August 2015 - U-Boot

[U-Boot] [PATCH v2 0/8] JFFS2 fixes and performance improvements
by Mark Tomlinson 13 Aug '15

13 Aug '15

In reply to comments from Wolfgang Denk and Heiko Schocher: My aim was to optimize U-Boot's loading of a JFFS2 file, since we needed to boot much more quickly on one of our devices than we currently were. We had discussed the possibility of abandoning JFFS2 entirely, but chose to see how quickly JFFS2 could be made to work. We also knew that Linux would mount this file system much quicker than u-boot did, so knew that at least some speed improvement was possible. The improvments was from three sources: (1) searching U-Boot mailing list archives, (2) looking at the linux kernel, and (3) our own measurements and improvements. For example, the merge sort (before my modifications) can be found here: http://lists.denx.de/pipermail/u-boot/2007-January/018777.html Two of the patches were inspired from the Linux kernel: Changing scansize, and recognising the CLEANMARKER. While trying to synchronize U-Boot code with Linux sounds like a good idea, I think this wouldn't be the right way to go, since Linux needs to also build lists of blocks which are empty, and a list of blocks that still need to be erased. Stripping this out is more work than just enhancing what U-Boot currently has. === Original cover message follows === These patches fix bugs and improve performance of JFFS2. Some of these improvements can already be found in old mailing lists, but for some reason they have not made their way into the u-boot source. I have the feeling that any one of these patches didn't show enough performance gain to warrant adding it to the source. I am hopeful that together, all these patches can be seen to make a big difference. One of these patches ("Only list each directory entry once") is a bug fix, all the rest are for performance. Although performance is not high on the priority list for a bootloader, the length of time it was taking to scan a JFFS2 filesystem was painfully slow. The code for mergesort was found in an abandoned u-boot patch, although I have refactored the code somewhat. The original author is still shown at the top of that file. The timings below are with jffs2_summary_support turned off. With these improvements, summary support was no longer useful - most of our time is now spent loading the actual release. Even without this feature turned on, the code will still load from a filesystem that has summary nodes. Due to not having other resources, I also have not done anything for NAND flash. At least some of these changes will be directly applicable to NAND as well as NOR. My own testing is on a system with a 1GHz PowerPC, and 256MB of NOR Flash. The flash accesses are slow compared with processing power and RAM, so minimising the number of flash accesses makes a huge difference. Here are the timing comparisons for three JFFS2 operations on this system: 1) Scanning the file system 2) Getting a director listing of the top level, and 3) Loading a 30MB file. Times are in seconds, and the contribution from each patch has been measured: Scan List Load Total Original: 266.0 17.3 23.4 306.7 Speed up comparison: 52.3 17.3 23.4 93.0 List dir entries once: 52.3 11.0 23.4 86.7 Improve read speed: 52.3 0.8 5.4 58.5 Optimize building lists: 31.9 0.8 5.4 38.1 Change scansize: 30.8 0.8 5.4 37.0 Use cleanmarker: 16.0 0.8 5.4 22.2 Use mergesort: 2.0 0.8 5.4 8.2 Note that "List dir entries once" is not a speed improvement as such, but because old versions of a file and deleted files are no longer scanned, there is an improvement on this filesystem (the flash is approx half full). Also, "change scansize" appears to do very little in this benchmark list. But without it, the "use cleanmarker" would not show as much improvement. Changes in v2: - Fix comment style - Remove extra {} pair. - Changed comment style. - Fixed some missing calls to put_fl_mem(). - Change comment style - Change comment style - Changed comment style - Changed copyright notice to use SPDX-Licence-Identifier. - Added URL to original mergesort code. - Removed #ifdef, using Makefile change instead. Mark Tomlinson (8): JFFS2: Return early when file read not necessary JFFS2: Speed up and fix comparison functions JFFS2: Only list each directory entry once JFFS2: Improve speed reading flash files JFFS2: Optimize building lists during scan JFFS2: Change scansize to match linux kernel JFFS2: Use CLEANMARKER to reduce scanning time JFFS2: Use merge sort when parsing filesystem fs/jffs2/Makefile | 1 + fs/jffs2/jffs2_1pass.c | 257 ++++++++++++++++++++++++++++++----------------- fs/jffs2/jffs2_private.h | 4 + fs/jffs2/mergesort.c | 52 ++++++++++ 4 files changed, 223 insertions(+), 91 deletions(-) create mode 100644 fs/jffs2/mergesort.c -- 1.9.1

3 17

[U-Boot] Setting up vexpress u-boot for distro configs and use in virt
by Dennis Gilmore 13 Aug '15

