Re: RE : cable modem
not be the one you'll notice - maybe somebody who will try such a service can comment on what actually happens (as far as I know, this service isn't widely available in Israel yet). It was not legal to sell the service (except on an experimental basis) in Israel until yesterday. Not all cable companies do it, but Golden Chanens (Arutzi Zahav) in Jerusalem do. Calling at 8am, my son was able to talk to a sales person who knew what they were talking about at 8pm. An installer is coming on Sunday. and get this, they SUPPORT LINUX!!! More info next week. Geoff. -- Geoffrey S. Mendelson Bloomberg L.P., BFM (Israel) 2 hours ahead of London, 7 hours ahead of New York. Tel: 972-(0)3-754-1158 Fax 972-(0)3-754-1236 Email: [EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
About fribidi PHP
Hi All I noticed that php added support for fribidi using a function called fribidi_log2vis What will be the benefit if Ill use fribidi_log2vis and not hebrevc/hebrev ? http://www.php.net/manual/en/function.fribidi-log2vis.php also just for curiosity, did any one tried the Mohawk session server ? -- Canaan Surfing Ltd. Internet Service Providers Ben-Nes Michael - Manager Tel: 972-4-6991122 http://sites.canaan.co.il -- = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
On Sun, 17 Mar 2002, Malcolm Kavalsky wrote:
I forgot to mention that I am running Redhat 7.2 with kernel 2.4.14. My
PIII has
384Mb dram, and runs at 667Mhz. The results are completely reproducible,
without
running anything in the background.
ok. obviously, we're doing something wrong with these measurements. lets
try to check a few more things:
1. how did you ocmpile the test program? give the command line you used. i
did 'gcc -O2 prog.c'. gcc is 2.91.66 - the same gcc was used to compile
both kernel, glibc and application. which compiler(s) did you use to
compile each of these?
2. CPUs - it is possible to take advantage of different assembly copy
routines in order to increase the copying speed. your CPUs are PIII.
i'm not aware of assembly insturctions on them which are faster then on
and AMD K6-2, which the optimizer might have used. or are there, and
they were used when compiling the kernel (read() and write() are using
the code compiled in the kernel, while memcpy uses the code compiled in
the appliction). thus, in order to make a fair comparison, both
application and kernel must be compiled using the same flags.
the kernel code (2.4.18) looks the same as the 2.4.4 code, so its not a
kernel change. what i see is that the copying routines go down to
mmx_copy_user and mmx_copy_user_zeroing, for chunks which are larger then
512 bytes (which are the chunks probably copied in our case). couldn't
find these functions in the sources, though. but this happens if
'CONFIG_X86_USE_3DNOW_AND_WORKS' is defined. otherwise, it uses
__copy_user and __copy_user_zeroing. so the question now is - when you
compiled your kernel, what CPU did you specify in the kernel's config?
I realised afterwards that I should have measured the time also in the
server, and attach
the fixed program, which still gives the same results.
I guess there have been some major improvements on the kernel code, running
the same program on my laptop ( PIII 500 Mhz, 128MB Dram, kernel 2.4.18)
results in:
Memcpy'ed 2000 blocks of size 1048576 in 11 seconds = 181 Mbytes/second
Sent 2000 blocks of size 1048576 in 5 seconds over unixsocket = 400
Mbytes/second
Received 2097152000 bytes in 5 seconds over unix socket =400
Mbytes/second
Even though zero-copy is not being done, isn't it surprising how much
faster it
is to send data over a socket than just to copy it from one buffer to
another ;)
don't you find it a bit odd, that it'll be faster with doing
double-copies, then when doing a single copy? or you tihnk virtual memory
somehow plays tricks on us? if so - why'd it be different?
--
guy
For world domination - press 1,
or dial 0, and please hold, for the creator. -- nob o. dy
-- Attached file included as plaintext by Listar --
-- File: socket_vs_memcpy.c
#include stdio.h
#include malloc.h
#include string.h
#include time.h
#include sys/socket.h
#include sys/un.h
#include sys/types.h
#include sys/wait.h
#include unistd.h
#define BUFSIZE 0x10 /* 1 Megabyte */
#define NBLOCKS 2000
#define PORT_NAME/tmp/foo
void server()
{
struct sockaddr_un sin,from;
int s,g,len,n;
char *buf;
float nbytes;
time_t start_time, elapsed_time;
buf = malloc( BUFSIZE );
/* Create an unbound socket */
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( bind( s, (struct sockaddr *)sin,
strlen(sin.sun_path) + sizeof(sin.sun_family)) 0){
printf( Bad bind\n);
return;
}
listen( s, 5 );
len = sizeof(from);
g = accept( s, (struct sockaddr *)from, len );
nbytes = read( g, buf, BUFSIZE );
start_time = time(0);
while( (n = read( g, buf, BUFSIZE )) 0 ) {
nbytes += n;
}
elapsed_time = time(0) - start_time;
close(g);
close(s);
unlink( PORT_NAME );
printf( Received %10.0f bytes in %d seconds over unix socket =,
nbytes, (int)elapsed_time );
printf( %4.0f Mbytes/second \n, nbytes / (0x10 * elapsed_time) );
}
void client()
{
struct sockaddr_un sin;
int s;
char *buf;
time_t start_time, elapsed_time;
int i;
buf = malloc( BUFSIZE );
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( connect( s, (struct sockaddr *)sin, sizeof(sin)) 0 ){
printf(Bad connect\n);
close(s);
return;
}
start_time = time(0);
for( i=0; i NBLOCKS write(s, buf, BUFSIZE) == BUFSIZE ; i++ );
elapsed_time = time(0) - start_time;
close(s);
printf( Sent %d blocks of size %d in %d seconds over unix socket =,
i, BUFSIZE, (int)elapsed_time );
printf( %d Mbytes/second \n, (NBLOCKS * BUFSIZE) / (0x10 * (int)elapsed_time)
);
}
void memcpy_benchmark()
{
char *src, *dst;
time_t start_time, elapsed_time;
int i;
src = malloc ( BUFSIZE );
dst = malloc ( BUFSIZE );
start_time =
Re: Cable modems
Nir Siminovich [EMAIL PROTECTED] writes: 1. You can't change the network card the ISP will give you, as it's MAC address is defined within the ISP LDAP server, and is required for obtaining an IP address. the mac's changeable. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Zope/Python Meeting
Is being setup. Details unclear. Make yourself heard! Go to the wiki! http://twistedmatrix.com/users/moshez.twistd/wiki/cgi-bin/moin.cgi Fill in name and preferences. Thanks. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
