Kbuild brought about many advantages for us but a significant
performance regression was reported by Simon Glass.
After some discussions and analysis, it turned out
its main cause is in $(call cc-option,...).
Historically, U-Boot parses all config.mk
(arch/*/config.mk and board/*/config.mk)
every time descending into subdirectories.
That means cc-options are evaluated over and over again.
$(call cc-option,...) is useful but costly.
So we want to evaluate them only in ./Makefile
and spl/Makefile and export compiler flags.
This commit changes the build system as follows:
- Modify scripts/Makefile.build to not include config.mk
Instead, add $(PLATFORM_CPPFLAGS) to asflags-y, ccflags-y,
cppflags-y.
- Export many variables
Going forward, Kbuild will not parse config.mk files
when it descends into subdirectories.
If we want to set variables in config.mk and use them
in subdirectories, they must be exported.
This is the list of variables to get exported:
PLATFORM_CPPFLAGS
CPUDIR
BOARDDIR
OBJCOPYFLAGS
LDFLAGS
LDFLAGS_FINAL
(used in nand_spl/board/*/*/Makefile)
CONFIG_STANDALONE_LOAD_ADDR
(used in examples/standalone/Makefile)
SYM_PREFIX
(used in examples/standalone/Makefile)
RELFLAGS
(used in examples/standalone/Makefile)
- Delete CPPFLAGS
This variable has been replaced with PLATFORM_CPPFLAGS
- Copy gcclibdir from example/standalone/Makefile
to arch/sparc/config.mk
The reference in CONFIG_STANDALONE_LOAD_ADDR must be
resolved before it is exported.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Reported-by: Simon Glass <sjg@chromium.org>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org> [on Sandbox]
Tested-by: Stephen Warren <swarren@nvidia.com> [on Tegra]
Before this commit, USE_PRIVATE_LIBGCC was defined in
arch-specific config.mk and referenced in
arch/$(ARCH)/lib/Makefile.
We are not happy about parsing config.mk again and again.
We have to keep the same behavior with a different way.
By adding "CONFIG_" prefix, this macro appears
in include/autoconf.mk, include/spl-autoconf.mk.
(And treating USE_PRIVATE_LIBGCC as CONFIG macro
is reasonable enough.)
Tegra SoC family defined USE_PRIVATE_LIBGCC as "yes"
in arch/arm/cpu/arm720t/tegra*/config.mk,
whereas did not define it in arch/arm/cpu/armv7/tegra*/config.mk.
It means Tegra enables PRIVATE_LIBGCC only for SPL.
We can describe the same behavior by adding
#ifdef CONFIG_SPL_BUILD
# define CONFIG_USE_PRIVATE_LIBGCC
#endif
to include/configs/tegra-common.h.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Tom Warren <twarren@nvidia.com>
Cc: Simon Glass <sjg@chromium.org>
Acked-by: Stephen Warren <swarren@nvidia.com>
We had switched to Kbuild so now we can specify
PLATFORM_LIBS/PLATFORM_LIBGCC with relative path.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Tom Rini <trini@ti.com>
Acked-by: Simon Glass <sjg@chromium.org>
Tested-by: Simon Glass <sjg@chromium.org>
This commit avoids generating ./SPL twice.
- Fist time descending to spl/
- Second time as a prerequisite of u-boot-with-spl.imx,
u-boot-with-nand-spl.imx.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Rename OBJCFLAGS to OBJCOPYFLAGS beforehand to use
"cmd_objcopy" in scripts/Makefile.lib in an upcoming commit.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Commit 6825a95 (kbuild: use Linux Kernel build scripts)
changed the behavior of linkage when USE_PRIAVATE_LIBGCC
is defined as "yes".
(It dropped arch/arm/lib/eabi_compat.o from the
target library.)
Affected boards are all Tegra boards.
This commit gets back the same behavior as before Kbuild series.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Tom Warren <twarren@nvidia.com>
Cc: Tom Rini <trini@ti.com>
This commit refactors rules of directory descending
and defines u-boot-dirs and u-boot-all-dirs.
(We will need u-boot-all-dirs when using
scripts/Makefile.clean)
Additionally, rename LIBS-y to libs-y.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Before this commit, {spl,tpl}-autoconf.mk was always generated
at the top Makefile even if SPL(TPL) build was not selected.
