<urja[m]>
Thanks (and pretty much as I figured, was just a "did I miss some place" after trying to look for just them..)
<esden[m]>
They will be available in the 1bitsquared.com and 1bitsquared.de stores in not too distant future. The products don't exist in the store yet.
<urja[m]>
Oh a random datapoint for your "shaving 0.5s from showing the help" conversation...
<urja[m]>
I have two ASUS C201 chromebooks, running the same binaries (afaik), and "time glasgow" (repeatedly done, so not a cold one) takes about 4 seconds on one, and about 7.5 seconds on another one .... and i cant think of any other meaningful difference than that the 7.5 second one is running from microSD (vs eMMC for the faster one)
redstarcomrade has quit [Read error: Connection reset by peer]
<purdeaandrei[m]>
It's not possible to control it from logic but it should be possible to control it during bitstream generation. And it would be useful have the pull-ups be enabled continuously for the selftest.
<purdeaandrei[m]>
I'm just thinking all pull-ups for all 16 data and 16 direction control pins enabled, could be useful for testing shorts
<whitequark[m]>
this has always been possible since amaranth's 0.1 release, what are you talking about?
<whitequark[m]>
that said, this requires modifying pin attributes and so adding a different resource, which will clash heavily with the target logic
bvernoux has joined #glasgow
<whitequark[m]>
purdeaandrei (@_discord_693328004947902465:catircservices.org) I think we should discuss the approach for the selftest before continuing; I feel like it doesn't fit very well into an applet any more
<whitequark[m]>
I think it might deserve its own target and a toplevel command, or something like that
redstarcomrade has joined #glasgow
redstarcomrade has quit [Changing host]
redstarcomrade has joined #glasgow
redstarcomrade has quit [Read error: Connection reset by peer]
bvernoux has quit [Read error: Connection reset by peer]
<Wanda[cis]>
hmm, say, what's the state of the memory-25x applet
<whitequark[cis]>
wdym?
<Wanda[cis]>
we're uhh
<Wanda[cis]>
having issues
<Wanda[cis]>
aside of the obvious UX ones (parameters not automatically taken from the ID block)
<Wanda[cis]>
it just hangs on programming?
<whitequark[cis]>
oh, check WP pin
<Wanda[cis]>
... hm?
<Wanda[cis]>
what would be the failure mode of that?
<Wanda[cis]>
pin is apparently high
<whitequark[cis]>
weird. try -vvv
<Wanda[cis]>
.... that fixed the issue
<Wanda[cis]>
I have no idea why.
<Wanda[cis]>
(I mean in the meantime elly also got annoyed with memory-25x and switched to spi-flashrom, but that also was hanging; also she got impatient and switched from 2.5MHz to 5MHz at the same time as trying -vvv)
<Wanda[cis]>
(I'm going to pin this on cosmic radiation)
<whitequark[cis]>
wtf
<whitequark[cis]>
wait
<whitequark[cis]>
HOLD?
<whitequark[cis]>
is that connected
<Wanda[cis]>
it is
<Wanda[cis]>
well
<Wanda[cis]>
it flashed correctly with spi-flashrom at second attempt
<Wanda[cis]>
and the target seems to be booting
<Wanda[cis]>
so... yay I guess?
<whitequark[cis]>
if you had it disconnected that could be random charge on HOLD or sth
<Wanda[cis]>
HOLD was connected the whole time
<Wanda[cis]>
we also ended up disconnecting WP at some point in attempts to diagnose this thing
<Wanda[cis]>
it worked without it in the end
<whitequark[cis]>
bizarre
<whitequark[cis]>
memory-25x is in general incredibly reliable
<Wanda[cis]>
.... I mean the target is a bit wonky