March 2019 - U-Boot

[U-Boot] [PATCH v2 0/5] Marvell DB-XC3-24G4XG board support
by Chris Packham 11 Apr '19

11 Apr '19

This series adds support for Marvell's Switches with integrated CPUs and the DB-XC3-24G4XG board. The CPU side is similar to the Armada range. For now the DDR training code needs to come from the Marvell bin_hdr. It's one area where the integrated SoCs differ from the Armada range so neither the Armada-XP nor Armada-38x training code will work as-is. I'm asking Marvell about the possibility of re-licensing the code under a Proprietary/BSD/GPL as they did with Armada-38x. I also have access to a DB-DXBC2-MM board with a different chip. I'll look at adding support for that as well at some point. It's harder to work with because it has no USB, but other than that it's similar to the DB-XC3. Changes in v2: - use CONFIG_ARMADA_MSYS instead of just CONFIG_MSYS - Disable MBUS Error proagation - new, split out from Add DB-XC3-24G4XG board with a better explanation - u-boot specific changes in u-boot.dtsi - remove unnecessary entries from board config.h - move some changes to earlier patches Chris Packham (5): arm: sync armada-xp dts files from Linux 5.0 arm: mvebu: Add Marvell's integrated CPUs arm: mvebu: NAND clock support for MSYS devices tools: kwbimage: don't adjust for image_header for Armada MSYS arm: mvebu: Add DB-XC3-24G4XG board arch/arm/dts/Makefile | 3 +- arch/arm/dts/armada-370-xp.dtsi | 133 +++---- arch/arm/dts/armada-xp-98dx3236.dtsi | 343 ++++++++++++++++++ arch/arm/dts/armada-xp-98dx3336.dtsi | 39 ++ arch/arm/dts/armada-xp-98dx4251.dtsi | 54 +++ .../dts/armada-xp-db-xc3-24g4xg-u-boot.dtsi | 24 ++ arch/arm/dts/armada-xp-db-xc3-24g4xg.dts | 110 ++++++ arch/arm/dts/armada-xp-gp.dts | 167 ++++----- arch/arm/dts/armada-xp-maxbcm.dts | 24 +- arch/arm/dts/armada-xp-mv78230.dtsi | 55 +-- arch/arm/dts/armada-xp-mv78260.dtsi | 58 +-- arch/arm/dts/armada-xp-mv78460.dtsi | 58 +-- arch/arm/dts/armada-xp-synology-ds414.dts | 199 +++++----- arch/arm/dts/armada-xp-theadorable.dts | 69 ++-- arch/arm/dts/armada-xp.dtsi | 214 ++++++----- arch/arm/mach-mvebu/Kconfig | 26 +- arch/arm/mach-mvebu/Makefile | 1 + arch/arm/mach-mvebu/cpu.c | 33 +- arch/arm/mach-mvebu/include/mach/config.h | 2 +- arch/arm/mach-mvebu/include/mach/cpu.h | 3 + arch/arm/mach-mvebu/include/mach/soc.h | 31 ++ arch/arm/mach-mvebu/mbus.c | 5 + board/Marvell/db-xc3-24g4xg/MAINTAINERS | 7 + board/Marvell/db-xc3-24g4xg/Makefile | 5 + board/Marvell/db-xc3-24g4xg/README | 4 + board/Marvell/db-xc3-24g4xg/binary.0 | 11 + board/Marvell/db-xc3-24g4xg/db-xc3-24g4xg.c | 68 ++++ board/Marvell/db-xc3-24g4xg/kwbimage.cfg | 12 + configs/db-xc3-24g4xg_defconfig | 55 +++ drivers/ddr/marvell/axp/xor_regs.h | 4 + include/configs/db-xc3-24g4xg.h | 41 +++ tools/Makefile | 4 + tools/kwbimage.c | 4 + 33 files changed, 1319 insertions(+), 547 deletions(-) create mode 100644 arch/arm/dts/armada-xp-98dx3236.dtsi create mode 100644 arch/arm/dts/armada-xp-98dx3336.dtsi create mode 100644 arch/arm/dts/armada-xp-98dx4251.dtsi create mode 100644 arch/arm/dts/armada-xp-db-xc3-24g4xg-u-boot.dtsi create mode 100644 arch/arm/dts/armada-xp-db-xc3-24g4xg.dts create mode 100644 board/Marvell/db-xc3-24g4xg/MAINTAINERS create mode 100644 board/Marvell/db-xc3-24g4xg/Makefile create mode 100644 board/Marvell/db-xc3-24g4xg/README create mode 100644 board/Marvell/db-xc3-24g4xg/binary.0 create mode 100644 board/Marvell/db-xc3-24g4xg/db-xc3-24g4xg.c create mode 100644 board/Marvell/db-xc3-24g4xg/kwbimage.cfg create mode 100644 configs/db-xc3-24g4xg_defconfig create mode 100644 include/configs/db-xc3-24g4xg.h -- 2.20.1

