Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Eugene Grosbein]

04.02.2018 6:42, Michael Voorhis wrote:
 
> 1 frame of your requested "top" output, sorted as specified:
> 
>> last pid: 47195;  load averages:  0.17,  0.37,  0.44 up 99+20:40:41  18:37:07
>> 369 processes: 1 running, 368 sleeping
>> CPU:  0.2% user,  0.0% nice,  0.2% system,  0.0% interrupt, 99.7% idle
>> Mem: 6989M Active, 79G Inact, 27G Laundry, 10G Wired, 1568M Buf, 1680M Free
>> Swap: 21G Total, 11G Used, 10G Free, 50% Inuse
>>
>>   PID USERNAMETHR PRI NICE   SIZERES STATE   C   TIMEWCPU COMMAND
>> 61839 pgsql 1  210 32665M 31349M select 13   3:13   0.00% 
>> postgres
>> 59354 pgsql 1  200 32659M 26625M select 12 288:57   0.00% 
>> postgres
>> 40354 pgsql 1  200 32660M 26182M select 12 220:15   0.00% 
>> postgres
>> 24708 pgsql 1  200 32659M 26144M select  3 204:14   0.00% 
>> postgres
>> 37664 pgsql 1  200 32649M 25967M select 12  48:55   0.00% 
>> postgres
>> 56094 pgsql 1  250 32659M 25734M select  5 189:00   0.00% 
>> postgres
>> 57255 pgsql 1  200 32661M 25414M select  6 143:53   0.00% 
>> postgres
>> 38662 pgsql 1  200 32659M 23669M select  1 229:03   0.00% 
>> postgres
>> 48035 pgsql 1  200 32663M 22195M select  3 149:13   0.00% 
>> postgres
>> 66831 pgsql 1  200 32659M 21944M select  2  61:00   0.13% 
>> postgres
>> 45832 pgsql 1  200 32659M 21427M select  4 158:26   0.00% 
>> postgres
>> 56976 pgsql 1  200 32661M 14110M select  1  56:26   0.00% 
>> postgres
>> 26207 pgsql 1  200 32659M 11620M select  8  34:21   0.00% 
>> postgres
>>   383 www  64  200 34010M  8860M nanslp 14  11:26   0.11% jsvc
>> 44067 pgsql 1  200 32659M  6194M select 11  15:16   0.00% 
>> postgres
>> 31431 pgsql 1  200 32661M  4928M select 15  15:41   0.00% 
>> postgres
>> 34641 pgsql 1  200 32650M  2383M select 14   2:04   0.00% 
>> postgres> [...]

Well, you have tens of processes each allocating about 32G of memory.
Some part of that memory may be shared but some is not not.

No wonder there are tens gigabytes of private pages shown processes have
and some of that pages were paged out to swap area in moments
when kernel was short of free pages (note small "Free" number).

Later when some memory was freed, those pages were NOT get back from the swap
because there were no demand for their contents. That's pretty normal.

For example, PIDs 56976 or 26207 or 26207 all allocated over 30G
and may be touched over 20G (let's assume that) but their RESident size
is considerably less, so rest was paged out to swap.

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Michael Voorhis]

On 02/03/2018 05:48 PM, Eugene Grosbein wrote:
> These 10G may be just several pages of several processes.
> Please show output of "top -ores -d1".

This just shows a bunch of hungry postgres processes (see below).

In response to your slightly-earlier email, swap_enabled is set (its
default) and swap_idle_enabled is NOT set.  No swap-related sysctl's are
set in /etc/sysctl.conf.

1 frame of your requested "top" output, sorted as specified:

