March 2021 - U-Boot

[PATCH] usb: mtu3: flush cache for next GPD
by Chunfeng Yun 03 Mar '21

03 Mar '21

When flush cache of the current GPD and resume QMU, the controller will try to access the next GPD after processing the current one, if not flush the next GPD, the controller may get wrong GPD status. Signed-off-by: Chunfeng Yun <chunfeng.yun(a)mediatek.com> --- drivers/usb/mtu3/mtu3_qmu.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/mtu3/mtu3_qmu.c b/drivers/usb/mtu3/mtu3_qmu.c index 801c2bc416..95eaf6d236 100644 --- a/drivers/usb/mtu3/mtu3_qmu.c +++ b/drivers/usb/mtu3/mtu3_qmu.c @@ -198,6 +198,7 @@ static int mtu3_prepare_tx_gpd(struct mtu3_ep *mep, struct mtu3_request *mreq) enq->flag &= ~GPD_FLAGS_HWO; gpd->next_gpd = cpu_to_le32((u32)gpd_virt_to_dma(ring, enq)); + mtu3_flush_cache((uintptr_t)enq, sizeof(*gpd)); if (req->zero) gpd->ext_flag |= GPD_EXT_FLAG_ZLP; @@ -234,6 +235,8 @@ static int mtu3_prepare_rx_gpd(struct mtu3_ep *mep, struct mtu3_request *mreq) enq->flag &= ~GPD_FLAGS_HWO; gpd->next_gpd = cpu_to_le32((u32)gpd_virt_to_dma(ring, enq)); + mtu3_flush_cache((uintptr_t)enq, sizeof(*gpd)); + gpd->flag |= GPD_FLAGS_IOC | GPD_FLAGS_HWO; mreq->gpd = gpd; -- 2.18.0

2 1

[PATCH 0/5] lmb: Add config for the number of memory and reserved regions
by Patrick Delaunay 03 Mar '21

03 Mar '21

Hi, I propose a update of the lmb library to allow the configuration with Kconfig of the number of memory and reserved regions in lmb libary: - CONFIG_LMB_RESERVED_REGIONS - CONFIG_LMB_MEMORY_REGIONS By default, I keep the default value of 8 regions. This serie avoids issue on stm32mp15 platform, as the kernel device tree defines many and non contiguous reserved regions. Patrick Delaunay (5): lmb: move CONFIG_LMB in Kconfig lmb: remove lmb_region.size lmb: Move lmb property arrays in struct lmb lmb: Add 2 config to define the max number of regions configs: stm32mp15: increase the number of reserved memory region in lmb arch/arc/include/asm/config.h | 2 -- arch/arm/include/asm/config.h | 1 - arch/m68k/include/asm/config.h | 1 - arch/microblaze/include/asm/config.h | 2 -- arch/mips/include/asm/config.h | 1 - arch/nds32/include/asm/config.h | 1 - arch/powerpc/include/asm/config.h | 1 - arch/riscv/include/asm/config.h | 1 - arch/sh/include/asm/config.h | 2 -- arch/x86/include/asm/config.h | 1 - arch/xtensa/include/asm/config.h | 2 -- configs/stm32mp15_basic_defconfig | 1 + configs/stm32mp15_trusted_defconfig | 1 + include/configs/10m50_devboard.h | 5 ---- include/configs/3c120_devboard.h | 5 ---- include/configs/sandbox.h | 2 -- include/configs/x86-common.h | 2 -- include/image.h | 2 +- include/lmb.h | 34 ++++++++++++++++++++++++---- lib/Kconfig | 23 +++++++++++++++++++ lib/lmb.c | 14 ++++++------ scripts/config_whitelist.txt | 1 - 22 files changed, 63 insertions(+), 42 deletions(-) -- 2.17.1

3 7

Commit 401d1c4f5d2 ("common: Drop asm/global_data.h from common header") broke CONFIG_CMD_SQUASHFS=y
by Pali Rohár 03 Mar '21

03 Mar '21

Hello! Commit 401d1c4f5d2d29c4bc4beaec95402ca23eb63295 ("common: Drop asm/global_data.h from common header") broke compilation of squashfs filesystem when CONFIG_CMD_SQUASHFS=y is enabled. Here is console output: fs/squashfs/sqfs_inode.c: In function ‘sqfs_find_inode’: fs/squashfs/sqfs_inode.c:121:18: warning: implicit declaration of function ‘le32_to_cpu’ [-Wimplicit-function-declaration] for (k = 0; k < le32_to_cpu(inode_count); k++) { ^~~~~~~~~~~ LD u-boot aarch64-linux-gnu-ld.bfd: fs/built-in.o: in function `sqfs_find_inode': /u-boot/fs/squashfs/sqfs_inode.c:126: undefined reference to `le32_to_cpu' aarch64-linux-gnu-ld.bfd: /u-boot/fs/squashfs/sqfs_inode.c:121: undefined reference to `le32_to_cpu' Segmentation fault make: *** [Makefile:1765: u-boot] Error 139 make: *** Deleting file 'u-boot' When compiling U-Boot prior that commit, there is no error. Simon and Tom: you have signed problematic mentioned commit. Could you please look at it?