2 18

[U-Boot] [PATCH v2 1/5] ppc: t2080qds: add eSDHC node
by Yinbo Zhu 11 Apr '19

11 Apr '19

From: Yinbo Zhu <yinbo.zhu(a)nxp.com> This patch is to add eSDHC node for t2080qds Signed-off-by: Yinbo Zhu <yinbo.zhu(a)nxp.com> --- Change in v2: add York sun for reviewing the patches set arch/powerpc/dts/t2080.dtsi | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/powerpc/dts/t2080.dtsi b/arch/powerpc/dts/t2080.dtsi index db65ea5725..f895dda4d7 100644 --- a/arch/powerpc/dts/t2080.dtsi +++ b/arch/powerpc/dts/t2080.dtsi @@ -58,5 +58,15 @@ device_type = "open-pic"; clock-frequency = <0x0>; }; + + esdhc: esdhc@114000 { + compatible = "fsl,esdhc"; + reg = <0x114000 0x1000>; + interrupts = <48 2 0 0>; + clock-frequency = <0>; + sdhci,auto-cmd12; + bus-width = <4>; + voltage-ranges = <1800 1800 3300 3300>; + }; }; }; -- 2.17.1

2 10

[U-Boot] [PATCH 0/6] ARM: meson: Add support for G12A based U200 board
by Neil Armstrong 10 Apr '19

