Can the ESP32 boot off of any SPI interface other than SPI (e.g. HSPI, VSPI)? The reason that I ask is that if not, and maybe I'm missing something here, it doesn't appear that there's a pin config that support SPI boot, SD, and JTAG access.
Using the function 5 values for SD_*, MTDI is shared with SD_DATA2 on the alternate function and MTMS is shared with SD_CLK.
So is the aforementioned config possible (SPI boot, SD, and JTAG simultaneously)?
SPI boot/pin mux config question
-
- Posts: 9709
- Joined: Thu Nov 26, 2015 4:08 am
Re: SPI boot/pin mux config question
It can boot from HSPI as well, if I recall correctly. Plus, you have the option of remapping pins after boot. I think you can even remap some _before_ boot, by using the efuses, but that's not documented yet.
Re: SPI boot/pin mux config question
So there are more strapping pins? Are we also going to get some details/quantifications on the effect of gpio matrix routing on high speed signals?ESP_Sprite wrote:It can boot from HSPI as well, if I recall correctly. Plus, you have the option of remapping pins after boot. I think you can even remap some _before_ boot, by using the efuses, but that's not documented yet.
Re: SPI boot/pin mux config question
I'm hoping for VSPI here -- HSPI gets in the way of JTAG. VSPI doesn't interfere with anything I care about. According to Greg, VSPI should work.ESP_Sprite wrote:It can boot from HSPI as well, if I recall correctly. Plus, you have the option of remapping pins after boot. I think you can even remap some _before_ boot, by using the efuses, but that's not documented yet.
-
- Posts: 9709
- Joined: Thu Nov 26, 2015 4:08 am
Re: SPI boot/pin mux config question
It's configurable by efuses. The effect of routing high-speed signals over the GPIO matrix is a latency of 2 clock cycles, iirc. It's somewhere in the techdocs, if memory serves.
Re: SPI boot/pin mux config question
Ah OK so program the efuse over serial first. I'm not sure if the docs specify the clock cycles, so thanks.ESP_Sprite wrote:It's configurable by efuses. The effect of routing high-speed signals over the GPIO matrix is a latency of 2 clock cycles, iirc. It's somewhere in the techdocs, if memory serves.
Who is online
Users browsing this forum: No registered users and 191 guests