| Age | Commit message (Collapse) | Author |
|---|
|
Write the FADT in common code since it's used on all architectures.
Since the FADT is mandatory all SoCs or mainboards must implement the
introduced function acpi_fill_fadt() and properly update the FADT.
Signed-off-by: Patrick Rudolph <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Cc: Simon Glass <[email protected]>
Cc: Bin Meng <[email protected]>
|
|
This moves the SPCR and DBG2 table generation into common code, so that
they can be used by architectures other than x86.
Signed-off-by: Maximilian Brune <[email protected]>
Signed-off-by: Patrick Rudolph <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Cc: Simon Glass <[email protected]>
Cc: Bin Meng <[email protected]>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-rockchip
CI: https://source.denx.de/u-boot/custodians/u-boot-rockchip/-/pipelines/22993
- New boards:
rk3566: Hardkernel ODROID-M1S
rk3588s: Hardkernel ODROID-M2
rk3588: NanoPC-T6 LTS
- Migrate to use USB_DWC3_GENERIC for rk3328
- Other board level config and dts update
|
|
Increase rng-seed size to make Linux happy and initialize rng pool instantly.
Linux 5.19+ requires 32 bytes of entropy to initialize random pool, but u-boot
currently provides only 8 bytes.
Linux 5.18 and probably some versions before it used to require 64 bytes.
Bump min value to 64 bytes to be on a safe side.
Boot with 8 byte rng-seed (Linux 6.11):
# dmesg | grep crng
[ 12.089286] random: crng init done
Boot with 32 byte rng-seed (Linux 6.11):
# dmesg | grep crng
[ 0.000000] random: crng init done
Linux source references:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/char/random.c?h=v5.19#n551
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/char/random.c?h=v5.18#n236
Signed-off-by: Alex Shumsky <[email protected]>
Fixes: d2048bafae40 ("rockchip: board: Add board_rng_seed() for all Rockchip devices")
Reviewed-by: Dragan Simic <[email protected]>
Reviewed-by: Marek Vasut <[email protected]>
Reviewed-by: Quentin Schulz <[email protected]>
Reviewed-by: Kever Yang <[email protected]>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-imx
CI: https://source.denx.de/u-boot/custodians/u-boot-imx/-/pipelines/22989
- Remove unneeded USB board code and fix reset on mx6ul_14x14_evk.
- Update fastboot buffer size/address for verdin-imx8m{m|p}.
- Fix imxrt1050-evk boot and convert it to standard boot.
- Fix imx8qxp-mek and imx8qm-mek boot.
- Add support for the i.MX93 9X9 QSB board.
- Make livetree API to work on i.MX.
- Set sane default value for i.MX8M SPL_LOAD_FIT_ADDRESS.
- Deduplicate DH i.MX8MP/i.MX6 DHSOM defconfigs.
- Select default TEXT_BASE for i.MX6/i.MX7.
- Several updates for DH i.MX8MP DRC02.
|
|
Enable standard boot support and add default environments for the
imxrt1050-evk board.
Signed-off-by: Jesse Taube <[email protected]>
|
|
Simon Glass <[email protected]> says:
This little series adds a new 'memmap' command, intended to show the
layout of memory within U-Boot and how much memory is available for
loading images.
Link: https://lore.kernel.org/r/[email protected]
|
|
The bootstage array includes pointers to strings but not the strings
themselves. The strings are added when stashing, but including them in
the size calculation gives an inflated view of the amount of space used
by the array.
Update this function so it can return the amount of memory used by the
bootstage structures themselves, without the strings which they point
to.
Signed-off-by: Simon Glass <[email protected]>
|
|
Add accessors for bloblist, bootstage, trace and video to avoid needing
more #ifdefs in the C code.
Signed-off-by: Simon Glass <[email protected]>
|
|
The call to malloc() is a bit strange. The naming of the arguments
suggests that an address is passed, but in fact it is a pointer, at
least in the board_init_r() function and SPL equivalent.
Update it to work as described. Add a function comment as well.
Note that this does adjustment does not extend into the malloc()
implementation itself, apart from changing mem_malloc_init(), since
there are lots of casts and pointers and integers are used
interchangeably.
