Leigh Brown [Wed, 27 Feb 2019 11:32:16 +0000 (11:32 +0000)]
Commit "ARM: CPU: arm926ejs: Consolidate cache routines to common file" breaks u-boot on Dreamplug
Hello,
Vagrant Cascadian asked for people to test the version of u-boot packaged
for Debian Buster. I tested u-boot on my Dreamplug and found it was not
working correctly. I raised a bug for Debian[1] but I have also tested
with the mainline version of u-boot and found the same issues.
The first issue is that the following commit caused u-boot to no longer
be able to access usb storage on the Dreamplug:
ARM: CPU: arm926ejs: Consolidate cache routines to common file
Four different boards had different options for enabling cache
that were virtually all the same. This consolidates these
common functions into arch/arm/cpu/arm926ejs/cache.c
This also has the positive side-effect of enabling cache on
the Davinci (da850) boards.
Signed-off-by: Adam Ford <aford173@gmail.com>
[trini: Add mach-at91 to the list of consolidations] Signed-off-by: Tom Rini <trini@konsulko.com>
I don't have much knowledge of ARM caching, but the following patch makes
it work again on my Dreamplug.
Gbp-Pq: Topic dreamplug
Gbp-Pq: Name Commit-ARM-CPU-arm926ejs-Consolidate-cache-routines-.patch
Chris Packham [Thu, 28 Feb 2019 07:53:23 +0000 (20:53 +1300)]
ARM: kirkwood: add spi0 alias for dreamplug
The conversion to DM_SPI managed to break accessing the environment on
dreamplug. This is because the environment code relies on being to able
to select the SPI device based on the sequence number. Add an alias so
that the spi0 bus gets sequence number 0.
Reported-by: Leigh Brown <leigh@solinno.co.uk> Signed-off-by: Chris Packham <judge.packham@gmail.com>
Gbp-Pq: Topic dreamplug
Gbp-Pq: Name ARM-kirkwood-add-spi0-alias-for-dreamplug.patch
Debian's flash-kernel expect the console variable to just contain the device,
because it will set the bootargs to "console=${console}". So revert adding
"console=" to the console parameter, but also adjust the shipped bootscripts
for exynos boards to cope with it.
Bug-Debian: https://bugs.debian.org/920116 Signed-off-by: Benjamin Drung <bdrung@debian.org>
Gbp-Pq: Topic exynos
Gbp-Pq: Name 0001-arm-config-fix-default-console-only-to-specify-the-d.patch
Marek Vasut [Wed, 6 Feb 2019 12:19:29 +0000 (13:19 +0100)]
fs: ext4: Unmount FS in do_fs_type()
Unlike other generic FS accessors, fs_get_info() does not call fs_close()
at the end of it's operation. Thus, using fs_get_info() in do_fs_type()
without calling fs_close() causes potential memory leak by creating new
filesystem structures on each call of do_fs_type().
The test case to trigger this problem is as follows. It is required to
have ext4 filesystem on the first partition of the SDMMC device, since
ext4 requires stateful mount and causes memory allocation.
=> while true ; do mmc rescan ; fstype mmc 1 ; done
Eventually, the mounting of ext4 will fail due to malloc failures
and the filesystem will not be correctly detected.
This patch fixes the problem by adding the missing fs_close().
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Simon Glass <sjg@chromium.org> Cc: Tom Rini <trini@konsulko.com>
Gbp-Pq: Topic upstream
Gbp-Pq: Name fs-ext4-Unmount-FS-in-do_fs_type.patch
Ensure that CONFIG_SANDBOX is set when running "make env", avoiding a
failure to build caused by config_distro_bootcmd.h following the wrong
codepath...
Gbp-Pq: Name ensure-config-sandbox-for-make-env.patch
[ Romain Perier ]
* [armel] u-boot-rpi: Add Raspberry Pi Zero W.
[ Vagrant Cascadian ]
* [armhf] u-boot-sunxi: Add bananapi_m2_berry target.
(Closes: #923443) Thanks to Lucas Nussbaum.
* [armhf] u-boot-imx: Add novena-rawsd target.
* Add patches to fix SPI and USB on Dreamplug (Closes: #923379).
Thanks to Leigh Brown and Chris Packham.
* u-boot-tools: Remove Synopsys lintian override.
