NetBSD Problem Report #39242
From Wolfgang.Stukenbrock@nagler-company.com Mon Jul 28 21:08:48 2008
Return-Path: <Wolfgang.Stukenbrock@nagler-company.com>
Received: from mail.netbsd.org (mail.netbsd.org [204.152.190.11])
by narn.NetBSD.org (Postfix) with ESMTP id 515D063B91E
for <gnats-bugs@gnats.NetBSD.org>; Mon, 28 Jul 2008 21:08:48 +0000 (UTC)
Message-Id: <20080728180121.6784479B8D@s040.nagler-company.com>
Date: Mon, 28 Jul 2008 20:01:21 +0200 (CEST)
From: Wolfgang.Stukenbrock@nagler-company.com
Reply-To: Wolfgang.Stukenbrock@nagler-company.com
To: gnats-bugs@gnats.NetBSD.org
Subject: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
X-Send-Pr-Version: 3.95
>Number: 39242
>Category: kern
>Synopsis: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
>Confidential: no
>Severity: critical
>Priority: high
>Responsible: kern-bug-people
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Mon Jul 28 21:10:00 +0000 2008
>Last-Modified: Thu Jul 31 11:00:02 +0000 2008
>Originator: Wolfgang Stukenbrock
>Release: NetBSD 4.0_STABLE
>Organization:
Dr. Nagler & Company GmbH
>Environment:
System: NetBSD s040 4.0_STABLE NetBSD 4.0_STABLE (NSW-S040) #7: Fri Jul 25 16:02:03 CEST 2008 root@s040:/usr/src/sys/arch/amd64/compile/NSW-S040 amd64
Architecture: x86_64
Machine: amd64
>Description:
After 4 GB main memory is used by the system, allocating another physical page fails,
the pagedeamon is kicked, but there is about 3,9 GB free memory - according to the statistic - and
th pagedeamon will do nothing.
The system comes to sudden stop at this point and nothing works anymore.
>How-To-Repeat:
Setup a machine with e.g. 8 GB RAM.
Then either start some large processes (e.g. "dd if=.. of=... bs=1024000k") until the need more than
4 GB memory or unpack large achives into the filesystem so that the filesystem cache will eat up
4 GB memory.
Run vmstat or top in parallel an you will see that there is something around 3,9 GB free memory, but
it is not allocated for unknown reasons.
The pagedaemon will show up 100% activity in top (on all CPU's in the system after a while) - if top
still gets updates, until everything freezes.
I've found no way to get the system out of this state without pressing the reset-button.
>Fix:
not known to me up to now.
I've tried VMHIST, but that does not realy help ...
Some additional printouts in the pagedeamon routines show up, that the pagedeamon is dooing nothing,
because there is still enought free physical memory.
The problem seems to be related to the 4 GB boundary - other systems with identical HW but only 4 GB RAM
doesn't show this behaviour.
>Audit-Trail:
From: Simon Burge <simonb@NetBSD.org>
To: Wolfgang.Stukenbrock@nagler-company.com, gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Tue, 29 Jul 2008 14:19:17 +1000
Wolfgang.Stukenbrock@nagler-company.com wrote:
> >Synopsis: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
What does "vmstat -s | grep colors" show on this machine? If it's a
reasonably recent Intel CPU with 6MB or 12MB of L2 cache, this will
probably not be a power of two and would explain what you're seeing.
In -current this was fixed by both fixing the cache detection stuff
and rev 1.32 of sys/arch/x86/x86/cpu.c.
Cheers,
Simon.
From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Tue, 29 Jul 2008 09:07:09 +0200
Hi,
yes it is a E3110 CPU with 6MB cache.
What files I need to catch from the current and integrate the changes
into my 4.0-version of netbsd?
You talk about the "cache detection stuff" AND the file named below.
I think it would be a real great idea to bring this fix into the
releases as soon as possible.
This CPU is "very" cheep compared to the other one's (at least in
germany) and therefore is very attractive for new systems.
W. Stukenbrock
Simon Burge wrote:
> The following reply was made to PR kern/39242; it has been noted by GNATS.
>
> From: Simon Burge <simonb@NetBSD.org>
> To: Wolfgang.Stukenbrock@nagler-company.com, gnats-bugs@NetBSD.org
> Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
> netbsd-bugs@netbsd.org
> Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
> Date: Tue, 29 Jul 2008 14:19:17 +1000
>
> Wolfgang.Stukenbrock@nagler-company.com wrote:
>
> > >Synopsis: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
>
> What does "vmstat -s | grep colors" show on this machine? If it's a
> reasonably recent Intel CPU with 6MB or 12MB of L2 cache, this will
> probably not be a power of two and would explain what you're seeing.
>
> In -current this was fixed by both fixing the cache detection stuff
> and rev 1.32 of sys/arch/x86/x86/cpu.c.
>
> Cheers,
> Simon.
>
>
From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc:
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Tue, 29 Jul 2008 09:14:27 +0200
Hi again,
sorry I've failed to insert the requested output ov vmstat ...
here it is:
s040# vmstat -s | grep colors
96 page colors
s040#
Simon Burge wrote:
> The following reply was made to PR kern/39242; it has been noted by GNATS.
>
> From: Simon Burge <simonb@NetBSD.org>
> To: Wolfgang.Stukenbrock@nagler-company.com, gnats-bugs@NetBSD.org
> Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
> netbsd-bugs@netbsd.org
> Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
> Date: Tue, 29 Jul 2008 14:19:17 +1000
>
> Wolfgang.Stukenbrock@nagler-company.com wrote:
>
> > >Synopsis: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
>
> What does "vmstat -s | grep colors" show on this machine? If it's a
> reasonably recent Intel CPU with 6MB or 12MB of L2 cache, this will
> probably not be a power of two and would explain what you're seeing.
>
> In -current this was fixed by both fixing the cache detection stuff
> and rev 1.32 of sys/arch/x86/x86/cpu.c.
>
> Cheers,
> Simon.
