Re: Does anyone use MATE (Gnome 2 fork)?
Hello, Just FYI: I tried it today on a newly installed Fedora 20 machine, with jdk-6u38-linux-amd64.rpm and with JEdit 4.2, which is considered a stable version, and again the same problem of losing the focus occurs immediately, So I again will probably move to KDE. Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Does anyone use MATE (Gnome 2 fork)?
Hi, The main issues I encountered in last months in KDE are: 1) loads application slowly compared to Gnome. 2) sometimes KDE plasma crashes when I press on the Kickoff application launcher 3) the kwallet pop up annoys me (though I am sure it is trivial to get rid of this pop up) I wonder which other problems did you have with it ? KDE is quite veteran as you know. I did not try the other Display Managers. In case this problem with Gnome and JEdit losing focus will not be solved somehow (which I doubt it will), I will consider one of them (starting with Xfce). Regards, Rami Rosen On Wed, Dec 25, 2013 at 5:49 PM, Steve Litt sl...@troubleshooters.comwrote: On Wed, 25 Dec 2013 15:38:31 +0200 Rami Rosen rosenr...@gmail.com wrote: Hello, Just FYI: I tried it today on a newly installed Fedora 20 machine, with jdk-6u38-linux-amd64.rpm and with JEdit 4.2, which is considered a stable version, and again the same problem of losing the focus occurs immediately, So I again will probably move to KDE. Rami, Would you consider Xfce, LXDE, OpenBox or IceWM? I've had *a lot* of trouble with KDE in the past. Thanks, SteveT Steve Litt* http://www.troubleshooters.com/ Troubleshooting Training * Human Performance ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Does anyone use MATE (Gnome 2 fork)?
Hi, The issue I with JEdit I had is probably not related to the fact it is Java based application. What I had is that when pressing ctrl-f for search in JEdit , which is a very frequent operation, the JEdit search box was opened, but the focus was not on the search box of the search dialog. So I had to manually go with the mouse there and click to obtain the focus. This issue did not occur on previous Fedora distros. The problem was persistent over several Fedora machines with Gnomer3. Regards, Rami Rosen http://ramirose.wix.com/ramirosen#!home/mainPage On Sat, Nov 16, 2013 at 6:17 PM, E.S. Rosenberg esr+linux...@g.jct.ac.ilwrote: I am puzzled how is a Java based program influenced in the slightest bit by what DE you're using? 2013/11/15 Rami Rosen rosenr...@gmail.com: Hi, I was also disappointed with Gnome 3 on recent Fedora releases. I tried using Gnome classic mode but still there were issues. Most disturbing was with the JEdit editor. I also tried using MATE, but I still had issues with the JEdit editor. So after years of using Gnome on various Fedora releases, I switched several months ago to KDE, after years of using Gnome 3. The issues with JEdit are solved with KDE. Note: 1) it seems to me that launching an application in KDE (from the application menu or from a desktop shortcut) is really slower than with Gnome. However I did not tried to delved into it - maybe there are some KDE plugins which can be disabled, or other tweaks. 2) The issues I had with JEdit might be due to that I am not using the latest release of JEdit (from my own/legacy reasons). I do not want to upgrade from my own reasons, which I can describe if anyone is interested to hear. Rami Rosen http://ramirose.wix.com/ramirosen On Fri, Nov 15, 2013 at 12:09 PM, Omer Zak w...@zak.co.il wrote: After reading Ilan Shavit's blog article about his disappointment from Gnome 3 (http://ilsh.info/archives/3084), I found about the Gnome 2 fork MATE (http://mate-desktop.org/). Before installing MATE on my Debian Wheezy system, I would like to know other people's experience with it. Thanks, --- Omer -- We will only have peace with the Arabs when they love their children more than they hate us. Golda Meir (Israeli Prime Minister between 1969-1974) My own blog is at http://www.zak.co.il/tddpirate/ My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with which I may be affiliated in any way. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Does anyone use MATE (Gnome 2 fork)?
Hi, I was also disappointed with Gnome 3 on recent Fedora releases. I tried using Gnome classic mode but still there were issues. Most disturbing was with the JEdit editor. I also tried using MATE, but I still had issues with the JEdit editor. So after years of using Gnome on various Fedora releases, I switched several months ago to KDE, after years of using Gnome 3. The issues with JEdit are solved with KDE. Note: 1) it seems to me that launching an application in KDE (from the application menu or from a desktop shortcut) is really slower than with Gnome. However I did not tried to delved into it - maybe there are some KDE plugins which can be disabled, or other tweaks. 2) The issues I had with JEdit might be due to that I am not using the latest release of JEdit (from my own/legacy reasons). I do not want to upgrade from my own reasons, which I can describe if anyone is interested to hear. Rami Rosen http://ramirose.wix.com/ramirosen On Fri, Nov 15, 2013 at 12:09 PM, Omer Zak w...@zak.co.il wrote: After reading Ilan Shavit's blog article about his disappointment from Gnome 3 (http://ilsh.info/archives/3084), I found about the Gnome 2 fork MATE (http://mate-desktop.org/). Before installing MATE on my Debian Wheezy system, I would like to know other people's experience with it. Thanks, --- Omer -- We will only have peace with the Arabs when they love their children more than they hate us. Golda Meir (Israeli Prime Minister between 1969-1974) My own blog is at http://www.zak.co.il/tddpirate/ My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with which I may be affiliated in any way. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Motherboards for new Ubuntu install
Hi, About four years ago I bought a laptop in one of the stores mentioned in the thread. It turned out that it came with a different wireless nic than that which appeared in the official manufacturer link! It came with a wireless nic without 80211n support. And it was not a link to a wrong model. Eventually it turned out that it was a rare error from the manufacturer (They said that the same model name has different ingredients for the middle east and for Europe). And this manufacturer is a big, well known and veteran company.Since this was important to me at that time, I asked to replace it, but the store did not want to negotiate this with the supplier, so I had to negotiate it by myself. So keep in mind that this kind of things also can happen! regards, Rami Rosen http://ramirose.wix.com/ramirosen On Thu, Dec 27, 2012 at 9:00 AM, Dotan Cohen dotanco...@gmail.com wrote: On Thu, Dec 27, 2012 at 1:29 AM, E.S. Rosenberg esr+linux...@g.jct.ac.il wrote: Note that that is a meta-package that installs the kernel specific package, so you'll need to download at least the kernel specific one and the meta package is good to make sure that after a kernel upgrade you still have the drivers. Regards, Eliyahu - אליהו Thanks. -- Dotan Cohen http://gibberish.co.il http://what-is-what.com ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Kernel module that used in MANET creation ?
