ECHO
As part of this, we move CMD_SAVES to be after CMD_LOADS as they are
logically related (load or save an s-record format file) and this makes
grouping of CONFIG_SYS_LOADS_BAUD_CHANGE easier.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sat, 29 Oct 2022 00:26:59 +0000 (20:26 -0400)]
fs: jffs2: Move SYS_JFFS2_SORT_FRAGMENTS to Kconfig
Move the symbol SYS_JFFS2_SORT_FRAGMENTS to Kconfig and use the only
remaining part of doc/README.JFFS2 that is still relevant and useful to
the help for this option.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Tom Rini [Sat, 29 Oct 2022 00:26:58 +0000 (20:26 -0400)]
Remove unused symbols CONFIG_SYS_JFFS2_FIRST_BANK et al
This removes the following symbols:
CONFIG_RTC_MCFRRTC
CONFIG_SYS_JFFS2_FIRST_BANK
CONFIG_SYS_JFFS2_FIRST_SECTOR
CONFIG_SYS_JFFS2_NUM_BANKS
CONFIG_SYS_LBC_CACHE_BASE
CONFIG_SYS_LIME_SIZE
CONFIG_SYS_MAMR
CONFIG_SYS_MCFRRTC_BASE
CONFIG_SYS_MONITOR_SEC
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Simon Glass [Sun, 30 Oct 2022 01:47:17 +0000 (19:47 -0600)]
dm: sandbox: Switch over to using the new host uclass
Update the sandbox implementation to use UCLASS_HOST and adjust all
the pieces to continue to work:
- Update the 'host' command to use the new API
- Replace various uses of UCLASS_ROOT with UCLASS_HOST
- Disable test_eficonfig since it doesn't work (this should have a unit
test to allow this to be debugged)
- Update the blk test to use the new API
- Drop the old header file
Unfortunately it does not seem to be possible to split this change up
further.
Simon Glass [Sun, 30 Oct 2022 01:47:15 +0000 (19:47 -0600)]
dm: sandbox: Create a new HOST uclass
Sandbox supports block devices which can access files on the host machine.
At present there is no uclass for this. The devices are attached to the
root devic. The block-device type is therefore set to UCLASS_ROOT which
is confusing.
Block devices should be attached to a 'media' device instead, something
which handles access to the actual media and provides the block driver
for the block device.
Create a new uclass to handle this. It supports two operations, to attach
and detach a file on the host machine.
Simon Glass [Sun, 30 Oct 2022 01:47:14 +0000 (19:47 -0600)]
dm: blk: Tidy up obtaining a block device from its parent
This function now finds its block-device child by looking for a child
device of the correct uclass (UCLASS_BLK). It cannot produce a device of
any other type, so drop the superfluous check.
Provide a version which does not probe the device, since that is often
needed when setting up the device's platdata.
Simon Glass [Sun, 30 Oct 2022 01:47:13 +0000 (19:47 -0600)]
test: Add a way to detect a test that breaks another
When running unit tests, some may have side effects which cause a
subsequent test to break. This can sometimes be seen when using 'ut dm'
or similar.
Add a new argument which allows a particular (failing) test to be run
immediately after a certain number of tests have run. This allows the
test causing the failure to be determined.
Simon Glass [Sun, 30 Oct 2022 01:47:08 +0000 (19:47 -0600)]
dm: test: Clear the block cache after running a test
Some tests access data in block devices and so cause the cache to fill
up. This results in memory being allocated.
Some tests check the malloc usage at the beginning and then again at the
end, to ensure there is no memory leak caused by the test. The block cache
makes this difficult, since the any test may cause entries to be allocated
or even freed, if the cache becomes full.
It is simpler to clear the block cache after each test. This ensures that
it will not introduce noise in tests which check malloc usage.
Add the logic to clear the cache, using the existing blkcache_invalidate()
function. Drop the duplicate code at the same time.
