Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 03:55:23PM -0400, Daniel Eischen wrote: David also has patches for debugging support at: http://people.freebsd.org/~davidxu/kse/dbg/ Unless David Xu completes full FSF paper work, we can't use his patches. Using them tants us forever in getting stock GDB to support our threading. -- -- David ([EMAIL PROTECTED]) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 01:48:17PM -0700, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:27:08PM -0600, Scott Long wrote: Doug Rabson also has basic TLS support working in perforce. What platforms? My understanding was that new binutils and gcc was needed for sparc64 at a minimum. Yes. It's i386 only and not even close to being complete. In fact, there has been discussions that the thread pointer on i386 needs to change. Whether that's the case or not, it's likely that TLS will complicate matters way too much to for it to ever work in 5.3. It'd be nice to get TLS and debugging in before 5.3-release. Yes. Be aware that there is a serious effort to get GDB 6.x into the tree for 5.3. I just asked cvs@ and core@ to advice about importing gdb 6.1 into src/contrib/gdb6. We need to break this vicious circle of people waiting on each other. This came up before and you were already asked not to import into src/contrib/gdb6 (and what email would be sent to Core if you did). Any GDB imports need to go into src/contrib/gdb/. -- -- David ([EMAIL PROTECTED]) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 02:12:49PM -0700, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. Wake up, people. This is quickly becoming a joke... Right... dons flak jacket Basically only i386 is a Tier 1. The rest is a joke if it is called Tier 1. If only because there are insufficient numbers of committers actively working on the arch. (maybe pc98 should be called Tier 1, I just don't know enough about that one). -- Wilko Bulte [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, 2004/06/06 at 14:59:21 -0700, Kris Kennaway wrote: On Sun, Jun 06, 2004 at 03:49:13PM -0600, Scott Long wrote: amd64 is approaching critical mass for tier-1. There are a number of developers that own amd64 hardware now, and a number of users who are asking about it on the mailing lists. Peter is finishing up the last blocking item for it (kld's) not including the observed KSE problems. It's very close and I _will_ hold up the release for it to get done. amd64 is the future of commodity computing and we aren't going to ignore it for 5-STABLE. amd64 has a bug with swapping - when something begins to access swap, the entire system becomes almost entirely unresponsive (e.g. no mouse response for up to 10 seconds) until it stops. Peter has some ideas about it, but it's a serious enough bug that it forced me to stop using amd64 as my desktop machine (hello, kde!). Hmmm, I have encountered a similar problem on sparc64 once; the reason was that vm_pageout_map_deactivate_pages() calls pmap_remove() for the range from the start to the end of the process's vm_map when a process is swapped out. Start and end are VM_MIN_ADDRESS and VM_MAXUSER_ADDRESS respectively, and on 64-bit architectures, that range is very large (128TB on ia64 if I'm not mistaken), so the iteration in pmap_remove() must be carefully designed to make as large steps as possible to avoid long run times (or to not iterate over the range at all if it becomes too large, which we did on sparc64). It seems that the amd64 version of pmap_remove() will essentially always iterate in 2MB (level 2 page table) steps, regardless of whether there is mapping for the respective level 2 table in the table levels above; that means that in the previously mentioned case, the outer loop will usually run for about 67 million iterations (the resident count guard may not be of much use here if a stack page is left at the very end of the address space). Since there are a few memory accesses needed in each iterations, that may already be the cause of such a delay. I have no hardware to test this, so all of the above is just a wild- assed guess; but maybe it is of use (and sorry for the spam if it is not). - Thomas -- Thomas Moestl [EMAIL PROTECTED] http://www.tu-bs.de/~y0015675/ [EMAIL PROTECTED] http://people.FreeBSD.org/~tmm/ I realized that the purpose of writing is to inflate weak ideas, obscure poor reasoning, and inhibit clarity. -- Calvin and Hobbes ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Garance A Drosihn wrote: At 3:46 PM -0600 6/6/04, Scott Long wrote: At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. ... As I said back then, demotion is not a terminal condition, and