>
>
From: Simon Burge <simonb@NetBSD.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org,
netbsd-bugs@netbsd.org, Wolfgang.Stukenbrock@nagler-company.com
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Tue, 29 Jul 2008 23:21:47 +1000
Wolfgang Stukenbrock wrote:
> The following reply was made to PR kern/39242; it has been noted by GNATS.
>
> From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
> To: gnats-bugs@NetBSD.org
> Cc:
> Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
> Date: Tue, 29 Jul 2008 09:07:09 +0200
>
> Hi,
>
> yes it is a E3110 CPU with 6MB cache.
> What files I need to catch from the current and integrate the changes
> into my 4.0-version of netbsd?
> You talk about the "cache detection stuff" AND the file named below.
>
> I think it would be a real great idea to bring this fix into the
> releases as soon as possible.
> This CPU is "very" cheep compared to the other one's (at least in
> germany) and therefore is very attractive for new systems.
Does rev 1.32 of sys/arch/x86/x86/cpu.c apply cleanly to netbsd-4 ? I
don't recall which bits of which files moved around with x86 recently.
If so, that at least guarantees that a bogus number doesn't get passed
deeper into UVM, and that'll be enough for netbsd-4.
Note also the problem was to do with "half memory used", not "4G of
memory used".
Cheers,
Simon.
From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc: Simon Burge <simonb@NetBSD.org>, kern-bug-people@NetBSD.org
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Wed, 30 Jul 2008 20:13:32 +0200
Hi again,
I've seen responce to my last mail directly to Simon, but I continue
testing my system.
here the patch I've added to /usr7src/sys/amd64/amd64/cpu.c:
s012# rcsdiff -c -r1.1 c*
===================================================================
RCS file: RCS/cpu.c,v
retrieving revision 1.1
diff -c -r1.1 cpu.c
*** cpu.c 2008/07/29 07:28:03 1.1
--- cpu.c 2008/07/29 08:05:21
***************
*** 217,222 ****
--- 217,239 ----
tcolors /= cai->cai_associativity;
}
ncolors = max(ncolors, tcolors);
+ /*
+ * If the desired number of colors is not a power of
+ * two, it won't be good. Find the greatest power of
+ * two which is an even divisor of the number of colors,
+ * to preserve even coloring of pages.
+ */
+ if (ncolors & (ncolors - 1) ) {
+ int try, picked = 1;
+ for (try = 1; try < ncolors; try *= 2) {
+ if (ncolors % try == 0) picked = try;
+ }
+ if (picked == 1) {
+ panic("desired number of cache colors %d
is "
+ " > 1, but not even!", ncolors);
+ }
+ ncolors = picked;
+ }
}
/*
Just some minutes ago, I've got two new kernel crashes.
1. the kernel process [scsibus0] starts looping and sleeps sometimes in
pglalloc. After reinstalling the system with the patch above, there was
no DDB in the kernel, so I could not get any other information.
2. now I've DDB in the kernel and tried to reproduce the problem. But it
crashes prior reaching this state in pagedaemon ...
Some output from the console below:
uvm_fault(0xffffffff80628800, 0x0, 1) -> e
kernel: page fault trap, code=0
Stopped in pid 26.1 (pagedaemon) at netbsd:uvm_rb_insert+0x37:
movq 0
x40(%rax),%rax
db{0}> trace
uvm_rb_insert() at netbsd:uvm_rb_insert+0x37
uvm_map_enter() at netbsd:uvm_map_enter+0x290
uvm_map() at netbsd:uvm_map+0xfe
uvm_pagermapin() at netbsd:uvm_pagermapin+0x92
uvm_swap_io() at netbsd:uvm_swap_io+0x3c
swapcluster_flush() at netbsd:swapcluster_flush+0x55
uvm_pageout() at netbsd:uvm_pageout+0x42b
db{0}> show uvmexp
Current UVM status:
pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
2039030 VM pages: 1333964 active, 651446 inactive, 2248 wired, 5 free
pages 1674503 anon, 311641 file, 1580 exec
freemin=64, free-target=85, wired-max=679676
faults=3420439, traps=3975116, intrs=8916183, ctxswitch=14697719
softint=710475, syscalls=12629494, swapins=177, swapouts=210
fault counts:
noram=5, noanon=0, pgwait=0, pgrele=0
ok relocks(total)=1068(1069), anget(retrys)=73451(286), amapcopy=31017
neighbor anon/obj pg=65282/474220, gets(lock/unlock)=116248/783
cases: anon=51370, anoncow=21184, obj=96126, prcopy=20121, przero=63798
daemon and swap counts:
woke=32081, revs=4228, scans=1204885, obscans=1161315, anscans=2987
busy=0, freed=1164012, reactivate=128, deactivate=1857899
pageouts=75209, pending=1088875, nswget=78493
nswapdev=1, swpgavail=6291455
swpages=6291455, swpginuse=1164021, swpgonly=1085489, paging=66
db{0}>
This patch seems to enable the system to work with the 6MB cache of the
E3110 CPU, but the kernel is not realy stable at all.
Any idea? What should I try next?
By the way "vmstat -s | grep colo" reports 32 colors now.
The system was busy again when the crash happens. raidframe sync on one
SATA-raid and one SCSI-raid, transfered something arund 12 GB int /tmp
(tmpfs) so nearly 5 GB of the 24 GB swap was used.
The kernel crashes at the moment where I've tried to copy on oth the
archives from /tmp to a filesystem on the raid just syncing.
OK that may be a lot of work for the system, and it may get slow, but it
may not crash!
I've failed to get a core image this time - sorry.
continue does not work and the system freezes in sync from DDB ...
Bx the way: I've saved the 8 GB core-image from the crash below, but
that one has still 96 page color active - the patch was missing. I think
it makes no sence to look at it at all.
If nobody came to me in the next few day with a request for it, I will
remove it.
W. Stukenbrock
Wolfgang Stukenbrock wrote:
> Hi,
>
> I've took the diff's from x86/x86/cpu.c rev 1.32 and merged them into
> amd64/amd64/cpu.c - there is no x86/x86/cpu.c in 4.0 ...