> last pid: 47195;  load averages:  0.17,  0.37,  0.44  
>up 99+20:40:41  18:37:07
> 369 processes: 1 running, 368 sleeping
> CPU:  0.2% user,  0.0% nice,  0.2% system,  0.0% interrupt, 99.7% idle
> Mem: 6989M Active, 79G Inact, 27G Laundry, 10G Wired, 1568M Buf, 1680M Free
> Swap: 21G Total, 11G Used, 10G Free, 50% Inuse
> 
>   PID USERNAMETHR PRI NICE   SIZERES STATE   C   TIMEWCPU COMMAND
> 61839 pgsql 1  210 32665M 31349M select 13   3:13   0.00% postgres
> 59354 pgsql 1  200 32659M 26625M select 12 288:57   0.00% postgres
> 40354 pgsql 1  200 32660M 26182M select 12 220:15   0.00% postgres
> 24708 pgsql 1  200 32659M 26144M select  3 204:14   0.00% postgres
> 37664 pgsql 1  200 32649M 25967M select 12  48:55   0.00% postgres
> 56094 pgsql 1  250 32659M 25734M select  5 189:00   0.00% postgres
> 57255 pgsql 1  200 32661M 25414M select  6 143:53   0.00% postgres
> 38662 pgsql 1  200 32659M 23669M select  1 229:03   0.00% postgres
> 48035 pgsql 1  200 32663M 22195M select  3 149:13   0.00% postgres
> 66831 pgsql 1  200 32659M 21944M select  2  61:00   0.13% postgres
> 45832 pgsql 1  200 32659M 21427M select  4 158:26   0.00% postgres
> 56976 pgsql 1  200 32661M 14110M select  1  56:26   0.00% postgres
> 26207 pgsql 1  200 32659M 11620M select  8  34:21   0.00% postgres
>   383 www  64  200 34010M  8860M nanslp 14  11:26   0.11% jsvc
> 44067 pgsql 1  200 32659M  6194M select 11  15:16   0.00% postgres
> 31431 pgsql 1  200 32661M  4928M select 15  15:41   0.00% postgres
> 34641 pgsql 1  200 32650M  2383M select 14   2:04   0.00% 
> postgres> [...]

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Eugene Grosbein]

04.02.2018 4:14, Michael Voorhis wrote:

> I've got an amd64 system running 11.1-STABLE r325027, with something
> like 20G of swap. "swapinfo" shows that half the swap is used.
> 
> So of course I'm curious to know which processes have been swapped
> out. I'm not using any "tmpfs" filesystems; no ZFS, no huge amounts of
> wired-down memory. The system's got 16 processors and 128G of RAM. "ps
> auxww" output shows *no* processes that are swapped out (2nd character
> in "STAT" field is "W"). Not a single one. The only process with a W in
> the stat field at all is the "[intr]" kernel thread.
> 
> What is using the swapspace?

These 10G may be just several pages of several processes.
Please show output of "top -ores -d1".

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Eugene Grosbein]

04.02.2018 5:32, Brandon Allbery wrote:

> Also worth noting is that likely candidates for such pageouts include
> long-lived daemons that are only needed, or which only need certain pages,
> during startup/shutdown. So evicting only those pages to swap allows
> optimal use of memory that would otherwise be wasted unnecessarily.
> 
> Studying demand paging and unified page management is worth the effort.
> Modern OSes, including Windows, make heavy use of this to optimize memory
> usage --- but it means that old-style notions of process memory usage will
> leave you wondering how the numbers make any sense. (I see this quite a
> lot; most people still seem to think the basic unit of memory management is
> a process, not a memory page, despite unified page management being over a
> decade old and basic demand paging going back to 4BSD days.)

FreeBSD kernel does not try to "swapout" even long-lived sleeping daemons as a 
whole 
when it needs some free pages. Even if it needs to free more pages by paging out
some of them, it tries to minimize such I/O operations by default
and writes only minimal necessary amount of pages keeping the rest intact,
when vm.swap_idle_enabled=0.

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Eugene Grosbein]

04.02.2018 5:09, Mark Millard via freebsd-stable wrote:

> I do not know if a W after the first letter in state (STAT) for
> "ps auxww" track the kernel-stacks' resident-vs-not status for the
> process or not. (Matching your not sure status.)

A process has specific flag P_INMEM that is normally 1 but is cleader by a 
kernel
when "swapout" procedure is executed for a process. Such process cannot get CPU 
time etc.
top shows process name like "" instead of just "name" in that case.

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Brandon Allbery]

Also worth noting is that likely candidates for such pageouts include
long-lived daemons that are only needed, or which only need certain pages,
during startup/shutdown. So evicting only those pages to swap allows
optimal use of memory that would otherwise be wasted unnecessarily.