Hi, * MANET - mobile ad hoc networks some times refers as mesh networks First, I suggest you will take a look in the mesh mode slides here: http://www.haifux.org/lectures/206/ Second, assuming that MANET works with IEEE 802.11s wireless mesh standard (please verify this) , then certainly mac80211 is the kernel module you should learn. Especially what you should look in is: net/mac80211/mesh.c net/mac80211/meshhwmp.c (Layer 2 Hybrid Wireless Mesh Protocol) net/mac80211/mesh_pathtbl.c net/mac80211/mesh_plink.c net/mac80211/mesh.h Keep in mind that very few drivers support mesh mode, so you must verify that your wireless network card does support mesh mode. ; moreover, you should make sure that your kernel version is new enough to support mesh mode. Rgs, Rami Rosen 2010/3/8 Boris shtrasman borissh1...@gmail.com: Hi , I'm on the last year of software Engineering. studies (B.Sc), and I wish to do the Engineering project in MANET area. While I found lots of theoretical information , I'm strugling with find a _maintained_ module in the kernel tree. Currently I study the mac80211 module(but not sure if this is the right one). Anyway what will be the right module to work in ? since I wish to integrate the work directly into a used module and not create a new one. P.S * MANET - mobile ad hoc networks some times refers as mesh networks Project information - Creating an adaptive network that will change the location of the nodes according to predefined rules (economic , reception , area coverage etc...) on calculated terrain. -- -- -- Boris Shtrasman |Gnu/Linux Software developer | | IM : bori...@jabber.org | | URL : myrtfm.blogspot.com| ___ ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: [SUMMARY, 95% SUCCESS] Re: Attempt to buy a Linux-compatible Bluetooth dongle
Hi, hcid daemon was already active, with the -x -s options. man hcid provided no explanation of the -x option -x means stand for experimental in hcid daemon of bluez-utils. This is mainly for development tests. Just a thought: can you test it with restarting the hcid daemon without this x flag? rgs, Rami Rosen On Wed, Jul 29, 2009 at 7:46 AM, Omer Zakw...@zak.co.il wrote: On Tue, 2009-07-28 at 14:19 +0300, Nitzan Brumer wrote: I bought the LVT-010 bluetooth dongle for my acer one. I use it with Ubuntu 8.10 and it works great out of the box. no problem there. http://www.bug.co.il/prodtxt.asp?id=6233perur=1t=10 Thanks also to Geoffrey Mendelson, Noam Rathaus and Rami Rosen. The message was that nowadays Linux compatibility of Bluetooth devices is not a problem. I ended up buying a no-name brand, which looks the same as LVT-010. It identified itself to lsusb as: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode) According to http://www.qbik.ch/usb/devices/search_res.php?pattern=bluetooth this USB dongle is Linux compatible. While this dongle was found by my Nokia E71 cellphone, initially I was not successful in doing things with it. However http://howto-pages.org/v600i_bluetooth/ provided me with useful advice how to test the connection, which proved to work out of the box (except for failure to pair and for not working under regular user): - '/sbin/lsmod' showed rfcomm,hci_usb,l2cap. - As root, 'hciconfig' displayed details about device hci0. In particular, I saw the line: UP RUNNING PSCAN ISCAN - 'hcitool scan' found my Nokia E71, in which I activated Bluetooth and set it to be discoverable. - 'l2ping xx:xx:xx:xx:xx:xx' (where xx:xx:xx:xx:xx:xx is the E71's address as reported by 'hcitool scan') worked. The document was not up-to-date when it came to dealing with PIN helpers. hcid daemon was already active, with the -x -s options. man hcid provided no explanation of the -x option, while -s option meant that an internal SDP server was activated (probably meaning that no sdpd daemon is necessary). - The command 'kbluetoothd' is now 'kbluetooth' and it did not discover my E71 when run under my user; but under root it did discover - if my E71's bluetooth was already active when 'kbluetooth' was invoked. Then when it accepted files from the E71, they were saved under root. - 'kbtobexclient' worked out of the box, and I was successful in copying a file from PC to the E71. - I was not successful in my attempts to pair the PC with the E71. P.S.: my system is Debian Lenny, with bluez installed from Debian repositories. --- Omer -- MS-Windows is the Pal-Kal of the PC world. My own blog is at http://www.zak.co.il/tddpirate/ My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with which I may be affiliated in any way. WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Attempt to buy a Linux-compatible Bluetooth dongle
Hi, You may find some info here about a problem with 64 bit linux machine and CSR based bluetooth adapter: (It talks about BT headset) http://technomosh.blogspot.com/2009/07/why-reboot-why.html Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Setting NFS server on Fedora Core 9
Ori, 1) can you post here the output of your mounting trial with verbose mode: (mount -v) ? (and the exact mount options you are using) 2) Also : can you post /etc/exports ? Best Regards, Rami Rosen On Thu, Jun 25, 2009 at 2:46 PM, Dotan Shavitdo...@shavitos.com wrote: On Thursday 25 June 2009, Ori Idan wrote: I have checked again and made sure firewall is disabled and still get the same problem. First try mounting from the same machine. And... check the log files... # ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Netfilter module name oddity
Hello, All this is true, but I just want to add these two points: In Ira mail, he mentions many modules which start with xt_* like: 10809 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_connmark.ko 12079 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_CONNMARK.ko 10066 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_dscp.ko 11193 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_DSCP.ko 9885/lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_mark.ko 10903 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_MARK.ko 10137 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_rateest.ko 13381 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_RATEEST.ko 9193/lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_tcpmss.ko 12506 /lib/modules/2.6.26-1-amd64/kernel/net/netfilter/xt_TCPMSS.ko First, it is true also for these modules, that when you see lower case it is a matching module, and when you see upper case it is a target module. (most of them, if not all, have also MODULE_ALIAS to a corresponding ipt_*.) And second: there are no ipt_* modules under net/netfilter; only xt_* modules. the difference between the ipt_* modules, which reside under net/ipv4/netfilter (like net/ipv4/netfilter/ipt_ecn.c, net/ipv4/netfilter/ipt_ECN.c, net/ipv4/netfilter/ipt_LOG.c and more) and the xt_* modules under net/netfilter is that (naturally) the ipt_* modules under net/ipv4/netfilter handle only IPV4 packets (The family member of the target (xt_target) or of the match (xt_match) is NFPROTO_IPV4), while the xt_* modules (under /net/netfilter) support is not limited to IPV4 (They have usually NFPROTO_UNSPEC or have targets/matches arrays with both NFPROTO_IPV4 and NFPROTO_IPV6). Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Hebrew FAQ
Hi, In Gnome, it is simple done thus: right click on the panel select Add to Panel select Keyboard inidcator. An indicator name USA will appear on the panel (assuming you use the default USA english). Then, in order to make this USA an english-hebrew toggle: Right click on the USA. Go to Keyboard prefernces. Go to layouts Choose add. Select Israel That's it. Left click on USA will change it to Isr and the language to Hebrew, clicking again will set it to english back. Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Re: Setting up a PBX for Israel-US communication
Hello, It is the most compressed out of all the audio codecs As a matter of fact and as far as I know, the G.729 codec and G.729A codec are more compressed than gsm. (It could be that it is not supported by twinkle or other clients, I don't know ) See: http://en.wikipedia.org/wiki/G.729 Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Windows Free Laptop (one more)
Hello, Just FYI: - I happened to be today at an Ivory shop (http://www.ivory.co.il/) - I asked about Laptops without Windows; they offered me some MSI laptops (which I didn't like so much). These MSI laptops are with freeDos. - When I asked about some Lenovo models, they told me that it is impossible to get them without Windows. - When I told them about Zvi Devir and the Windows refund case with the Dell laptop (which appeared in the newspaper) they told me that they did not hear about it before ; they added that if the same case happened with a Lenovo laptop, there might be a place for talking about it. I want to add that I am not affiliated with Ivory or any other PC Vendor. Regards, Rami Rosen ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
