Tom Rini [Wed, 1 Mar 2023 01:24:25 +0000 (20:24 -0500)]
Merge tag 'u-boot-rockchip-20230228' of https://source.denx.de/u-boot/custodians/u-boot-rockchip
Please pull the updates for rockchip platform:
- Add support for rk3588 soc;
- Add rk3588 Edgeble Neu6 board and Radxa ROCK5B board;
- Add rk3308 Radxa ROCK Pi S board;
- Add rk3568 Radxa ROCK 3 board,
- Add rk3566 Radxa Compute Module 3 board;
- Add support for sdram reg info version 3
- Refactor rockchip OTP/eFuse driver and add more soc support;
- Add external TPL support for binman;
binman support for mkimage ignore missing entry is based on [1];
Eugen Hristev [Wed, 22 Feb 2023 09:05:12 +0000 (11:05 +0200)]
board: rockchip: add Radxa ROCK5B Rk3588 board
ROCK 5B is a Rockchip RK3588 based SBC (Single Board Computer) by Radxa.
There are tree variants depending on the DRAM size : 4G, 8G and 16G.
Specification:
Rockchip Rk3588 SoC
4x ARM Cortex-A76, 4x ARM Cortex-A55
4/8/16GB memory LPDDR4x
Mali G610MC4 GPU
MIPI CSI 2 multiple lanes connector
eMMC module connector
uSD slot (up to 128GB)
2x USB 2.0, 2x USB 3.0
2x HDMI output, 1x HDMI input
Ethernet port
40-pin IO header including UART, SPI, I2C and 5V DC power in
USB PD over USB Type-C
Size: 85mm x 54mm
Kernel commits: a1d3281450ab ("arm64: dts: rockchip: Add rock-5b board") 6fb13f888f2a ("arm64: dts: rockchip: Update sdhci alias for rock-5b")
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
clk: rockchip: rk3568: add more supported clk rates for sdmmc and emmc
SDHCI driver may attempt to set 26MHz clock, but clk_rk3568
will return error in this case. Apparently, SDHCI silently ignores the
error and as a result eMMC initialization fails.
Add 25 MHz and 26 MHz clk rates for sdmmc and emmc on rk3568 to fix that.
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Booting from SDMMC is one of the fast and easy booting methods
for initial support of any SoC to upstream more features.
This patch is trying to add the sdmmc node for rk3588 and added
as u-boot specific node in -u-boot.dtsi as upstream Linux is not
supporting yet.
As soon as Linux supports it, a sync of the Linux device tree
would eventually drop this node.
Clock properties as added according to the rockchip mmc driver
but the actual definition might add scmi clocks into 0 and 1
indexes. This is due to scmi clock are not supporting in upstream
U-Boot. Properly addition of scmi clock would eventually follow
sdmmc clock definition of Linux once they upstreamed.
Jagan Teki [Mon, 30 Jan 2023 14:57:45 +0000 (20:27 +0530)]
arm: rockchip: Add RK3588 arch core support
The Rockchip RK3588 is a ARM-based SoC with quad-core Cortex-A76
and quad-core Cortex-A55 including NEON and GPU, 6TOPS NPU,
Mali-G610 MP4, HDMI Out, HDMI In, DP, eDP, MIPI DSI, MIPI CSI2,
LPDDR4/4X/5, eMMC5.1, SD3.0/MMC4.5, USB OTG 3.0, Type-C, USB 2.0,
PCIe 3.0, SATA 3, Ethernet, SDIO3.0 I2C, UART, SPI, GPIO and PWM.
Add arch core support for it.
Signed-off-by: Jagan Teki <jagan@edgeble.ai> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
Jagan Teki [Mon, 30 Jan 2023 14:57:42 +0000 (20:27 +0530)]
arm64: dts: rockchip: Add base DT for rk3588 SoC
This initial version supports CPU, dma, interrupts, timers, UART and
SDHCI (everything necessary to boot Linux on this system on chip) as
well as Ethernet, I2C, PWM and SPI.