Studying demand paging and unified page management is worth the effort.
Modern OSes, including Windows, make heavy use of this to optimize memory
usage --- but it means that old-style notions of process memory usage will
leave you wondering how the numbers make any sense. (I see this quite a
lot; most people still seem to think the basic unit of memory management is
a process, not a memory page, despite unified page management being over a
decade old and basic demand paging going back to 4BSD days.)

On Sat, Feb 3, 2018 at 5:09 PM, Mark Millard via freebsd-stable <
freebsd-stable@freebsd.org> wrote:

> Brandon Allbery allbery.b at gmail.com wrote on
> Sat Feb 3 21:18:53 UTC 2018 :
>
> > Swapping whole processes out is not really a thing any more. Individual
> > pages are paged to/from memory; if a memory page has no backing file, it
> > will be allocated a block in swap space as its backing storage.
> >
> > (I'm not sure "W" status even means swap; I thought whole-process
> swapping
> > wasn't even supported any more.)
>
> From what I've seen on the lists there is a technical distinction
> made between "kernel stacks for the process no longer memory resident"
> (swapped out) and other pages for the process having paged to disk and
> not being resident.
>
> But many tools do not seem to present that point of view and still
> reflect an older view in the terminology used, including in
> documentation. One has to interpret what one is shown as I understand.
>
> As an example, top can show RES being zero despite the kernel stacks
> for the process not having been moved to disk. RES zero might not
> mean what one might expect about "swapped out".
>
> I do not know if a W after the first letter in state (STAT) for
> "ps auxww" track the kernel-stacks' resident-vs-not status for the
> process or not. (Matching your not sure status.)
>
>
> > On Sat, Feb 3, 2018 at 4:14 PM, Michael Voorhis 
> wrote:
> >
> > > Hi all,
> > >
> > > I've got an amd64 system running 11.1-STABLE r325027, with something
> > > like 20G of swap. "swapinfo" shows that half the swap is used.
> > >
> > > So of course I'm curious to know which processes have been swapped
> > > out. I'm not using any "tmpfs" filesystems; no ZFS, no huge amounts of
> > > wired-down memory. The system's got 16 processors and 128G of RAM. "ps
> > > auxww" output shows *no* processes that are swapped out (2nd character
> > > in "STAT" field is "W"). Not a single one. The only process with a W in
> > > the stat field at all is the "[intr]" kernel thread.
> > >
> > > What is using the swapspace
>
> The so-called swapspace is really the paging/swap-space with
> most of the use being paging typically. (As Brandon indicated.)
>
> Once a page is paged out, if the process sticks around but
> does not use or free the page, the page likely stays
> paged-out. (I'm guessing some at the intended results for
> default tuning --and that you probably are using default
> tuning.) So the in-use swapspace is likely from one or
> more existing processes that did page-outs earlier.
>
> (Expect my descriptions to be over simplified, but hopefully
> pointing in the right general direction.)
>
> > > Please educate me.
> > >
>
>
> ===
> Mark Millard
> marklmi at yahoo.com
> ( markmi at dsl-only.net is
> going away in 2018-Feb, late)
>
> ___
> freebsd-stable@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
>



-- 
brandon s allbery kf8nh   sine nomine associates
allber...@gmail.com  ballb...@sinenomine.net
unix, openafs, kerberos, infrastructure, xmonadhttp://sinenomine.net
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Mark Millard via freebsd-stable]

Brandon Allbery allbery.b at gmail.com wrote on
Sat Feb 3 21:18:53 UTC 2018 :

> Swapping whole processes out is not really a thing any more. Individual
> pages are paged to/from memory; if a memory page has no backing file, it
> will be allocated a block in swap space as its backing storage.
> 
> (I'm not sure "W" status even means swap; I thought whole-process swapping
> wasn't even supported any more.)

>From what I've seen on the lists there is a technical distinction
made between "kernel stacks for the process no longer memory resident"
(swapped out) and other pages for the process having paged to disk and
not being resident.

But many tools do not seem to present that point of view and still
reflect an older view in the terminology used, including in
documentation. One has to interpret what one is shown as I understand.

As an example, top can show RES being zero despite the kernel stacks
for the process not having been moved to disk. RES zero might not
mean what one might expect about "swapped out".

