Re: [m5-dev] stable has been updated
I did. I also found a bug in SPARC which has been in there since November of 2006. The parenthesis in inUserMode are wrong so that when tracing in FS mode, it tries to access the wrong miscreg which happens to panic. I'm running SPARC regressions right now, and assuming they pass, is that something I should add to stable? Gabe nathan binkert wrote: Can you at least update that regression? I think it's just very slight stats diffs and Korey thought they were fine. Nate On Sat, Jul 4, 2009 at 8:23 PM, Gabe Blackgbl...@eecs.umich.edu wrote: nathan binkert wrote: The stable repository is now up to date with the development repository. Not that I think it's a bad thing to update stable, but did you verify that all regressions pass? For example, I think that there are still two inorder regressions that fail. Also, are there any known bugs at this point that need to be fixed? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev I didn't, and looking at the last full regressions I see the failing in order runs. I'm not aware of any other known bugs in anything that would be expected to work other than the simple timing trace bug that Geoff found a while ago. Gabe ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable has been updated
I did. I also found a bug in SPARC which has been in there since November of 2006. The parenthesis in inUserMode are wrong so that when tracing in FS mode, it tries to access the wrong miscreg which happens to panic. I'm running SPARC regressions right now, and assuming they pass, is that something I should add to stable? In my opinion, any time a bug is fixed, doing it in stable is OK. It's just that as stable and dev diverge, it's more work to do that. Generally, I'd say that if you did want to do such a thing, you should commit the fix in stable and merge it into -dev. If there haven't been any intervening -dev commits, as I believe is the case right now, it doesn't matter. By the way. Thanks a ton for doing this! Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
[m5-dev] stable has been updated
The stable repository is now up to date with the development repository. Gabe ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable has been updated
The stable repository is now up to date with the development repository. Not that I think it's a bad thing to update stable, but did you verify that all regressions pass? For example, I think that there are still two inorder regressions that fail. Also, are there any known bugs at this point that need to be fixed? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable has been updated
nathan binkert wrote: The stable repository is now up to date with the development repository. Not that I think it's a bad thing to update stable, but did you verify that all regressions pass? For example, I think that there are still two inorder regressions that fail. Also, are there any known bugs at this point that need to be fixed? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev I didn't, and looking at the last full regressions I see the failing in order runs. I'm not aware of any other known bugs in anything that would be expected to work other than the simple timing trace bug that Geoff found a while ago. Gabe ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable has been updated
Can you at least update that regression? I think it's just very slight stats diffs and Korey thought they were fine. Nate On Sat, Jul 4, 2009 at 8:23 PM, Gabe Blackgbl...@eecs.umich.edu wrote: nathan binkert wrote: The stable repository is now up to date with the development repository. Not that I think it's a bad thing to update stable, but did you verify that all regressions pass? For example, I think that there are still two inorder regressions that fail. Also, are there any known bugs at this point that need to be fixed? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev I didn't, and looking at the last full regressions I see the failing in order runs. I'm not aware of any other known bugs in anything that would be expected to work other than the simple timing trace bug that Geoff found a while ago. Gabe ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
[m5-dev] stable
I just tried to compile m5 on a fc9 machine and it didn't work We shouldn't push stable until I get that fixed... Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable
That's odd. I thought I had done that. What compiler is it? What swig, etc? On Wed, Oct 8, 2008 at 1:01 PM, Ali Saidi [EMAIL PROTECTED] wrote: I just tried to compile m5 on a fc9 machine and it didn't work We shouldn't push stable until I get that fixed... Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable
It's the mysql version string in compare versions. I thought I fixed it, but I'm looking at it now... Ali On Oct 8, 2008, at 4:09 PM, nathan binkert wrote: That's odd. I thought I had done that. What compiler is it? What swig, etc? On Wed, Oct 8, 2008 at 1:01 PM, Ali Saidi [EMAIL PROTECTED] wrote: I just tried to compile m5 on a fc9 machine and it didn't work We shouldn't push stable until I get that fixed... Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable
I just pushed a fix I wrote it and I thought I had committed it, but apparently not. Ali On Oct 8, 2008, at 6:18 PM, Ali Saidi wrote: It's the mysql version string in compare versions. I thought I fixed it, but I'm looking at it now... Ali On Oct 8, 2008, at 4:09 PM, nathan binkert wrote: That's odd. I thought I had done that. What compiler is it? What swig, etc? On Wed, Oct 8, 2008 at 1:01 PM, Ali Saidi [EMAIL PROTECTED] wrote: I just tried to compile m5 on a fc9 machine and it didn't work We shouldn't push stable until I get that fixed... Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable
Seems ready to me. On Wed, Oct 8, 2008 at 5:28 PM, [EMAIL PROTECTED] wrote: So is everything ready for stable now? Gabe Quoting Ali Saidi [EMAIL PROTECTED]: I just pushed a fix I wrote it and I thought I had committed it, but apparently not. Ali On Oct 8, 2008, at 6:18 PM, Ali Saidi wrote: It's the mysql version string in compare versions. I thought I fixed it, but I'm looking at it now... Ali On Oct 8, 2008, at 4:09 PM, nathan binkert wrote: That's odd. I thought I had done that. What compiler is it? What swig, etc? On Wed, Oct 8, 2008 at 1:01 PM, Ali Saidi [EMAIL PROTECTED] wrote: I just tried to compile m5 on a fc9 machine and it didn't work We shouldn't push stable until I get that fixed... Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
[m5-dev] stable
I'm running a full set of regressions and I will commit all of the output so that there will be no differences. Other than that, the only thing left is the problem in the mips register file. Diff #3 that I sent the other day is not actually correct, but it would at least generate an assertion and allow m5 to compile under 4.3. If someone would like to fix it, that would be great. Anything else to do? I personally don't want this to be 2.0, we need to plan a bit ahead for that. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] stable
1.99? Or did we already have one of those? Gabe nathan binkert wrote: I'm running a full set of regressions and I will commit all of the output so that there will be no differences. Other than that, the only thing left is the problem in the mips register file. Diff #3 that I sent the other day is not actually correct, but it would at least generate an assertion and allow m5 to compile under 4.3. If someone would like to fix it, that would be great. Anything else to do? I personally don't want this to be 2.0, we need to plan a bit ahead for that. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
Do you want to hold off on letting a stable out for this stuff? I'd say that we need to get stable done to allow things into the tree, and we can try to make this a part of the next stable release which we can target as 2.0. Nate I think high on the list needs to be a clean-up of the statistics in models. Some are fine, most exist but their precise meaning (mostly in the CPU and cache models) isn't clear, a few might be wrong, and some don't exist, but they need to (bus and bridge). I don't the the statistics are presently any worse than they are in stable, however I do think we should spend some time working on them in the near term and call that 2.0. I'll volunteer to take care of the bus and the bridge. Ali On Sep 26, 2008, at 16:41, nathan binkert [EMAIL PROTECTED] wrote: Ok, now that kevin has fixed the O3 problems that we've had, what's left? 1) I've got a couple more outstanding changes that get GCC 4.3 fully functional 2) I'd like to update all regressions so there are no differences. Anything else? Someone else willing to do the leg work to announce everything? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
I could go either way. If it forced us to just get it done in 2 or 3 days then it would be worth it. Otherwise, we should push it off, but not too far. Ali On Sep 26, 2008, at 11:51 PM, nathan binkert wrote: Do you want to hold off on letting a stable out for this stuff? I'd say that we need to get stable done to allow things into the tree, and we can try to make this a part of the next stable release which we can target as 2.0. Nate I think high on the list needs to be a clean-up of the statistics in models. Some are fine, most exist but their precise meaning (mostly in the CPU and cache models) isn't clear, a few might be wrong, and some don't exist, but they need to (bus and bridge). I don't the the statistics are presently any worse than they are in stable, however I do think we should spend some time working on them in the near term and call that 2.0. I'll volunteer to take care of the bus and the bridge. Ali On Sep 26, 2008, at 16:41, nathan binkert [EMAIL PROTECTED] wrote: Ok, now that kevin has fixed the O3 problems that we've had, what's left? 1) I've got a couple more outstanding changes that get GCC 4.3 fully functional 2) I'd like to update all regressions so there are no differences. Anything else? Someone else willing to do the leg work to announce everything? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