13 Aug '15

Hi all, The attached patches fix up a minor noise issue if a board defines its own value for CONFIG_BOOTP_VCI_STRING, makes sure we do not hit any noise on vexpress and port to the distro boot configs while maintaining the existing boot setup. I have built and tested for vexpress_ca15_tc2 using qemu-system-arm, however I hit a snag with an illegal instruction. U-Boot 2015.07-rc2-00140-gef0f2f5-dirty (Jun 27 2015 - 06:42:35 -0500) DRAM: 1 GiB WARNING: Caches not enabled Flash: 128 MiB MMC: MMC: 0 *** Warning - bad CRC, using default environment In: serial Out: serial Err: serial Net: smc911x-0 Warning: smc911x-0 using MAC address from net device Hit any key to stop autoboot: 0 MMC Device 1 not found no mmc device at slot 1 switch to partitions #0, OK mmc0 is current device env - environment handling commands Usage: env default [-f] -a - [forcibly] reset default environment env default [-f] var [...] - [forcibly] reset variable(s) to their default values env delete [-f] var [...] - [forcibly] delete variable(s) env export [-t | -b | -c] [-s size] addr [var ...] - export environment env import [-d] [-t [-r] | -b | -c] addr [size] - import environment env print [-a | name ...] - print environment env run var [...] - run commands in an environment variable env save - save environment env set [-f] name [arg ...] Scanning mmc 0:1... Found /extlinux/extlinux.conf Retrieving file: /extlinux/extlinux.conf 486 bytes read in 32 ms (14.6 KiB/s) Ignoring unknown command: ui Ignoring malformed menu command: autoboot Ignoring malformed menu command: hidden Ignoring unknown command: totaltimeout Fedora-Server-armhfp-22-3 Boot Options. 1: Fedora-Server-armhfp-22-3 (4.0.4-301.fc22.armv7hl) Enter choice: 1 1: Fedora-Server-armhfp-22-3 (4.0.4-301.fc22.armv7hl) Retrieving file: /initramfs-4.0.4-301.fc22.armv7hl.img 38558585 bytes read in 4209 ms (8.7 MiB/s) Retrieving file: /vmlinuz-4.0.4-301.fc22.armv7hl 5645832 bytes read in 688 ms (7.8 MiB/s) append: ro root=UUID=fe0fbdbc-e734-4f07-9b1c-7564e3a356b9 console=ttyAMA0,38400n8 Kernel image @ 0xa0008000 [ 0x000000 - 0x562608 ] Starting kernel ... undefined instruction pc : [<9ffac060>] lr : [<9ff83b24>] sp : 9fee1be8 ip : a0008000 fp : 9ff83b50 r10: 00000000 r9 : 9fee2f00 r8 : 9fee1ce4 r7 : 00000000 r6 : 0000001a r5 : 00000131 r4 : 00000000 r3 : 80002000 r2 : 80002000 r1 : 000008e0 r0 : 00000000 Flags: nzcv IRQs off FIQs off Mode UK6_32 Resetting CPU ... resetting ... In the end my goal here is to have a u-boot that works with qemu-system-arm so that the user never needs to extract files out of the system in order to get a booting vm. The answer may well be to make a virt specific u-boot that can set the amount of ram, cpu, etc based on what qemu tells it is there. Regards Dennis

3 7

[U-Boot] [PATCH v2 1/2] fdt: prevent clearing memory node if there are no banks
by Andre Przywara 13 Aug '15

13 Aug '15

Avoid clearing the reg property in the memory DT node if no memory banks have been specified for a board (CONFIG_NR_DRAM_BANKS == 0). This allows boards to let U-Boot skip the DT memory tinkering in case other firmware has already setup the node properly before. This should be safe as all callers of fdt_fixup_memory_banks that use a computed <banks> value put at least 1 in there. Add some documentation comments to the header file. Signed-off-by: Andre Przywara <osp(a)andrep.de> --- Hi Tom, if you haven't applied the last version of this patch, consider using this version instead, which has Rob's comment addressed (bailing out early). Cheers, Andre. common/fdt_support.c | 3 +++ include/fdt_support.h | 26 ++++++++++++++++++++++++++ 2 files changed, 29 insertions(+) diff --git a/common/fdt_support.c b/common/fdt_support.c index 10648b5..f86365e 100644 --- a/common/fdt_support.c +++ b/common/fdt_support.c @@ -454,6 +454,9 @@ int fdt_fixup_memory_banks(void *blob, u64 start[], u64 size[], int banks) return err; } + if (!banks) + return 0; + len = fdt_pack_reg(blob, tmp, start, size, banks); err = fdt_setprop(blob, nodeoffset, "reg", tmp, len); diff --git a/include/fdt_support.h b/include/fdt_support.h index 0edc4fa..296add0 100644 --- a/include/fdt_support.h +++ b/include/fdt_support.h @@ -67,8 +67,34 @@ void do_fixup_by_compat(void *fdt, const char *compat, const char *prop, const void *val, int len, int create); void do_fixup_by_compat_u32(void *fdt, const char *compat, const char *prop, u32 val, int create); +/** + * Setup the memory node in the DT. Creates one if none was existing before. + * Calls fdt_fixup_memory_banks() to populate a single reg pair covering the + * whole memory. + * + * @param blob FDT blob to update + * @param start Begin of DRAM mapping in physical memory + * @param size Size of the single memory bank + * @return 0 if ok, or -1 or -FDT_ERR_... on error + */ int fdt_fixup_memory(void *blob, u64 start, u64 size); + +/** + * Fill the DT memory node with multiple memory banks. + * Creates the node if none was existing before. + * If banks is 0, it will not touch the existing reg property. This allows + * boards to not mess with the existing DT setup, which may have been + * filled in properly before. + * + * @param blob FDT blob to update + * @param start Array of size <banks> to hold the start addresses. + * @param size Array of size <banks> to hold the size of each region. + * @param banks Number of memory banks to create. If 0, the reg + * property will be left untouched. + * @return 0 if ok, or -1 or -FDT_ERR_... on error + */ int fdt_fixup_memory_banks(void *blob, u64 start[], u64 size[], int banks); + void fdt_fixup_ethernet(void *fdt); int fdt_find_and_setprop(void *fdt, const char *node, const char *prop, const void *val, int len, int create); -- 1.8.4