I would be thrilled if someone comes forward in the future and brings the platform back up to date. I think you have to officially demote it, with emphasis on the point that demotion is not a terminal condition. Then, if some developer(s) show up and implement all the missing pieces, we can happily announce it back in tier 1. But for now, say that it *IS* demoted. Not that you're advocating that we think about maybe demoting it in the future unless someone offers to start looking into the missing pieces. At the moment, it probably also makes sense to demote sparc64, even though I own one of those. Not that we have anything against it, but as a practical matter we haven't hit critical mass on it just yet. Since I am interested in sparc64, I can take that as a goal to help make it a tier-1 platform by 5.4-release... One thing to note is that whatever platforms get dropped from tier-1 status will have a high probablility of not getting updated with the upcoming binutils/gcc/gdb update that is coming. Therefore, if we are going to drop a platform, we had better be very serious about it since bringing it back up might be hard. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Mon, Jun 07, 2004 at 09:37:12AM -0600, Scott Long wrote: One thing to note is that whatever platforms get dropped from tier-1 status will have a high probablility of not getting updated with the upcoming binutils/gcc/gdb update that is coming. Logic dictates that the probability should be 0. One cannot reach tier 1 from any tier 1 if binutils, gcc and gdb are not working. Hence, any tier 2 platform must already be supported by binutils, gcc and gdb. -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Monday 07 June 2004 07:33 am, Thomas Moestl wrote: On Sun, 2004/06/06 at 14:59:21 -0700, Kris Kennaway wrote: On Sun, Jun 06, 2004 at 03:49:13PM -0600, Scott Long wrote: amd64 is approaching critical mass for tier-1. There are a number of developers that own amd64 hardware now, and a number of users who are asking about it on the mailing lists. Peter is finishing up the last blocking item for it (kld's) not including the observed KSE problems. It's very close and I _will_ hold up the release for it to get done. amd64 is the future of commodity computing and we aren't going to ignore it for 5-STABLE. amd64 has a bug with swapping - when something begins to access swap, the entire system becomes almost entirely unresponsive (e.g. no mouse response for up to 10 seconds) until it stops. Peter has some ideas about it, but it's a serious enough bug that it forced me to stop using amd64 as my desktop machine (hello, kde!). Hmmm, I have encountered a similar problem on sparc64 once; the reason was that vm_pageout_map_deactivate_pages() calls pmap_remove() for the range from the start to the end of the process's vm_map when a process is swapped out. Start and end are VM_MIN_ADDRESS and VM_MAXUSER_ADDRESS respectively, and on 64-bit architectures, that range is very large (128TB on ia64 if I'm not mistaken), so the iteration in pmap_remove() must be carefully designed to make as large steps as possible to avoid long run times (or to not iterate over the range at all if it becomes too large, which we did on sparc64). It seems that the amd64 version of pmap_remove() will essentially always iterate in 2MB (level 2 page table) steps, regardless of whether there is mapping for the respective level 2 table in the table levels above; that means that in the previously mentioned case, the outer loop will usually run for about 67 million iterations (the resident count guard may not be of much use here if a stack page is left at the very end of the address space). Since there are a few memory accesses needed in each iterations, that may already be the cause of such a delay. You know, this sounds spot-on! Thanks for the tip! -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] All of this is for nothing if we don't go to the stars - JMS/B5 ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
At 9:37 AM -0600 6/7/04, Scott Long wrote: Garance A Drosihn wrote: I think you have to officially demote it, with emphasis on the point that demotion is not a terminal condition. Then, if some developer(s) show up and implement all the missing pieces, we can happily announce it back in tier 1. But for now, say that it *IS* demoted. Not that you're advocating that we think about maybe demoting it in the future unless someone offers to start looking into the missing pieces. One thing to note is that whatever platforms get dropped from tier-1 status will have a high probablility of not getting updated with the upcoming binutils/gcc/gdb update that is coming. If that is the case, then you can not say demotion is not a terminal condition. That sounds pretty terminal to me. -- Garance Alistair Drosehn= [EMAIL PROTECTED] Senior Systems Programmer or [EMAIL PROTECTED] Rensselaer Polytechnic Instituteor [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
