| Message ID | 20191206213936.v6.11.If3f82501255df1ffcf086953cbd1ed9b65432073@changeid |
|---|---|
| State | Accepted |
| Commit | 1653b6a4bf2edf101c2e3bf632f60e857ec8c038 |
| Delegated to: | Bin Meng |
| Headers | show |
| Series | x86: Add initial support for apollolake | expand |
On Sat, Dec 7, 2019 at 12:46 PM Simon Glass <sjg@chromium.org> wrote: > > When device-tree compilation fails it is sometimes tricky to see which > line is broken, since the input file to dtc is a pre-processed version > of the device tree. > > Add a line that points to the file that needs to be checked: > > When the error is in the main .dts file, output is something like this: > > output: 'Error: arch/x86/dts/.chromebook_coral.dtb.pre.tmp:478.46-47 > syntax error > FATAL ERROR: Unable to parse input tree > > but in fact looking at that file shows nothing useful: > > PAD_CFG_NF_IOSSTATE_IOSTERM(GPIO_157, UP_20K, DEEP, NF1, HIZCRX1, DISPUPD) > > Instead we need to look at the preprocessed file, which shows: > > 163 ((1U << 30) | (1 << 10)) ((0xb << 10) | PAD_CFG1_IOSSTATE_HIZCRX1) > > Here it is clear that PAD_CFG1_IOSSTATE_HIZCRX1 is not defined and so is > not being resolved by the preprocessor. > > This commit adds an additional useful message: > > Check arch/x86/dts/.chromebook_coral.dtb.dts.tmp for errors > > Note that if the error is reported in an included file, such as > u-boot.dtsi then the output is the following: > > Error: arch/x86/dts/u-boot.dtsi:137.14-15 syntax error > FATAL ERROR: Unable to parse input tree > > But again, if the error is due to a preprocessor failure, like this: > > filename = CONFIG_IFW_INPUT_FILE; > > then you can't tell what the problem is by looking at the source. All you > see is the original code: > > intel-ifwi { > filename = CONFIG_IFW_INPUT_FILE; > ... > }; > }; > intel-fsp-m { > filename = CONFIG_FSP_FILE_M; > }; > > Everything looks fine. But looking at the output of the preprocessor: > > intel-ifwi { > filename = CONFIG_IFW_INPUT_FILE; > ... > }; > intel-fsp-m { > filename = "fsp_m.bin"; > }; > > This shows that the filename (normally "fitimage.bin") has not been > inserted the preprocess, leading to the realisation that the value should > be CONFIG_IFWI_INPUT_FILE. > > If the above does not make sense, I encourage people to try introducing > errors in the device tree preprocessed values. > > Signed-off-by: Simon Glass <sjg@chromium.org> > Reviewed-by: Bin Meng <bmeng.cn@gmail.com> > --- > > Changes in v6: None > Changes in v5: None > Changes in v4: > - One last desperate attempt to try to explain the purpose of this commit > - Update the message to mention the preprocessed file, not un-preprocessed > > Changes in v3: > - Update example error message to better show the intended purpose > > Changes in v2: None > > scripts/Makefile.lib | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > applied to u-boot-x86/next, thanks!
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index ef116e0e0a..c10cd83a0a 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -300,7 +300,9 @@ cmd_dtc = mkdir -p $(dir ${dtc-tmp}) ; \ $(CPP) $(dtc_cpp_flags) -x assembler-with-cpp -o $(dtc-tmp) $(pre-tmp) ; \ $(DTC) -O dtb -o $@ -b 0 \ -i $(dir $<) $(DTC_FLAGS) \ - -d $(depfile).dtc.tmp $(dtc-tmp) ; \ + -d $(depfile).dtc.tmp $(dtc-tmp) || \ + (echo "Check $(shell pwd)/$(pre-tmp) for errors" && false) \ + ; \ cat $(depfile).pre.tmp $(depfile).dtc.tmp > $(depfile) ; \ sed -i "s:$(pre-tmp):$(<):" $(depfile)
When device-tree compilation fails it is sometimes tricky to see which line is broken, since the input file to dtc is a pre-processed version of the device tree. Add a line that points to the file that needs to be checked: When the error is in the main .dts file, output is something like this: output: 'Error: arch/x86/dts/.chromebook_coral.dtb.pre.tmp:478.46-47 syntax error FATAL ERROR: Unable to parse input tree but in fact looking at that file shows nothing useful: PAD_CFG_NF_IOSSTATE_IOSTERM(GPIO_157, UP_20K, DEEP, NF1, HIZCRX1, DISPUPD) Instead we need to look at the preprocessed file, which shows: 163 ((1U << 30) | (1 << 10)) ((0xb << 10) | PAD_CFG1_IOSSTATE_HIZCRX1) Here it is clear that PAD_CFG1_IOSSTATE_HIZCRX1 is not defined and so is not being resolved by the preprocessor. This commit adds an additional useful message: Check arch/x86/dts/.chromebook_coral.dtb.dts.tmp for errors Note that if the error is reported in an included file, such as u-boot.dtsi then the output is the following: Error: arch/x86/dts/u-boot.dtsi:137.14-15 syntax error FATAL ERROR: Unable to parse input tree But again, if the error is due to a preprocessor failure, like this: filename = CONFIG_IFW_INPUT_FILE; then you can't tell what the problem is by looking at the source. All you see is the original code: intel-ifwi { filename = CONFIG_IFW_INPUT_FILE; ... }; }; intel-fsp-m { filename = CONFIG_FSP_FILE_M; }; Everything looks fine. But looking at the output of the preprocessor: intel-ifwi { filename = CONFIG_IFW_INPUT_FILE; ... }; intel-fsp-m { filename = "fsp_m.bin"; }; This shows that the filename (normally "fitimage.bin") has not been inserted the preprocess, leading to the realisation that the value should be CONFIG_IFWI_INPUT_FILE. If the above does not make sense, I encourage people to try introducing errors in the device tree preprocessed values. Signed-off-by: Simon Glass <sjg@chromium.org> Reviewed-by: Bin Meng <bmeng.cn@gmail.com> --- Changes in v6: None Changes in v5: None Changes in v4: - One last desperate attempt to try to explain the purpose of this commit - Update the message to mention the preprocessed file, not un-preprocessed Changes in v3: - Update example error message to better show the intended purpose Changes in v2: None scripts/Makefile.lib | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)