* Re: Freezing.. (was Re: Intel P6 vs P7 system call performance)
From: Russell King @ 2002-12-19 9:13 UTC (permalink / raw)
To: Adam J. Richter; +Cc: linux-kernel
In-Reply-To: <200212190108.RAA03649@adam.yggdrasil.com>
On Wed, Dec 18, 2002 at 05:08:45PM -0800, Adam J. Richter wrote:
> I don't currently use bugzilla (just due to inertia), but the
> whole world doesn't have to switch to something overnight in order for
> that facility to end up saving more time and resources than it has
> cost. Adoption can grow gradually, and it's probably easier to work
> out bugs (in bugzilla) and improvements that way anyhow.
I'm not asking the world to switch to it overnight. Just one person
would be nice. 8)
--
Russell King (rmk@arm.linux.org.uk) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
^ permalink raw reply
* [BENCHMARK] scheduler tunables with contest - max_sleep_average
From: Con Kolivas @ 2002-12-19 9:14 UTC (permalink / raw)
To: linux kernel mailing list; +Cc: Robert Love, Ingo Molnar
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Here are a set of contest results for 2.5.52-mm1 with max_sleep_average
scheduler settings (default is 2000) using the osdl hardware
(http://www.osdl.org)
noload:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [8] 39.7 180 0 0 1.10
max_sle1000 [3] 40.0 179 0 0 1.10
max_sle4000 [4] 39.7 180 0 0 1.10
max_sle500 [3] 39.6 180 0 0 1.09
cacherun:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 36.9 194 0 0 1.02
max_sle1000 [3] 36.8 194 0 0 1.02
max_sle4000 [4] 36.6 194 0 0 1.01
max_sle500 [3] 36.7 194 0 0 1.01
process_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 49.0 144 10 50 1.35
max_sle1000 [3] 52.8 133 14 61 1.46
max_sle4000 [4] 46.5 154 8 39 1.28
max_sle500 [3] 48.1 150 9 44 1.33
ctar_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 55.5 156 1 10 1.53
max_sle1000 [3] 53.2 161 1 10 1.47
max_sle4000 [4] 51.7 159 1 9 1.43
max_sle500 [3] 54.7 161 1 10 1.51
xtar_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 77.4 122 1 8 2.14
max_sle1000 [3] 82.6 117 1 9 2.28
max_sle4000 [3] 66.0 127 1 8 1.82
max_sle500 [3] 67.0 131 1 8 1.85
io_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 80.5 108 10 19 2.22
max_sle1000 [3] 89.9 94 14 21 2.48
max_sle4000 [3] 65.2 121 8 18 1.80
max_sle500 [3] 77.4 110 11 21 2.14
io_other:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 60.1 131 7 18 1.66
max_sle1000 [3] 66.3 127 9 20 1.83
max_sle4000 [3] 80.0 111 10 20 2.21
max_sle500 [3] 106.6 101 19 25 2.94
read_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 49.9 149 5 6 1.38
max_sle1000 [3] 50.2 148 5 6 1.39
max_sle4000 [3] 50.4 149 5 6 1.39
max_sle500 [3] 50.1 148 5 6 1.38
list_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 43.8 167 0 9 1.21
max_sle1000 [3] 43.7 167 0 9 1.21
max_sle4000 [3] 43.6 167 0 9 1.20
max_sle500 [3] 44.1 167 0 9 1.22
mem_load:
Kernel [runs] Time CPU% Loads LCPU% Ratio
2.5.52-mm1 [7] 71.1 123 36 2 1.96
max_sle1000 [3] 103.3 81 36 2 2.85
max_sle4000 [3] 104.4 79 36 2 2.88
max_sle500 [3] 101.3 76 35 2 2.80
Pretty weird. No clear relationship with changes here except perhaps a
proportional (small) change in ctar_load.
Con
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.0 (GNU/Linux)
iD8DBQE+AY3qF6dfvkL3i1gRAj1sAKCf80vYSSfEjl9SI3WFSOZE1HV6cgCggFgj
tivVtTQa/1wiytrsNyNAjow=
=rd8R
-----END PGP SIGNATURE-----
^ permalink raw reply
* MVME5100 boot from flash
From: Stefano Coluccini @ 2002-12-19 9:03 UTC (permalink / raw)
To: Linuxppc-Embedded
Hi all,
I have a MVME5100 and i can use Linux on it booting from the network, now
I'm trrying to boot from flash but without success. I have thought to simply
put the image that is good for the network boot in the flash bank with
pflash, but it doesn't work. Anyone knows what i have to do or where i can
found documentation about this topic ?
Thanks,
Stefano.
+--------------------------------------------------------+
! Stefano Coluccini ! CAEN SpA - Computing Div. !
! Via Vetraia, 11 ! 55049 Viareggio (LU)-ITALY !
! Tel. +39 0584 388 398 ! Fax +39 0584 388 959 !