10 Apr '19

This patchset adds basic support for the recent Amlogic G12A SoC family. This SoC family embeds 4xCortex A53 CPUs, and is very similar to the AXG SoC Family with the multimedia features of previous families. This patchset adds : - pinctrl support - clk driver atfer a rework of the drivers files - import Linux 5.1-rc1 G12A & U200 DT files - add board/amlogic/u200 to support the U200 reference design Jerome Brunet (5): pinctrl: meson: add g12a support clk: create meson directory and move related drivers clk: meson: add g12a support ARM: meson: add G12a support boards: meson: add g12a u200 Neil Armstrong (1): ARM: dts: Import Amlogic G12A u200 DT from Linux 5.1-rc1 MAINTAINERS | 2 +- arch/arm/dts/Makefile | 3 +- arch/arm/dts/meson-g12a-u200.dts | 29 + arch/arm/dts/meson-g12a.dtsi | 192 +++ arch/arm/include/asm/arch-meson/clock-g12a.h | 104 ++ arch/arm/include/asm/arch-meson/g12a.h | 66 + arch/arm/mach-meson/Kconfig | 9 +- arch/arm/mach-meson/Makefile | 1 + arch/arm/mach-meson/board-g12a.c | 150 ++ board/amlogic/u200/MAINTAINERS | 5 + board/amlogic/u200/Makefile | 6 + board/amlogic/u200/README | 128 ++ board/amlogic/u200/u200.c | 22 + configs/u200_defconfig | 40 + drivers/clk/Kconfig | 1 + drivers/clk/Makefile | 2 +- drivers/clk/meson/Kconfig | 23 + drivers/clk/meson/Makefile | 9 + drivers/clk/{clk_meson_axg.c => meson/axg.c} | 0 drivers/clk/{ => meson}/clk_meson.h | 0 drivers/clk/meson/g12a.c | 315 ++++ drivers/clk/{clk_meson.c => meson/gxbb.c} | 0 drivers/pinctrl/meson/Kconfig | 4 + drivers/pinctrl/meson/Makefile | 1 + drivers/pinctrl/meson/pinctrl-meson-g12a.c | 1294 +++++++++++++++++ drivers/pinctrl/meson/pinctrl-meson.c | 17 +- drivers/pinctrl/meson/pinctrl-meson.h | 1 + include/dt-bindings/clock/g12a-aoclkc.h | 34 + include/dt-bindings/clock/g12a-clkc.h | 135 ++ include/dt-bindings/gpio/meson-g12a-gpio.h | 114 ++ .../reset/amlogic,meson-g12a-reset.h | 134 ++ include/dt-bindings/reset/g12a-aoclkc.h | 18 + 32 files changed, 2850 insertions(+), 9 deletions(-) create mode 100644 arch/arm/dts/meson-g12a-u200.dts create mode 100644 arch/arm/dts/meson-g12a.dtsi create mode 100644 arch/arm/include/asm/arch-meson/clock-g12a.h create mode 100644 arch/arm/include/asm/arch-meson/g12a.h create mode 100644 arch/arm/mach-meson/board-g12a.c create mode 100644 board/amlogic/u200/MAINTAINERS create mode 100644 board/amlogic/u200/Makefile create mode 100644 board/amlogic/u200/README create mode 100644 board/amlogic/u200/u200.c create mode 100644 configs/u200_defconfig create mode 100644 drivers/clk/meson/Kconfig create mode 100644 drivers/clk/meson/Makefile rename drivers/clk/{clk_meson_axg.c => meson/axg.c} (100%) rename drivers/clk/{ => meson}/clk_meson.h (100%) create mode 100644 drivers/clk/meson/g12a.c rename drivers/clk/{clk_meson.c => meson/gxbb.c} (100%) create mode 100644 drivers/pinctrl/meson/pinctrl-meson-g12a.c create mode 100644 include/dt-bindings/clock/g12a-aoclkc.h create mode 100644 include/dt-bindings/clock/g12a-clkc.h create mode 100644 include/dt-bindings/gpio/meson-g12a-gpio.h create mode 100644 include/dt-bindings/reset/amlogic,meson-g12a-reset.h create mode 100644 include/dt-bindings/reset/g12a-aoclkc.h -- 2.21.0

1 7

[U-Boot] [PATCH] reset-meson: Add AXG reset compatible
by Neil Armstrong 10 Apr '19

10 Apr '19

Signed-off-by: Neil Armstrong <narmstrong(a)baylibre.com> --- drivers/reset/reset-meson.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/reset/reset-meson.c b/drivers/reset/reset-meson.c index 92f04695ec..31aa4d41e8 100644 --- a/drivers/reset/reset-meson.c +++ b/drivers/reset/reset-meson.c @@ -69,6 +69,7 @@ struct reset_ops meson_reset_ops = { static const struct udevice_id meson_reset_ids[] = { { .compatible = "amlogic,meson-gxbb-reset" }, + { .compatible = "amlogic,meson-axg-reset" }, { } }; -- 2.21.0

1 1

[U-Boot] [PATCH 0/2] ARM: meson: Add support for Libretech-AC
by Neil Armstrong 10 Apr '19

