OS Halting after about two minutes?

Jan 10, 2012 at 3:20 AM
Edited Jan 10, 2012 at 3:47 AM

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

 

Note: When running pear select "Release" instead of "Debug" and hit the green arrow. Thanks

Jan 10, 2012 at 6:24 AM
civilwarrock wrote:

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

 

Note: When running pear select "Release" instead of "Debug" and hit the green arrow. Thanks

Fixed, it I had memory leaks and it ran out of ram. I fixed it by seeing what was doing it. Fixed. Thanks - Matt

Coordinator
Jan 10, 2012 at 6:56 AM
Do we still have the memory display on top? If not, it might very well be running out of memory...

On Tue, Jan 10, 2012 at 5:20 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Jan 10, 2012 at 6:15 PM
mterwoord wrote:
Do we still have the memory display on top? If not, it might very well be running out of memory...

On Tue, Jan 10, 2012 at 5:20 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


I'm pretty sure it was. See one of my devsaccidenty was defining something every kernel run event and after awhile the os would just halt. So I am pretty surethats what it was. I fixed it and now it works fine. Is there anyway to display how much RAM your using? And is there anyway to do RAM management? Thanks - Matt

Coordinator
Jan 11, 2012 at 6:46 AM
We dont have any garbage collector yet, and it's probably better to hold off with that for a bit (till after coming compiler rework)..


On Tue, Jan 10, 2012 at 8:16 PM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

mterwoord wrote:
Do we still have the memory display on top? If not, it might very well be running out of memory...

On Tue, Jan 10, 2012 at 5:20 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


I'm pretty sure it was. See one of my devsaccidenty was defining something every kernel run event and after awhile the os would just halt. So I am pretty surethats what it was. I fixed it and now it works fine. Is there anyway to display how much RAM your using? And is there anyway to do RAM management? Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Jan 11, 2012 at 7:54 AM
mterwoord wrote:
We dont have any garbage collector yet, and it's probably better to hold off with that for a bit (till after coming compiler rework)..


On Tue, Jan 10, 2012 at 8:16 PM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

mterwoord wrote:
Do we still have the memory display on top? If not, it might very well be running out of memory...

On Tue, Jan 10, 2012 at 5:20 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


I'm pretty sure it was. See one of my devsaccidenty was defining something every kernel run event and after awhile the os would just halt. So I am pretty surethats what it was. I fixed it and now it works fine. Is there anyway to display how much RAM your using? And is there anyway to do RAM management? Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Oh ok, no problem just wondering. I'll be fine, I'll I have to do is ensure that my os is not defining something every Kernel run method. I recently coded multitasking for it. So hope fully I wont run into any leaks there. But I should be ok. Anyway, I was gonna ask you mterwood, is there anything in Cosmos to get how much ram has been used? I know you can get how much ram you computer has, but I would like to also see how much I've used. Anyway, Thanks - Matt

Coordinator
Jan 11, 2012 at 8:01 AM
You can access the Heap i think.. Probably the same class you use to detect how much ram you have..


On Wed, Jan 11, 2012 at 9:54 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

mterwoord wrote:
We dont have any garbage collector yet, and it's probably better to hold off with that for a bit (till after coming compiler rework)..


On Tue, Jan 10, 2012 at 8:16 PM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

mterwoord wrote:
Do we still have the memory display on top? If not, it might very well be running out of memory...

On Tue, Jan 10, 2012 at 5:20 AM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

Hey guys, if you go to: PearOs and download the current source and run it. The os halts after two mins or so on live boot and virtual. Any reason why - Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


I'm pretty sure it was. See one of my devsaccidenty was defining something every kernel run event and after awhile the os would just halt. So I am pretty surethats what it was. I fixed it and now it works fine. Is there anyway to display how much RAM your using? And is there anyway to do RAM management? Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Oh ok, no problem just wondering. I'll be fine, I'll I have to do is ensure that my os is not defining something every Kernel run method. I recently coded multitasking for it. So hope fully I wont run into any leaks there. But I should be ok. Anyway, I was gonna ask you mterwood, is there anything in Cosmos to get how much ram has been used? I know you can get how much ram you computer has, but I would like to also see how much I've used. Anyway, Thanks - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Developer
Jan 11, 2012 at 3:23 PM