HEADS UP! KSE needs more attention
All, We are about 4-6 weeks away from starting the 5.3 release cycle. As it stands, KSE still only works reliably on i386. There are reports of significant instability on amd64, and it doesn't work at all on alpha and sparc64. I'm willing to drop the alpha requirement and maybe even the sparc64 requirement, but there absolutely will not be a 5.3 until amd64 is solid. Please contact myself, Dan Eischen, and David Xu if you are interested in helping out. Thanks, Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 01:14:57PM -0600, Scott Long wrote: All, We are about 4-6 weeks away from starting the 5.3 release cycle. As it stands, KSE still only works reliably on i386. I don't have any problems on ia64. ... I'm willing to drop the alpha requirement and maybe even the sparc64 requirement, but there absolutely will not be a 5.3 until amd64 is solid. I think sparc64 should have KSE too. If we already accept that sparc64 is feature incomplete, we set/acknowledge a really bad precedence. -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, 6 Jun 2004, Scott Long wrote: All, We are about 4-6 weeks away from starting the 5.3 release cycle. As it stands, KSE still only works reliably on i386. There are reports of significant instability on amd64, and it doesn't work at all on alpha and sparc64. I'm willing to drop the alpha requirement and maybe even the sparc64 requirement, but there absolutely will not be a 5.3 until amd64 is solid. Please contact myself, Dan Eischen, and David Xu if you are interested in helping out. amd64 looks to be a problem in readline which doesn't seem to redispatch signal handlers with SA_SIGINFO arguments. David also has patches for debugging support at: http://people.freebsd.org/~davidxu/kse/dbg/ Doug Rabson also has basic TLS support working in perforce. It'd be nice to get TLS and debugging in before 5.3-release. -- Dan Eischen ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Daniel Eischen wrote: On Sun, 6 Jun 2004, Scott Long wrote: All, We are about 4-6 weeks away from starting the 5.3 release cycle. As it stands, KSE still only works reliably on i386. There are reports of significant instability on amd64, and it doesn't work at all on alpha and sparc64. I'm willing to drop the alpha requirement and maybe even the sparc64 requirement, but there absolutely will not be a 5.3 until amd64 is solid. Please contact myself, Dan Eischen, and David Xu if you are interested in helping out. amd64 looks to be a problem in readline which doesn't seem to redispatch signal handlers with SA_SIGINFO arguments. David also has patches for debugging support at: http://people.freebsd.org/~davidxu/kse/dbg/ Doug Rabson also has basic TLS support working in perforce. What platforms? My understanding was that new binutils and gcc was needed for sparc64 at a minimum. It'd be nice to get TLS and debugging in before 5.3-release. Yes. Be aware that there is a serious effort to get GDB 6.x into the tree for 5.3. That presents us with a bit of a dilemma since David's work is against GDB 5.x. It's hard to back off on the GDB 6 requirement since it is needed for amd64 and sparc64. We need to get David, Marcel, David O'brien, and Alexander Kabaev together to work out the combined picture. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 01:14:57PM -0600, Scott Long wrote: All, We are about 4-6 weeks away from starting the 5.3 release cycle. As it stands, KSE still only works reliably on i386. I don't have any problems on ia64. Good to hear =-) ... I'm willing to drop the alpha requirement and maybe even the sparc64 requirement, but there absolutely will not be a 5.3 until amd64 is solid. I think sparc64 should have KSE too. If we already accept that sparc64 is feature incomplete, we set/acknowledge a really bad precedence. I do too, but there is the difficult fact in that there are few people out there that are willing to work on sparc64. One person offered to try to learn it and tackle it, but that's a lot to ask. As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 02:27:08PM -0600, Scott Long wrote: Doug Rabson also has basic TLS support working in perforce. What platforms? My understanding was that new binutils and gcc was needed for sparc64 at a minimum. Yes. It's i386 only and not even close to being complete. In