Why is the binary generated by my gcc cross compiler so large?

After getting a gcc toolchain for ARM built the next step was to build a simple "Hello world" application. I put together something simple to set an output and toggle it:

#include "stm32f10x.h"
#include "stm32f10x_rcc.h"
#include "stm32f10x_gpio.h"

const static int DelayCount = 2500000;

void Delay()
{
volatile int delayValue;

delayValue = DelayCount;
while(delayValue)
{
delayValue--;
}
}

int main()
{
// enable gpio c clock, IOPCEN bit
RCC_APB2PeriphClockCmd(RCC_APB2Periph_GPIOC, ENABLE);

// configure PC12 as output, open drain
GPIO_TypeDef* gpio = GPIOC;
GPIO_InitTypeDef gpioInit;
GPIO_StructInit(&gpioInit);
gpioInit.GPIO_Mode = GPIO_Mode_Out_OD;
gpioInit.GPIO_Pin = GPIO_Pin_12;
GPIO_Init(gpio, &gpioInit);

// loop forever
while(1)
{
// output 1 on PC12
GPIO_SetBits(gpio, GPIO_Pin_12);

Delay();

// output 0 on PC12
GPIO_ResetBits(gpio, GPIO_Pin_12);

Delay();
}

return 1;
}

After building this program in Eclipse, this site was helpful among others, it was ~770k. Figuring it was due to debugging symbols I tried running 'strip' on the elf and using 'objcopy -O binary in.elf out.bin'.

Ran readelf -s on the elf file to see what symbols were present and saw tons of functions that I didn't expect to see, things like malloc.

Turns out there were a few issues.

- Eclipse wasn't seeing the assembly files as they had '.s' extension instead of '.S'. Renaming the appropriate assembly file resolved this issue.

- Gcc/linker wasn't being properly configured to remove unused symbols

Options to add for the compiler:

-fdata-sections -ffunction-sections
Causes the compiler to place code and data in separate sections.

Options to add for the linker:

--gc-sections
Causes the linker to remove unused sections.

After these changes the elf went from ~770k down to ~589k and after the conversion to binary file the file was down to under 2k.

Comments

UTM (qemu on OSX) only boots into uefi shell after installing Ubuntu 22.10 (solved!)

I've been meaning to try out running Linux and Windows virtual machines on my x86 MacBook Pro using qemu . I've been using Parallels for several years. Parallels is a nice application and works well but I'm always getting bugged to purchase upgrades. Parallels provides free updates on minor version upgrades only, say 13.0 to 13.1. With so many prompts to purchase upgrades it feels like Parallels could release a bit more minor versions and a bit fewer major versions. Rather than just purchase an upgrade to Parallels I figured I'd try out qemu via the UTM application . UTM wraps qemu with a helpful gui and comes as a packages Mac application. And UTM supports both x86 and M1 systems (as qemu also supports these systems). Creating an Ubuntu VM with UTM didn't go as smoothly as with VirtualBox or Parallels but it was the settings I chose that messed me up. Here are two things I ran into when installing Ubuntu on UTM. #1. Configure the qemu instance to use a 'Display...

Debugging an imprecise bus access fault on a Cortex-M3

This information may apply to other cortex series processors but is written from practical experience with the Cortex-M3. Imprecise bus access faults are ambiguous, as noted by the term "imprecise". Compared to precise bus errors, imprecise errors are much trickier to debug and especially so without a deep understanding of arm processors and assembly language. Imprecise and precise flags are found in the BusFault status register, a byte in the CFSR (Configurable Fault Status Register). BusFault status register bits The definition for imprecise and precise bits is: [2] IMPRECISERR Imprecise data bus error: 0 = no imprecise data bus error 1 = a data bus error has occurred, but the return address in the stack frame is not related to the instruction that caused the error. When the processor sets this bit to 1, it does not write a fault address to the BFAR. This is an asynchronous fault. Therefore, if it is detected when the priority of the current pr...

Graco Swing By Me - Battery to AC wall adapter modification

If you have one of these Graco battery powered swings you are probably familiar with the cost of C batteries! The swing takes four of them and they only last a handful of days. I'm not sure if the newer models support being plugged into the wall but ours didn't. If you are a little familiar with electronics and soldering, here is a rough guide on how you can modify yours to plug in! I wasn't sure how exactly to disassemble the swing side where the batteries were. I was able to open up the clamshell a bit but throughout this mod I was unable to determine how to fully separate the pieces. I suspect that there is some kind of a slip plate on the moving arm portion. The two parts of the plastic are assembled and the moving arm portion with the slip plate is slid onto the shaft. Because of the tension in that slip plate it doesn't want to back away, and because of the mechanicals that portion of the assembly doesn't appear accessible in order to free it. I was...

Chris M.

Search This Blog