Links and some info about IPv6 in Linux lecture in Herzelinux yesterday (6.11.08)
Hello, As I promised, here are some links and some additional info about IPv6 in Linux lecture, which was given in Herzelinux yesterday (6.11.08) by me: 1) The article about IPv6 I wrote, which was published in linuxdevices two days ago, is here: http://www.linuxdevices.com/articles/AT7843301253.html 2) Slides for today IPv6 lecture are available in tuxology web site are here: http://tuxology.net/lectures/ipv6-in-the-linux-kernel/ 3) Network namespaces and sysfs: This issue was raised at the end of the lecture. For those who wonder what is the tagged sysfs patch and what it has to do with network namespaces and why it was dropped, here is a short explanation and links: The problem with network namespaces is that when using some processes, each with its own network stack (using network namespaces), it causes a problem regarding sysfs; the reason is that one process has, for example, /sys/class/net/eth0 entry, but a second process, with a different namespace, can also have a (different) /sys/class/net/eth0 entry. The entry /sys/class/net/eth0 has an important data :flags, statistics, and other parameters which are unique for each process with a different network stack. A solution for it was proposed by tagged sysfs, and patches were sent already in August 2008. They were accepted into GregKH tree, but as I said, recently GregKH announced that they are dropped because of a BUG. In this thread in LKML (Linux Kernel Mailing List), GregKH talks about dropping the sysfs patch from his tree: http://marc.info/?l=linux-kernelm=122318517411596w=2 the patch is: driver-core/sysfs-implement-sysfs-tagged-directory-support.patch see also: Sysfs and namespaces article in: http://lwn.net/Articles/295587/ 4) The links to the two vpn solutions I mentioned (and do not appear in the slides) are: http://www.openswan.org/ http://www.openvpn.org/ Best Regards, Rami Rosen = 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: SCTP and netstat
Hello, Noam, The official last net-tools version (from this URL: http://www.tazenda.demon.co.uk/phil/net-tools/) does not include support for sctp. Probably the Debian net-tools package you use does not include support for sctp, too. In fedora core 9, (maybe also in older fedora distros) there is support for sctp via some patches made by Redhat, for supporting sctp. On fedora 9 , the netstat which is part of the net-tools, works with sctp. I created a simple listening sctp socket; When running netstat -anl | grep we get: sctp0.0.0.0:LISTEN (BTW, netstat -nl | grep does not give anyhting, while on TCP I believe netstat -nl is enough). If you must work on Debian, you can try the following: get the source and Fedora patches (including SCTP patches). You can get it ,from bash shell, thus: CVSROOT=:pserver:[EMAIL PROTECTED]:/cvs/pkgs cvs checkout net-tools (look for net-tools-1.60-sctp.patch net-tools-1.60-sctp-quiet.patch). And then try to apply these patches; it may require some additional work due to some Fedora quirks. Rgs, Rami Rosen On Tue, Oct 28, 2008 at 9:21 AM, Noam Rathaus [EMAIL PROTECTED] wrote: Hi, I am trying to 'upgrade' my netstat and make it display SCTP related stream/listeners/etc Any idea how to do this? I am using the latest (unstable) net-tools package from Debian but it still doesn't show. -- Noam Rathaus CTO [EMAIL PROTECTED] http://www.beyondsecurity.com Know that you are safe. Beyond Security Finalist for the Red Herring 100 Global Awards 2007 = 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] On Tue, Oct 28, 2008 at 9:21 AM, Noam Rathaus [EMAIL PROTECTED] wrote: Hi, I am trying to 'upgrade' my netstat and make it display SCTP related stream/listeners/etc Any idea how to do this? I am using the latest (unstable) net-tools package from Debian but it still doesn't show. -- Noam Rathaus CTO [EMAIL PROTECTED] http://www.beyondsecurity.com Know that you are safe. Beyond Security Finalist for the Red Herring 100 Global Awards 2007 = 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: [Herzelinux] Linux Kernel Networking - the Routing Subsystem (Rami Rosen)
Hello, It does not, naturally. However, I'd expect other people who come from Tel Aviv or souther by bus and possibly by car to experience similar problems. So you need to take it into account. I just wanted to note that in that (historic!) day, when the first Herzelinux lecture was given, July 17, 2008 (Thursady) , there were no trains at all during all day due to maintainence railway work (all lines from Tel Aviv to Nahariya). This probably contributed to the fact thay there were traffic jams, etc (though it is hard to estimate in numbers). I remember this no-trains day without any doubt because I had to skip the idea of taking a train that day. See also: http://glz.msn.co.il/NewsArticle.aspx?NewsId=24389 Regards, Rami Rosen On Wed, Aug 6, 2008 at 4:09 PM, Shlomi Fish [EMAIL PROTECTED] wrote: Hi! On Wednesday 06 August 2008, Gilad Ben-Yossef wrote: Shlomi Fish wrote: On Wednesday 06 August 2008, Gilad Ben-Yossef wrote: On Thursday, 7/8, Hezelinux will hold it's second meeting. Topic: Linux Kernel Networking - the Routing Subsystem Lecturer: Rami Rosen Where: Hi Tech College, Maskit 27, Herzelia Pituch. Time: Thursday, 7/8, 18:30. One should note that this meeting co-incides with the PHP-Israel meeting: http://xrl.us/omvcn I learned about both meetings today, and now I'm contemplating which meeting to go to. Can you please next time make sure these meeting don't fall on one another? Well, this is most unfortunate indeed, but I just learned of the PHP meeting myself. I usually check that there are no conflicts with Haifux and Telux. I'll now added the PHP club to the list. Any other clubs I should check? There's a list of them here: http://www.shlomifish.org/open-source/resources/israel/guide-to-israeli-foss-resources/#clubs Here's what I know of their status: 1. Haifux is active and meets on Mondays. 2. Telux is active and meets on Sundays. 3. The Jerusalem Linux Club is not active. 4. Herzelinu is naturally active and meets on Thursdays. 5. Perl-Israel have merged with Telux into OSDClub Tel Aviv. 6. Python-Israel are not active. (Possibly also part of OSDClub-TA now). 7. Ruby-Israel are not active. (Possibly also part of OSDClub-TA now). 8. PHP-IL are active and meet on Thursdays. 9. I don't know anything about the Israeli Grid Association. 10. Don't think Hebrew SUSE have meetings. 11. Redhat.org.il don't meet either I think. 12. Mandriva-IL don't have meetings of their own from what I know. 13. Debian-IL, Ubuntu-IL and Hackers-IL don't either - they're just online communities. - There's now also: http://www.hackers.org.il/mediawiki/index.php/Rehovot_LUG But I don't know anything about them. They said they'll meet on Wednesdays. - The meeting takes place in Herzelia Pituch (Maskit 27, In hi-tech college), a short walking distance from most of Silicon Wadi companies, to the pleasure of the lazy Hi Tech bums :-) Well, the problem with that is that I still recall the trouble it took me to go to Herzelia Pitu'akh from my home in Ramat Aviv Gimel on a Thursday. Most buses and service cabs didn't stop at my station, there was a huge traffic jam, and the bus that I could finally ride was fully packed. Well, amazingly enough, the world does not revolve around you, Shlomi :-) It does not, naturally. However, I'd expect other people who come from Tel Aviv or souther by bus and possibly by car to experience similar problems. So you need to take it into account. I think I'll go to the PHP meeting tomorrow instead to save this inconvenience. Perhaps you'd like to choose a different day than Thursday, or relocate? No, I don't think I will. As you wish - it's just that I and other people will have to accomodate with this awkward choice of location and timing. Next time, please give us some time to announce it. This is for tomorrow, so I don't have a lot of time to publicise it. This was just a reminder. The event was published over a week ago on the club site and facebook group. I see. However, I can't tell how to receive an RSS feed for the events. This feed - http://tuxology.net/feed/ - does not contain any announcements of events. And the Google Calendar does not seem to provide any RSS feeds. I have my own Calendar at home, so I don't need one at Google. I registered as a fan of the facebook group, and hope I'll receive updates by email from there, but there should be a way to do it using RSS too. So, no wonder I didn't know about it and learned about it only now. With OSDClub Tel Aviv, I'm trying to publicise events in Whatsup (newswire + calendar), Linmagazine (events feed), relevant mailing lists and now Facebook at least a week in advance. (Sometimes I forget, etc. but that's the ideal) A previous day reminder to Linux-IL is not enough for people to learn about your event. Regards, Shlomi Fish For more details
