It sounds like there are at least two issues involved.
The first could be a buffer cache starvation issue due to the load on
the filesystem from the tar. If the usb program is doing any filesystem
operation at all, even at low bandwidths, it could be hitting blockages
due to the
Matthew Dillon
dil...@backplane.com
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to freebsd-hackers
by
the buffer cache.
That limit is completely irrelevant now and should probably be set to
0x7FFFLLU (since seek offsets are signed).
-Matt
Matthew Dillon
dil
:On Sat, Dec 18, 2010 at 02:24:34PM -0800, Matthew Dillon wrote:
: Does anyone know if an IRET cancels/triggers a MONITOR event?
:
:AMD's Architecture Programmer's Manual explicitly contains:
:
:Events that cause an exit from the monitor event pending state include:
:...
:- Any far control
Does anyone know if an IRET cancels/triggers a MONITOR event? Here's
the problem:
(1) main line kernel code is executing a MONITOR/MWAIT sequence.
It executes its MONITOR but has not yet gotten to the MWAIT.
(2) An interrupt occurs inbetween the MONITOR and the MWAIT.
:cronfy cro...@gmail.com wrote:
:
: And also, maybe there are other ways to create incremental backups
: instead of using rsync/hardlinks?
:
:Yes. Use dump(8) -- that's what it's for. It reads the inodes,
:directories, and files directly from the disk device, thereby
:eliminating stat()
Matthew Dillon
dil...@backplane.com
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
.
It is possible for files to be caught mid-change but also fairly
easy to detect the case if it winds up being a problem. And, of
course, more sophisticated methodologies can be built on top.
-Matt
Matthew Dillon
:void
:waitrunningbufspace(void)
:{
:/*
:mtx_lock(rbreqlock);
:while (runningbufspace hirunningspace) {
:++runningbufreq;
:msleep(runningbufreq, rbreqlock, PVM, wdrain, 0);
:}
:mtx_unlock(rbreqlock);
:*/
:}
:
:so far, I can't
:All of these tests have been apples vs. oranges for years.
:
:The following seems to be true, though:
:
:a) FreeBSD sequential write performance in UFS has always been less than
:optimal.
If there's no read activity sequential write performance should be
maximal with UFS. The
: It would be interesting to see a blogbench comparison between UFS
: and ZFS on the same hw/disk.
:
:
:I'll do it, just tell me how do you want to run the tests.
:
:The system params are:
:
:8GB Memory
:2x72GB SCSI HDD
:2x3.4Ghz Xeon
:Overall: Dell Poweredge 1850. With no raid installed.
We experimented a bit with aligning fdisk (dos slices) by changing
the sector offset to 2 but I came to the conclusion that it was better
to do the alignment in disklabel / gpt / whatever higher-level
partitioner floats your boat and not mess with anything the BIOS
uses to boot
I'll note one last thing with regards to write combining within the
drive's zone cache. Drive zone caches work very well for combing
adjacent sectors when the write zones are perfectly linear (when
the writes within each zone being tracked are perfectly linear).
But the drive
: There is a sysctl, md_compress, that I turned out in my tests, but not
:working as expected.
: Why using gnop -S 4096 works well?
:
:Thiago
You are setting the sector size to 4K with gnop -S 4096 so presumably
ZFS will not do any fragmented writes smaller than that. I'm not
sure
Play with the read-ahead mount options for NFS, but it might require
more work with that kind of latency. You need to be able to have
a lot of RPC's in-flight to maintain the pipeline with higher latencies.
At least 16 and possibly more.
It might be easier to investigate why
I'm just covering all the bases. To be frank, half the time when
someone posts they are doing something a certain way it turns out that
they actually aren't. I've learned that covering the bases tends to
lead to solutions more quickly than assuming a perfect rendition.
For
Oh, one more thing... I'm assuming you haven't used tcpdump with
NFS much. tcpdump has issues parsing the NFS RPC's out of a TCP
stream. For the purposes of testing you may want to temporarily
use a UDP NFS mount. tcpdump can parse the NFS RPCs out of the UDP
stream far more
:Hello
:
:This is regarding the dump utility cache efficiency analysis post made on
:February '07 by Peter Jeremy [
:http://lists.freebsd.org/pipermail/freebsd-hackers/2007-February/019666.html]
:and if this project is still open. I would be interested to begin exploring
:FreeBSD (and
:On Thursday 21 May 2009 23:37:20 Nate Eldredge wrote:
: Of course all these problems are solved, under any policy, by having more
: memory or swap. =A0But overcommit allows you to do more with less.
