November 2023 - U-Boot

[PATCH v3 0/5] cmd: add scmi command
by AKASHI Takahiro 29 Nov '23

29 Nov '23

"Scmi" command will be re-introduced per Michal's request. The functionality is the same as I put it in my patch set of adding SCMI base protocol support, but made some tweak to make UT, "ut dm scmi_cmd," more flexible and tolerable when enabling/disabling a specific SCMI protocol for test purpose. Each commit may have some change history inherited from the preceding patch series. Test ==== The patch series was tested on the following platforms: * sandbox Prerequisite: ============= * This patch series is based on the latest master. Changes: ======== v3(Nov 14, 2023) * return -EAGAIN if we want to skip a test. * use CONFIG_IS_ENABLED() rather than IS_ENABLED(). * remove goto by introducing a function in sandbox implementation. v2(Nov 13, 2023) * localize global variables to avoid pytest errors. AKASHI Takahiro (5): test: dm: skip scmi tests against disabled protocols firmware: scmi: support protocols on sandbox only if enabled cmd: add scmi command for SCMI firmware doc: cmd: add documentation for scmi test: dm: add scmi command test cmd/Kconfig | 9 + cmd/Makefile | 1 + cmd/scmi.c | 384 +++++++++++++++++++ configs/sandbox_defconfig | 1 + doc/usage/cmd/scmi.rst | 126 ++++++ doc/usage/index.rst | 1 + drivers/firmware/scmi/sandbox-scmi_agent.c | 30 +- drivers/firmware/scmi/sandbox-scmi_devices.c | 78 ++-- test/dm/scmi.c | 93 +++++ 9 files changed, 686 insertions(+), 37 deletions(-) create mode 100644 cmd/scmi.c create mode 100644 doc/usage/cmd/scmi.rst -- 2.34.1

3 8

[PATCH v2 0/3] Xilinx SPI driver enhancements
by Mayuresh Chitale 29 Nov '23

29 Nov '23

This series makes the Xilinx XPS SPI driver compatible with the MMC_SPI driver and also includes an enhacement from the corresponding Linux driver. Changes in v2: - Improve comment and add reviewed-by tag in patch 3 Mayuresh Chitale (3): drivers: xilinx_spi: Use udevice in start_tranfer drivers: xilinx_spi: Add xfer callback drivers: xilinx_spi: Probe fifo_depth at runtime drivers/spi/xilinx_spi.c | 68 ++++++++++++++++++++++++++++++---------- 1 file changed, 52 insertions(+), 16 deletions(-) -- 2.34.1

3 6

[PATCH v2 1/3] arm: mvebu: Espressobin: move FDT fixup into a separate function
by Robert Marko 29 Nov '23

29 Nov '23

Currently, Esspresobin FDT is being fixed up directly in ft_board_setup() which makes it hard to add support for any other board to be fixed up. So, lets just move the FDT fixup code to a separate function and call it if compatible matches, there should be no functional change. Signed-off-by: Robert Marko <robert.marko(a)sartura.hr> Reviewed-by: Stefan Roese <sr(a)denx.de> --- board/Marvell/mvebu_armada-37xx/board.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/board/Marvell/mvebu_armada-37xx/board.c b/board/Marvell/mvebu_armada-37xx/board.c index 04124d8014..1471caa9a6 100644 --- a/board/Marvell/mvebu_armada-37xx/board.c +++ b/board/Marvell/mvebu_armada-37xx/board.c @@ -363,18 +363,14 @@ EVENT_SPY_SIMPLE(EVT_LAST_STAGE_INIT, last_stage_init); #endif #ifdef CONFIG_OF_BOARD_SETUP -int ft_board_setup(void *blob, struct bd_info *bd) +static int espressobin_fdt_setup(void *blob) { -#ifdef CONFIG_ENV_IS_IN_SPI_FLASH int ret; int spi_off; int parts_off; int part_off; /* Fill SPI MTD partitions for Linux kernel on Espressobin */ - if (!of_machine_is_compatible("globalscale,espressobin")) - return 0; - spi_off = fdt_node_offset_by_compatible(blob, -1, "jedec,spi-nor"); if (spi_off < 0) return 0; @@ -459,6 +455,14 @@ int ft_board_setup(void *blob, struct bd_info *bd) return 0; } + return 0; +} + +int ft_board_setup(void *blob, struct bd_info *bd) +{ +#ifdef CONFIG_ENV_IS_IN_SPI_FLASH + if (of_machine_is_compatible("globalscale,espressobin")) + return espressobin_fdt_setup(blob); #endif return 0; } -- 2.41.0

