www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Bounty Increase on Issue #1325927

reply Shammah Chancellor <anonymous coward.com> writes:
I've increased the bounty on this bug.   Fast CTFE is very important.

https://www.bountysource.com/issues/1325927-ctfe-copy-on-write-is-slow-and-caus
s-huge-memory-usage 


-Shammah
Jun 26 2014
parent reply Nick Sabalausky <SeeWebsiteToContactMe semitwist.com> writes:
On 6/26/2014 7:02 AM, Shammah Chancellor wrote:
 I've increased the bounty on this bug.   Fast CTFE is very important.

 https://www.bountysource.com/issues/1325927-ctfe-copy-on-write-is-slow-and-causes-huge-memory-usage

This is great news, and I'm sure very much appreciated by all. I can't help being a little concerned over issue ownership, though. My understanding is that Don's already done a large amount of work towards this issue. I wonder if that could actually be holding people back from contributing to the issue, for fear of taking whole pot unfairly (ie, swooping in and just doing the last little bit, or being perceived as attempting that), or fear of stirring up disagreement over money?
Jun 26 2014
next sibling parent Jonathan M Davis via Digitalmars-d <digitalmars-d puremagic.com> writes:
 Sent: Thursday, June 26, 2014 at 10:52 AM
 From: "Nick Sabalausky via Digitalmars-d" <digitalmars-d puremagic.com>
 To: digitalmars-d puremagic.com
 Subject: Re: Bounty Increase on Issue #1325927

 On 6/26/2014 7:02 AM, Shammah Chancellor wrote:
 I've increased the bounty on this bug.   Fast CTFE is very important.

 https://www.bountysource.com/issues/1325927-ctfe-copy-on-write-is-slow-and-causes-huge-memory-usage

This is great news, and I'm sure very much appreciated by all. I can't help being a little concerned over issue ownership, though. My understanding is that Don's already done a large amount of work towards this issue. I wonder if that could actually be holding people back from contributing to the issue, for fear of taking whole pot unfairly (ie, swooping in and just doing the last little bit, or being perceived as attempting that), or fear of stirring up disagreement over money?

I suspect that the bigger issue is simply that someone else needs to take the time to actually dig into the code and figure out how to do it. I don't think that anyone else has Don's level of expertise in this area, and I don't know how long it will take someone else to do it, but regardless of their skill level, it still requires time and initiative, and AFAIK, no one has stepped up to attempt it. - Jonathan M Davis
Jun 26 2014
prev sibling next sibling parent reply "Joakim" <dlang joakim.airpost.net> writes:
On Thursday, 26 June 2014 at 17:52:13 UTC, Nick Sabalausky wrote:
 On 6/26/2014 7:02 AM, Shammah Chancellor wrote:
 I've increased the bounty on this bug.   Fast CTFE is very 
 important.

 https://www.bountysource.com/issues/1325927-ctfe-copy-on-write-is-slow-and-causes-huge-memory-usage

This is great news, and I'm sure very much appreciated by all. I can't help being a little concerned over issue ownership, though. My understanding is that Don's already done a large amount of work towards this issue. I wonder if that could actually be holding people back from contributing to the issue, for fear of taking whole pot unfairly (ie, swooping in and just doing the last little bit, or being perceived as attempting that), or fear of stirring up disagreement over money?

Don's a senior developer at a company that just got bought for $200 million. I doubt he's stressing over a $400 bounty, ;) especially if it takes some work off his plate.
Jun 26 2014
next sibling parent Etienne <etcimon gmail.com> writes:
On 2014-06-27 5:53 AM, safety0ff wrote:
 Perhaps the bounty won't stimulate anybody who doesn't have other
 motivations to improve the situation, but more information about the
 scope of the issue would be helpful to both backers and potential
 claimants.

From what I've seen writing an ASN.1 compiler with D is, sometimes you just don't know if some part of the tree structure generated from the ctfe is referenced anywhere else (which is somewhat possible to track with reference counts). The garbage collector solves a potential 500+ hours of work making tree structures referentially self-aware. My guess is that the compiler doesn't know if parts of the CTFE function will be used at runtime, no matter how obvious it is that it won't, there's just no information kept lying around about it and it gets confused with the tree structures used and sent to the backend for the runtime routines.
Jun 27 2014
prev sibling parent Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 6/27/14, 8:54 AM, Andrej Mitrovic via Digitalmars-d wrote:
 That's a pretty big bounty though. I bet it would be motivating for
 the jobless. :P

 On 6/27/14, Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> wrote:
 On 27 June 2014 10:42, Don via Digitalmars-d
 <digitalmars-d puremagic.com> wrote:
 I doubt bounties are effective as a motivation for this kind of thing.

