Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-09 Thread Tharen Abela
I managed to resolve the issue by allocating in non-GC memory, using persistentalloc. Should anyone want to look into it in detail, here is the commit On Monday, 8 April 2019 20:50:48 UTC+2, Tharen Abela

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-08 Thread Michael Jones
I have no experience with this code...but a lot of debugging experience. Since you think you know the general problem and have both hands inside the patient already, why not (for now) modify the GC scanner to always look at your batch’s pointer list and explicitly avoid messing with them. Then

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-08 Thread Tharen Abela
I am keeping an `allb` slice, and with that I did see it occasionally succeed. I am using the binarytree test, since it is an issue regarding the GC. In fact running it

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-08 Thread Ian Lance Taylor
On Sun, Apr 7, 2019 at 12:30 PM Tharen Abela wrote: > > The gist of the problem is that I am allocating an object in the runtime, > (which I refer to as the batch struct), and the GC is deallocating the > object, even though a reference is being kept in a slice (similar to allp and > allm). >

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-07 Thread Tharen Abela
Should be public now, thanks for bringing that to my attention. On Sunday, 7 April 2019 20:41:41 UTC+2, robert engels wrote: > > The link to the repo does not allow not credentialed access... > > On Apr 7, 2019, at 5:58 AM, Tharen Abela > > wrote: > > I'll preface this by saying this is for

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-07 Thread Tharen Abela
The gist of the problem is that I am allocating an object in the runtime, (which I refer to as the batch struct), and the GC is deallocating the object, even though a reference is being kept in a slice (similar to *allp* and *allm*). While allocating, I call *acquirem* to prevent the GC being

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-07 Thread robert engels
I think it would be helpful to if you providee a summary of the issue/idea before the detailed information - it makes it nearly impossible to understand - at least for me. > On Apr 7, 2019, at 1:41 PM, robert engels wrote: > > The link to the repo does not allow not credentialed access... >

Re: [go-nuts] Modifying the Runtime Scheduler

2019-04-07 Thread robert engels
The link to the repo does not allow not credentialed access... > On Apr 7, 2019, at 5:58 AM, Tharen Abela wrote: > > I'll preface this by saying this is for academic purposes. > > I am working in the following repository > , but the following files are