! s.coluccini@caen.it ! www.caen.it/computing !
+--------------------------------------------------------+
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
^ permalink raw reply
* [LARTC] linux bridging and htb.init / cbq.init
From: Maik Ellinger @ 2002-12-19 9:02 UTC (permalink / raw)
To: lartc
Hello,
I'am successful using a linux (2.4.20) bridge and the cbq.init_v0.7
script for traffic shaping. But I can only use the queuing discipline
tbf (LEAF=tbf). If I try use queuing discipline sfq or none then tc
applies the rules but no shaping occurs.
Is this problem a normal behavior when using bridging ?
I tried to use htb.init (using patched tc binary). It accepts
the rules but also no shaping occurs (tested with all queuing
disciplines (LEAF=none|sfq|pfifo).
Why it doesn't work together with bridging ?
Thanks, Maik
--
VFS: Busy inodes after unmount. Self-destruct in 5 seconds. Have a nice
day...
GPG-Key: http://www.m-ellinger.de/output/mael.gpg
Mitglied der Zwickau Linux User Group zLUG e.V. -> http://www.zlug.org
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply
* EDE-0.0.5
From: Neil Holmes @ 2002-12-19 9:01 UTC (permalink / raw)
To: Linux 8086
Something else I forgot.
I have put a floppy disk label template for EDE in the 0.0.5 package too.
Its a word doc and fits Avery L7666 flopy disk label stationery.
Have fun.
Neil
^ permalink raw reply
* Re: 2.5.52-mm2
From: William Lee Irwin III @ 2002-12-19 8:54 UTC (permalink / raw)
To: Andrew Morton; +Cc: lkml, linux-mm
In-Reply-To: <3E015ECE.9E3BD19@digeo.com>
On Wed, Dec 18, 2002 at 09:53:18PM -0800, Andrew Morton wrote:
> url: http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.52/2.5.52-mm2/
Kernel compile on ramfs, shpte off, overcommit on (probably more like a
stress test for shpte):
c0116810 187174 0.528821 pfn_to_nid
c01168b8 192912 0.545032 x86_profile_hook
c0163c0c 267362 0.755375 d_lookup
c014e530 286920 0.810632 get_empty_filp
c01b0b18 287959 0.813567 __copy_from_user
c01320cc 307597 0.86905 find_get_page
c011a6b0 331219 0.935789 scheduler_tick
c0140890 342427 0.967455 vm_enough_memory
c013fc60 353705 0.999319 handle_mm_fault
c014eb49 358710 1.01346 .text.lock.file_table
c011a1f8 379521 1.07226 load_balance
c01b0ab0 840162 2.3737 __copy_to_user
c013f5a0 1040429 2.93951 do_anonymous_page
c01358c8 1056576 2.98513 __get_page_state
c014406c 1260931 3.56249 page_remove_rmap
c0143e68 1265355 3.57499 page_add_rmap
c0106f38 21236731 59.9999 poll_idle
shpte on will follow.
Bill
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/
^ permalink raw reply
* Minicom fails
From: Peter @ 2002-12-19 8:50 UTC (permalink / raw)
To: linux
Hi,
After I run a RH7.3 upgrade I can't access anymore minicom as a user.
I get the following message:
]$ minicom
Device /dev/ttyS0 lock failed: Operation not permitted.
No problem accessing it as su and /dev/ttyS0 is set to chmod a+rw.
How will I get my minicom back as a user?
Thanks & regards
--
Peter
-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs
^ permalink raw reply
* Minicom fails
From: Peter @ 2002-12-19 8:48 UTC (permalink / raw)
To: linux
Hi,
After I run a RH7.3 upgrade I can't access anymore minicom as a user.
I get the following message:
]$ minicom
Device /dev/ttyS0 lock failed: Operation not permitted.
No problem accessing it as su and /dev/ttyS0 is set to chmod a+rw.
How will I get my minicom back as a user?
Thanks & regards
--
Peter
-
To unsubscribe from this list: send the line "unsubscribe linux-newbie" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.linux-learn.org/faqs
^ permalink raw reply
* Re: 2.5.52-mm2
From: William Lee Irwin III @ 2002-12-19 8:54 UTC (permalink / raw)
To: Andrew Morton; +Cc: lkml, linux-mm
In-Reply-To: <3E015ECE.9E3BD19@digeo.com>
On Wed, Dec 18, 2002 at 09:53:18PM -0800, Andrew Morton wrote:
> url: http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.52/2.5.52-mm2/
Kernel compile on ramfs, shpte off, overcommit on (probably more like a
stress test for shpte):
c0116810 187174 0.528821 pfn_to_nid
c01168b8 192912 0.545032 x86_profile_hook
c0163c0c 267362 0.755375 d_lookup
c014e530 286920 0.810632 get_empty_filp
c01b0b18 287959 0.813567 __copy_from_user
c01320cc 307597 0.86905 find_get_page
c011a6b0 331219 0.935789 scheduler_tick
c0140890 342427 0.967455 vm_enough_memory
c013fc60 353705 0.999319 handle_mm_fault
c014eb49 358710 1.01346 .text.lock.file_table
c011a1f8 379521 1.07226 load_balance
c01b0ab0 840162 2.3737 __copy_to_user
c013f5a0 1040429 2.93951 do_anonymous_page
c01358c8 1056576 2.98513 __get_page_state
c014406c 1260931 3.56249 page_remove_rmap
c0143e68 1265355 3.57499 page_add_rmap
c0106f38 21236731 59.9999 poll_idle
shpte on will follow.