2 6

Re: [PATCH v2 1/3] drivers: xilinx_spi: Use udevice in start_tranfer
by Kumar, Love 29 Nov '23

29 Nov '23

Tested-by: Love Kumar <love.kumar(a)amd.com> On 16/11/23, 10:13 PM, "U-Boot" <u-boot-bounces(a)lists.denx.de> wrote: Modify start_transfer and related functions to take a udevice parameter as input instead of spi_slave. This is needed so that start_transfer can be used directly via the xfer callback. Also fix a compiler warning. Signed-off-by: Mayuresh Chitale <mchitale(a)ventanamicro.com<mailto:mchitale@ventanamicro.com>> --- drivers/spi/xilinx_spi.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) diff --git a/drivers/spi/xilinx_spi.c b/drivers/spi/xilinx_spi.c index b58a3f632a..d9faceec18 100644 --- a/drivers/spi/xilinx_spi.c +++ b/drivers/spi/xilinx_spi.c @@ -67,7 +67,7 @@ /* SPI Slave Select Register (spissr), [1] p13, [2] p13 */ #define SPISSR_MASK(cs) (1 << (cs)) #define SPISSR_ACT(cs) ~SPISSR_MASK(cs) -#define SPISSR_OFF ~0UL +#define SPISSR_OFF (~0U) /* SPI Software Reset Register (ssr) */ #define SPISSR_RESET_VALUE 0x0a @@ -217,9 +217,9 @@ static u32 xilinx_spi_read_rxfifo(struct udevice *bus, u8 *rxp, u32 rxbytes) return i; } -static int start_transfer(struct spi_slave *spi, const void *dout, void *din, u32 len) +static int start_transfer(struct udevice *dev, const void *dout, void *din, u32 len) { - struct udevice *bus = spi->dev->parent; + struct udevice *bus = dev->parent; struct xilinx_spi_priv *priv = dev_get_priv(bus); struct xilinx_spi_regs *regs = priv->regs; u32 count, txbytes, rxbytes; @@ -259,10 +259,9 @@ static int start_transfer(struct spi_slave *spi, const void *dout, void *din, u3 return 0; } -static void xilinx_spi_startup_block(struct spi_slave *spi) +static void xilinx_spi_startup_block(struct udevice *dev) { - struct dm_spi_slave_plat *slave_plat = - dev_get_parent_plat(spi->dev); + struct dm_spi_slave_plat *slave_plat = dev_get_parent_plat(dev); unsigned char txp; unsigned char rxp[8]; @@ -270,13 +269,13 @@ static void xilinx_spi_startup_block(struct spi_slave *spi) * Perform a dummy read as a work around for * the startup block issue. */ - spi_cs_activate(spi->dev, slave_plat->cs); + spi_cs_activate(dev, slave_plat->cs); txp = 0x9f; - start_transfer(spi, (void *)&txp, NULL, 1); + start_transfer(dev, (void *)&txp, NULL, 1); - start_transfer(spi, NULL, (void *)rxp, 6); + start_transfer(dev, NULL, (void *)rxp, 6); - spi_cs_deactivate(spi->dev); + spi_cs_deactivate(dev); } static int xilinx_spi_mem_exec_op(struct spi_slave *spi, @@ -294,14 +293,15 @@ static int xilinx_spi_mem_exec_op(struct spi_slave *spi, * as QSPI provides command. So first command fails. */ if (!startup) { - xilinx_spi_startup_block(spi); + xilinx_spi_startup_block(spi->dev); startup++; } spi_cs_activate(spi->dev, slave_plat->cs); if (op->cmd.opcode) { - ret = start_transfer(spi, (void *)&op->cmd.opcode, NULL, 1); + ret = start_transfer(spi->dev, (void *)&op->cmd.opcode, + NULL, 1); if (ret) goto done; } @@ -313,7 +313,7 @@ static int xilinx_spi_mem_exec_op(struct spi_slave *spi, addr_buf[i] = op->addr.val >> (8 * (op->addr.nbytes - i - 1)); - ret = start_transfer(spi, (void *)addr_buf, NULL, + ret = start_transfer(spi->dev, (void *)addr_buf, NULL, op->addr.nbytes); if (ret) goto done; @@ -322,16 +322,16 @@ static int xilinx_spi_mem_exec_op(struct spi_slave *spi, dummy_len = (op->dummy.nbytes * op->data.buswidth) / op->dummy.buswidth; - ret = start_transfer(spi, NULL, NULL, dummy_len); + ret = start_transfer(spi->dev, NULL, NULL, dummy_len); if (ret) goto done; } if (op->data.nbytes) { if (op->data.dir == SPI_MEM_DATA_IN) { - ret = start_transfer(spi, NULL, + ret = start_transfer(spi->dev, NULL, op->data.buf.in, op->data.nbytes); } else { - ret = start_transfer(spi, op->data.buf.out, + ret = start_transfer(spi->dev, op->data.buf.out, NULL, op->data.nbytes); } if (ret) -- 2.34.1