10 Apr '19

LibreTech AC is a single board computer manufactured by Libre Technology with the following specifications: - Amlogic S805X ARM Cortex-A53 quad-core SoC @ 1.2GHz - ARM Mali 450 GPU - 512MiB DDR4 SDRAM - 10/100 Ethernet - HDMI 2.0 4K/60Hz display - 40-pin GPIO header - 4 x USB 2.0 Host - eMMC, SPI NOR Flash - Infrared receiver The u-boot specific code is the same as the P212 support, so use the P212 board support code with a distinct defconfig and config include files. The patchset inports the DT from Linux 5.0 and add the corresponding configs and update the MAINTAINERS files. Neil Armstrong (2): ARM: dts: Import libretech-ac DT from Linux 5.0 boards: Amlogic: Add support for Libretech-AC arch/arm/dts/Makefile | 1 + arch/arm/dts/meson-gxl-s805x-libretech-ac.dts | 248 ++++++++++++++++++ arch/arm/dts/meson-gxl.dtsi | 70 ++++- board/amlogic/p212/MAINTAINERS | 1 + board/amlogic/p212/README.libretech-ac | 103 ++++++++ configs/libretech-ac_defconfig | 74 ++++++ include/configs/libretech-ac.h | 24 ++ 7 files changed, 512 insertions(+), 9 deletions(-) create mode 100644 arch/arm/dts/meson-gxl-s805x-libretech-ac.dts create mode 100644 board/amlogic/p212/README.libretech-ac create mode 100644 configs/libretech-ac_defconfig create mode 100644 include/configs/libretech-ac.h -- 2.21.0

1 3

[U-Boot] [PATCH] i2c: meson: add configurable divider factors
by Guillaume La Roque 10 Apr '19

10 Apr '19

This patch add support for I2C controller in Meson-AXG SoC, Due to the IP changes between I2C controller, we need to introduce a compatible data to make the divider factor configurable. backport from linux: 931b18e92cd0 ("2c: meson: add configurable divider factors") Signed-off-by: Guillaume La Roque <glaroque(a)baylibre.com> --- drivers/i2c/meson_i2c.c | 30 +++++++++++++++++++++++++----- 1 file changed, 25 insertions(+), 5 deletions(-) diff --git a/drivers/i2c/meson_i2c.c b/drivers/i2c/meson_i2c.c index 7d06d95cf3..ee59bac123 100644 --- a/drivers/i2c/meson_i2c.c +++ b/drivers/i2c/meson_i2c.c @@ -41,7 +41,12 @@ struct i2c_regs { u32 tok_rdata1; }; +struct meson_i2c_data { + unsigned char div_factor; +}; + struct meson_i2c { + const struct meson_i2c_data *data; struct clk clk; struct i2c_regs *regs; struct i2c_msg *msg; /* Current I2C message */ @@ -229,7 +234,7 @@ static int meson_i2c_set_bus_speed(struct udevice *bus, unsigned int speed) if (IS_ERR_VALUE(clk_rate)) return -EINVAL; - div = DIV_ROUND_UP(clk_rate, speed * 4); + div = DIV_ROUND_UP(clk_rate, speed * i2c->data->div_factor); /* clock divider has 12 bits */ if (div >= (1 << 12)) { @@ -253,6 +258,8 @@ static int meson_i2c_probe(struct udevice *bus) struct meson_i2c *i2c = dev_get_priv(bus); int ret; + i2c->data = (const struct meson_i2c_data *)dev_get_driver_data(bus); + ret = clk_get_by_index(bus, 0, &i2c->clk); if (ret < 0) return ret; @@ -272,11 +279,24 @@ static const struct dm_i2c_ops meson_i2c_ops = { .set_bus_speed = meson_i2c_set_bus_speed, }; +static const struct meson_i2c_data i2c_meson6_data = { + .div_factor = 4, +}; + +static const struct meson_i2c_data i2c_gxbb_data = { + .div_factor = 4, +}; + +static const struct meson_i2c_data i2c_axg_data = { + .div_factor = 3, +}; + static const struct udevice_id meson_i2c_ids[] = { - { .compatible = "amlogic,meson6-i2c" }, - { .compatible = "amlogic,meson-gx-i2c" }, - { .compatible = "amlogic,meson-gxbb-i2c" }, - { } + {.compatible = "amlogic,meson6-i2c", .data = (ulong)&i2c_meson6_data}, + {.compatible = "amlogic,meson-gx-i2c", .data = (ulong)&i2c_gxbb_data}, + {.compatible = "amlogic,meson-gxbb-i2c", .data = (ulong)&i2c_gxbb_data}, + {.compatible = "amlogic,meson-axg-i2c", .data = (ulong)&i2c_axg_data}, + {} }; U_BOOT_DRIVER(i2c_meson) = { -- 2.17.1