Signed-off-by: Simon Glass <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-microblaze
AMD/Xilinx changes for v2025.01-rc1 v2
.mailmap:
- Switch Padmarao's email to AMD one
zynq_spi:
- Make update_stripe static
xilinx:
- Update DT description for EMMCs
zynqmp:
- Update logic around RPUs and tcm handling
- Update bootmenu selection for Kria
- Add description for SC vm-p-b1369
- Fix comment about file location in zynqmp-p-a2197-00-revA.dts
versal:
- Fix logic around USB boot
versal2:
- Disable useless features for Mini configurations
versal-net:
- Get rid of current-speed DT property from mini configuration
microblaze:
- Fix scriptaddr location
# -----BEGIN PGP SIGNATURE-----
#
# iF0EABECAB0WIQQbPNTMvXmYlBPRwx7KSWXLKUoMIQUCZxuweAAKCRDKSWXLKUoM
# IY1iAKCH/GKJHEXFfLvr0OGuO6c1SX9+ZQCfTjRAHrL186X6LUgjOpmtmsrVK1c=
# =4gY0
# -----END PGP SIGNATURE-----
# gpg: Signature made Fri 25 Oct 2024 08:51:36 AM CST
# gpg: using DSA key 1B3CD4CCBD79989413D1C31ECA4965CB294A0C21
# gpg: Good signature from "Michal Simek <[email protected]>" [full]
# gpg: aka "Michal Simek (Xilinx) <[email protected]>" [full]
# gpg: aka "Michal Simek (Xilinx) <[email protected]>" [full]
# gpg: aka "Michal Simek (AMD) <[email protected]>" [unknown]
|
|
https://source.denx.de/u-boot/custodians/u-boot-dfu
CI: https://source.denx.de/u-boot/custodians/u-boot-dfu/-/pipelines/22962
Dfu:
- Rely on device tree for spi speed/mode on spi flash
Android Image:
- Fix booting on platforms having > 4GiB of memory
- Decompress boot image to kernel_addr_r when compression is enabled
- Honor CONFIG_SYS_LOAD_ADDR when mkbootimg uses default address
Bcb:
- Rework bcb command to use U_BOOT_LONGHELP
- Move ab_select cmd to bcb cmd
- Implement ab_dump command in bcb
- bcb: Write '_<slot>' instead of '<slot>' to misc partition
|
|
The Hardkernel ODROID-M2 is a single-board computer based on Rockchip
RK3588S2 SoC. It features e.g. 8/16 GB LPDDR5 RAM, 64 GB eMMC, SD-card,
GbE LAN, HDMI 2.0, M.2 NVMe and USB 2.0/3.0/Type-C.
Features tested on a ODROID-M2 16GB rev1.0 20240611:
- SD-card boot
- eMMC boot
- Ethernet
- PCIe/NVMe
- USB gadget
- USB host
Signed-off-by: Jonas Karlman <[email protected]>
Reviewed-by: Kever Yang <[email protected]>
|
|
For unknown reason USB host hasn't been described for distro boot that's
why wire it now.
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/9715605c232ce38a36a200204c14208b9d76cd83.1727694230.git.michal.simek@amd.com
|
|
There is only one USB controller that's why remove description for USB1.
Signed-off-by: Michal Simek <[email protected]>
Link: https://lore.kernel.org/r/f148161ee489d19d708df1b4fa7477821e06bb55.1727694230.git.michal.simek@amd.com
|
|
Add i.MX93 9x9 Quick Start Board support.
- Two ddr scripts included w/o inline ecc feature.
- SDHC/NETWORK/I2C/UART supported
- PCA9450 supported, default over drive mode
- Documentation added.
