avsm changed the topic of #mirage to: mirage 2 released! party on!
smondet has quit [Ping timeout: 260 seconds]
apache2 has joined #mirage
ahf has quit [Ping timeout: 268 seconds]
apache2_ has quit [Ping timeout: 268 seconds]
ahf has joined #mirage
brson has quit [Quit: leaving]
mcclurmc has joined #mirage
jermar has quit [Remote host closed the connection]
rgrinberg has quit [Ping timeout: 260 seconds]
noddy has quit [Ping timeout: 260 seconds]
mcclurmc has quit [Quit: Leaving]
copy` has quit [Quit: Connection closed for inactivity]
andreas23 has quit [Quit: Leaving.]
andreas23 has joined #mirage
jermar has joined #mirage
mort___ has joined #mirage
mort___1 has joined #mirage
mort___ has quit [Ping timeout: 258 seconds]
mort___ has joined #mirage
mort___1 has quit [Ping timeout: 258 seconds]
yomimono has joined #mirage
smkz has quit [Quit: x]
mort___ has quit [Quit: Leaving.]
mort___ has joined #mirage
mort___ has quit [Quit: Leaving.]
yomimono has quit [Ping timeout: 258 seconds]
mort___ has joined #mirage
mort___ has quit [Quit: Leaving.]
mort___ has joined #mirage
mort___ has quit [Client Quit]
mort___ has joined #mirage
mort___ has quit [Client Quit]
mort___ has joined #mirage
mort___ has quit [Client Quit]
mort___ has joined #mirage
mort___ has quit [Client Quit]
jermar has quit [Remote host closed the connection]
rgrinberg has joined #mirage
mort___ has joined #mirage
<demonimin>
zut, ocaml-fat is possibly broken
agarwal1975 has joined #mirage
<hannes>
demonimin: on which platform/scenario?
<hannes>
demonimin: I remember someone said that using it inside of xen didn't work reliable (the writing bits)... would be great to have it tested and fixed ;)
<hannes>
mato: maybe the easy way out would be to replace zarith with sth more sane? I've been looking into PolarSSL bignums which seems to be fine for many things (most operations in nocrypto, apart from DH group generation which nobody really should do in the first place)
<mato>
hannes: Yeah, you mentioned that... entirely up to you/the rest of the nocrypto folks.
<mato>
hannes: Would the PolarSSL code replace both zarith and gmp or just the former?
<hannes>
mato: it is actually up to David (who is doing nocrypto). either scenario is possible. but for the time being, we will need to life with zarith..
<hannes>
mato: the changes for zarith-freestanding include a make -> gmake in mirage-build.sh -- clearly this will break on any other platform than BSD...
<hannes>
but I guess an uname -s == FreeBSD can be arranged there..
<mato>
hannes: perhaps $(MAKE) is set? or do the uname -s thing....
<mato>
hannes: anyhow, are you ok with merging the solo5 and ocaml-freestanding PRs?
<hannes>
mato: I have not tested them right now, they look fine to me, though
<mato>
hannes: well they work on 11.0-RELEASE and the solo5 changes also check out on 10.3-RELEASE (don't have opam on that box), so it looks safe to me...
<mato>
hannes: anyhow, no hurry, can merge on monday...
<hannes>
mato: if they work for you, please just merge. I'll test them at some point and complain (or PR) if they don't work for me.
<mato>
hannes: hmm, in the zarith patch, do we not need the cclib= ?
<hannes>
mato: isn't a gmp-freestanding added automagically?
<hannes>
mato: the LD stuff... I think when using solo5 it already uses $LD for linking, thus to get it on FreeBSD you can setenv/export LD /usr/local/bin/ld and be fine (obviously the pkg-config thingy looks better, since no user-interaction is needed)
<hannes>
it == mirage
<mato>
hannes: yeah, it's added automatically in the META hack. as long as the build process itself doesn't need cclib= then we're fine
<mato>
hannes: for LD, that was the idea (no user interaction needed)
<hannes>
mato: travis is happy without the cclib :)
<mato>
ok then :)
<hannes>
and I have actually tested those patches on my FreeBSD when I compiled some solo5 unikernels :)
<mato>
ok. i'll wait for you to test my newest changes, will merge them on monday.
<hannes>
ok
apache2_ has joined #mirage
apache2 has quit [Ping timeout: 260 seconds]
mort___ has quit [Quit: Leaving.]
mort___ has joined #mirage
copy` has joined #mirage
yomimono has joined #mirage
<mort___>
mato: re pinning — isn't it the difference between explicitly specifying `-k git` or not?
<mort___>
(whether it takes files in working dir or the current HEAD)