:
:Or to put it another way, 90% of the problems that could be solved by havin=
:g=20
:more
There is no such thing as a graceful way to deal with running out of
memory. What is a program supposed to do? Even if it gracefully exits
it still fails to perform the function for which it was designed. If
such a program is used in a script then the script fails as well.
I'll put in a plug for using DragonFly's pluggable scheduler
framework :-). We (DragonFly) also offer shell accounts, git
integration and publishing, a virtual kernel build/run environment
for doing kernel projects, and help over IRC and email.
Someone with the gumption to
any more.
They just have '/' and maybe a small boot partition, and that's it.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd
Go here:
http://forums.mysql.com/read.php?28,197644,197644
There are a ton of ways to maintain mysql backups without having to
replay the entire log. Google some keywords.
With regards to solutions based on filesystem snapshots, such as partial
log replaying
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL
: -vfs.nfs.realign_test: 22141777
: +vfs.nfs.realign_test: 498351
:
: -vfs.nfsrv.realign_test: 5005908
: +vfs.nfsrv.realign_test: 0
:
: +vfs.nfsrv.commit_miss: 0
: +vfs.nfsrv.commit_blks: 0
:
: changing them did nothing - or at least with respect to nfs throughput
automatically but I'm not sure).
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
:well, not really, at least not the name by which it was looked up.
:you MIGHT (sometimes) be able to use the directory name cache to work
:it out.. At one stage it was possible to do this for some percentage
:of the files but I dont remember if it was possible in 4.x.
:
:the idea is that you
:On Friday 27 June 2008 10:43:29 Roman Divacky wrote:
: hi
:
: I have two questions:
:
: 1) is kmem_alloc_wait() expensive operation? I believe it's not
: very cheap looking at the code but I want confirmation
:
: 2) is there a support for memory pools in FreeBSD?
:
: to give you a little
You can do it for outgoing connections fairly easily using the NAT
trick (with PF), but you can't really load balance multiple links
without support from some outside entity. If one of the tunnels goes
down you can fail-over but any pre-existing connections will die and
have
... how can that possibly be reliable?
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL
vm_map_protect()
as long as you properly handle the vm_map_entry.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
:Matthew Dillon wrote:
: :Matt,
:...
: :Don't you use something like VMWare for development and debugging?
:
: We use vkernel's for development and debugging. Pretty much everything
: except hardware device driver development can be done using a vkernel...
:
:Does that include trying
:Matt,
:
:We use VMWare Server at work. It does not have the same nice image management
interface and/or video capture as commercial counterparts. However, it is is
free and testing on it helps us out big time. We never concluded whether it
maked sense to pay for VMWare licenses, instead of
programmers trying to read and understand it.
All in all, it was a very good move for the project and I would
strongly recommend that FreeBSD do the same thing.
-Matt
Matthew Dillon
:Matt, I'm sorry I'm not trying to hijack this thread but isn't the vkernel
:approach very similar to VMWare's hosted architecture products (such as
:Fusion for the Mac and Client Workstation for windows)?
:
:As I understand it, they have a regular process like vkernel called
:vmware-vmx which
are always
welcome in our corner of the woods.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http
.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send
. That would cut the overhead in half.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http
?
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail
Basically DragonFly has a syscall API that allows a userland process
to create and completely control any number of VM spaces, including
the ability to pass execution control to a VM space and get it back,
and control memory mappings within that VM space (and in the virtual
:
:Given the fact that there are not as many developers as needed, what would be
a practical purpose of vkernel?
:
:UML is typically used to debug drivers and/or for hosting. Now that Linux
about to have or already has container technology, hosting on UML makes little
sense.
The single
for the
emulated processes you need the code to swap VM spaces for a process.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers
*() functions.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo
booting
the live CD so those bits don't have to be re-read. Doing a direct
copy has always felt 'faster' to me then unpacking split up tar files.
-Matt
Matthew Dillon
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
We added it basically because doing all the junk described in
previous postings in this thread in userland is a ridiculously huge
eyesore that doesn't scale and doesn't make sense when 5 minutes of
programming nets you a shiny new system call which does it all for you.
If you
for allocations up to and including PAGE_SIZE*2 bytes.
Fun, eh?
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org
to point the MBR at sector 40.
