[Buildroot] [RFC] ti-gfx: add new package
Arnout Vandecappelle
arnout at mind.be
Thu Jun 13 16:49:47 UTC 2013
Hi Spenser,
On 07/06/13 22:18, Spenser Gilliland wrote:
> This is WIP of adding the TI Graphics SDK to buildroot.
>
> TODO:
> Install startup files for sgx core.
> Install khronos gles/egl/vg libraries.
> Create .pc files to simplfy additional packaging.
> Testing, lots of testing.
>
> ---
> package/Config.in | 1 +
> package/opengl/libegl/libegl.mk | 4 +
> package/opengl/libgles/libgles.mk | 4 +
> package/opengl/libopenvg/libopenvg.mk | 4 +
> package/ti-gfx/Config.in | 54 ++++++++++++++++++
> package/ti-gfx/ti-gfx-km_install_modules.patch | 14 +++++
> package/ti-gfx/ti-gfx-newclkapi.patch | 55 ++++++++++++++++++
> package/ti-gfx/ti-gfx.mk | 71 ++++++++++++++++++++++++
> 8 files changed, 207 insertions(+), 0 deletions(-)
> create mode 100644 package/ti-gfx/Config.in
> create mode 100644 package/ti-gfx/ti-gfx-km_install_modules.patch
> create mode 100644 package/ti-gfx/ti-gfx-newclkapi.patch
> create mode 100644 package/ti-gfx/ti-gfx.mk
>
> diff --git a/package/Config.in b/package/Config.in
> index ce82e99..8c40b56 100644
> --- a/package/Config.in
> +++ b/package/Config.in
> @@ -290,6 +290,7 @@ source "package/snowball-hdmiservice/Config.in"
> source "package/sredird/Config.in"
> source "package/statserial/Config.in"
> source "package/sysstat/Config.in"
> +source "package/ti-gfx/Config.in"
Don't we usually put this kind of stuff under Libraries / Graphics?
> source "package/ti-utils/Config.in"
> source "package/uboot-tools/Config.in"
> source "package/udev/Config.in"
[snip]
> diff --git a/package/ti-gfx/Config.in b/package/ti-gfx/Config.in
> new file mode 100644
> index 0000000..c3ac984
> --- /dev/null
> +++ b/package/ti-gfx/Config.in
> @@ -0,0 +1,54 @@
> +config BR2_PACKAGE_TI_GFX
> + bool "ti-gfx"
> + select BR2_LINUX_KERNEL
> + select BR2_PACKAGE_DEVMEM2
Why is this needed?
> + select BR2_PACKAGE_HAS_OPENGL_EGL
> + select BR2_PACKAGE_HAS_OPENGL_ES
> + depends on BR2_TOOLCHAIN_EXTERNAL_GLIBC || BR2_TOOLCHAIN_CTNG_eglibc || BR2_TOOLCHAIN_CTNG_glibc
> + help
> + Graphics libraries for TI boards.
> +
> + http://downloads.ti.com/dsps/dsps_public_sw/gfxsdk/
> +
> +if BR2_PACKAGE_TI_GFX
> +
> +config BR2_PACKAGE_TI_GFX_DEBUG
> + bool "enable debug support"
> + help
> + Turn on debugging in kernel module and install libraries built with
> + debugging enabled
> +
> +choice
> + prompt "Target"
> + default BR2_PACKAGE_TI_GFX_OMAP3
I guess you mean _ES3
> + help
> + Select the SOC for which you would like to install drivers. Please
> + use the chart at
> + http://processors.wiki.ti.com/index.php/OMAP35x_Graphics_SDK_Getting_Started_Guide
> +
> +config BR2_PACKAGE_TI_GFX_ES3
> + bool "es3.x"
> + help
> + OMAP35xx, AM35xx Rev 3.1+
> +
> +config BR2_PACKAGE_TI_GFX_ES5
> + bool "es5.x"
> + help
> + AM37xx, DM37xx
> +
> +config BR2_PACKAGE_TI_GFX_ES6
> + bool "es6.x"
> + help
> + AM387x, C6A814x, AM389x, C6A816x
> +
> +config BR2_PACKAGE_TI_GFX_ES8
> + bool "es8.x"
> + help
> + AM335x
> +
> +endchoice
> +
> +endif
> +
> +comment "requires an external eglibc/glibc based toolchain"
You must say that ti-gfx requires it, because the config option is not
visible when there's no glibc.
> + depends on !(BR2_TOOLCHAIN_EXTERNAL_GLIBC || BR2_TOOLCHAIN_CTNG_eglibc || BR2_TOOLCHAIN_CTNG_glibc)
We really should refactor this into a single symbol...