Chris Packham [Sat, 5 Nov 2022 04:24:00 +0000 (17:24 +1300)]
arm: mvebu: Add RD-AC5X board
The RD-AC5X-32G16HVG6HLG-A0 development board main components and
features include:
* Main 12V/54V power supply
* 270 Gbps throughput packet processor on the main board
* DDR4:
* SR1: 2GB DDR4 2400MT/S(1GB x 2 pcs ) with ECC(1GB x 1 pcs)
* SR2: 4GB DDR4 2400MT/S(2GB x 2 pcs ) with ECC(2GB x 1 pcs)
* PCB co-layout with 4GB device to support 8GB (Dual CS) requirement
* 16GB eMMC (Samsung KLMAG1JETD-B041006)
* 16MB SPI NOR(GD25Q127C)
* 32 x 1000 Base-T interfaces
* 16 x 2500 Base-T interfaces
* SR1: 88E2540*4
* SR2: 88E2580*1+88E2540*2
* Six (6) x 25G Base-R SFP28 interfaces
* One (1) x RJ-45 console connector, interfacing to the on board UART
* One (1) x USB Type-A connector, interfacing to the USB 2.0 port (0)
* One (1) x USB Type-mini B connector, interfacing to the USB 2.0 port (1)
* One (1) x RJ-45 1G Base-T Management port, interfacing to the host
port (shared with PCIe) Connected to 88E1512 Gigabit Ethernet Phy
* One (1) x Oculink port, interfacing to the PCIe port for external CPU
connection
* POE 802.3AT support on Port 1 ~ Port 32, 802.3BT support on Port 33 ~
Port 48 (Microsemi PD69208T4, PD69208M or TI TPS2388,TPS23881
solution)
* POE total power budget 780W
* LED interfaces per network port/POE
* LED interfaces (common) showing system status
* PTP TC mode Supported (Reserved M.2 connector to support BC mode)
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Chris Packham [Sat, 5 Nov 2022 04:23:59 +0000 (17:23 +1300)]
arm: mvebu: Support for 98DX25xx/98DX35xx SoC
Add support for the Allecat5/Alleycat5X SoC. These are L3 switches with
an integrated CPU (referred to as the CnM block in Marvell's
documentation). These have dual ARMv8.2 CPUs (Cortex-A55). This support
has been ported from Marvell's SDK which is based on a much older
version of U-Boot.
Signed-off-by: Chris Packham <judge.packham@gmail.com>
Chris Packham [Sat, 5 Nov 2022 04:23:57 +0000 (17:23 +1300)]
usb: ehci: ehci-marvell: Support for marvell,ac5-ehci
Unlike the other 64-bit mvebu SoCs the AlleyCat5 uses the older ehci
block from the 32-bit SoCs. Adapt the ehci-marvell.c driver to cope with
the fact that the ac5 does not have the mbus infrastructure the 32-bit
SoCs have and ensure USB_EHCI_IS_TDI is selected.
Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de>
Chris Packham [Sat, 5 Nov 2022 04:23:56 +0000 (17:23 +1300)]
net: mvneta: Add support for AlleyCat5
Add support for the AlleyCat5 SoC. This lacks the mbus from the other
users of the mvneta.c driver so a new compatible string is needed to
allow for a different window configuration.
Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de>
Chris Packham [Sat, 5 Nov 2022 04:23:55 +0000 (17:23 +1300)]
arm: mvebu: Don't use CONFIG_TIMER on ARM64
The 64-bit mvebu SoCs don't have a suitable timer driver so add a !ARM64
condition to the select.
Fixes: 7b530bb19e ("arm: mvebu: Use CONFIG_TIMER on all MVEBU & KIRKWOOD platforms") Signed-off-by: Chris Packham <judge.packham@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de>
Pali Rohár [Wed, 2 Nov 2022 17:51:28 +0000 (18:51 +0100)]
Makefile: Rename u-boot-spl.kwb to u-boot-with-spl.kwb
File name with pattern u-boot-spl* is used on all places except in kwb
image for binary with SPL-only code. Combined binary with both SPL and
proper U-Boot in other places has file name pattern u-boot-with-spl*.
Make it consistent also for kwb image and rename u-boot-spl.kwb to
u-boot-with-spl.kwb as this image contains both SPL and proper U-Boot code.
Also update documentation about file name changes.
Signed-off-by: Pali Rohár <pali@kernel.org> Reviewed-by: Stefan Roese <sr@denx.de>
Hamish Martin [Thu, 20 Oct 2022 21:21:59 +0000 (10:21 +1300)]
arm: armada: dts: Add clock to armada-ap80x uart1
The uart1 node was missing the 'clock-frequency' property. This meant
the driver for this device would fail at probe.
The clock for uart1 is fed from the same source as uart0 and is a fixed
200MHz clock. This is confirmed via documentation for the CN9130 SoC
and from the equivalent code in Linux at:
<linux>/arch/arm64/boot/dts/marvell/armada-ap80x.dtsi
where uart0 and uart1 share a common 'clocks' definition.
Signed-off-by: Hamish Martin <hamish.martin@alliedtelesis.co.nz> Reviewed-by: Stefan Roese <sr@denx.de>
AllocatePages() can be called with Type=AllocateAddress. Such a call can
only succeed if *Memory points to the address of an unallocated page range.
