Attention is currently required from: Nico Huber, Edward O'Callaghan, Samir Ibradžić. Simon Buhrow has posted comments on this change. ( https://review.coreboot.org/c/flashrom/+/40477 )
Change subject: ft2232_spi.c: Implement spi_send_multicommand() ......................................................................
Patch Set 32:
(1 comment)
Patchset:
PS32:
I wondered what device was used for testing so far and […]
Thanks for your effort! hm, I´m not sure what´s wrong. Saving a few seconds is not what I was looking for...
Some information:
* the 1,36MHz on the screenshot is wrong (sorry for missleading!!) It´s the distance between the cursors which may be placed arbitrarily * I´m using default divisor and on my osci screenshots taken last week that results in something between 25-30 MHz (did not measure, but estimate from screenshots) * lsusb says bcdDevice 7.00 as well * I run: `make CONFIG_ENABLE_LIBUSB1_PROGRAMMERS=no`. So I´m not using libusb 1.0 maybe this leads to some difference? * I´m doing tests on my side with an N25Q128..3E chip (has quite the same parameters as W25Q128FV") * I´m running flashrom on a virutal machine (don´t think this makes any difference)
* at SPI-Bus level the multicmd+chunk4k looks much better and faster (i.e. less bursts) so I have no idea why this should not lead to an overall speed increase... (regardless of the setup)
NB: yepp, I noticed that with the erase block size, as well and have it on my todo list. I was thinking about something like comandline parameter to select block size or an extra parameter at flashchips.c to mark the preferred block size. May be we should open a new patch for that to discuss ideas... (it was just because of lack of support for this patch for month why I did not step further)
-
Simon Buhrow (Code Review)