This commit moves the build rule of {spl,tpl}-autoconf.mk
from the top Makefile to spl/Makefile.
It prevents unnecessary {spl,tpl}-autoconf.mk from being
generated.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
This commit changes the location of include directives
of board configuration files.
The purpose of this change is:
- Slim down $(TOPDIR)/config.mk
- Prevent $(TOPDIR)/Makefile from including the same
configuration file twice
- Do not include include/config.mk multiple times
because ARCH, CPU, BOARD, VENDOR, SOC are exported
Before this commit:
- include/autoconf.mk was included from $(TOPDIR)/Makefile
and $(TOPDIR)/config.mk
(This means $(TOPDIR)/Makefile included include/autoconf.mk twice)
- include/{spl,tpl}-autoconf.mk was included from $(TOPDIR)/config.mk
- include/config.mk was included from $(TOPDIR)/Makefile
and $(TOPDIR)/config.mk
(This means $(TOPDIR)/Makefile included include/config.mk twice)
After this commit:
- include/autoconf.mk is included from $(TOPDIR)/Makefile
and $(TOPDIR)/scripts/Makefile.build
- include/{spl,tpl}-autoconf.mk is included from $(TOPDIR)/spl/Makefile
and $(TOPDIR)/scripts/Makefile.build
- include/config.mk is included from $(TOPDIR)/config.mk and
$(TOPDIR)/spl/Makefile
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Now we are ready to switch over to real Kbuild.
This commit disables temporary scripts:
scripts/{Makefile.build.tmp, Makefile.host.tmp}
and enables real Kbuild scripts:
scripts/{Makefile.build,Makefile.host,Makefile.lib}.
This switch is triggered by the line in scripts/Kbuild.include
-build := -f $(if $(KBUILD_SRC),$(srctree)/)scripts/Makefile.build.tmp obj
+build := -f $(if $(KBUILD_SRC),$(srctree)/)scripts/Makefile.build obj
We need to adjust some build scripts for U-Boot.
But smaller amount of modification is preferable.
Additionally, we need to fix compiler flags which are
locally added or removed.
In Kbuild, it is not allowed to change CFLAGS locally.
Instead, ccflags-y, asflags-y, cppflags-y,
CFLAGS_$(basetarget).o, CFLAGS_REMOVE_$(basetarget).o
are prepared for that purpose.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Tested-by: Gerhard Sittig <gsi@denx.de>
This commit changes the working directory
where the build process occurs.
Before this commit, build process occurred under the source
tree for both in-tree and out-of-tree build.
That's why we needed to add $(obj) prefix to all generated
files in makefiles like follows:
$(obj)u-boot.bin: $(obj)u-boot
Here, $(obj) is empty for in-tree build, whereas it points
to the output directory for out-of-tree build.
And our old build system changes the current working directory
with "make -C <sub-dir>" syntax when descending into the
sub-directories.
On the other hand, Kbuild uses a different idea
to handle out-of-tree build and directory descending.
The build process of Kbuild always occurs under the output tree.
When "O=dir/to/store/output/files" is given, the build system
changes the current working directory to that directory and
restarts the make.
Kbuild uses "make -f $(srctree)/scripts/Makefile.build obj=<sub-dir>"
syntax for descending into sub-directories.
(We can write it like "make $(obj)=<sub-dir>" with a shorthand.)
This means the current working directory is always the top
of the output directory.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Tested-by: Gerhard Sittig <gsi@denx.de>
Some flags are used for SPL (and TPL) build only.
This commit moves them from config.mk to spl/Makefile.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
This commit adjusts some files to use Kbuild.include.
- Use cc-option defined in Kbuild.include
(Delete cc-option in config.mk)
- Use cc-version defined in
(Delete cc-version in config.mk)
- Move binutils-version and dtc-version to Kbuild.include
by analogy to cc-version
This commit also adds srctree (same as SRCTREE)
to use Kbuild scripts.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Some Samsung boards have their own tools under board/samsung/<board>/tools/.
This commit refactor more makefiles with "hostprogs-y".