Signed-off-by: Peng Fan <[email protected]>
|
|
The function prototypes that are defined in the header most likely has
to be marked as static inline. This helps avoiding the compiler warnings:
include/mux.h:120:14: warning: no previous prototype for ‘mux_control_states’ [-Wmissing-prototypes]
include/mux.h:125:18: warning: no previous prototype for ‘mux_control_select’ [-Wmissing-prototypes]
include/mux.h:133:5: warning: no previous prototype for ‘mux_control_deselect’ [-Wmissing-prototypes]
include/mux.h:138:21: warning: no previous prototype for ‘mux_control_get’ [-Wmissing-prototypes]
include/mux.h:143:6: warning: no previous prototype for ‘mux_control_put’ [-Wmissing-prototypes]
include/mux.h:147:21: warning: no previous prototype for ‘devm_mux_control_get’ [-Wmissing-prototypes]
include/mux.h:153:5: warning: no previous prototype for ‘dm_mux_init’ [-Wmissing-prototype ]
Signed-off-by: Andy Shevchenko <[email protected]>
|
|
There are no users of the blk_create_device() function outside the uclass.
Let's make it static. This will ensure that new block drivers will use
blk_create_devicef().
Signed-off-by: Heinrich Schuchardt <[email protected]>
Reviewed-by: Ilias Apalodimas <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
When booting with platforms having > 4GiB of memory,
the kernel physical address can be more than 32bits.
Use ulong like all the other addresses, and fix the
print to show the > 32bits address numbers.
Signed-off-by: Neil Armstrong <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Tested-by: Guillaume La Roque <[email protected]>
Link: https://lore.kernel.org/r/20241017-topic-fastboot-fixes-mkbootimg-v2-1-c3927102d931@linaro.org
Signed-off-by: Mattijs Korpershoek <[email protected]>
|
|
It's really helpful to have the ability to dump BCB block for debugging
A/B logic on the board supported this partition schema.
Command 'bcb ab_dump' prints all fields of bootloader_control struct
including slot_metadata for all presented slots.
Output example:
=====
> board# bcb ab_dump ubi 0#misc
> Read 512 bytes from volume misc to 000000000bf07580
> Read 512 bytes from volume misc to 000000000bf42f40
> Bootloader Control: [misc]
> Active Slot: _a
> Magic Number: 0x42414342
> Version: 1
> Number of Slots: 2
> Recovery Tries Remaining: 0
> CRC: 0x2c8b50bc (Valid)
>
> Slot[0] Metadata:
> - Priority: 15
> - Tries Remaining: 0
> - Successful Boot: 1
> - Verity Corrupted: 0
>
> Slot[1] Metadata:
> - Priority: 14
> - Tries Remaining: 7
> - Successful Boot: 0
> - Verity Corrupted: 0
====
The ab_dump command allows you to display ABC data directly on the
U-Boot console. During an A/B test execution, this test verifies the
accuracy of each field within the ABC data.
Signed-off-by: Dmitry Rokosov <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Tested-by: Mattijs Korpershoek <[email protected]> # vim3_android
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mattijs Korpershoek <[email protected]>
|
|
To enhance code organization, it is beneficial to consolidate all A/B
BCB management routines into a single super-command.
The 'bcb' command is an excellent candidate for this purpose.
This patch integrates the separate 'ab_select' command into the 'bcb'
group as the 'ab_select' subcommand, maintaining the same parameter list
for consistency.
Signed-off-by: Dmitry Rokosov <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Tested-by: Mattijs Korpershoek <[email protected]> # vim3_android
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mattijs Korpershoek <[email protected]>
|
|
There are new function documentation requirements in U-Boot, so apply
these changes for android_ab.
Reviewed-by: Mattijs Korpershoek <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Tested-by: Guillaume La Roque <[email protected]>
Signed-off-by: Dmitry Rokosov <[email protected]>
Tested-by: Mattijs Korpershoek <[email protected]> # vim3_android
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Mattijs Korpershoek <[email protected]>
|
|
To smoothly handle the transition from the legacy SPI FLASH
API to the driver model API, add the DM functions
as dummy inline functions.
Today, client code uses #if/#else conditionals, but it's better
to use if(IS_ENABLED()) to make sure all code builds fine
and avoid configuration hell, leaving the compiler remove
the dead code.
An example is cmd/sf, which could make use of those dummy
functions to drop the conditional compilation.