It is possible to access the heap, but due to the absence of a GC, reclaiming memory isn't currently possible. (As soon as I finish getting the debugger situated with elements of arrays as children (I have it setup so that all of the array types can actually call a generic method which will sort them out :P (I still need to set that up for the non-array types, but they aren't in need of it nearly as bad as the arrays were (40-lines of code each, and their almost identical)), I also need to see if I can manage to setup infinite array depth, but not sure quite how to do that, aside from parsing the type name itself.), I'll see if I can't manage to port the GC we wrote for Proton (It's a compacting, generational, partially-delayed reference counting GC, it also has the concept of a large object heap, but the entire GC still needs a lot of testing so :P), over to C#, for use in Cosmos.)

Jan 11, 2012 at 4:40 PM
blah38621 wrote:

It is possible to access the heap, but due to the absence of a GC, reclaiming memory isn't currently possible. (As soon as I finish getting the debugger situated with elements of arrays as children (I have it setup so that all of the array types can actually call a generic method which will sort them out :P (I still need to set that up for the non-array types, but they aren't in need of it nearly as bad as the arrays were (40-lines of code each, and their almost identical)), I also need to see if I can manage to setup infinite array depth, but not sure quite how to do that, aside from parsing the type name itself.), I'll see if I can't manage to port the GC we wrote for Proton (It's a compacting, generational, partially-delayed reference counting GC, it also has the concept of a large object heap, but the entire GC still needs a lot of testing so :P), over to C#, for use in Cosmos.)

Ok cool, I'll see if I can access the heap. Thanks guys - Matt

Coordinator
Jan 12, 2012 at 7:04 AM
Orvid, could you have a look at field support as well for locals/watches?
GC is huge, and likely porting existing ones will be hard, as most require us to use hardware paging, or things like that...


On Wed, Jan 11, 2012 at 6:40 PM, civilwarrock <notifications@codeplex.com> wrote:

From: civilwarrock

blah38621 wrote:

It is possible to access the heap, but due to the absence of a GC, reclaiming memory isn't currently possible. (As soon as I finish getting the debugger situated with elements of arrays as children (I have it setup so that all of the array types can actually call a generic method which will sort them out :P (I still need to set that up for the non-array types, but they aren't in need of it nearly as bad as the arrays were (40-lines of code each, and their almost identical)), I also need to see if I can manage to setup infinite array depth, but not sure quite how to do that, aside from parsing the type name itself.), I'll see if I can't manage to port the GC we wrote for Proton (It's a compacting, generational, partially-delayed reference counting GC, it also has the concept of a large object heap, but the entire GC still needs a lot of testing so :P), over to C#, for use in Cosmos.)

Ok cool, I'll see if I can access the heap. Thanks guys - Matt

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com


Developer
Jan 12, 2012 at 1:00 PM

Actually, the GC for Proton is actually only 13kb of C code so :P Biggest thing will be modifying the places needed so that the GC is compatable with the way Cosmos handles objects :P (The proton GC was designed with a small layer of indirection, the reference type objects pointing to headers for objects in the heap rather than the objects themselves.) (I'll probably end up modifying the GC to be compatable rather than modifying Cosmos to be compatable, as there will be enough bugs to deal with as is without re-writing most of IL2CPU :P)

Coordinator
Jan 12, 2012 at 1:04 PM
Before implementing a GC: i think we should have some more kernel-parts implemented, like for example tcp/ip and some more...


On Thu, Jan 12, 2012 at 3:00 PM, blah38621 <notifications@codeplex.com> wrote:

From: blah38621

Actually, the GC for Proton is actually only 13kb of C code so :P Biggest thing will be modifying the places needed so that the GC is compatable with the way Cosmos handles objects :P (The proton GC was designed with a small layer of indirection, the reference type objects pointing to headers for objects in the heap rather than the objects themselves.) (I'll probably end up modifying the GC to be compatable rather than modifying Cosmos to be compatable, as there will be enough bugs to deal with as is without re-writing most of IL2CPU :P)

Read the full discussion online.

To add a post to this discussion, reply to this email (Cosmos@discussions.codeplex.com)

To start a new discussion for this project, email Cosmos@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com