The more I think about it, the more sense it makes.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd
.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo
:No.
:The first partition is the EFI GPT (0xee):
:
:% fdisk -1
:*** Working on device /dev/ad0 ***
:...
:parameters to be used for BIOS calculations are:
:cylinders=116280 heads=16 sectors/track=63 (1008 blks/cyl)
:
:Media sector size is 512
:Warning: BIOS sector numbering starts with
I'm having to tackle this issue right now in DFly. With GPT having to
start at sector 1 (no choice there), the compatible MBR in sector 0
presumably must have a slice (#1) which covers the entire disk.
But do we have to make slice #1 bootable? Could we also create a
slice #2
The real culprit here is passing held mutexes to unrelated procedures
in the first place because those procedures might have to block, in
order so those procedures can release and reacquire the mutex.
That's just bad coding in my view. The unrelated procedure has no
clue as to
The TSCs for each individual cpu core can drift relative to each other,
even on multi-core chips like AMD X2s. This only effects code which
uses the TSC, which isn't a whole lot. They need to be synchronized
with each other (by calculating the drift and correcting for it) when
.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail
scratching my head a bit. This may be just as broken
as before, just in a different (and safer) way.
-Matt
Matthew Dillon
[EMAIL PROTECTED
It is interesting to note that DragonFly has had variant symlnks for
a long time, but we haven't actually found a use for them yet. No
smoking killer app has presented itself. We use varsyms in our RC
implementation kind of like uber-environment variables (rclist, rcstart,
:...
:BSD behaviour:
:- OpenBSD handles hardlinks since 3.3:
: -P Overwrite regular files before deleting them. Files
:are overwritten three times, first with the byte pattern
:0xff, then 0x00, and then 0xff again, before they are
:deleted. Files
:
:* Matthew Dillon ([EMAIL PROTECTED]) wrote:
:
: felt that 8 partitions is restrictive. My main home server has 10
: and the main DragonFly box has 11.
:
: There is another solution for FreeBSD folks, however. You *DO* have
: four slices to play with. You can put
:
: Dmitry == Dmitry Marakasov [EMAIL PROTECTED] writes:
:
:Dmitry * Matthew Dillon ([EMAIL PROTECTED]) wrote:
: felt that 8 partitions is restrictive. My main home server has 10
: and the main DragonFly box has 11.
:
: There is another solution for FreeBSD folks, however. You *DO*
: have four
.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail
:AFAIR the DFly FPU rework allows to use FPU/XMM instructions in their
:kernel without the need to do some manual state preserving (it's done
:...
:
:Bye,
:Alexander.
That actually isn't quite how it works. If the userland had active
FP state then the kernel still has to save it
:Hello Matt!
:
:I hope you'll make the materials available on the Net.
Yah, I'm putting some slides together and will make them available
after the talk.
-Matt
___
freebsd-hackers@freebsd.org
Hello everyone! I will be giving a DragonFly talk at the next
Bay Lisa. The primary focus of my talk will be a physical
characterization (latencies, overheads, etc) of MP mechanisms
and algorithms implemented by DragonFly. I'll be explaining how
the algorithms work and
to resolve a file handle containing an inode which cannot be found in
the original directory it was opened in.
-Matt
Matthew Dillon
[EMAIL PROTECTED
At the cost of drawing ire from FreeBSD core developers, I will
point out that reverse-resolution is hardly a black-and-white issue.
There are many shades of grey, and there is a huge problem set that
can either be solved or 99.99% of the way solved (greatly reducing the
time
I think this applies to FreeBSD, too and I could use some help in figuring
out how best to solve the problem.
I have a user reporting an interesting filesystem/VM deadlock in DragonFly
when running rtorrent. rtorrent has the particular effect of issuing
socket read()'s
:
:Matt,
:
:Thank you very much for response. This is a general solution, but it
:not sufficient for our needs. I guess I should have been more clear
:while explaining what we need.
:
:We want list of these locks for a group of processes.
:
:We made an implementation based on your suggestion, but
Hi Kirk, hackers!
I'm trying to track down a bug that is causing a buffer to be left
in a locked state and then causes the filesystem to lock up because
of that.
The symptoms are that a heavily used filesystem suddenly starts running
out of space. It isn't due to deleted
Addendum: My user reports that the problem also occurs on
FreeBSD 4.10 and 4.11, on uniprocessor builds (other builds
and 5.x/6.x have not been tested).
I took a look at the 4.x and 6.x softupdates code and didn't see any
commits that might address the problem.
This
TOK=0xc043e500 TD=0xcc2b7200
3 0 78286413140279 pmap_remove_pages+455 zfree+83
tokens_release /usr/src/sys/kern/lwkt_token.c:466 REF=0xd71f2a54
TOK=0xc043e500 TD=0xcc2b7200
-Matt
Matthew
continue iterating until F_GETLK tells you that
there are no more locks.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers
You think that is bad, try running 'rain' on an xterm!
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
I've been tracking down a crash one of our users gets occassionally.
He has a quad Intel(R) XEON(TM) CPU 2.00GHz (1996.61-MHz 686-class CPU)
system.
After getting a few of these crashes he pulled three of the four cpus
out. But with just one physical cpu, with HTT turned on
(but only one
or two on the HT cpu)... even though they share the same pipeline.
-Matt
Matthew Dillon
[EMAIL PROTECTED
and it
isn't an instruction/memory ordering issue.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http
... ]
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe
Here is the core of the FPU setup and restoration code for the kernel
bcopy in DragonFly, from i386/bcopy.s.
DragonFly uses the TD_SAVEFPU-is-a-pointer method that was outlined in
the original comment in the FreeBSD code. I further enhance the
algorithm to guarentee that the
All I really did was implement a comment that DG had made many years
ago in the PCB structure about making the FPU save area a pointer rather
then hardwiring it into the PCB.
This greatly reduces the complexity of work required to allow
the kernel to 'borrow' the FPU. It
it shouldn't be a problem.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org
Well, I don't want do disuade you from trying, but I think you are
seriously underestimating the effort required to restore device state.
You basically would either have to make all device drivers support a new
hibernation/restore API (because it is not really possible to restore
there is for a process.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman
Matthew Dillon
[EMAIL PROTECTED]
___
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL
If you guys are interested in the checkpointing code, now is the time
to port it. And maybe someone could donate the last little bit required
to make it reasonably secure when used by a non-root user. That bit being
to have the kernel record the file handles and creds in a
:Unfortunately, it is the binary driver from Nvidia. Maybe someone using
:DragonFly is having similar problems?
Not that I know of. There's not much that can be done with binary-only
drivers short of throwing them away and finding hardware that works
with normal drivers.
:I ran
reused before it is reusable.
If you have either of these two symptoms in FreeBSD-5/6 then you need
to take a very close look at your contigmalloc() code, even if you think
it is correct.
-Matt
Matthew Dillon
Minor correction. The 'start = 1' part of the patch is not required,
I missed the fact that the second loop was starting at start + 1.
-Matt
___
[EMAIL PROTECTED] mailing list
preclude an error path to this
panic from the pmap code. However, pmap panics could be related to
corrupted VM pages.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
:The first PR
Here is the final commit I made to DFly. I cleaned up the confirmation
message somewhat to make it more useful and correct the grammer.
I'm not saying that this should or should not be done in FreeBSD, but
I cannot think of any negatives and the -I option does allow for a far
then '/' that you would want to ask confirmation for that just
as obviously cannot be made default operation for rm.
-Matt
Matthew Dillon
[EMAIL PROTECTED
I think I'll commit something like this to DragonFly (you might
get patch errors w/ FreeBSD but this is the basic idea).
-Matt
Index: rm.1
===
RCS file: /cvs/src/bin/rm/rm.1,v
:pass more arguments and added a KASSERT in trap.c that is only enabled for
:kernels compiled with INVARIANTS turned on?
:...
:
:A KASSERT() wrapped in #ifdef INVARIANTS has zero overhead for normal,
:non-debugging kernels. The developers who are responsible for writing and
:testing new system
Well, wait a second... are we talking about a lot of packets being
discarded by the filter in 'normal' operation, or are we talking about
an attack? Because if we are takling about an attack the LAST ethernet
device anyone would ever want to use would be ED. i.e. they would be
:Hi,
:giving a look to interrupt 0x80 handling code (i386/i386/exception.s), I've
:met FAKE_MCOUNT/MEXITCOUNT system. FAKE_MCOUNT in the end calls _mcount
:(libkern/mcount.c) function which seems to be used for profiling purposes
:IMHO. It seems (reading comments in sys/gmon.h) that sysctl could
Matthew Dillon
[EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to [EMAIL PROTECTED]
1 - 100 of 1295 matches
Mail list logo