fact, there has been discussions that the thread pointer on i386 needs to change. Whether that's the case or not, it's likely that TLS will complicate matters way too much to for it to ever work in 5.3. It'd be nice to get TLS and debugging in before 5.3-release. Yes. Be aware that there is a serious effort to get GDB 6.x into the tree for 5.3. I just asked cvs@ and core@ to advice about importing gdb 6.1 into src/contrib/gdb6. We need to break this vicious circle of people waiting on each other. That presents us with a bit of a dilemma since David's work is against GDB 5.x. And not even complete. There are still issues that haven't found a solution or even compromise and it too is only for i386. -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. Wake up, people. This is quickly becoming a joke... -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, 6 Jun 2004, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. Not to take away from the tremendous effort that jake had done for sparc64, but it should really take more than one or two supporting developers to obtain tier 1 support. People come and go, and tierness should take that into account. As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. We shouldn't keep an arch at tier 1 just to save face. Better to just lower it to tier 2 and be done with it. My $.02, FWIW. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Hey all, Daniel Eischen wrote: On Sun, 6 Jun 2004, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: Not to take away from the tremendous effort that jake had done for sparc64, but it should really take more than one or two supporting developers to obtain tier 1 support. People come and go, and tierness should take that into account. We shouldn't keep an arch at tier 1 just to save face. Better to just lower it to tier 2 and be done with it. My $.02, FWIW. I agree with Daniel, although i am not developping for freebsd (lack of knowledge). I think that what Daniel says is true, you cannot say that you support a product in a tier 1 status, while you have way to less people to resolve things. In my opinion it's also better to lower it then, and perhaps upgrade it again when more supporters are available to resolve issues. It takes more time to recover from failed support then by honestly saying that you don't have the manpower to support it in a tier 1 branch. Besides that i can understand that this will hit certain people, depending on the sparc64 tier 1 status, but perhaps they can support as well.. We need to resolve this with all of us, and if we cannot find enough people to help , then it should -sadly enough- be degraded in tier status. my $0.02 :) -- Kind regards, Remko Lodder Elvandar.org/DSINet.org www.mostly-harmless.nl Dutch community for helping newcomers on the hackerscene ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. Wake up, people. This is quickly becoming a joke... It's not that there is face to loose on alpha, it's that every time I announce that alpha is going to be killed unless issues X, Y, and Z are fixed, someone comes along and fixes X, part of Y, and promises to fix Z. There is nothing wrong with this, and I definitely appreciate it when people step up to fix things. However, it does prolong the process and doesn't leave us with 100% of what we need. At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. I made it very clear last winter that alpha needed a full time maintainer in order to stay viable, and that really never happened. As I said back then, demotion is not a terminal condition, and I would be thrilled if someone comes forward in the future and brings the platform back up to date. If anyone wants to claim it now and keep it alive for 5.3, they need to both finish KSE, make KSE work reliably, and be very responsive to Kris Kenneway about ports issues. This needs to happen in no more than 4 weeks. After that, I will turn away even the best of intentions. Anyways, moving on, KSE needs attention. Please figure out what can be done for sparc64 and amd64 and do it. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Wilko Bulte wrote: On Sun, Jun 06, 2004 at 02:12:49PM -0700, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. Wake up, people. This is quickly becoming a joke... Right... dons flak jacket Basically only i386 is a Tier 1. The rest is a joke if it is called Tier 1. If only because there are insufficient numbers of committers actively working on the arch. (maybe pc98 should be called Tier 1, I just don't know enough about that one). amd64 is approaching critical mass for tier-1. There are a number of developers that own amd64 hardware now, and a number of users who are asking about it on the mailing lists. Peter is finishing up the last blocking item for it (kld's) not including the observed KSE problems. It's very close and I _will_ hold up the release for it to get done. amd64 is the future of commodity computing and we aren't going to ignore it for 5-STABLE. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On 6 Jun, Daniel Eischen wrote: On Sun, 6 Jun 2004, Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 02:31:56PM -0600, Scott Long wrote: As with Alpha, the fate of a platform rests on the people who are willing to work on it, not on whether it is in a particular list. Agreed, but it's the projects responsibility to take the tierness and the intend to support multiple platforms serious and not to chicken out at the first signs of complications or hurdles. We labeled sparc64 as a tier 1 platform and we better deal with the consequences. Not to take away from the tremendous effort that jake had done for sparc64, but it should really take more than one or two supporting developers to obtain tier 1 support. People come and go, and tierness should take that into account. I've got some sparc64 hardware that recently became available for FreeBSD develpment. Unfortunately my time available to FreeBSD is likely to be the limiting factor. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 03:49:13PM -0600, Scott Long wrote: amd64 is approaching critical mass for tier-1. There are a number of developers that own amd64 hardware now, and a number of users who are asking about it on the mailing lists. Peter is finishing up the last blocking item for it (kld's) not including the observed KSE problems. It's very close and I _will_ hold up the release for it to get done. amd64 is the future of commodity computing and we aren't going to ignore it for 5-STABLE. amd64 has a bug with swapping - when something begins to access swap, the entire system becomes almost entirely unresponsive (e.g. no mouse response for up to 10 seconds) until it stops. Peter has some ideas about it, but it's a serious enough bug that it forced me to stop using amd64 as my desktop machine (hello, kde!). Kris pgpOd3IUDPnRF.pgp Description: PGP signature
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 05:27:56PM -0400, Daniel Eischen wrote: As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. We shouldn't keep an arch at tier 1 just to save face. Better to just lower it to tier 2 and be done with it. My $.02, FWIW. You misunderstand my statement. Lowering alpha to tier 2 is what I suggested multiple times before. The point is that we haven't yet and alpha is degenerating more and more while we fail to adjust the tierness. That's where we fail to save face. -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sun, Jun 06, 2004 at 03:46:44PM -0600, Scott Long wrote: As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. It's not that there is face to loose on alpha, it's that every time I announce that alpha is going to be killed unless issues X, Y, and Z are fixed, someone comes along and fixes X, part of Y, and promises to fix Z. I don't read anything I haven't read before. People have made promises before and good intentions notwithstanding, you cannot keep a project hostage this way. Degrade alpha and only promote it after people have actually demonstrated that alpha is tier 1 material after all. At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. Just *do* it. You've been advocating for way too long. -- Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Marcel Moolenaar wrote: On Sun, Jun 06, 2004 at 03:46:44PM -0600, Scott Long wrote: As for alpha, we don't even seem to be able to degrade it to tier 2 without losing face. kris@ has already stopped package builds for it for his own sake. It's not that there is face to loose on alpha, it's that every time I announce that alpha is going to be killed unless issues X, Y, and Z are fixed, someone comes along and fixes X, part of Y, and promises to fix Z. I don't read anything I haven't read before. People have made promises before and good intentions notwithstanding, you cannot keep a project hostage this way. Degrade alpha and only promote it after people have actually demonstrated that alpha is tier 1 material after all. At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. Just *do* it. You've been advocating for way too long. There are always people with good intentions, and they often want to find some way to help. I appreciate this very much and I don't want to turn them away. We also can't make a snap decision in an afternoon and risk missing people who