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Before this commit, makefiles under tools/ directory
were implemented with their own way.
This commit refactors them by using "hostprogs-y" variable.
Several C sources have been added to wrap other C sources
to simplify Makefile.
For example, tools/crc32.c includes lib/crc32.c
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Add SPL support to be able to detect a USB Mass Storage device
connected to a USB host. Once a USB Mass storage device is detected
the SPL will load the u-boot.img from a FAT partition to target address.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Using the TPL method for nand boot by sram was already
supported. Here add some code for mpc85xx ifc nand boot.
- For ifc, elbc, esdhc, espi, all need the SPL without
section .resetvec.
- Use a clear function name for nand spl boot.
- Add CONFIG_SPL_DRIVERS_MISC_SUPPORT to compile the fsl_ifc.c
in spl/Makefile;
Signed-off-by: Po Liu <Po.Liu@freescale.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Reviewed-by: York Sun <yorksun@freescale.com>
Some targets will build fine but not boot if sections .hash and
.got.plt are not present in the binary. Add them back.
Also, Exynos machines require .machine_param section in SPL.
Add it.
Signed-off-by: Albert ARIBAUD <albert.u.boot@aribaud.net>
Tested-by: Rajeshwari S Shinde <rajeshwari.s@samsung.com>
When variable size SPL is used, the BL1 expects the SPL to be
encapsulated differently: instead of putting the checksum at a fixed
offset in the SPL blob, prepend the blob with a header including the
size and the checksum.
The enhancements include
- adding a command line option, '--vs' to indicate the need for the
variable size encapsulation
- padding the fixed size encapsulated blob with 0xff instead of random
memory contents
- do not silently truncate the input file, report error instead
- no need to explicitly closing files/freeing memory, this all happens
on exit; removing cleanups it makes code clearer
- profuse commenting
- modify Makefile to allow enabling the new feature per board
Signed-off-by: Vadim Bendebury <vbendeb@chromium.org>
Signed-off-by: Rajeshwari S Shinde <rajeshwari.s@samsung.com>
Acked-by: Simon Glass <sjg@chromium.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Convert like follows:
CPU mpc83xx -> CONFIG_MPC83xx
CPU mpc85xx -> CONFIG_MPC85xx
CPU mpc86xx -> CONFIG_MPC86xx
CPU mpc5xxx -> CONFIG_MPC5xxx
CPU mpc8xx -> CONFIG_8xx
CPU mpc8260 -> CONFIG_8260
CPU ppc4xx -> CONFIG_4xx
CPU x86 -> CONFIG_X86
ARCH x86 -> CONFIG_X86
ARCH powerpc -> CONFIG_PPC
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Before this commit, output files under tpl/ directry
were not ignored.
This commit fixes this problem.
And we have only one source file under spl/ directory:
spl/Makefile
So, we can describe .gitignore more simply.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Some editors such as Emacs can highlight source files.
But their parser algorithm is not perfect.
If you use one double-quotation alone, some editor cannot
handle it nicely and mark source lines as a string by mistake.
It is preferable to use two double-quotations as a pair.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Freescale DDR driver has been used for mpc83xx, mpc85xx, mpc86xx SoCs.
The similar DDR controllers will be used for ARM-based SoCs.
Signed-off-by: York Sun <yorksun@freescale.com>
We have converted all makefiles needed to build $(LIBS).
Until this commit we used to grep switch so that U-Boot style
and Kbuild style makefiles coexist.
But we do not need any more.
Goint forward, use always Kbuild style Makefile when adding
a new Makefile
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
This patch moves S5PC, EXYNOS specific directory entries
from the toplevel Makefile to arch/arm/cpu/armv7/Makefile
using Kbuild descending feature.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Minkyu Kang <mk7.kang@samsung.com>
This patch moves OMAP specific directory entries
from the toplevel Makefile and spl/Makefile
to arch/arm/cpu/armv7/Makefile using Kbuild descending feature.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Tom Rini <trini@ti.com>
This patch moves Tegra specific directory entries
from the toplevel Makefile and spl/Makefile
to arch/arm/cpu/*/Makefile using Kbuild descending feature.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Tom Warren <TWarren@nvidia.com>
Multiple targets are included in arch/arm/imx-common/Makefile
In order to refactor it,
we need to tweak Makefile and spl/Makefile.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Every makefile in sub directories has common lines
at the top and the bottom.