The DT is split into rk3588 and rk3588s, which is a reduced version
(i.e. with less peripherals) of the former.
Jagan Teki [Mon, 30 Jan 2023 14:57:41 +0000 (20:27 +0530)]
arm: rockchip: Add ioc header for rk3588
Add IOC unit header include for rk3588.
Signed-off-by: Steven Liu <steven.liu@rock-chips.com> Signed-off-by: Joseph Chen <chenjh@rock-chips.com> Signed-off-by: Jagan Teki <jagan@edgeble.ai> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com> Signed-off-by: Jagan Teki <jagan@edgeble.ai> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jagan Teki [Mon, 30 Jan 2023 14:57:38 +0000 (20:27 +0530)]
ram: rockchip: Add rk3588 ddr driver support
Add ddr driver for rk3588 to get the ram capacity.
Co-developed-by: Jonas Karlman <jonas@kwiboo.se> Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Signed-off-by: Jagan Teki <jagan@edgeble.ai> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Chris Morgan [Mon, 13 Feb 2023 22:27:39 +0000 (16:27 -0600)]
rockchip: rk3568: enable automatic power savings
It enables automatic clock gating on idle, disables the eDP phy by
default, and sets the core pvtpll ring length. It is reported this
lowers the temperature on at least one SoC by 7C.
Signed-off-by: Peter Geis <pgwipeout@gmail.com> Signed-off-by: Chris Morgan <macromorgan@hotmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Chris Morgan [Mon, 13 Feb 2023 22:27:38 +0000 (16:27 -0600)]
rockchip: rk3568: add boot device detection
Enable spl to detect which device it was booted from.
Signed-off-by: Peter Geis <pgwipeout@gmail.com> Signed-off-by: Chris Morgan <macromorgan@hotmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Chris Morgan [Mon, 13 Feb 2023 22:27:35 +0000 (16:27 -0600)]
dts: rockchip: px30: add gpio-ranges property to gpio nodes
Add the gpio-ranges property to each GPIO node for use in deriving
the correct bank ID. Note that invoking "gpio status -a" no longer
causes the board to hit a "Synchronous Abort".
Fixes: 537b1a277479 ("rockchip: add px30 devicetrees") Signed-off-by: Chris Morgan <macromorgan@hotmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Chris Morgan [Mon, 13 Feb 2023 22:27:34 +0000 (16:27 -0600)]
gpio: gpio-rockchip: parse gpio-ranges for bank id
Use the new devicetree property of gpio-ranges to determine the GPIO
bank ID. Preserve the "old" way of doing things too, so that boards
can be migrated and tested gradually (I only have a 3566 and 3326 to
test).
Signed-off-by: Chris Morgan <macromorgan@hotmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Manoj Sai [Fri, 17 Feb 2023 11:58:45 +0000 (17:28 +0530)]
rk3566: radxa-cm3: Enable USB2.0, USB3.0 support
=> usb start
starting USB...
Bus usb@fd000000: Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.10
Bus usb@fd800000: USB EHCI 1.00
scanning bus usb@fd000000 for devices... cannot reset port 1!?
2 USB Device(s) found
scanning bus usb@fd800000 for devices... 4 USB Device(s) found
scanning usb for storage devices... 2 Storage Device(s) found
=> usb tree
USB device tree:
1 Hub (5 Gb/s, 0mA)
| U-Boot XHCI Host Controller
|
+-2 Mass Storage (5 Gb/s, 224mA)
SanDisk Dual Drive 04019c9b2e1a58f24ee318c3c123aa5
1 Hub (480 Mb/s, 0mA)
| u-boot EHCI Host Controller
|
+-2 Hub (480 Mb/s, 100mA)
| USB 2.0 Hub
|
+-3 Mass Storage (480 Mb/s, 500mA)
| JetFlash Mass Storage Device 19M7I4ZQFTSC08SU
|
+-4 Human Interface (12 Mb/s, 98mA)
Logitech USB Receiver
Co-developed-by: Suniel Mahesh <sunil@amarulasolutions.com> Signed-off-by: Suniel Mahesh <sunil@amarulasolutions.com> Signed-off-by: Manoj Sai <abbaraju.manojsai@amarulasolutions.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jagan Teki [Fri, 17 Feb 2023 11:58:41 +0000 (17:28 +0530)]
drivers: phy: add naneng combphy for rk3568
RK3568 has three combo phys, and PCIe/USB3/SATA/QSGMII controllers
share one pipe interface for each combo phy, here is the diagram
of the complex connection.
