05:59 <hqyang> I use jtag to connect riscv-013. It took 2s to write 4k Bytes of data to sram with target_write_buffer. Is this normal?

05:59 <hqyang> can someone help me

12:29 <Guest19> Hi folks, im struggling with flashing my external nor flash (over qspi) on a stm32h750b-dk board. it seems, the flashing via openocd can be performed with success but flashing again lead to timeout errors. i know thats very unspecific, but after run in application (qspi init routines are called) i can flash via openocd again. one important thing:

12:29 <Guest19> once this error state occurs, i only get out of that state using the stm qube programmer or do a power cycle.

12:44 <vfazio> any chance someone with an RPi can take a stab at: https://review.openocd.org/c/openocd/+/7258 ?

12:55 <karlp> vfazio: what do you mean?

12:55 <vfazio> test it? sign off on it? review it?

12:55 <karlp> it's already merged...

12:56 <karlp> october last year

12:56 <vfazio> fuck

12:56 <vfazio> wrong change request

12:56 <karlp> vfazio: what do you mean?np :)

12:56 <vfazio> i haven't had coffee yet, just a sec

12:56 <karlp> no problem :)

12:57 <vfazio> karlp, https://review.openocd.org/c/openocd/+/7732

12:59 <vfazio> it's been working for us, but would like formal review and to see it merged at some point, maybe the next tagged version? we have another patch to submit too, but we're still testing it

13:09 <vfazio> karlp, is there a standard way to collect performance data for these types of changes? or are we just looking at how long it takes to program a device?

15:04 <karlp> just how long it takes to program something reasonable sized, both write and read. (write is often limited by flash of the target, reads less so)

15:04 <karlp> and big enough that the setup isn't the dominant factor

15:04 <karlp> given that you know the macros are independent, you may want to consider doing that in a separate patch too, though I personally don't really mind.

15:04 <karlp> (I don't have anything to test this with fo ryou, I'm just a commenter, not a committer ;)

16:07 <vfazio> I had originally split this up but massaging the individual commits to play nice with checkpatch became overly burdensome. I appreciate the input karlp

21:02 <vfazio> borneoa___, i'll try to get the data you requested next week

21:03 <vfazio> i need to look at the asm being generated, i'm a little surprised at our early testing

21:13 <borneoa___> vfazio: thanks