Re: NIC woes with Debian MSI MS-7507
Hi, It is probably a BUG in the driver. Google didn't find anything useful Build your own search engine ; don't use google :) Accoding to this link, the same issue also occurred (ethtool returns FIBRE for r8168). http://www.spinics.net/lists/netdev/msg43754.html Since that this message is from 2007, I would consider getting the linux driver from the vendor site: http://www.realtek.com.tw/downloads/downloadsView.aspx?Langid=1PNid=13PFid=5Level=5Conn=4DownTypeID=3GetDown=false version 8.006.00is quite recent ( 22/4/2008). The second , less probable option, is that the motherboard does not support this chipset; There were (very rare ) cases in which such was the case; but these things happen. This can easily be checked with their support. Regards, Rami Rosen On Tue, Jul 15, 2008 at 5:14 PM, ronys [EMAIL PROTECTED] wrote: Hi, Debian etch (latest stable) distro installed on the above PC results in an unusable onboard NIC. /var/log/messages tell me that the NIC is Jul 15 19:15:54 hostname kernel: eth0: RTL8168b/8111b at 0xf8822000, 00:1d:92:a1:96:19, IRQ 177 The module for it is found and loaded: Jul 15 19:15:54 hostname kernel: r8169 Gigabit Ethernet driver 2.2LK-NAPI loaded However, the NIC never gets an IP address from DHCP. Even worse, ethtool shows that it's totally confused as to its identity: # ethtool eth0 Settings for eth0: Supported ports: [ FIBRE ] Supported link modes: 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: Not reported Advertised auto-negotiation: Yes Speed: 1000Mb/s Duplex: Full Port: FIBRE PHYAD: 0 Transceiver: internal Auto-negotiation: on Supports Wake-on: pumbg Wake-on: pumbg Current message level: 0x0033 (51) Link detected: yes (FIBRE port when it really should be TP - Twisted Pair) Google didn't find anything useful. My current workaround is adding another NIC, which works fine, but is unacceptable in the long term. This occurs on two different PCs that have the same motherboard, so it's not a fluke hardware issue. Also, if I play around with ifup/ifdown AFTER booting it, I can sometimes get it to work, but only at 10MB/sec (connected to a 100 MB/s switch). Any ideas? Rony 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: Linux and video multicasting
Hello, I suggest that you try the pim daemon of Xorp: http://www.xorp.org/ Xorp is a large project; it contains the PIM-SM daemon (Protocol Independent Multicast-Sparse Mode) to handle multicasting in Linux. Make sure that your kernel is built with multicast support (recent kernels in most distros come with it). Regards, Rami Rosen On Wed, Jul 9, 2008 at 3:06 PM, Arie Skliarouk [EMAIL PROTECTED] wrote: Hi, My friend needs to broadcast an event live for 15 viewers. Unicast broadcasting requires tremendous bandwidth and hence - money. Thus he asked me to check multicast option. There are instructions on doing multicast using VLC: http://www.youtube.com/watch?v=rpkmdi1gKvM but during my researches I discovered that few deployed routers support multicasting technolody, and even then, the resulting video stream would suffer from slightest packet drop. Can anybody comment? If I understand correctly, every multicast-enabled router or firewall must monitor subscribed multicasts and periodically poll the clients, whether they are still interested in getting the multicast stream. I have not found a way to forward multicast packets on Linux. Once there was package mrouterd, but it is not shipped anymore. Have anybody worked with multicasted video on Linux? -- Arie = 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: linux vserver vs. Xen?
Hi, maybe using vservers instead of a fully para-virtualized Xen Question: are you talking about lvs (http://www.linuxvirtualserver.org/) or about Linux-VServer (http://linux-vserver.org/Welcome_to_Linux-VServer.org) or maybe something else? Regards, Rami Rosen On Mon, May 5, 2008 at 9:07 AM, Amos Shapira [EMAIL PROTECTED] wrote: Hello, We are currently use Xen on CentOS 5.1 to host several machines of ours (i.e. everything on that physical box is in our control). Since we decided to standardise on CentOS (it was decided for us - hard to find hosters who support Debian without extra hassle), which means that all virtual hosts run the same OS, I though that maybe using vservers instead of a fully para-virtualized Xen machine for each function would let us squeeze out that much more out of our hardware (memory, cpu, io). Two questions: 1. Is this correct (that vserver will run more efficiently in that situation)? 2. How does the vserver support in CentOS 5.1 compares to Xen? I mean - with Xen it's just a matter of installing a couple of packages and off you go (especially now that the learning curve is mostly behind me), with vserver - I can't even find a package which mentions this string in its name or description (yum search vserver). So is it worth the hassle? We expect to use the servers with very high IO (lots of disk access) and CPU utilization. Thanks, --Amos = 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: Programatically configure VLAN
Hi, If you are talking only about tasks as create and remove vlan-devices (8021.q) , then this is merely matter of building a request and sending it by ioctls (like SIOCSIFVLAN for adding /removing with ADD_VLAN_CMD or DEL_VLAN_CMD). See: /usr/include/linux/if_vlan.h and /usr/include/linux/sockios.h As far as I understand, cfgsh can be easily extended using this mechanism. Or are you talking about doing more than that ? Regards, Rami Rosen On Wed, Apr 30, 2008 at 2:35 PM, Noam Rathaus [EMAIL PROTECTED] wrote: Hi, I am trying to extend the software cfgsh by Gilad to support VLAN, anyone can point me to Linux API that would allow me to do this? I rather not access /proc/net/vlan/ directly if possible. -- Noam Rathaus CTO [EMAIL PROTECTED] http://www.beyondsecurity.com Know that you are safe. Beyond Security Finalist for the Red Herring 100 Global Awards 2007 = 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: Porting Code from IPv4 to IPv6
Hi, This HowTo, written by Eva M. Castro, may help you as a start: http://gsyc.es/%7Eeva/IPv6-web/ipv6.html Regards, Rami Rosen On Wed, Apr 16, 2008 at 8:56 PM, Noam Rathaus [EMAIL PROTECTED] wrote: Hi, Anyone has a link/site/guide on how to port code written for IPv4 to support IPv6? I have found numerous documentation for Windows - with guides and samples (courtesy of Microsoft), but couldn't find one for Linux that is of good quality. -- Noam Rathaus CTO [EMAIL PROTECTED] http://www.beyondsecurity.com Know that you are safe. Beyond Security Finalist for the Red Herring 100 Global Awards 2007 = 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]
IPv6 in the Linux Kernel lecture slides are available now
Hi, For all those who were interested: The slides of the IPV6 in the Linux Kernel lecture are on Haifux Website now: see: http://www.haifux.org/lectures/187/netLec3.pdf This lecture is the third lecture in the advanced Linux Kernel Networking series, dealing with Linux Kernel implementation of IPv6, ICMPv6, Autoconfiguration, Router Advertisements and Router Solicitations, RADVD, MLDv1/MLDv2, DHCPv6, IPv6 Addresses and more. Regards, Rami Rosen = 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: IPv6 support in old (2.4) kernels
Hi,
Following is a fifth comment: (better later than never)
It occurred to me that the IPv6 Stateless Address Autoconfiguration
(SLAAC) RFC you mention here (RFC2462, from 1971 )
is obsolete; it was replaced by RFC 4862, from 2007. There were some
changes and clarifications, especially regarding DAD (Duplicate
Address Detection); for example, the section that describes what
happens when DAD fails is much more detailed. (Section 5.4.5)
The new RFC removes references to site-local addresses (This type of
IPV6 addresses does not exist anymore); and more.
Appendix C of the new RFC (RFC 4862) details the changes since RFC 2462.
Regards,
Rami Rosen
On Sun, Mar 16, 2008 at 5:38 PM, Oleg Goldshmidt [EMAIL PROTECTED] wrote:
Hi folks,
I suppose there are some people here who for one reason or another still run
kernel 2.4 on servers or embedded systems and are willing to admit to it.
Can anyone tell me how well kernels circa 2.4.19 (or later) support the
*current* IPv6 specification? I see IPv6 support in the code
(http://lxr.linux.no/linux-old+v2.4.19/net/ipv6/) which is marked as
EXPERIMENTAL and carries warnings in the configuration, but it looks like
this is associated with some unloading bug when IPv6 is configured as a
module (http://lxr.linux.no/linux-old+v2.4.19/net/Config.in). The same
warning appears in 2.4.31 which is the last 2.4 in LXR at least.