>
> It looks like it will solve the problem.
> Thanks
>
> The system will no longer freeze after using something around 4 GB
> memory (of the 8 GB installed ...).
>
> But I've recognized, that under (very) heavy load the system will panic
> with "out of memory" in the pagedaemon.
> I have a 8GB core file here - anybody interested in analysing ???. I
> think it will not pass through any mailing system ... (the bzip2
> compresseed version is still larger 2 GB (compression still running ...)
>
> In DDB there was exactly on page stated to be free in "show uvmexp" -
> the output follows:
>
> db{1}> show uvmexp
> Current UVM status:
> pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
> 2039036 VM pages: 1119946 active, 547095 inactive, 4031 wired, 1 free
> pages 1328086 anon, 341555 file, 1764 exec
> freemin=64, free-target=85, wired-max=679678
> faults=13246446, traps=18516116, intrs=17316633, ctxswitch=64031864
> softint=6934242, syscalls=65761334, swapins=431, swapouts=1372
> fault counts:
> noram=2008, noanon=0, pgwait=0, pgrele=0
> ok relocks(total)=2409(2418), anget(retrys)=12931894(1119),
> amapcopy=365963
> neighbor anon/obj pg=702333/4863720, gets(lock/unlock)=1135392/1299
> cases: anon=6302100, anoncow=256103, obj=947341, prcopy=188007,
> przero=24372
> 47
> daemon and swap counts:
> woke=9819, revs=6729, scans=2176414, obscans=1733276, anscans=143716
> busy=0, freed=1876669, reactivate=14780, deactivate=3760043
> pageouts=9330, pending=134669, nswget=1111
> nswapdev=1, swpgavail=6291455
> swpages=6291455, swpginuse=143978, swpgonly=142546, paging=323
> db{1}> trace
> cpu_Debugger() at netbsd:cpu_Debugger+0x5
> panic() at netbsd:panic+0x1f5
> pmap_growkernel() at netbsd:pmap_growkernel+0x446
> uvm_map_prepare() at netbsd:uvm_map_prepare+0x371
> uvm_map() at netbsd:uvm_map+0xae
> uvm_km_alloc() at netbsd:uvm_km_alloc+0x73
> vmem_xalloc() at netbsd:vmem_xalloc+0x130
> vmem_alloc() at netbsd:vmem_alloc+0x86
> amap_alloc() at netbsd:amap_alloc+0xdb
> uvm_map_enter() at netbsd:uvm_map_enter+0x24b
> uvm_map() at netbsd:uvm_map+0xfe
> sys_obreak() at netbsd:sys_obreak+0x106
> syscall_plain() at netbsd:syscall_plain+0x1fc
> uvm_fault(0xffff8000a666f440, 0x6907000, 1) -> e
> kernel: page fault trap, code=0
> Faulted in DDB; continuing...
> db{1}> continue
> syncing disks... 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 giving up
>
> dumping to dev 18,1 offset 33560487
> dump 8189 8188 8187 8186 8185 8184 8183 8182 8181 8180 8179 8178 8177 .....
>
> I've started more than 1000 processes to see what will happen if "some"
> memory is needed for processes. (cat ... | dd obs=... | dd obs=... | ...
> >/dev/null)
> The systems reduces the amount of memory used by the file-cache to
> something about 1,2 GB of the 8 GB main memory - as expected.
> At the time of the crash there was something around 800 MB of 24 GB swap
> space used.
>
> I know that this is not related to the previous problem. Does it make
> sence to create another bug report for that? I'm not shure about it.
>
> W. Stukenbrock
>
> Simon Burge wrote:
>
>> Wolfgang Stukenbrock wrote:
>>
>>
>>> The following reply was made to PR kern/39242; it has been noted by
>>> GNATS.
>>>
>>> From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
>>> To: gnats-bugs@NetBSD.org
>>> Cc: Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang
>>> on machines with more than 4 GB memory
>>> Date: Tue, 29 Jul 2008 09:07:09 +0200
>>>
>>> Hi,
>>>
>>> yes it is a E3110 CPU with 6MB cache.
>>> What files I need to catch from the current and integrate the changes
>>> into my 4.0-version of netbsd?
>>> You talk about the "cache detection stuff" AND the file named below.
>>>
>>> I think it would be a real great idea to bring this fix into the
>>> releases as soon as possible.
>>> This CPU is "very" cheep compared to the other one's (at least in
>>> germany) and therefore is very attractive for new systems.
>>>
>>
>> Does rev 1.32 of sys/arch/x86/x86/cpu.c apply cleanly to netbsd-4 ? I
>> don't recall which bits of which files moved around with x86 recently.
>> If so, that at least guarantees that a bogus number doesn't get passed
>> deeper into UVM, and that'll be enough for netbsd-4.
>>
>> Note also the problem was to do with "half memory used", not "4G of
>> memory used".
>>
>> Cheers,
>> Simon.
>>
>
>
From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Thu, 31 Jul 2008 10:24:28 +0200
Hi,
I think I've located the problem with the looping scsibus0 kernel process!
I've added some print in the uvm_plistalloc_simple() routine that will
start if the routine is gooing to sleep for more memory.
The problem happend after unpacking a tar file in the filesystem on the
SCSI-disks and then call sync to bring the cache to the disk.