I do not know if a W after the first letter in state (STAT) for
"ps auxww" track the kernel-stacks' resident-vs-not status for the
process or not. (Matching your not sure status.)


> On Sat, Feb 3, 2018 at 4:14 PM, Michael Voorhis  wrote:
> 
> > Hi all,
> >
> > I've got an amd64 system running 11.1-STABLE r325027, with something
> > like 20G of swap. "swapinfo" shows that half the swap is used.
> >
> > So of course I'm curious to know which processes have been swapped
> > out. I'm not using any "tmpfs" filesystems; no ZFS, no huge amounts of
> > wired-down memory. The system's got 16 processors and 128G of RAM. "ps
> > auxww" output shows *no* processes that are swapped out (2nd character
> > in "STAT" field is "W"). Not a single one. The only process with a W in
> > the stat field at all is the "[intr]" kernel thread.
> >
> > What is using the swapspace

The so-called swapspace is really the paging/swap-space with
most of the use being paging typically. (As Brandon indicated.)

Once a page is paged out, if the process sticks around but
does not use or free the page, the page likely stays
paged-out. (I'm guessing some at the intended results for
default tuning --and that you probably are using default
tuning.) So the in-use swapspace is likely from one or
more existing processes that did page-outs earlier.

(Expect my descriptions to be over simplified, but hopefully
pointing in the right general direction.)

> > Please educate me.
> >


===
Mark Millard
marklmi at yahoo.com
( markmi at dsl-only.net is
going away in 2018-Feb, late)

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Eugene Grosbein]

04.02.2018 4:18, Brandon Allbery wrote:

> (I'm not sure "W" status even means swap; I thought whole-process swapping
> wasn't even supported any more.)

Kernel may decide to swap out entire processes if vm.swap_enabled=1 (default)
and its free page pool heavily stressed or system was configured by 
administrator
to swap out idle processes proactively using vm.swap_idle_enabled=1 (not 
default).

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Michael Voorhis]

On 02/03/2018 04:18 PM, Brandon Allbery wrote:
> Swapping whole processes out is not really a thing any more. 
> Individual pages are paged to/from memory; if a memory page has no 
> backing file, it will be allocated a block in swap space as its 
> backing storage.

Is there a method to determine what swap contents are connected to, if
looking at processes is no longer "a thing"? I have great confidence in
the wonderful FreeBSD documentation, but I find nothing (quickly) in the
manual pages.

> (I'm not sure "W" status even means swap; I thought whole-process 
> swapping wasn't even supported any more.)

The manpage for ps(1), (which I'm sure you're aware of!) describes the
"state" field and its multiple characters and their meanings... that's
what I used for reference.

"W" as 1st character means "idle interrupt thread [of the kernel]."
Subsequent W characters imply swapped-out processes. In subsequent
characters a W indicates that a PID is swapped out.

Thanks for your reply,

--MCV.
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: 50 percent swap used, but "ps auxww" output shows no processes swapped out

[Brandon Allbery]

Swapping whole processes out is not really a thing any more. Individual
pages are paged to/from memory; if a memory page has no backing file, it
will be allocated a block in swap space as its backing storage.

(I'm not sure "W" status even means swap; I thought whole-process swapping
wasn't even supported any more.)

On Sat, Feb 3, 2018 at 4:14 PM, Michael Voorhis  wrote:

> Hi all,
>
> I've got an amd64 system running 11.1-STABLE r325027, with something
> like 20G of swap. "swapinfo" shows that half the swap is used.
>
> So of course I'm curious to know which processes have been swapped
> out. I'm not using any "tmpfs" filesystems; no ZFS, no huge amounts of
> wired-down memory. The system's got 16 processors and 128G of RAM. "ps
> auxww" output shows *no* processes that are swapped out (2nd character
> in "STAT" field is "W"). Not a single one. The only process with a W in
> the stat field at all is the "[intr]" kernel thread.
>
> What is using the swapspace?
>
> Please educate me.
>
> Thanks,
>
> --MCV.
> ___
> freebsd-stable@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
>



-- 
brandon s allbery kf8nh   sine nomine associates
allber...@gmail.com  ballb...@sinenomine.net
unix, openafs, kerberos, infrastructure, xmonadhttp://sinenomine.net
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

50 percent swap used, but "ps auxww" output shows no processes swapped out

[Michael Voorhis]

Hi all,

I've got an amd64 system running 11.1-STABLE r325027, with something
like 20G of swap. "swapinfo" shows that half the swap is used.