have better things to do on their weekends/holidays than track FreeBSD mail =-) But yes, Alpha has been languishing for too long. I'll say it in a loud voice now: This is the last call for Alpha. It cannot maintain minimum tier-1 requirements and it will be demoted for 5.3. Unless you currently have access to a representative selection of hardware, the knowledge to track down platform-specific issues, the willingness to handle the current outstanding issues, and the time to do it for the next 18 months, please accept my regrets on it. Again, while I appreciate people who can pick away at some of the lesser issues or want to learn the bigger things, we need a 100% maintenance and development solution. Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
At 3:46 PM -0600 6/6/04, Scott Long wrote: At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. ... As I said back then, demotion is not a terminal condition, and I would be thrilled if someone comes forward in the future and brings the platform back up to date. I think you have to officially demote it, with emphasis on the point that demotion is not a terminal condition. Then, if some developer(s) show up and implement all the missing pieces, we can happily announce it back in tier 1. But for now, say that it *IS* demoted. Not that you're advocating that we think about maybe demoting it in the future unless someone offers to start looking into the missing pieces. At the moment, it probably also makes sense to demote sparc64, even though I own one of those. Not that we have anything against it, but as a practical matter we haven't hit critical mass on it just yet. Since I am interested in sparc64, I can take that as a goal to help make it a tier-1 platform by 5.4-release... -- Garance Alistair Drosehn= [EMAIL PROTECTED] Senior Systems Programmer or [EMAIL PROTECTED] Rensselaer Polytechnic Instituteor [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
Garance A Drosihn wrote: At 3:46 PM -0600 6/6/04, Scott Long wrote: At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. ... As I said back then, demotion is not a terminal condition, and I would be thrilled if someone comes forward in the future and brings the platform back up to date. I think you have to officially demote it, with emphasis on the point that demotion is not a terminal condition. Then, if some developer(s) show up and implement all the missing pieces, we can happily announce it back in tier 1. But for now, say that it *IS* demoted. Not that you're advocating that we think about maybe demoting it in the future unless someone offers to start looking into the missing pieces. At the moment, it probably also makes sense to demote sparc64, even though I own one of those. Not that we have anything against it, but as a practical matter we haven't hit critical mass on it just yet. Since I am interested in sparc64, I can take that as a goal to help make it a tier-1 platform by 5.4-release... Ok, thanks for all of the input. I'm going to wait a few more days for anyone else to chime in and then discuss it with re@ and [EMAIL PROTECTED] We'll likely have a formal announcement in a week. Back to the topic at hand, who is looking that the other KSE issues? Scott ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Mon, Jun 07, 2004 at 02:32:11AM +0200, Matthias Andree wrote:
Marcel Moolenaar [EMAIL PROTECTED] writes:
As for alpha, we don't even seem to be able to degrade it to tier 2
without losing face. kris@ has already stopped package builds for it
for his own sake.
Alpha is special, with what seems to me like a GDB bug. Try this:
echo '#include stdlib.h
int main() {abort();}' abortme.c
gcc -O2 -o abortme abortme.c
./abortme
gdb ./abortme ./core.abortme
(inside gdb:) backtrace
(inside gdb:) backtrace full
(inside gdb:) quit
This stuff is run as part of the ports/mail/bogofilter test suite (which
is part of the build) determine if core dumps from the build logs are
usable.
A couple of days ago, the backtrace of the trivial program ended up in
an unterminated loop on the build cluster, GDB kept repeating stack
frame #0. Whoops. The other architectures appeared fine though.
This is fixed in gdb 6.x:
GNU gdb 6.1.0.90_20040413 [FreeBSD]
Copyright 2004 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type show copying to see the conditions.
There is absolutely no warranty for GDB. Type show warranty for details.
This GDB was configured as alpha-intree-freebsd...
Core was generated by `abortme'.
Program terminated with signal 6, Aborted.
Reading symbols from /lib/libc.so.5...done.
Loaded symbols for /lib/libc.so.5
Reading symbols from /libexec/ld-elf.so.1...done.