3 4

[U-Boot] Pull request: u-boot-mmc 12082015
by Pantelis Antoniou 13 Aug '15

13 Aug '15

Hi Tom, The following changes since commit e2c1c5bae619d2e87505de99f907a26237640bc9: Merge git://git.denx.de/u-boot-dm (2015-08-10 10:06:07 -0400) are available in the git repository at: git://git.denx.de/u-boot-mmc.git master for you to fetch changes up to 956f662b1f79ae7455bc6aaa9defc2a756e8e9ca: mmc_spi: Big-endian support (2015-08-12 10:41:14 +0300) ---------------------------------------------------------------- Marek Vasut (3): mmc: dw_mmc: Stop bounce buffer even in case of failure mmc: dw_mmc: Zap endless timeout mmc: dw_mmc: Improve handling of data transfer failure Simon Glass (3): mmc: dw_mmc: Avoid using printf() for errors mmc: dw_mmc: Support bypass mode with the get_mmc_clk() method dw_mmc: Calculate dwmmc FIFO threshold size if not provided Yoshinori Sato (1): mmc_spi: Big-endian support drivers/mmc/dw_mmc.c | 61 +++++++++++++++++++++++++++++++++++++++++++------------------ drivers/mmc/exynos_dw_mmc.c | 2 +- drivers/mmc/mmc_spi.c | 10 +++++----- include/dwmmc.h | 16 +++++++++++++++- 4 files changed, 64 insertions(+), 25 deletions(-) Regards — Pantelis

3 3

[U-Boot] Pull request: u-boot-net.git master
by Joe Hershberger 13 Aug '15

13 Aug '15

The following changes since commit e2c1c5bae619d2e87505de99f907a26237640bc9: Merge git://git.denx.de/u-boot-dm (2015-08-10 10:06:07 -0400) are available in the git repository at: git://git.denx.de/u-boot-net.git master for you to fetch changes up to faa765d40760d84bff0de26a5a5f605621dbff39: net: e1000: Increase autoneg timeout to 8 seconds (2015-08-12 14:15:29 -0500) ---------------------------------------------------------------- Clemens Gruber (2): net: Improve 88E151x PHY initialization net: Add support for Marvell 88E1510 PHY Daniel Inderbitzin (1): qoriq eth.c bugfix: handle received corrupted frames correctly Jeroen Hofstee (1): net: davinci_emac: don't teardown inactive rx channel Jiandong Zheng (1): net: phy: broadcom: Add BCM Cygnus PHY Jörg Krause (2): net: phy: fix data type of phy_id net: phy: delay only if reset handler is registered Max Krummenacher (1): tftp.c: fix CONFIG_TFTP_TSIZE for small files Shengzhou Liu (1): net/phy: set led for rtl8211f phy Stefan Roese (1): net: e1000: Increase autoneg timeout to 8 seconds Tim Harvey (3): e1000: releasing semaphore once no longer needed Revert "e1000: fix sw fw sync on igb i210/i211" e1000: remove unnecessary clearing of SWSM.SWSM_SMBI Vladimir Zapolskiy (4): net: lpc32xx: correct command register reset value net: lpc32xx: connect MAC to phy with CONFIG_PHY_ADDR id net: lpc32xx: improve MAC configuration on reset and initialization net: lpc32xx: add RMII phy mode support Wu, Josh (1): net: macb: add gmac multi-queue support arch/arm/cpu/arm926ejs/lpc32xx/devices.c | 7 +++- drivers/net/davinci_emac.c | 3 +- drivers/net/e1000.c | 34 +++++++++++----- drivers/net/e1000.h | 3 +- drivers/net/fm/eth.c | 5 ++- drivers/net/lpc32xx_eth.c | 55 +++++++++++++++----------- drivers/net/macb.c | 33 ++++++++++++++++ drivers/net/macb.h | 9 +++++ drivers/net/phy/broadcom.c | 29 ++++++++++++++ drivers/net/phy/marvell.c | 67 +++++++++++++++++++++++++------- drivers/net/phy/phy.c | 10 +++-- drivers/net/phy/realtek.c | 8 ++++ net/tftp.c | 2 + 13 files changed, 210 insertions(+), 55 deletions(-)