Maybe we should start by going through flyspray and determining what issues still exist and which need to be closed. It would be nice to get a 2.0 release done, but it seems hard to say that we could do that in 2 or 3 days. I could put some serious effort into it after ISCA is over and try to really get something going by sometime in January. Any other opinions? Nate I could go either way. If it forced us to just get it done in 2 or 3 days then it would be worth it. Otherwise, we should push it off, but not too far. Ali On Sep 26, 2008, at 11:51 PM, nathan binkert wrote: Do you want to hold off on letting a stable out for this stuff? I'd say that we need to get stable done to allow things into the tree, and we can try to make this a part of the next stable release which we can target as 2.0. Nate I think high on the list needs to be a clean-up of the statistics in models. Some are fine, most exist but their precise meaning (mostly in the CPU and cache models) isn't clear, a few might be wrong, and some don't exist, but they need to (bus and bridge). I don't the the statistics are presently any worse than they are in stable, however I do think we should spend some time working on them in the near term and call that 2.0. I'll volunteer to take care of the bus and the bridge. Ali On Sep 26, 2008, at 16:41, nathan binkert [EMAIL PROTECTED] wrote: Ok, now that kevin has fixed the O3 problems that we've had, what's left? 1) I've got a couple more outstanding changes that get GCC 4.3 fully functional 2) I'd like to update all regressions so there are no differences. Anything else? Someone else willing to do the leg work to announce everything? Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
nathan binkert wrote: I'll give it a look and see if I can figure out what's wrong. It would also be awesome if you can take a look at the perlbmk problem. You can talk to Gabe about it, but I don't think he was able to figure it out. Hope you had a relaxing trip! Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev Not to be pushy or anything (ok, yes, to be pushy), but I just counted and I've got 41 outstanding patches that need to make it into the tree, and that's before I separate out a lot of little things that should be in their own patches I was too lazy to worry about at the time. It's starting to get to the point where it feels like I'm working on a fork... Gabe ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
how long has SMT regressions been broke? Is this a new thing? I havent looked at O3 SMT in awhile, but I would assume the O3 changes over the past year or so take my fix time way up! Is there a particular changeset or marker where we can say it broke here...? On Sat, Sep 6, 2008 at 1:25 PM, nathan binkert [EMAIL PROTECTED] wrote: What's still left to do for stable? There's the SMT regression failure. Korey or Kevin can you guys look into that so we can get this done? Is there anything else? I have stuff just waiting to get into the tree and I'm holding off for this stable release. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev -- -- Korey L Sewell Graduate Student - PhD Candidate Computer Science Engineering University of Michigan ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
alright, i'll dig a little and see what's up. On Mon, Sep 8, 2008 at 10:27 AM, nathan binkert [EMAIL PROTECTED] wrote: how long has SMT regressions been broke? Is this a new thing? I havent looked at O3 SMT in awhile, but I would assume the O3 changes over the past year or so take my fix time way up! Is there a particular changeset or marker where we can say it broke here...? It hasn't been broken for all that long, and the break is just that an assertion is failing now that never used to fail. The assertion started firing when Steve changed the default latency for main memory to something longer and more realistic. I think Ali sent an e-mail a few weeks ago with the exact changeset number, but given that it's an assertion that now fails because a simple latency changed, I don't know that the changeset helps all that much. The problem is that the rest of us don't know enough to understand the assertion. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev -- -- Korey L Sewell Graduate Student - PhD Candidate Computer Science Engineering University of Michigan ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