I've got the following output:
plistalloc - waiting orig num 1 - num 1 low 0x1000000 high 0x100000000 -
free 245 pd_res 1 kres 5
plistalloc - loop fl 0 psi 3 ps-fl 1 num 1
plistalloc - loop fl 0 psi 2 ps-fl 1 num 1
plistalloc - loop fl 0 psi 1 ps-fl 0 num 1
plistalloc - loop fl 0 psi 0 ps-fl 0 num 1
plistalloc - loop fl 1 psi 3 ps-fl 1 num 1
plistalloc - loop fl 1 psi 2 ps-fl 1 num 1
plistalloc - loop fl 1 psi 1 ps-fl 0 num 1
plistalloc - loop fl 1 psi 0 ps-fl 0 num 1
plistalloc - waiting orig num 1 - num 1 low 0x1000000 high 0x100000000 -
free 245 pd_res 1 kres 5
plistalloc - loop fl 0 psi 3 ps-fl 1 num 1
plistalloc - loop fl 0 psi 2 ps-fl 1 num 1
plistalloc - loop fl 0 psi 1 ps-fl 0 num 1
plistalloc - loop fl 0 psi 0 ps-fl 0 num 1
plistalloc - loop fl 1 psi 3 ps-fl 1 num 1
plistalloc - loop fl 1 psi 2 ps-fl 1 num 1
plistalloc - loop fl 1 psi 1 ps-fl 0 num 1
plistalloc - loop fl 1 psi 0 ps-fl 0 num 1
plistalloc - waiting orig num 1 - num 1 low 0x1000000 high 0x100000000 -
free 245 pd_res 1 kres 5
plistalloc - loop fl 0 psi 3 ps-fl 1 num 1
plistalloc - loop fl 0 psi 2 ps-fl 1 num 1
plistalloc - loop fl 0 psi 1 ps-fl 0 num 1
plistalloc - loop fl 0 psi 0 ps-fl 0 num 1
plistalloc - loop fl 1 psi 3 ps-fl 1 num 1
plistalloc - loop fl 1 psi 2 ps-fl 1 num 1
plistalloc - loop fl 1 psi 1 ps-fl 0 num 1
plistalloc - loop fl 1 psi 0 ps-fl 0 num 1
...
endless gooing on ....
The controller seems to request one additional page in the range fom
0x1000000 to 0x100000000. This means an address below 4GB.
The page daemon is kicked, but does nothing, because pagedaemon does not
know anything about the range in which the required memory must reside.
This looks like a conceptual problem in 4.0 to me.
some additional information from DDB - after I get the system into the
debugger ...
Stopped in pid 28.1 (aiodoned) at netbsd:cpu_Debugger+0x5:
leave
db{0}> trace
cpu_Debugger() at netbsd:cpu_Debugger+0x5
comintr() at netbsd:comintr+0x6e0
Xintr_ioapic_edge4() at netbsd:Xintr_ioapic_edge4+0xd4
--- interrupt ---
_kernel_lock() at netbsd:_kernel_lock+0xad
intr_biglock_wrapper() at netbsd:intr_biglock_wrapper+0x16
Xintr_ioapic_level10() at netbsd:Xintr_ioapic_level10+0xd8
--- interrupt ---
Xspllower() at netbsd:Xspllower+0xe
DDB lost frame for netbsd:Xsoftclock+0x1a, trying 0xffff800056ff4db8
Xsoftclock() at netbsd:Xsoftclock+0x1a
--- interrupt ---
0x56ff4e30:
db{0}> show uvmexp
Current UVM status:
pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
2039030 VM pages: 1276109 active, 623406 inactive, 2284 wired, 372 free
pages 1470432 anon, 429766 file, 1604 exec
freemin=64, free-target=85, wired-max=679676
faults=6949833, traps=7652305, intrs=16596032, ctxswitch=30047566
softint=6211694, syscalls=17975420, swapins=287, swapouts=306
fault counts:
noram=112, noanon=0, pgwait=11, pgrele=0
ok relocks(total)=1590(1591), anget(retrys)=1256742(505),
amapcopy=452013
neighbor anon/obj pg=900772/6909020, gets(lock/unlock)=1606322/1086
cases: anon=966859, anoncow=288741, obj=1336907, prcopy=269411,
przero=86586
0
daemon and swap counts:
woke=67411, revs=5048, scans=1391672, obscans=1386891, anscans=2492
busy=0, freed=1389383, reactivate=103, deactivate=2015339
pageouts=89708, pending=1299675, nswget=99734
nswapdev=1, swpgavail=6291455
swpages=6291455, swpginuse=1389021, swpgonly=1289440, paging=0
db{0}>
I failed to get a stack-listing of the scsibus0 process
ps output:
9 0 0 0 2 0x20200 1 scsibus0
but "trace/t 9" hangs up and I'm not able to get back into DDB
For better understanding - the printf() statments I've inserted into
uvm_plistalloc_simple():
XXX - start of modified routine ...
static int
uvm_pglistalloc_simple(int num, paddr_t low, paddr_t high,
struct pglist *rlist, int waitok)
{
int fl, psi, s, error;
struct vm_physseg *ps;
int o_num = num;
int xx = 0;
/* Default to "lose". */
error = ENOMEM;
again:
/*
* Block all memory allocation and lock the free list.
*/
s = uvm_lock_fpageq();
/* Are there even any free pages? */
if (uvmexp.free <= (uvmexp.reserve_pagedaemon +
uvmexp.reserve_kernel))
goto out;
for (fl = 0; fl < VM_NFREELIST; fl++) {
#if (VM_PHYSSEG_STRAT == VM_PSTRAT_BIGFIRST)
for (psi = vm_nphysseg - 1 ; psi >= 0 ; psi--)
#else
for (psi = 0 ; psi < vm_nphysseg ; psi++)
#endif
{
ps = &vm_physmem[psi];
if (xx != 0) printf("plistalloc - loop fl %d psi %d ps-fl %d num %d\n",
fl, psi, ps->free_list, num);
if (ps->free_list != fl)
continue;
num -= uvm_pglistalloc_s_ps(ps, num, low, high,
rlist);
if (num == 0) {
error = 0;
goto out;
}
}
}
out:
/*
* check to see if we need to generate some free pages waking
* the pagedaemon.
