00:00
<
heller >
hkaiser: I can push the constexpr fix
00:01
<
heller >
in combination with another small fix for comparing signed and unsigned
00:01
<
hkaiser >
heller: sure, feel free
00:02
<
hkaiser >
heller: I have the inspect fixes done locally
00:03
<
heller >
I'll wait until you pushed your stuff
00:13
<
hkaiser >
heller: done, I hope it does not create conflicts
00:13
ste||ar-github has joined #ste||ar
00:13
<
ste||ar-github >
hpx/thread_local_allocator 8c55e9a Hartmut Kaiser: Fixing inspect problems...
00:13
ste||ar-github has left #ste||ar [#ste||ar]
00:15
<
heller >
hkaiser: ok, I either messed something up when copying over the results, or the changes didn't change the overall timings for my benchmarks significantly (at least I can't spot visiual differrence between the graphs)
00:15
<
hkaiser >
heller: could be that the distributed overheads overshadow everything
00:16
<
heller >
no, even my future overhead tests
00:16
<
hkaiser >
heller: I see massive speedup in Phylanx
00:16
<
heller >
or local async
00:16
<
hkaiser >
or tcmalloc is that good
00:16
<
hkaiser >
or jemalloc
00:17
ste||ar-github has joined #ste||ar
00:17
<
ste||ar-github >
hpx/thread_local_allocator 26d37e6 Thomas Heller: Fixing gcc compilation errors
00:17
ste||ar-github has left #ste||ar [#ste||ar]
00:19
<
heller >
hkaiser: yeah, those tests should be fairly allocator friendly
00:19
<
heller >
at least there is no performance regression ;)
00:20
<
hkaiser >
heller: would be worth chnaging the thread-allocator benchmark
00:20
<
hkaiser >
to compare things
00:20
<
khuck >
heller: is there a branch I can pull to benchmark with phylanx?
00:21
<
heller >
khuck: yes. thread_local_allocator
00:21
<
hkaiser >
K-ballo: use the thread_allocator branch for Phylanx as well
00:21
<
hkaiser >
khuck: ^^
00:21
<
hkaiser >
sorry K-ballo
00:21
<
khuck >
do I do anything else differently? with/without tc/jemalloc?
00:22
<
hkaiser >
just use the hpx and phylanx branches
00:24
<
heller >
still failing
00:26
<
hkaiser >
ok, new errors
00:26
<
hkaiser >
typename's missing ... I hate the MS compiler for not reporting those
00:28
ste||ar-github has joined #ste||ar
00:28
<
ste||ar-github >
hpx/thread_local_allocator 134adbf Hartmut Kaiser: Adding missing typename keyword
00:28
ste||ar-github has left #ste||ar [#ste||ar]
00:29
<
heller >
hkaiser: gcc didn't either :/
00:32
<
khuck >
heller: not even with -Wall -Werror?
00:47
V|r has joined #ste||ar
01:27
khuck has quit [Remote host closed the connection]
01:31
khuck has joined #ste||ar
02:01
khuck has quit [Remote host closed the connection]
02:29
hkaiser has quit [Quit: bye]
02:35
khuck has joined #ste||ar
02:45
khuck has quit [Ping timeout: 252 seconds]
03:09
khuck has joined #ste||ar
03:10
khuck has quit [Read error: Connection reset by peer]
03:10
khuck_ has joined #ste||ar
03:15
khuck_ has quit [Ping timeout: 272 seconds]
03:51
K-ballo has quit [Quit: K-ballo]
03:53
khuck has joined #ste||ar
03:54
khuck has quit [Read error: Connection reset by peer]
03:54
khuck_ has joined #ste||ar
03:55
nanashi55 has quit [Ping timeout: 272 seconds]
03:55
nanashi55 has joined #ste||ar
03:59
khuck_ has quit [Ping timeout: 252 seconds]
04:01
khuck has joined #ste||ar
04:06
khuck has quit [Ping timeout: 244 seconds]
04:27
khuck has joined #ste||ar
04:31
khuck has quit [Ping timeout: 244 seconds]
04:51
khuck has joined #ste||ar
04:52
khuck has quit [Read error: Connection reset by peer]
04:52
khuck_ has joined #ste||ar
04:57
khuck_ has quit [Ping timeout: 272 seconds]
05:41
khuck has joined #ste||ar
05:42
khuck has quit [Read error: Connection reset by peer]
05:42
khuck_ has joined #ste||ar