[m5-dev] Stable
What's still left to do for stable? There's the SMT regression failure. Korey or Kevin can you guys look into that so we can get this done? Is there anything else? I have stuff just waiting to get into the tree and I'm holding off for this stable release. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable
Statistics cleanup? gcc 4.2? Ali On Sep 6, 2008, at 1:25 PM, nathan binkert wrote: What's still left to do for stable? There's the SMT regression failure. Korey or Kevin can you guys look into that so we can get this done? Is there anything else? I have stuff just waiting to get into the tree and I'm holding off for this stable release. Nate ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable Release
I'm going to be gone starting on Wednesday until Sept. 6th, so I won't have a chance to take a look at the regression errors before hand. I'll try to take a look when I come back. Kevin Quoting Ali Saidi [EMAIL PROTECTED]: From the release notes the following are still outstanding for a 2.0 release. I don't really care about the Cygwin problems and I'm not sure that anyone else does, so I move to strike those. Additionally, we made the repository public so that one is gone. I think the three things that need to be fixed are: O3 CPU -- SMT SE regression hits an assert; bug causing perlbmk to fail (Gabe reported this in April 2007 and I don't know if it's still a problem http://www.m5sim.org/flyspray/task/221); NACKs/coherence messages Statistics -- The CPU statistics are confusing. I think the idle time in the simple CPU can still be wrong after a CPU switch and it's at least not clear to me which idles mean quiesced time or idle because of other things (blocked in the O3 case, not switched out in the simple case); The bus needs statistics; I don't know if the cache statistics are reasonable. Testing -- I've at least done a fair amount of testing over the last month or two, so I'm pretty happy with that. I think some validation needs to be done on an statistics changes though. Ali 1. Fix O3 CPU bug in SE 40.perlbmk fails 2. Fix O3 processing nacks/coherence messages 3. Better statistics for the caches. 4. FS mode doesn't work under Cygwin 5. memtest regression crashes under Cygwin 6. Make repository public 7. Testing 8. Validation On Aug 24, 2008, at 9:55 AM, Ali Saidi wrote: Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
[m5-dev] Stable Release
Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable Release
From the release notes the following are still outstanding for a 2.0 release. I don't really care about the Cygwin problems and I'm not sure that anyone else does, so I move to strike those. Additionally, we made the repository public so that one is gone. I think the three things that need to be fixed are: O3 CPU -- SMT SE regression hits an assert; bug causing perlbmk to fail (Gabe reported this in April 2007 and I don't know if it's still a problem http://www.m5sim.org/flyspray/task/221); NACKs/coherence messages Statistics -- The CPU statistics are confusing. I think the idle time in the simple CPU can still be wrong after a CPU switch and it's at least not clear to me which idles mean quiesced time or idle because of other things (blocked in the O3 case, not switched out in the simple case); The bus needs statistics; I don't know if the cache statistics are reasonable. Testing -- I've at least done a fair amount of testing over the last month or two, so I'm pretty happy with that. I think some validation needs to be done on an statistics changes though. Ali 1. Fix O3 CPU bug in SE 40.perlbmk fails 2. Fix O3 processing nacks/coherence messages 3. Better statistics for the caches. 4. FS mode doesn't work under Cygwin 5. memtest regression crashes under Cygwin 6. Make repository public 7. Testing 8. Validation On Aug 24, 2008, at 9:55 AM, Ali Saidi wrote: Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable Release
We should also figure out why the all regressions failed completely last night. Gabe Black wrote: I'm pretty sure that perlbmk bug is still there and we just got rid of the regression. It might have gone away on it's own, but I don't think anyone actively tried to fix it. Gabe Ali Saidi wrote: From the release notes the following are still outstanding for a 2.0 release. I don't really care about the Cygwin problems and I'm not sure that anyone else does, so I move to strike those. Additionally, we made the repository public so that one is gone. I think the three things that need to be fixed are: O3 CPU -- SMT SE regression hits an assert; bug causing perlbmk to fail (Gabe reported this in April 2007 and I don't know if it's still a problem http://www.m5sim.org/flyspray/task/221); NACKs/coherence messages Statistics -- The CPU statistics are confusing. I think the idle time in the simple CPU can still be wrong after a CPU switch and it's at least not clear to me which idles mean quiesced time or idle because of other things (blocked in the O3 case, not switched out in the simple case); The bus needs statistics; I don't know if the cache statistics are reasonable. Testing -- I've at least done a fair amount of testing over the last month or two, so I'm pretty happy with that. I think some validation needs to be done on an statistics changes though. Ali 1. Fix O3 CPU bug in SE 40.perlbmk fails 2. Fix O3 processing nacks/coherence messages 3. Better statistics for the caches. 