March 2021 - U-Boot

OMAP3 clocks with ARM vs Thumb
by Adam Ford 19 Mar '21

19 Mar '21

The arch/arm/mach-omap2/omap3/clock.c is being forced to compile with an ARM target even when thumb2 is the preferred target. The note in the makefile states it's due to a bootup issue with OMAP35. Since Thumb2 can help reduce code size, I thought I'd run some tests. The DM3730 can boot with clock.c built for Thumb2, but I can confirm that omap3530 still fails. Instead of tuning the entire file for ARM, I thought I might try to identify which function needs to be compiled for ARM. I was able to set the target attribute for prcm_init() to build for ARM while compiling the rest of the file for Thumb, and the OMAP3530 was able to successfully boot. Unfortunately, doing that broke the DM3730. I started looking at the differences between the OM36xx and OM34xx inside prcm_init() it appears as if there are a bunch of static functions that are unique to either omap34 or omap36, but they are not drastically different except for one area where some math is done to a pointer on the OMAP35 and not on the OMAP36/DM7. /* Moving to the right sysclk and ES rev base */ ptr = ptr + (3 * clk_index) + sil_index; I could be nothing, but I am not well versed in Thumb2 assembly. If someone has any suggestions to try, it would be nice to be able to let the file be built for Thumb2 for both OMAP35 and OMAP36/DM37. adam

1 1

Please pull u-boot-video
by Anatolij Gustschin 19 Mar '21

19 Mar '21

Hi Tom, The following changes since commit bced796e2dadc708ce213170d36be7f0428618a1: Merge tag 'ti-v2021.04-rc3' of https://gitlab.denx.de/u-boot/custodians/u-boot-ti (2021-02-22 09:12:04 -0500) are available in the Git repository at: https://source.denx.de/u-boot/custodians/u-boot-video.git tags/video-for-v2021.04 for you to fetch changes up to fc3241feb9d76fc88a2794b2fc8f82555f2be8b8: video: Drop unprintable characters from video_font_data.h (2021-02-22 20:06:43 +0100) ---------------------------------------------------------------- - drop unprintable characters from video font header ---------------------------------------------------------------- Simon Glass (1): video: Drop unprintable characters from video_font_data.h include/video_font_data.h | 258 +++++++++++++++++++++++----------------------- 1 file changed, 129 insertions(+), 129 deletions(-) gitlab CI: https://source.denx.de/u-boot/custodians/u-boot-video/-/pipelines/6824 Thanks, Anatolij

2 1

[PATCH v7 0/4] Move qfw to DM, add Arm support
by Asherah Connor 19 Mar '21

19 Mar '21

This series moves the QFW driver into a uclass, UCLASS_QFW, and splits the driver into qfw_pio and qfw_mmio. Each driver is selected on the appropriate QEMU board. A sandbox driver is also added, and a DM unit test against that driver. The qfw command is tested in QEMU, and documentation added. Version 7 ensures the qfw test assertions work both against real QEMU instances and the sandbox and removes CMD_QFW from the sandbox_spl config. To view the changes online, see: https://git.src.kameliya.ee/~kameliya/u-boot/log/qfw-priv Changes in v7: * Make the qfw tests agnostic to running against a real QEMU instance or in the sandbox. * Remove CONFIG_CMD_QFW=y from sandbox_spl_defconfig, per trini's request. Changes in v6: * Sandbox driver and tests are split into own commit. Changes in v5: * Split conversion of existing x86-only QFW to DM into its own patch. * Split MMIO driver into its own commit. * Split adding MMIO driver to QEMU arm/64 into own commit. Asherah Connor (4): x86: qemu: move QFW to its own uclass test: qemu: add qfw sandbox driver, dm tests, qemu tests qemu: add MMIO driver for QFW qemu: arm: select QFW, MMIO on qemu-arm arch/x86/cpu/qemu/cpu.c | 9 +- arch/x86/cpu/qemu/qemu.c | 49 +------ arch/x86/cpu/qfw_cpu.c | 11 +- board/emulation/qemu-arm/Kconfig | 2 + board/emulation/qemu-x86/Kconfig | 1 + cmd/qfw.c | 56 ++++--- common/Makefile | 2 + common/qfw.c | 105 +++++++++++++ configs/sandbox_spl_defconfig | 1 - drivers/misc/Kconfig | 18 ++- drivers/misc/Makefile | 7 +- drivers/misc/qfw.c | 243 +++++++++++-------------------- drivers/misc/qfw_mmio.c | 119 +++++++++++++++ drivers/misc/qfw_pio.c | 69 +++++++++ drivers/misc/qfw_sandbox.c | 128 ++++++++++++++++ include/dm/uclass-id.h | 1 + include/qfw.h | 200 +++++++++++++++++++++---- test/dm/Makefile | 1 + test/dm/qfw.c | 42 ++++++ test/py/tests/test_qfw.py | 26 ++++ 20 files changed, 817 insertions(+), 273 deletions(-) create mode 100644 common/qfw.c create mode 100644 drivers/misc/qfw_mmio.c create mode 100644 drivers/misc/qfw_pio.c create mode 100644 drivers/misc/qfw_sandbox.c create mode 100644 test/dm/qfw.c create mode 100644 test/py/tests/test_qfw.py -- 2.20.1