guy keren wrote: On Sun, 17 Mar 2002, Malcolm Kavalsky wrote: I forgot to mention that I am running Redhat 7.2 with kernel 2.4.14. My PIII has 384Mb dram, and runs at 667Mhz. The results are completely reproducible, without running anything in the background. ok. obviously, we're doing something wrong with these measurements. lets try to check a few more things: 1. how did you ocmpile the test program? give the command line you used. i did 'gcc -O2 prog.c'. gcc is 2.91.66 - the same gcc was used to compile both kernel, glibc and application. which compiler(s) did you use to compile each of these? I have compiled using kgcc, and gcc, with optimization O2, and without, and see no difference. kgcc version : gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) gcc version:gcc version 2.96 2731 (Red Hat Linux 7.1 2.96-98) I didn't recompile glibc. 2. CPUs - it is possible to take advantage of different assembly copy routines in order to increase the copying speed. your CPUs are PIII. i'm not aware of assembly insturctions on them which are faster then on and AMD K6-2, which the optimizer might have used. or are there, and they were used when compiling the kernel (read() and write() are using the code compiled in the kernel, while memcpy uses the code compiled in the appliction). thus, in order to make a fair comparison, both application and kernel must be compiled using the same flags. Ok. So, I'll try using kernel flags on the application. the kernel code (2.4.18) looks the same as the 2.4.4 code, so its not a kernel change. what i see is that the copying routines go down to mmx_copy_user and mmx_copy_user_zeroing, for chunks which are larger then 512 bytes (which are the chunks probably copied in our case). couldn't find these functions in the sources, though. but this happens if 'CONFIG_X86_USE_3DNOW_AND_WORKS' is defined. otherwise, it uses __copy_user and __copy_user_zeroing. so the question now is - when you compiled your kernel, what CPU did you specify in the kernel's config? PIII Coppermine I realised afterwards that I should have measured the time also in the server, and attach the fixed program, which still gives the same results. I guess there have been some major improvements on the kernel code, running the same program on my laptop ( PIII 500 Mhz, 128MB Dram, kernel 2.4.18) results in: Memcpy'ed 2000 blocks of size 1048576 in 11 seconds = 181 Mbytes/second Sent 2000 blocks of size 1048576 in 5 seconds over unixsocket = 400 Mbytes/second Received 2097152000 bytes in 5 seconds over unix socket =400 Mbytes/second Even though zero-copy is not being done, isn't it surprising how much faster it is to send data over a socket than just to copy it from one buffer to another ;) don't you find it a bit odd, that it'll be faster with doing double-copies, then when doing a single copy? or you tihnk virtual memory somehow plays tricks on us? if so - why'd it be different? I asked one of the top Unix hackers that I know, and he said: I would guess that if you do large af_unix transfers that are page aligned then the system doesn't have to actually copy the data rather it can share the page and do a copy on write. This preserves the socket semantics and can be faster than memcpy. This was done many years ago in Solaris. I wonder if digging deep enough in the kernel sources, will reveal this ... _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
On Mon, Mar 18, 2002, Malcolm Kavalsky wrote about Re: pthreads question: I asked one of the top Unix hackers that I know, and he said: I would guess that if you do large af_unix transfers that are page aligned then the system doesn't have to actually copy the data rather it can share the page and do a copy on write. This preserves the socket semantics and can be faster than memcpy. This was done many years ago in Solaris. I wonder if digging deep enough in the kernel sources, will reveal this ... You can try to check if this is the case, by following each send or memcpy by a memset() of the buffer. If the memcpy method suddenly becomes quicker, this explanation might be true. Strange though - how come malloc() returns page-aligned buffers? Does the Linux code really checks for this rare and rather esoteric case (if you write to the buffer after sending it, and the kernel can't know you're writing whole pages, it will have to do a copy-on- write and do the copy anyway). -- Nadav Har'El|Monday, Mar 18 2002, 5 Nisan 5762 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |An error? Impossible! My modem is error http://nadav.harel.org.il |correcting. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: About fribidi PHP
The result will be Unicode compliant. On Mon, 18 Mar 2002, Ben-Nes Michael wrote: Hi All I noticed that php added support for fribidi using a function called fribidi_log2vis What will be the benefit if Ill use fribidi_log2vis and not hebrevc/hebrev ? http://www.php.net/manual/en/function.fribidi-log2vis.php also just for curiosity, did any one tried the Mohawk session server ? -- Canaan Surfing Ltd. Internet Service Providers Ben-Nes Michael - Manager Tel: 972-4-6991122 http://sites.canaan.co.il -- = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] -- Behdad Esfahbod 27 Esfand 1380, 2002 Mar 18 behdad at bamdad dot org [Finger for Geek Code] Ritchie's Rule: (1) Everything has some value -- if you use the right currency. (2) Paint splashes last longer than the paint job. (3) Search and ye shall find -- but make sure it was lost. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: RE : cable modem
first I want to apologize for the mistakes I made last night ( 10x Nadav :-) now some more info: I joined the cable test Arutzi Zahav running more then 6 month ago. I've connected my Linux box which work both as a server/workstation and a router for another box running win. as for the speed, you can UL and DL at the same speed theoreticly ( the same speed limit is for both directions ) that mean that I made some DL and UL with some servers that kept a steady speed transferring 1 Mb in 9-10 sec. as for the future when the number of user will increase I assume the speed will decrease ( it's much like adding more stations on a local net that works on 10Mb cards) as I write this mail I'm DL from iglu at ~100k/s via FTP client running on the win box. -- Yehuda Drori http://whatsup.co.il On Monday 18 March 2002 10:17, Geoffrey S. Mendelson wrote: not be the one you'll notice - maybe somebody who will try such a service can comment on what actually happens (as far as I know, this service isn't widely available in Israel yet). It was not legal to sell the service (except on an experimental basis) in Israel until yesterday. Not all cable companies do it, but Golden Chanens (Arutzi Zahav) in Jerusalem do. Calling at 8am, my son was able to talk to a sales person who knew what they were talking about at 8pm. An installer is coming on Sunday. and get this, they SUPPORT LINUX!!! More info next week. Geoff. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: About fribidi PHP
Ben-Nes Michael wrote: Hi All I noticed that php added support for fribidi using a function called fribidi_log2vis What will be the benefit if Ill use fribidi_log2vis and not hebrevc/hebrev ? Working on the Haayal Hakore system and some related sites, we encountered several bugs in hebrev() and had to concoct workarounds and patches. I reported one of the simpler ones (handling of square parentheses) to the PHP bug tracking system but it was never acknowledged. I haven't tested FriBiDi yet (I was too lazy to write the PHP wrapper), but it definitely seems like a more serious effort. Regards, Eran Tromer = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
Eureka!