Suppose I will be happy to compile IPv6 into the kernel (not as a module) -
I am still interested in how well 2.4 supports today's IPv6. More
specifically, after some research I have come to a conclusion that I am
interested in assessment of 2.4 support for (at least) the following:
* IPv6 (RFC2460)
* ICMPv6 (RFC 4443)
* Neighbor discovery for IPv6 (RFC2461)
* Path MTU discovery for IP6 (RFC1981)
* Address configuration - either SLAAC (RFC2462) or DHCPv6 (RFC3315)
* IPv6 addressing architecture (RFC4291)
* Scoped address architecture (RFC4007)
* Unique local IPv6 unicast addresses (RFC4193)
* Multicast listener discovery (MLD) for IPv6 (RFC2710)
Can anyone shed light on the above (any V's or X's will help, as will don't
even think of it)? Is anyone running 2.4 in IPv6 environments? How
mature/up-to-date is the support?
If IPv6 is a requirement, does it absolutely mandate moving to
2.6.{latest,recent-enough} or will 2.4 be possible? Is there any version of
2.4 from which IPv6 support is markedly better than in earlier ones?
I have seen http://www.tldp.org/HOWTO/Linux+IPv6-HOWTO/ and
http://www.deepspace6.net/docs/best_ipv6_support.html, among others. It is
not clear to me how updated the info related to old kernels is. If anyone
knows that I can trust these documents that will be great.
Let's not go into the question why 2.4 is important, OK?
Thanks a lot in advance,
--
Oleg Goldshmidt | [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: IPv6 support in old (2.4) kernels
Hello,
I don't have direct answer to your questions as I don't know much about
the 2.4 kernel code.
However, I just want to give these four comments about IPV6 in 2.4 Linux kernel:
1) Also in the latest 2.6 kernels you cannot rmmod ipv6 when it is
configured as a module, although work is done to enable this in the future.
So you can ignore or not treat too seriously this
unloading bug when IPv6 is configured as a module note about the 2.4
kernel you mentioned
2) As far as I know, most test suites for IPV6 are not compatible with
the 2.4 kernels (for example, TAHI Conformance Test:
http://www.tahi.org/conformance/).
I did not check this point though.
3) The 2.4 kernel does not have any IPSec support (not for IPV6 ; I think that
also not for IPV4).
IPSec is mandatory according to the latest IPV6 spec and is
optional in IPV4 spec (although most operating systems, including Linux and
OpenSolaris had implemented it also as part of IPV4).
See, for example, in RFC 2401, (Security Architecture for the Internet
Protocol),
although it is **not** mentioned in your list.
from rfc 2401:
...
...
10. Conformance Requirements
All IPv4 systems that claim to implement IPsec MUST comply with all
requirements of the Security Architecture document. All IPv6 systems
MUST comply with all requirements of the Security Architecture
document..
..
..
4) The only tunnel you have in 2.4 IPV6 is sit (IPv6 in IPv4 tunnel).
At least in the
link for 2.4 code you mention in your mail. (maybe it does exist in
newer versions of 2.4)
You don't have the ip6_tunnel (ip6tnl) which is IPV4 or IPV6 in IPV6.
(Though this is relatively easy to implement, based on the 2.6 implementation)
Regards,
Rami Rosen
On Sun, Mar 16, 2008 at 4:38 PM, Oleg Goldshmidt [EMAIL PROTECTED] wrote:
Hi folks,
I suppose there are some people here who for one reason or another still run
kernel 2.4 on servers or embedded systems and are willing to admit to it.
Can anyone tell me how well kernels circa 2.4.19 (or later) support the
*current* IPv6 specification? I see IPv6 support in the code
(http://lxr.linux.no/linux-old+v2.4.19/net/ipv6/) which is marked as
EXPERIMENTAL and carries warnings in the configuration, but it looks like
this is associated with some unloading bug when IPv6 is configured as a
module (http://lxr.linux.no/linux-old+v2.4.19/net/Config.in). The same
warning appears in 2.4.31 which is the last 2.4 in LXR at least.
Suppose I will be happy to compile IPv6 into the kernel (not as a module) -
I am still interested in how well 2.4 supports today's IPv6. More
specifically, after some research I have come to a conclusion that I am
interested in assessment of 2.4 support for (at least) the following:
* IPv6 (RFC2460)
* ICMPv6 (RFC 4443)
* Neighbor discovery for IPv6 (RFC2461)
* Path MTU discovery for IP6 (RFC1981)
* Address configuration - either SLAAC (RFC2462) or DHCPv6 (RFC3315)
* IPv6 addressing architecture (RFC4291)
* Scoped address architecture (RFC4007)
* Unique local IPv6 unicast addresses (RFC4193)
* Multicast listener discovery (MLD) for IPv6 (RFC2710)
Can anyone shed light on the above (any V's or X's will help, as will don't
even think of it)? Is anyone running 2.4 in IPv6 environments? How
mature/up-to-date is the support?
If IPv6 is a requirement, does it absolutely mandate moving to
2.6.{latest,recent-enough} or will 2.4 be possible? Is there any version of
2.4 from which IPv6 support is markedly better than in earlier ones?
I have seen http://www.tldp.org/HOWTO/Linux+IPv6-HOWTO/ and
http://www.deepspace6.net/docs/best_ipv6_support.html, among others. It is
not clear to me how updated the info related to old kernels is. If anyone
knows that I can trust these documents that will be great.
Let's not go into the question why 2.4 is important, OK?
Thanks a lot in advance,
--
Oleg Goldshmidt | [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: Processing time spent in IRQ handling and what to do about it
Hi, You cannot turn it on/off. The driver may support this optional API or not. If it supports it, it's the driver sole decision when it's better to use polling/interrupt-per-packet according it its hardware specifics. I doubt whether this is exactly so for all NICS, as one might understand from your answer. For example, with e1000 NICs, you can select to build the driver with or without polling support. See, while configuring the kernel: Device Drivers-Network Device Support-Ethernet 1000 Mbit - Intel (R) PRO/1000 Gigabit Ethernet Support-Use RX polling (NAPI). selecting it sets the CONFIG_E1000_NAPI to y. By default, in newer kernels, for e1000, it comes with support for NAPI by default, but you can also build it without this support. And if you will look at the code of the driver, you will find in e1000_main.c module the following: #ifdef CONFIG_E1000_NAPI netdev-poll = e1000_clean; netdev-weight = 64; #endif Which means that , when building without CONFIG_E1000_NAPI set, you will not have the poll method and therefore no polling/NAPI. You have also the ability to choose NAPI for other nics; for example, Tulip; see Device Drivers-Network Device Support-Ethernet 10 or 100 Mbit - Tulip family network device support-Use NAPI RX polling. It could be that on other NICs you cannot turn it on/off. Broadcom was the first to release the tg3 driver with support for NAPI for Linux. So they have probably a lot of experience with it, and it could be that there NAPI support is built in and you cannot avoid it. BTW, with Open Solaris, this is exactly the situation: the NAPI support is in the core automatically; the driver start as interrupt driver, and changes to polling when there is a high load of interrupts.The drivers need not be built with any NAPI special support. The driver binary is the same when working with/without NAPI.There is a way, however, to configure kernel-wide NAPI parameters. Regards, Rami Rosen On Dec 18, 2007 10:14 PM, Oron Peled [EMAIL PROTECTED] wrote: On Tuesday, 18 בDecember 2007, Yedidyah Bar-David wrote: I am not an expert on this, but what you want might be NAPI - a new network driver infrastructure designed to solve just that. Google a bit - I do not know exactly when it entered 2.6 (and you did not state your kernel version) and which drivers use it already. 1. NAPI was new at kernel 2.3.x when it was developed towards 2.4 2. It gives the *driver* the option to toggle between interrupt driven and polling mode at runtime. E.g: - A GB ethernet at full speed may better poll the hardware every once in a while. - The same card is better off using interrupt driven mode if the trafic is low. 3. You cannot turn it on/off. The driver may support this optional API or not. If it supports it, it's the driver sole decision when it's better to use polling/interrupt-per-packet according it its hardware specifics. 4. I don't think a single fast ethernet card can severely affect your hardware interrupt load. So either: - You have a GB (or maybe 2GB?) ethernet with high load. - You have several fast-ethernet cards working at full speed. 5. A far better suspect would be the disk controller (e.g: working without DMA etc.) 6. Why guess? watch -n10 -d cat /proc/interrupts And calculate how many interrupts per-sec occured for various devices. That would give you a rough idea who are the possible suspects. -- Oron Peled Voice/Fax: +972-4-8228492 [EMAIL PROTECTED] http://www.actcom.co.il/~oron ICQ UIN: 16527398 Linux lasts longer! -- Kim J. Brand [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: RedHat Cluster Suite as a replacement for linux-ha?