Manoj Sai [Fri, 17 Feb 2023 11:58:40 +0000 (17:28 +0530)]
phy: rockchip-inno-usb2: Add USB2 PHY for rk3568
RK3568 has two USB 2.0 PHYs, and each PHY has two ports, the OTG port
of PHY0 support OTG mode with charging detection function, they are
similar to previous Rockchip SoCs.
However, there are three different designs for RK3568 USB 2.0 PHY.
1. RK3568 uses independent USB GRF module for each USB 2.0 PHY.
2. RK3568 accesses the registers of USB 2.0 PHY IP directly by APB.
3. The two ports of USB 2.0 PHY share one interrupt.
This patch only PHY1 with necessary attributes required to function
USBPHY1 on U-Boot.
Co-developed-by: Ren Jianing <jianing.ren@rock-chips.com> Signed-off-by: Ren Jianing <jianing.ren@rock-chips.com> Co-developed-by: Jagan Teki <jagan@amarulasolutions.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Signed-off-by: Manoj Sai <abbaraju.manojsai@amarulasolutions.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jagan Teki [Fri, 17 Feb 2023 11:58:39 +0000 (17:28 +0530)]
phy: rockchip: inno-usb2: Add support #address_cells = 2
New Rockchip devices have the usb phy nodes as standalone devices.
These nodes have register nodes with #address_cells = 2, but only
use 32 bit addresses.
Adjust the driver to check if the returned address is "0", and adjust
the index in that case.
Derived and adjusted the similar change from linux-next with below
commit <9c19c531dc98> ("phy: phy-rockchip-inno-usb2: support
#address_cells = 2")
Co-developed-by: Manoj Sai <abbaraju.manojsai@amarulasolutions.com> Signed-off-by: Manoj Sai <abbaraju.manojsai@amarulasolutions.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Radxa Compute Module 3(CM3) is one of the modules from a series
System On Module based on the Radxa ROCK 3 series and is compatible
with Raspberry Pi CM4 pinout and form factor.
Specification:
- Rockchip RK3566
- up to 8GB LPDDR4
- up to 128GB high performance eMMC
- Optional wireless LAN, 2.4GHz and 5.0GHz IEEE 802.11b/g/n/ac wireless,
BT 5.0, BLE with onboard and external antenna.
- Gigabit Ethernet PHY
Radxa CM3 needs to mount on top of this IO board in order to create
complete Radxa CM3 IO board platform.
Since Radxa CM3 is compatible with Raspberry Pi CM4 pinout so it is
possible to mount Radxa CM3 on top of the Rasberry Pi CM4 IO board.
linux-next commit for the same,
commit <8f19828844f2> ("arm64: dts: rockchip: Fix compatible for Radxa
CM3")
Add support for Radxa CM3.
Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Co-developed-by: FUKAUMI Naoki <naoki@radxa.com> Signed-off-by: FUKAUMI Naoki <naoki@radxa.com> Signed-off-by: Jagan Teki <jagan@amarulasolutions.com>
include: rk3328: Add default env for compressed kernel images
Add default memory addresses for kernel_comp_addr_r and kernel_comp_size
to enable booting from a compressed kernel image. This area is temporarily
used to decompress the kernel image on-the-fly.
Signed-off-by: Christopher Obbard <chris.obbard@collabora.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Akash Gajjar [Tue, 14 Feb 2023 15:31:09 +0000 (21:01 +0530)]
arm64: dts: rockchip: rk3308: Add Radxa ROCK Pi S support
Add Radxa ROCK 3 Model A support. sync rk3308-rock-pi-s.dts from
Linux 6.2.0-rc7.