1 0

[PATCH 0/6] board: siemens: clean up subfolders
by Enrico Leto 29 Nov '23

29 Nov '23

This serie depends on the serie: [PATCH 0/6] siemens,am335x: clean up the draco board family The common folder was initialially created for the common parts of the products based on draco-am355x board family. We have the product lines 'pxm2', 'rut' and the base line unfortunately named 'draco'! Adding the new capricorn-imx8 board family, the files were enhanced without cleanup. Simplify first EEPROM probe and access that implements both i2c with & without driver model. Use abstraction functions for this. Move all am355x specifics to a new file 'board_am335x'. Clean-up includes, config checks, maintainer. Signed-off-by: Enrico Leto <enrico.leto(a)siemens.com> --- Enrico Leto (6): siemens: eeprom: clean up definitions siemens: eeprom: simplify setup & read siemens draco: i2c: use driver model for u-boot siemens: board: clean up draco products vs common siemens: board: clean up includes siemens: factoryset: use correct config for soc specific implementation board/siemens/capricorn/board.c | 5 +- board/siemens/common/board.c | 110 ++++++++++++---------------- board/siemens/common/board.h | 24 ++++++ board/siemens/common/board_am335x.c | 58 +++++++++++++++ board/siemens/common/board_am335x.h | 38 ++++++++++ board/siemens/common/factoryset.c | 82 ++++----------------- board/siemens/draco/Makefile | 2 + board/siemens/draco/board.c | 59 ++++----------- board/siemens/draco/board.h | 32 ++------ board/siemens/draco/mux.c | 6 +- board/siemens/pxm2/MAINTAINERS | 2 +- board/siemens/pxm2/Makefile | 2 + board/siemens/pxm2/board.c | 34 +++------ board/siemens/pxm2/board.h | 21 ------ board/siemens/pxm2/mux.c | 6 +- board/siemens/rut/MAINTAINERS | 2 +- board/siemens/rut/Makefile | 2 + board/siemens/rut/board.c | 33 ++------- board/siemens/rut/board.h | 21 ------ board/siemens/rut/mux.c | 5 +- configs/draco-etamin_defconfig | 4 +- configs/draco-rastaban_defconfig | 4 +- configs/draco-thuban_defconfig | 4 +- include/configs/draco-etamin.h | 3 - include/configs/draco-rastaban.h | 3 - include/configs/draco-thuban.h | 3 - 26 files changed, 245 insertions(+), 320 deletions(-) create mode 100644 board/siemens/common/board.h create mode 100644 board/siemens/common/board_am335x.c create mode 100644 board/siemens/common/board_am335x.h delete mode 100644 board/siemens/pxm2/board.h delete mode 100644 board/siemens/rut/board.h -- 2.30.2