Hello, I had made some tests with RedHat cluster on RedHat Enterprise and it was ok. I had used it in a two node configuration with gfs file system. My main goal was using a safe shared file system which can be accesses from the 2 nodes without fear of collisions which will cause a loss of data; using gfs file system with its DLM (Distributed Lock Manager) enables the file system to be shared safely. There is also gfs2, but I did not try it yet. (Apart from gfs , there is also OCFS2, the oracle ext3 based file system, which supports safe file sharing, and is part of the Linux kernel). BTW, you can use DRBD in conjunction with hearbeat. See, for example, http://www.linuxjournal.com/node/9074. DRBD was submiitted for the Linux kernel on July 2007, so there are chances that you will encounter it in the near future as part of your distro. I think DRBD concept (RAID over a network using block device) is a very good one and I wonder why Microsoft did not adopt it yet (AFAIK). Regards, Rami Rosen On Nov 30, 2007 11:31 AM, Amos Shapira [EMAIL PROTECTED] wrote: Hello, I'm beginning to give up on making Linux-HA's heartbeat work for my environment (CentOS x86_64) and am wondering what other option have I got to help me: 1. Use IPVS to maintain a cluster of servers serving virtual IP, either master/slave or load-balanced. 2. Use DRBD in master/slave fashion to keep a home-grown application highly-available. The first thing I stumbled upon is RedHat Cluster Suite (http://www.linuxtopia.org/online_books/centos_linux_guides/centos_cluster_configuration_and_management/ ), from which I also saw some packages on my CentOS servers. I've never heard of it before and am just starting to dig its docs, but if someone here can confirm/deny that this is a possible route to take it might save me some time or doubts. Thanks, --Amos = 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: Linux 2.6 Vs Solaris 10
Hi, I was involved in porting code from Linux to Open Solaris. I can say talk about two points, from my personal experience: First, it seems to me that some companies will prefer to develop code with SUN CDDL license, since this way they can keep the code only for themselves and SUN. On the other hand, this is not so good for the Open Source community; and this is in contradiction to the Linux spirit. Second, though Open Solaris is aimed to be an open source project, like Linux, it seems to me that still there are many differences when comparing the Linux open source model and SUN open source model; for example, most of the code in Open Solaris is written by Open Solaris employees, as far as I understand. If you are not employed by SUN, you cannot just send a patch to a mailing list (as you can do with linux LKML). First you have to contact SUN project manager and negotiate a bit with him, etc. Regards, Rami Rosen On Nov 9, 2007 3:41 PM, Nadav Har'El [EMAIL PROTECTED] wrote: On Fri, Nov 09, 2007, Ohad Levy wrote about Linux 2.6 Vs Solaris 10: I've heard quite a lot from many EDA vendors in the last few months that they are much happier with Solaris 10 platforms on X86_64 than Linux as the system is more stable on a high load. I was aware in the past that 2.4 kernel scheduler was not as optimized as the 2.6 is today . .. I can't answer your specific question (on whether Solaris is better than Linux or not), but I'd just like to point out that performance, and load, is a multi-dimensional variable, and its quite possible that Linux is better at some types of load, and Solaris is better at others. In some types of applications, you have many processes running concurrently, and the quality of the scheduler is indeed important. In others, there may be a lot of threads so the lightness of these threads is a problem (there is a big difference between how Solaris and Linux - especially the 2.4 kernel - do threads). Other servers (the really efficient ones ;-)) have just one or few processes, but can handle thousands of clients using a poll loop - in such cases an efficient poll alternative (such as the new epoll in Linux) is very important. For some servers reading from the disk is very important, in which case an efficient file-system and handling of io (prefetching, caching, request reordering, etc.) is paramount. Some servers deal with a huge amount of network traffic, in which case the efficiency of the network stack and the way the kernel passes the data to the application (zero copy being the holy grail here). On SMP machines (and it appears that today, you can't even buy a modern machine which isn't an SMP!), it is important for the kernel to utilize all processors in everything it does (for example, if a certain kernel runs the network stack just on one CPU, a machine with a huge packet load will only utilize one CPU) but on the other hand must do so efficiently (when the kernel itself is running on several CPUs, all sorts of locking inefficiencies can start). In any one of these parameters (and you can probably think of more), Solaris is probably different from Linux (and the Linux 2.6 kernel is probably different from 2.4), and quite possibly one is better in certain parameters and the other is better in other parameters. If you know exactly what *your* application needs (e.g., many concurrent processes, modest bandwidth, small data set so everything from disk should be cached in memory), then you can compare the two OSes. Of course, it is *possible* that one of these OSes is better in every parameter, but (again, without checking) I doubt it. I especially doubt that Linux - which is constantly being developed by dozens of Non-Idiots - is the one that will turn out worse on every parameter. -- Nadav Har'El| Friday, Nov 9 2007, 28 Heshvan 5768 [EMAIL PROTECTED] |- Phone +972-523-790466, ICQ 13349191 |If you notice this notice, you'll notice http://nadav.harel.org.il |it's not worth noticing but is noticable. = 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: finding core size limit of running process?
Hi, I suppose the existance of this patch means that this isn't possible in the standard kernel. You are right; it is not in the kernel , also not in the git unstable tree . In fact , as you can see in that link, it is from January 2005; So I suppose it will not be integrated into the kernel in the near future; (don't know why) Regards, Rami Rosen On Nov 7, 2007 1:57 PM, Amos Shapira [EMAIL PROTECTED] wrote: On 07/11/2007, Rami Rosen [EMAIL PROTECTED] wrote: Amos, I do not know if this is possible to probe for an RLIMIT resoruces of a process; however, there is a patch of Bill Rugolsky which enables it; see:http://lwn.net/Articles/119703/ this patch is against a single file: /proc/fs/base.c Thanks. I suppose the existance of this patch means that this isn't possible in the standard kernel. Thanks for the reference, I'll keep it around (or actually - know to get to the mailing list archives when I need it :). Cheers, --Amos = 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: NETDEV WATCHDOG transmit timeout?