2 1

[U-Boot] Pull request: u-boot-pmic
by Przemyslaw Marczak 13 Aug '15

13 Aug '15

Hello Tom, The following changes since commit e2c1c5bae619d2e87505de99f907a26237640bc9: Merge git://git.denx.de/u-boot-dm (2015-08-10 10:06:07 -0400) are available in the git repository at: git://git.denx.de/u-boot-pmic.git master for you to fetch changes up to e6b606d6572893a9f6fb6339b6a09a4b6942f989: dm: pmic: max77686: Correct two typos in a comment (2015-08-12 11:17:45 +0200) ---------------------------------------------------------------- Peng Fan (7): power: pfuze100 correct SWBST macro definition power: regulator use node name when no regulator-name power: regulator: update comments for regulator-name power: pmic: pfuze100 support driver model power: regulator: add pfuze100 support fsl: common: pfuze: no use original pfuze code if DM_PMIC power: regulator: max77686 correct variable type Simon Glass (1): dm: pmic: max77686: Correct two typos in a comment board/freescale/common/pfuze.c | 2 + doc/device-tree-bindings/regulator/regulator.txt | 19 +- drivers/power/pmic/Makefile | 1 + drivers/power/regulator/Kconfig | 8 + drivers/power/regulator/Makefile | 1 + drivers/power/regulator/max77686.c | 20 +- drivers/power/regulator/pfuze100.c | 568 +++++++++++++++++++++++ drivers/power/regulator/regulator-uclass.c | 4 +- include/power/pfuze100_pmic.h | 26 +- include/power/regulator.h | 1 + 10 files changed, 623 insertions(+), 27 deletions(-) create mode 100644 drivers/power/regulator/pfuze100.c Best regards, -- Przemyslaw Marczak Samsung R&D Institute Poland Samsung Electronics p.marczak(a)samsung.com

2 1

[U-Boot] [PATCH] ARM AM43XX: make boot-params available to U-Boot, save to env
by Holleis, Edgar 13 Aug '15

13 Aug '15

This patch makes the Boot Parameter Structure of AM43XX available to U-Boot. The feature is already available for OMAP3 and OMAP4, but not AM43XX. It also adds the value of the boot device to the U-Boot environment. Signed-off-by: Edgar Holleis <edgar.holleis(a)zuehlke.com> --- arch/arm/cpu/armv7/am33xx/board.c | 11 +++++++++++ arch/arm/cpu/armv7/omap-common/boot-common.c | 10 +++++++--- board/ti/am43xx/board.c | 5 +++++ 3 files changed, 23 insertions(+), 3 deletions(-) diff --git a/arch/arm/cpu/armv7/am33xx/board.c b/arch/arm/cpu/armv7/am33xx/board.c index 67bef23..7149990 100644 --- a/arch/arm/cpu/armv7/am33xx/board.c +++ b/arch/arm/cpu/armv7/am33xx/board.c @@ -203,6 +203,17 @@ int arch_misc_init(void) return 0; } +#ifdef CONFIG_ARCH_CPU_INIT +/* + * SOC specific cpu init + */ +int arch_cpu_init(void) +{ + save_omap_boot_params(); + return 0; +} +#endif /* CONFIG_ARCH_CPU_INIT */ + #ifndef CONFIG_SKIP_LOWLEVEL_INIT /* * In the case of non-SPL based booting we'll want to call these diff --git a/arch/arm/cpu/armv7/omap-common/boot-common.c b/arch/arm/cpu/armv7/omap-common/boot-common.c index bbc6bed..d1053eb 100644 --- a/arch/arm/cpu/armv7/omap-common/boot-common.c +++ b/arch/arm/cpu/armv7/omap-common/boot-common.c @@ -64,9 +64,13 @@ void save_omap_boot_params(void) #endif { dev_desc = *((u32 *)(rom_params + DEV_DESC_PTR_OFFSET)); - dev_data = *((u32 *)(dev_desc + DEV_DATA_PTR_OFFSET)); - gd->arch.omap_boot_params.omap_bootmode = - *((u32 *)(dev_data + BOOT_MODE_OFFSET)); + if (dev_desc != 0) { + dev_data = *((u32 *)(dev_desc + DEV_DATA_PTR_OFFSET)); + gd->arch.omap_boot_params.omap_bootmode = + *((u32 *)(dev_data + BOOT_MODE_OFFSET)); + } else { + gd->arch.omap_boot_params.omap_bootmode = MMCSD_MODE_UNDEFINED; + } } } diff --git a/board/ti/am43xx/board.c b/board/ti/am43xx/board.c index d7b9e5a..ecd6c56 100644 --- a/board/ti/am43xx/board.c +++ b/board/ti/am43xx/board.c @@ -670,6 +670,11 @@ int board_late_init(void) safe_string[sizeof(header.version)] = 0; setenv("board_rev", safe_string); #endif + + char tmpstr[5]; + sprintf(tmpstr, "0x%02x", gd->arch.omap_boot_params.omap_bootdevice); + setenv("omap_bootdevice", tmpstr); + return 0; } #endif -- 1.9.1