4. FS mode doesn't work under Cygwin 5. memtest regression crashes under Cygwin 6. Make repository public 7. Testing 8. Validation On Aug 24, 2008, at 9:55 AM, Ali Saidi wrote: Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable Release
There was a change that Ke Meng sent that fixed some problems with O3 in Alpha deadlocking. That fixed quite possibly fixed this as well. Ali On Aug 24, 2008, at 12:24 PM, Gabe Black wrote: I'm pretty sure that perlbmk bug is still there and we just got rid of the regression. It might have gone away on it's own, but I don't think anyone actively tried to fix it. Gabe Ali Saidi wrote: From the release notes the following are still outstanding for a 2.0 release. I don't really care about the Cygwin problems and I'm not sure that anyone else does, so I move to strike those. Additionally, we made the repository public so that one is gone. I think the three things that need to be fixed are: O3 CPU -- SMT SE regression hits an assert; bug causing perlbmk to fail (Gabe reported this in April 2007 and I don't know if it's still a problem http://www.m5sim.org/flyspray/task/221); NACKs/coherence messages Statistics -- The CPU statistics are confusing. I think the idle time in the simple CPU can still be wrong after a CPU switch and it's at least not clear to me which idles mean quiesced time or idle because of other things (blocked in the O3 case, not switched out in the simple case); The bus needs statistics; I don't know if the cache statistics are reasonable. Testing -- I've at least done a fair amount of testing over the last month or two, so I'm pretty happy with that. I think some validation needs to be done on an statistics changes though. Ali 1. Fix O3 CPU bug in SE 40.perlbmk fails 2. Fix O3 processing nacks/coherence messages 3. Better statistics for the caches. 4. FS mode doesn't work under Cygwin 5. memtest regression crashes under Cygwin 6. Make repository public 7. Testing 8. Validation On Aug 24, 2008, at 9:55 AM, Ali Saidi wrote: Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
Re: [m5-dev] Stable Release
Because the machines were busy, which meant that the jobs timed out. Ali On Aug 24, 2008, at 12:45 PM, Gabe Black wrote: We should also figure out why the all regressions failed completely last night. Gabe Black wrote: I'm pretty sure that perlbmk bug is still there and we just got rid of the regression. It might have gone away on it's own, but I don't think anyone actively tried to fix it. Gabe Ali Saidi wrote: From the release notes the following are still outstanding for a 2.0 release. I don't really care about the Cygwin problems and I'm not sure that anyone else does, so I move to strike those. Additionally, we made the repository public so that one is gone. I think the three things that need to be fixed are: O3 CPU -- SMT SE regression hits an assert; bug causing perlbmk to fail (Gabe reported this in April 2007 and I don't know if it's still a problem http://www.m5sim.org/flyspray/task/221); NACKs/coherence messages Statistics -- The CPU statistics are confusing. I think the idle time in the simple CPU can still be wrong after a CPU switch and it's at least not clear to me which idles mean quiesced time or idle because of other things (blocked in the O3 case, not switched out in the simple case); The bus needs statistics; I don't know if the cache statistics are reasonable. Testing -- I've at least done a fair amount of testing over the last month or two, so I'm pretty happy with that. I think some validation needs to be done on an statistics changes though. Ali 1. Fix O3 CPU bug in SE 40.perlbmk fails 2. Fix O3 processing nacks/coherence messages 3. Better statistics for the caches. 4. FS mode doesn't work under Cygwin 5. memtest regression crashes under Cygwin 6. Make repository public 7. Testing 8. Validation On Aug 24, 2008, at 9:55 AM, Ali Saidi wrote: Everyone should hold off pushing patches to the dev tree for the next two weeks so we can create a new stable release. Please only push fixes to known problems and test the repository in the next two weeks. Thanks, Ali ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev ___ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev
