The loops waiting for the card to leave its busy state started their yield
backoff at 100 ms. A card is typically busy for a few milliseconds after a
write, so the backoff never fired and every write was followed by hundreds
of back-to-back CMD13 commands. Occupying the host controller like this
slows down unrelated work on both cores, not just the calling task.
Delay between polls instead, starting at CONFIG_SD_READY_POLL_PERIOD_START_US
(100 us) and doubling. Both the delay and the configured start period are
capped at one FreeRTOS tick period, where vTaskDelay() already yields and one
command per tick is not a storm. A typical wait now costs a handful of
commands instead of hundreds.
Applies to sdmmc_wait_for_idle(), sdmmc_init_sd_wait_data_ready() and
read_tuning_block().
Closes https://github.com/espressif/esp-idf/issues/19034
- don't do swapping if result is thrown away
- use correct return code if csd_ver is unknown
- use correct release function
When `heap_caps_malloc`/`heap_caps_calloc` are used for allocation,
`heap_caps_free` should be used to release allocated block.
This has the potential of speeding up SD card access significantly.
Reusing the DMA aligned buffer gives the additional option of having to
allocate the transaction buffer only once instead of for every
transaction, while still keeping the improved transaction time.
To give users the maximum amount of control, the Kconfig option for the
transaction buffer size applies only to the temporary buffer, which is
only allocated if the DMA aligned buffer has not been pre-allocated.
Closes https://github.com/espressif/esp-idf/pull/17642
Co-authored-by: Adam Múdry <adam.mudry@espressif.com>
sdmmc: UHS-I support for SDR50 (100MHz, 50MB/s), DDR50 (50MHz, 50MB/s)
Closes IDF-10294, IDF-11447, and IDF-8886
See merge request espressif/esp-idf!34441
Previous commit has enabled buffers in PSRAM for ESP32-P4. But this
also caused a regression for ESP32-S3, where PSRAM is not DMA capable.
This commit re-introduces the check for esp_ptr_external_ram in case
SOC_SDMMC_PSRAM_DMA_CAPABLE is not set.
According to the application note in SD Card Physical Specification:
> The host shall set ACMD41 timeout more than 1 second to abort repeat
of issuing ACMD41 when the card does not indicate ready. The timeout
count starts from the first ACMD41 which is set voltage window
in the argument.
Previously, the timeout was exactly one second, and this caused
certain larger-capacity cards to "time out", because they couldn't
finish initialization process in time.
ACMD41 argument is different between SD mode and SPI mode.
In SPI mode, the only non-zero bit may be the HCS bit. Unlike the SD
mode, the bits reflecting the host's OCR should be zero.
Previously, we used to set these bits the same way as for the SD mode.
This has caused certain cards to fail initializing, apparently their
controllers have checked the ACMD41 argument more strictly and refused
to finish initialization, resulting in an error such as
sdmmc_common: sdmmc_init_ocr: send_op_cond (1) returned 0x107
(Note that this error may have other causes than the one fixed in
this commit. For example, if the card doesn't have a sufficient and
stable power supply, it may also fail to complete the internal
initialization process, and will never clear the busy flag in R1
response.)
Closes https://github.com/espressif/esp-idf/issues/6686
Closes https://github.com/espressif/esp-idf/issues/10542
Adam Múdry
Andrey Starodubtsev
morris
sonika.rathi
jofrevandAdam Múdry
Alexey Lapshin
Armando
Ivan Grokhotkov
Martin Vychodil
Steffen Beyer
Michael (XIAO Xufeng)
gaoxu