2 1

[U-Boot] [PATCH v2 00/15] Add macros to ease our life with independent CONFIGs between U-Boot and SPL
by Masahiro Yamada 13 Aug '15

13 Aug '15

Refer to Simon's question, too: http://lists.denx.de/pipermail/u-boot/2015-July/219598.html Since U-boot introduced SPL (not since Kconfig), enabling features for U-boot and SPL independently is always a PITA. - decide if each feature should be supported for SPL or not - Add CONFIG_SPL_FRED_SUPPORT into Makefile.spl - Add #undef include/config_uncmd_spl.h to disable features we do not want to support on SPL - Add "ifdef CONFIG_SPL_BUILD ... endif" here and there to adjust things - Add "#ifdef CONFIG_SPL_BUILD ... #endif" here and there to fix things up Things are getting more and more crappy. When U-boot switched to Kconfig, first I introduced separate .config (.config, spl/.config, tpl/.config) to clean up them. But it turned out to be a pain. So, I believe the current single .config is much better. But I also admit we need something systematic to subdue our PITA. One possibility is to support "spl-y" in makefiles. (This idea is cribbed from barebox.) obj-$(CONFIG_FOO) += foo.o spl-$(CONFIG_SPL_FOO) += foo.o is cleaner than ifdef CONFIG_SPL_BUILD obj-$(CONFIG_SPL_FOO) += foo.o else obj-$(CONFIG_FOO) += foo.o endif It is a nice improvement in makefile side. But we still need to do something with C files. Another option is something like CONFIG_FOO=yyn (yes for U-boot, yes for SPL, no for TPL) To achieve this, I think a big operation is needed in Kconfig core. I cannot do that. (Of course, Patches are welcome if someone else can do that.) So, I was thinking of something different. My idea was inspired by IS_ENABLED() macro in include/linux/kconfig.h. Linux defines different macros for built-in and module, and it is possible to write #if IS_ENABLED(CONFIG_FOO) ... #endif instead of #if defined(CONFIG_FOO) || defined(CONFIG_FOO_MODULE) ... #endif So, I'd like to propose new macros to write code like #if CONFIG_IS_ENABLED(FOO) ... #endif instead of #if (!defined(CONFIG_SPL_BUILD) && defined(CONFIG_FOO)) || \ (defined(CONFIG_SPL_BUILD) && defined(CONFIG_SPL_FOO)) ... #endif I hope this series will make our life easier. The result of buildman: Summary of 16 commits for 1092 boards (8 threads, 1 job per thread) 01: Merge branch 'master' of git://git.denx.de/u-boot-tegra sh: + sh7753evb sh7785lcr_32bit sh7785lcr nios2: + nios2-generic arm: + openrd_base axm openrd_ultimate openrd_client zynq_zc70x taurus powerpc: + ebony taihu ocotea TQM834x taishan katmai alpr 02: kbuild: fixdep: optimize code slightly 03: kbuild: add a makefile macro useful with per-image config options 04: linux/kconfig.h: add CPP macros useful for per-image config options 05: spl: move SPL driver entries to driver/Makefile 06: dm: unify obj-$(CONFIG_DM) and obj-$(CONFIG_SPL_DM) entries 07: clk: rename CONFIG_SPL_CLK_SUPPORT to CONFIG_SPL_CLK 08: clk: unify obj-$(CONFIG_CLK) and obj-$(CONFIG_SPL_CLK) entries 09: ram: rename CONFIG_SPL_RAM_SUPPORT to CONFIG_SPL_RAM 10: ram: unify obj-$(CONFIG_RAM) and obj-$(CONFIG_SPL_RAM) entries 11: led: rename CONFIG_SPL_LED_SUPPORT to CONFIG_SPL_LED 12: led: unify obj-$(CONFIG_LED) and obj-$(CONFIG_SPL_LED) entries 13: dm: drop CONFIG_DM_DEVICE_REMOVE from uncmd list 14: fdtdec: fix OF_CONTROL switch 15: of: flip CONFIG_SPL_DISABLE_OF_CONTROL into CONFIG_SPL_OF_CONTROL 16: of: clean up OF_CONTROL ifdef conditionals Masahiro Yamada (15): kbuild: fixdep: optimize code slightly kbuild: add a makefile macro useful with per-image config options linux/kconfig.h: add CPP macros useful for per-image config options spl: move SPL driver entries to