*/
uvm_kick_pdaemon();
uvm_unlock_fpageq(s);
if (error) {
if (waitok) {
/* XXX perhaps some time limitation? */
#ifdef DEBUG
printf("pglistalloc waiting\n");
#endif
printf("plistalloc - waiting orig num %d - num %d low 0x%lx high 0x%lx -
free %d pd_res %d kres %d\n",
o_num, num, low, high, uvmexp.free, uvmexp.reserve_pagedaemon,
uvmexp.reserve_kernel); xx=1;
uvm_wait("pglalloc");
goto again;
} else
uvm_pglistfree(rlist);
}
#ifdef PGALLOC_VERBOSE
if (!error)
printf("pgalloc: %lx..%lx\n",
VM_PAGE_TO_PHYS(TAILQ_FIRST(rlist)),
VM_PAGE_TO_PHYS(TAILQ_LAST(rlist, pglist)));
#endif
return (error);
}
XXX - end of modified routine ...
I thing the we need something like a list of ranges where memory is need
and the pagedaemon should free some memory in that range.
At the moment, any system with more than 4 GB RAM may get into this
problem - the SCSI-controller is an Adaptec 29160A-R
I will remove 4 GB of memory from the system in order to get it stable,
but that cannot be the final sollution.
The problem switches back from "CPU-cache-problem" to my initial subject
for some supported controlers.
best regards
W. Stukenbrock
Wolfgang Stukenbrock wrote:
> The following reply was made to PR kern/39242; it has been noted by GNATS.
>
> From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
> To: gnats-bugs@NetBSD.org
> Cc: Simon Burge <simonb@NetBSD.org>, kern-bug-people@NetBSD.org
> Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
> Date: Wed, 30 Jul 2008 20:13:32 +0200
>
> Hi again,
>
> I've seen responce to my last mail directly to Simon, but I continue
> testing my system.
>
> here the patch I've added to /usr7src/sys/amd64/amd64/cpu.c:
>
> s012# rcsdiff -c -r1.1 c*
> ===================================================================
> RCS file: RCS/cpu.c,v
> retrieving revision 1.1
> diff -c -r1.1 cpu.c
> *** cpu.c 2008/07/29 07:28:03 1.1
> --- cpu.c 2008/07/29 08:05:21
> ***************
> *** 217,222 ****
> --- 217,239 ----
> tcolors /= cai->cai_associativity;
> }
> ncolors = max(ncolors, tcolors);
> + /*
> + * If the desired number of colors is not a power of
> + * two, it won't be good. Find the greatest power of
> + * two which is an even divisor of the number of colors,
> + * to preserve even coloring of pages.
> + */
> + if (ncolors & (ncolors - 1) ) {
> + int try, picked = 1;
> + for (try = 1; try < ncolors; try *= 2) {
> + if (ncolors % try == 0) picked = try;
> + }
> + if (picked == 1) {
> + panic("desired number of cache colors %d
> is "
> + " > 1, but not even!", ncolors);
> + }
> + ncolors = picked;
> + }
> }
>
> /*
>
>
> Just some minutes ago, I've got two new kernel crashes.
>
> 1. the kernel process [scsibus0] starts looping and sleeps sometimes in
> pglalloc. After reinstalling the system with the patch above, there was
> no DDB in the kernel, so I could not get any other information.
>
> 2. now I've DDB in the kernel and tried to reproduce the problem. But it
> crashes prior reaching this state in pagedaemon ...
> Some output from the console below:
>
>
> uvm_fault(0xffffffff80628800, 0x0, 1) -> e
> kernel: page fault trap, code=0
> Stopped in pid 26.1 (pagedaemon) at netbsd:uvm_rb_insert+0x37:
> movq 0
> x40(%rax),%rax
> db{0}> trace
> uvm_rb_insert() at netbsd:uvm_rb_insert+0x37
> uvm_map_enter() at netbsd:uvm_map_enter+0x290
> uvm_map() at netbsd:uvm_map+0xfe
> uvm_pagermapin() at netbsd:uvm_pagermapin+0x92
> uvm_swap_io() at netbsd:uvm_swap_io+0x3c
> swapcluster_flush() at netbsd:swapcluster_flush+0x55
> uvm_pageout() at netbsd:uvm_pageout+0x42b
> db{0}> show uvmexp
> Current UVM status:
> pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
> 2039030 VM pages: 1333964 active, 651446 inactive, 2248 wired, 5 free
> pages 1674503 anon, 311641 file, 1580 exec
> freemin=64, free-target=85, wired-max=679676
> faults=3420439, traps=3975116, intrs=8916183, ctxswitch=14697719
> softint=710475, syscalls=12629494, swapins=177, swapouts=210
> fault counts:
> noram=5, noanon=0, pgwait=0, pgrele=0
> ok relocks(total)=1068(1069), anget(retrys)=73451(286), amapcopy=31017
> neighbor anon/obj pg=65282/474220, gets(lock/unlock)=116248/783
> cases: anon=51370, anoncow=21184, obj=96126, prcopy=20121, przero=63798
> daemon and swap counts:
> woke=32081, revs=4228, scans=1204885, obscans=1161315, anscans=2987
> busy=0, freed=1164012, reactivate=128, deactivate=1857899
> pageouts=75209, pending=1088875, nswget=78493
> nswapdev=1, swpgavail=6291455
> swpages=6291455, swpginuse=1164021, swpgonly=1085489, paging=66
> db{0}>
>
>
> This patch seems to enable the system to work with the 6MB cache of the
> E3110 CPU, but the kernel is not realy stable at all.
> Any idea? What should I try next?
>
> By the way "vmstat -s | grep colo" reports 32 colors now.
> The system was busy again when the crash happens. raidframe sync on one
> SATA-raid and one SCSI-raid, transfered something arund 12 GB int /tmp
> (tmpfs) so nearly 5 GB of the 24 GB swap was used.
> The kernel crashes at the moment where I've tried to copy on oth the
> archives from /tmp to a filesystem on the raid just syncing.
>
> OK that may be a lot of work for the system, and it may get slow, but it
> may not crash!
> I've failed to get a core image this time - sorry.
> continue does not work and the system freezes in sync from DDB ...