So of course I'm curious to know which processes have been swapped
out. I'm not using any "tmpfs" filesystems; no ZFS, no huge amounts of
wired-down memory. The system's got 16 processors and 128G of RAM. "ps
auxww" output shows *no* processes that are swapped out (2nd character
in "STAT" field is "W"). Not a single one. The only process with a W in
the stat field at all is the "[intr]" kernel thread.

What is using the swapspace?

Please educate me.

Thanks,

--MCV.
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: panic when loading mlxen

[Daniel Braniss]

> On 3 Feb 2018, at 12:16, Hans Petter Selasky  wrote:
> 
> Hi,
> 
> I think Alexander came ahead of me:
> 
> https://svnweb.freebsd.org/base?view=revision=328805
> 
> Can you try r328805 ?
> 
> --HPS

yup, it works, well it doesn’t panic.

thanks
danny

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: panic when loading mlxen

[Hans Petter Selasky]

Hi,

I think Alexander came ahead of me:

https://svnweb.freebsd.org/base?view=revision=328805

Can you try r328805 ?

--HPS
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: panic when loading mlxen

[Daniel Braniss]

> On 3 Feb 2018, at 11:34, Hans Petter Selasky  wrote:
> 
> On 02/03/18 08:34, Daniel Braniss wrote:
>>> On 2 Feb 2018, at 20:47, K. Macy  wrote:
>>> 
>>> That's odd since it doesn't use any of taskqgroup stuff. I take it you
>>> can't get a core?
>> no core but some more info:
>> db> bt
>> Tracing pid 0 tid 10 td 0x81e0e500
>> taskqgroup_attach_cpu() at taskqgroup_attach_cpu+0x4f/frame 
>> 0x822e4c30
>> tasklet_subsystem_init() at tasklet_subsystem_init+0xde/frame 
>> 0x822e4c90
>> mi_startup() at mi_startup+0x9c/frame 0x822e4cb0
>> btext() at btext+0x2c
>>> 
>>> Also, why are you loading it in loader.conf (slower) as opposed to rc.conf?
>> sometimes it’s booted diskless, and the driver is needed early.
>> and btw, this box doesn’t even have a mellanox card.
>>> -M
>>> 
>>> 
>>> 
>>> On Fri, Feb 2, 2018 at 4:46 AM, Daniel Braniss  wrote:
 with latest stable (r328769) when I have
mlxen_load=“YES”
 in my loader.conf it panics:
 
 KDB: debugger backends: ddbsize 0x4638 at 0x22d6000
  f
 KDB: current backend: ddb
 Copyright (c) 1992-2018 The FreeBSD Project.
 Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
   The Regents of the University of California. All rights reserved.
 FreeBSD is a registered trademark of The FreeBSD Foundation.
 FreeBSD 11.1-STABLE #18: Fri Feb  2 10:46:12 IST 2018
   danny@pe-44:/home/obj/pe-44/net/rnd/r+d/stable/11/sys/HUJI amd64
 FreeBSD clang version 5.0.1 (tags/RELEASE_501/final 320880) (based on LLVM 
 5.0.1)
 VT(vga): resolution 640x480
 CPU: Intel(R) Xeon(R) CPU   E5507  @ 2.27GHz (2261.04-MHz K8-class 
 CPU)
 Origin="GenuineIntel"  Id=0x106a5  Family=0x6  Model=0x1a  Stepping=5
 Features=0xbfebfbff
 Features2=0x9ce3bd
 AMD Features=0x28100800
 AMD Features2=0x1
 VT-x: PAT,HLT,MTF,PAUSE,EPT,VPID
 TSC: P-state invariant, performance statistics
 real memory  = 25769803776 (24576 MB)
 avail memory = 24931561472 (23776 MB)
 Event timer "LAPIC" quality 100
 ACPI APIC Table: 
 FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs
 FreeBSD/SMP: 2 package(s) x 4 core(s)
 ioapic1: Changing APIC ID to 1
 ioapic0  irqs 0-23 on motherboard
 ioapic1  irqs 32-55 on motherboard
 