> diff --git a/package/ti-gfx/ti-gfx-km_install_modules.patch b/package/ti-gfx/ti-gfx-km_install_modules.patch
> new file mode 100644
> index 0000000..63bfd19
> --- /dev/null
> +++ b/package/ti-gfx/ti-gfx-km_install_modules.patch
> @@ -0,0 +1,14 @@
> +Index: ti-gfx-4_09_00_01/GFX_Linux_KM/Makefile
> +===================================================================
> +--- ti-gfx-4_09_00_01.orig/GFX_Linux_KM/Makefile 2013-03-07 11:00:11.000000000 -0600
> ++++ ti-gfx-4_09_00_01/GFX_Linux_KM/Makefile 2013-05-23 01:36:29.356676281 -0500
> +@@ -479,6 +479,9 @@
> + all:
> + $(MAKE) -C $(KERNELDIR) M=`pwd` $*
> +
> ++install:
> ++ $(MAKE) -C $(KERNELDIR) M=`pwd` modules_install
> ++
Actually, instead of patching the Makefile, you could just run the
kernel make command directly from buildroot's .mk file. I think there are
a few module packages that do that.
> + clean:
> + @find . -name "*.o" -exec rm -r {} \;
> + @find . -name "*.ko" -exec rm -r {} \;
> diff --git a/package/ti-gfx/ti-gfx-newclkapi.patch b/package/ti-gfx/ti-gfx-newclkapi.patch
> new file mode 100644
> index 0000000..b35ac31
> --- /dev/null
> +++ b/package/ti-gfx/ti-gfx-newclkapi.patch
> @@ -0,0 +1,55 @@
This patch is missing an explanation and a SoB.
> +Index: ti-gfx-4_09_00_01/GFX_Linux_KM/services4/system/omap3630/sysutils_linux.c
> +===================================================================
> +--- ti-gfx-4_09_00_01.orig/GFX_Linux_KM/services4/system/omap3630/sysutils_linux.c 2013-03-04 08:41:35.000000000 -0600
> ++++ ti-gfx-4_09_00_01/GFX_Linux_KM/services4/system/omap3630/sysutils_linux.c 2013-06-05 12:00:08.778605221 -0500
> +@@ -166,11 +166,27 @@
> + }
> +
> + PVR_DPF((PVR_DBG_MESSAGE, "EnableSGXClocks: Enabling SGX Clocks"));
> +-
> ++ res=clk_prepare(psSysSpecData->psSGX_FCK);
> ++ if (res < 0)
> ++ {
> ++ PVR_DPF((PVR_DBG_ERROR, "EnableSGXClocks: Couldn't enable SGX functional clock (%d)", res));
> ++ clk_unprepare(psSysSpecData->psSGX_FCK);
> ++ return PVRSRV_ERROR_UNABLE_TO_ENABLE_CLOCK;
> ++ }
> ++
> + res=clk_enable(psSysSpecData->psSGX_FCK);
> + if (res < 0)
> + {
> + PVR_DPF((PVR_DBG_ERROR, "EnableSGXClocks: Couldn't enable SGX functional clock (%d)", res));
> ++ clk_unprepare(psSysSpecData->psSGX_FCK);
> ++ return PVRSRV_ERROR_UNABLE_TO_ENABLE_CLOCK;
> ++ }
> ++
> ++ res=clk_prepare(psSysSpecData->psSGX_ICK);
> ++ if(res < 0)
> ++ {
> ++ PVR_DPF((PVR_DBG_ERROR, "EnableSGXClocks: Couldn't enable SGX functional clock (%d)", res));
> ++ clk_unprepare(psSysSpecData->psSGX_FCK);
> + return PVRSRV_ERROR_UNABLE_TO_ENABLE_CLOCK;
> + }
> +
> +@@ -178,8 +194,9 @@
> + if (res < 0)
> + {
> + PVR_DPF((PVR_DBG_ERROR, "EnableSGXClocks: Couldn't enable SGX interface clock (%d)", res));
> +-
> ++ clk_unprepare(psSysSpecData->psSGX_ICK);
> + clk_disable(psSysSpecData->psSGX_FCK);
> ++ clk_unprepare(psSysSpecData->psSGX_FCK);
> + return PVRSRV_ERROR_UNABLE_TO_ENABLE_CLOCK;
> + }
> +
> +@@ -247,8 +264,9 @@
> + PVR_DPF((PVR_DBG_MESSAGE, "DisableSGXClocks: Disabling SGX Clocks"));
> +
> + clk_disable(psSysSpecData->psSGX_FCK);
> +-
> ++ clk_unprepare(psSysSpecData->psSGX_FCK);
> + clk_disable(psSysSpecData->psSGX_ICK);
> ++ clk_unprepare(psSysSpecData->psSGX_ICK);
> +
> + // SysDisableSGXInterrupts(psSysData);
> +
> diff --git a/package/ti-gfx/ti-gfx.mk b/package/ti-gfx/ti-gfx.mk
> new file mode 100644
> index 0000000..154c792
> --- /dev/null
> +++ b/package/ti-gfx/ti-gfx.mk
> @@ -0,0 +1,71 @@
> +###############################################################################
> +#
> +# ti-gfx
> +#
> +###############################################################################
> +
> +TI_GFX_VERSION = 4_09_00_01
> +TI_GFX_SOURCE = Graphics_SDK_setuplinux_$(TI_GFX_VERSION)_minimal_demos.bin
> +TI_GFX_SITE = http://software-dl.ti.com/dsps/dsps_public_sw/sdo_sb/targetcontent/gfxsdk/$(TI_GFX_VERSION)/exports/
> +TI_GFX_LICENSE = TSPA
This is not one of the well-known licenses, so should be expanded to
"Texas Instruments Software Piracy Agreement" or whatever it stands for.