* Add patch to fix spelling of "mismatched".
* Add patch to fix spelling of "download".
Debian's flash-kernel expect the console variable to just contain the device,
because it will set the bootargs to "console=${console}". So revert adding
"console=" to the console parameter, but also adjust the shipped bootscripts
for exynos boards to cope with it.
Bug-Debian: https://bugs.debian.org/920116 Signed-off-by: Benjamin Drung <bdrung@debian.org>
Gbp-Pq: Topic exynos
Gbp-Pq: Name 0001-arm-config-fix-default-console-only-to-specify-the-d.patch
Marek Vasut [Wed, 6 Feb 2019 12:19:29 +0000 (13:19 +0100)]
fs: ext4: Unmount FS in do_fs_type()
Unlike other generic FS accessors, fs_get_info() does not call fs_close()
at the end of it's operation. Thus, using fs_get_info() in do_fs_type()
without calling fs_close() causes potential memory leak by creating new
filesystem structures on each call of do_fs_type().
The test case to trigger this problem is as follows. It is required to
have ext4 filesystem on the first partition of the SDMMC device, since
ext4 requires stateful mount and causes memory allocation.
=> while true ; do mmc rescan ; fstype mmc 1 ; done
Eventually, the mounting of ext4 will fail due to malloc failures
and the filesystem will not be correctly detected.
This patch fixes the problem by adding the missing fs_close().
Signed-off-by: Marek Vasut <marek.vasut+renesas@gmail.com> Cc: Simon Glass <sjg@chromium.org> Cc: Tom Rini <trini@konsulko.com>
Gbp-Pq: Topic upstream
Gbp-Pq: Name fs-ext4-Unmount-FS-in-do_fs_type.patch
Ensure that CONFIG_SANDBOX is set when running "make env", avoiding a
failure to build caused by config_distro_bootcmd.h following the wrong
codepath...
Gbp-Pq: Name ensure-config-sandbox-for-make-env.patch
[ Vagrant Cascadian ]
* debian/patches: Add patch from upstream to unmount ext4 filesystems
preventing a memory leak.
* Add patch to fix console setting on exynos platforms. Thanks to
Dongjin Kim and Benjamin Drung. (Closes: #920116).
* Use wildcards for lintian overrides.
* Update tester for p2371-2180 (Jetson-TX1).
* Update tester for novena.
Ensure that CONFIG_SANDBOX is set when running "make env", avoiding a
failure to build caused by config_distro_bootcmd.h following the wrong
codepath...
Gbp-Pq: Name ensure-config-sandbox-for-make-env.patch
Pinebook is a laptop produced by Pine64, with USB-connected keyboard,
USB-connected touchpad and an eDP LCD panel connected via a RGB-eDP
bridge from Analogix.
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Tested-by: Maxime Ripard <maxime.ripard@bootlin.com> Cc: Vagrant Cascadian <vagrant@debian.org> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0010-sunxi-DT-add-support-for-Pinebook.patch
Andre Przywara [Mon, 29 Oct 2018 00:56:48 +0000 (00:56 +0000)]
sunxi: A64: Re-add syscon to DT node
The sun50i-a64.dtsi changes introduced in Linux v4.19-rc1 changed the
compatible name for the syscon controller, dropping the generic "syscon"
fallback. Using this new DT node will make the Ethernet driver in every
older kernel (or non-Linux kernels) fail to initialise the MAC device.
To allow booting distribution kernels (from installer images via UEFI,
for instance), re-add the syscon compatible string as a fallback. This
works with both older and newer kernels.
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0004-sunxi-A64-Re-add-syscon-to-DT-node.patch
Andre Przywara [Mon, 29 Oct 2018 00:56:47 +0000 (00:56 +0000)]
sunxi: A64: Update .dts/.dtsi files
Update the .dts/.dtsi file from the Linux sunxi/dt64-for-4.20 tree:
commit 679294497be31596e1c9c61507746d72b6b05f26
Author: Rodrigo Exterckötter Tjäder <rodrigo@tjader.xyz>
Date: Wed Sep 26 19:48:24 2018 +0000
arm64: dts: allwinner: a64: a64-olinuxino: set the PHY TX delay
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0003-sunxi-A64-Update-.dts-.dtsi-files.patch
Ensure that CONFIG_SANDBOX is set when running "make env", avoiding a
failure to build caused by config_distro_bootcmd.h following the wrong
codepath...
