* New Debian Kernel Packages @ 2013-07-29 22:46 John David Anglin 2013-08-01 22:20 ` Helge Deller 0 siblings, 1 reply; 7+ messages in thread From: John David Anglin @ 2013-07-29 22:46 UTC (permalink / raw) To: Parisc List The following kernel packages are now in the parisc-linux.org archive: -rw-r--r-- 1 deller ftpadmin 19931220 Jul 29 22:01 linux- image-3.10-1-parisc-smp_3.10.3-1_hppa.deb -rw-r--r-- 1 deller ftpadmin 21374850 Jul 29 22:01 linux- image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb -rw-r--r-- 1 deller ftpadmin 20766562 Jul 29 22:01 linux- image-3.10-1-parisc64_3.10.3-1_hppa.deb -rw-r--r-- 1 deller ftpadmin 19290152 Jul 29 22:01 linux- image-3.10-1-parisc_3.10.3-1_hppa.deb Dave -- John David Anglin dave.anglin@bell.net ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-07-29 22:46 New Debian Kernel Packages John David Anglin @ 2013-08-01 22:20 ` Helge Deller 2013-08-01 22:39 ` John David Anglin 0 siblings, 1 reply; 7+ messages in thread From: Helge Deller @ 2013-08-01 22:20 UTC (permalink / raw) To: John David Anglin; +Cc: Parisc List Hi Dave, On 07/30/2013 12:46 AM, John David Anglin wrote: > The following kernel packages are now in the parisc-linux.org > archive: > > linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb > linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb > linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb > linux-image-3.10-1-parisc_3.10.3-1_hppa.deb That's fantastic! This brings us one big step further to being able to build a real debian-unstable-installer boot CD. I'm still hoping the C8000 patches I pushed for inclusion into 3.11 will then show up in stable 3.10 series soon as well. If that happens we will have a kernel which should boot on all machines - including the c8000. Helge ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-08-01 22:20 ` Helge Deller @ 2013-08-01 22:39 ` John David Anglin 2013-08-01 22:52 ` James Bottomley 0 siblings, 1 reply; 7+ messages in thread From: John David Anglin @ 2013-08-01 22:39 UTC (permalink / raw) To: Helge Deller; +Cc: Parisc List On 1-Aug-13, at 6:20 PM, Helge Deller wrote: > Hi Dave, > > On 07/30/2013 12:46 AM, John David Anglin wrote: >> The following kernel packages are now in the parisc-linux.org >> archive: >> >> linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb >> linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb >> linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb >> linux-image-3.10-1-parisc_3.10.3-1_hppa.deb > > That's fantastic! > This brings us one big step further to being able to build a real > debian-unstable-installer boot CD. > > I'm still hoping the C8000 patches I pushed for inclusion into 3.11 > will then > show up in stable 3.10 series soon as well. If that happens we will > have a kernel > which should boot on all machines - including the c8000. I have bootstrap tested the linux-image-3.10-1-parisc64- smp_3.10.3-1_hppa.deb package. We'll have to see if 3.10 will be the Debian choice for the next release. As I mentioned in private, I have found that flush_cache_all() is the principal problem causing random segmentation faults on SMP systems. Unfortunately, the only fix so far is to not use flush_cache_all() in flush_cache_mm() and flush_cache_range(), and this really slows things down running GCC testsuite. It doesn't seem to hurt much for other stuff. My thought that this might be a timing issue didn't work. I tried masking all external interrupts except for IPI on all processors first, but it didn't help. It may be this can be narrowed somewhat. I suspect that we have an inconsistent cache state during fork. I think if this issue is fixed, c8000's will be worth a bit more than 100 Euros... Dave -- John David Anglin dave.anglin@bell.net ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-08-01 22:39 ` John David Anglin @ 2013-08-01 22:52 ` James Bottomley 2013-08-01 23:29 ` John David Anglin 2013-08-14 23:43 ` John David Anglin 0 siblings, 2 replies; 7+ messages in thread From: James Bottomley @ 2013-08-01 22:52 UTC (permalink / raw) To: John David Anglin; +Cc: Helge Deller, Parisc List On Thu, 2013-08-01 at 18:39 -0400, John David Anglin wrote: > On 1-Aug-13, at 6:20 PM, Helge Deller wrote: > > > Hi Dave, > > > > On 07/30/2013 