build fails if the input exceeds 8000 symbols

[Amomum]

That's a bug similar in manifestation and conditions to this https://answers.launchpad.net/gcc-arm-embedded/+question/236792.

I am using arm gcc 4.9 2015q1 on windows 7. I'm using eclipse cdt with GNU Make (Sourcery CodeBench Lite 2013.05-23) 3.81.
I'm trying to compile ARM CMSIS DSP Lib which has a lot of files:

arm-none-eabi-g++ -o "Flutilizer" ./src/Startup/system_stm32f4xx.o ./src/STM_Discovery_Audio/stm32f4_discovery_audio_codec.o ./src/STM32F4xx_StdPeriph_Driver/src/misc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_adc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_can.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_crc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_cryp.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_cryp_aes.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_cryp_des.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_cryp_tdes.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_dac.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_dbgmcu.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_dcmi.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_dma.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_exti.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_flash.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_fsmc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_gpio.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_hash.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_hash_md5.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_hash_sha1.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_i2c.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_iwdg.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_pwr.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_rcc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_rng.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_rtc.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_sdio.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_spi.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_syscfg.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_tim.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_usart.o ./src/STM32F4xx_StdPeriph_Driver/src/stm32f4xx_wwdg.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_bitreversal.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_init_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_init_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_init_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix2_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_init_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_init_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_init_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix4_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_cfft_radix8_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_init_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_init_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_init_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_dct4_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_fast_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_fast_init_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_init_f32.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_init_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_init_q31.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_q15.o ./src/CMSIS/DSP_Lib/TransformFunctions/arm_rfft_q31.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_copy_f32.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_copy_q15.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_copy_q31.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_copy_q7.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_fill_f32.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_fill_q15.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_fill_q31.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_fill_q7.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_float_to_q15.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_float_to_q31.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_float_to_q7.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q15_to_float.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q15_to_q31.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q15_to_q7.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q31_to_float.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q31_to_q15.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q31_to_q7.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q7_to_float.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q7_to_q15.o ./src/CMSIS/DSP_Lib/SupportFunctions/arm_q7_to_q31.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_cos_f32.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_cos_q15.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_cos_q31.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_sin_f32.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_sin_q15.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_sin_q31.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_sqrt_q15.o ./src/CMSIS/DSP_Lib/FastMathFunctions/arm_sqrt_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_conj_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_conj_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_conj_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_dot_prod_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_dot_prod_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_dot_prod_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_squared_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_squared_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mag_squared_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_cmplx_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_cmplx_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_cmplx_q31.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_real_f32.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_real_q15.o ./src/CMSIS/DSP_Lib/ComplexMathFunctions/arm_cmplx_mult_real_q31.o ./src/CMSIS/DSP_Lib/CommonTables/arm_common_tables.o ./src/CMSIS/DSP_Lib/CommonTables/arm_const_structs.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_abs_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_abs_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_abs_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_abs_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_add_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_add_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_add_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_add_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_dot_prod_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_dot_prod_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_dot_prod_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_dot_prod_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_mult_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_mult_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_mult_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_mult_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_negate_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_negate_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_negate_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_negate_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_offset_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_offset_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_offset_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_offset_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_scale_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_scale_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_scale_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_scale_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_shift_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_shift_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_shift_q7.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_sub_f32.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_sub_q15.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_sub_q31.o ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_sub_q7.o ./src/main.o ./src/waverecorder.

And I get the error:
arm-none-eabi-g++: error: ./src/CMSIS/DSP_Lib/BasicMathunctions/arm_sub_q15.o: No such file or directory
the correct path to the file is:
 ./src/CMSIS/DSP_Lib/BasicMathFunctions/arm_sub_q7.o

Note please the lack of "F" in BasicMathunctions in the error message. The letter F is 8281st symbol (not the expected 8192nd); and yet for some reason this letter is ignored.
I have tried to exclude some files from the build but some symbol always get ignored alghout it's not always 8281st, it's always more the 8000nd.

[Tony Liu (mrtoniliu)]

Hi,

Would you please follow the comments in the question you found as the workaround?
It seems I can't reproduce this problem on the Linux machine and I guess it may be some kind of problems only exist in Windows.

Regards,
Tony

[Amomum (amomum)]

Well, that is not a very nice workaround. Right now Eclipse generates makefile for me, I really don't want to edit it by hand.

Even if it is a windows-only bug, it's still a bug, isn't it?

[Terry Guo (terry.guo)]

I am checking on this. Another possible workaround is archiving those arm_sub_q7.o objects into library like libmath.a first and then use it in final link step.

[Terry Guo (terry.guo)]

You can also try solution from http://mcuoneclipse.com/2015/03/29/solving-the-8192-character-command-line-limit-on-windows/. You are going to install some other tools which can support long command line.

[Amomum (amomum)]

Yes, fortunaterly DSP Library is already available as a bunch of .a files for different platforms. Another workaround that I found is to use Eclipse internal builder instead of GNU make.

But it's still a bug.

[Amomum (amomum)]

Oh, so that's a problem with cmd.exe, not with gcc?

[Terry Guo (terry.guo)]

I guess your project is a Makefile based project. If so, it is a problem of your make.exe. The process should be the Eclipse call make.exe to handle Makefile, the make.exe then call arm-none-eabi-gcc.exe to do final link. When make.exe pass arguments to arm-none-eabi-gcc.exe, the make.exe cuts the arguments itself, so the arm-none-eabi-gcc.exe gets corrupt options. I suggest you to backup your current make.exe and use make.exe provided from http://gnuarmeclipse.livius.net/blog/build-tools-windows/.

[Amomum (amomum)]

Okay, thank you very much.
Since it seems not to be a bug in gcc, I suppose, we can close it?

[Terry Guo (terry.guo)]

Closed as Invalid. Feel free to reopen this for any further concerns.