05:47
khuck has joined #ste||ar
05:47
khuck_ has quit [Ping timeout: 252 seconds]
05:47
khuck has quit [Read error: No route to host]
05:48
khuck has joined #ste||ar
05:53
khuck has quit [Ping timeout: 272 seconds]
05:57
khuck has joined #ste||ar
05:58
khuck has quit [Read error: Connection reset by peer]
05:58
khuck_ has joined #ste||ar
05:59
khuck_ has quit [Read error: Connection reset by peer]
05:59
khuck has joined #ste||ar
06:00
khuck has quit [Read error: Connection reset by peer]
06:00
khuck_ has joined #ste||ar
06:04
khuck_ has quit [Ping timeout: 244 seconds]
07:48
khuck has joined #ste||ar
07:53
khuck has quit [Ping timeout: 252 seconds]
07:53
khuck has joined #ste||ar
07:54
khuck has quit [Read error: Connection reset by peer]
07:54
khuck_ has joined #ste||ar
07:59
khuck_ has quit [Ping timeout: 272 seconds]
09:06
ste||ar-github has joined #ste||ar
09:06
<
ste||ar-github >
hpx/gh-pages 3e1c159 StellarBot: Updating docs
09:06
ste||ar-github has left #ste||ar [#ste||ar]
10:03
mcopik_ has quit [Ping timeout: 264 seconds]
10:43
khuck has joined #ste||ar
10:48
khuck has quit [Ping timeout: 252 seconds]
11:35
K-ballo has joined #ste||ar
13:55
Anushi1998 has joined #ste||ar
13:55
eschnett has joined #ste||ar
13:55
<
K-ballo >
my log branch makes things slower somehow, as I suspected
14:19
khuck has joined #ste||ar
14:23
eschnett has quit [Quit: eschnett]
14:24
khuck has quit [Ping timeout: 246 seconds]
14:29
Anushi1998 has quit [Remote host closed the connection]
14:29
Anushi1998 has joined #ste||ar
14:32
khuck has joined #ste||ar
14:33
khuck has quit [Read error: Connection reset by peer]
14:33
khuck_ has joined #ste||ar
14:38
khuck_ has quit [Ping timeout: 252 seconds]
14:40
khuck has joined #ste||ar
14:43
hkaiser has joined #ste||ar
14:45
khuck has quit [Ping timeout: 252 seconds]
15:05
mcopik_ has joined #ste||ar
15:07
khuck has joined #ste||ar
15:14
mcopik_ has quit [Ping timeout: 250 seconds]
15:16
eschnett has joined #ste||ar
15:18
khuck has quit [Ping timeout: 246 seconds]
15:19
eschnett has quit [Client Quit]
15:22
Anushi1998 has quit [Quit: Bye]
15:24
ste||ar-github has joined #ste||ar
15:24
<
ste||ar-github >
hpx/logging 412a259 Agustin K-ballo Berge: pruning util/logging
15:24
ste||ar-github has left #ste||ar [#ste||ar]
15:44
<
nikunj97 >
hkaiser, see pm please
15:48
* zao
eats all nikunj97's PMs
16:00
eschnett has joined #ste||ar
16:13
eschnett has quit [Quit: eschnett]
17:21
mcopik has joined #ste||ar
17:29
mcopik has quit [Ping timeout: 252 seconds]
18:24
eschnett has joined #ste||ar
18:29
mcopik has joined #ste||ar
18:44
eschnett has quit [Quit: eschnett]
18:46
eschnett has joined #ste||ar
18:50
eschnett has quit [Ping timeout: 245 seconds]
19:51
ste||ar-github has joined #ste||ar
19:51
<
ste||ar-github >
hpx/thread_local_allocator c00c89a Hartmut Kaiser: Modifying allocator test to print timing information, adding std::allocator as a reference...
19:51
ste||ar-github has left #ste||ar [#ste||ar]
19:51
<
hkaiser >
heller: I added timings to the allocator test as well as measuring base line timings using std::allocator
19:51
<
hkaiser >
heller: would be interesting to see how jemalloc/tcmalloc perform compared to what we have now
19:52
<
hkaiser >
this will also tell us whether its worth going ahead with the thread local allocator
20:36
zao_ has joined #ste||ar
21:48
zao_ has quit [Quit: Up, up, and away!]
22:33
ste||ar-github has joined #ste||ar
22:33
ste||ar-github has left #ste||ar [#ste||ar]
22:33
<
ste||ar-github >
hpx/logging cc90463 Agustin K-ballo Berge: pruning util/logging
23:03
<
K-ballo >
now I get the same numbers I get for master