+1


There are always students and un(der)employed people who have a passion for something, but need to mind other things to make ends meet. Bounties allow them to work on what they like and also make some money. Facebook granted me some additional budget for bounties. I am looking for ideas on allocating it. Andrei
Jun 27 2014
prev sibling next sibling parent "Don" <x nospam.com> writes:
On Thursday, 26 June 2014 at 21:20:04 UTC, Joakim wrote:
 On Thursday, 26 June 2014 at 17:52:13 UTC, Nick Sabalausky 
 wrote:
 On 6/26/2014 7:02 AM, Shammah Chancellor wrote:
 I've increased the bounty on this bug.   Fast CTFE is very 
 important.

 https://www.bountysource.com/issues/1325927-ctfe-copy-on-write-is-slow-and-causes-huge-memory-usage

This is great news, and I'm sure very much appreciated by all. I can't help being a little concerned over issue ownership, though. My understanding is that Don's already done a large amount of work towards this issue. I wonder if that could actually be holding people back from contributing to the issue, for fear of taking whole pot unfairly (ie, swooping in and just doing the last little bit, or being perceived as attempting that), or fear of stirring up disagreement over money?

Don's a senior developer at a company that just got bought for $200 million. I doubt he's stressing over a $400 bounty, ;) especially if it takes some work off his plate.

Yes, of course I'm not interested in bounties. But note that that issue is not really a "bug", it's a project. I put hundreds of hours of work into this, to get to the point where we are now - fixing the compiler structure to the point where a JIT is possible. That work was funded by an insolvency payout :). Daniel Murphy has done some work on it, as well. I doubt bounties are effective as a motivation for this kind of thing.
Jun 27 2014
prev sibling next sibling parent "safety0ff" <safety0ff.dev gmail.com> writes:
On Friday, 27 June 2014 at 09:42:22 UTC, Don wrote:
 Yes, of course I'm not interested in bounties. But note that 
 that issue is not really a "bug", it's a project.
 I put hundreds of hours of work into this, to get to the point 
 where we are now - fixing the compiler structure to the point 
 where a JIT is possible. That work was funded by an insolvency 
 payout :). Daniel Murphy has done some work on it, as well.

 I doubt bounties are effective as a motivation for this kind of 
 thing.

Is there any chance you could offer a brief summary of the state of things w.r.t. this issue? I.e. expanding on this comment: "Upgrading severity. I've done several commits to move towards a solution but I still need to do more restructuring to properly fix this." Perhaps the bounty won't stimulate anybody who doesn't have other motivations to improve the situation, but more information about the scope of the issue would be helpful to both backers and potential claimants.
Jun 27 2014
prev sibling next sibling parent Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> writes:
On 27 June 2014 10:42, Don via Digitalmars-d
<digitalmars-d puremagic.com> wrote:
 I doubt bounties are effective as a motivation for this kind of thing.

+1
Jun 27 2014
prev sibling next sibling parent Andrej Mitrovic via Digitalmars-d <digitalmars-d puremagic.com> writes:
That's a pretty big bounty though. I bet it would be motivating for
the jobless. :P

On 6/27/14, Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> wrote:
 On 27 June 2014 10:42, Don via Digitalmars-d
 <digitalmars-d puremagic.com> wrote:
 I doubt bounties are effective as a motivation for this kind of thing.

+1

Jun 27 2014
prev sibling next sibling parent =?UTF-8?B?Ik5vcmRsw7Z3Ig==?= <per.nordlow gmail.com> writes:
 I doubt bounties are effective as a motivation for this kind of 
 thing.

If so couldn't you make your code public?
Jun 28 2014
prev sibling next sibling parent Iain Buclaw via Digitalmars-d <digitalmars-d puremagic.com> writes:
On 28 June 2014 15:21, "Nordlöw" <digitalmars-d puremagic.com> wrote:
 I doubt bounties are effective as a motivation for this kind of thing.

If so couldn't you make your code public?

I doubt there is any code to be made public that isn't already. :)
Jun 28 2014
prev sibling parent "Dicebot" <public dicebot.lv> writes:
On Saturday, 28 June 2014 at 14:21:03 UTC, Nordlöw wrote:
 I doubt bounties are effective as a motivation for this kind 
 of thing.

If so couldn't you make your code public?

Don does not have any code that actually fixes that issue. He as done a lot of job on CTFE refactoring and unification making it actually possible to _start_ fixing the issue. And all this work is now part of DMD master.
Jun 28 2014