* [Adeos-main] port 2.6.19 x86_64 to 2.6.20 @ 2007-02-25 23:50 trem 2007-02-26 5:56 ` Philippe Gerum 0 siblings, 1 reply; 8+ messages in thread From: trem @ 2007-02-25 23:50 UTC (permalink / raw) To: adeos-main Hi I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find it here : http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't tested it yet (I mean run kernel). I hope this patch is good enough to be used. Otherwise, I'd be pleased to fix it (maybe with a bit of help). regards, trem ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-02-25 23:50 [Adeos-main] port 2.6.19 x86_64 to 2.6.20 trem @ 2007-02-26 5:56 ` Philippe Gerum 2007-03-03 22:24 ` trem 0 siblings, 1 reply; 8+ messages in thread From: Philippe Gerum @ 2007-02-26 5:56 UTC (permalink / raw) To: trem; +Cc: adeos-main On Mon, 2007-02-26 at 00:50 +0100, trem wrote: > Hi > > I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find > it here : > http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch The good news is that moving to 2.6.20 does not seem to be too much of a hell. Right? > > I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't > tested it yet (I mean run kernel). > > I hope this patch is good enough to be used. Otherwise, I'd be pleased > to fix it (maybe with a bit of help). > We are hopefully close to have a reasonably good patch for 2.6.19. Once we achieve this, porting to 2.6.20 should be the next step. Thanks. -- Philippe. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-02-26 5:56 ` Philippe Gerum @ 2007-03-03 22:24 ` trem 2007-03-03 22:58 ` Philippe Gerum 0 siblings, 1 reply; 8+ messages in thread From: trem @ 2007-03-03 22:24 UTC (permalink / raw) To: rpm; +Cc: adeos-main Philippe Gerum wrote: > On Mon, 2007-02-26 at 00:50 +0100, trem wrote: > >> Hi >> >> I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find >> it here : >> http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch >> > > The good news is that moving to 2.6.20 does not seem to be too much of a > hell. Right? > > Yes, there are few conflict so it's quite easy to port xenomai to 2.6.20. >> I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't >> tested it yet (I mean run kernel). >> >> I hope this patch is good enough to be used. Otherwise, I'd be pleased >> to fix it (maybe with a bit of help). >> >> > > We are hopefully close to have a reasonably good patch for 2.6.19. Once > we achieve this, porting to 2.6.20 should be the next step. Thanks. > > I've seen ths in the Changelog : "this closes the implementation phase of the x86_64 port." So I suppose that the patch for 2.6.19 is finished. I've tried it and it works fine. The result of latency is : [trem@domain.hid bin]$ sudo ./latency -T 10 == Sampling period: 100 us == Test mode: periodic user-mode task == All results in microseconds warming up... RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat worst RTD| 0.760| 1.073| 3.660| 0| 0.760| 3.660 RTD| 0.737| 1.063| 3.012| 0| 0.737| 3.660 RTD| 0.741| 1.083| 3.396| 0| 0.737| 3.660 RTD| 0.755| 1.088| 2.344| 0| 0.737| 3.660 RTD| 0.754| 1.093| 3.431| 0| 0.737| 3.660 RTD| 0.755| 1.089| 5.119| 0| 0.737| 5.119 RTD| 0.724| 1.091| 3.126| 0| 0.724| 5.119 RTD| 0.762| 1.089| 3.171| 0| 0.724| 5.119 RTD| 0.751| 1.097| 5.508| 0| 0.724| 5.508 ---|------------|------------|------------|--------|------------------------- RTS| 0.724| 1.085| 5.508| 0| 00:00:10/00:00:10 If I don't mistake, I see a max latency of 5us. Is it realistic ? Now, the port to 2.6.20 is open ? we can work on ? trem ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-03-03 22:24 ` trem @ 2007-03-03 22:58 ` Philippe Gerum 2007-03-04 17:32 ` trem 0 siblings, 1 reply; 8+ messages in thread From: Philippe Gerum @ 2007-03-03 22:58 