Loaded symbols for /libexec/ld-elf.so.1
#0 0x0001601b29bc in kill () from /lib/libc.so.5
(gdb) l
1 #include stdlib.h
2 int main()
3 {
4 abort();
5 }
(gdb) bt
#0 0x0001601b29bc in kill () from /lib/libc.so.5
#1 0x0001601a5298 in raise () from /lib/libc.so.5
#2 0x000160233f88 in abort () from /lib/libc.so.5
#3 0x000128a0 in main () at abortme.c:4
(gdb)
An import of gdb 6.1 or gdb 6.1.1 will resolve this.
--
Marcel Moolenaar USPA: A-39004 [EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
In message: [EMAIL PROTECTED] Marcel Moolenaar [EMAIL PROTECTED] writes: : On Sun, Jun 06, 2004 at 05:27:56PM -0400, Daniel Eischen wrote: : : As for alpha, we don't even seem to be able to degrade it to tier 2 : without losing face. kris@ has already stopped package builds for it : for his own sake. : : We shouldn't keep an arch at tier 1 just to save face. Better to just : lower it to tier 2 and be done with it. My $.02, FWIW. : : You misunderstand my statement. Lowering alpha to tier 2 is what I : suggested multiple times before. The point is that we haven't yet : and alpha is degenerating more and more while we fail to adjust : the tierness. That's where we fail to save face. I tend to agree. Tierness is a combination of politics and technical reality. The rality of the situation with alpha is that it has had no clothes long enough that it no longer reflects the Tier-1 ideas that we strive to attain. Warner ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: HEADS UP! KSE needs more attention
On Sunday 06 June 2004 16:49, Scott Long wrote: Garance A Drosihn wrote: At 3:46 PM -0600 6/6/04, Scott Long wrote: At this point, I'm going to advocate that Alpha be dropped from Tier-1 status for 5.3 and 5-STABLE and no longer be a blocking item for releases. ... As I said back then, demotion is not a terminal condition, and I would be thrilled if someone comes forward in the future and brings the platform back up to date. I think you have to officially demote it, with emphasis on the point that demotion is not a terminal condition. Then, if some developer(s) show up and implement all the missing pieces, we can happily announce it back in tier 1. But for now, say that it *IS* demoted. Not that you're advocating that we think about maybe demoting it in the future unless someone offers to start looking into the missing pieces. At the moment, it probably also makes sense to demote sparc64, even though I own one of those. Not that we have anything against it, but as a practical matter we haven't hit critical mass on it just yet. Since I am interested in sparc64, I can take that as a goal to help make it a tier-1 platform by 5.4-release... Ok, thanks for all of the input. I'm going to wait a few more days for anyone else to chime in and then discuss it with re@ and [EMAIL PROTECTED] We'll likely have a formal announcement in a week. Back to the topic at hand, who is looking that the other KSE issues? The tierness of a particular platform wasn't all that important as long as FreeBSD 5 remained a development branch, but with 5-STABLE rapidly approaching, it will become much more important. When we first documented the tier approach, we selected sparc64 as the reference 64-bit platform, because it was at that time the most viable and the most actively developed platform. This was shortly before the Athlon64 release, which has turned into an avalanche. None of the current tiers are set in stone, in fact, they're not really very firm at all. The stability and completeness of each will have to be evaluated as the 5.3 release becomes real and adjustments made. It wouldn't surprise me to see a new 64-bit reference platform chosen; the momentum in that arena has definitely shifted. As an aside, I would sadden me greatly to see sparc64 or alpha abandoned. Even sadder (to me) is how little I've been able to contribute to either. I don't have time to bring my sparc skills up to working on FreeBSD internals and I'm not likely to suddenly grow more time soon. I do have an Ultra 5 workstation that is free to any developer in North America who will use it to further FreeBSD development. I'll contact dlo@ with details. sparc64 and alpha need a few champions. Without them, these ports will quickly be relegated to the bit-bucket. At any rate, FreeBSD will continue to move, in a direction we've decided to call forward. -- Where am I, and what am I doing in this handbasket? Wes Peters [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to [EMAIL PROTECTED]
