January 2016 - U-Boot

[U-Boot] [PATCH 1/2] pci: imx: Adjust the return value when imx_pcie_addr_valid() fails
by Bin Meng 09 Jan '16

09 Jan '16

When trying to access non-existent/unsupported PCI devices in imx_pcie_read_config(), when imx_pcie_addr_valid() fails it returns error code and fills in the result with 0xffffffff manually. But it really should return zero to upper layer codes. Reported-by: Fabio Estevam <fabio.estevam(a)nxp.com> Signed-off-by: Bin Meng <bmeng.cn(a)gmail.com> Tested-by: Fabio Estevam <fabio.estevam(a)nxp.com> --- drivers/pci/pcie_imx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pci/pcie_imx.c b/drivers/pci/pcie_imx.c index f1e189e..c14bb0a 100644 --- a/drivers/pci/pcie_imx.c +++ b/drivers/pci/pcie_imx.c @@ -381,7 +381,7 @@ static int imx_pcie_read_config(struct pci_controller *hose, pci_dev_t d, ret = imx_pcie_addr_valid(d); if (ret) { *val = 0xffffffff; - return ret; + return 0; } va_address = get_bus_address(d, where); -- 1.7.9.5

2 3

[U-Boot] Asymmetry in setting of "fileaddr" variable
by David Müller (ELSOFT AG) 09 Jan '16

09 Jan '16

Hello In net/net.c::net_loop(), the variables "fileaddr" and "filesize" are both set in case of a successful transfer. In fs/fs.c::do_load(), just the variable "filesize" is set in case of success. Is this asymmetry intentional or a mistake? Dave

3 3

[U-Boot] [RFC v2 PATCH] get_maintainer: fix perl 5.22/5.24 deprecated/incompatible "\C" use
by Heiko Schocher 09 Jan '16

09 Jan '16

from linux commit ce8155f7a3d5: Perl 5.22 emits a deprecated message when "\C" is used in a regex. Perl 5.24 will disallow it altogether. Fix it by using [A-Z] instead of \C. >From linux adapted to U-Boot by: Signed-off-by: Heiko Schocher <hs(a)denx.de> Reviewed-by: Andre Przywara <andre.przywara(a)arm.com> --- running get_maintainer.pl drops warnings with PERL > 5.22 . In linux this is fixed though above commit. But ... I get differences when running the python script: import os import subprocess root_dir = '.' for directory, subdirectories, files in os.walk(root_dir): for file in files: tmp=os.path.join(directory, file) print(tmp) ret = subprocess.call(["scripts/get_maintainer.pl", "-f", tmp]) and comparing the output from it, based on the original u-boot code and the code with this patch ... I have no PERL experience, so made this patch as RFC, maybe someone with better PERL skills can fix it correct. Changes in v2: - add Reviewed-by from Andre Przywara - remove wrong backslash before [A-Z] - remove change in MAINTAINERS as no longer needed scripts/get_maintainer.pl | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/scripts/get_maintainer.pl b/scripts/get_maintainer.pl index 4707dfd..83a4e5b 100755 --- a/scripts/get_maintainer.pl +++ b/scripts/get_maintainer.pl @@ -298,7 +298,7 @@ sub read_maintainers { while (<$maint>) { my $line = $_; - if ($line =~ m/^(\C):\s*(.*)/) { + if ($line =~ m/^([A-Z]):\s*(.*)/) { my $type = $1; my $value = $2; @@ -533,7 +533,7 @@ sub range_is_maintained { for (my $i = $start; $i < $end; $i++) { my $line = $typevalue[$i]; - if ($line =~ m/^(\C):\s*(.*)/) { + if ($line =~ m/^([A-Z]):\s*(.*)/) { my $type = $1; my $value = $2; if ($type eq 'S') { @@ -551,7 +551,7 @@ sub range_has_maintainer { for (my $i = $start; $i < $end; $i++) { my $line = $typevalue[$i]; - if ($line =~ m/^(\C):\s*(.*)/) { + if ($line =~ m/^([A-Z]):\s*(.*)/) { my $type = $1; my $value = $2; if ($type eq 'M') { @@ -600,7 +600,7 @@ sub get_maintainers { for ($i = $start; $i < $end; $i++) { my $line = $typevalue[$i]; - if ($line =~ m/^(\C):\s*(.*)/) { + if ($line =~ m/^([A-Z]):\s*(.*)/) { my $type = $1; my $value = $2; if ($type eq 'X') { @@ -615,7 +615,7 @@ sub get_maintainers { if (!$exclude) { for ($i = $start; $i < $end; $i++) { my $line = $typevalue[$i]; - if ($line =~ m/^(\C):\s*(.*)/) { + if ($line =~ m/^([A-Z]):\s*(.*)/) { my $type = $1; my $value = $2; if ($type eq 'F') { @@ -917,7 +917,7 @@ sub find_first_section { while ($index < @typevalue) { my $tv = $typevalue[$index]; - if (($tv =~ m/^(\C):\s*(.*)/)) { + if (($tv =~ m/^([A-Z]):\s*(.*)/)) { last; } $index++; @@ -931,7 +931,7 @@ sub find_starting_index { while ($index > 0) { my $tv = $typevalue[$index]; - if (!($tv =~ m/^(\C):\s*(.*)/)) { + if (!($tv =~ m/^([A-Z]):\s*(.*)/)) { last; } $index--; @@ -945,7 +945,7 @@ sub find_ending_index { while ($index < @typevalue) { my $tv = $typevalue[$index]; - if (!($tv =~ m/^(\C):\s*(.*)/)) { + if (!($tv =~ m/^([A-Z]):\s*(.*)/)) { last; } $index++; @@ -971,7 +971,7 @@ sub get_maintainer_role { for ($i = $start + 1; $i < $end; $i++) { my $tv = $typevalue[$i]; - if ($tv =~ m/^(\C):\s*(.*)/) { + if ($tv =~ m/^([A-Z]):\s*(.*)/) { my $ptype = $1; my $pvalue = $2; if ($ptype eq "S") { @@ -1030,7 +1030,7 @@ sub add_categories { for ($i = $start + 1; $i < $end; $i++) { my $tv = $typevalue[$i]; - if ($tv =~ m/^(\C):\s*(.*)/) { + if ($tv =~ m/^([A-Z]):\s*(.*)/) { my $ptype = $1; my $pvalue = $2; if ($ptype eq "L") { @@ -1072,7 +1072,7 @@ sub add_categories { if ($name eq "") { if ($i > 0) { my $tv = $typevalue[$i - 1]; - if ($tv =~ m/^(\C):\s*(.*)/) { + if ($tv =~ m/^([A-Z]):\s*(.*)/) { if ($1 eq "P") { $name = $2; $pvalue = format_email($name, $address, $email_usename); -- 2.5.0