UTC (permalink / raw) To: trem; +Cc: adeos-main On Sat, 2007-03-03 at 23:24 +0100, trem wrote: > Philippe Gerum wrote: > > On Mon, 2007-02-26 at 00:50 +0100, trem wrote: > > > >> Hi > >> > >> I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find > >> it here : > >> http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch > >> > > > > The good news is that moving to 2.6.20 does not seem to be too much of a > > hell. Right? > > > > > Yes, there are few conflict so it's quite easy to port xenomai to 2.6.20. > >> I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't > >> tested it yet (I mean run kernel). > >> > >> I hope this patch is good enough to be used. Otherwise, I'd be pleased > >> to fix it (maybe with a bit of help). > >> > >> > > > > We are hopefully close to have a reasonably good patch for 2.6.19. Once > > we achieve this, porting to 2.6.20 should be the next step. Thanks. > > > > > I've seen ths in the Changelog : "this closes the implementation phase > of the x86_64 port." > So I suppose that the patch for 2.6.19 is finished. I've tried it and it > works fine. > The result of latency is : > > [trem@domain.hid bin]$ sudo ./latency -T 10 > == Sampling period: 100 us > == Test mode: periodic user-mode task > == All results in microseconds > warming up... > RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) > RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat > worst > RTD| 0.760| 1.073| 3.660| 0| 0.760| > 3.660 > RTD| 0.737| 1.063| 3.012| 0| 0.737| > 3.660 > RTD| 0.741| 1.083| 3.396| 0| 0.737| > 3.660 > RTD| 0.755| 1.088| 2.344| 0| 0.737| > 3.660 > RTD| 0.754| 1.093| 3.431| 0| 0.737| > 3.660 > RTD| 0.755| 1.089| 5.119| 0| 0.737| > 5.119 > RTD| 0.724| 1.091| 3.126| 0| 0.724| > 5.119 > RTD| 0.762| 1.089| 3.171| 0| 0.724| > 5.119 > RTD| 0.751| 1.097| 5.508| 0| 0.724| > 5.508 > ---|------------|------------|------------|--------|------------------------- > RTS| 0.724| 1.085| 5.508| 0| 00:00:10/00:00:10 > > If I don't mistake, I see a max latency of 5us. Is it realistic ? > Yes. X-related problems Paul told us about being put aside (likely MTRR issues, still to be chased), I see < 10 us under SMP test load on a 2 x dual core Opteron 285 here (test load meaning two parallel kernel compilations (-j5), and some significant network load). We still have to benchmark this port more exhaustively and aggressively (e.g. different I/O loads and severe cache trashing, for longer than 8 hours), but the foundations look sane, especially since the Xenomai/x86_64 port currently runs with a default latency calibration set to 500 ns over this. > Now, the port to 2.6.20 is open ? we can work on ? > Yep. Have fun. > > trem > > > -- Philippe. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-03-03 22:58 ` Philippe Gerum @ 2007-03-04 17:32 ` trem 2007-03-07 23:31 ` Philippe Gerum 2007-03-25 17:19 ` Philippe Gerum 0 siblings, 2 replies; 8+ messages in thread From: trem @ 2007-03-04 17:32 UTC (permalink / raw) To: rpm; +Cc: adeos-main Hi As I've got your start now, I've ported (at least tried) the 2.6.19 x86_64 adeos patch to 2.6.20. I've tried it and it seems to works fine, here the latency : [trem@domain.hid ~]$ cat latency_xenomai_2_6_20.txt [trem@domain.hid bin]$ sudo ./latency -T 10 == Sampling period: 100 us == Test mode: periodic user-mode task == All results in microseconds warming up... RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat worst RTD| 0.737| 1.067| 3.332| 0| 0.737| 3.332 RTD| 0.744| 1.067| 3.568| 0| 0.737| 3.568 RTD| 0.724| 1.080| 3.362| 0| 0.724| 3.568 RTD| 0.730| 1.070| 2.900| 0| 0.724| 3.568 RTD| 0.711| 1.071| 3.441| 0| 0.711| 3.568 RTD| 0.733| 1.066| 3.547| 0| 0.711| 3.568 RTD| 0.735| 1.079| 4.523| 0| 0.711| 4.523 RTD| 0.716| 