2 1

[U-Boot] [PATCH] Fix ext4 block group descriptor sizing
by Tom Rini 10 Apr '19

10 Apr '19

From: Benjamin Lim <jarsp.ctf(a)gmail.com> Signed-off-by: Benjamin Lim <jarsp.ctf(a)gmail.com> --- fs/ext4/ext4_common.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-) diff --git a/fs/ext4/ext4_common.c b/fs/ext4/ext4_common.c index 67e2471bd3..8638ef3b68 100644 --- a/fs/ext4/ext4_common.c +++ b/fs/ext4/ext4_common.c @@ -1584,7 +1584,7 @@ static int ext4fs_blockgroup int ext4fs_read_inode(struct ext2_data *data, int ino, struct ext2_inode *inode) { - struct ext2_block_group blkgrp; + struct ext2_block_group *blkgrp; struct ext2_sblock *sblock = &data->sblock; struct ext_filesystem *fs = get_fs(); int log2blksz = get_fs()->dev_desc->log2blksz; @@ -1592,17 +1592,28 @@ int ext4fs_read_inode(struct ext2_data *data, int ino, struct ext2_inode *inode) long int blkno; unsigned int blkoff; + /* Allocate blkgrp based on gdsize (for 64-bit support). */ + blkgrp = zalloc(get_fs()->gdsize); + if (!blkgrp) + return 0; + /* It is easier to calculate if the first inode is 0. */ ino--; status = ext4fs_blockgroup(data, ino / le32_to_cpu - (sblock->inodes_per_group), &blkgrp); - if (status == 0) + (sblock->inodes_per_group), blkgrp); + if (status == 0) { + free(blkgrp); return 0; + } inodes_per_block = EXT2_BLOCK_SIZE(data) / fs->inodesz; - blkno = ext4fs_bg_get_inode_table_id(&blkgrp, fs) + + blkno = ext4fs_bg_get_inode_table_id(blkgrp, fs) + (ino % le32_to_cpu(sblock->inodes_per_group)) / inodes_per_block; blkoff = (ino % inodes_per_block) * fs->inodesz; + + /* Free blkgrp as it is no longer required. */ + free(blkgrp); + /* Read the inode. */ status = ext4fs_devread((lbaint_t)blkno << (LOG2_BLOCK_SIZE(data) - log2blksz), blkoff, -- 2.17.1

1 1

[U-Boot] [PATCH 1/5] test/py: mmc: Factor out device selection
by Marek Vasut 10 Apr '19

10 Apr '19

Factor out the 'mmc dev' call so it can be recycled by other tests. Signed-off-by: Marek Vasut <marek.vasut+renesas(a)gmail.com> Cc: Tom Rini <trini(a)konsulko.com> Cc: Simon Glass <sjg(a)chromium.org> --- test/py/tests/test_mmc_rd.py | 38 +++++++++++++++++++++++++----------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/test/py/tests/test_mmc_rd.py b/test/py/tests/test_mmc_rd.py index a13bc0a5f6..40ae1a8459 100644 --- a/test/py/tests/test_mmc_rd.py +++ b/test/py/tests/test_mmc_rd.py @@ -57,6 +57,32 @@ env__mmc_rd_configs = ( ) """ +def mmc_dev(u_boot_console, is_emmc, devid, partid): + """Run the "mmc dev" command. + + Args: + u_boot_console: A U-Boot console connection. + is_emmc: Whether the device is eMMC + devid: Device ID + partid: Partition ID + + Returns: + Nothing. + """ + + # Select MMC device + cmd = 'mmc dev %d' % devid + if is_emmc: + cmd += ' %d' % partid + response = u_boot_console.run_command(cmd) + assert 'no card present' not in response + if is_emmc: + partid_response = '(part %d)' % partid + else: + partid_response = '' + good_response = 'mmc%d%s is current device' % (devid, partid_response) + assert good_response in response + @pytest.mark.buildconfigspec('cmd_mmc') def test_mmc_rd(u_boot_console, env__mmc_rd_config): """Test the "mmc read" command. @@ -86,17 +112,7 @@ def test_mmc_rd(u_boot_console, env__mmc_rd_config): addr = '0x%08x' % ram_base # Select MMC device - cmd = 'mmc dev %d' % devid - if is_emmc: - cmd += ' %d' % partid - response = u_boot_console.run_command(cmd) - assert 'no card present' not in response - if is_emmc: - partid_response = '(part %d)' % partid - else: - partid_response = '' - good_response = 'mmc%d%s is current device' % (devid, partid_response) - assert good_response in response + mmc_dev(u_boot_console, is_emmc, devid, partid) # Clear target RAM if expected_crc32: -- 2.20.1