2 10

[PATCH v2] efi_loader: Make DisconnectController follow the EFI spec
by Ilias Apalodimas 29 Nov '23

29 Nov '23

commit 239d59a65e20 ("efi_loader: reconnect drivers on failure") tried to fix the UninstallProtocol interface which must reconnect any controllers it disconnected by calling ConnectController() in case of failure. However, the reconnect functionality was wired in efi_disconnect_all_drivers() instead of efi_uninstall_protocol(). As a result some SCT tests started failing. Specifically, BBTestOpenProtocolInterfaceTest333CheckPoint3() test - Calls ConnectController for DriverImageHandle1 - Calls DisconnectController for DriverImageHandle1 which will disconnect everything apart from TestProtocol4. That will remain open on purpose. - Calls ConnectController for DriverImageHandle2. TestProtocol4 which was explicitly preserved was installed wth BY_DRIVER attributes. The new protocol will call DisconnectController since its attributes are BY_DRIVER|EXCLUSIVE, but TestProtocol4 will not be removed. The test expects EFI_ACCESS_DENIED which works fine. The problem is that DisconnectController, will eventually call EFI_DRIVER_BINDING_PROTOCOL.Stop(). But on the aforementioned test this will call CloseProtocol -- the binding protocol is defined in 'DBindingDriver3.c' and the .Stop function uses CloseProtocol. If that close protocol call fails with EFI_NOT_FOUND, the current code will try to mistakenly reconnect all drivers and the subsequent tests that rely on the device being disconnected will fail. Move the reconnection in efi_uninstall_protocol() were it belongs. Fixes: commit 239d59a65e20 ("efi_loader: reconnect drivers on failure") Signed-off-by: Ilias Apalodimas <ilias.apalodimas(a)linaro.org> --- Apologies for the fast resend, but since Heinrich reviewed it and we want it in 2024.01 resending Changes since v1: - return ret instead of (ret != EFI_SUCCESS ? ret : EFI_SUCCESS) which does the same thing... - Add a comment about reconnecting all controllers lib/efi_loader/efi_boottime.c | 27 ++++++++++----------------- 1 file changed, 10 insertions(+), 17 deletions(-) diff --git a/lib/efi_loader/efi_boottime.c b/lib/efi_loader/efi_boottime.c index 0b7579cb5af1..ea711919630b 100644 --- a/lib/efi_loader/efi_boottime.c +++ b/lib/efi_loader/efi_boottime.c @@ -1339,7 +1339,7 @@ static efi_status_t efi_disconnect_all_drivers const efi_guid_t *protocol, efi_handle_t child_handle) { - efi_uintn_t number_of_drivers, tmp; + efi_uintn_t number_of_drivers; efi_handle_t *driver_handle_buffer; efi_status_t r, ret; @@ -1350,27 +1350,13 @@ static efi_status_t efi_disconnect_all_drivers if (!number_of_drivers) return EFI_SUCCESS; - tmp = number_of_drivers; while (number_of_drivers) { - ret = EFI_CALL(efi_disconnect_controller( + r = EFI_CALL(efi_disconnect_controller( handle, driver_handle_buffer[--number_of_drivers], child_handle)); - if (ret != EFI_SUCCESS) - goto reconnect; - } - - free(driver_handle_buffer); - return ret; - -reconnect: - /* Reconnect all disconnected drivers */ - for (; number_of_drivers < tmp; number_of_drivers++) { - r = EFI_CALL(efi_connect_controller(handle, - &driver_handle_buffer[number_of_drivers], - NULL, true)); if (r != EFI_SUCCESS) - EFI_PRINT("Failed to reconnect controller\n"); + ret = r; } free(driver_handle_buffer); @@ -1409,6 +1395,13 @@ static efi_status_t efi_uninstall_protocol r = efi_disconnect_all_drivers(handle, protocol, NULL); if (r != EFI_SUCCESS) { r = EFI_ACCESS_DENIED; + /* + * This will reconnect all controllers of the handle, even ones + * that were not connected before. This can be done better + * but we are following the EDKII implementation on this for + * now + */ + EFI_CALL(efi_connect_controller(handle, NULL, NULL, true)); goto out; } /* Close protocol */ -- 2.37.2