1.072| 3.307| 0| 0.711| 4.523 RTD| 0.715| 1.066| 3.507| 0| 0.711| 4.523 ---|------------|------------|------------|--------|------------------------- RTS| 0.711| 1.071| 4.523| 0| 00:00:10/00:00:10 The patch (2.6.20 x86_64) can be found here : http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-07-trem.patch I hope it could be used. regards, trem Philippe Gerum wrote: > On Sat, 2007-03-03 at 23:24 +0100, trem wrote: > >> Philippe Gerum wrote: >> >>> On Mon, 2007-02-26 at 00:50 +0100, trem wrote: >>> >>> >>>> Hi >>>> >>>> I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find >>>> it here : >>>> http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch >>>> >>>> >>> The good news is that moving to 2.6.20 does not seem to be too much of a >>> hell. Right? >>> >>> >>> >> Yes, there are few conflict so it's quite easy to port xenomai to 2.6.20. >> >>>> I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't >>>> tested it yet (I mean run kernel). >>>> >>>> I hope this patch is good enough to be used. Otherwise, I'd be pleased >>>> to fix it (maybe with a bit of help). >>>> >>>> >>>> >>> We are hopefully close to have a reasonably good patch for 2.6.19. Once >>> we achieve this, porting to 2.6.20 should be the next step. Thanks. >>> >>> >>> >> I've seen ths in the Changelog : "this closes the implementation phase >> of the x86_64 port." >> So I suppose that the patch for 2.6.19 is finished. I've tried it and it >> works fine. >> The result of latency is : >> >> [trem@domain.hid bin]$ sudo ./latency -T 10 >> == Sampling period: 100 us >> == Test mode: periodic user-mode task >> == All results in microseconds >> warming up... >> RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) >> RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat >> worst >> RTD| 0.760| 1.073| 3.660| 0| 0.760| >> 3.660 >> RTD| 0.737| 1.063| 3.012| 0| 0.737| >> 3.660 >> RTD| 0.741| 1.083| 3.396| 0| 0.737| >> 3.660 >> RTD| 0.755| 1.088| 2.344| 0| 0.737| >> 3.660 >> RTD| 0.754| 1.093| 3.431| 0| 0.737| >> 3.660 >> RTD| 0.755| 1.089| 5.119| 0| 0.737| >> 5.119 >> RTD| 0.724| 1.091| 3.126| 0| 0.724| >> 5.119 >> RTD| 0.762| 1.089| 3.171| 0| 0.724| >> 5.119 >> RTD| 0.751| 1.097| 5.508| 0| 0.724| >> 5.508 >> ---|------------|------------|------------|--------|------------------------- >> RTS| 0.724| 1.085| 5.508| 0| 00:00:10/00:00:10 >> >> If I don't mistake, I see a max latency of 5us. Is it realistic ? >> >> > > Yes. X-related problems Paul told us about being put aside (likely MTRR > issues, still to be chased), I see < 10 us under SMP test load on a 2 x > dual core Opteron 285 here (test load meaning two parallel kernel > compilations (-j5), and some significant network load). > > We still have to benchmark this port more exhaustively and aggressively > (e.g. different I/O loads and severe cache trashing, for longer than 8 > hours), but the foundations look sane, especially since the > Xenomai/x86_64 port currently runs with a default latency calibration > set to 500 ns over this. > > >> Now, the port to 2.6.20 is open ? we can work on ? >> >> > > Yep. Have fun. > > >> trem >> >> >> >> ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-03-04 17:32 ` trem @ 2007-03-07 23:31 ` Philippe Gerum 2007-03-25 17:19 ` Philippe Gerum 1 sibling, 0 replies; 8+ messages in thread From: Philippe Gerum @ 2007-03-07 23:31 UTC (permalink / raw) To: trem; +Cc: adeos-main On Sun, 2007-03-04 at 18:32 +0100, trem wrote: > Hi > > As I've got your start now, I've ported (at least tried) the 2.6.19 > x86_64 adeos patch to 2.6.20. > I've tried it and it seems to works fine, here the latency : > > [trem@domain.hid ~]$ cat latency_xenomai_2_6_20.txt > [trem@domain.hid bin]$ sudo ./latency -T 10 > == Sampling