2 5

[PATCH 0/3 v5] Add ESRT and test ESRT creation
by Jose Marinho 03 Mar '21

03 Mar '21

The following 3 commits add the ESRT and provide a test of the functionality. The first commit adds the ESRT as defined in the UEFI 2.8 specification. An empty ESRT is created during the execution of the efi_init_obj_list(). The ESRT is updated when: 1) a FMP protocol is installed in the system: this will add the corresponding entries to the ESRT. 2) a capsule is installed via UpdateCapsule: this should update entries already present in the ESRT. This implementation of the ESRT creation takes input from FMP only. It is assumed that the FMP will maintain the following values across reboot: - LastAttemptVersion. - LastAttemptStatus. The second commit introduces a helper cmd function to print the ESRT in the u-boot shell. The third commit enables testing the ESRT creation in the sandbox platform. That commit is composed of 2 tests. - Test 1 executes from the u-boot shell with "bootefi selftest". - Test 2 executes in the pytest environment. Patch v5: - Address v4 comments. - split the v4 2/2 commit into the v5 2/3 (debug print ESRT) and 3/3 ESRT tests. Patch v4: - update stale [Patch 1/2 v3] commit message. Patch v3: - Address v2 comments. Patch v2: - The ESRT is now regenerated from scratch at every FMP EVT_NOTIFY_SIGNAL and whenever a capsule is updated. - Extended TestEfiCapsuleFirmwareFit::test_efi_capsule_fw3 to verify that the ESRT is correctly populated after an UpdateCapsule. - Addressed v1 comments. Patch v1: - reworked the ESRT creation code, allowing table to resize as FMPs are installed. - registered a callback for the FMP protocol install. - Created a unit test running on the sandbox platform. rfc: initial patch submission CC: Heinrich Schuchardt <xypron.glpk(a)gmx.de> CC: Sughosh Ganu <sughosh.ganu(a)linaro.org> CC: AKASHI Takahiro <takahiro.akashi(a)linaro.org> CC: Ilias Apalodimas <ilias.apalodimas(a)linaro.org> CC: Andre Przywara <andre.przywara(a)arm.com> CC: Alexander Graf <agraf(a)csgraf.de> CC: nd(a)arm.com Jose Marinho (3): efi: Add ESRT to the EFI system table cmd: efi: ESRT table debug print efi: ESRT creation tests cmd/efidebug.c | 68 +++ include/efi_api.h | 21 + include/efi_loader.h | 24 + lib/efi_loader/Kconfig | 7 + lib/efi_loader/Makefile | 1 + lib/efi_loader/efi_boottime.c | 7 +- lib/efi_loader/efi_capsule.c | 8 + lib/efi_loader/efi_esrt.c | 510 ++++++++++++++++++ lib/efi_loader/efi_setup.c | 6 + lib/efi_selftest/Makefile | 2 + lib/efi_selftest/efi_selftest_esrt.c | 227 ++++++++ .../test_efi_capsule/test_capsule_firmware.py | 4 + 12 files changed, 881 insertions(+), 4 deletions(-) create mode 100644 lib/efi_loader/efi_esrt.c create mode 100644 lib/efi_selftest/efi_selftest_esrt.c -- 2.17.1

4 9

[PATCH] Revert "mmc: fsl_esdhc_imx: remove the 1ms delay before sending command"
by Fabio Estevam 03 Mar '21