Signed-off-by: Neil Armstrong <[email protected]>
Reviewed-by: Mattijs Korpershoek <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
[mkorpershoek: removed duplicate "the" from commit msg]
Signed-off-by: Mattijs Korpershoek <[email protected]>
|
|
Nothing in cyclic.h is needed to define struct global_data, so do not
include that header.
If any .c file relies on getting cyclic.h through asm/global_data.h,
it needs to include it itself.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Tom Rini <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
Nobody relies on getting the cyclic API declared by including the
watchdog.h header, but for historical reasons, many TUs include
watchdog.h to get a declaration of schedule(). Now that we have a
dedicated header for just that, include that header instead of
cyclic.h.
Eventually, all TUs that call schedule() should themselves include
u-boot/schedule.h, but this is a step towards getting rid of
unnecessary include statements in cyclic.h and global_data.h.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
The only caller left is schedule(); everybody outside cyclic.c now
calls or references schedule().
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
I noticed an "unnecessary" include of <cyclic.h> in
global_data.h, in the sense that nothing in cyclic.h is needed in
order to define 'struct global_data'.
Well, it's not unnecessary, as it implicitly ensures that everybody
gets a declaration of schedule(), and schedule() is (obviously) called
all over the tree. Almost none of those places directly include
<cyclic.h>, but for historical reasons, many do include
<watchdog.h> (most schedule() instances are replacements of
WATCHDOG_RESET()).
However, very few TUs actually need the declarations of the
cyclic_register() and struct cyclic_info, and they also don't really
need anything from the watchdog.h header.
So introduce a new header which just contains a declaration of
schedule(), which can then be included from all the places that do
call schedule(). I removed the direct reference to cyclic_run(),
because we shouldn't have two public functions for doing roughly the
same without being very explicit about when one should call one or the
other.
Testing of later patches that explicitly include <schedule.h> when
schedule() is used revealed a problem with host tool build on win32,
which apparently picked up a host <schedule.h>. To avoid that problem,
put the new header in include/u-boot/ and hence make the include
statements say <u-boot/schedule.h>.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
Reviewed-by: Stefan Roese <[email protected]>
|
|
The VNBYTES() macro has been updated to silence possible warnings
regarding authorized (but unusual) uses of this macro, but the comment
was kept unchanged. A year has passed so let's fix the comment now to
avoid confusions.
Fixes: cc05d352fbc ("video: Add parentheses around VNBYTES() macro")
Suggested-by: Tom Rini <[email protected]>
Link: https://lore.kernel.org/u-boot/20240906183432.GG3879073@bill-the-cat/
Signed-off-by: Miquel Raynal <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
Chia-Wei Wang <[email protected]> says:
Aspeed AST2700 SoCs integrates the Caliptra secure IP, where an ECDSA384
signature verification HW interface is exported for SoC crypto needs.
This patch series firstly extends the FIT image signing/verify common
code to support the ECDSA384 algorithm. For better convenience, the
device tree for ECDSA public key storage is also revised by referring
to RSA implementations.
After the FIT common code revision, the driver is implemented for
AST2700 to leverage the Caliptra ECDSA384 signature verification.
These are verified by signed FIT images with the algorithm "sha384,ecdsa384".
Link: https://lore.kernel.org/r/[email protected]
|
|
Add ECDSA384 algorithm support for image signing and verification.
Signed-off-by: Chia-Wei Wang <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-dm
A few new x86 commands and minor improvements
expo improvements
binman support for signing FIT images
|
|
Move the SR3 bit definition in the right place. Fix
what is likely a rebase artifact. No functional change.
Fixes: 215f1d5794c6 ("mtd: spi-nor: Clear Winbond SR3 WPS bit on boot")
Signed-off-by: Marek Vasut <[email protected]>
|
|
Some files may include this header file without first including dm.h
so add a forward declaration.
Signed-off-by: Simon Glass <[email protected]>
|
|
This function is really just an assignment, so serves no useful
purpose. Drop it.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present menu items are stored according to their sequence number in
the menu. In some cases we may want to have holes in that sequence, or
not use a sequence at all.
Add a new 'value' property for menu items. This will be used for
reading and writing, if present. If there is no 'value' property, then
the normal sequence number will be used instead.
