| Message ID | 1392964470-24720-1-git-send-email-yamada.m@jp.panasonic.com |
|---|---|
| State | Accepted |
| Delegated to: | Tom Rini |
| Headers | show |
On 20 February 2014 22:34, Masahiro Yamada <yamada.m@jp.panasonic.com> wrote: > Commit 6ab6b2af deleted ./u-boot.dtb because it was a copy > of dts/dt.dtb. > > But Simon suggested to keep u-boot.dtb at the top directory > because it is a build output. > > After discussions, we agreed on revival of ./u-boot.dtb. > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > Cc: Simon Glass <sjg@chromium.org> > Cc: Tom Rini <trini@ti.com> Acked-by: Simon Glass <sjg@chromium.org> > --- > > Hello Simon, > > In our discussions, I memtioned to post version 4 > of "dts: re-write dts/Makefile more simply with Kbuild". > > But, before that, version 3 has been applied. > > Anyway, I promised to get back ./u-boot.dtb, > so I'm posting this patch. > > To avoid a conflict with other patches, > this patch is rebased on > "Follow-up for Kbuild series: more misc targets and short logs" series. > (And using Kbuild style to display short log) > > BTW, another question hit me when I was writing this patch. > Please let me ask you a little more questions. > > I guess the reason why you want device tree at the top directory > is Sandbox. Is this right? > If so, I am 100% convinced and I'd like to suggest to modify > > ALL-$(CONFIG_OF_SEPARATE) += u-boot.dtb u-boot-dtb.bin > > to > > ALL-$(CONFIG_OF_SEPARATE) += u-boot-dtb.bin > ALL-$(CONFIG_OF_HOSTFILE) += u-boot.dtb > > Is this better? > For CONFIG_OF_SEPARATE, I think there is no opportunity to use u-boot.dtb > because we have u-boot-dtb.bin. Well some build systems may take the u-boot.dtb and package it up with U-Boot in a different way. Also they may look inside this file to find out things about the board they are targeting. For example, you might imagine a flash program which gets flash programming parameters from this file so it knows how to flash the board. I don't consider something in the dts/ directory to be a build product really. I suspect one day we might support multiple .dtb files in U-Boot and select between them based on a setting or a hardware feature, and in that case perhaps everything will be put in a FIT, but we are not there yet. But for now I see the .dtb file as a first-class build product just like the main U-Boot binary. Regards, Simon
Hello Simon, > Well some build systems may take the u-boot.dtb and package it up with > U-Boot in a different way. Also they may look inside this file to find > out things about the board they are targeting. For example, you might > imagine a flash program which gets flash programming parameters from > this file so it knows how to flash the board. > > I don't consider something in the dts/ directory to be a build product > really. I suspect one day we might support multiple .dtb files in > U-Boot and select between them based on a setting or a hardware > feature, and in that case perhaps everything will be put in a FIT, but > we are not there yet. > > But for now I see the .dtb file as a first-class build product just > like the main U-Boot binary. Then, I agreed to include this version to the main line. Thanks for your answering. I confirmed this patch can apply on commit 1674df60 + Follow-up for Kbuild series v2. Best Regards Masahiro Yamada
On Fri, Feb 21, 2014 at 03:34:30PM +0900, Masahiro Yamada wrote: > Commit 6ab6b2af deleted ./u-boot.dtb because it was a copy > of dts/dt.dtb. > > But Simon suggested to keep u-boot.dtb at the top directory > because it is a build output. > > After discussions, we agreed on revival of ./u-boot.dtb. > > Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> > Cc: Simon Glass <sjg@chromium.org> > Cc: Tom Rini <trini@ti.com> > Acked-by: Simon Glass <sjg@chromium.org> Applied to u-boot/master, thanks!
diff --git a/Makefile b/Makefile index b91381c..db263ef 100644 --- a/Makefile +++ b/Makefile @@ -720,7 +720,7 @@ ALL-$(CONFIG_RAMBOOT_PBL) += u-boot.pbl ALL-$(CONFIG_SPL) += spl/u-boot-spl.bin ALL-$(CONFIG_SPL_FRAMEWORK) += u-boot.img ALL-$(CONFIG_TPL) += tpl/u-boot-tpl.bin -ALL-$(CONFIG_OF_SEPARATE) += u-boot-dtb.bin +ALL-$(CONFIG_OF_SEPARATE) += u-boot.dtb u-boot-dtb.bin ifneq ($(CONFIG_SPL_TARGET),) ALL-$(CONFIG_SPL) += $(CONFIG_SPL_TARGET:"%"=%) endif @@ -766,6 +766,12 @@ dtbs dts/dt.dtb: checkdtc u-boot u-boot-dtb.bin: u-boot.bin dts/dt.dtb FORCE $(call if_changed,cat) +quiet_cmd_copy = COPY $@ + cmd_copy = cp $< $@ + +u-boot.dtb: dts/dt.dtb + $(call cmd,copy) + OBJCOPYFLAGS_u-boot.hex := -O ihex OBJCOPYFLAGS_u-boot.srec := -O srec
Commit 6ab6b2af deleted ./u-boot.dtb because it was a copy of dts/dt.dtb. But Simon suggested to keep u-boot.dtb at the top directory because it is a build output. After discussions, we agreed on revival of ./u-boot.dtb. Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com> Cc: Simon Glass <sjg@chromium.org> Cc: Tom Rini <trini@ti.com> --- Hello Simon, In our discussions, I memtioned to post version 4 of "dts: re-write dts/Makefile more simply with Kbuild". But, before that, version 3 has been applied. Anyway, I promised to get back ./u-boot.dtb, so I'm posting this patch. To avoid a conflict with other patches, this patch is rebased on "Follow-up for Kbuild series: more misc targets and short logs" series. (And using Kbuild style to display short log) BTW, another question hit me when I was writing this patch. Please let me ask you a little more questions. I guess the reason why you want device tree at the top directory is Sandbox. Is this right? If so, I am 100% convinced and I'd like to suggest to modify ALL-$(CONFIG_OF_SEPARATE) += u-boot.dtb u-boot-dtb.bin to ALL-$(CONFIG_OF_SEPARATE) += u-boot-dtb.bin ALL-$(CONFIG_OF_HOSTFILE) += u-boot.dtb Is this better? For CONFIG_OF_SEPARATE, I think there is no opportunity to use u-boot.dtb because we have u-boot-dtb.bin. Makefile | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)