3 14

[U-Boot] [PATCH] fs: fat: fix reading non-cluster-aligned root directory
by Anssi Hannula 10 Apr '19

10 Apr '19

FAT16 root directory might not start at a cluster boundary (though it always ends on one). However, next_dent() always starts reading the directory at the beginning of the cluster, causing no files to be found (at least if the beginning of the cluster contains zeroes). Fix that by skipping to the actual start of the root directory in such a case. This is a relatively common case as at least the filesystem formatter on Win7 seems to create such filesystems by default on 2GB USB sticks (cluster size 64 sectors, rootdir size 32 sectors). dosfstools mkfs.vfat does not seem to create such filesystems. Signed-off-by: Anssi Hannula <anssi.hannula(a)bitwise.fi> --- fs/fat/fat.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/fs/fat/fat.c b/fs/fat/fat.c index b08949d370..4a0d4bb8bc 100644 --- a/fs/fat/fat.c +++ b/fs/fat/fat.c @@ -782,6 +782,20 @@ static dir_entry *next_dent(fat_itr *itr) itr->remaining = nbytes / sizeof(dir_entry) - 1; itr->dent = dent; + + if (itr->fsdata->fatsize != 32 && + itr->clust == itr->fsdata->root_cluster) { + /* rootdir might not start at cluster boundary */ + u32 sect_offset = itr->fsdata->rootdir_sect - + clust_to_sect(itr->fsdata, + itr->fsdata->root_cluster); + u32 dirent_offset = itr->fsdata->sect_size / + sizeof(dir_entry) * sect_offset; + + itr->remaining -= dirent_offset; + itr->dent += dirent_offset; + } + } else { itr->remaining--; itr->dent++; -- 2.17.1

3 4

[U-Boot] [U-BOOT] fs: ext4: Problem with ext4load and sparse files
by Gero Schumacher 10 Apr '19

10 Apr '19

Hi, when I try to load a sparse file via ext4load, I am getting the error message 'invalid extent' After a deeper look in the code, it seems to be an issue in the function ext4fs_get_extent_block in fs/ext4/ext4_common.c: The file starts with 1k of zeros. The blocksize is 1024. So the first extend block contains the following information: eh_entries: 1 eh_depth: 1 ei_block 1 When the upper layer (ext4fs_read_file) asks for fileblock 0, we are running in the 'invalid extent' error message. For me it seems, that the code is not prepared for handling a sparse block at the beginning of the file. The following change, solved my problem: diff --git a/fs/ext4/ext4_common.c b/fs/ext4/ext4_common.c index 67e2471bd3..7c8080d740 100644 --- a/fs/ext4/ext4_common.c +++ b/fs/ext4/ext4_common.c @@ -1546,8 +1546,10 @@ static struct ext4_extent_header *ext4fs_get_extent_block break; } while (fileblock >= le32_to_cpu(index[i].ei_block)); - if (--i < 0) - return NULL; + /* if first logical block number is higher than requested fileblock, + it is a sparse file. This is handled on upper layer */ + if (i > 0) + i--; block = le16_to_cpu(index[i].ei_leaf_hi); block = (block << 32) + le32_to_cpu(index[i].ei_leaf_lo); I am really not an expert in ext4 filesystems. Can somebody please have a look at this issue and give me a feedback, if I am totally wrong or not?

2 1