Signed-off-by: Simon Glass <[email protected]>
|
|
Set aside some expo IDs for 'save' and 'discard' buttons. This avoids
needing to store the IDs for these. Adjust the documentation and expo
tool for the new EXPOID_BASE_ID value.
Ignore these objects when saving and loading the cedit, since they do
not contain real data.
Adjust 'cedit run' to return failure when the user exits the expo
without saving. Update the test for this change as well.
Signed-off-by: Simon Glass <[email protected]>
|
|
At present a fixed position is used for menu items, 200 pixels to the
right of the left side of the labels. This means that a menu item with
a very long label may overlap the items.
It seems better to calculate the maximum label width and then place the
items to the right of all of them.
To implement this, add a new struct to containing arrangement
information. Calculate it before doing the actual arrangement. Add a
new style item which sets the amount of space from the right side of
the labels to left side of the items.
Signed-off-by: Simon Glass <[email protected]>
|
|
When starting a new text line, an assumption is made that the current
vertical position is a multiple of the character height. When this is
not true, characters can be written after the end of the framebuffer.
This can causes crashes and strange errors from QEMU.
Adjust the scrolling check when processing a newline character, to
avoid any problems.
Add some comments to make things a little clearer.
Signed-off-by: Simon Glass <[email protected]>
|
|
This is useful for highlighting something with a black background, as
is needed with cedit when using a white-on-black console. Add this as
a new colour.
Signed-off-by: Simon Glass <[email protected]>
|
|
Some boards still don't enable BLK but we want to be able to at least
compile the code which relies on this. For example, bootstd includes
calls to blk_...() functions, albeit with a check for BLK so that the
code is eliminated by the compiler.
Reduce the scope of the BLK #ifdef to help with this.
Signed-off-by: Simon Glass <[email protected]>
|
|
https://gitlab.denx.de/u-boot/custodians/u-boot-imx
CI: https://source.denx.de/u-boot/custodians/u-boot-imx/-/pipelines/22796
- Switch to using upstream DT on DH i.MX8MP DHCOM PDK2/PDK3.
- Add ability to build fallback DTBOs from arch/$(ARCH)/dts.
- Remove fdt_high and initrd_high env variables from imx6-dhcom.
- Add dummy clk for imx8.
- Fix DT corruption in imx8_cpu.
- Improve DDR stability on pico-imx7d.
|
|
Remove both "fdt_high" and "initrd_high" environment variables
in favor of "bootm_size" to safely contain a kernel, device tree
and initrd for relocation inside of 256 MiB region of DRAM.
Signed-off-by: Marek Vasut <[email protected]>
Reviewed-by: Fabio Estevam <[email protected]>
|
|
Move do_irqinfo() prototype to a header file, otherwise compiler is not
happy:
arch/x86/lib/interrupts.c:130:5: warning: no previous prototype for ‘do_irqinfo’ [-Wmissing-prototypes]
Signed-off-by: Andy Shevchenko <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
[trini: Add 'struct cmd_tbl;' to irq_func.h]
Signed-off-by: Tom Rini <[email protected]>
|
|
https://source.denx.de/u-boot/custodians/u-boot-stm
CI: https://source.denx.de/u-boot/custodians/u-boot-stm/-/pipelines/22732
- stm32mp: fix boot issue with OP-TEE
- stm32mp: Add script to install U-Boot from SD/eMMC to SPI NOR on DH STM32MP15xx
- stm32mp: Switch to using upstream DT on DH STM32 DHSOM
- stm32mp: Generate u-boot.itb using binman on DH STM32 DHSOM
|
|
|
|
This adds TCPM framework in preparation for fusb302 support, which can
handle USB power delivery messages. This is needed to solve issues with
devices, that are running from a USB-C port supporting USB-PD, but not
having a battery.
Such a device currently boots to the kernel without interacting with
the power-supply at all. If there are no USB-PD message replies within
5 seconds, the power-supply assumes the peripheral is not capable of
USB-PD. It usually takes more than 5 seconds for the system to reach
the kernel and probe the I2C based fusb302 chip driver. Thus the
system always runs into this state. The power-supply's solution to
fix this error state is a hard reset, which involves removing the
power from VBUS. Boards without a battery (or huge capacitors) will
reset at this point resulting in a boot loop.