>
> Bx the way: I've saved the 8 GB core-image from the crash below, but
> that one has still 96 page color active - the patch was missing. I think
> it makes no sence to look at it at all.
> If nobody came to me in the next few day with a request for it, I will
> remove it.
>
>
> W. Stukenbrock
>
> Wolfgang Stukenbrock wrote:
>
> > Hi,
> >
> > I've took the diff's from x86/x86/cpu.c rev 1.32 and merged them into
> > amd64/amd64/cpu.c - there is no x86/x86/cpu.c in 4.0 ...
> >
> > It looks like it will solve the problem.
> > Thanks
> >
> > The system will no longer freeze after using something around 4 GB
> > memory (of the 8 GB installed ...).
> >
> > But I've recognized, that under (very) heavy load the system will panic
> > with "out of memory" in the pagedaemon.
> > I have a 8GB core file here - anybody interested in analysing ???. I
> > think it will not pass through any mailing system ... (the bzip2
> > compresseed version is still larger 2 GB (compression still running ...)
> >
> > In DDB there was exactly on page stated to be free in "show uvmexp" -
> > the output follows:
> >
> > db{1}> show uvmexp
> > Current UVM status:
> > pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
> > 2039036 VM pages: 1119946 active, 547095 inactive, 4031 wired, 1 free
> > pages 1328086 anon, 341555 file, 1764 exec
> > freemin=64, free-target=85, wired-max=679678
> > faults=13246446, traps=18516116, intrs=17316633, ctxswitch=64031864
> > softint=6934242, syscalls=65761334, swapins=431, swapouts=1372
> > fault counts:
> > noram=2008, noanon=0, pgwait=0, pgrele=0
> > ok relocks(total)=2409(2418), anget(retrys)=12931894(1119),
> > amapcopy=365963
> > neighbor anon/obj pg=702333/4863720, gets(lock/unlock)=1135392/1299
> > cases: anon=6302100, anoncow=256103, obj=947341, prcopy=188007,
> > przero=24372
> > 47
> > daemon and swap counts:
> > woke=9819, revs=6729, scans=2176414, obscans=1733276, anscans=143716
> > busy=0, freed=1876669, reactivate=14780, deactivate=3760043
> > pageouts=9330, pending=134669, nswget=1111
> > nswapdev=1, swpgavail=6291455
> > swpages=6291455, swpginuse=143978, swpgonly=142546, paging=323
> > db{1}> trace
> > cpu_Debugger() at netbsd:cpu_Debugger+0x5
> > panic() at netbsd:panic+0x1f5
> > pmap_growkernel() at netbsd:pmap_growkernel+0x446
> > uvm_map_prepare() at netbsd:uvm_map_prepare+0x371
> > uvm_map() at netbsd:uvm_map+0xae
> > uvm_km_alloc() at netbsd:uvm_km_alloc+0x73
> > vmem_xalloc() at netbsd:vmem_xalloc+0x130
> > vmem_alloc() at netbsd:vmem_alloc+0x86
> > amap_alloc() at netbsd:amap_alloc+0xdb
> > uvm_map_enter() at netbsd:uvm_map_enter+0x24b
> > uvm_map() at netbsd:uvm_map+0xfe
> > sys_obreak() at netbsd:sys_obreak+0x106
> > syscall_plain() at netbsd:syscall_plain+0x1fc
> > uvm_fault(0xffff8000a666f440, 0x6907000, 1) -> e
> > kernel: page fault trap, code=0
> > Faulted in DDB; continuing...
> > db{1}> continue
> > syncing disks... 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 giving up
> >
> > dumping to dev 18,1 offset 33560487
> > dump 8189 8188 8187 8186 8185 8184 8183 8182 8181 8180 8179 8178 8177 .....
> >
> > I've started more than 1000 processes to see what will happen if "some"
> > memory is needed for processes. (cat ... | dd obs=... | dd obs=... | ...
> > >/dev/null)
> > The systems reduces the amount of memory used by the file-cache to
> > something about 1,2 GB of the 8 GB main memory - as expected.
> > At the time of the crash there was something around 800 MB of 24 GB swap
> > space used.
> >
> > I know that this is not related to the previous problem. Does it make
> > sence to create another bug report for that? I'm not shure about it.
> >
> > W. Stukenbrock
> >
> > Simon Burge wrote:
> >
> >> Wolfgang Stukenbrock wrote:
> >>
> >>
> >>> The following reply was made to PR kern/39242; it has been noted by
> >>> GNATS.
> >>>
> >>> From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
> >>> To: gnats-bugs@NetBSD.org
> >>> Cc: Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang
> >>> on machines with more than 4 GB memory
> >>> Date: Tue, 29 Jul 2008 09:07:09 +0200
> >>>
> >>> Hi,
> >>>
> >>> yes it is a E3110 CPU with 6MB cache.
> >>> What files I need to catch from the current and integrate the changes
> >>> into my 4.0-version of netbsd?
> >>> You talk about the "cache detection stuff" AND the file named below.
> >>>
> >>> I think it would be a real great idea to bring this fix into the
> >>> releases as soon as possible.
> >>> This CPU is "very" cheep compared to the other one's (at least in
> >>> germany) and therefore is very attractive for new systems.
> >>>
> >>
> >> Does rev 1.32 of sys/arch/x86/x86/cpu.c apply cleanly to netbsd-4 ? I
> >> don't recall which bits of which files moved around with x86 recently.
> >> If so, that at least guarantees that a bogus number doesn't get passed
> >> deeper into UVM, and that'll be enough for netbsd-4.
> >>
> >> Note also the problem was to do with "half memory used", not "4G of
> >> memory used".
> >>
> >> Cheers,
> >> Simon.
> >>
> >
> >
>
>
>
From: Wolfgang Stukenbrock <Wolfgang.Stukenbrock@nagler-company.com>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/39242: NetBSD 4.0 will start busy-loop an hang on machines with more than 4 GB memory
Date: Thu, 31 Jul 2008 12:56:33 +0200
Hi - once again ...
