Perl 6 - the future is here, just unevenly distributed

IRC log for #rosettacode, 2015-07-22

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

All times shown according to UTC.

Time Nick Message
00:18 eMBee joined #rosettacode
01:11 Birlinn[RCo] joined #rosettacode
01:12 Birlinn[RCo] any racket contributors present?
01:42 fedaykin joined #rosettacode
01:45 sora joined #rosettacode
02:15 mwn3d joined #rosettacode
02:21 eMBee they are busy making a racket elsewhere
03:03 mwn3d1 joined #rosettacode
03:04 mwn3d joined #rosettacode
03:04 mwn3d1 joined #rosettacode
03:04 mwn3d joined #rosettacode
03:08 Rainb joined #rosettacode
04:08 Wright_ joined #rosettacode
06:04 Rainb joined #rosettacode
07:43 sora joined #rosettacode
08:25 sora joined #rosettacode
09:13 sora joined #rosettacode
10:23 Rainb joined #rosettacode
11:45 mwn3d1 joined #rosettacode
12:14 kuzetsa goodness, this ubuntu VM I spun up JUST for android development (nothing else gets built in it... and specifically had to be ubuntu 'cause it's sort of the defacto reference standard for android development and reduces my hassle) has ccache set up for 60 gigabytes...
12:15 kuzetsa only building android for a single target at the moment, and this is the first time I built since setting up the ccache
12:15 kuzetsa already 8 gigabytes of ccache in use
12:16 kuzetsa [14:55:21] <+kuzetsa> compiling android from source is tedious <<< I swear, so far this has taken longer than a stage 1 gentoo build I did back in 2003 on a rather slow single-core system O_O
12:26 pdurbin kuzetsa: why are you compiling android again? what's your ultimate goal?
12:27 kuzetsa pdurbin: there's a limit to the battery-savings and footprint-reduction the android devs were able to do when they came up with the spec after project svelte / project volta
12:28 pdurbin so you want longer battery life
12:28 kuzetsa my goal is to exceed 100 hours standby while still having a gentoo userland running concurrently along side the android userland (on the same kernel)
12:29 kuzetsa the android kernel doesn't flawlessly run certain non-android linux sofware, so I had to make some patches there even if nothing else got touched :)
12:32 kuzetsa an example that stood out as requiring a mandatory kernel patch was getting full kernel-level semaphore support set up in a way which would've been slow to emulate and reduced the performance of ... I think it was python (it was months ago, but whatever it was I'm certain it needed some kernel modification to work properly)
12:34 kuzetsa either way... support for /dev/shm isn't normally a thing in android
12:34 pdurbin gotcha
12:35 kuzetsa https://bugzilla.mozilla.org/show_bug.cgi?id=594878 [quote] "android has decided not to support SysV shmem (OK) or POSIX shmem slash /dev/shm (sigh), and instead created two new shmem implementations in their kernel.  One is /dev/ashmem, which is non-physically-contiguous shmem that can be thrown out at any time when there's memory pressure (that is, just like /dev/shm except the
12:35 kuzetsa throw-out-at-arbitrary-times bit)."
12:35 fedaykin "594878 - Use /dev/ashmem for Shmem on android" http://rldn.net/4zrB
12:35 kuzetsa ^ insane version exist, but porting / switching to that one seems suboptimal and kinda undesirable O_O
15:47 Rainb joined #rosettacode
17:53 Shayanjm joined #rosettacode
18:11 fedaykin joined #rosettacode
18:31 sora joined #rosettacode
19:15 fedaykin joined #rosettacode
21:50 bb010g joined #rosettacode
22:17 mwn3d joined #rosettacode
23:37 Wright joined #rosettacode

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