03 Mar '21

Removing the 1ms delay before sending command causes a regression on imx8mq-evk where the SD card cannot be accessed. Since this 1ms delay has been present since the driver introduction in 2008, keep it to avoid regressions. This reverts commit 9098682200e6cca4b776638a51200dafa16f50fb. Reported-by: Ye Li <ye.li(a)nxp.com> Signed-off-by: Fabio Estevam <festevam(a)gmail.com> --- drivers/mmc/fsl_esdhc_imx.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/mmc/fsl_esdhc_imx.c b/drivers/mmc/fsl_esdhc_imx.c index e0e132698e30..a0a5c8960563 100644 --- a/drivers/mmc/fsl_esdhc_imx.c +++ b/drivers/mmc/fsl_esdhc_imx.c @@ -463,6 +463,13 @@ static int esdhc_send_cmd_common(struct fsl_esdhc_priv *priv, struct mmc *mmc, while (esdhc_read32(&regs->prsstat) & PRSSTAT_DLA) ; + /* Wait at least 8 SD clock cycles before the next command */ + /* + * Note: This is way more than 8 cycles, but 1ms seems to + * resolve timing issues with some cards + */ + udelay(1000); + /* Set up for a data transfer if we have one */ if (data) { err = esdhc_setup_data(priv, mmc, data); -- 2.25.1

2 1

[PATCH 1/1] video: buffer overrun in TrueType console
by Heinrich Schuchardt 03 Mar '21

03 Mar '21

When scrolling the TrueType console a buffer overrun occurs. Fixes: a29b012037cc ("video: Add a console driver that uses TrueType fonts") Signed-off-by: Heinrich Schuchardt <xypron.glpk(a)gmx.de> --- drivers/video/console_truetype.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) diff --git a/drivers/video/console_truetype.c b/drivers/video/console_truetype.c index b9d467ac90..6d40be6482 100644 --- a/drivers/video/console_truetype.c +++ b/drivers/video/console_truetype.c @@ -128,38 +128,36 @@ static int console_truetype_set_row(struct udevice *dev, uint row, int clr) struct video_priv *vid_priv = dev_get_uclass_priv(dev->parent); struct console_tt_priv *priv = dev_get_priv(dev); void *end, *line; - int pixels = priv->font_size * vid_priv->line_length; - int i, ret; + int ret; line = vid_priv->fb + row * priv->font_size * vid_priv->line_length; + end = line + priv->font_size * vid_priv->line_length; + switch (vid_priv->bpix) { #ifdef CONFIG_VIDEO_BPP8 case VIDEO_BPP8: { - uint8_t *dst = line; + u8 *dst; - for (i = 0; i < pixels; i++) - *dst++ = clr; - end = dst; + for (dst = line; dst < (u8 *)end; ++dst) + *dst = clr; break; } #endif #ifdef CONFIG_VIDEO_BPP16 case VIDEO_BPP16: { - uint16_t *dst = line; + u16 *dst = line; - for (i = 0; i < pixels; i++) - *dst++ = clr; - end = dst; + for (dst = line; dst < (u16 *)end; ++dst) + *dst = clr; break; } #endif #ifdef CONFIG_VIDEO_BPP32 case VIDEO_BPP32: { - uint32_t *dst = line; + u32 *dst = line; - for (i = 0; i < pixels; i++) - *dst++ = clr; - end = dst; + for (dst = line; dst < (u32 *)end; ++dst) + *dst = clr; break; } #endif -- 2.30.1

2 1

[PATCH] MAINTAINERS: Update info for Raspberry Pi
by matthias.bgg＠kernel.org 03 Mar '21

03 Mar '21

From: Matthias Brugger <mbrugger(a)suse.com> Add RPi config files and custodian repository. Signed-off-by: Matthias Brugger <mbrugger(a)suse.com> --- MAINTAINERS | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index b82b7adbeb..15929bf4ee 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -160,7 +160,7 @@ F: drivers/clk/aspeed/ F: drivers/pinctrl/aspeed/ N: aspeed -ARM BROADCOM BCM283X +ARM BROADCOM BCM283X / BCM27XX M: Matthias Brugger <mbrugger(a)suse.com> S: Maintained F: arch/arm/dts/bcm283* @@ -175,6 +175,8 @@ F: drivers/video/bcm2835.c F: include/dm/platform_data/serial_bcm283x_mu.h F: include/dt-bindings/pinctrl/bcm2835.h F: drivers/pinctrl/broadcom/ +F: configs/rpi_* +T: git https://source.denx.de/u-boot/custodians/u-boot-arm.git ARM BROADCOM BCMSTB M: Thomas Fitzsimmons <fitzsim(a)fitzsim.org> -- 2.30.0

2 1

[PATCH 0/5] cmd: Add support for command substitution
by Sean Anderson 03 Mar '21