2 11

Ficus EE not working / supported ?
by Ezequiel Garcia 19 Mar '21

19 Mar '21

Hi Mani, Kever, I tried to bringup (again) my Ficus board, but I've been unable to boot a kernel. Seems my old u-boot branch doesn't want to boot, and the only U-Boot I managed to boot is some recent one (which I guess is better than having an ancient one). Now, doc/board/rockchip/rockchip.rst mentions TPL but Ficus defconfig isn't updated to use that. So I'm using rock960-rk3399_defconfig, with ethernet enabled, the issue is I can't boot any kernel with that. Seems it hangs up very early, and doesn't even want to output anything on earlyconsole. I was wondering if maybe this rings a bell? Maybe DDR setup? U-Boot TPL 2021.01-dirty (Mar 15 2021 - 17:46:11) Channel 0: DDR3, 800MHz BW=32 Col=10 Bk=8 CS0 Row=15 CS=1 Die BW=16 Size=1024MB Channel 1: DDR3, 800MHz BW=32 Col=10 Bk=8 CS0 Row=15 CS=1 Die BW=16 Size=1024MB 256B stride Trying to boot from BOOTROM Returning to boot ROM... U-Boot SPL 2021.01-dirty (Mar 15 2021 - 17:46:11 -0300) Trying to boot from MMC1 cannot find image node 'atf_1': -1 U-Boot 2021.01-dirty (Mar 15 2021 - 17:46:11 -0300) SoC: Rockchip rk3399 Reset cause: POR Model: 96boards RK3399 Ficus DRAM: 2 GiB PMIC: RK808 MMC: mmc@fe310000: 2, mmc@fe320000: 1, sdhci@fe330000: 0 Loading Environment from MMC... *** Warning - bad CRC, using default environment In: serial Out: serial Err: serial Model: 96boards RK3399 Ficus Net: Warning: ethernet@fe300000 (eth0) using random MAC address - f2:a3:4d:58:42:dd eth0: ethernet@fe300000 starting USB... .... .... rock960 => setenv ipaddr 192.168.0.20 rock960 => setenv ipaddr 192.168.0.200 rock960 => setenv bootcmd "tftpboot 0x02000000 rk3399-ficus/Image; tftpboot 0x01f00000 rk3399-ficus/rk3399-ficus.dtb; booti 0x02000000 - 0x01f00000" rock960 => boot Speed: 1000, full duplex Using ethernet@fe300000 device TFTP from server 192.168.0.20; our IP address is 192.168.0.200 Filename 'rk3399-ficus/Image'. Load address: 0x2000000 Loading: ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ################################################################# ############## 5.1 MiB/s done Bytes transferred = 9736704 (949200 hex) Speed: 1000, full duplex Using ethernet@fe300000 device TFTP from server 192.168.0.20; our IP address is 192.168.0.200 Filename 'rk3399-ficus/rk3399-ficus.dtb'. Load address: 0x1f00000 Loading: #### 3.7 MiB/s done Bytes transferred = 50911 (c6df hex) Moving Image from 0x2000000 to 0x2080000, end=2a1f000 ## Flattened Device Tree blob at 01f00000 Booting using the fdt blob at 0x1f00000 Host not halted after 16000 microseconds. Loading Device Tree to 0000000079f16000, end 0000000079f256de ... OK Starting kernel ... Thanks, Ezequiel