A call with *Memory being an address that is not page aligned must not
succeed. The UEFI specification requires returning EFI_OUT_OF_RESOURCES
if the requested pages cannot be allocated.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
AKASHI Takahiro [Fri, 21 Oct 2022 03:01:57 +0000 (12:01 +0900)]
efi_loader: remove CONFIG_EFI_SETUP_EARLY
Since the commit a9bf024b2933 ("efi_loader: disk: a helper function to
create efi_disk objects from udevice"), CONFIG_EFI_SETUP_EARLY option is
by default on and will never be turned off.
Jan Kiszka [Fri, 14 Oct 2022 16:10:06 +0000 (18:10 +0200)]
efi_loader: Let networking support depend on NETDEVICES
CONFIG_NET does not imply that there are actually network devices
available, only CONFIG_NETDEVICES does. Changing to this dependency
obsoletes the check in Kconfig because NETDEVICES means DM_ETH.
Fixes: 0efe1bcf5c2c ("efi_loader: Add network access support") Suggested-by: Tom Rini <trini@konsulko.com> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Tom Rini [Thu, 3 Nov 2022 18:25:44 +0000 (14:25 -0400)]
docs: Add a basic security document
Based loosely on the Linux kernel
Documentation/admin-guide/security-bugs.rst file, create a basic
security document for U-Boot. In sum, security issues should be
disclosed in public on the mailing list if at all possible as an initial
position.
Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Marek Vasut [Sun, 30 Oct 2022 22:38:35 +0000 (23:38 +0100)]
usb: Add 1ms delay after first Get Descriptor request
Logitech Unifying Receiver 046d:c52b bcdDevice 12.10 seems
sensitive about the first Get Descriptor request. If there
are any other requests in the same microframe, the device
reports bogus data, first of the descriptor parts is not
sent to the host. Wait over one microframe duration before
issuing subsequent requests to avoid probe failure with
this device, since it can be used to connect USB keyboards.
Signed-off-by: Marek Vasut <marex@denx.de> Tested-by: Janne Grunau <j@jannau.net>
Padmarao Begari [Thu, 27 Oct 2022 06:02:02 +0000 (11:32 +0530)]
riscv: Update Microchip MPFS Icicle Kit support
This patch updates Microchip MPFS Icicle Kit support. For now,
add Microchip QSPI driver and a small 4MB reservation is
made at the end of 32-bit DDR to provide some memory for
the HSS to use.
Padmarao Begari [Thu, 27 Oct 2022 06:01:59 +0000 (11:31 +0530)]
riscv: dts: Update memory configuration
In the v2022.10 Icicle reference design, the seg registers have been
changed, resulting in a required change to the memory map.
A small 4MB reservation is made at the end of 32-bit DDR to provide some
memory for the HSS to use, so that it can cache its payload between
reboots of a specific context.
As PLICSW is used to trigger the software interrupt, we should rename
Andes PLIC configuration and file name to reflect the usage. This patch
also updates PLMT and PLICSW compatible strings to be consistent with
OpenSBI fdt driver.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com> Reviewed-by: Rick Chen <rick@andestech.com>
Stefan Roese [Fri, 28 Oct 2022 12:46:29 +0000 (14:46 +0200)]
mips: mtmips: spl/Kconfig: Set CONFIG_SPL_PAD_TO to 0x0 for ARCH_MTMIPS
It was noticed that while converting CONFIG_SPL_PAD_TO to Kconfig its
value for the MIPS MT762x/8x targets got not ported correctly. Its
default is not 0x10000 instead of 0x0. This patch fixes this issue.
Fixes: ca8a329a1b7f ("Convert CONFIG_SPL_PAD_TO et al to Kconfig") Signed-off-by: Stefan Roese <sr@denx.de> Cc: Ruben Winters <Ruben.Winters@gooiland-elektro.nl> Cc: Weijie Gao <weijie.gao@mediatek.com> Cc: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Cc: Tom Rini <trini@konsulko.com>
Resolve all uses of CONFIG_SYS_MHZ with the currently defined value.
Remove code which depends on CONFIG_SYS_MHZ but where no board configs
actually use that code.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com> Reviewed-by: Stefan Roese <sr@denx.de>
Baruch Siach [Sun, 23 Oct 2022 09:28:12 +0000 (12:28 +0300)]
cmd: eeprom: don't truncate target address at 32-bit
On 64-bit platforms where int is 32-bit wide, the eeprom command
parse_numeric_param() routine truncates the memory address parameter to
the lower 32-bit. Make parse_numeric_param() return long to allow
read/write of addresses beyond the lower 4GB.