 
 Fatal trap 12: page fault while in kernel mode
 cpuid = 0; apic id = 10
 fault virtual address   = 0x1818
 fault code  = supervisor write data, page not present
 instruction pointer = 0x20:0x80ad427f
 stack pointer   = 0x28:0x822e3be0
 frame pointer   = 0x28:0x822e3c30
 code segment= base 0x0, limit 0xf, type 0x1b
   = DPL 0, pres 1, long 1, def32 0, gran 1
 processor eflags= interrupt enabled, resume, IOPL = 0
 current process = 0 (swapper)
 [ thread pid 0 tid 10 ]
 Stopped at  taskqgroup_attach_cpu+0x4f: lock cmpxchgq   %r12,(%rdi)
> 
> Hi,
> 
> It should work if you "kldload mlxen" after boot or add it to kld_list in 
> /etc/rc.conf. Looks like I have one more combination to test after the 
> LinuxKPI upgrade in 11-stable. Thanks for notifying me.
> 
Hi,
it’s ok, i don’t need it yet, I was just surprised that a simple upgrade got 
the panic.
let me know if you need me to test.

thanks,

danny

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: panic when loading mlxen

[Hans Petter Selasky]

On 02/03/18 08:34, Daniel Braniss wrote:




On 2 Feb 2018, at 20:47, K. Macy  wrote:

That's odd since it doesn't use any of taskqgroup stuff. I take it you
can't get a core?


no core but some more info:
db> bt
Tracing pid 0 tid 10 td 0x81e0e500
taskqgroup_attach_cpu() at taskqgroup_attach_cpu+0x4f/frame 0x822e4c30
tasklet_subsystem_init() at tasklet_subsystem_init+0xde/frame 0x822e4c90
mi_startup() at mi_startup+0x9c/frame 0x822e4cb0
btext() at btext+0x2c



Also, why are you loading it in loader.conf (slower) as opposed to rc.conf?

sometimes it’s booted diskless, and the driver is needed early.
and btw, this box doesn’t even have a mellanox card.



-M



On Fri, Feb 2, 2018 at 4:46 AM, Daniel Braniss  wrote:

with latest stable (r328769) when I have
mlxen_load=“YES”
in my loader.conf it panics:

KDB: debugger backends: ddbsize 0x4638 at 0x22d6000 
f
KDB: current backend: ddb
Copyright (c) 1992-2018 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
   The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 11.1-STABLE #18: Fri Feb  2 10:46:12 IST 2018
   danny@pe-44:/home/obj/pe-44/net/rnd/r+d/stable/11/sys/HUJI amd64
FreeBSD clang version 5.0.1 (tags/RELEASE_501/final 320880) (based on LLVM 
5.0.1)
VT(vga): resolution 640x480
CPU: Intel(R) Xeon(R) CPU   E5507  @ 2.27GHz (2261.04-MHz K8-class CPU)
Origin="GenuineIntel"  Id=0x106a5  Family=0x6  Model=0x1a  Stepping=5
Features=0xbfebfbff
Features2=0x9ce3bd
AMD Features=0x28100800
AMD Features2=0x1
VT-x: PAT,HLT,MTF,PAUSE,EPT,VPID
TSC: P-state invariant, performance statistics
real memory  = 25769803776 (24576 MB)
avail memory = 24931561472 (23776 MB)
Event timer "LAPIC" quality 100
ACPI APIC Table: 
FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs
FreeBSD/SMP: 2 package(s) x 4 core(s)
ioapic1: Changing APIC ID to 1
ioapic0  irqs 0-23 on motherboard
ioapic1  irqs 32-55 on motherboard


Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 10
fault virtual address   = 0x1818
fault code  = supervisor write data, page not present
instruction pointer = 0x20:0x80ad427f
stack pointer   = 0x28:0x822e3be0
frame pointer   = 0x28:0x822e3c30
code segment= base 0x0, limit 0xf, type 0x1b
   = DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags= interrupt enabled, resume, IOPL = 0
current process = 0 (swapper)
[ thread pid 0 tid 10 ]
Stopped at  taskqgroup_attach_cpu+0x4f: lock cmpxchgq   %r12,(%rdi)


Hi,

It should work if you "kldload mlxen" after boot or add it to kld_list 
in /etc/rc.conf. Looks like I have one more combination to test after 
the LinuxKPI upgrade in 11-stable. Thanks for notifying me.