ROCK Pi S is RK3308 based SBC from radxa.com. ROCK Pi S has a,
- 256MB/512MB DDR3 RAM
- SD, NAND flash (optional on board 1/2/4/8Gb)
- 100MB ethernet, PoE (optional)
- Onboard 802.11 b/g/n wifi + Bluetooth 4.0 Module
- USB2.0 Type-A HOST x1
- USB3.0 Type-C OTG x1
- 26-pin expansion header
- USB Type-C DC 5V Power Supply
Linux commit commit for the same,
<2e04c25b1320> ("arm64: dts: rockchip: add ROCK Pi S DTS support")
Signed-off-by: Akash Gajjar <gajjar04akash@gmail.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Sat, 25 Feb 2023 19:01:35 +0000 (19:01 +0000)]
binman: Mark mkimage entry missing when its subnodes is missing
Using the mkimage entry with the multiple-data-files prop and having a
missing external blob result in an unexpected ValueError exception using
the --allow-missing flag.
ValueError: Filename 'missing.bin' not found in input path (...)
Fix this by using _pathname that is resolved by ObtainContents for blob
entries, ObtainContents also handles allow missing for external blobs.
Mark mkimage entry as missing and return without running mkimage when
missing entries is reported by CheckMissing.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Sat, 25 Feb 2023 19:01:34 +0000 (19:01 +0000)]
rockchip: Use an external TPL binary on RK3568
Rockchip SoCs typically use U-Boot TPL to initialize DRAM, then jumps
back to BootRom to load next stage, U-Boot SPL, into DRAM. BootRom then
jumps to U-Boot SPL to continue the normal boot flow.
However, there is no support to initialize DRAM on RK35xx SoCs using
U-Boot TPL and instead an external TPL binary must be used to generate a
bootable u-boot-rockchip.bin image.
Add CONFIG_ROCKCHIP_EXTERNAL_TPL to indicate that an external TPL should
be used. Build U-Boot with ROCKCHIP_TPL=/path/to/ddr.bin to generate a
bootable u-boot-rockchip.bin image for RK3568.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Tested-by: Eugen Hristev <eugen.hristev@collabora.com>
Jonas Karlman [Sat, 25 Feb 2023 19:01:33 +0000 (19:01 +0000)]
binman: Add support for a rockchip-tpl entry
The rockchip-tpl entry can be used when an external TPL binary should be
used instead of the normal U-Boot TPL.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Tested-by: Eugen Hristev <eugen.hristev@collabora.com>
Jonas Karlman [Fri, 10 Feb 2023 18:30:05 +0000 (18:30 +0000)]
rockchip: dts: rk3328: fix sdram params
The rk3328 sdram driver read sdram parameters from the devicetree into a
struct rk3328_sdram_params using dev_read_u32_array.
After commit 5ab30c3176bf ("ram: rockchip: Update ddr pctl regs for px30")
changed the size of struct ddr_pctl_regs, a member of struct
rk3328_sdram_params, U-Boot TPL can no longer initialize DRAM on RK3328.
Add ten u32 to the sdram parameter array in devicetree to align with
this size change. This fixes DRAM initialization on RK3328.
Fixes: 5ab30c3176bf ("ram: rockchip: Update ddr pctl regs for px30") Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Simon Glass <sjg@chromium.org> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Reviewed-by: Jagan Teki <jagan@amarulasolutions.com> Tested-by: Jagan Teki <jagan@amarulasolutions.com> # roc-rk3328-cc
Jonas Karlman [Tue, 7 Feb 2023 17:27:11 +0000 (17:27 +0000)]
rockchip: sdram: add dram bank with usable memory beyond 4GB
Add a second dram bank of usable memory beyond the blob of space for
peripheral near 4GB. Any memory that exists beyond the 4GB mark is added
to the second bank.
Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
Jonas Karlman [Tue, 7 Feb 2023 17:27:10 +0000 (17:27 +0000)]
rockchip: sdram: add basic support for sdram reg info version 3
Newer DRAM initialization blobs from vendor can encode sdram info in a
new version 3 format. The new format makes use of more bits in sys_reg3
compared to the version 2 format.