Bill
^ permalink raw reply
* Re: [LARTC] HTB steals bandwidth
From: Stef Coene @ 2002-12-19 8:45 UTC (permalink / raw)
To: lartc
In-Reply-To: <marc-lartc-104005823313721@msgid-missing>
> If I reduce rate and ceil, it doesn't work properly either. I tried setting
> the rate to 8kbit and the ceil to 16kbit and the upload did get less than
> 2kbyte/s, it was app. 1,3kbyte/s.
16kbit is 2 kbyte and 1,3 is allmost 2kbyte. So it seems to me that you are
shaping what you want. But what worried me was the fact that in you first
post, the bandwidth dropped after 5 sec from 128kbit to 80kbit. Do you
still have the same problem with the lower rate/ceil?
> Any ideas, why the rate crashes down with these rules? May the
> processor-power be the problem? The router is a Pentium-200, 64mb ram.
What's the load and memory usage on that box ?
Stef
--
stef.coene@docum.org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.oftc.net
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply
* Re: [LARTC] WonderShaper on LAN link kills to-host speed
From: Stef Coene @ 2002-12-19 8:39 UTC (permalink / raw)
To: lartc
In-Reply-To: <marc-lartc-104016341620263@msgid-missing>
On Wednesday 18 December 2002 23:22, Kenneth Porter wrote:
> --On Wednesday, December 18, 2002 10:43 PM +0100 Stef Coene
>
> <stef.coene@docum.org> wrote:
> > I'm not sure, but the policer can calculate the rate in the class in 2
> > ways. And maybe your CPU can't handle the calculations. What CPU do
> > you have and what's the load on the sstem?
>
> It's a P2-233 with 128 MB memory (Dell PowerEdge 4200). It's a bit
> memory-starved but otherwise seems to handle the load. It plays router,
> mail server, DNS, and file server. (Long-term plan is to offload
> non-gateway functions, once another box is freed up.)
>
> What are the "2 ways"? A pointer to source code would be fine, I just need
> to know where to start looking.
It's in the lartc howto. You can use a tbf alike function or a rate
estimator.
Stef
--
stef.coene@docum.org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.oftc.net
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply
* Re: QOS
From: Cedric Blancher @ 2002-12-19 8:38 UTC (permalink / raw)
To: Paulo Andre; +Cc: netfilter
In-Reply-To: <1040280504.11113.140.camel@bigblue>
Le jeu 19/12/2002 à 07:48, Paulo Andre a écrit :
> Which is the better package to use for QOS??
> HTB or CBQ..???
Aren't we discussing Netfilter in this list ?
You should consider STFW and see http://lartc.org/howto/...
"As your HTB configuration gets more complex, your configuration scales
well. With CBQ it is already complex even in simple cases!"
--
Cédric Blancher <blancher@cartel-securite.fr>
IT systems and networks security expert - Cartel Sécurité
Phone : +33 (0)1 44 06 97 87 - Fax: +33 (0)1 44 06 97 99
PGP KeyID:157E98EE FingerPrint:FA62226DA9E72FA8AECAA240008B480E157E98EE
^ permalink raw reply
* What do you think, is compatible locking solvable at all?
From: Michal Samek @ 2002-12-19 8:30 UTC (permalink / raw)
To: Linux-MSDOS Mailing list
Hi,
after 2 years of searching for the stable dosemu/samba shared clipper
application environment (maybe you remember my posts and a lot of other
locking-related posts) I'm thinking about my last try. I know I have to
find someone with the dos/linux low-level development skills since it's
not clear where the problem is - maybe mfs, maybe smbfs, probably both
are causing locking incompatibilities. Before I start searching for
someone skilled enough who can try to solve it I would like to ask you,
dosemu developers, what do you think, is my problem (I mean totally
compatible locking of files/records between samba/win sessions and
samba/smbfs/dosemu sessions) solvable at all? Maybe there are some
desing issues which make it really impossible to hack? Thanks for your
opinions.
--
Michal Samek <webmaster@tony.cz>
^ permalink raw reply
* [BENCHMARK] TIObench performance of mm2 patch. Latencies come down.
From: Aniruddha M Marathe @ 2002-12-19 8:24 UTC (permalink / raw)
To: linux-kernel
Hi,
Here are the results of comparison of kernel 2.5.52 with mm2 and 2.5.52 with mm1.
on TIObench. key findings are listed in the table. Values in the table indicate approximate percentage change with respect to previous result.
* Maximum latency has come down for this patch. This was on rise from last 5.52 release.
-------------------------------------------------------------
test 2.5.52-mm2 (as compared to
2.5.52 mm1) APPRXIMATE % change
-------------------------------------------------------------
rate (megabytes per second) 5% increase
CPU % utilization lessthan 2% increase
Average Latency 10% decrease
Maximum latency 10 % decrease
CPU efficiency less than 5% increase
-------------------------------------------------------------
Here are the complete results.
No size specified, using 252 MB
Unit information
================
File size = megabytes
Blk Size = bytes
Rate = megabytes per second
CPU% = percentage of CPU used during the test
Latency = milliseconds
Lat% = percent of requests that took longer than X seconds
CPU Eff = Rate divided by CPU% - throughput per cpu load
Sequential Reads
File Blk Num Avg Maximum Lat% Lat% CPU
Identifier Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
---------------------------- ------ ----- --- ------ ------ --------- ----------- -------- -------- -----
2.5.52 252 4096 10 9.08 5.545% 11.806 1837.19 0.00000 0.00000 164
Random Reads
File Blk Num Avg Maximum Lat% Lat% CPU
Identifier Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
---------------------------- ------ ----- --- ------ ------ --------- ----------- -------- -------- -----
2.5.52 252 4096 10 0.48 0.770% 215.442 1346.07 0.00000 0.00000 62
Sequential Writes
File Blk Num Avg Maximum Lat% Lat% CPU
Identifier Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
---------------------------- ------ ----- --- ------ ------ --------- ----------- -------- -------- -----
2.5.52 252 4096 10 17.59 32.71% 3.810 23958.10 0.05625 0.00156 54
Random Writes
File Blk Num Avg Maximum Lat% Lat% CPU
Identifier Size Size Thr Rate (CPU%) Latency Latency >2s >10s Eff
---------------------------- ------ ----- --- ------ ------ --------- ----------- -------- -------- -----
2.5.52 252 4096 10 0.77 1.067% 0.710 1113.98 0.00000 0.00000 72
Regards,
Aniruddha Marathe
WIPRO Technologies, India
aniruddha.marathe@wipro.com
+91-80-5502001 extn 5092
^ permalink raw reply
* Re: Invalid PBLK length
From: Jens Haug @ 2002-12-19 8:13 UTC (permalink / raw)
To: ducrot-kk6yZipjEM5g9hUCZPvPmw; +Cc: acpi-devel-pyega4qmqnRoyOMFzWx49A
> Ouch! Sorry. FACP is the signature of FADT in fact.
> FACS is the sig. of the FACS table, which contains the global lock and
> other things.
Ok, I see.
> ./acpidmp FACP | ./acpitbl
>
> Should give something like:
(...)
> OEMID: TOSCPL
> OEM Table ID: 888M1
This is ASUS and L2000D for me.
> S4BIOS_REQ: 0xf2
This is 0x00 for me. I had checked that before (when you
wrote about s4bios the last time).
> PM1a_EVT_BLK: 0x00001000
0x0000e400.
Now I only have to find out what that means. Guess I'll have
to read some specs.
> PM1b_EVT_BLK: 0x00000000
Same here.
> PM1a_CNT_BLK: 0x00001004
0x0000e404.
> PM1b_CNT_BLK: 0x00000000
Same here.
> PM2_CNT_BLK: 0x00001020
0x00000000.
> PM_TMR_BLK: 0x00001008
0x0000e408.
> GPE0_BLK: 0x00001028
0x0000e420
> GPE1_BLK: 0x0000102c
0x0000e430.
> PM1_EVT_LEN: 4
> PM1_CNT_LEN: 2
Same here.
> PM2_CNT_LEN: 1
0.
> PM_TM_LEN: 4
> GPE0_BLK_LEN: 4
> GPE1_BLK_LEN: 4
> GPE1_BASE: 16
Same here.
> P_LVL2_LAT: 10
Is this the latency for C2? This is 90 for me.
> P_LVL3_LAT: 1001
1900.
> FLUSH_SIZE: 0
> FLUSH_STRIDE: 0
> DUTY_OFFSET: 1
> DUTY_WIDTH: 3
> DAY_ALRM: 0x0d
Same.
> MON_ALRM: 0x08
> CENTURY: 0x32
Both 0x00.
> Flags: 0x00000025
0x000000a6.
I'll look this up in the specs and see what I can learn. Well,
at least I'm gonna try to find this in the specs.
Jens
--
Jens Haug
IKFF Universität Stuttgart Tel. 0711/685-6422
Pfaffenwaldring 9 Fax 0711/685-6356
70550 Stuttgart haug-X6ztD3ggwzuBAmxm6OvjtTjhTm2NLCe8@public.gmane.org
-------------------------------------------------------
This SF.NET email is sponsored by: Order your Holiday Geek Presents Now!
Green Lasers, Hip Geek T-Shirts, Remote Control Tanks, Caffeinated Soap,
MP3 Players, XBox Games, Flying Saucers, WebCams, Smart Putty.
T H I N K G E E K . C O M http://www.thinkgeek.com/sf/
^ permalink raw reply
* Re: Sequencer Input
From: Clemens Ladisch @ 2002-12-19 8:13 UTC (permalink / raw)
To: Brian Victor; +Cc: alsa-devel
In-Reply-To: <20021218000609.GB12303@bhv1.adelphia.net>
Brian Victor wrote:
> I'm using a MidiSport 1x1 and devfs. The modules are loading, pmidi is
> displaying the MidiSport and playing MIDI files through it just fine, so
> I presume output through /dev/snd/seq is working. cat /dev/snd/midiC0D0
> displays approrpriate "garbage" characters, but cat /dev/snd/seq does
> nothing,
This is OK, /dev/snd/seq cannot be accessed as a byte stream, unlike some
other devices. You have to use a program using the ALSA sequencer API to
test sequencer input.
> and the seqdemo.c sample just sits there.
seqdemo.c prints only those events sent to its port. Use aconnect to
connect the 1x1 port to the seqdemo port.
HTH
Clemens
-------------------------------------------------------
This SF.NET email is sponsored by: Order your Holiday Geek Presents Now!
Green Lasers, Hip Geek T-Shirts, Remote Control Tanks, Caffeinated Soap,
MP3 Players, XBox Games, Flying Saucers, WebCams, Smart Putty.
T H I N K G E E K . C O M http://www.thinkgeek.com/sf/
^ permalink raw reply
* boot meet deadlock at pppd link, what may cause wrong?
From: eric @ 2002-12-12 8:14 UTC (permalink / raw)
To: linux-kernel
Dear Linux kernel programmers:
I use Debian distribution, after some recent update, upgrade, reboot, at
boot time into 2.4.20, it deadlock at pppd link
What may cause wrong? and in addition fsck /dev/hdc2
what else I should repair it(since fsck is no help at nextboot time dead
lock)
Highly appreciate your help
Sincere Eric
www.linuxspice.com
linux/window pc for sale
^ permalink raw reply
* Re: [LARTC] HTB steals bandwidth
From: Robert Brueckmann @ 2002-12-19 8:08 UTC (permalink / raw)
To: lartc
In-Reply-To: <marc-lartc-104005823313721@msgid-missing>
> On Monday 16 December 2002 18:02, Robert Brueckmann wrote:
> > Hi!
> >
> > I just tested my HTB setup. But I have a problem. Let's assume I run the
> > following script (even if it might do nothing useful, just for
> > demonstrating my problem):
> >
> > #!/bin/bash
> > /usr/sbin/tc qdisc add dev ppp0 root handle 1: htb default 12
> > /usr/sbin/tc class add dev ppp0 parent 1: classid 1:1 htb rate 125kbit
ceil
> > 125kbit
> > /usr/sbin/tc class add dev ppp0 parent 1:1 classid 1:14 htb rate 125kbit
> > ceil 125kbit prio 0
> > iptables -A POSTROUTING -t mangle -o ppp0 -p tcp --dport ftp-data -j
> > MARK --set-mark 14
> > tc filter add dev ppp0 parent 1:0 prio 0 protocol ip handle 14 fw flowid
> > 1:14
> >
> > I have an adsl-connection (768kbit down/128kbit up), Linux kernel
2.4.20.
> > The script should do nothing to an outgoing ftp-upload, since I grant
all
> > the available bandwith to it. No other traffic is happending during all
> > that, only one ftp-upload from a computer inside the LAN. I start the
> > upload without the rules above, and the upload is at a constant maximum
of
> > 128kbit/sec. After running the script above and waiting for say 5
seconds,
> > the upload speed drops down to app. 80 kbit/s! After removing the rules
> > above, the speed climbs up again to top speed.
> Have you tried with other rates and ceil values?
> And you defined a default class 12, but there is no such class.
> Ftp-data can use dynamic ports. So can you check that the iptables line
with
> "--dport ftp-data" is really catching the ftp packets??
>
If I reduce rate and ceil, it doesn't work properly either. I tried setting
the rate to 8kbit and the ceil to 16kbit and the upload did get less than
2kbyte/s, it was app. 1,3kbyte/s.
I modified the script to satisfy the need for the default class 12:
#!/bin/bash
/usr/sbin/tc qdisc add dev ppp0 root handle 1:0 htb default 12
/usr/sbin/tc class add dev ppp0 parent 1:0 classid 1:1 htb rate 125kbit ceil
125kbit
/usr/sbin/tc class add dev ppp0 parent 1:1 classid 1:12 htb rate 25kbit ceil
125kbit prio 1
/usr/sbin/tc class add dev ppp0 parent 1:1 classid 1:14 htb rate 100kbit
ceil 125kbit prio 0
iptables -A POSTROUTING -t mangle -o ppp0 -p tcp --dport ftp-data -j
MARK --set-mark 14
tc filter add dev ppp0 parent 1:0 prio 0 protocol ip handle 14 fw flowid
1:14
The ftp-data-port rule works for all active ftp-connections, and the packet
dounter of the rule increases just as I expected.
Any ideas, why the rate crashes down with these rules? May the
processor-power be the problem? The router is a Pentium-200, 64mb ram.
Thanks,
Robert
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
^ permalink raw reply
* kazaa
From: Paulo Andre @ 2002-12-19 8:08 UTC (permalink / raw)
To: netfilter
I know that this is an old topic but I have done some reading and am
still stuck.
I have opened tcp and udp for port 1214 outgoing, but Kazaa is still not
connecting am I missing something..???
Paulo
^ permalink raw reply
* Re: QOS
From: Dharmendra.T @ 2002-12-19 8:05 UTC (permalink / raw)
To: Paulo Andre; +Cc: netfilter
In-Reply-To: <1040280504.11113.140.camel@bigblue>
HTB has very good options than CBQ. But bit confusing...
Regards
Dharmendra.T
Linux Security Expert
On Thu, 2002-12-19 at 12:18, Paulo Andre wrote:
> Which is the better package to use for QOS??
> HTB or CBQ..???
>
> Paulo
>
>
>
>
^ permalink raw reply
* RE: Docs for the new style dpalloc/hostalloc patch.
From: Joakim Tjernlund @ 2002-12-19 8:04 UTC (permalink / raw)
To: Pantelis Antoniou, linuxppc-embedded
In-Reply-To: <3E008DE0.2040902@intracom.gr>
Hi
This looks good to me. It's only a bit big, but I guess that's the price one
has to pay. I haven't tested it yet though.
Jocke
>
>
>
> Hello.
>
> Since the first mail I've sent hit the mailing list limit
> and got deleted, the following with only the patches
> are rather confusing.
>
> Here are some pointers about what the patch does and the
> rational behind it's creation...
>
> The patch is against the current top linuxppc_2_4_devel tree,
> and provides better functions for allocation of dual ported
> memory and uncached host memory for the 8xx family of processors.
>
> The problem with the current allocation routines are
> that are very simplistic. They only keep a top
> pointer to the next memory range to allocate
> and never free the allocated memory.
>
> This is fine for any drivers compiled into the kernel,
> but brakes down when you want to have your driver
> as a module.
>
> The requirement to have a driver as a module is
> peculiar when talking about so small systems, but
> it is the only way if you want to avoid GPLing all
> your design.
>
> As I mentioned we will submit all changes done to
> the kernel and the standard drivers, but
> we cannot do that for our proprietary hardware.
>
> The new routines get around the problem by
> managing the memory properly.
>
> That is the memory allocated can be freed and
> then reused for any following requests.
>
> The implementation is done by using something
> called a remote heap.
> It is called remote because in constrast to
> the normal heap management schemes we cannot
> use the memory we allocate to implement lists
> holding the allocated and free region.
> A nice side effect is that there is no overhead
> on the managed address ranges.
>
> The enable switches are under 'MPC8xx CPM Options'.
> If you don't enable then everything works the same as
> before.
>
> The functions are:
>
> void *new_m8xx_cpm_dpalloc(unsigned int size, const char *owner);
>
> Allocate the given amount of memory from the dual port ram
> and assign a text string describing the owner.
> A future patch that will add a proc interface to the cpm.
> The size is rounded to the next 8 bytes.
>
> Returns the va of the allocated address on success.
> It is always aligned to an 8 byte boundary.
> NULL otherwise
>
> int new_m8xx_cpm_dpfree(void *start);
>
> Free the previously allocated memory.
> Returns the size of the freed block, or -EINVAL in
> case the start argument does not lie in any allocated
> block.
>
> void *new_m8xx_cpm_dpalloc_fixed(void *start, int size,
> const char *owner);
>
> The same thins as the previous dpalloc function, but
> you can request memory from a specific address.
> In a future patch, this is used to allocate the
> microcode dpram areas.
>
> void *new_m8xx_cpm_hostalloc(unsigned int size, const char *owner);
>
> Allocate the given amount of uncached host memory for use by
> the cpm. Typically this memory will used for FIFOs, BDs and
> such when the cost of explicitly calling the invalidate_dcache
> and flush_dcache function is provibitive.
> If no available free memory is already present, it will allocate
> enough pages to satisfy the request. These pages
> will be tracked and when a later free releases all the
> memory allocated from them, they will be freed.
> Return NULL on out of memory.
>
> int new_m8xx_cpm_hostfree(void *start);
>
> Frees the given memory by a previous call to the hostalloc function.
> Returns the size of the free memory block or -EINVAL on error.
>
>
> That's it, please contact my for any more info.
>
> Pantelis
>
>
>
>
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
^ permalink raw reply
* EDE 0.0.5
From: Neil Holmes @ 2002-12-19 8:00 UTC (permalink / raw)
To: Linux 8086
I forgot to mention, when announcing EDE 0.0.5 yesterday, I have also put a
front install screen with simple progress bar on the installation. There is
no scrolling text noting activity any more. In the absence of "curses", I
have sought to provide a more attractive "first sight" for the new ELKS
user. Please try it and let me know what you think. If you don't want to
play with your existing desktops then it runs quite happily in DOSEMU. I did
all the testing in DOSEMU before moving to my desktop.
I have still not been able to upload. I believe Harry is making arrangements
such that we can put EDE in with the rest of the project on Sourceforge
(where it should be now). Please email if you want a copy of the image file
to build your install floppy in advance of this.
Have fun
Neil
^ permalink raw reply
* [BENCHMARK] Lmbench result for 5.52 mm2 patch.
From: Aniruddha M Marathe @ 2002-12-19 8:08 UTC (permalink / raw)
To: linux-kernel
Hi,
Here are the results of Lmbench on 2.5.52-mm2. I have made a table of prominent changes. The table entries are median values of 5 results.
* mmap latency that was on rise for some 5.52 and mm1 patches, is now decreased.
* also fork proc is giving better performance now.
2.5.52-mm2 2.5.52-mm1
==============================================================================
Processor, Processes - times in microseconds - smaller is better
1.fork proc 441 400
------------------------------------------------------------------------------
Context switching - times in microseconds - smaller is better
1. 2p/0K ctxsw 1.390 1.290
2. 2p/16k ctxsw 5.08 4.62
------------------------------------------------------------------------------
*Local* Communication latencies in microseconds - smaller is better
1. 2p/0K 1.390 1.290
------------------------------------------------------------------------------
File & VM system latencies in microseconds - smaller is better
1. mmap latency 597 640
------------------------------------------------------------------------------
*Local* Communication bandwidths in MB/s - bigger is better
1. pipe 404 335
==============================================================================
Here are the complete results for mm2 patch.
L M B E N C H 2 . 0 S U M M A R Y
------------------------------------
(Alpha software, do not distribute)
Basic system parameters
----------------------------------------------------
Host OS Description Mhz
--------- ------------- ----------------------- ----
benchtest Linux 2.5.52 i686-pc-linux-gnu 790
benchtest Linux 2.5.52 i686-pc-linux-gnu 790
benchtest Linux 2.5.52 i686-pc-linux-gnu 790
benchtest Linux 2.5.52 i686-pc-linux-gnu 790
benchtest Linux 2.5.52 i686-pc-linux-gnu 790
Processor, Processes - times in microseconds - smaller is better
----------------------------------------------------------------
Host OS Mhz null null open selct sig sig fork exec sh
call I/O stat clos TCP inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ----- ---- ---- ---- ---- ----
benchtest Linux 2.5.52 790 0.44 0.82 28 29 1.28 5.20 458 1546 7942
benchtest Linux 2.5.52 790 0.44 0.80 28 29 34 1.28 5.21 445 1564 8118
benchtest Linux 2.5.52 790 0.44 0.80 28 29 32 1.28 5.42 351 1550 8021
benchtest Linux 2.5.52 790 0.46 0.82 28 30 32 1.30 5.43 441 1547 8096
benchtest Linux 2.5.52 790 0.46 0.83 28 30 34 1.30 5.42 381 1572 8048
Context switching - times in microseconds - smaller is better
-------------------------------------------------------------
Host OS 2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
--------- ------------- ----- ------ ------ ------ ------ ------- -------
benchtest Linux 2.5.52 1.390 5.0800 15 8.2900 178 43 180
benchtest Linux 2.5.52 1.360 5.2000 14 8.8000 178 42 180
benchtest Linux 2.5.52 1.430 4.7800 17 6.9300 181 46 181
benchtest Linux 2.5.52 1.370 5.1800 14 6.0600 179 45 180
benchtest Linux 2.5.52 1.450 5.2200 14 9.0300 179 42 180
*Local* Communication latencies in microseconds - smaller is better
-------------------------------------------------------------------
Host OS 2p/0K Pipe AF UDP RPC/ TCP RPC/ TCP
ctxsw UNIX UDP TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
benchtest Linux 2.5.52 1.390 8.173 21 34 59 125 156 171
benchtest Linux 2.5.52 1.360 7.991 21 35 60 127 156 171
benchtest Linux 2.5.52 1.430 8.242 21 35 59 126 156 175
benchtest Linux 2.5.52 1.370 8.415 21 35 59 125 158 174
benchtest Linux 2.5.52 1.450 8.335 21 35 60 125 156 173
File & VM system latencies in microseconds - smaller is better
--------------------------------------------------------------
Host OS 0K File 10K File Mmap Prot Page
Create Delete Create Delete Latency Fault Fault
--------- ------------- ------ ------ ------ ------ ------- ----- -----
benchtest Linux 2.5.52 118 58 373 118 586 0.912 4.00000
benchtest Linux 2.5.52 118 58 381 129 599 0.899 4.00000
benchtest Linux 2.5.52 119 58 381 127 587 0.891 4.00000
benchtest Linux 2.5.52 119 58 379 129 597 0.903 4.00000
benchtest Linux 2.5.52 119 58 377 126 601 0.980 4.00000
*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------
Host OS Pipe AF TCP File Mmap Bcopy Bcopy Mem Mem
UNIX reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
benchtest Linux 2.5.52 335 137 23 299 355 124 113 355 171
benchtest Linux 2.5.52 432 130 22 295 352 123 112 353 169
benchtest Linux 2.5.52 322 129 22 295 352 123 112 352 169
benchtest Linux 2.5.52 407 134 22 293 352 123 112 352 168
benchtest Linux 2.5.52 292 131 22 290 350 124 113 350 168
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
---------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Guesses
--------- ------------- ---- ----- ------ -------- -------
benchtest Linux 2.5.52 790 3.799 8.8840 174
benchtest Linux 2.5.52 790 3.807 8.8830 176
benchtest Linux 2.5.52 790 3.798 8.8830 176
benchtest Linux 2.5.52 790 3.808 8.8740 176
benchtest Linux 2.5.52 790 3.808 56 177
Regards,
Aniruddha Marathe
WIPRO Ttechnologies, India
Aniruddha.marathe@wipro.com
+91-80-5502001 extn 5092.
^ permalink raw reply
* [Linux-ia64] Can't boot in SMP with kernel 2.5.50 ia64
From: Xavier Bru @ 2002-12-19 7:53 UTC (permalink / raw)
To: linux-ia64
Hello,
Booting 2.5.50 with David's patch, it seems we can't boot in SMP on an
ia64 machine. We get the message: SMP mode deactivated.
Problem is due to smp_prepare_cpus() declaring max_cpus as "unsigned
int" and testing against the -1 value.
Problem was not seen in 2.5.45 because max_cpus was initialized to
UINT_MAX.
Boot in SMP is OK with the following patch:
diff --exclude-from /home/xb/proc/diff.exclude -Nur 2.5.50.ref/arch/ia64/kernel/smpboot.c 2.5.50/arch/ia64/kernel/smpboot.c
--- 2.5.50.ref/arch/ia64/kernel/smpboot.c Wed Dec 18 11:15:48 2002
+++ 2.5.50/arch/ia64/kernel/smpboot.c Wed Dec 18 15:11:52 2002
@@ -472,7 +472,7 @@
* Cycle through the APs sending Wakeup IPIs to boot each.
*/
void __init
-smp_prepare_cpus (unsigned int max_cpus)
+smp_prepare_cpus (int max_cpus)
{
int boot_cpu_id = hard_smp_processor_id();
diff --exclude-from /home/xb/proc/diff.exclude -Nur 2.5.50.ref/include/linux/smp.h 2.5.50/include/linux/smp.h
--- 2.5.50.ref/include/linux/smp.h Wed Dec 18 11:15:48 2002
+++ 2.5.50/include/linux/smp.h Wed Dec 18 15:28:47 2002
@@ -34,7 +34,7 @@
/*
* Prepare machine for booting other CPUs.
*/
-extern void smp_prepare_cpus(unsigned int max_cpus);
+extern void smp_prepare_cpus(int max_cpus);
/*
* Bring a CPU up
--
Sincères salutations.
_____________________________________________________________________
Xavier BRU BULL ISD/R&D/INTEL office: FREC B1-422
tel : +33 (0)4 76 29 77 45 http://www-frec.bull.fr
fax : +33 (0)4 76 29 77 70 mailto:Xavier.Bru@bull.net
addr: BULL, 1 rue de Provence, BP 208, 38432 Echirolles Cedex, FRANCE
_____________________________________________________________________
^ permalink raw reply
* [LARTC] telnet on demand
From: it eintz @ 2002-12-19 7:50 UTC (permalink / raw)
To: lartc
[-- Attachment #1: Type: text/plain, Size: 845 bytes --]
Hello,
Hopefully this subject is under the scope of this mailing list.
I have the following arrangement:
[Windows] --- [Linux] ---[Router]---[Modem]---[DOS]
The windows is running PCanywhere and so does the DOS
I can telnet into the router and make it dial the modem into the DOS
The windows needs to use telnet to access the DOS.
The question is, How do i make the linux accept a telnet session and when it connects automatically telnet itself into the Router and dial the modem to transfer the content of that telnet session into its session with the Windows.
making the router dial and connect is easy, but i have no idea how to inject lines to one telnet session and then transfer it into another telnet session.
Thank you.
---------------------------------
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now
[-- Attachment #2: Type: text/html, Size: 1048 bytes --]
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.