period: 100 us > == Test mode: periodic user-mode task > == All results in microseconds > warming up... > RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) > RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat > worst > RTD| 0.737| 1.067| 3.332| 0| 0.737| > 3.332 > RTD| 0.744| 1.067| 3.568| 0| 0.737| > 3.568 > RTD| 0.724| 1.080| 3.362| 0| 0.724| > 3.568 > RTD| 0.730| 1.070| 2.900| 0| 0.724| > 3.568 > RTD| 0.711| 1.071| 3.441| 0| 0.711| > 3.568 > RTD| 0.733| 1.066| 3.547| 0| 0.711| > 3.568 > RTD| 0.735| 1.079| 4.523| 0| 0.711| > 4.523 > RTD| 0.716| 1.072| 3.307| 0| 0.711| > 4.523 > RTD| 0.715| 1.066| 3.507| 0| 0.711| > 4.523 > ---|------------|------------|------------|--------|------------------------- > RTS| 0.711| 1.071| 4.523| 0| 00:00:10/00:00:10 > > > The patch (2.6.20 x86_64) can be found here : > http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-07-trem.patch > s,1.7-07,1.0-07, > I hope it could be used. > Nice. Will use this work as the upcoming 2.6.20/x86_64 baseline in my git repo. Thanks. -- Philippe. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-03-04 17:32 ` trem 2007-03-07 23:31 ` Philippe Gerum @ 2007-03-25 17:19 ` Philippe Gerum 2007-03-30 10:16 ` Paul 1 sibling, 1 reply; 8+ messages in thread From: Philippe Gerum @ 2007-03-25 17:19 UTC (permalink / raw) To: trem; +Cc: adeos-main On Sun, 2007-03-04 at 18:32 +0100, trem wrote: > Hi > > As I've got your start now, I've ported (at least tried) the 2.6.19 > x86_64 adeos patch to 2.6.20. > I've tried it and it seems to works fine, here the latency : > > [trem@domain.hid ~]$ cat latency_xenomai_2_6_20.txt > [trem@domain.hid bin]$ sudo ./latency -T 10 > == Sampling period: 100 us > == Test mode: periodic user-mode task > == All results in microseconds > warming up... > RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) > RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat > worst > RTD| 0.737| 1.067| 3.332| 0| 0.737| > 3.332 > RTD| 0.744| 1.067| 3.568| 0| 0.737| > 3.568 > RTD| 0.724| 1.080| 3.362| 0| 0.724| > 3.568 > RTD| 0.730| 1.070| 2.900| 0| 0.724| > 3.568 > RTD| 0.711| 1.071| 3.441| 0| 0.711| > 3.568 > RTD| 0.733| 1.066| 3.547| 0| 0.711| > 3.568 > RTD| 0.735| 1.079| 4.523| 0| 0.711| > 4.523 > RTD| 0.716| 1.072| 3.307| 0| 0.711| > 4.523 > RTD| 0.715| 1.066| 3.507| 0| 0.711| > 4.523 > ---|------------|------------|------------|--------|------------------------- > RTS| 0.711| 1.071| 4.523| 0| 00:00:10/00:00:10 > > > The patch (2.6.20 x86_64) can be found here : > http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-07-trem.patch > > I hope it could be used. > Merged, thanks. This upgrade is now available from the Adeos GIT repo and Xenomai's SVN trunk. > regards, > trem > > > Philippe Gerum wrote: > > On Sat, 2007-03-03 at 23:24 +0100, trem wrote: > > > >> Philippe Gerum wrote: > >> > >>> On Mon, 2007-02-26 at 00:50 +0100, trem wrote: > >>> > >>> > >>>> Hi > >>>> > >>>> I've tried to port the 2.6.19 x86_64 adeos patch to 2.6.20. you can find > >>>> it here : > >>>> http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-03-trem.patch > >>>> > >>>> > >>> The good news is that moving to 2.6.20 does not seem to be too much of a > >>> hell. Right? > >>> > >>> > >>> > >> Yes, there are few conflict so it's quite easy to port xenomai to 2.6.20. > >> > >>>> I can apply it with with prepare_kernel.sh (from xenomai svn), but I haven't > >>>> tested it yet (I mean run kernel). > >>>> > >>>> I hope this patch is good enough to be used. Otherwise, I'd be pleased > >>>> to fix it (maybe with a bit of help). > >>>> > >>>> > >>>> > >>> We are hopefully close to have a reasonably good patch for 2.6.19. Once > >>> we achieve this, porting to 2.6.20 should be the next