I've removed 4 GB of memory now and the SCSI-controller seems to work
now, but ....
kernel: protection fault trap, code=0
Stopped in pid 28.1 (aiodoned) at netbsd:uvm_tree_RB_REMOVE+0x50:
movq %
r14,0x10(%r15)
db{0}> trace
uvm_tree_RB_REMOVE() at netbsd:uvm_tree_RB_REMOVE+0x50
uvm_rb_remove() at netbsd:uvm_rb_remove+0x1c
uvm_unmap_remove() at netbsd:uvm_unmap_remove+0x179
uvm_pagermapout() at netbsd:uvm_pagermapout+0x110
uvm_aio_aiodone() at netbsd:uvm_aio_aiodone+0xc4
uvm_aiodone_daemon() at netbsd:uvm_aiodone_daemon+0xd2
db{0}>
At the time of the crash top reports:
load averages: 1.03, 0.47, 0.19
up 0 days, 0:38 12:27:08
67 processes: 1 runnable, 64 sleeping, 2 on processor
CPU0 states: 0.0% user, 0.0% nice, 11.9% system, 2.5% interrupt,
85.6% idle
CPU1 states: 0.0% user, 0.0% nice, 3.0% system, 0.0% interrupt,
97.0% idle
Memory: 2565M Act, 1253M Inact, 8924K Wired, 6296K Exec, 1316M File,
228K Free
Swap: 24G Total, 9435M Used, 15G Free
PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
10722 root -5 0 868K 1792K biowai/0 0:13 78.93% 7.52% tar
23 root -6 0 0K 28M RUN/1 0:10 5.37% 5.37%
[raidio0]
26 root -18 0 0K 28M pgdaem/0 0:04 1.71% 1.71%
[pagedaemon]
28 root -18 0 0K 28M aiodon/0 0:04 1.17% 1.17%
[aiodoned]
21 root -6 0 0K 28M raidio/1 0:05 1.07% 1.07%
[raidio1]
9957 root 28 0 120K 820K CPU/1 0:00 0.00% 0.98% cp
22 root -6 0 0K 28M rfwcon/0 0:01 0.68% 0.68% [raid0]
20 root -6 0 0K 28M rfwcon/0 0:00 0.05% 0.05% [raid1]
9 root -6 0 0K 28M sccomp/1 0:02 0.00% 0.00%
[scsibus0]
198 ncadmin 28 0 572K 1588K CPU/0 0:00 0.00% 0.00% top
27 root 18 0 0K 28M syncer/0 0:00 0.00% 0.00%
[ioflush]
816 root 18 0 988K 4188K pause/0 0:00 0.00% 0.00% ntpd
9303 wgstuken 18 0 260K 1168K pause/1 0:00 0.00% 0.00% <csh>
10024 root 18 0 236K 1144K pause/1 0:00 0.00% 0.00% <csh>
1515 root 18 0 240K 1088K pause/1 0:00 0.00% 0.00% <csh>
18 root 14 0 0K 28M crypto/0 0:00 0.00% 0.00%
[cryptoret]
13 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb7]
12 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb6]
11 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb5]
10 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb4]
571 root 10 0 0K 28M nfsidl/0 0:00 0.00% 0.00% [nfsio]
570 root 10 0 0K 28M nfsidl/0 0:00 0.00% 0.00% [nfsio]
545 root 10 0 0K 28M nfsidl/0 0:00 0.00% 0.00% [nfsio]
499 root 10 0 0K 28M nfsidl/0 0:00 0.00% 0.00% [nfsio]
3 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb0]
4 root 10 0 0K 28M usbtsk/0 0:00 0.00% 0.00%
[usbtask-hc]
5 root 10 0 0K 28M usbtsk/0 0:00 0.00% 0.00%
[usbtask-dr]
6 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb1]
7 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb2]
8 root 10 0 0K 28M usbevt/0 0:00 0.00% 0.00% [usb3]
10303 root 10 0 484K 2768K wait/0 0:00 0.00% 0.00% <login>
195 root 10 0 620K 1944K wait/1 0:00 0.00% 0.00% <login>
1566 root 10 0 476K 1936K wait/0 0:00 0.00% 0.00% <login>
196 ncadmin 10 0 284K 1048K wait/0 0:00 0.00% 0.00% <sh>
1443 root 10 0 268K 880K nanosl/1 0:00 0.00% 0.00% <cron>
1 root 10 0 100K 812K wait/0 0:00 0.00% 0.00% <init>
557 root 2 0 1060K 3988K select/0 0:00 0.00% 0.00% <amd>
10740 root 2 0 492K 1536K poll/0 0:00 0.00% 0.00% <rlogind>
872 root 2 0 344K 1456K select/0 0:00 0.00% 0.00% <sshd>
1385 postfix 2 0 628K 1216K kqread/0 0:00 0.00% 0.00% <qmgr>
There seems to be a bigger problem in uvm as expected by me before!