Nadav Har'El wrote:
On Mon, Mar 18, 2002, Malcolm Kavalsky wrote about Re: pthreads question:
I asked one of the top Unix hackers that I know, and he said:
I would guess that if you do large af_unix transfers that are page
aligned then the system doesn't have to actually copy the data rather it
can share the page and do a copy on write. This preserves the socket
semantics and can be faster than memcpy. This was done many years ago in
Solaris.
I wonder if digging deep enough in the kernel sources, will reveal this ...
You can try to check if this is the case, by following each send or memcpy
by a memset() of the buffer. If the memcpy method suddenly becomes quicker,
this explanation might be true.
Strange though - how come malloc() returns page-aligned buffers? Does the
Linux code really checks for this rare and rather esoteric case (if you
write to the buffer after sending it, and the kernel can't know you're
writing whole pages, it will have to do a copy-on- write and do the copy
anyway).
This is exactly what happened! I added in memset after memcpy, and also
after sending
the buffer, the results are:
Memcpy'ed and memsetted 1000 blocks of size 1048576 in 18 seconds = 55
Mbytes/second
Started receiving at Mon Mar 18 13:41:13 2002
Received 1048576000 bytes in 17 seconds over unix socket = 59
Mbytes/second
Started sending at Mon Mar 18 13:41:13 2002
Sent and memsetted 1000 blocks of size 1048576 in 17 seconds over unix
socket = 58 Mbytes/second
(You notice that I also added printing exact time that send and receive
started, to ensure no
delay between the two)
I also attach the source file for reference.
#include stdio.h
#include malloc.h
#include string.h
#include time.h
#include sys/socket.h
#include sys/un.h
#include sys/types.h
#include sys/wait.h
#include unistd.h
#define BUFSIZE 0x10 /* 1 Megabyte */
#define NBLOCKS 1000
#define PORT_NAME/tmp/foo
void server()
{
struct sockaddr_un sin,from;
int s,g,len,n;
char *buf;
float nbytes;
time_t start_time, elapsed_time;
buf = malloc( BUFSIZE );
/* Create an unbound socket */
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( bind( s, (struct sockaddr *)sin,
strlen(sin.sun_path) + sizeof(sin.sun_family)) 0){
printf( Bad bind\n);
return;
}
listen( s, 5 );
len = sizeof(from);
g = accept( s, (struct sockaddr *)from, len );
nbytes = read( g, buf, BUFSIZE );
start_time = time(0);
while( (n = read( g, buf, BUFSIZE )) 0 ) {
nbytes += n;
}
elapsed_time = time(0) - start_time;
close(g);
close(s);
unlink( PORT_NAME );
printf(\nStarted receiving at %s, ctime( start_time ));
printf( Received %10.0f bytes in %d seconds over unix socket =,
nbytes, (int)elapsed_time );
printf( %4.0f Mbytes/second \n, nbytes / (0x10 * elapsed_time) );
}
void client()
{
struct sockaddr_un sin;
int s;
char *buf;
time_t start_time, elapsed_time;
int i;
buf = malloc( BUFSIZE );
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( connect( s, (struct sockaddr *)sin, sizeof(sin)) 0 ){
printf(Bad connect\n);
close(s);
return;
}
start_time = time(0);
for( i=0; i NBLOCKS write(s, buf, BUFSIZE) == BUFSIZE ; i++ ) {
memset( buf, 'A', BUFSIZE );
}
elapsed_time = time(0) - start_time;
close(s);
printf(\nStarted sending at %s, ctime( start_time ));
printf( Sent and memsetted %d blocks of size %d in %d seconds over unix socket =,
i, BUFSIZE, (int)elapsed_time );
printf( %d Mbytes/second \n, (NBLOCKS * BUFSIZE) / (0x10 * (int)elapsed_time)
);
}
void memcpy_benchmark()
{
char *src, *dst;
time_t start_time, elapsed_time;
int i;
src = malloc ( BUFSIZE );
dst = malloc ( BUFSIZE );
start_time = time(0);
for( i=0; i NBLOCKS; i++ ){
memcpy( dst, src, BUFSIZE );
memset( dst, 'A', BUFSIZE );
}
elapsed_time = time(0) - start_time;
printf( Memcpy'ed and memsetted %d blocks of size %d in %d seconds =,
NBLOCKS, BUFSIZE, (int)elapsed_time );
printf( %d Mbytes/second\n, (NBLOCKS * BUFSIZE) / (0x10 * (int)elapsed_time)
);
}
void socket_benchmark()
{
int status;
if ( fork() == 0 ) {
server();
} else {
sleep(1); /* Dirty, but ensures client runs after server is ready */
client();
}
wait(status);
}
int main()
{
memcpy_benchmark();
socket_benchmark();
return 0;
}
Re: pthreads question
Eureka!
Nadav Har'El wrote:
On Mon, Mar 18, 2002, Malcolm Kavalsky wrote about Re: pthreads
question:
I asked one of the top Unix hackers that I know, and he said:
I would guess that if you do large af_unix transfers that are page
aligned then the system doesn't have to actually copy the data rather it
can share the page and do a copy on write. This preserves the socket
semantics and can be faster than memcpy. This was done many years ago in
Solaris.
I wonder if digging deep enough in the kernel sources, will reveal
this ...
You can try to check if this is the case, by following each send or
memcpy
by a memset() of the buffer. If the memcpy method suddenly becomes
quicker,
this explanation might be true.
Strange though - how come malloc() returns page-aligned buffers? Does the
Linux code really checks for this rare and rather esoteric case (if you
write to the buffer after sending it, and the kernel can't know you're
writing whole pages, it will have to do a copy-on- write and do the copy
anyway).