03 Mar '21

This series adds support for command substitution using $(). The motivation for this is that we have many commands which have a version which prints some information, and one which sets an environmental variable with that info. This is a bit clunky, since every time someone wants to grab info from a command (e.g. [1], [2]), they have to add a version which sets a variable. Every command also has subtle differences in the syntax of how the environmental is specified. The classic way to do this is to redirect the output of the command into a variable using command substitution. Unfortunately, this approach will not work in U-Boot for a few reasons. First, we don't have any jobs, so we can't use pipelines to filter output. In addition, the command output tends to be very "human-oriented" and would be hard to parse, even if something like console recording were used. Instead, commands just set gd->cmd_result. This result is then printed by the caller of cmd_process. Unfortunately, this means that to support command substitution, commands must be modified to set cmd_result. However, they already would have needed modification to support setting an environmental variable. Printing out the result of the command may not be the best approach here. If a command which prints out many messages already returns a result, just printing the result on its own does not clearly convey what the result means. Perhaps a prefix should be added to printed results? Though this would require echo to be changed. This initial series just converts two commands: echo and part uuid. This showcases the manner in which this new functionality could be implemented. I think that some helper functions like for env may be useful so that commands don't have to do some much bookkeeping (like in the current echo command). This series depends on [3]. [1] https://patchwork.ozlabs.org/project/uboot/patch/20210226181733.19307-1-far… [2] https://patchwork.ozlabs.org/project/uboot/patch/20201215165439.13165-1-m.s… [3] https://patchwork.ozlabs.org/project/uboot/patch/20210228212951.1175231-1-s… Sean Anderson (5): hush: Print syntax error line with DEBUG_SHELL cmd: Add support for (limited) command substitution cmd: Convert echo to use cmd_result test: hush: Add test for command substitution cmd: Convert part uuid to use cmd_result cmd/echo.c | 33 +++++++++++++++---- cmd/part.c | 18 ++++++++--- common/cli_hush.c | 53 +++++++++++++++++++++++++------ common/cli_simple.c | 8 ++++- common/command.c | 3 ++ include/asm-generic/global_data.h | 4 +++ test/cmd/test_echo.c | 10 +++++- 7 files changed, 107 insertions(+), 22 deletions(-) -- 2.30.1

3 14

[BUG] segmentation fault in efi_firmware_fit_get_image_info()
by Heinrich Schuchardt 03 Mar '21

03 Mar '21

Hello Takahiro, Jose's ESRT implementation calls efi_firmware_fit_get_image_info() multiple times. This results in a segmentation fault in the depths of dfu_free_entities() involving device_chld_remove(). I used env set dfu_alt_info "sf 0:0=u-boot-bin raw 0x100000 0x50000;u-boot-env raw 0x150000 0x200000" for testing on sandbox_defconfig. If I comment out the dfu_free_entities() calls the segmentation fault is avoided. Could you, please, check with Lukasz when and if you should call dfu_free_entities() in efi_get_dfu_info(). Best regards Heinrich

2 1

[GIT PULL] rpi: second round of updates for v2021.04
by Matthias Brugger 02 Mar '21

02 Mar '21

Hi Tom, Below are the second round of updates for Raspberry Pi aimed for the v2021.04 release. The green CI loop can be found here: https://source.denx.de/u-boot/custodians/u-boot-raspberrypi/-/pipelines/6579 Please have a look. Regards, Matthias --- The following changes since commit c5219c4a18f2b27547ecd799914f94e48b0fa86f: Prepare v2021.04-rc3 (2021-03-01 22:46:41 -0500) are available in the Git repository at: https://source.denx.de/u-boot/custodians/u-boot-raspberrypi.git tags/rpi-next-2021.04.2 for you to fetch changes up to fae165b5a1991ca55acfb3dc78f3c473275b3faf: configs: rpi4_32: Enable iProc RNG200 (2021-03-02 13:53:53 +0100) ---------------------------------------------------------------- - Disable Grub workaround for RPi2 - enable HS200 mode for iproc sdhci - add armv7 support for iproc_rng200 ---------------------------------------------------------------- Bharat Gooty (1): drivers: mmc: iproc_sdhci: enable HS200 mode Matthias Brugger (4): configs: RPi2: Disable EFI Grub workaround MAINTAINERS: Update info for Raspberry Pi rng: iproc_rng200: Enable support for RPi4 armv7 configs: rpi4_32: Enable iProc RNG200 MAINTAINERS | 4 +- configs/rpi_2_defconfig | 1 + configs/rpi_4_32b_defconfig | 2 + drivers/mmc/iproc_sdhci.c | 92 ++++++++++++++++++++++++++++++++++++++++----- drivers/rng/iproc_rng200.c | 8 ++-- 5 files changed, 93 insertions(+), 14 deletions(-)

2 1