<GitHub75>
[jruby] trejkaz opened issue #4612: REGRESSION - File.realpath fails on classpath paths https://git.io/v9hL0
Talltree has quit [Remote host closed the connection]
Talltree has joined #jruby
Talltree has quit [Ping timeout: 240 seconds]
Talltree has joined #jruby
Specialist has joined #jruby
_whitelogger has joined #jruby
Specialist has quit [Ping timeout: 240 seconds]
Specialist has joined #jruby
rsim has joined #jruby
rsim has quit [Client Quit]
rsim has joined #jruby
rsim has quit [Client Quit]
vtunka has joined #jruby
drbobbeaty has joined #jruby
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
bbrowning_away is now known as bbrowning
Specialist has quit [Ping timeout: 245 seconds]
Specialist has joined #jruby
Specialist has quit [Ping timeout: 240 seconds]
drbobbeaty has joined #jruby
drbobbeaty has quit [Ping timeout: 240 seconds]
Aethenelle has joined #jruby
vtunka has quit [Quit: Leaving]
olle has joined #jruby
enebo has joined #jruby
* olle
waves to enebo
* enebo
waves back
<olle>
(9.1.9.0! Huzzah, etc!(
<olle>
))
<enebo>
olle: yeah good to have it out
olle has quit [Quit: olle]
kegster has joined #jruby
etehtsea has joined #jruby
hobodave has joined #jruby
hobodave_ has joined #jruby
hobodave has quit [Ping timeout: 255 seconds]
<nirvdrum>
Aethenelle: Do you have any thoughts on "sparcv9" vs "sparc64"? It looks like FFI gets the CPU name from RbConfig::CONFIG['host_cpu'], which returns "sparc64" in MRI. But just about everything else uses "sparcv9".
<nirvdrum>
I can normalize to "sparcv9" in lib/ffi/platform.rb, but I'm not sure if that's the correct thing to do or not.
lanceball is now known as lance|afk
ruurd has joined #jruby
ruurd is now known as Guest58547
etehtsea has quit [Ping timeout: 240 seconds]
<Aethenelle>
yeah... that's an odd one... is there something else we can take a look at?
<Aethenelle>
also, you happen to know where host_cpu comes from?
<Aethenelle>
what does yocto use? debian? demu?
<nirvdrum>
I'll look at the MRI source.
<nirvdrum>
FFI reports "sparc" as the ARCH on Solaris.
<nirvdrum>
The immediate problem is the JNR projects use sparcv9 and key off FFI::Platform::ARCH, so package names don't match.
hobodave has joined #jruby
hobodave_ has quit [Ping timeout: 268 seconds]
subbu is now known as subbu|lunch
camlow325 has joined #jruby
lance|afk is now known as lanceball
<nirvdrum>
Aethenelle: It looks like that comes from the configure script. So, something in the build toolchain.
<nirvdrum>
My Solaris build issue has to do with Solaris make being picked up rather than gmake.
<Aethenelle>
yeah... there's a bug in the solaris stuff... a couple PRs for it but neither fully solve it... maybe something using RbConfig::CONFIG['cflags'] instead?
<nirvdrum>
I think some of it is just the Rakefile being a bit off. "rake compile" seems fine. But "rake install" seems to ignore the GMAKE setting.
<nirvdrum>
enebo: Here's a fun one for you. The jnr-constants lookup strategy can take the arch into account, but it's ordered as OS.ARCH. JNR-FFI orders things as ARCH.OS.