2 2

[PATCH v6 0/4] Move qfw to DM, add Arm support
by Asherah Connor 19 Mar '21

19 Mar '21

This series moves the QFW driver into a uclass, UCLASS_QFW, and splits the driver into qfw_pio and qfw_mmio. Each driver is selected on the appropriate QEMU board. A sandbox driver is also added, and a DM unit test against that driver. The qfw command is tested in QEMU, and documentation added. Version 6 splits the sandbox driver and qemu tests out of the initial conversion commit. To view the changes online, see: https://git.src.kameliya.ee/~kameliya/u-boot/log/qfw-priv Changes in v6: * Sandbox driver and tests are split into own commit. Changes in v5: * Split conversion of existing x86-only QFW to DM into its own patch. * Split MMIO driver into its own commit. * Split adding MMIO driver to QEMU arm/64 into own commit. Asherah Connor (4): x86: qemu: move QFW to its own uclass test: qemu: add qfw sandbox driver, dm tests, qemu tests qemu: add MMIO driver for QFW qemu: arm: select QFW, MMIO on qemu-arm arch/x86/cpu/qemu/cpu.c | 9 +- arch/x86/cpu/qemu/qemu.c | 49 +------ arch/x86/cpu/qfw_cpu.c | 11 +- board/emulation/qemu-arm/Kconfig | 2 + board/emulation/qemu-x86/Kconfig | 1 + cmd/qfw.c | 56 ++++--- common/Makefile | 2 + common/qfw.c | 105 +++++++++++++ drivers/misc/Kconfig | 18 ++- drivers/misc/Makefile | 7 +- drivers/misc/qfw.c | 243 +++++++++++-------------------- drivers/misc/qfw_mmio.c | 119 +++++++++++++++ drivers/misc/qfw_pio.c | 69 +++++++++ drivers/misc/qfw_sandbox.c | 128 ++++++++++++++++ include/dm/uclass-id.h | 1 + include/qfw.h | 200 +++++++++++++++++++++---- test/dm/Makefile | 1 + test/dm/qfw.c | 42 ++++++ test/py/tests/test_qfw.py | 21 +++ 19 files changed, 812 insertions(+), 272 deletions(-) create mode 100644 common/qfw.c create mode 100644 drivers/misc/qfw_mmio.c create mode 100644 drivers/misc/qfw_pio.c create mode 100644 drivers/misc/qfw_sandbox.c create mode 100644 test/dm/qfw.c create mode 100644 test/py/tests/test_qfw.py -- 2.20.1

3 9

[PATCH] Avoid delay when initializing USB peripheral by dwc2
by João Loureiro 19 Mar '21

19 Mar '21

When `usb start` is called on the terminal, the dwc2 driver will try to start every USB device as host first, even if it is explicitly configured as peripheral in the device tree (dr_mode = "peripheral"). So to avoid an unwanted 15 seconds delay when initializing the usb (one second per channel = 1s x 15), this patch adds a check to the initialization, and will skip host initialization of the device is explicitly set as peripheral. The checking is already done similarly in the `drivers/usb/gadget/dwc2_udc_otg.c` driver. Signed-off-by: João Loureiro <joaofl(a)gmail.com> --- drivers/usb/host/dwc2.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/usb/host/dwc2.c b/drivers/usb/host/dwc2.c index ec643e9f45..43cc2e0433 100644 --- a/drivers/usb/host/dwc2.c +++ b/drivers/usb/host/dwc2.c @@ -21,6 +21,7 @@ #include <asm/io.h> #include <dm/device_compat.h> #include <linux/delay.h> +#include <linux/usb/otg.h> #include <power/regulator.h> #include <reset.h> @@ -1204,7 +1205,13 @@ static int dwc2_init_common(struct udevice *dev, struct dwc2_priv *priv) #endif dwc_otg_core_init(dev); - dwc_otg_core_host_init(dev, regs); + + if (usb_get_dr_mode(dev_ofnode(dev)) == USB_DR_MODE_PERIPHERAL) { + dev_dbg(dev, "USB device %s dr_mode set to %d. Skipping host_init.\n", + dev->name, usb_get_dr_mode(dev_ofnode(dev))); + } else { + dwc_otg_core_host_init(dev, regs); + } clrsetbits_le32(&regs->hprt0, DWC2_HPRT0_PRTENA | DWC2_HPRT0_PRTCONNDET | DWC2_HPRT0_PRTENCHNG | -- 2.25.1

2 1

[PATCH 1/2] binman: Use a unique number for the symbols test file
by Simon Glass 19 Mar '21

19 Mar '21

Two test devicetree files currently have 192 as their unique number. Fix this by separating them out. Signed-off-by: Simon Glass <sjg(a)chromium.org> --- tools/binman/ftest.py | 2 +- .../test/{192_symbols_nodtb.dts => 196_symbols_nodtb.dts} | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename tools/binman/test/{192_symbols_nodtb.dts => 196_symbols_nodtb.dts} (100%) diff --git a/tools/binman/ftest.py b/tools/binman/ftest.py index 81c213a908a..34449553ca0 100644 --- a/tools/binman/ftest.py +++ b/tools/binman/ftest.py @@ -1377,7 +1377,7 @@ class TestFunctional(unittest.TestCase): def testSymbolsNoDtb(self): """Test binman can assign symbols embedded in U-Boot SPL""" - self.checkSymbols('192_symbols_nodtb.dts', + self.checkSymbols('196_symbols_nodtb.dts', U_BOOT_SPL_NODTB_DATA + U_BOOT_SPL_DTB_DATA, 0x38) diff --git a/tools/binman/test/192_symbols_nodtb.dts b/tools/binman/test/196_symbols_nodtb.dts similarity index 100% rename from tools/binman/test/192_symbols_nodtb.dts rename to tools/binman/test/196_symbols_nodtb.dts -- 2.31.0.rc2.261.g7f71774620-goog