2 2

[PATCH v3 0/9] fs: fat: calculate FAT type based on cluster count
by christian.taedcke-oss＠weidmueller.com 29 Nov '23

29 Nov '23

From: Christian Taedcke <christian.taedcke(a)weidmueller.com> This series fixes an issue where the FAT type (FAT12, FAT16) is not correctly detected, e.g. when the BPB field BS_FilSysType contains the valid value "FAT ". This issue occures, for example, if a partition is formatted by swupdate using its diskformat handler. swupdate uses the FAT library from http://elm-chan.org/fsw/ff/ internally. See https://groups.google.com/g/swupdate/c/7Yc3NupjXx8 for a discussion in the swupdate mailing list. Please refer to the commit messages for more details. 1. Added bootsector checks Most tests from https://www.win.tue.nl/~aeb/linux/fs/fat/fat-2.html are added in the commit 'fs: fat: add bootsector validity check'. Only the tests VIII, IX and X are not implemented. I also checked the Linux kernel code (v6.6) and did not find any checks on 'vistart->fs_type'. This is the reason why is skipped them here. See section '2. Size comparisons' for the impact on the binary size. 2. Size comparisons I executed bloat-o-meter from the Linux kernel for an arm64 target (config xilinx_zynqmp_mini_emmc0_defconfig): Comparison of the binary spl/u-boot-spl between master (rev e17d174773e9ba9447596708e702b7382e47a6cf) and this patch series (including the added validity checks of the boot sector): add/remove: 0/0 grow/shrink: 1/1 up/down: 100/-12 (88) Function old new delta read_bootsectandvi 308 408 +100 fat_itr_root 444 432 -12 Total: Before=67977, After=68065, chg +0.13% When compare the size of the binary spl/u-boot-spl between master this series without the the validity checks of the boot sector: add/remove: 0/0 grow/shrink: 0/2 up/down: 0/-24 (-24) Function old new delta read_bootsectandvi 308 296 -12 fat_itr_root 444 432 -12 Total: Before=67977, After=67953, chg -0.04% So the size of the spl on this arm64 target increases by 88 bytes for this series. When i remove the validity check the size decreases by 24 bytes. Changes in v3: - Enable some fs tests for FAT12 - Add test to detect fat fs type Changes in v2: - Use get_unaligned_le16 instead of custom macro - Move code change in test/image/spl_load_fs.c into separate commit - Use get_unaligned_le16 instead of custom macro - Extract FAT32 logic from determine_fat_bits() into read_bootsectandvi() - Rename determine_fat_bits() to determine_legacy_fat_bits() - Keep one goto label instead of removing both - Move changing gotos from read_bootsectandvi() into separate commit - Add validity checks for boot sector Christian Taedcke (9): fs: fat: use get_unaligned_le16 to convert u8[2] to u16 test: spl: Remove usage of FAT32_SIGN fs: fat: calculate FAT type based on cluster count fs: fat: simplify gotos from read_bootsectandvi fs: fat: add bootsector validity check test: Add support to create a fat12 fs test: fs: Add fat12 to supported fs of some tests test: Add size granularity parameter to mk_fs test: fs: Add test to detect fat type fs/fat/fat.c | 114 ++++++++++++++++++++------- include/fat.h | 6 -- test/image/spl_load_fs.c | 2 +- test/py/tests/fs_helper.py | 11 ++- test/py/tests/test_fs/conftest.py | 53 ++++++++++++- test/py/tests/test_fs/test_fs_fat.py | 25 ++++++ 6 files changed, 170 insertions(+), 41 deletions(-) create mode 100644 test/py/tests/test_fs/test_fs_fat.py -- 2.34.1