Rony, This sounds toe me like a driver bug . The sky2 drivers had many problems and many patches, by Stephen Hemminger and others. I would suggest to check if this problem occurs with 2.6.20 or higher kernel (on this machine, or, if not possible, and just to verify that this indeed is the problem , on another machine with both 2.6.20 (or higher) and the kernel version you have, and that nic on it. Regards, Rami Rosen On Oct 28, 2007 1:49 PM, ronys [EMAIL PROTECTED] wrote: Hi, Under heavy traffic, I'm getting the above syslog message every day or two. Unfortunately, the failure is far from graceful: ifconfig up/down doesn't help, and the system is occasionally hung so that it's unreachable via other interfaces, forcing a reboot. Some more details: Dual core 3.4GHz Intel CPU # cat /proc/version Linux version 2.6.18.5abc ([EMAIL PROTECTED]) (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) #3 SMP Fri Sep 7 21:32:00 IDT 2007 # lspci -v [...] 01:00.0 Ethernet controller: Marvell Technology Group Ltd. 88E8053 PCI-E Gigabit Ethernet Controller (rev 19) Subsystem: SysKonnect Unknown device 4340 Flags: bus master, fast devsel, latency 0, IRQ 74 Memory at d002 (64-bit, non-prefetchable) [size=16K] I/O ports at 9000 [size=256] [virtual] Expansion ROM at 6000 [disabled] [size=128K] Capabilities: [48] Power Management version 2 Capabilities: [50] Vital Product Data Capabilities: [5c] Message Signalled Interrupts: Mask- 64bit+ Queue=0/1 Enable+ Capabilities: [e0] Express Legacy Endpoint IRQ 0 Capabilities: [100] Advanced Error Reporting [...] # cat /var/log/messages [...] Oct 26 09:39:45 abc kernel: NETDEV WATCHDOG: eth3: transmit timed out Oct 26 09:39:45 abc kernel: sky2 hardware hung? flushing Oct 26 09:48:35 abc kernel: NETDEV WATCHDOG: eth3: transmit timed out Oct 26 09:48:35 abc kernel: sky2 status report lost? [...] A quick google didn't find anything blindingly relevant. Any ideas, suggestions or hints humbly accepted. Thanks, Rony 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: Help with Linux on sparc64
Hello, I've installed at least 3 linux distros on my 64-bit ultra5 years before Ubuntu was created Regarding ultra5 - you are right. The Ubuntu distribution was to first to run on a Niagra SPARC64 machines. I should have mention this. Regards, Rami Rosen On 7/31/07, Mike Tewner [EMAIL PROTECTED] wrote: On 7/31/07, Rami Rosen [EMAIL PROTECTED] wrote: Hello Nadav, What binary distribution has the best support for Sun machines? The first distribution which was equipped with support for Sun SPARC64 is Ubuntu; this happened after some attempts to boot Fedora Core (by Dave Miller and others) failed giving panic. This of course does not imply the Ubuntu is the best one , but it is the most veteran in this field. So I suggest you try it (you can try Ubuntu live-cd first to see what are the results). That's a very dangerous statement - I've installed at least 3 linux distros on my 64-bit ultra5 years before Ubuntu was created. Granted, one of them was debian :-) but after boot the keyboard stops working (It only has USB keyboardmouse). to make sure this keyboard is identified please try : tail -f /var/log/messages, and then unplug the USB keyboard and plug it back in. Do you see any messages ? Regards, Rami Rosen On 7/26/07, Nadav Shemer [EMAIL PROTECTED] wrote: Hi. I have a Sun Ultra 45 and I've tried (unsuccessfully) to install some kind of Linux distribution on it. I've tried Debian netinst (both stable and testing) and Gentoo (stage2) CDs, but after boot the keyboard stops working (It only has USB keyboardmouse). Booting from the serial console, It hangs with the following message: su: Cannot register IRQ 1 Has anyone had any luck with this particular machine? What binary distribution has the best support for Sun machines? Can anyone help me? Nadav Shemer Tehuti Networks [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] = 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: Help with Linux on sparc64
Hello Nadav, What binary distribution has the best support for Sun machines? The first distribution which was equipped with support for Sun SPARC64 is Ubuntu; this happened after some attempts to boot Fedora Core (by Dave Miller and others) failed giving panic. This of course does not imply the Ubuntu is the best one , but it is the most veteran in this field. So I suggest you try it (you can try Ubuntu live-cd first to see what are the results). but after boot the keyboard stops working (It only has USB keyboardmouse). to make sure this keyboard is identified please try : tail -f /var/log/messages, and then unplug the USB keyboard and plug it back in. Do you see any messages ? Regards, Rami Rosen On 7/26/07, Nadav Shemer [EMAIL PROTECTED] wrote: Hi. I have a Sun Ultra 45 and I've tried (unsuccessfully) to install some kind of Linux distribution on it. I've tried Debian netinst (both stable and testing) and Gentoo (stage2) CDs, but after boot the keyboard stops working (It only has USB keyboardmouse). Booting from the serial console, It hangs with the following message: su: Cannot register IRQ 1 Has anyone had any luck with this particular machine? What binary distribution has the best support for Sun machines? Can anyone help me? Nadav Shemer Tehuti Networks [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: Xen paravirtualized domU doesn't start
Oded ,
I have Fedora Core 6 on an i386 mahine ; I had used a Virtual
Machine Monitor to create and start a domU. (which is para virtualized
and FC6 based). This Virtual Machine Monitor is part of the Fedora
Core 6 installation.
(Applications-Systems Tools-Virtual Machine Monitor).
It works without any problems.
Installation is quite simple, and you can find a step by step description here:
http://wiki.xensource.com/xenwiki/XenIntro#head-b92d8b84fba2c72099710afd1027e8eafdad0f98
Regards,
Rami Rosen
On 11/29/06, Oded Arbel [EMAIL PROTECTED] wrote:
--=-3LD+5Pxn+/qCKd/a9nM3
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Hi list - a Xen question for the Xen masters out there, if you please.
I have a Fedora Core 6 with Xen 3.0.3 installed which is running a Cent
OS 4.4 on a fully virtualized domU (all this on an EM64T dual-cpu). The
cpuinfo claims I have VT support, but I'm not very happy with the
performance I'm getting (the VM stutters a lot and is generally very
slow and painful).
So I decided to try to paravirtualize it. I downloaded a RHEL 4.4 kernel
RPM from XenSource web site, and installed in on both the disk image the
vm is running from, and in my host. I then modified the vm config file
to look like this:
snip
name = sub1
## builder = hvm
memory = 512
disk = [ 'file:/var/xen/sub1/hda,hda,w' ]
vif = [ 'mac=00:16:3e:35:fe:9c, bridge=xenbr0', ]
uuid = 7fb7a5f3-199c-df57-3556-ba823c98b372
## device_model = /usr/lib64/xen/bin/qemu-dm
## kernel = /usr/lib/xen/boot/hvmloader
vnc=1
vncunused=1
#pae=1
#bootloader=/usr/bin/pygrub
initrd = /boot/xeninitrd
kernel = /boot/xenkernel
root = /dev/hda1 ro
vcpus=1
## serial = pty # enable serial console
on_reboot = 'restart'
on_crash= 'restart'
snap
(commented out lines is the values used for the fully-virtualized
setup).
Now when I try to start the VM, I get:
# xm create -c sub1
Using config file /etc/xen/sub1.