Add basic support for detecting the version 3 format and decoding the
high bits used for ddrtype.
This fixes decode of sdram size on my RK3568 boards that have LPDDR4X.
Details on the new format was deciphered from vendor u-boot commit [1].
Quentin Schulz [Tue, 17 Jan 2023 17:26:54 +0000 (18:26 +0100)]
rockchip: ringneck: fix SDRAM init fail
CONFIG_RAM_PX30_DDR4 got renamed to CONFIG_RAM_ROCKCHIP_DDR4 in commit 26f92be07e2a ("ram: rockchip: Add common ddr type configs"). Since both
patchsets were merged unbeknownst to the other, the conflict wasn't
detected while testing each patchset individually and could only be
observed after a merge to master branch.
Fixes: c925be73a0a8 ("rockchip: add support for PX30 Ringneck SoM on Haikou Devkit") Cc: Quentin Schulz <foss+uboot@0leil.net> Signed-off-by: Quentin Schulz <quentin.schulz@theobroma-systems.com> Reviewed-by: Kever Yang <kever.yang@rock-chips.com> Signed-off-by: Kever Yang <kever.yang@rock-chips.com>
The CLKDIV register should always be programmed with a value
higher than zero (0); that is, a clock divider should always be
used for 8-bit DDR mode.
In Linux, the driver applies this logic for all SoCs using the driver
and does not distinguish RK3288, so presumably this requirement is the
same for all other Rockchip SoCs using this IP.
Add the necessary code to double the clock frequency when 8-bit DDR is
selected. The dw_mmc core already handles setting CLKDIV correctly
given the input clock and desired bus clock.
Signed-off-by: John Keeping <john@metanate.com> Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
Mihai Sain [Thu, 23 Feb 2023 09:12:58 +0000 (11:12 +0200)]
ARM: dts: at91: sama7g5: fix clock id for ebi node
The PMC_MCK1 clock id for the ebi node is 23.
Fixes: 746b738224ad ("ARM: dts: at91: sama7g5/sama7g5ek: align with Linux DT") Reviewed-by: Eugen Hristev <eugen.hristev@collabora.com> Signed-off-by: Mihai Sain <mihai.sain@microchip.com>
Tom Rini [Thu, 23 Feb 2023 22:30:49 +0000 (17:30 -0500)]
Merge branch '2023-02-23-assorted-fixes'
- btrfs bugfix, silence a bunch of gcc-12.2 linker warnings finally,
relax one of the trace test time requirements (so CI doesn't fail due
to test being slightly slow, but still correct), and correct env on
MMC and checking for where GPT can be
Maxim Schwalm [Tue, 14 Feb 2023 17:35:30 +0000 (19:35 +0200)]
ARM: tegra: provide default USB gadget setup
All Nvidia boards use the same manufacturer, vendor ID and product ID
for the gadgets. Make them the defaults to remove some boilerplate from
the defconfigs.
Inspired by commit e02687bda96c ("sunxi: provide default USB gadget
setup") which did the same for Allwinner boards.
Tested-by: Thierry Reding <treding@nvidia.com> # T30 and T124 Signed-off-by: Maxim Schwalm <maxim.schwalm@gmail.com> Signed-off-by: Tom <twarren@nvidia.com>
Svyatoslav Ryhel [Tue, 14 Feb 2023 17:35:29 +0000 (19:35 +0200)]
spi: tegra20_slink: accept any word length
Original t20 slink could work with commands only
fully divisible by 8. This patch removes such
restriction, so commands of any bitlength now
can be passed and processed.
Tested-by: Andreas Westman Dorcsak <hedmoo@yahoo.com> # ASUS TF600T T30 Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # LG P895 T30 Tested-by: Thierry Reding <treding@nvidia.com> # T30 and T124 Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com> Signed-off-by: Tom <twarren@nvidia.com>
Svyatoslav Ryhel [Tue, 14 Feb 2023 17:35:26 +0000 (19:35 +0200)]
ARM: tegra: clock: add clock_decode_pair helper
Get periph clock id and its parent from device tree.