driver/Makefile dm: unify obj-$(CONFIG_DM) and obj-$(CONFIG_SPL_DM) entries clk: rename CONFIG_SPL_CLK_SUPPORT to CONFIG_SPL_CLK clk: unify obj-$(CONFIG_CLK) and obj-$(CONFIG_SPL_CLK) entries ram: rename CONFIG_SPL_RAM_SUPPORT to CONFIG_SPL_RAM ram: unify obj-$(CONFIG_RAM) and obj-$(CONFIG_SPL_RAM) entries led: rename CONFIG_SPL_LED_SUPPORT to CONFIG_SPL_LED led: unify obj-$(CONFIG_LED) and obj-$(CONFIG_SPL_LED) entries dm: drop CONFIG_DM_DEVICE_REMOVE from uncmd list fdtdec: fix OF_CONTROL switch of: flip CONFIG_SPL_DISABLE_OF_CONTROL into CONFIG_SPL_OF_CONTROL of: clean up OF_CONTROL ifdef conditionals arch/arm/Kconfig | 3 -- arch/arm/cpu/armv7/am33xx/board.c | 2 +- arch/arm/cpu/armv7/exynos/Kconfig | 8 ---- arch/arm/cpu/armv7/exynos/pinmux.c | 2 +- arch/arm/cpu/armv7/s5pc1xx/Kconfig | 2 - arch/arm/include/asm/arch-exynos/dwmmc.h | 2 - arch/arm/include/asm/arch-exynos/mipi_dsim.h | 2 - arch/arm/include/asm/arch-exynos/mmc.h | 2 - arch/arm/mach-tegra/Kconfig | 1 - arch/arm/mach-tegra/clock.c | 4 +- arch/arm/mach-tegra/tegra114/clock.c | 4 +- arch/arm/mach-tegra/tegra124/clock.c | 4 +- arch/arm/mach-tegra/tegra20/clock.c | 4 +- arch/arm/mach-tegra/tegra30/clock.c | 4 +- board/vpac270/u-boot-spl.lds | 2 +- .../xilinx/microblaze-generic/microblaze-generic.c | 2 +- board/xilinx/zynq/board.c | 2 +- common/cli.c | 4 +- common/spl/spl.c | 3 +- configs/am335x_boneblack_vboot_defconfig | 1 - configs/arches_defconfig | 1 - configs/canyonlands_defconfig | 1 - configs/galileo_defconfig | 1 - configs/microblaze-generic_defconfig | 1 - configs/odroid_defconfig | 1 - configs/origen_defconfig | 1 - configs/s5pc210_universal_defconfig | 1 - configs/socfpga_socrates_defconfig | 1 - configs/trats2_defconfig | 1 - configs/trats_defconfig | 1 - drivers/Makefile | 41 ++++++++++++++++-- drivers/clk/Kconfig | 2 +- drivers/core/Makefile | 8 ++-- drivers/core/device.c | 10 ++--- drivers/core/lists.c | 2 +- drivers/core/root.c | 6 +-- drivers/core/uclass.c | 4 +- drivers/gpio/mxc_gpio.c | 2 +- drivers/gpio/vybrid_gpio.c | 2 +- drivers/i2c/s3c24x0_i2c.c | 4 +- drivers/input/Makefile | 2 +- drivers/input/tegra-kbc.c | 2 +- drivers/led/Kconfig | 2 +- drivers/mmc/exynos_dw_mmc.c | 2 +- drivers/mmc/s5p_sdhci.c | 2 +- drivers/mmc/tegra_mmc.c | 2 +- drivers/mmc/zynq_sdhci.c | 2 +- drivers/mtd/spi/sf_probe.c | 6 +-- drivers/net/xilinx_emaclite.c | 2 +- drivers/net/zynq_gem.c | 2 +- drivers/power/exynos-tmu.c | 2 +- drivers/power/pmic/pmic_max77686.c | 4 +- drivers/ram/Kconfig | 2 +- drivers/serial/ns16550.c | 2 +- drivers/serial/serial-uclass.c | 4 +- drivers/serial/serial_omap.c | 2 +- drivers/serial/serial_pl01x.c | 2 +- drivers/serial/serial_tegra.c | 4 +- drivers/serial/serial_uniphier.c | 2 +- drivers/serial/serial_zynq.c | 2 +- drivers/sound/max98095.c | 2 +- drivers/sound/wm8994.c | 2 +- drivers/tpm/tpm_tis_i2c.c | 2 +- drivers/video/exynos_dp.c | 4 +- drivers/video/exynos_dp_lowlevel.c | 2 +- drivers/video/exynos_fb.c | 8 ++-- drivers/video/exynos_fimd.c | 4 +- drivers/video/exynos_mipi_dsi.c | 4 +- drivers/video/tegra.c | 2 +- dts/Kconfig | 6 +-- include/cli.h | 2 +- include/config_uncmd_spl.h | 4 -- include/configs/microblaze-generic.h | 3 +- include/configs/socfpga_common.h | 2 +- include/dm/device-internal.h | 10 ++--- include/dm/device.h | 4 +- include/dm/uclass-internal.h | 4 +- include/fdtdec.h | 10 ----- include/linux/kconfig.h | 48 ++++++++++++++++++++++ lib/Makefile | 13 +++--- lib/fdtdec.c | 2 +- scripts/Kbuild.include | 6 +++ scripts/Makefile.spl | 30 +------------- scripts/Makefile.uncmd_spl | 5 --- scripts/basic/fixdep.c | 33 +++++++++++++-- 85 files changed, 225 insertions(+), 186 deletions(-) -- 1.9.1