Signed-off-by: Baruch Siach <baruch@tkos.co.il> Reviewed-by: Ramon Fried <rfried.dev@gmail.com>
Andre Przywara [Thu, 20 Oct 2022 22:10:25 +0000 (23:10 +0100)]
highbank: switch to use the Arm SP804 DM_TIMER driver
So far the Calxeda machines were using the CONFIG_SYS_TIMER_* macros to
simply hardcode the address of the counter register of the SP804 timer.
This method is deprecated and scheduled for removal.
Use the newly introduced SP804 DM_TIMER driver to provide timer
functionality on Highbank and Midway machines. The base address and base
frequency are taken from the devicetree.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Thu, 20 Oct 2022 22:10:24 +0000 (23:10 +0100)]
highbank: scan into hb_sregs DT subnodes
The DT used for Calxeda Highbank and Midway systems exposes a "system
registers" block, modeled as a DT subnode.
This includes several clocks, including the two fixed clocks for the
main oscillator and timer.
So far U-Boot was ignorant of this special construct (a "clocks" node
within the "hb-sregs" node), as it didn't need the PLL clocks in there.
But that also meant we lost the fixed clocks, which form the base for
the UART baudrate generator and also the SP804 timer.
To allow the generic PL011 and SP804 driver to read the clock rate,
add a simple bus driver, which triggers the DT node discovery inside this
special node. As we only care about the fixed clocks (we don't have
drivers for the PLLs anyway), just ignore the address translation (for
now).
The binding is described in bindings/arm/calxeda/hb-sregs.yaml, the DT
snippet in question looks like:
Andre Przywara [Thu, 20 Oct 2022 22:10:23 +0000 (23:10 +0100)]
timer: add SP804 UCLASS timer driver
The "Arm Ltd. Dual-Timer Module (SP804)" is a simple 32-bit count-down
timer IP with interrupt functionality, and is used in some SoCs from
various vendors.
Add a simple DM compliant timer driver, to allow users of the SP804 to
switch to DM_TIMER.
This relies on the input clock to be accessible via the DM clock
framework, which should be fine as we probably look at fixed-clock's
here anyway.
We re-program the control register in the probe() function, but keep
the divider in place, in case this has been set to something on purpose
before.
The TRM for the timer IP can be found here:
https://developer.arm.com/documentation/ddi0271/latest
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Wed, 5 Oct 2022 16:38:49 +0000 (17:38 +0100)]
arm: smh: Allow semihosting trap calls to be inlined
Currently our semihosting trap function is somewhat fragile: we rely
on the current compiler behaviour to assign the second inline assembly
argument to the next free register (r1/x1), which happens to be the
"addr" argument to the smh_trap() function (per the calling convention).
I guess this is also the reason for the noinline attribute.
Make it explicit what we want: the "addr" argument needs to go into r1,
so we add another register variable. This allows to drop the "noinline"
attribute, so now the compiler beautifully inlines just the trap
instruction directly into the calling function.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Wed, 5 Oct 2022 16:38:48 +0000 (17:38 +0100)]
arm: smh: Make semihosting trap calls more robust
Commit f4b540e25c5c("arm: smh: Fix uninitialized parameters with newer
GCCs") added a memory clobber to the semihosting inline assembly trap
calls, to avoid too eager GCC optimisation: when passing a pointer, newer
compilers couldn't be bothered to actually fill in the structure that it
pointed to, as this data would seemingly never be used (at least from the
compiler's point of view).
But instead of the memory clobber we need to tell the compiler that we are
passing an *array* instead of some generic pointer, this forces the
compiler to actually populate the data structure.
This involves some rather hideous cast, which is best hidden in a macro.
But regardless of that, we actually need the memory clobber, but for two
different reasons: explain them in comments.
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Andre Przywara [Wed, 5 Oct 2022 16:38:47 +0000 (17:38 +0100)]
arm: smh: specify Thumb trap instruction
The ARM semihosting interface uses different trap instructions for
different architectures and instruction sets. So far we were using
AArch64 and ARMv7-M, and had an untested v7-A entry. The latter does
not work when building for Thumb, as can be verified by using
qemu_arm_defconfig, then enabling SEMIHOSTING and SYS_THUMB_BUILD:
==========
{standard input}:35: Error: invalid swi expression
{standard input}:35: Error: value of 1193046 too large for field of 2 bytes at 0
==========
Fix this by providing the recommended instruction[1] for Thumb, and
using the ARM instruction only when not building for Thumb. This also
removes some comment, as QEMU for ARM allows to now test this case.
Also use the opportunity to clean up the inline assembly, and just define
the actual trap instruction inside #ifdef's, to improve readability.