> +TI_GFX_LICENSE_FILES = TSPA.txt
> +TI_GFX_INSTALL_STAGING = YES
> +
> +TI_GFX_DEPENDENCIES = linux
> +
> +ifeq ($(BR2_PACKAGE_TI_GFX_ES3),y)
> +TI_GFX_OMAPES = 3.x
> +TI_GFX_PLATFORM = omap3
> +endif
> +ifeq ($(BR2_PACKAGE_TI_GFX_ES5),y)
> +TI_GFX_OMAPES = 5.x
> +TI_GFX_PLATFORM = omap3630
> +endif
> +ifeq ($(BR2_PACKAGE_TI_GFX_ES6),y)
> +TI_GFX_OMPAES = 6.x
> +TI_GFX_PLATFORM = ti81xx
> +endif
> +ifeq ($(BR2_PACKAGE_TI_GFX_ES8),y)
> +TI_GFX_OMAPES = 8.x
> +TI_GFX_PLATFORM = ti335x
> +endif
Personally I prefer to do "case statements" like this in the Config.in,
but it's a minor thing.
> +
> +define TI_GFX_EXTRACT_CMDS
> + $(RM) -rf $(TI_GFX_DIR)
We normally don't do that. Is it needed (i.e. does it fail if you don't
and the source was already extracted before)?
Also, use $(@D) instead of $(TI_GFX_DIR).
> + chmod +x $(DL_DIR)/$(TI_GFX_SOURCE)
> + printf "Y\nY\n qY\n\n" | $(DL_DIR)/$(TI_GFX_SOURCE) \
> + --prefix $(TI_GFX_DIR) \
> + --mode console
> +endef
> +
> +define TI_GFX_CONFIGURE_CMDS
> + echo "no configure required binary only package"
Just don't define it then.
> +endef
> +
> +TI_GFX_MAKE_CMD = cd $(TI_GFX_SRCDIR)/GFX_Linux_KM && \
> + $(MAKE) $(LINUX_MAKE_FLAGS) \
> + BUILD=$(BR2_PACKAGE_TI_GFX_DEBUG) \
> + TI_PLATFORM=$(TI_GFX_PLATFORM) \
> + OMAPES=$(TI_GFX_OMAPES) \
> + SUPPORT_XORG=0 \
> + KERNELDIR=$(LINUX_DIR)
We normally only set the make flags, e.g.
TI_GFX_MAKE_OPTS = $(LINUX_MAKE_FLAGS) \
...
and the build commands become
$(MAKE) -C $(@D) $(TI_GFX_MAKE_OPTS) all
> +
> +define TI_GFX_BUILD_CMDS
> + ( $(TI_GFX_MAKE_CMD) clean && \
Is the clean really needed? We normally don't.
Also don't connect commands with && unless they really need to be
executed in the same sub-shell.
> + $(TI_GFX_MAKE_CMD) all )
> +endef
> +
> +define TI_GFX_INSTALL_STAGING_CMDS
> +
If you have nothing to install than you shouldn't install to staging :-)
> +endef
> +
> +define TI_GFX_INSTALL_TARGET_CMDS
> + ( $(TI_GFX_MAKE_CMD) install ) || \
Don't you need to pass TARGET_DIR anywhere?
> + echo "Your kernel configuration must include FB_DA8XX"
I think that should be checked by observing the kernel's .config in the
CONFIGURE_CMDS.
Ideally, it would be enabled automatically but I don't think we can do
that at the moment.
> +endef
> +
> +define TI_GFX_CLEAN_CMDS
> + ( $(TI_GFX_MAKE_CMD) clean )
> +endef
Clean is going to be deprecated so you don't need to add it. But I
guess you're using it a lot now...
> +
> +$(eval $(generic-package))
>
--
Arnout Vandecappelle arnout at mind be
Senior Embedded Software Architect +32-16-286500
Essensium/Mind http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F
More information about the buildroot
mailing list