2 10

[PATCH 00/13] Import "string" I/O functions from Linux
by Igor Prusov 29 Nov '23

29 Nov '23

This series imports generic versions of ioread_rep/iowrite_rep and reads/writes from Linux. Some cleanup is done to make sure that all platforms have proper defines for implemented functions and there are no redefinitions. Igor Prusov (13): sandbox: move asm-generic include to the end of file x86: Add defines for ins/outs functions mips: io.h: Add const to reads functions params mips: io.h: Add defines for read/write/in/out functions riscv: io.h: Add defines for reads/writes functions riscv: io.h: Fix signatures of reads/writes functions nios2: io.h: Add defines for ins/outs functions powerpc: io.h: Add defines for __raw_{read,write} functions xtensa: io.h: Add defines for ins/outs functions asm-generic: Import functions from Linux spi: meson_spifc_a1: Switch to io{read,write}32_rep() treewide: Include linux/io.h instead of asm-generic/io.h musb-new: Remove implementation of io.h functions arch/mips/include/asm/io.h | 35 +- arch/nios2/include/asm/io.h | 6 + arch/powerpc/include/asm/io.h | 11 + arch/riscv/include/asm/io.h | 26 +- arch/sandbox/include/asm/io.h | 28 +- arch/x86/include/asm/io.h | 6 + arch/xtensa/include/asm/io.h | 6 + drivers/mtd/nand/raw/atmel/nand-controller.c | 34 -- drivers/mtd/nand/raw/nand_base.c | 33 -- drivers/spi/cadence_qspi.c | 2 +- drivers/spi/meson_spifc_a1.c | 4 +- drivers/usb/cdns3/cdns3-ti.c | 1 - drivers/usb/dwc3/dwc3-meson-g12a.c | 2 +- drivers/usb/dwc3/dwc3-meson-gxl.c | 2 +- drivers/usb/musb-new/musb_io.h | 24 -- include/asm-generic/io.h | 348 +++++++++++++++++++ 16 files changed, 450 insertions(+), 118 deletions(-) -- 2.34.1

3 15

[PATCH] efi_loader: Make DisconnectController follow the EFI spec
by Ilias Apalodimas 28 Nov '23