This works by looking up the peripheral's 'clocks' node and
reading out the second and fourth cells, which are the
peripheral and PLL clock numbers.
Tested-by: Andreas Westman Dorcsak <hedmoo@yahoo.com> # ASUS TF T30 Tested-by: Robert Eckelmann <longnoserob@gmail.com> # ASUS TF101 T20 Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # HTC One X Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com> Signed-off-by: Tom <twarren@nvidia.com>
Svyatoslav Ryhel [Tue, 14 Feb 2023 17:35:25 +0000 (19:35 +0200)]
ARM: tegra: clock: add clk_id_to_pll_id helper
This function allows to convert a device tree clock ID to PLL ID.
Tested-by: Andreas Westman Dorcsak <hedmoo@yahoo.com> # ASUS TF T30 Tested-by: Robert Eckelmann <longnoserob@gmail.com> # ASUS TF101 T20 Tested-by: Svyatoslav Ryhel <clamor95@gmail.com> # HTC One X Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com> Signed-off-by: Tom <twarren@nvidia.com>
Marek Vasut [Thu, 9 Feb 2023 12:30:10 +0000 (13:30 +0100)]
env: mmc: Apply GPT only on eMMC user HW partition
Apply the GPT U-Boot environment GUID type look up only on eMMC user
HW partition, do not apply the look up on eMMC boot HW partitions as
mmc_offset_try_partition() assumes either SD partitions or eMMC user
HW partition.
This fixes environment operation on systems where CONFIG_SYS_MMC_ENV_PART
is non-zero and CONFIG_SYS_REDUNDAND_ENVIRONMENT is set.
Fixes: 80105d8fd52 ("env: mmc: select GPT env partition by type guid") Signed-off-by: Marek Vasut <marex@denx.de> Reviewed-by: Tom Rini <trini@konsulko.com>
Simon Glass [Fri, 17 Feb 2023 22:45:16 +0000 (15:45 -0700)]
trace: Relax test requirements
We expect the profile and bootstage to agree on timing, but when
running on slow machines there can be a larger descrepency. Increase the
tolerance to fix this.
Fixes: 9cea4797aeb ("trace: Add a test") Signed-off-by: Simon Glass <sjg@chromium.org>
Tom Rini [Tue, 14 Feb 2023 20:09:31 +0000 (15:09 -0500)]
Makefile: Link with -z noexectack
When moving to gcc-12.2 we started trying to quiet some of the new
linker warnings, that are not relevant to us. However, a
misunderstanding of the mechanics at play meant that I intentionally
omitted passing -z noexecstack to the linker, when we do need to. Add
this flag and in turn remove warnings from the linker.
Fixes: 1e1c51f8ace8 ("Makefile: link with --no-warn-rwx-segments") Signed-off-by: Tom Rini <trini@konsulko.com> Reviewed-by: Simon Glass <sjg@chromium.org>
Note the second and the last line, the @cur_len is 450+MiB, which is
almost a chunk size.
And inside __btrfs_map_block(), we limits the returned value to stripe
length, but that's depending on the chunk type:
if (map->type & (BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID1 |
BTRFS_BLOCK_GROUP_RAID1C3 | BTRFS_BLOCK_GROUP_RAID1C4 |
BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6 |
BTRFS_BLOCK_GROUP_RAID10 |
BTRFS_BLOCK_GROUP_DUP)) {
/* we limit the length of each bio to what fits in a stripe */
*length = min_t(u64, ce->size - offset,
map->stripe_len - stripe_offset);
} else {
*length = ce->size - offset;
}
This means, if the chunk is SINGLE profile, then we don't limit the
returned length at all, and even for other profiles, we can still return
a length much larger than the requested one.
[FIX]
Properly clamp the returned length, preventing it from returning a much
larger range than expected.
Reported-by: Andreas Schwab <schwab@linux-m68k.org> Signed-off-by: Qu Wenruo <wqu@suse.com>
projects / u-boot.git / log