--HPS
___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

Re: panic when loading mlxen

[Daniel Braniss]

> On 2 Feb 2018, at 20:47, K. Macy  wrote:
> 
> That's odd since it doesn't use any of taskqgroup stuff. I take it you
> can't get a core?

no core but some more info:
db> bt
Tracing pid 0 tid 10 td 0x81e0e500
taskqgroup_attach_cpu() at taskqgroup_attach_cpu+0x4f/frame 0x822e4c30
tasklet_subsystem_init() at tasklet_subsystem_init+0xde/frame 0x822e4c90
mi_startup() at mi_startup+0x9c/frame 0x822e4cb0
btext() at btext+0x2c

> 
> Also, why are you loading it in loader.conf (slower) as opposed to rc.conf?
sometimes it’s booted diskless, and the driver is needed early.
and btw, this box doesn’t even have a mellanox card.


> -M
> 
> 
> 
> On Fri, Feb 2, 2018 at 4:46 AM, Daniel Braniss  wrote:
>> with latest stable (r328769) when I have
>>mlxen_load=“YES”
>> in my loader.conf it panics:
>> 
>> KDB: debugger backends: ddbsize 0x4638 at 0x22d6000  
>>f
>> KDB: current backend: ddb
>> Copyright (c) 1992-2018 The FreeBSD Project.
>> Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
>>   The Regents of the University of California. All rights reserved.
>> FreeBSD is a registered trademark of The FreeBSD Foundation.
>> FreeBSD 11.1-STABLE #18: Fri Feb  2 10:46:12 IST 2018
>>   danny@pe-44:/home/obj/pe-44/net/rnd/r+d/stable/11/sys/HUJI amd64
>> FreeBSD clang version 5.0.1 (tags/RELEASE_501/final 320880) (based on LLVM 
>> 5.0.1)
>> VT(vga): resolution 640x480
>> CPU: Intel(R) Xeon(R) CPU   E5507  @ 2.27GHz (2261.04-MHz K8-class 
>> CPU)
>> Origin="GenuineIntel"  Id=0x106a5  Family=0x6  Model=0x1a  Stepping=5
>> Features=0xbfebfbff
>> Features2=0x9ce3bd
>> AMD Features=0x28100800
>> AMD Features2=0x1
>> VT-x: PAT,HLT,MTF,PAUSE,EPT,VPID
>> TSC: P-state invariant, performance statistics
>> real memory  = 25769803776 (24576 MB)
>> avail memory = 24931561472 (23776 MB)
>> Event timer "LAPIC" quality 100
>> ACPI APIC Table: 
>> FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs
>> FreeBSD/SMP: 2 package(s) x 4 core(s)
>> ioapic1: Changing APIC ID to 1
>> ioapic0  irqs 0-23 on motherboard
>> ioapic1  irqs 32-55 on motherboard
>> 
>> 
>> Fatal trap 12: page fault while in kernel mode
>> cpuid = 0; apic id = 10
>> fault virtual address   = 0x1818
>> fault code  = supervisor write data, page not present
>> instruction pointer = 0x20:0x80ad427f
>> stack pointer   = 0x28:0x822e3be0
>> frame pointer   = 0x28:0x822e3c30
>> code segment= base 0x0, limit 0xf, type 0x1b
>>   = DPL 0, pres 1, long 1, def32 0, gran 1
>> processor eflags= interrupt enabled, resume, IOPL = 0
>> current process = 0 (swapper)
>> [ thread pid 0 tid 10 ]
>> Stopped at  taskqgroup_attach_cpu+0x4f: lock cmpxchgq   %r12,(%rdi)
>> ___
>> freebsd-stable@freebsd.org mailing list
>> https://lists.freebsd.org/mailman/listinfo/freebsd-stable
>> To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"

___
freebsd-stable@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"