This commit pushes the common parts into script/Makefile.build.
Going forward sub-makefiles only need to describe this part:
COBJS := ...
COBJS += ...
SOBJS := ...
But using obj-y is preferable to prepare for switching to Kbuild.
The conventional (non-Kbuild) Makefile style is still supported.
This is achieved by greping the Makefile before entering into it.
U-Boot conventional sub makefiles always include some other makefiles.
So the build system searches a line beginning with "include" keyword
in the makefile in order to distinguish which style it is.
If the Makefile include a "include" line, we assume it is a conventional
U-Boot style. Otherwise, it is treated as a Kbuild-style makefile.
With this tweak, we can switch sub-makefiles
from U-Boot style to Kbuild style little by little.
obj-y := foo/
syntax (descending into the sub directory) is not supportd yet.
It will be implemented in the upcomming commit.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Cc: Simon Glass <sjg@chromium.org>
Cc: Tom Rini <trini@ti.com>
Due to the nand SPL on some board(e.g. P1022DS)has a size limit, it can
not be more than 4K. So, the SPL cannot initialize the DDR with the SPD
code. This patch introduces TPL to enable a loader stub that is loaded
by the code from the SPL. It initializes the DDR with the SPD or other
operations.
The TPL's size is sizeable, the maximum size is decided by the memory's
size that TPL runs. It initializes the DDR through SPD code, and copys
final uboot image to DDR. So there are three stage uboot images:
* spl_boot, * tpl_boot, * final uboot image
Signed-off-by: Ying Zhang <b40530@freescale.com>
Acked-by: York Sun <yorksun@freescale.com>
The code from the internal on-chip ROM. It loads the final uboot image
into DDR, then jump to it to begin execution.
The SPL's size is sizeable, the maximum size must not exceed the size of L2
SRAM. It initializes the DDR through SPD code, and copys final uboot image
to DDR. So there are two stage uboot images:
* spl_boot, 96KB size. The env variables are copied to L2 SRAM, so that
ddr spd code can get the interleaving mode setting in env. It loads
final uboot image from offset 96KB.
* final uboot image, size is variable depends on the functions enabled.
Signed-off-by: Ying Zhang <b40530@freescale.com>
Acked-by: York Sun <yorksun@freescale.com>
SPL defines CONFIG_SPL_BUILD but this does not percolate to the
autoconf.mk Makefile. As a result the build breaks when
CONFIG_SPL_BUILD is used in the board-specific include header file. With
this, there is a possibility of having a CONFIG option defined in the
header file but not defined in the Makefile causing all kinds of build
failure and problems.
It also messes things for up, for example, when one might want to
undefine options to keep the SPL small and doesn't want to be stuck with
the CONFIG options used for U-boot. Lastly, this also avoids defining
special CONFIG_SPL_ variables for cases where some options are required
in U-boot but not in SPL.
We add a spl-autoconf.mk rule that is generated for SPL with the
CONFIG_SPL_BUILD flag and conditionally include it for SPL builds.
Signed-off-by: Joel Fernandes <joelf@ti.com>
Signed-off-by: Ying Zhang <b40530@freescale.com>
Adding a new CONFIG_OMAP_COMMON which is included by all boards
that needs to build cpu/armv7/omap-common folder.
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
This requires that cpu_is_exynos4/5 should be made available before tzpc_init.
Hence this patch also makes necessary changes to have cpu_info in spl and
invokes arch_cpu_init before tzpc_init in low_level_init.S for smdk5250.
Signed-off-by: Inderpal Singh <inderpal.singh@linaro.org>
Acked-by: Chander Kashyap <chander.kashyap@linaro.org>
Signed-off-by: Minkyu Kang <mk7.kang@samsung.com>
Masahiro Yamada
Dan Murphy
Po Liu
Albert ARIBAUD
Rajeshwari Birje
York Sun
Tom Rini
Heiko Schocher
Ying Zhang
Joel Fernandes
Lokesh Vutla
Wolfgang Denk
Inderpal Singh
Benoît Thébaudeau