FATFS memory leak when calling fopen
Posted: Mon Nov 30, 2020 11:53 am
Hi everyone, I have implemented a FATFS for a partition on my flash memory and have been observing a strange behavior after awhile where I can no longer create a file using the command fopen(filepath, "wb"). The call to fopen returns null and the error code is 2.
Calling perror("error") returns: Error: No such file or directory. Then when I tried to get the current memory from f_getfree, I realised that my free sectors have been reduced down to 0. In my code typically i call fopen with the mode set to wb and each time before I write to the file, I will call the unlink method.
It is strange because I cannot seem to isolate the scenario where this will happen however it will trigger maybe after several days or weeks of developing on the ESP32. Additionally when I reflash my ESP32 my original assumptions were that it would clear/format the FATfs partition but it seems that given this is happening, it seems my assumption is incorrect.
During my testing when I run my code, the data that I am writing to the FATFS is typically always the same length and I have not touched this part of the code at all for quite awhile and it will trigger this issue without much warning after flashing/working on the code for a while.
Has anyone encountered this before or is able to give some input to identify where the issue could be coming from?
Note: Flash encryption is not enabled and I am writing to a partition on my external flash
An extract of the code I use to add the file:
How I mount my partition
Calling perror("error") returns: Error: No such file or directory. Then when I tried to get the current memory from f_getfree, I realised that my free sectors have been reduced down to 0. In my code typically i call fopen with the mode set to wb and each time before I write to the file, I will call the unlink method.
It is strange because I cannot seem to isolate the scenario where this will happen however it will trigger maybe after several days or weeks of developing on the ESP32. Additionally when I reflash my ESP32 my original assumptions were that it would clear/format the FATfs partition but it seems that given this is happening, it seems my assumption is incorrect.
During my testing when I run my code, the data that I am writing to the FATFS is typically always the same length and I have not touched this part of the code at all for quite awhile and it will trigger this issue without much warning after flashing/working on the code for a while.
Has anyone encountered this before or is able to give some input to identify where the issue could be coming from?
Note: Flash encryption is not enabled and I am writing to a partition on my external flash
An extract of the code I use to add the file:
Code: Select all
// Remove file before adding
unlink(filepath);
// Open file to write
FILE *f = fopen(filepath, "wb");
if (f == NULL)
{
ESP_LOGE(TAG, "Failed to open file for writing. Error code: %d", errno);
perror("Error");
return ESP_FAIL;
}
else
{
// Write data to the file here
if (fprintf(f, "%s", inBuffer) == EOF)
{
ESP_LOGE(TAG, "File Write Error, Memory Might be full");
fclose(f);
return ESP_FAIL;
}
}
// Remember to close the file
fclose(f);How I mount my partition
Code: Select all
const esp_vfs_fat_mount_config_t mount_config = {
.max_files = 4,
.format_if_mount_failed = true,
.allocation_unit_size = CONFIG_WL_SECTOR_SIZE};
esp_err_t err = esp_vfs_fat_spiflash_mount(ROOT, NAME, &mount_config, &s_wl_handle_FileStorage);
if (err != ESP_OK)
{
ESP_LOGE(TAG, "Failed to mount FATFS (%s)", esp_err_to_name(err));
return ESP_FAIL;
}