28 Nov '23

commit 239d59a65e20 ("efi_loader: reconnect drivers on failure") tried to fix the UninstallProtocol interface which must reconnect any controllers it disconnected by calling ConnectController() in case of failure. However, the reconnect functionality was wired in efi_disconnect_all_drivers() instead of efi_uninstall_protocol(). As a result some SCT tests started failing. Specifically, BBTestOpenProtocolInterfaceTest333CheckPoint3() test - Calls ConnectController for DriverImageHandle1 - Calls DisconnectController for DriverImageHandle1 which will disconnect everything apart from TestProtocol4. That will remain open on purpose. - Calls ConnectController for DriverImageHandle2. TestProtocol4 which was explicitly preserved was installed wth BY_DRIVER attributes. The new protocol will call DisconnectController since its attributes are BY_DRIVER|EXCLUSIVE, but TestProtocol4 will not be removed. The test expects EFI_ACCESS_DENIED which works fine. The problem is that DisconnectController, will eventually call EFI_DRIVER_BINDING_PROTOCOL.Stop(). But on the aforementioned test this will call CloseProtocol -- the binding protocol is defined in 'DBindingDriver3.c' and the .Stop function uses CloseProtocol. If that close protocol call fails with EFI_NOT_FOUND, the current code will try to mistakenly reconnect all drivers and the subsequent tests that rely on the device being disconnected will fail. Move the reconnection in efi_uninstall_protocol() were it belongs. Fixes: commit 239d59a65e20 ("efi_loader: reconnect drivers on failure") Signed-off-by: Ilias Apalodimas <ilias.apalodimas(a)linaro.org> --- Heinrich this is critical. Although it doesn't break anything on our normal use cases it does break the ACS testing for SystemReady-IR and I prefer landing it in -master before the 2024.01 release lib/efi_loader/efi_boottime.c | 23 +++++------------------ 1 file changed, 5 insertions(+), 18 deletions(-) diff --git a/lib/efi_loader/efi_boottime.c b/lib/efi_loader/efi_boottime.c index 0b7579cb5af1..370eaee8ce1a 100644 --- a/lib/efi_loader/efi_boottime.c +++ b/lib/efi_loader/efi_boottime.c @@ -1339,7 +1339,7 @@ static efi_status_t efi_disconnect_all_drivers const efi_guid_t *protocol, efi_handle_t child_handle) { - efi_uintn_t number_of_drivers, tmp; + efi_uintn_t number_of_drivers; efi_handle_t *driver_handle_buffer; efi_status_t r, ret; @@ -1350,31 +1350,17 @@ static efi_status_t efi_disconnect_all_drivers if (!number_of_drivers) return EFI_SUCCESS; - tmp = number_of_drivers; while (number_of_drivers) { - ret = EFI_CALL(efi_disconnect_controller( + r = EFI_CALL(efi_disconnect_controller( handle, driver_handle_buffer[--number_of_drivers], child_handle)); - if (ret != EFI_SUCCESS) - goto reconnect; - } - - free(driver_handle_buffer); - return ret; - -reconnect: - /* Reconnect all disconnected drivers */ - for (; number_of_drivers < tmp; number_of_drivers++) { - r = EFI_CALL(efi_connect_controller(handle, - &driver_handle_buffer[number_of_drivers], - NULL, true)); if (r != EFI_SUCCESS) - EFI_PRINT("Failed to reconnect controller\n"); + ret = r; } free(driver_handle_buffer); - return ret; + return (ret != EFI_SUCCESS ? ret : EFI_SUCCESS); } /** @@ -1409,6 +1395,7 @@ static efi_status_t efi_uninstall_protocol r = efi_disconnect_all_drivers(handle, protocol, NULL); if (r != EFI_SUCCESS) { r = EFI_ACCESS_DENIED; + EFI_CALL(efi_connect_controller(handle, NULL, NULL, true)); goto out; } /* Close protocol */ -- 2.37.2

2 3

[PATCH 0/3] Unify K3 initial memory map
by Andrew Davis 28 Nov '23

28 Nov '23

Hello all, This was an RFC as it might still break remoteproc loading, but posted now for merge as worst case this would expose the issue in remoteproc loading better and force someone to go fix it. :) Changes for non-RFC: - Included dependency patch from here[0] into this series - Add Tested-bys - Rebase on latest -next Thanks, Andrew [0] https://lore.kernel.org/all/20231122211218.607757-1-afd@ti.com/ Andrew Davis (3): arm: mach-k3: Let the compiler size the mem_map lists arm: mach-k3: Remove non-cached memory map areas arm: mach-k3: Merge initial memory maps arch/arm/mach-k3/arm64-mmu.c | 275 +---------------------------------- 1 file changed, 2 insertions(+), 273 deletions(-) -- 2.39.2

3 6