step. Thanks. > >>> > >>> > >>> > >> I've seen ths in the Changelog : "this closes the implementation phase > >> of the x86_64 port." > >> So I suppose that the patch for 2.6.19 is finished. I've tried it and it > >> works fine. > >> The result of latency is : > >> > >> [trem@domain.hid bin]$ sudo ./latency -T 10 > >> == Sampling period: 100 us > >> == Test mode: periodic user-mode task > >> == All results in microseconds > >> warming up... > >> RTT| 00:00:01 (periodic user-mode task, 100 us period, priority 99) > >> RTH|-----lat min|-----lat avg|-----lat max|-overrun|----lat best|---lat > >> worst > >> RTD| 0.760| 1.073| 3.660| 0| 0.760| > >> 3.660 > >> RTD| 0.737| 1.063| 3.012| 0| 0.737| > >> 3.660 > >> RTD| 0.741| 1.083| 3.396| 0| 0.737| > >> 3.660 > >> RTD| 0.755| 1.088| 2.344| 0| 0.737| > >> 3.660 > >> RTD| 0.754| 1.093| 3.431| 0| 0.737| > >> 3.660 > >> RTD| 0.755| 1.089| 5.119| 0| 0.737| > >> 5.119 > >> RTD| 0.724| 1.091| 3.126| 0| 0.724| > >> 5.119 > >> RTD| 0.762| 1.089| 3.171| 0| 0.724| > >> 5.119 > >> RTD| 0.751| 1.097| 5.508| 0| 0.724| > >> 5.508 > >> ---|------------|------------|------------|--------|------------------------- > >> RTS| 0.724| 1.085| 5.508| 0| 00:00:10/00:00:10 > >> > >> If I don't mistake, I see a max latency of 5us. Is it realistic ? > >> > >> > > > > Yes. X-related problems Paul told us about being put aside (likely MTRR > > issues, still to be chased), I see < 10 us under SMP test load on a 2 x > > dual core Opteron 285 here (test load meaning two parallel kernel > > compilations (-j5), and some significant network load). > > > > We still have to benchmark this port more exhaustively and aggressively > > (e.g. different I/O loads and severe cache trashing, for longer than 8 > > hours), but the foundations look sane, especially since the > > Xenomai/x86_64 port currently runs with a default latency calibration > > set to 500 ns over this. > > > > > >> Now, the port to 2.6.20 is open ? we can work on ? > >> > >> > > > > Yep. Have fun. > > > > > >> trem > >> > >> > >> > >> > > -- Philippe. ^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Adeos-main] port 2.6.19 x86_64 to 2.6.20 2007-03-25 17:19 ` Philippe Gerum @ 2007-03-30 10:16 ` Paul 0 siblings, 0 replies; 8+ messages in thread From: Paul @ 2007-03-30 10:16 UTC (permalink / raw) To: adeos-main, rpm On Sunday 25 March 2007 18:19, Philippe Gerum wrote: > > The patch (2.6.20 x86_64) can be found here : > > http://zarb.org/~trem/adeos-ipipe-2.6.20-x86_64-1.7-07-trem.patch > > > > I hope it could be used. > > Merged, thanks. This upgrade is now available from the Adeos GIT repo > and Xenomai's SVN trunk. Attempting to compile against 2.6.20.4, hit a couple of minor snags.. Two includes in kernel/ipipe/tracer.c. linux/ipipe.h for __ipipe_pipeline and linux/sched.h for tasklist_lock - Also needed to insert an extern for cpu_khz in asm-x86_64/ipipe.h (no diffs available I'm afraid).. It looks like the kernel developers are starting to clean up the rats nest of includes, so I suspect there are going to be more "undefined variable" errors cropping up in the future. The 2.6.20.4 packages are available on http://zathras.tuxcnc.org/xenomai if anyone would like to test - Latest builds use r2340 trunk from SVN. Regards, Paul. ^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2007-03-30 10:16 UTC | newest] Thread overview: 8+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2007-02-25 23:50 [Adeos-main] port 2.6.19 x86_64 to 2.6.20 trem 2007-02-26 5:56 ` Philippe Gerum 2007-03-03 22:24 ` trem 2007-03-03 22:58 ` Philippe Gerum 2007-03-04 17:32 ` trem 2007-03-07 23:31 ` Philippe Gerum 2007-03-25 17:19 ` Philippe Gerum 2007-03-30 10:16 ` Paul
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.