00:48
parsa has joined #ste||ar
01:12
parsa has joined #ste||ar
01:48
parsa has joined #ste||ar
01:49
parsa has quit [Client Quit]
01:49
parsa has joined #ste||ar
01:52
parsa has quit [Client Quit]
01:57
parsa has joined #ste||ar
03:16
hkaiser has quit [Read error: Connection reset by peer]
05:00
bita has joined #ste||ar
06:24
bita has quit [Ping timeout: 264 seconds]
07:21
parsa has joined #ste||ar
07:25
parsa has quit [Client Quit]
07:45
parsa has joined #ste||ar
07:45
parsa has quit [Client Quit]
08:57
parsa has joined #ste||ar
09:01
parsa has quit [Client Quit]
09:06
parsa has joined #ste||ar
10:42
<
gonidelis[m] >
"Performing operations on a forward iterator that is dereferenceable never makes its iterator value non-dereferenceable"
10:42
<
gonidelis[m] >
what's that supposed to mena?
10:42
<
gonidelis[m] >
mean^^
11:46
parsa has joined #ste||ar
11:47
parsa has quit [Client Quit]
11:50
parsa has joined #ste||ar
12:01
parsa has joined #ste||ar
12:08
parsa has joined #ste||ar
12:09
parsa has quit [Client Quit]
12:11
parsa has joined #ste||ar
12:12
parsa has quit [Client Quit]
12:22
parsa has joined #ste||ar
12:35
hkaiser has joined #ste||ar
12:55
<
gonidelis[m] >
hkaiser: 5 minute call?
12:56
<
hkaiser >
gonidelis[m]: didn't have coffee yet - I doubt I'd be able to answer questions right now ;-)
12:56
<
hkaiser >
can we do it at 8.45 Central?
12:56
<
gonidelis[m] >
hkaiser: yeah I just realised it's 7am there... what the hell ;p
12:57
<
gonidelis[m] >
give me a sec to calculate the difference
12:57
<
hkaiser >
16.45 your time
12:58
<
gonidelis[m] >
ahh... sorry I cannot. Could I ping you at some point after 9.30 central? Or is your program full today?
12:58
<
hkaiser >
yah, sorry
12:58
<
hkaiser >
1pm central would work
12:59
<
gonidelis[m] >
ok that's great
12:59
<
gonidelis[m] >
i will be extremely quick
12:59
<
hkaiser >
hope it's not too late for you
12:59
<
gonidelis[m] >
no it's not
13:52
<
gnikunj[m] >
hkaiser: yt?
14:03
<
hkaiser >
gnikunj[m]: here
14:04
<
gnikunj[m] >
hkaiser: I was checking the failing tests. It says 'BAD_COMMAND' there. What does it mean?
14:04
<
hkaiser >
no idea :/
14:04
<
hkaiser >
I wanted to spend some time looking into the windows problems today
14:04
<
gnikunj[m] >
it works on linux and it worked there perfectly. I haven't tried on windows, should I?
14:06
<
hkaiser >
something is off on the github windows builder
14:06
<
gnikunj[m] >
so that's not something on my end, right?
14:06
<
gnikunj[m] >
I was trying to reproduce it yesterday and failed
14:07
<
hkaiser >
I think you're fine (for now ;-)
14:07
<
gnikunj[m] >
sounds good. I'll wait for your lead then ;-)
14:12
<
gnikunj[m] >
it works without having the typename _Tp = ... part
14:13
<
gnikunj[m] >
ok got it, I should do a std::remove_reference to make the typename _Tp part work
14:42
parsa has joined #ste||ar
14:43
parsa has quit [Client Quit]
14:46
parsa has joined #ste||ar
14:48
parsa has quit [Client Quit]
14:49
parsa has joined #ste||ar
14:49
parsa has quit [Client Quit]
14:54
parsa has joined #ste||ar
14:56
parsa has quit [Client Quit]
15:05
parsa has joined #ste||ar
15:14
parsa has joined #ste||ar
15:15
parsa has quit [Client Quit]
15:15
weilewei has joined #ste||ar
15:16
parsa has joined #ste||ar
15:18
parsa has quit [Client Quit]
15:29
parsa has joined #ste||ar
15:29
parsa has quit [Client Quit]
15:47
parsa has joined #ste||ar
16:01
hkaiser has quit [Read error: Connection reset by peer]
16:03
hkaiser has joined #ste||ar
16:15
<
weilewei >
When I turned on HPX_WITH_THREAD_IDLE_RATES in cmake, how do I print hpx idle rate when running the application?
16:15
<
weilewei >
--hpx:print-counter=/threads/idle-rate ?