db{0}> show uvmexp
Current UVM status:
pagesize=4096 (0x1000), pagemask=0xfff, pageshift=12
1018418 VM pages: 656891 active, 320807 inactive, 2230 wired, 5 free
pages 634067 anon, 344367 file, 1574 exec
freemin=64, free-target=85, wired-max=339472
faults=3609406, traps=3722187, intrs=7567575, ctxswitch=9890365
softint=380503, syscalls=14268860, swapins=245, swapouts=270
fault counts:
noram=24, noanon=0, pgwait=24, pgrele=0
ok relocks(total)=1148(1150), anget(retrys)=114762(394), amapcopy=55274
neighbor anon/obj pg=115362/837538, gets(lock/unlock)=198035/756
cases: anon=77487, anoncow=37248, obj=164994, prcopy=33039,
przero=98761
daemon and swap counts:
woke=15991, revs=10979, scans=2430998, obscans=2420929, anscans=2524
busy=0, freed=2423039, reactivate=730, deactivate=2791838
pageouts=156825, pending=2266288, nswget=297605
nswapdev=1, swpgavail=6291455
swpages=6291455, swpginuse=2422999, swpgonly=2125395, paging=78
db{0}> ps
PID PPID PGRP UID S FLAGS LWPS COMMAND
WAIT
9957 1515 9957 0 2 0x4002 1 cp
uvn_fp1
10722 10024 10722 0 2 0x4002 1 tar
10024 9303 10024 0 2 0x4002 1 csh
pause
9303 10303 9303 1002 2 0x4002 1 csh
pause
10303 10740 10303 0 2 0x4103 1 login
wait
10740 1262 1262 0 2 0x4100 1 rlogind
poll
198 196 198 500 2 0x4002 1 top
poll
196 195 196 500 2 0x4002 1 sh
wait
195 194 195 0 2 0x4103 1 login
wait
194 1262 1262 0 2 0x4100 1 rlogind
poll
1515 1566 1515 0 2 0x4002 1 csh
pause
1566 1 1566 0 2 0x4102 1 login
wait
1443 1 1443 0 2 0 1 cron
nanosle
1262 1 1262 0 2 0 1 inetd
kqread
1385 1263 1263 12 2 0x4108 1 qmgr
kqread
950 1263 1263 12 2 0x4108 1 pickup
kqread
1263 1 1263 0 2 0x4108 1 master
kqread
872 1 872 0 2 0 1 sshd
select
816 1 816 0 2 0 1 ntpd
pause
757 1 757 0 2 0 1 lpd
poll
631 1 631 0 2 0 1 rpc.lockd
select
672 1 672 0 2 0xa0008 1 rpc.statd
select
682 614 614 0 2 0 1 nfsd
nfsd
677 614 614 0 2 0 1 nfsd
nfsd
637 614 614 0 2 0 1 nfsd
nfsd
671 614 614 0 2 0 1 nfsd
nfsd
614 1 614 0 2 0 1 nfsd
poll
615 1 615 0 2 0 1 mountd
select
571 0 0 0 2 0x20200 1 nfsio
nfsidl
570 0 0 0 2 0x20200 1 nfsio
nfsidl
545 0 0 0 2 0x20200 1 nfsio
nfsidl
499 0 0 0 2 0x20200 1 nfsio
nfsidl
557 1 557 0 2 0 1 amd
select
505 1 505 0 2 0 1 ypbind
select
497 1 497 0 2 0 1 rpcbind
poll
574 1 574 0 2 0 1 syslogd
kqread
248 1 248 0 2 0 1 routed
select
108 0 0 0 2 0x20200 1 physiod
physiod
>28 0 0 0 2 0x20200 1 aiodoned
27 0 0 0 2 0x20200 1 ioflush
syncer
26 0 0 0 2 0x20200 1 pagedaemon
pgdaemo
25 0 0 0 2 0x20200 1 raidio2
raidiow
24 0 0 0 2 0x20200 1 raid2
rfwcond
23 0 0 0 2 0x20200 1 raidio0
22 0 0 0 2 0x20200 1 raid0
21 0 0 0 2 0x20200 1 raidio1
20 0 0 0 2 0x20200 1 raid1
rfwcond
19 0 0 0 2 0x20200 1 atapibus0
sccomp
18 0 0 0 2 0x20200 1 cryptoret
crypto_
17 0 0 0 2 0x20200 1 atabus3
atath
16 0 0 0 2 0x20200 1 atabus2
atath
15 0 0 0 2 0x20200 1 atabus1
atath
14 0 0 0 2 0x20200 1 atabus0
atath
13 0 0 0 2 0x20200 1 usb7
usbevt
12 0 0 0 2 0x20200 1 usb6
usbevt
11 0 0 0 2 0x20200 1 usb5
usbevt
10 0 0 0 2 0x20200 1 usb4
usbevt
9 0 0 0 2 0x20200 1 scsibus0
sccomp
8 0 0 0 2 0x20200 1 usb3
usbevt
7 0 0 0 2 0x20200 1 usb2
usbevt
6 0 0 0 2 0x20200 1 usb1
usbevt
5 0 0 0 2 0x20200 1 usbtask-dr
usbtsk
4 0 0 0 2 0x20200 1 usbtask-hc
usbtsk
3 0 0 0 2 0x20200 1 usb0
usbevt
2 0 0 0 2 0x20200 1 sysmon
smtaskq
1 0 1 0 2 0x4001 1 init
wait
0 -1 0 0 2 0x20200 1 swapper
schedul
db{0}>
At the time of the crash "tar" is extracting an archive into one
filesystem and "cp" is copying a large file into another filesystem -
both on the same raid-device (raid1 on SATA) - but that has caused no
problem in the past.
The source of both commands is in /tmp (tmpfs) - ca. 11 GB resides in /tmp.
Is tmpfs known to be stable? Or may the problem be there?
Any Idea how to debug further on?
I will see if I'm able to replace tmpfs by a real filesystem and try
again ...
By the way: I've recognized another problem in uvm_plistalloc_simple()
by thinking about the strategie used there.
If some processes tries to allocate more more memory in sum as the
system has at all, the strategie used there may deadlock the whole system.
If e.g. 30 % of memory is given to each of tree processes that requests
each 50% of the whole memory, it will be impossible for any of them to
complete because the required memory is locked by the other processes
and cannot be stolen there (there is no way to manipulate the local
variable "num" from outside ...). It would be possible to satisfy the
request of the first on and after it has freed the memory again the next
and so on. Therefore a way is needed to "steal" steal the memory from
waiting processes again.
This would be a very rare case and strange situation, but it is not
detected by the system at the moment. I'm not shure if there is an
inexpensive and easy way to detect such a situation, but this "possible"
problem should be at least documented in a comment in the source file.
best regards
W. Stukenbrock
PS. I've removed the privious mail content - it is in the gnats system
and can be reviewed there.
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.39 2013/11/01 18:47:49 spz Exp $
$NetBSD: gnats_config.sh,v 1.8 2006/05/07 09:23:38 tsutsui Exp $
Copyright © 1994-2007
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.