Rasmus Villemoes [Fri, 28 Oct 2022 11:50:54 +0000 (13:50 +0200)]
cyclic: get rid of cyclic_init()
Currently, we must call cyclic_init() at some point before
cyclic_register() becomes possible. That turns out to be somewhat
awkward, especially with SPL, and has resulted in a watchdog callback
not being registered, thus causing the board to prematurely reset.
We already rely on gd->cyclic reliably being set to NULL by the asm
code that clears all of gd. Now that the cyclic list is a hlist, and
thus an empty list is represented by a NULL head pointer, and struct
cyclic_drv has no other members, we can just as well drop a level of
indirection and put the hlist_head directly in struct
global_data. This doesn't increase the size of struct global_data,
gets rid of an early malloc(), and generates slightly smaller code.
But primarily, this avoids having to call cyclic_init() early; the cyclic
infrastructure is simply ready to register callbacks as soon as we
enter C code.
We can still end up with schedule() being called from asm very early,
so we still need to check that gd itself has been properly initialized
[*], but once it has, gd->cyclic_list is perfectly fine to access, and
will just be an empty list.
As for cyclic_uninit(), it was never really the opposite of
cyclic_init() since it didn't free the struct cyclic_drv nor set
gd->cyclic to NULL. Rename it to cyclic_unregister_all() and use that
in test/, and also insert a call at the end of the board_init_f
sequence so that gd->cyclic_list is a fresh empty list before we enter
board_init_r().
A small piece of ugliness is that I had to add a cast in
cyclic_get_list() to silence a "discards 'volatile' qualifier"
warning, but that is completely equivalent to the existing handling of
the uclass_root_s list_head member.
[*] I'm not really sure where we guarantee that the register used for
gd contains 0 until it gets explicitly initialized, but that must be
the case, otherwise testing gd for being NULL would not make much sense.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Tim Harvey <tharvey@gateworks.com> # imx8mm-venice-*
Rasmus Villemoes [Fri, 28 Oct 2022 11:50:53 +0000 (13:50 +0200)]
cyclic: switch to using hlist instead of list
A hlist is headed by just a single pointer, so can only be traversed
forwards, and insertions can only happen at the head (or before/after
an existing list member). But each list node still consists of two
pointers, so arbitrary elements can still be removed in O(1).
This is precisely what we need for the cyclic_list - we never need to
traverse it backwards, and the order the callbacks appear in the list
should really not matter.
One advantage, and the main reason for doing this switch, is that an
empty list is represented by a NULL head pointer, so unlike a
list_head, it does not need separate C code to initialize - a
memset(,0,) of the containing structure is sufficient.
This is mostly mechanical:
- The iterators are updated with an h prefix, and the type of the
temporary variable changed to struct hlist_node*.
- Adding/removing is now just hlist_add_head (and not tail) and
hlist_del().
- struct members and function return values updated.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Tim Harvey <tharvey@gateworks.com> # imx8mm-venice-*
Currently, there is only one "user" of any of these, namely in
fs/ubifs/super.c, but that actually uses the "new-style" form, and
is (obviously, or it wouldn't have built) inside #ifndef __UBOOT__.
Before adding actual users of these, import the version as of linux
v6.1-rc1, including the hlist_entry_safe() helper used by the new
versions.
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Tim Harvey <tharvey@gateworks.com> # imx8mm-venice-*
Rasmus Villemoes [Fri, 28 Oct 2022 11:50:51 +0000 (13:50 +0200)]
cyclic: drop redundant cyclic_ready flag
We're already relying on gd->cyclic being NULL before cyclic_init() is
called - i.e., we're relying on all of gd being zeroed before entering
any C code. And when we do populate gd->cyclic, its ->cyclic_ready
member is automatically set to true. So we can actually just rely on
testing gd->cyclic itself.
The only wrinkle is that cyclic_uninit() actually did set
->cyclic_ready to false. However, since it doesn't free gd->cyclic,
the cyclic infrastructure is actually still ready (i.e., the list_head
is properly initialized as an empty list).
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Tim Harvey <tharvey@gateworks.com> # imx8mm-venice-*
Rasmus Villemoes [Fri, 28 Oct 2022 11:50:50 +0000 (13:50 +0200)]
cyclic: use a flag in gd->flags for recursion protection
As a preparation for future patches, use a flag in gd->flags rather
than a separate member in (the singleton) struct cyclic_drv to keep
track of whether we're already inside cyclic_run().
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk> Reviewed-by: Stefan Roese <sr@denx.de> Tested-by: Stefan Roese <sr@denx.de> Tested-by: Tim Harvey <tharvey@gateworks.com> # imx8mm-venice-*
projects / u-boot.git / log