Gbp-Pq: Name ensure-config-sandbox-for-make-env.patch
[MIPS] Fix little-endian build with non-ELDK toolchains
We've been in trouble for a long time when cross compiling with non-ELDK
toolchains. This is caused by -EB passed to CPPFLAGS incorrectly, by the
lack of an endian specifier to LDFLAGS, and by wrong OUTPUT_FORMATs.
We're going to implement two workarounds. One is the endianness specifier
bugfix not to pass -EB / -EL to CPPFLAGS unless ELDK toolchain is used.
Note that ELDK and non-ELDK toolchains know their default endianness, so
the endianness specifier may not be necessary in principle.
The other is removal of OUTPUT_FORMAT in *.lds files. If we have this,
it doesn't work unless an endianness specifier is added to LDFLAGS. As
we haven't added that to LDFLAGS so far, it must have not worked properly,
except ELDK; I don't know why and how ELDK works, though.
With these two changes, all objects will be generated and linked in the
toolchain's default endianness. Then MAKEALL mips_el will work even with
non-ELDK toolchain.
Note that Linux/MIPS kernel has CONFIG_CPU_BIG_ENDIAN and
CONFIG_CPU_LITTLE_ENDIAN alternatives to allow users to compile kernels
with a toolchain for the other endianness. But U-Boot does not have such
feature for now, and it's another story.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
Gbp-Pq: Name mipsel-native-endianness.diff
* u-boot-install-sunxi64:
- Detect target based on running device-tree.
- Prefer BL31 from arm-trusted-firmware over atf-allwinner.
- Error out when mkimage is not found and Recommend u-boot-tools
(Closes: #913879). Thanks to Nicolas Schier.
* [arm64] u-boot-sunxi: Update Recommends to use arm-trusted-firmware
instead of atf-allwinner.
* [arm64] u-boot-amlogic: Add Recommends on arm-trusted-firmware.
* Update upstream signing key.
Pinebook is a laptop produced by Pine64, with USB-connected keyboard,
USB-connected touchpad and an eDP LCD panel connected via a RGB-eDP
bridge from Analogix.
Signed-off-by: Vasily Khoruzhick <anarsoul@gmail.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Tested-by: Maxime Ripard <maxime.ripard@bootlin.com> Cc: Vagrant Cascadian <vagrant@debian.org> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0010-sunxi-DT-add-support-for-Pinebook.patch
Andre Przywara [Mon, 29 Oct 2018 00:56:48 +0000 (00:56 +0000)]
sunxi: A64: Re-add syscon to DT node
The sun50i-a64.dtsi changes introduced in Linux v4.19-rc1 changed the
compatible name for the syscon controller, dropping the generic "syscon"
fallback. Using this new DT node will make the Ethernet driver in every
older kernel (or non-Linux kernels) fail to initialise the MAC device.
To allow booting distribution kernels (from installer images via UEFI,
for instance), re-add the syscon compatible string as a fallback. This
works with both older and newer kernels.
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0004-sunxi-A64-Re-add-syscon-to-DT-node.patch
Andre Przywara [Mon, 29 Oct 2018 00:56:47 +0000 (00:56 +0000)]
sunxi: A64: Update .dts/.dtsi files
Update the .dts/.dtsi file from the Linux sunxi/dt64-for-4.20 tree:
commit 679294497be31596e1c9c61507746d72b6b05f26
Author: Rodrigo Exterckötter Tjäder <rodrigo@tjader.xyz>
Date: Wed Sep 26 19:48:24 2018 +0000
arm64: dts: allwinner: a64: a64-olinuxino: set the PHY TX delay
Signed-off-by: Andre Przywara <andre.przywara@arm.com> Acked-by: Maxime Ripard <maxime.ripard@bootlin.com> Reviewed-by: Jagan Teki <jagan@openedev.com>
Gbp-Pq: Topic pinebook
Gbp-Pq: Name 0003-sunxi-A64-Update-.dts-.dtsi-files.patch
Ensure that CONFIG_SANDBOX is set when running "make env", avoiding a
failure to build caused by config_distro_bootcmd.h following the wrong
codepath...