3 3

[U-Boot] [PATCH] fix always succesful memory test
by Rasmus Villemoes 09 Jan '16

09 Jan '16

Since 51209b1f42cb ("Use common mtest iteration counting"), do_mem_mtest has always reported 0 errors and hence returned 0, even if errors were detected. Fix the helpers mem_test_alt() and mem_test_quick() to return the number of errors found. Signed-off-by: Rasmus Villemoes <rasmus.villemoes(a)prevas.dk> --- common/cmd_mem.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/common/cmd_mem.c b/common/cmd_mem.c index 9fb2584..efa3929 100644 --- a/common/cmd_mem.c +++ b/common/cmd_mem.c @@ -931,7 +931,7 @@ static ulong mem_test_alt(vu_long *buf, ulong start_addr, ulong end_addr, addr[offset] = 0; } - return 0; + return errs; } static ulong mem_test_quick(vu_long *buf, ulong start_addr, ulong end_addr, @@ -990,7 +990,7 @@ static ulong mem_test_quick(vu_long *buf, ulong start_addr, ulong end_addr, val += incr; } - return 0; + return errs; } /* -- 2.5.0

2 1

[U-Boot] [PATCH] cyrus: Enable STDIO deregister
by Andy Fleming 09 Jan '16

09 Jan '16

If CONFIG_SYS_STDIO_DEREGISTER is not enabled, usb reset will fail every time. Enabling it allows usb reset to work. Signed-off-by: Andy Fleming <afleming(a)gmail.com> --- include/configs/cyrus.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/configs/cyrus.h b/include/configs/cyrus.h index bc7cac4..99b3aef 100644 --- a/include/configs/cyrus.h +++ b/include/configs/cyrus.h @@ -476,6 +476,7 @@ #define CONFIG_CMD_EXT2 #define CONFIG_EHCI_IS_TDI #define CONFIG_USB_KEYBOARD +#define CONFIG_SYS_STDIO_DEREGISTER #define CONFIG_SYS_USB_EVENT_POLL /* _VIA_CONTROL_EP */ #define CONFIG_CONSOLE_MUX -- 1.9.1

2 1

[U-Boot] [PATCH] MAINTAINERS, .mailmap: Update my email address
by York Sun 09 Jan '16

09 Jan '16

Change from yorksun(a)freescale.com to york.sun(a)nxp.com. Signed-off-by: York Sun <york.sun(a)nxp.com> --- .mailmap | 1 + MAINTAINERS | 6 +++--- board/freescale/ls2080a/MAINTAINERS | 2 +- board/freescale/mpc8572ds/MAINTAINERS | 2 +- 4 files changed, 6 insertions(+), 5 deletions(-) diff --git a/.mailmap b/.mailmap index 2f32fe8..c7bb0ef 100644 --- a/.mailmap +++ b/.mailmap @@ -28,4 +28,5 @@ Stefano Babic <sbabic(a)denx.de> TsiChung Liew <Tsi-Chung.Liew(a)freescale.com> Wolfgang Denk <wdenk> York Sun <yorksun(a)freescale.com> +York Sun <york.sun(a)nxp.com> Łukasz Majewski <l.majewski(a)samsung.com> diff --git a/MAINTAINERS b/MAINTAINERS index 5b3c93a..607bc9c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -239,7 +239,7 @@ S: Maintained T: git git://git.denx.de/u-boot-freebsd.git FREESCALE QORIQ -M: York Sun <yorksun(a)freescale.com> +M: York Sun <york.sun(a)nxp.com> S: Maintained T: git git://git.denx.de/u-boot-fsl-qoriq.git @@ -308,13 +308,13 @@ F: arch/powerpc/cpu/mpc83xx/ F: arch/powerpc/include/asm/arch-mpc83xx/ POWERPC MPC85XX -M: York Sun <yorksun(a)freescale.com> +M: York Sun <york.sun(a)nxp.com> S: Maintained T: git git://git.denx.de/u-boot-mpc85xx.git F: arch/powerpc/cpu/mpc85xx/ POWERPC MPC86XX -M: York Sun <yorksun(a)freescale.com> +M: York Sun <york.sun(a)nxp.com> S: Maintained T: git git://git.denx.de/u-boot-mpc86xx.git F: arch/powerpc/cpu/mpc86xx/ diff --git a/board/freescale/ls2080a/MAINTAINERS b/board/freescale/ls2080a/MAINTAINERS index 03ca168..975ea2d 100644 --- a/board/freescale/ls2080a/MAINTAINERS +++ b/board/freescale/ls2080a/MAINTAINERS @@ -1,5 +1,5 @@ LS2080A BOARD -M: York Sun <yorksun(a)freescale.com> +M: York Sun <york.sun(a)nxp.com> S: Maintained F: board/freescale/ls2080a/ F: include/configs/ls2080a_emu.h diff --git a/board/freescale/mpc8572ds/MAINTAINERS b/board/freescale/mpc8572ds/MAINTAINERS index 4c9b968..46c2903 100644 --- a/board/freescale/mpc8572ds/MAINTAINERS +++ b/board/freescale/mpc8572ds/MAINTAINERS @@ -1,5 +1,5 @@ MPC8572DS BOARD -M: York Sun <yorksun(a)freescale.com> +M: York Sun <york.sun(a)nxp.com> S: Maintained F: board/freescale/mpc8572ds/ F: include/configs/MPC8572DS.h -- 1.7.9.5

2 1

[U-Boot] [PATCH] am33xx/am43xx: Add platform data for GPIOs
by Tom Rini 09 Jan '16

09 Jan '16

On these platforms we have many cases of boards that enable device model and GPIO support but do not enable OF_CONTROL and pass in a device tree with the binary. We need to bring in the platform data here as well. Tested on Beaglebone Black. Reported-by: Robert Nelson <robertcnelson(a)gmail.com> Reported-by: Francisco Aguerre <franciscoaguerre(a)gmail.com> Reported-by: Jason Kridner <jkridner(a)beagleboard.org> Signed-off-by: Tom Rini <trini(a)konsulko.com> --- arch/arm/cpu/armv7/am33xx/board.c | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) diff --git a/arch/arm/cpu/armv7/am33xx/board.c b/arch/arm/cpu/armv7/am33xx/board.c index 466348f..e8d5be3 100644 --- a/arch/arm/cpu/armv7/am33xx/board.c +++ b/arch/arm/cpu/armv7/am33xx/board.c @@ -64,8 +64,31 @@ U_BOOT_DEVICES(am33xx_uarts) = { # endif # endif }; + +#ifdef CONFIG_DM_GPIO +static const struct omap_gpio_platdata am33xx_gpio[] = { + { 0, AM33XX_GPIO0_BASE }, + { 1, AM33XX_GPIO1_BASE }, + { 2, AM33XX_GPIO2_BASE }, + { 3, AM33XX_GPIO3_BASE }, +#ifdef CONFIG_AM43XX + { 4, AM33XX_GPIO4_BASE }, + { 5, AM33XX_GPIO5_BASE }, #endif +}; +U_BOOT_DEVICES(am33xx_gpios) = { + { "gpio_omap", &am33xx_gpio[0] }, + { "gpio_omap", &am33xx_gpio[1] }, + { "gpio_omap", &am33xx_gpio[2] }, + { "gpio_omap", &am33xx_gpio[3] }, +#ifdef CONFIG_AM43XX + { "gpio_omap", &am33xx_gpio[4] }, + { "gpio_omap", &am33xx_gpio[5] }, +#endif +}; +#endif +#endif #ifndef CONFIG_DM_GPIO static const struct gpio_bank gpio_bank_am33xx[] = { -- 1.7.9.5