6 52

[U-Boot] [PATCH v3] net: TFTP: add timer count environment variable
by Albert ARIBAUD (3ADEV) 13 Aug '15

13 Aug '15

There already is an environment variable 'tftptimeout' for setting individuabl block query timeout value, but there is none for setting the maximum TFTP timeout count, which can also help with unreliable network, TFTP server, or client hardware. Add environment variable 'tftptimeoutcountmax'. As per the comments about the global variable tftp_timeout_count_max, make sure tftptimeoutcountmax is nonnegative. Signed-off-by: Albert ARIBAUD (3ADEV) <albert.aribaud(a)3adev.fr> --- Changes in v3: - fixed mixup between tctp_timeout_count_max and timeout_count_max. Changes in v2: - fixed env var name in commit message README | 8 ++++++++ net/tftp.c | 14 ++++++++++++-- 2 files changed, 20 insertions(+), 2 deletions(-) diff --git a/README b/README index 4e0ff9f..56784b1 100644 --- a/README +++ b/README @@ -5477,6 +5477,14 @@ List of environment variables (most likely not complete): faster in networks with high packet loss rates or with unreliable TFTP servers. + tftptimeoutcountmax - maximum count of TFTP timeouts (no + unit, minimum value = 0). Defines how many timeouts + can happen during a single file transfer before that + transfer is aborted. The default is 10, and 0 means + 'no timeouts allowed'. Increasing this value may help + downloads succeed with high packet loss rates, or with + unreliable TFTP servers or client hardware. + vlan - When set to a value < 4095 the traffic over Ethernet is encapsulated/received over 802.1q VLAN tagged frames. diff --git a/net/tftp.c b/net/tftp.c index 3e99e73..715fe04 100644 --- a/net/tftp.c +++ b/net/tftp.c @@ -600,7 +600,7 @@ static void tftp_handler(uchar *pkt, unsigned dest, struct in_addr sip, } tftp_prev_block = tftp_cur_block; - timeout_count_max = TIMEOUT_COUNT; + timeout_count_max = tftp_timeout_count_max; net_set_timeout_handler(timeout_ms, tftp_timeout_handler); store_block(tftp_cur_block - 1, pkt + 2, len); @@ -715,6 +715,16 @@ void tftp_start(enum proto_t protocol) timeout_ms = 1000; } + ep = getenv("tftptimeoutcountmax"); + if (ep != NULL) + tftp_timeout_count_max = simple_strtol(ep, NULL, 10); + + if (tftp_timeout_count_max < 0) { + printf("TFTP timeout count max (%d ms) negative, set to 0\n", + tftp_timeout_count_max); + tftp_timeout_count_max = 0; + } + debug("TFTP blocksize = %i, timeout = %ld ms\n", tftp_block_size_option, timeout_ms); @@ -840,7 +850,7 @@ void tftp_start_server(void) puts("Loading: *\b"); - timeout_count_max = TIMEOUT_COUNT; + timeout_count_max = tftp_timeout_count_max; timeout_count = 0; timeout_ms = TIMEOUT; net_set_timeout_handler(timeout_ms, tftp_timeout_handler); -- 2.1.4

3 3

[U-Boot] [PATCH v6] mtd/nand/ubi: assortment of alignment fixes
by Marcel Ziswiler 13 Aug '15