Gbp-Pq: Name ensure-config-sandbox-for-make-env.patch
[MIPS] Fix little-endian build with non-ELDK toolchains
We've been in trouble for a long time when cross compiling with non-ELDK
toolchains. This is caused by -EB passed to CPPFLAGS incorrectly, by the
lack of an endian specifier to LDFLAGS, and by wrong OUTPUT_FORMATs.
We're going to implement two workarounds. One is the endianness specifier
bugfix not to pass -EB / -EL to CPPFLAGS unless ELDK toolchain is used.
Note that ELDK and non-ELDK toolchains know their default endianness, so
the endianness specifier may not be necessary in principle.
The other is removal of OUTPUT_FORMAT in *.lds files. If we have this,
it doesn't work unless an endianness specifier is added to LDFLAGS. As
we haven't added that to LDFLAGS so far, it must have not worked properly,
except ELDK; I don't know why and how ELDK works, though.
With these two changes, all objects will be generated and linked in the
toolchain's default endianness. Then MAKEALL mips_el will work even with
non-ELDK toolchain.
Note that Linux/MIPS kernel has CONFIG_CPU_BIG_ENDIAN and
CONFIG_CPU_LITTLE_ENDIAN alternatives to allow users to compile kernels
with a toolchain for the other endianness. But U-Boot does not have such
feature for now, and it's another story.
Signed-off-by: Shinya Kuribayashi <skuribay@ruby.dti.ne.jp>
Gbp-Pq: Name mipsel-native-endianness.diff
* New upstream release.
* [armhf] u-boot-imx: Drop udoo patches, full support for distro_bootcmd
is enabled upstream.
* [armhf] u-boot-rockchip: Drop firefly-rk3288 target (Closes: #898520).
* [arm64] u-boot-sunxi: Enable a64-olinuxino target (Closes: #881564).
Thanks to Rodrigo Exterckötter Tjäder.
* Add Pinebook support patches from sunxi maintainer tree.
* [arm64] u-boot-sunxi: Add pinebook target.
* [armel] Drop openrd targets, which FTBFS and are orphaned upstream.
Jonathan Gray [Tue, 8 May 2018 10:43:01 +0000 (20:43 +1000)]
rockchip: set SYS_NS16550_MEM32 for all SoCs
Add back part of patch send out as
'rockchip: enable SYS_NS16550 for all SoCs by default' that seems to have
gotten lost when it got merged to set SYS_NS16550_MEM32.
Signed-off-by: Kever Yang <kever.yang@rock-chips.com> Signed-off-by: Jonathan Gray <jsg@jsg.id.au> Cc: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Reviewed-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Acked-by: Philipp Tomsich <philipp.tomsich@theobroma-systems.com> Tested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Gbp-Pq: Topic rockchip
Gbp-Pq: Name 0001-rockchip-set-SYS_NS16550_MEM32-for-all-SoCs.patch
Peter Lebbing [Thu, 28 Sep 2017 12:40:24 +0000 (14:40 +0200)]
Bug#877074: u-boot-exynos: Default environment for Odroid XU3 has wrong console var
Package: u-boot-exynos
Version: 2017.09+dfsg1-1
Severity: normal
Tags: patch
The odroid-xu3 platform passes a wrong "console" argument on the kernel commandline:
console=console=ttySAC2,115200n8
>From the source, it's clear where this comes from:
include/configs/odroid_xu3.h:
#define CONFIG_DEFAULT_CONSOLE "console=ttySAC2,115200n8\0"
#define CONFIG_EXTRA_ENV_SETTINGS \
[...]
"console=" CONFIG_DEFAULT_CONSOLE \
The patch below makes the situation equal to what Debian patches in odroid.h: drop it from CONFIG_DEFAULT_CONSOLE. Judging from a grep in include/configs, I think the same situation exists for s5pc210_universal.h, s5p_goni.h, trats2.h and trats.h, but I don't have these devices. There are many more header files that have a "console=" in their CONFIG_DEFAULT_CONSOLE, but I don't see that definition being subsequently used somewhere, so I don't know whether it's being used wrongly or not.
Thanks for your efforts,
Peter.
Gbp-Pq: Topic odroid-xu3
Gbp-Pq: Name 877074-u-boot-exynos-Default-environment-for-Odr.patch
projects / u-boot.git / log