12:46 AM, John David Anglin wrote: > >> The following kernel packages are now in the parisc-linux.org > >> archive: > >> > >> linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb > >> linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb > >> linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb > >> linux-image-3.10-1-parisc_3.10.3-1_hppa.deb > > > > That's fantastic! > > This brings us one big step further to being able to build a real > > debian-unstable-installer boot CD. > > > > I'm still hoping the C8000 patches I pushed for inclusion into 3.11 > > will then > > show up in stable 3.10 series soon as well. If that happens we will > > have a kernel > > which should boot on all machines - including the c8000. > > > I have bootstrap tested the linux-image-3.10-1-parisc64- > smp_3.10.3-1_hppa.deb package. We'll > have to see if 3.10 will be the Debian choice for the next release. > > As I mentioned in private, I have found that flush_cache_all() is the > principal problem causing random > segmentation faults on SMP systems. That would tend to indicate the architectural flush code is wrong: it isn't flushing all as it should be. James ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-08-01 22:52 ` James Bottomley @ 2013-08-01 23:29 ` John David Anglin 2013-08-14 23:43 ` John David Anglin 1 sibling, 0 replies; 7+ messages in thread From: John David Anglin @ 2013-08-01 23:29 UTC (permalink / raw) To: James Bottomley; +Cc: Helge Deller, Parisc List On 1-Aug-13, at 6:52 PM, James Bottomley wrote: > On Thu, 2013-08-01 at 18:39 -0400, John David Anglin wrote: >> On 1-Aug-13, at 6:20 PM, Helge Deller wrote: >> >>> Hi Dave, >>> >>> On 07/30/2013 12:46 AM, John David Anglin wrote: >>>> The following kernel packages are now in the parisc-linux.org >>>> archive: >>>> >>>> linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc_3.10.3-1_hppa.deb >>> >>> That's fantastic! >>> This brings us one big step further to being able to build a real >>> debian-unstable-installer boot CD. >>> >>> I'm still hoping the C8000 patches I pushed for inclusion into 3.11 >>> will then >>> show up in stable 3.10 series soon as well. If that happens we will >>> have a kernel >>> which should boot on all machines - including the c8000. >> >> >> I have bootstrap tested the linux-image-3.10-1-parisc64- >> smp_3.10.3-1_hppa.deb package. We'll >> have to see if 3.10 will be the Debian choice for the next release. >> >> As I mentioned in private, I have found that flush_cache_all() is the >> principal problem causing random >> segmentation faults on SMP systems. > > That would tend to indicate the architectural flush code is wrong: it > isn't flushing all as it should be. That thought had occurred to me. It's also possible that it doesn't work properly with the L3 added in the PA8800/PA8900 systems. What seems to work is flushing all the user mappings. I should note that I have also seen corruption of space register %sr1 (added checks in the routines that use pa_memcpy). I think the only way that can happen is kernel stack corruption. Probably, this is also caused by flush_cache_all. I know this issue has been around for a long time. I'll look at the flush code again. Dave -- John David Anglin dave.anglin@bell.net ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-08-01 22:52 ` James Bottomley 2013-08-01 23:29 ` John David Anglin @ 2013-08-14 23:43 ` John David Anglin 2013-08-15 0:38 ` James Bottomley 1 sibling, 1 reply; 7+ messages in thread From: John David Anglin @ 2013-08-14 23:43 UTC (permalink / raw) To: James Bottomley; +Cc: Helge Deller, Parisc List On 1-Aug-13, at 6:52 PM, James Bottomley wrote: > On Thu, 2013-08-01 at 18:39 -0400, John David Anglin wrote: >> On 1-Aug-13, at 6:20 PM, Helge Deller wrote: >> >>> Hi Dave, >>> >>> On 07/30/2013 12:46 AM, John David Anglin wrote: >>>> The following kernel packages are now in the parisc-linux.org >>>> archive: >>>> >>>> linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb >>>> linux-image-3.10-1-parisc_3.10.3-1_hppa.deb >>> >>> That's fantastic! >>> This brings us one big step further to being able to build a real >>> debian-unstable-installer boot CD. >>> >>> I'm still hoping the C8000 patches I pushed for inclusion into 3.11 >>> will then >>> show up in stable 3.10 series soon as well. If that happens we will >>> have a kernel >>> which should boot on all machines - including the c8000. >> >> >> I have bootstrap tested the linux-image-3.10-1-parisc64- >> smp_3.10.3-1_hppa.deb package. We'll >> have to see if 3.10 will be the Debian choice for the next release. >> >> As I mentioned in private, I have found that flush_cache_all() is the >> principal problem causing random >> segmentation faults on SMP systems. > > That would tend to indicate the architectural flush code is wrong: it > isn't flushing all as it should be. After looking at this, I think there is an irq problem. It looks like we loose IPI interrupts on occasion, or there is a sequencing issue in processing them. In particular, it would be bad if we simultaneously tried to purge both theTLB and cache at the same time. However, this isn't the whole answer. Dave -- John David Anglin dave.anglin@bell.net ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: New Debian Kernel Packages 2013-08-14 23:43 ` John David Anglin @ 2013-08-15 0:38 ` James Bottomley 0 siblings, 0 replies; 7+ messages in thread From: James Bottomley @ 2013-08-15 0:38 UTC (permalink / raw) To: John David Anglin; +Cc: Helge Deller, Parisc List On Wed, 2013-08-14 at 19:43 -0400, John David Anglin wrote: > On 1-Aug-13, at 6:52 PM, James Bottomley wrote: > > > On Thu, 2013-08-01 at 18:39 -0400, John David Anglin wrote: > >> On 1-Aug-13, at 6:20 PM, Helge Deller wrote: > >> > >>> Hi Dave, > >>> > >>> On 07/30/2013 12:46 AM, John David Anglin wrote: > >>>> The following kernel packages are now in the parisc-linux.org > >>>> archive: > >>>> > >>>> linux-image-3.10-1-parisc-smp_3.10.3-1_hppa.deb > >>>> linux-image-3.10-1-parisc64-smp_3.10.3-1_hppa.deb > >>>> linux-image-3.10-1-parisc64_3.10.3-1_hppa.deb > >>>> linux-image-3.10-1-parisc_3.10.3-1_hppa.deb > >>> > >>> That's fantastic! > >>> This brings us one big step further to being able to build a real > >>> debian-unstable-installer boot CD. > >>> > >>> I'm still hoping the C8000 patches I pushed for inclusion into 3.11 > >>> will then > >>> show up in stable 3.10 series soon as well. If that happens we will > >>> have a kernel > >>> which should boot on all machines - including the c8000. > >> > >> > >> I have bootstrap tested the linux-image-3.10-1-parisc64- > >> smp_3.10.3-1_hppa.deb package. We'll > >> have to see if 3.10 will be the Debian choice for the next release. > >> > >> As I mentioned in private, I have found that flush_cache_all() is the > >> principal problem causing random > >> segmentation faults on SMP systems. > > > > That would tend to indicate the architectural flush code is wrong: it > > isn't flushing all as it should be. > > > After looking at this, I think there is an irq problem. It looks like > we loose IPI > interrupts on occasion, I don't really think this can be the case. Our flushes are all done via the generic function on_each_cpu() with wait set to true. If we were losing IPIs, this would cause the wait to wait forever and you should see a system hang (or at least one CPU spinning in cpu_relax while it waits). on_each_cpu() uses a list to process and store invocations, so even two simultaneous calls to the flush functions should be strictly sequenced on each cpu. > or there is a sequencing issue in processing > them. > In particular, it would be bad if we simultaneously tried to purge > both theTLB > and cache at the same time. However, this isn't the whole answer. I can't really see how: the architectural TLB flush will, of course, purge TLBs in the tmpalias region, but they'll just refill on the next access for a flush, so I could see simultaneous TLB flush and cache flush slowing each other down, but I can't see a problem arising, unless I've missed something? James ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2013-08-15 0:38 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2013-07-29 22:46 New Debian Kernel Packages John David Anglin 2013-08-01 22:20 ` Helge Deller 2013-08-01 22:39 ` John David Anglin 2013-08-01 22:52 ` James Bottomley 2013-08-01 23:29 ` John David Anglin 2013-08-14 23:43 ` John David Anglin 2013-08-15 0:38 ` James Bottomley
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.