13 Aug '15

From: Marcel Ziswiler <marcel.ziswiler(a)toradex.com> Various U-Boot adoptions/extensions to MTD/NAND/UBI did not take buffer alignment into account which led to failures of the following form: ERROR: v7_dcache_inval_range - start address is not aligned - 0x1f7f0108 ERROR: v7_dcache_inval_range - stop address is not aligned - 0x1f7f1108 Signed-off-by: Marcel Ziswiler <marcel.ziswiler(a)toradex.com> Reviewed-by: Simon Glass <sjg(a)chromium.org> Acked-by: Scott Wood <scottwood(a)freescale.com> --- Changes in v6: fix assembler build breakage caused by including regular header file like malloc.h by wrapping my wrapper in __ASSEMBLY__ Changes in v5: split up into separate patches to be picked up by the various subsystem maintainers as suggested by Marek Changes in v4: move wrapper to common.h as suggested by Scott Changes in v3: introduce malloc_cache_aligned() as suggested by Scott Changes in v2: run it through checkpatch.pl and fix long lines common/cmd_ubi.c | 2 +- drivers/mtd/nand/nand_util.c | 2 +- fs/ubifs/super.c | 5 +++-- fs/ubifs/ubifs.c | 4 ++-- include/common.h | 9 +++++++++ lib/gzip.c | 2 +- 6 files changed, 17 insertions(+), 7 deletions(-) diff --git a/common/cmd_ubi.c b/common/cmd_ubi.c index cbc10c5..10eea65 100644 --- a/common/cmd_ubi.c +++ b/common/cmd_ubi.c @@ -363,7 +363,7 @@ int ubi_volume_read(char *volume, char *buf, size_t size) tbuf_size = vol->usable_leb_size; if (size < tbuf_size) tbuf_size = ALIGN(size, ubi->min_io_size); - tbuf = malloc(tbuf_size); + tbuf = malloc_cache_aligned(tbuf_size); if (!tbuf) { printf("NO MEM\n"); return ENOMEM; diff --git a/drivers/mtd/nand/nand_util.c b/drivers/mtd/nand/nand_util.c index ee2c24d..21b4a61 100644 --- a/drivers/mtd/nand/nand_util.c +++ b/drivers/mtd/nand/nand_util.c @@ -839,7 +839,7 @@ int nand_torture(nand_info_t *nand, loff_t offset) patt_count = ARRAY_SIZE(patterns); - buf = malloc(nand->erasesize); + buf = malloc_cache_aligned(nand->erasesize); if (buf == NULL) { puts("Out of memory for erase block buffer\n"); return -ENOMEM; diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 10f8fff..0bf52db 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -57,7 +57,8 @@ struct inode *iget_locked(struct super_block *sb, unsigned long ino) { struct inode *inode; - inode = (struct inode *)malloc(sizeof(struct ubifs_inode)); + inode = (struct inode *)malloc_cache_aligned( + sizeof(struct ubifs_inode)); if (inode) { inode->i_ino = ino; inode->i_sb = sb; @@ -104,7 +105,7 @@ void iput(struct inode *inode) /* * Allocate and use new inode */ - ino = (struct inode *)malloc(sizeof(struct ubifs_inode)); + ino = (struct inode *)malloc_cache_aligned(sizeof(struct ubifs_inode)); memcpy(ino, inode, sizeof(struct ubifs_inode)); /* diff --git a/fs/ubifs/ubifs.c b/fs/ubifs/ubifs.c index 6dd6174..4daa7fa 100644 --- a/fs/ubifs/ubifs.c +++ b/fs/ubifs/ubifs.c @@ -108,7 +108,7 @@ static inline struct crypto_comp *crypto_alloc_comp(const char *alg_name, struct crypto_comp *ptr; int i = 0; - ptr = malloc(sizeof(struct crypto_comp)); + ptr = malloc_cache_aligned(sizeof(struct crypto_comp)); while (i < UBIFS_COMPR_TYPES_CNT) { comp = ubifs_compressors[i]; if (!comp) { @@ -723,7 +723,7 @@ static int do_readpage(struct ubifs_info *c, struct inode *inode, * destination area to a multiple of * UBIFS_BLOCK_SIZE. */ - buff = malloc(UBIFS_BLOCK_SIZE); + buff = malloc_cache_aligned(UBIFS_BLOCK_SIZE); if (!buff) { printf("%s: Error, malloc fails!\n", __func__); diff --git a/include/common.h b/include/common.h index fcc9ae7..2d5f24c 100644 --- a/include/common.h +++ b/include/common.h @@ -1010,6 +1010,15 @@ int cpu_release(int nr, int argc, char * const argv[]); #define DEFINE_CACHE_ALIGN_BUFFER(type, name, size) \ DEFINE_ALIGN_BUFFER(type, name, size, ARCH_DMA_MINALIGN) +#ifndef __ASSEMBLY__ +#include <malloc.h> + +static inline void *malloc_cache_aligned(size_t size) +{ + return memalign(ARCH_DMA_MINALIGN, ALIGN(size, ARCH_DMA_MINALIGN)); +} +#endif + /* * check_member() - Check the offset of a structure member * diff --git a/lib/gzip.c b/lib/gzip.c index ff37d4f..cd8e9fe 100644 --- a/lib/gzip.c +++ b/lib/gzip.c @@ -25,7 +25,7 @@ static void *zalloc(void *x, unsigned items, unsigned size) size *= items; size = (size + ZALLOC_ALIGNMENT - 1) & ~(ZALLOC_ALIGNMENT - 1); - p = malloc (size); + p = malloc_cache_aligned(size); return (p); } -- 2.4.3

1 0