1 1

[PATCH resend] mmc: mtk-sd: don't ignore max-frequency from device tree
by Daniel Golle 18 Mar '21

18 Mar '21

commit e58e68d9 ("mmc: mtk-sd: assign plat->cfg.f_max with a correct value") wrongly assumed that plat->cfg.f_max is always unset at the time mscd_drv_probe() is run. This is not true in case max-frequency being defined in device tree, as it is then already set by mmc_of_parse() in msdc_of_to_plat(). Only set plat->cfg.f_max to the default maximum value in case it is not already set to a sane value. Fixes: e58e68d93e ("mmc: mtk-sd: assign plat->cfg.f_max with a correct value") Cc: Stefan Roese <sr(a)denx.de> Cc: Weijie Gao <weijie.gao(a)mediatek.com> Signed-off-by: Daniel Golle <daniel(a)makrotopia.org> --- drivers/mmc/mtk-sd.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/mtk-sd.c b/drivers/mmc/mtk-sd.c index 3b9c12266a..48a764be82 100644 --- a/drivers/mmc/mtk-sd.c +++ b/drivers/mmc/mtk-sd.c @@ -1639,7 +1639,8 @@ static int msdc_drv_probe(struct udevice *dev) else cfg->f_min = host->src_clk_freq / (4 * 4095); - cfg->f_max = host->src_clk_freq; + if (cfg->f_max < cfg->f_min || cfg->f_max > host->src_clk_freq) + cfg->f_max = host->src_clk_freq; cfg->b_max = 1024; cfg->voltages = MMC_VDD_32_33 | MMC_VDD_33_34; -- 2.30.2

2 1

[PATCH] avb: Fix error when partition not found
by schspa 18 Mar '21

18 Mar '21

part_get_info_by_name will return -1 on error, and >0 on success. Signed-off-by: schspa <schspa(a)gmail.com> Cc: Igor Opaniuk <igor.opaniuk(a)gmail.com> --- common/avb_verify.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/common/avb_verify.c b/common/avb_verify.c index db10d0f21f..0520a71455 100644 --- a/common/avb_verify.c +++ b/common/avb_verify.c @@ -369,7 +369,7 @@ static struct mmc_part *get_partition(AvbOps *ops, const char *partition) } ret = part_get_info_by_name(mmc_blk, partition, &part->info); - if (!ret) { + if (ret < 0) { printf("Can't find partition '%s'\n", partition); goto err; } -- 2.24.3 (Apple Git-128)

3 2

[PATCH] image: Avoid -ENODATA in host tools
by Simon Glass 18 Mar '21

18 Mar '21

Unfortunately -ENODATA is not available in OpenBSD. Use -EBADMSG instead, to indicate a missing timestamp. Fixes: c5819701a3d image: Adjust the workings of fit_check_format() Signed-off-by: Simon Glass <sjg(a)chromium.org> --- common/image-fit.c | 2 +- include/image.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/common/image-fit.c b/common/image-fit.c index 28b3d2b1911..94501b10716 100644 --- a/common/image-fit.c +++ b/common/image-fit.c @@ -1651,7 +1651,7 @@ int fit_check_format(const void *fit, ulong size) /* mandatory / node 'timestamp' property */ if (!fdt_getprop(fit, 0, FIT_TIMESTAMP_PROP, NULL)) { log_debug("Wrong FIT format: no timestamp\n"); - return -ENODATA; + return -EBADMSG; } } diff --git a/include/image.h b/include/image.h index 138c83dd28d..b4b284d52b7 100644 --- a/include/image.h +++ b/include/image.h @@ -1158,7 +1158,7 @@ int fit_image_check_comp(const void *fit, int noffset, uint8_t comp); * @fit: pointer to the FIT format image header * @return 0 if OK, -ENOEXEC if not an FDT file, -EINVAL if the full FDT check * failed (e.g. due to bad structure), -ENOMSG if the description is - * missing, -ENODATA if the timestamp is missing, -ENOENT if the /images + * missing, -EBADMSG if the timestamp is missing, -ENOENT if the /images * path is missing */ int fit_check_format(const void *fit, ulong size); -- 2.30.0.617.g56c4b15f3c-goog

3 2