1 1

[U-Boot] [PATCH v3] serial: zynq: Use static inline for _debug_uart_init()
by Michal Simek 09 Jan '16

09 Jan '16

Mark _debug_uart_init() as static to avoid sparse warning and inline it to debug_uart_init(). Reported-by: Thomas Chou <thomas(a)wytron.com.tw> Signed-off-by: Michal Simek <michal.simek(a)xilinx.com> Reviewed-by: Thomas Chou <thomas(a)wytron.com.tw> Reviewed-by: Simon Glass <sjg(a)chromium.org> Reviewed-by: Bin Meng <bmeng.cn(a)gmail.com> --- Changes in v3: - Separate uartlite from this patch Changes in v2: - Add tags - Extend commit message reported by Bin drivers/serial/serial_zynq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/serial/serial_zynq.c b/drivers/serial/serial_zynq.c index b2b98dea1561..3430482f8d8b 100644 --- a/drivers/serial/serial_zynq.c +++ b/drivers/serial/serial_zynq.c @@ -192,7 +192,7 @@ U_BOOT_DRIVER(serial_zynq) = { }; #ifdef CONFIG_DEBUG_UART_ZYNQ -void _debug_uart_init(void) +static inline void _debug_uart_init(void) { struct uart_zynq *regs = (struct uart_zynq *)CONFIG_DEBUG_UART_BASE; -- 1.9.1

3 2

[U-Boot] [PATCH] arm64: Juno/FVP: adjust kernel load address
by Andre Przywara 09 Jan '16

09 Jan '16

The default kernel load offset for an arm64 kernel is 0x80000, so U-Boot takes cares of moving the loaded kernel to a matching memory location just before booting it. Since we run with caches off, this takes a while for any decently sized kernel - with no output explaining the reason for the delay (unless one uses a DEBUG build). By adjusting the default load offset for Juno and VFP to be 512K aligned in the first place we can skip this copying and boot much faster. Tested on Juno. Signed-off-by: Andre Przywara <andre.przywara(a)arm.com> --- include/configs/vexpress_aemv8a.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/configs/vexpress_aemv8a.h b/include/configs/vexpress_aemv8a.h index c767f90..814934a 100644 --- a/include/configs/vexpress_aemv8a.h +++ b/include/configs/vexpress_aemv8a.h @@ -189,7 +189,7 @@ #define CONFIG_EXTRA_ENV_SETTINGS \ "kernel_name=norkern\0" \ "kernel_alt_name=Image\0" \ - "kernel_addr=0x80000000\0" \ + "kernel_addr=0x80080000\0" \ "initrd_name=ramdisk.img\0" \ "initrd_addr=0x84000000\0" \ "fdt_name=board.dtb\0" \ @@ -233,7 +233,7 @@ #elif CONFIG_TARGET_VEXPRESS64_BASE_FVP #define CONFIG_EXTRA_ENV_SETTINGS \ "kernel_name=Image\0" \ - "kernel_addr=0x80000000\0" \ + "kernel_addr=0x80080000\0" \ "initrd_name=ramdisk.img\0" \ "initrd_addr=0x88000000\0" \ "fdt_name=devtree.dtb\0" \ -- 2.6.4

4 3

[U-Boot] [PATCH] arm, Makefile: correct compilation flag for u-boot-dtb
by Aneesh Bansal 09 Jan '16

09 Jan '16

The compilation of u-boot-dtb.img should be controlled by CONFIG_OF_CONTROL and not CONFIG_DM. CONFIG_DM may be defined even without Device Tree requirement. This was added in commit 947cee1127c5fa97529c8cda4f0b48d141f92560. Signed-off-by: Ruchika Gupta <ruchika.gupta(a)freescale.com> Signed-off-by: Aneesh Bansal <aneesh.bansal(a)freescale.com> CC: Alison Wang <alison.wang(a)freescale.com> --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index b58f283..a6af95d 100644 --- a/Makefile +++ b/Makefile @@ -1136,7 +1136,7 @@ spl/u-boot-spl.pbl: spl/u-boot-spl.bin FORCE $(call if_changed,mkimage) ifeq ($(ARCH),arm) -ifdef CONFIG_DM +ifdef CONFIG_OF_CONTROL UBOOT_BINLOAD := u-boot-dtb.img else UBOOT_BINLOAD := u-boot.img -- 1.8.1.4

2 1