This is exactly what happened! I added in memset after memcpy, and also
after sending
the buffer, the results are:
Memcpy'ed and memsetted 1000 blocks of size 1048576 in 18 seconds = 55
Mbytes/second
Started receiving at Mon Mar 18 13:41:13 2002
Received 1048576000 bytes in 17 seconds over unix socket = 59
Mbytes/second
Started sending at Mon Mar 18 13:41:13 2002
Sent and memsetted 1000 blocks of size 1048576 in 17 seconds over unix
socket = 58 Mbytes/second
(You notice that I also added printing exact time that send and receive
started, to ensure no
delay between the two)
I also attach the source file for reference.
Malcolm
#include stdio.h
#include malloc.h
#include string.h
#include time.h
#include sys/socket.h
#include sys/un.h
#include sys/types.h
#include sys/wait.h
#include unistd.h
#define BUFSIZE 0x10 /* 1 Megabyte */
#define NBLOCKS 1000
#define PORT_NAME/tmp/foo
void server()
{
struct sockaddr_un sin,from;
int s,g,len,n;
char *buf;
float nbytes;
time_t start_time, elapsed_time;
buf = malloc( BUFSIZE );
/* Create an unbound socket */
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( bind( s, (struct sockaddr *)sin,
strlen(sin.sun_path) + sizeof(sin.sun_family)) 0){
printf( Bad bind\n);
return;
}
listen( s, 5 );
len = sizeof(from);
g = accept( s, (struct sockaddr *)from, len );
nbytes = read( g, buf, BUFSIZE );
start_time = time(0);
while( (n = read( g, buf, BUFSIZE )) 0 ) {
nbytes += n;
}
elapsed_time = time(0) - start_time;
close(g);
close(s);
unlink( PORT_NAME );
printf(\nStarted receiving at %s, ctime( start_time ));
printf( Received %10.0f bytes in %d seconds over unix socket =,
nbytes, (int)elapsed_time );
printf( %4.0f Mbytes/second \n, nbytes / (0x10 * elapsed_time) );
}
void client()
{
struct sockaddr_un sin;
int s;
char *buf;
time_t start_time, elapsed_time;
int i;
buf = malloc( BUFSIZE );
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
printf( Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( connect( s, (struct sockaddr *)sin, sizeof(sin)) 0 ){
printf(Bad connect\n);
close(s);
return;
}
start_time = time(0);
for( i=0; i NBLOCKS write(s, buf, BUFSIZE) == BUFSIZE ; i++ ) {
memset( buf, 'A', BUFSIZE );
}
elapsed_time = time(0) - start_time;
close(s);
printf(\nStarted sending at %s, ctime( start_time ));
printf( Sent and memsetted %d blocks of size %d in %d seconds over unix socket =,
i, BUFSIZE, (int)elapsed_time );
printf( %d Mbytes/second \n, (NBLOCKS * BUFSIZE) / (0x10 * (int)elapsed_time)
);
}
void memcpy_benchmark()
{
char *src, *dst;
time_t start_time, elapsed_time;
int i;
src = malloc ( BUFSIZE );
dst = malloc ( BUFSIZE );
start_time = time(0);
for( i=0; i NBLOCKS; i++ ){
memcpy( dst, src, BUFSIZE );
memset( dst, 'A', BUFSIZE );
}
elapsed_time = time(0) - start_time;
printf( Memcpy'ed and memsetted %d blocks of size %d in %d seconds =,
NBLOCKS, BUFSIZE, (int)elapsed_time );
printf( %d Mbytes/second\n, (NBLOCKS * BUFSIZE) / (0x10 * (int)elapsed_time)
);
}
void socket_benchmark()
{
int status;
if ( fork() == 0 ) {
server();
} else {
sleep(1); /* Dirty, but ensures client runs after server is ready */
client();
}
wait(status);
}
int main()
{
memcpy_benchmark();
socket_benchmark();
return 0;
}
Telnet by any other time smells like shit just the same
See: http://theregister.co.uk/content/55/24447.html Title: Back Orifice for Unix flaw emerges from obscurity Apparently, it is possible to effectively telnet to a remote Unix box by anonymous XDMCP UDP connections. Analysis for hype, commercial exploitation potentials and exposure of clueless sysadmins - is left to the reader as homework. --- Omer There is no IGLU Cabal. No candidate to the IGLU Cabal could answer the question Who is John Galt?. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
On Mon, Mar 18, 2002 at 01:52:36PM +0200, Malcolm Kavalsky wrote:
Eureka!
Nadav Har'El wrote:
On Mon, Mar 18, 2002, Malcolm Kavalsky wrote about Re: pthreads
question:
I asked one of the top Unix hackers that I know, and he said:
I would guess that if you do large af_unix transfers that are page
aligned then the system doesn't have to actually copy the data rather it
can share the page and do a copy on write. This preserves the socket
semantics and can be faster than memcpy. This was done many years ago in
Solaris.
I wonder if digging deep enough in the kernel sources, will reveal
this ...
You can try to check if this is the case, by following each send or
memcpy
by a memset() of the buffer. If the memcpy method suddenly becomes
quicker,
this explanation might be true.
Strange though - how come malloc() returns page-aligned buffers? Does the
Linux code really checks for this rare and rather esoteric case (if you
write to the buffer after sending it, and the kernel can't know you're
writing whole pages, it will have to do a copy-on- write and do the copy
anyway).
This is exactly what happened! I added in memset after memcpy, and also
after sending the buffer, the results are:
Memcpy'ed and memsetted 1000 blocks of size 1048576 in 18 seconds = 55
Mbytes/second
Started receiving at Mon Mar 18 13:41:13 2002
Received 1048576000 bytes in 17 seconds over unix socket = 59
Mbytes/second
Started sending at Mon Mar 18 13:41:13 2002
Sent and memsetted 1000 blocks of size 1048576 in 17 seconds over unix
socket = 58 Mbytes/second
i decided to play too. i took your code and modified it, so that the
tests are run seperately (since i didnt want the after effects from
fork's COW behaviour to affect the memcpy case). i also modified it to
use getrusage().
here are my results:
[mulix@alhambra tmp]$ for arg in 1 2 3; do ./b memcpy ; done ;
memcpy'ed 1000 blocks of size 1048576. user time: 16.07 secs, system time:
0.06 secs
memcpy'ed 1000 blocks of size 1048576. user time: 15.96 secs, system time:
0.04 secs
memcpy'ed 1000 blocks of size 1048576. user time: 15.92 secs, system time:
0.06 secs
[mulix@alhambra tmp]$ for arg in 1 2 3; do ./b send ; done ;
sent1000 blocks of size 1048576. user time: 6.99 secs, system time:
10.02 secs
sent1000 blocks of size 1048576. user time: 7.42 secs, system time:
10.33 secs
sent1000 blocks of size 1048576. user time: 7.11 secs, system time:
10.38 secs
kernel is 2.4.18rc1, and here's the modified code:
#include stdio.h
#include malloc.h
#include string.h
#include time.h
#include sys/socket.h
#include sys/time.h
#include sys/resource.h
#include sys/un.h
#include sys/types.h
#include sys/wait.h
#include unistd.h
#include assert.h
#define BUFSIZE 0x10 /* 1 Megabyte */
#define NBLOCKS 1000
#define PORT_NAME/tmp/foo
void server(), client();
void socket_benchmark()
{
pid_t rc;
if ( (rc = fork()) == 0 ) {
server();
waitpid(rc, NULL, 0);
} else {
sleep(1); /* Dirty, but ensures client runs after server is ready */
client();
}
}
void server()
{
struct sockaddr_un sin,from;
int s,g,len;
char *buf;
buf = malloc( BUFSIZE );
/* Create an unbound socket */
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
perror(Bad socket\n);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( bind( s, (struct sockaddr *)sin,
sizeof(sin) ) 0){
perror(bind);
return;
}
listen( s, 5 );
len = sizeof(from);
g = accept( s, (struct sockaddr *)from, len );
while( read( g, buf, BUFSIZE ) 0 ); /* sink all data received */
close(g);
close(s);
unlink( PORT_NAME );
}
void client()
{
struct rusage r = {{0},};
struct sockaddr_un sin;
int s;
char *buf;
time_t start_time, elapsed_time;
int i;
assert(!(r.ru_utime.tv_sec | r.ru_utime.tv_usec |
r.ru_stime.tv_sec | r.ru_stime.tv_usec));
buf = malloc( BUFSIZE );
if( (s=socket( PF_UNIX, SOCK_STREAM, 0 )) 0 ){
perror(socket);
return;
}
strcpy( sin.sun_path, PORT_NAME );
sin.sun_family = PF_UNIX;
if( connect( s, (struct sockaddr *)sin, sizeof(sin)) 0 ){
perror(connect);
close(s);
return;
}
start_time = time(0);
for( i=0; i NBLOCKS write(s, buf, BUFSIZE) == BUFSIZE ; i++ )
memset( buf, 'A', BUFSIZE );;
elapsed_time = time(0) - start_time;
close(s);
#if 0
printf(
Re: Telnet by any other time smells like shit just the same
On Mon, Mar 18, 2002, Omer Zak wrote about Telnet by any other time smells like shit just the same: Apparently, it is possible to effectively telnet to a remote Unix box by anonymous XDMCP UDP connections. How is that more dangerous than running a telnet, or ssh server on your machine? If the attacker can guess (or sniff, or whatever) your password, they can simply ssh your server. Why would they bother with XDMCP? Of course, if you thought you disabled the telnet and ssh servers, you might still have a XDMCP server running you didn't mean to have running, which may be the problem they are referring to. To see if you have one running, do netstat -a and see if you have anything listening on port 177 (or xdmcp). On Redhat Linux, for example, this doesn't seem to be enabled by default. There is no IGLU Cabal. No candidate to the IGLU Cabal could answer the question Who is John Galt?. You're mixing up two different mailing lists ;) -- Nadav Har'El|Monday, Mar 18 2002, 5 Nisan 5762 [EMAIL PROTECTED] |- Phone: +972-53-245868, ICQ 13349191 |Always remember you're unique, just like http://nadav.harel.org.il |everyone else. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: About fribidi PHP
Ben-Nes Michael wrote: Hi All I noticed that php added support for fribidi using a function called fribidi_log2vis What will be the benefit if Ill use fribidi_log2vis and not hebrevc/hebrev ? Hebrev(c) pros: built in in PHP, no extra compile options. Documented. cons: bugs, not completely standard-compliant. Charset is only ISO8859-8/Windows-1255. Maintenance status: unmaintained, last time I asked. Fribidi: pros: unicode standard compliant. charset selectable. cons: undocumented, needs a compile option (--with-fribidi) and the fribidi library. Maintenance status: recently introduced, therefore alive. Herouth = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Telnet by any other time smells like shit just the same
On Mon, 18 Mar 2002, Nadav Har'El wrote: On Mon, Mar 18, 2002, Omer Zak wrote about Telnet by any other time smells like shit just the same: Apparently, it is possible to effectively telnet to a remote Unix box by anonymous XDMCP UDP connections. How is that more dangerous than running a telnet, or ssh server on your machine? If the attacker can guess (or sniff, or whatever) your password, they can simply ssh your server. Why would they bother with XDMCP? One item in their advisory: the default configuration of some versions of Mandrake has not only allowed remote root login, but also gives a list of valid users. This is more than what you typically get from telnet or ssh. I suspect that Redhat and Mandrake was too general.They also haven't checked SuSE or other distros. Of course, if you thought you disabled the telnet and ssh servers, you might still have a XDMCP server running you didn't mean to have running, which may be the problem they are referring to. To see if you have one running, do netstat -a netstat -l -n --ip and see if you have anything listening on port 177 (or xdmcp). On Redhat Linux, for example, this doesn't seem to be enabled by default. -- Tzafrir Cohen mailto:[EMAIL PROTECTED] http://www.technion.ac.il/~tzafrir = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
mulix wrote: i decided to play too. i took your code and modified it, so that the tests are run seperately (since i didnt want the after effects from fork's COW behaviour to affect the memcpy case). i also modified it to use getrusage(). here are my results: [mulix@alhambra tmp]$ for arg in 1 2 3; do ./b memcpy ; done ; memcpy'ed 1000 blocks of size 1048576. user time: 16.07 secs, system time: 0.06 secs memcpy'ed 1000 blocks of size 1048576. user time: 15.96 secs, system time: 0.04 secs memcpy'ed 1000 blocks of size 1048576. user time: 15.92 secs, system time: 0.06 secs [mulix@alhambra tmp]$ for arg in 1 2 3; do ./b send ; done ; sent 1000 blocks of size 1048576. user time: 6.99 secs, system time: 10.02 secs sent 1000 blocks of size 1048576. user time: 7.42 secs, system time: 10.33 secs sent 1000 blocks of size 1048576. user time: 7.11 secs, system time: 10.38 secs Interesting, I compiled and ran your code with results: memcpy'ed 1000 blocks of size 1048576. user time: 17.83 secs, system time: 0.01 secs sent1000 blocks of size 1048576. user time: 8.19 secs, system time: 5.67 secs The sum of user and system time is pretty much equal (just as in yours) I then commented out the memset commands and got: memcpy'ed 1000 blocks of size 1048576. user time: 8.90 secs, system time: 0.04 sec sent1000 blocks of size 1048576. user time: 0.00 secs, system time: 0.62 secs This is a dramatic difference. Did you try this ? _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
On Mon, Mar 18, 2002 at 02:39:09PM +0200, Malcolm Kavalsky wrote: mulix wrote: i decided to play too. i took your code and modified it, so that the tests are run seperately (since i didnt want the after effects from fork's COW behaviour to affect the memcpy case). i also modified it to use getrusage(). here are my results: [mulix@alhambra tmp]$ for arg in 1 2 3; do ./b memcpy ; done ; memcpy'ed1000 blocks of size 1048576. user time: 16.07 secs, system time: 0.06 secs memcpy'ed1000 blocks of size 1048576. user time: 15.96 secs, system time: 0.04 secs memcpy'ed1000 blocks of size 1048576. user time: 15.92 secs, system time: 0.06 secs [mulix@alhambra tmp]$ for arg in 1 2 3; do ./b send ; done ; sent 1000 blocks of size 1048576. user time: 6.99 secs, system time: 10.02 secs sent 1000 blocks of size 1048576. user time: 7.42 secs, system time: 10.33 secs sent 1000 blocks of size 1048576. user time: 7.11 secs, system time: 10.38 secs Interesting, I compiled and ran your code with results: memcpy'ed 1000 blocks of size 1048576. user time: 17.83 secs, system time: 0.01 secs sent1000 blocks of size 1048576. user time: 8.19 secs, system time: 5.67 secs The sum of user and system time is pretty much equal (just as in yours) I then commented out the memset commands and got: memcpy'ed 1000 blocks of size 1048576. user time: 8.90 secs, system time: 0.04 sec sent1000 blocks of size 1048576. user time: 0.00 secs, system time: 0.62 secs This is a dramatic difference. Did you try this ? based on nadav's suggestion, i added getrusage() in the server as well. here are the results: [mulix@alhambra tmp]$ echo without memsets: ; ./b memcpy; ./b send without memsets: memcpy'ed 1000 blocks of size 1048576. user time: 7.98 secs, system time: 0.01 secs client sent 1000 blocks of size 1048576. user time: 0.01 secs, system time: 5.07 secs server received 1048576000 bytes. user time: 0.02 secs, system time: 6.75 secs [mulix@alhambra tmp]$ echo with memsets: ; ./b memcpy; ./b send with memsets: memcpy'ed 1000 blocks of size 1048576. user time: 15.54 secs, system time: 0.09 secs client sent 1000 blocks of size 1048576. user time: 7.40 secs, system time: 9.56 secs server received 1048576000 bytes. user time: 0.04 secs, system time: 7.67 secs personally, i'm very curious why the memcpy case takes so much user time. objdump to the rescue. -- The ill-formed Orange Fails to satisfy the eye: http://vipe.technion.ac.il/~mulix/ Segmentation fault. http://syscalltrack.sf.net/ = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: pthreads question
based on nadav's suggestion, i added getrusage() in the server as well. here are the results: [mulix@alhambra tmp]$ echo without memsets: ; ./b memcpy; ./b send without memsets: memcpy'ed 1000 blocks of size 1048576. user time: 7.98 secs, system time: 0.01 secs client sent1000 blocks of size 1048576. user time: 0.01 secs, system time: 5.07 secs server received1048576000 bytes. user time: 0.02 secs, system time: 6.75 secs [mulix@alhambra tmp]$ echo with memsets: ; ./b memcpy; ./b send with memsets: memcpy'ed 1000 blocks of size 1048576. user time: 15.54 secs, system time: 0.09 secs client sent1000 blocks of size 1048576. user time: 7.40 secs, system time: 9.56 secs server received1048576000 bytes. user time: 0.04 secs, system time: 7.67 secs personally, i'm very curious why the memcpy case takes so much user time. objdump to the rescue. What version of kernel are you running ? _ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Telnet by any other time smells like shit just the same
On Mon, 18 Mar 2002, Nadav Har'El wrote: On Mon, Mar 18, 2002, Omer Zak wrote about Telnet by any other time smells like shit just the same: Apparently, it is possible to effectively telnet to a remote Unix box by anonymous XDMCP UDP connections. How is that more dangerous than running a telnet, or ssh server on your machine? If the attacker can guess (or sniff, or whatever) your password, they can simply ssh your server. Why would they bother with XDMCP? This is exactly what I hinted at in the Subject line and in the homework assignment (snipped out of your reply). There is no IGLU Cabal.No candidate to the IGLU Cabal could answer the question Who is John Galt?. You're mixing up two different mailing lists ;) .. whose audience has a big overlap. --- Omer There is no IGLU Cabal. There is an IGLU mailing list. Both concepts must not be mixed up. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
AOL and RH - True ??
Hi IGLU's, We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Hebrew post at Whatsup.co.il : http://whatsup.co.il/article.php?sid=77 Story at reuters.com : http://www.reuters.com/news_article.jhtml?type=searchStoryID=687652 Amir Tal, System Administrator Whatsup - Linux related news And support - in hebrew ! icq : 15748705 http://www.whatsup.co.il = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RE: AOL and RH - True ??
-Original Message- From: Herouth Maoz [mailto:[EMAIL PROTECTED]] Sent: Monday, March 18, 2002 4:41 PM To: Amir Tal; [EMAIL PROTECTED] Subject: Re: AOL and RH - True ?? Amir Tal wrote: We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Story 1: AOL takes over Redhat. Meaning - it makes policy decisions for RedHat from now on. Story 2: AOL hires a group of professionals from a reputable Linux company, to help it transition its machines (servers, employee desktops, who knows) from whatever operating system was previously on them, to RedHat Linux. Compare: 1. AOL takes over Pizza Hut. 2. AOL makes a deal with Pizza Hut to have a restaurant inside the AOL headquarters, for the benefit of the employees. Two different stories entirely. Though I can see how story 2 may have become story 1 as it went along the grapevine. Story 1. no doubt. (if you ask me) There was to much noise about this so called merge in the past few weeks, so I doubt its story 2. AOL wants to switch to Linux, and also wants to get into the open-source community\market . If you wanted to get into a motorcycle's club, how would you arrive to the first meeting ? On a black heavy Harley, or by bus ? Tal. Herouth = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: AOL and RH - True ??
Tal, At least get your news straight.. AOL will NOT buy RedHat - they don't need to buy them as AOL got some shares of RedHat already (from the days Netscape invested in Redhat) RedHat will help AOL moving SOME of their servers from Sun to Linux but will not do the job themselves - There will be a hardware vendor (my guess is either IBM or HP) that will do the dirty job and RedHat will give site support as well as other support needed. That doesn't mean AOL will not have Windows servers there - there will be tons of them which will not move to Linux - streaming servers, and other servers which are depends on Windows. Hetz On Monday 18 March 2002 15:51, Amir Tal wrote: Hi IGLU's, We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Hebrew post at Whatsup.co.il : http://whatsup.co.il/article.php?sid=77 Story at reuters.com : http://www.reuters.com/news_article.jhtml?type=searchStoryID=687652 Amir Tal, System Administrator Whatsup - Linux related news And support - in hebrew ! icq : 15748705 http://www.whatsup.co.il = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: AOL and RH - True ??
Amir Tal wrote: We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Story 1: AOL takes over Redhat. Meaning - it makes policy decisions for RedHat from now on. Story 2: AOL hires a group of professionals from a reputable Linux company, to help it transition its machines (servers, employee desktops, who knows) from whatever operating system was previously on them, to RedHat Linux. Compare: 1. AOL takes over Pizza Hut. 2. AOL makes a deal with Pizza Hut to have a restaurant inside the AOL headquarters, for the benefit of the employees. Two different stories entirely. Though I can see how story 2 may have become story 1 as it went along the grapevine. Herouth = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RE: AOL and RH - True ??
-Original Message- From: Hetz Ben Hamo [mailto:[EMAIL PROTECTED]] Sent: Monday, March 18, 2002 4:51 PM To: Amir Tal; [EMAIL PROTECTED] Subject: Re: AOL and RH - True ?? Tal, At least get your news straight.. AOL will NOT buy RedHat - they don't need to buy them as AOL got some shares of RedHat already (from the days Netscape invested in Redhat) I didn't say buy, said AOL indeed hire a team from RH as a first step of converting some of their systems to Linux Do you say the word buy in there ?? I am getting my news just fine. The buying issue was discussed last time, not now. I never said that. RedHat will help AOL moving SOME of their servers from Sun to Linux but will not do the job themselves - There will be a hardware vendor (my guess is either IBM or HP) that will do the dirty job and RedHat will give site support as well as other support needed. That doesn't mean AOL will not have Windows servers there - there will be tons of them which will not move to Linux - streaming servers, and other servers which are depends on Windows. This is why I said some of their systems ... I didn't say AOL are dumping windows and moving to linux Tal. Hetz On Monday 18 March 2002 15:51, Amir Tal wrote: Hi IGLU's, We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Hebrew post at Whatsup.co.il : http://whatsup.co.il/article.php?sid=77 Story at reuters.com : http://www.reuters.com/news_article.jhtml?type=searchStoryID=687652 Amir Tal, System Administrator Whatsup - Linux related news And support - in hebrew ! icq : 15748705 http://www.whatsup.co.il = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
JLC First Meeting - Final !!
Hi IGLUs, The first meeting of the JLC is underway!! We shall meet on Thu, March 21 @ 18:30 at the IBM Bldg in Har Chotzvim. We shall gather at said time. I guess we will actually start @ 19:00. As you remember, this meeting will be an introduction meeting. We shall discuss what are the issues that interest us and who will lecture about them. Getting-there directions: Softel / IGSI Building 6 11 Kiryat Mada, Har Hotzvim Jerusalem 91450 From the entrance to Har Hotzvim, you pass the Paz gas station and turn in the first street to the left, this is Kiryat Mada street. Our building will be about the fourth on the left side of the street. The building has one entrance, but 2 wings (with associated stairs and elevators). We are in wing B, Floor 2. Turn to the right after exiting the elevator, you will see our door. There is a buzzer button and an intercom on the right side of the door. If it is not open, buzz until somebody is tired of the noise and comes to open the door :-) I hope to see you all there! Amichai Rotman Jerusalem Linux Club Founder = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: JLC First Meeting - Final !!
for those of us who doesn't have cars I think busses 35,36,32,16 are getting there (not sure about 16) non of the busses enters inside so be ready for some walk Ely Levy System group Hebrew University Jerusalem Israel On Mon, 18 Mar 2002, Amichai Rotman wrote: Hi IGLUs, The first meeting of the JLC is underway!! We shall meet on Thu, March 21 @ 18:30 at the IBM Bldg in Har Chotzvim. We shall gather at said time. I guess we will actually start @ 19:00. As you remember, this meeting will be an introduction meeting. We shall discuss what are the issues that interest us and who will lecture about them. Getting-there directions: Softel / IGSI Building 6 11 Kiryat Mada, Har Hotzvim Jerusalem 91450 From the entrance to Har Hotzvim, you pass the Paz gas station and turn in the first street to the left, this is Kiryat Mada street. Our building will be about the fourth on the left side of the street. The building has one entrance, but 2 wings (with associated stairs and elevators). We are in wing B, Floor 2. Turn to the right after exiting the elevator, you will see our door. There is a buzzer button and an intercom on the right side of the door. If it is not open, buzz until somebody is tired of the noise and comes to open the door :-) I hope to see you all there! Amichai Rotman Jerusalem Linux Club Founder = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: X problems
On Mon 2002-03-18, Shai Bentin wrote: Hi list, Lately I've been having X freeze problems. what happens is during work, suddenly the mouse events are not captured, soon after that the keyboard events are gone, and although I know that the machine still functions there is nothing I can do. I can think of 2 options: (1) telnet to it from another machine (2) uses the SysRq key With the second option you can recover from many kinds of trouble, for example ALT-SysRq-r will probably give you the keyboard back. Or ALT-SysRq-i will kill all your processes except init, etc. see linux/Documentation/sysrq.txt for the details. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: X problems
On 18 Mar 2002, Shai Bentin wrote: One more thing, is there any way to break X into regular command line mode, when starting the system in runlevel 5? First thing to try: Ctrl-Alt-F1 (to switch to a text-mode console, login, and try to handle things) Typically it won't work when the X server is badly stuck Ctrl-Alt-Backspace (Kill the X server) Typically won't work under the same circumstances If you're in a network: connect from a different computer (you have an sshd listening, right?) magic sysrq: I quote another message: With the second option you can recover from many kinds of trouble, for example ALT-SysRq-r will probably give you the keyboard back. Or ALT-SysRq-i will kill all your processes except init, etc. see linux/Documentation/sysrq.txt for the details. ALT-SysRq-h should give you a Help: a short list of available commands. Beware, as those commands allow you to kill processes, and reboot your system even without a sync of the disks, not to mention a proper shutdown. -- Tzafrir Cohen mailto:[EMAIL PROTECTED] http://www.technion.ac.il/~tzafrir = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Executing a Script on KDE's startup
Ilya Konstantinov [EMAIL PROTECTED] writes: On Sat, 2002-03-16 at 09:39, Shlomi Fish wrote: I'd like to execute a script automatically when KDE's starts. How do I do that? Alternatively it can be something that X starts after everything was initialized. Place the executable or a symbolic link to it in ~/.kde/Autostart/ I never figured out (asked on this list and elsewhere a while ago) regarding execution of scripts on KDE (or GNOME) logout. If anyone stumbles on it, please let me know. -- Oleg Goldshmidt | [EMAIL PROTECTED] If it aint't broken it hasn't got enough features yet. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: heb fonts solved, previously hebrew fonts under linux
On Wednesday 27 February 2002 16:02, Shai Bentin wrote: I have successfully installed hebrew fonts under abiword 0.99 with bidi. I'll write a small how-to and publish it here in a few days time. Anybody who can't wait can try and e-mail me directly. Shai Hi shai (list) I managed to install hebrew fonts under abiword 0.9.2, and I saw that the one I have is not bidi enabled. I want to d/l a bidi-enabled binary for mandrake. are they available? How about the how-to? will it be available soon? - diego -- You should never bet against anything in science at odds of more than about 10^12 to 1. -- Ernest Rutherford = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: AOL and RH - True ??
you all forgot one of the most interesting part of this story: AOL is developing a web browser based on gecko. It is been told that that is what is going to take a little bit of the IE market (about 39 million users to be more exact). - diego On Monday 18 March 2002 16:47, Amir Tal wrote: -Original Message- From: Herouth Maoz [mailto:[EMAIL PROTECTED]] Sent: Monday, March 18, 2002 4:41 PM To: Amir Tal; [EMAIL PROTECTED] Subject: Re: AOL and RH - True ?? Amir Tal wrote: We already discussed this matter a few weeks ago, when ira posted news he saw at The times (I think) about an upcoming merge between AOL and Redhat. This issue was discussed on the list for a few days, and we ended up thinking that it will not happen, at least not in the near future..remember ? Well - its happening ! AOL indeed hire a team from RH as a first step of converting some of their systems to Linux ! I guess that this is not as far as we first figured . Story 1: AOL takes over Redhat. Meaning - it makes policy decisions for RedHat from now on. Story 2: AOL hires a group of professionals from a reputable Linux company, to help it transition its machines (servers, employee desktops, who knows) from whatever operating system was previously on them, to RedHat Linux. Compare: 1. AOL takes over Pizza Hut. 2. AOL makes a deal with Pizza Hut to have a restaurant inside the AOL headquarters, for the benefit of the employees. Two different stories entirely. Though I can see how story 2 may have become story 1 as it went along the grapevine. Story 1. no doubt. (if you ask me) There was to much noise about this so called merge in the past few weeks, so I doubt its story 2. AOL wants to switch to Linux, and also wants to get into the open-source community\market . If you wanted to get into a motorcycle's club, how would you arrive to the first meeting ? On a black heavy Harley, or by bus ? Tal. Herouth = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] -- The smiling Spring comes in rejoicing, And surly Winter grimly flies. Now crystal clear are the falling waters, And bonnie blue are the sunny skies. Fresh o'er the mountains breaks forth the morning, The ev'ning gilds the oceans's swell: All creatures joy in the sun's returning, And I rejoice in my bonnie Bell. The flowery Spring leads sunny Summer, The yellow Autumn presses near; Then in his turn come gloomy Winter, Till smiling Spring again appear. Thus seasons dancing, life advancing, Old Time and Nature their changes tell; But never ranging, still unchanging, I adore my bonnie Bell. -- Robert Burns, My Bonnie Bell = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: GPLed flash plugin?
On Monday 18 March 2002 20:55, Tzafrir Cohen wrote: http://www.swift-tools.com/Flash/ Anybody tried it? It's quite old (been since the KDE-1.x days) support is on par up to flash 3.. Hetz = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Busses - Thanks
Hi all, I thank Ely for mentioning the buses. I forgot about them, which is starange, due to the fact I might use one to get there myself... Amichai. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: GPLed flash plugin?
On Mon, 18 Mar 2002, Hetz Ben Hamo wrote: On Monday 18 March 2002 20:55, Tzafrir Cohen wrote: http://www.swift-tools.com/Flash/ Anybody tried it? It's quite old (been since the KDE-1.x days) support is on par up to flash 3.. Any idea if it really matters? Rational: I really don't need all of those cute flash animations. I usually get annoyed by flash animations in some very annoying adds. However certain sites rely on flash for navigation, and no browser that I know allows me to toggle flash support on runtime. So I'm looking for thin or lynx-like flash support -- Tzafrir Cohen mailto:[EMAIL PROTECTED] http://www.technion.ac.il/~tzafrir = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
