ATA-Trim support

[SybexX]

copied the ESP IDF component fatfs to our project and modified it to:

1. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> ATA-Trim is supported -> run through ATA-Trim routine
2. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> ATA-Trim is not supported -> do nothing
3. SD initialize -> read SD parameters -> if FF_USE_TRIM = 0 -> do nothing

Original (official ESP-IDF behaviour):

1. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> no matter whether ATA-Trim is supported or not -> run through ATA-Trim routine
2. SD initialize -> read SD parameters -> if FF_USE_TRIM = 0 -> do nothing

See #2710 for details

[jomjol]

Is this doing the same as: #2779 ???

@SybexX and @caco3:

My question: one or the other pull request or both?

[SybexX]

Yes, only I have now adapted fatfs and sdmmc from the ESP-IDF so that FF_USE_TRIM can remain at 1 and cards that support it can also use it.
I don't know whether ATA Trim has any advantages at all, but I still want to keep it and not deactivate it completely.

---

Process in the ESP-IDF:

1. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> no matter whether ATA-Trim is supported or not -> run through ATA-Trim routine
2. SD initialize -> read SD parameters -> if FF_USE_TRIM = 0 -> do nothing

---

changed to:

1. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> ATA-Trim is supported -> run through ATA-Trim routine
2. SD initialize -> read SD parameters -> if FF_USE_TRIM = 1 -> ATA-Trim is not supported -> do nothing
3. SD initialize -> read SD parameters -> if FF_USE_TRIM = 0 -> do nothing

[caco3]

@jomjol Yes, the other PR was the initial one, but we should go with the one here as it is advanced.
@SybexX Can you please in the readme and in the code mark your changes? This makes it easier when we want to update the file sin the future.

I updated the PR description with your nice explanation.

@friedpa and @Rak3tenmann Can you also test this PR and check it works for you?
Get the build from here: https://github.com/jomjol/AI-on-the-edge-device/actions?query=event%3Apull_request (use the latest [SybexX:ATA_TRIM_SUPPORT](https://github.com/SybexX/AI-on-the-edge-device/tree/ATA_TRIM_SUPPORT) build)

[friedpa]

Good Afternoon Gents,
the AI-on-the-edge-device__update__2781_merge_(9a3f3de) works as tested before.
Happy new year!

[caco3]

Thanks @SybexX for your contribution!

[SybexX]

Since this problem really affects all 5. versions, I have now created a "Issues" in espressif, let's see if something happens there.

espressif/esp-idf#12912

[olegarch]

Linker complains about duplicates:

packages/toolchain-xtensa-esp32/bin/../lib/gcc/xtensa-esp32-elf/12.2.0/../../../../xtensa-esp32-elf/bin/ld.exe: .pio/build/esp32cam/esp-idf/esp-sdmmc/libesp-sdmmc.a(sdmmc_common_mh.o): in function sdmmc_init_ocr': AI-on-the-edge-device\code/components\esp-sdmmc/sdmmc_common_mh.c:25: multiple definition of sdmmc_init_ocr'; .pio/build/esp32cam/esp-idf/sdmmc/libsdmmc.a(sdmmc_common.o):C:\Users\olegk.platformio\packages\framework-espidf\components\sdmmc/sdmmc_common.c:23: first defined here

[SybexX]

Unfortunately, this is due to Visual Studio Code, and linking problems occasionally occur.
As far as I could read on the internet, this problem only affects Windows users^^