<avsm>
noddy: more code review on the website than the tls eh ;-)
djs551 has quit [Client Quit]
djs55 has joined #mirage
djs55 has quit [Client Quit]
djs55 has joined #mirage
<avsm>
dsheets: lucid ppa seems unreliable
<avsm>
dsheets: maybe the rest are easier
<dsheets>
avsm, unreliable??
<avsm>
dsheets: transient debhelper dependenncy failures, but not always
<avsm>
some of their build workers are broken i think
<dsheets>
weird
<avsm>
yeah
<dsheets>
well, i made an image with a manual build anyway but it would be nice to have a ppa oh well
<avsm>
dsheets: will look later, best not block on that so manual build is useful
<avsm>
dsheets: rest of xen stuff workign?
<dsheets>
yeah, xen stuff is fine now... just paying off the posix syscall binding debt
<avsm>
heh
<dsheets>
i'm not really sure why the values of macros are in the lem spec :-(
<avsm>
values? shouldnt be there unless posix specifies them, surely
<dsheets>
yeah... linux does some stupid stuff with O_RDONLY = 0 and so forth
<dsheets>
but it's not clear that these issues shouldn't live at the syscall abstraction
<avsm>
aye, abi vs posix there
<avsm>
ok, the conduit resolver now just calculates a connection strategy
<avsm>
and hands that off to the conduit_backend which executes it
<avsm>
type endp = [
<avsm>
| `TCP of Ipaddr.t * int (** ipaddr and dst port *)
<avsm>
| `Vchan of string list (** xenstore path *)
<avsm>
| `TLS of endp (** wrap in a TLS channel *)
<avsm>
| `Unix_domain_socket of string (** unix file path *)
<avsm>
| `Unknown of string (** failed resolution *)
<avsm>
] with sexp
<avsm>
and you can hand in a rewrite trie to the resolver to make it not default to TCP
thomasga has quit [Quit: Leaving.]
dsheets has quit [Read error: Connection reset by peer]
dsheets has joined #mirage
AltGr has left #mirage [#mirage]
AltGr has joined #mirage
thomasga has joined #mirage
<dsheets>
thomasga, AltGr, i have some depopts on a package but i'd like to make one depopt "primary" and require the others as a dependency if the primary is present. Is this possible to express?
<AltGr>
in short: no :/
<AltGr>
well it used to be possible to _express_ but it wouldn't work
<dsheets>
i want logical implication in my depopt expressions
<dsheets>
but i guess we don't even have constraints any more?
<AltGr>
we won't but actually that patch has been postponed just today
<AltGr>
because adding the conflicts may make inconsistent existing installations
<AltGr>
and it's not handled very well by opam 1.1
<dsheets>
it's not impossible, though? it seems like opam could solve for the depopts and then check which clauses are satisfied and add any implied dependencies to the required installation target?
<AltGr>
in the new encoding, that would be a formula like "a && !b" in the conflicts field -- but that's not allowed for solvability reasons
<AltGr>
it's probably possible to encode that with a couple virtual packages, but I'm sure you don't want to
<dsheets>
i'd rather have one package and a sort of fixpoint in the solver driver
<AltGr>
with the new "features" field, you'd be able to use any formula to advertise availability of your feature
<AltGr>
then pass the appropriate flag to your `./configure`
<AltGr>
Nooooooooooooooooooooooooooooooo, no fixpoint !
<avsm>
right at the end, that's a bit of a harsh slot
<yomimono>
luckily it's not terribly inconvenient for me, but I do wonder how well-attended it'll be
<avsm>
I think that should be main track, not a bof
<avsm>
will mail lars
<yomimono>
I think it is main-track and there's just a BoF scheduled alongside it?
* yomimono
may be reading the schedule improperly
<avsm>
oh yes, i think you're right, if the colors are anything to go by
<avsm>
i would move one of the popular talks to the end though, and keep panel in middle
<avsm>
xen on ARM status for example
<yomimono>
say, do you know whether anyone is going to OSBridge in Portland next week?
<avsm>
adam wick might be, from galois
<yomimono>
cool. I got talked into going at the last minute
<avsm>
looks like a fun schedule — lots of privacy talks I notice
<avsm>
speaking there?
<yomimono>
nope! Some friends are, though
<avsm>
Ah yes, that's where I heard about it — O'Reilly has a stand and asked if any authors are showing up
* avsm
is losing track of all these open source conferences now!
<yomimono>
I kept getting confused between OSBridge and OSCon - I think OSCon is the one with the really big O'Reilly presence (maybe they are the major sponsor, in fact?). Same city but OSCon is a month later
<avsm>
Yeah, OSCon is very commercial (and expensive)
<avsm>
OSBridge seems to be more about the cultural end of things
dsheets has quit [Ping timeout: 264 seconds]
djs55 has quit [Quit: Leaving.]
rgrinberg has quit [Quit: Leaving.]
rgrinberg has joined #mirage
thomasga has quit [Ping timeout: 272 seconds]
rgrinberg has quit [Quit: Leaving.]
thomasga has joined #mirage
rgrinberg has joined #mirage
avsm has quit [Quit: Leaving.]
avsm has joined #mirage
avsm has quit [Quit: Leaving.]
thomasga has quit [Quit: Leaving.]
mcclurmc has quit [Read error: Connection reset by peer]
thomasga has joined #mirage
mcclurmc has joined #mirage
yomimono has quit [Ping timeout: 272 seconds]
thomasga has quit [Quit: Leaving.]
djs55 has joined #mirage
avsm has joined #mirage
tlockney is now known as tlockney_away
tlockney_away is now known as tlockney
tlockney is now known as tlockney_away
tlockney_away is now known as tlockney
avsm has quit [Quit: Leaving.]
avsm has joined #mirage
thomasga has joined #mirage
thomasga has quit [Client Quit]
rgrinberg has quit [Quit: Leaving.]
jonludlam has quit [Ping timeout: 245 seconds]
thomasga has joined #mirage
jonludlam has joined #mirage
djs55 has quit [Quit: Leaving.]
thomasga has quit [Quit: Leaving.]
dsheets has joined #mirage
osmano807 has joined #mirage
thomasga has joined #mirage
avsm has quit [Quit: Leaving.]
<osmano807>
Is Mirage done strictly to run on a hypervisor or are there plans to run directly on the hardware?