Perl 6 - the future is here, just unevenly distributed

IRC log for #moarvm, 2016-05-30

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:45 diakopter timotimo: thanks for the link
00:45 timotimo YW
01:48 ilbot3 joined #moarvm
01:48 Topic for #moarvm is now https://github.com/moarvm/moarvm | IRC logs at  http://irclog.perlgeek.de/moarvm/today
05:35 domidumont joined #moarvm
05:40 domidumont joined #moarvm
06:15 domidumont joined #moarvm
06:52 domidumont joined #moarvm
07:18 nwc10 good *, #moarvm
07:40 brrt joined #moarvm
07:50 brrt good * #moarvm
08:08 jnthn moarning o/
08:30 brrt jnthn, how worthwhile do you think it will be to allow the interpreter to get the JIT entry label?
08:31 brrt eh, well, to be exact
08:31 jnthn brrt: Can it not today? :)
08:31 brrt very well
08:31 brrt :-)
08:31 brrt busy day today?
08:31 jnthn heh, that wasn't the parse I expected :P
08:32 jnthn "Can the interpreter not already discover that today?" :)
08:32 jnthn ->jit_entry_label on frame or so
08:33 jnthn I should probably have waited for the more exact :)
08:33 brrt oh, that way
08:33 brrt yeah, it can
08:34 brrt what i mean is, it would be nice if the JIT didn't have to do the bookkeeping, but the interpreter could call a function (downstack from the JIT), which would give you that offset, just from the return address of the function
08:34 brrt i.e. MVM_jit_get_reentry_label(MVMThreadContext *tc, void *jit_code_base, size_t jit_code_size)
08:35 brrt the downside is that we would have to compile with -fno-omit-frame-pointer
08:35 brrt and the equivalent in MSVC
08:35 brrt i can write that function is assembly easily enough
08:35 brrt getting it to link correctly on multiple platforms, that is one minor annoyance
08:36 jnthn How much of a saving do you think it would be?
08:36 brrt i don't know. we currently insert 'store reentry labels' every basic block, every invokish, every throwish thing
08:37 brrt even if they may not throw or invoke at all
08:37 jnthn I'm assuming -fno-omit-frame-pointer costs a little bit too, though maybe not so much on x64 as on x86...
08:37 brrt that is quite a bit of memory traffic
08:37 jnthn Since the former is less register starved
08:37 brrt agreed
08:37 brrt only way to know is to measure i think....
08:37 jnthn And since we don't JIT on x86 we could only set the flag on x64
08:38 brrt it might hurt the interpreter a bit, but i don't know that
08:38 brrt how much
08:40 jnthn Yeah. It's hard to guess these things, but while it may be quite a bit of memory traffic, it's also (a) likely to be a cache hit most of the time, and (b) doens't have a data dep
08:40 brrt also gtrue
08:41 brrt same is true for stack walking of course
08:58 zakharyas joined #moarvm
09:09 brrt i wonder if there is a practical way to make cross platform c-linkable assembly
09:10 brrt joined #moarvm
11:49 zakharyas joined #moarvm
12:52 _longines joined #moarvm
12:59 cognominal joined #moarvm
15:09 zakharyas joined #moarvm
17:39 domidumont joined #moarvm
18:23 FROGGS joined #moarvm
20:05 zakharyas joined #moarvm
22:03 LLamaRider joined #moarvm

| Channels | #moarvm index | Today | | Search | Google Search | Plain-Text | summary