16:16
<
weilewei >
yea, correct, now I see it
16:18
parsa has joined #ste||ar
16:19
parsa has quit [Client Quit]
16:20
parsa has joined #ste||ar
16:23
parsa has quit [Client Quit]
16:23
bita has joined #ste||ar
16:42
parsa has joined #ste||ar
16:43
parsa has quit [Client Quit]
17:03
parsa has joined #ste||ar
17:03
parsa has quit [Client Quit]
17:06
parsa has joined #ste||ar
17:08
parsa has quit [Client Quit]
17:10
parsa has joined #ste||ar
17:13
parsa has quit [Client Quit]
17:51
weilewei has quit [Quit: Connection closed]
18:05
parsa has joined #ste||ar
18:07
parsa has quit [Client Quit]
18:17
<
K-ballo >
gnikunj[m]: yep, you got it
18:17
K-ballo has quit [Quit: K-ballo]
18:17
<
gnikunj[m] >
K-ballo :D
18:20
K-ballo has joined #ste||ar
18:21
<
gonidelis[m] >
why can't I use a `hpx::ranges::remove_if` inside my remove_if `tag_invoke` overload implementation?
18:27
parsa has joined #ste||ar
18:28
parsa has quit [Client Quit]
18:48
parsa has joined #ste||ar
19:11
<
gonidelis[m] >
hkaiser: you have 5 minutes?
19:11
<
hkaiser >
sorry, I missed the time
19:11
<
gonidelis[m] >
i missed it too
19:11
<
gonidelis[m] >
no worries
19:11
<
gonidelis[m] >
link?
19:37
weilewei has joined #ste||ar
19:43
parsa has joined #ste||ar
20:21
parsa has joined #ste||ar
20:22
parsa has quit [Client Quit]
20:26
<
gonidelis[m] >
K-ballo: do you have any idea why `remove` overloads that take a range as an input return a `borrowed_subrange_t` while `remove`s that take `iterator-sentinel` input return `surange` ?
20:29
<
K-ballo >
how could the iterator-sentinel overloads detecting dangling?
20:30
<
K-ballo >
the idea behind borrowed is that the iterators can survive the death of the range, so you need the range to be able to tell
20:31
<
gonidelis[m] >
ok I get your first sentence. that's what I have read
20:32
<
gonidelis[m] >
but why is dangling a danger in range overloads and not in iterator-sentinel overloads
20:32
<
K-ballo >
it's not, it's always a danger
20:33
<
gonidelis[m] >
in it-sent?
20:33
<
K-ballo >
yes, it is a danger for the iterator-sentinel overload too
20:33
<
K-ballo >
remove(std::vector{1, 2, 3, 0}.begin(), sentinel_stop_at(0), ...) dangles
20:34
<
K-ballo >
but std::vector{1, 2, 3, 0}.begin() would (could) be simply int*, so how can you tell?
20:34
<
K-ballo >
vs: std::vector v{1, 2, 3, 0}; remove(v.begin(), sentinel_stop_at(0), ...) // borrow from v
20:41
<
gonidelis[m] >
hmm...
20:43
parsa has joined #ste||ar
21:05
<
gonidelis[m] >
K-ballo: so i am still trying to figure out: is it like in ranges we can "contain" this dangle danger, and so we do?
21:05
<
gonidelis[m] >
but in iter-sent we cannot contain it at all?
21:06
<
K-ballo >
if you have an iterator-sentinel pair, you do not know the range they come from
21:06
<
K-ballo >
the range specific information, including whether you can borrow from it or not, is not available to you
21:07
<
K-ballo >
so it's impossible to know whether the resulting subrange will dangle or not
21:08
<
K-ballo >
both `std::vector v; foo(v.begin());` and `foo(std::vector{}.begin());` correspond to the same instantiation of `foo`, and yet one is fine and borrows while the other one dangles
21:09
<
K-ballo >
on the other hand, `std::vector v; bar(b);` and `bar(std::vector{})` correspond to different instantiations of `bar`
21:17
hkaiser has quit [Quit: bye]
21:39
weilewei has quit [Quit: Connection closed]
21:40
parsa has joined #ste||ar
21:47
parsa has joined #ste||ar
21:50
parsa has quit [Client Quit]
21:51
parsa has joined #ste||ar
21:52
parsa has quit [Client Quit]
21:53
parsa has joined #ste||ar
21:58
parsa has quit [Client Quit]
21:58
parsa has joined #ste||ar
22:08
parsa has joined #ste||ar
22:09
parsa has quit [Client Quit]
22:10
parsa has joined #ste||ar
22:19
parsa has joined #ste||ar
22:42
weilewei has joined #ste||ar
22:48
parsa has joined #ste||ar
22:51
parsa has quit [Client Quit]
22:53
parsa has joined #ste||ar
22:54
parsa has quit [Client Quit]
22:56
parsa has joined #ste||ar
22:58
parsa has quit [Client Quit]
23:02
parsa has joined #ste||ar
23:04
parsa has quit [Client Quit]
23:08
parsa has joined #ste||ar
23:09
parsa has quit [Client Quit]
23:10
parsa has joined #ste||ar
23:11
parsa has quit [Client Quit]
23:12
weilewei has quit [Quit: Connection closed]
23:15
hkaiser has joined #ste||ar
23:50
parsa has joined #ste||ar