This imports the TCPM framework from the kernel. The porting has
originally been done by Rockchip using hardware timers and the Linux
kernel's TCPM code from some years ago.
I had a look at upgrading to the latest TCPM kernel code, but that
beast became a lot more complex due to adding more USB-C features.
I believe these features are not needed in U-Boot and with multiple
kthreads and hrtimers being involved it is non-trivial to port them.
Instead I worked on stripping down features from the Rockchip port
to an even more basic level. Also the TCPM code has been reworked
to avoid complete use of any timers (Rockchip used SoC specific
hardware timers + IRQ to implement delayed work mechanism). Instead
the delayed state changes are handled directly from the poll loop.
Note, that (in contrast to the original Rockchip port) the state
machine has the same hard reset quirk, that the kernel has - i.e.
it avoids disabling the CC pin resistors for devices that are not
self-powered. Without that quirk, the Radxa Rock 5B will not just
end up doing a machine reset when a hard reset is triggered, but will
not even recover, because the CPU will loose power and the FUSB302
will keep this state because of leak voltage arriving through the RX
serial pin (assuming a serial adapter is connected).
This also includes a 'tcpm' command, which can be used to get
information about the current state and the negotiated voltage
and current.
Co-developed-by: Wang Jie <[email protected]>
Signed-off-by: Wang Jie <[email protected]>
Tested-by: Soeren Moch <[email protected]>
Tested-by: Anand Moon <[email protected]>
Reviewed-by: Jonas Karlman <[email protected]>
Signed-off-by: Sebastian Reichel <[email protected]>
|
|
Rasmus Villemoes <[email protected]> says:
Some small improvements to the serial rx buffer feature.
CI seems happy: https://github.com/u-boot/u-boot/pull/674
Link: https://lore.kernel.org/r/[email protected]
|
|
The initialization of upriv->buf doesn't check for a NULL return. But
there's actually no point in doing a separate, unconditional malloc()
in post_probe; we can just make serial_dev_priv contain the rx buffer
itself, and let the (larger) allocation be handled by the driver core
when it allocates the ->per_device_auto. The total run-time memory
used is mostly the same, we reduce the code size a little, and as a
bonus, struct serial_dev_priv does not contain the unused members when
!SERIAL_RX_BUFFER.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
The current implementation of the circular rx buffer falls into a
common trap with circular buffers: It keeps the head/tail indices
reduced modulo the buffer size. The problem with that is that it makes
it impossible to distinguish "buffer full" from "buffer empty",
because in both situations one has head==tail.
This can easily be demonstrated: Build sandbox with RX_BUFFER enabled,
set the RX_BUFFER_SIZE to 32, and try pasting the string
01234567890123456789012345678901
Nothing seems to happen, but in reality, all characters have been read
and put into the buffer, but then tstc ends up believing nothing is in
the buffer anyway because upriv->rd_ptr == upriv->wr_ptr.
A better approach is to let the indices be free-running, and only
reduce them modulo the buffer size when accessing the array. Then
"empty" is head-tail==0 and "full" is head-tail==size. This does rely
on the buffer size being a power-of-two and the free-running
indices simply wrapping around to 0 when incremented beyond the
maximal positive value.
Incidentally, that change from signed to unsigned int also improves
code generation quite a bit: In C, (signed int)%(signed int) is
defined to have the sign of the dividend (so (-35) % 32 is -3, not
29), and hence despite the modulus being a power-of-two, x % 32 does
not actually compile to the same as a simple x & 31 - on x86 with -Os,
it seems that gcc ends up emitting an idiv instruction, which is quite
expensive.
Signed-off-by: Rasmus Villemoes <[email protected]>
Reviewed-by: Simon Glass <[email protected]>
|
|
This is cosmetic change.
Remove the empty comment blocks remaining after conversion to Kconfig
of CONFIG_SYS_MAX_NAND_DEVICE and CONFIG_SERVERIP.
Signed-off-by: Patrick Delaunay <[email protected]>
|