Error: (22, 'Invalid argument')
and this log in the xend.log:
snip
[2006-11-29 12:39:43 xend.XendDomainInfo 2655] DEBUG
(XendDomainInfo:190) XendDomainInfo.create(['vm', ['name', 'sub1'],
['memory', '512'], ['on_reboot', 'restart'], ['on_crash', 'restart'],
['vcpus', 1], ['uuid', '7fb7a5f3-199c-df57-3556-ba823c98b372'],
['image', ['linux', ['kernel', '/boot/xenkernel'], ['root', '/dev/hda1
ro'], ['vnc', 1], ['vncunused', 1], ['display', 'localhost:10.0'],
['xauthority', '/root/.Xauthority']]], ['device', ['vbd', ['uname',
'file:/var/xen/sub1/hda'], ['dev', 'hda'], ['mode', 'w']]], ['device',
['vif', ['bridge', 'xenbr0'], ['mac', '00:16:3e:35:fe:9c')
[2006-11-29 12:39:43 xend.XendDomainInfo 2655] DEBUG
(XendDomainInfo:296) parseConfig: config is ['vm', ['name', 'sub1'],
['memory', '512'], ['on_reboot', 'restart'], ['on_crash', 'restart'],
['vcpus', 1], ['uuid', '7fb7a5f3-199c-df57-3556-ba823c98b372'],
['image', ['linux', ['kernel', '/boot/xenkernel'], ['root', '/dev/hda1
ro'], ['vnc', 1], ['vncunused', 1], ['display', 'localhost:10.0'],
['xauthority', '/root/.Xauthority']]], ['device', ['vbd', ['uname',
'file:/var/xen/sub1/hda'], ['dev', 'hda'], ['mode', 'w']]], ['device',
['vif', ['bridge', 'xenbr0'], ['mac', '00:16:3e:35:fe:9c'
[2006-11-29 12:39:43 xend.XendDomainInfo 2655] DEBUG
(XendDomainInfo:395) parseConfig: result is {'shadow_memory': None,
'uuid': '7fb7a5f3-199c-df57-3556-ba823c98b372', 'on_crash': 'restart',
'on_reboot': 'restart', 'localtime': None, 'image': ['linux', ['kernel',
'/boot/xenkernel'], ['root', '/dev/hda1 ro'], ['vnc', 1], ['vncunused',
1], ['display', 'localhost:10.0'], ['xauthority', '/root/.Xauthority']],
'on_poweroff': None, 'bootloader_args': None, 'cpus': None, 'name':
'sub1', 'backend': [], 'vcpus': 1, 'cpu_weight': None, 'features': None,
'vcpu_avail': None, 'memory': 512, 'device': [('vbd', ['vbd', ['uname',
'file:/var/xen/sub1/hda'], ['dev', 'hda'], ['mode', 'w']]), ('vif',
['vif', ['bridge', 'xenbr0'], ['mac', '00:16:3e:35:fe:9c']])],
'bootloader': None, 'cpu': None, 'maxmem': None}
[2006-11-29 12:39:43 xend.XendDomainInfo 2655] DEBUG
(XendDomainInfo:1253) XendDomainInfo.construct: None
[2006-11-29 12:39:43 xend.XendDomainInfo 2655] DEBUG
(XendDomainInfo:1285) XendDomainInfo.initDomain: 4 1.0
[2006-11-29 12:39:43 xend 2655] INFO (image:214) configuring linux guest
[2006-11-29 12:39:43 xend 2655] INFO (image:232) setting use_graphics
[2006-11-29 12:39:43 xend 2655] DEBUG (balloon:127) Balloon: 524780 KiB
free; need 524288; done.
[2006-11-29 12:39:43 xend 2655] INFO (image:138) buildDomain os=linux
dom=4 vcpus=1
[2006-11-29 12:39:43 xend 2655] DEBUG (image:193) dom= 4
[2006-11-29 12:39:43 xend 2655] DEBUG (image:194) image
= /boot/xenkernel
[2006-11-29 12:39:43 xend 2655] DEBUG (image:195) store_evtchn = 1
[2006-11-29 12:39:43 xend 2655] DEBUG (image:196) console_evtchn = 2
[2006-11-29 12:39:43 xend 2655] DEBUG (image:197) cmdline=
root=/dev/hda1 ro
[2006-11-29 12:39:43 xend 2655] DEBUG (image:198) ramdisk=
[2006-11-29 12:39:43 xend 2655] DEBUG (image:199
Re: memory leak in Solaris process
Yahav, I have to admit my sin in doing some kernel programming and user space development recently in open solaris . In fact, some of it is porting from linux to solaris which even may sound worth in this mailing list... Solaris is known for it's dynamic tracing functionality, also known as DTrace. You can also use DTrace with java prcoesses on solaris. Please take a look at https://solaris10-dtrace-vm-agents.dev.java.net You should act according to the java version you have and the instructions there. I had Java HotSpot(TM) Server VM (build 1.5.0_07-b03, mixed mode). So I had to download dvm.zip and install it and configure it according to the instructions there. Running dtrace on a java process is simple: After installing dvm, run the java process like this: java -Xrundvmti:all SendCmdLoop then find out your processId run: dtrace -n'dvmProcessId:::' -p ProcessId thats it: you get many messages, like track_allocation:object-alloc, _method_exit:method-return and more. you can also activate DTrace with less probes (instead of -Xrundvmti:all). You may want also to look at the DTrace manual , which is quite long. In case you have a Mustang ( java 6/jdk 1.6 prerelease) the provider is different is a bit different, see in that web site. There is a syscall and signales tracker utility which is built in in solaris called truss. It may aslo help you. Activation is simple: truss -p processId. Hope this helps. Regards, Rami Rosen On 10/11/06, Yahav Biran [EMAIL PROTECTED] wrote: Hi, I'm running an instance of Weblogic on Solaris 8. when I monitoring the process resource usage using prstat I can see that it after a week the RSS size is growing until the service is really bad and then I restart the application. The application is a web application that connected to oracle DB using OCI client. When I view the process map (pmap PID) I can see that the size of the (UNIX process)heap is growing. BUT when I monitoring the memory usage in the JVM I can see that there is no memory leak as no OutOfMemoryError is arising from the GC. i would like to have a way to monitor the usage of the UNIX process heap. I any body knows on such utility? Thanks yahav = 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: netif_rx and dev stack
Hi, Is it possible to post the code you wrote to this list? Rami On 10/7/06, Chava Leviatan [EMAIL PROTECTED] wrote: 2) just for the sake of deubuggin : can you allocate one skb only in that module, comment the skb_free() call, perform insmod and afterwards rmmod, and see if it crashes ? Did that already ... Further more I have allocated 100 skb, did not free them and the rmmod worked well - no crashes. And I have also allocated the skb, and free it from the thread itself - that also worked fine - Original Message - From: Rami Rosen [EMAIL PROTECTED] To: Chava Leviatan [EMAIL PROTECTED] Cc: linux-il@linux.org.il Sent: Saturday, October 07, 2006 7:46 AM Subject: Re: netif_rx and dev stack Chava , , when i do the rmmod the machine crashes , i.e., performs a reset !! 1) - when there is such reset , is there antyhing written in the kernel log ? can you please check and post it ? 2) just for the sake of deubuggin : can you allocate one skb only in that module, comment the skb_free() call, perform insmod and afterwards rmmod, and see if it crashes ? Regards, Rami Rosen On 10/5/06, Chava Leviatan [EMAIL PROTECTED] wrote: Hello, I have a 2.4.18 machine with 2 ethernet interfaces. I am trying to simulate a scenario where packet has just arrived from one interface and going to the other. I have build a kernel module that runs a kernel thread. That kernel thread is invoked every second, and then it does the following ; -alloc_skb -Fill the proper sk_buff fields -Fill IP header fields -Fill ICMP header fields -skb_put with the data length -call to netif_rx I am able to see the packets with ethreal (on the correct interface). I have also put some debug prints into the kernel code (dev.c) and I am able to see those packets getting into __kfree_skb. However, when i do the rmmod the machine crashes , i.e., performs a reset !! When I remove the netif_rx the rmmod returns ok. Can anyone please explain why does it crash ?? Any help is apreciated, Thanks, Chava = 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: netif_rx and dev stack
Chava , , when i do the rmmod the machine crashes , i.e., performs a reset !! 1) - when there is such reset , is there antyhing written in the kernel log ? can you please check and post it ? 2) just for the sake of deubuggin : can you allocate one skb only in that module, comment the skb_free() call, perform insmod and afterwards rmmod, and see if it crashes ? Regards, Rami Rosen On 10/5/06, Chava Leviatan [EMAIL PROTECTED] wrote: Hello, I have a 2.4.18 machine with 2 ethernet interfaces. I am trying to simulate a scenario where packet has just arrived from one interface and going to the other. I have build a kernel module that runs a kernel thread. That kernel thread is invoked every second, and then it does the following ; -alloc_skb -Fill the proper sk_buff fields -Fill IP header fields -Fill ICMP header fields -skb_put with the data length -call to netif_rx I am able to see the packets with ethreal (on the correct interface). I have also put some debug prints into the kernel code (dev.c) and I am able to see those packets getting into __kfree_skb. However, when i do the rmmod the machine crashes , i.e., performs a reset !! When I remove the netif_rx the rmmod returns ok. Can anyone please explain why does it crash ?? Any help is apreciated, Thanks, Chava = 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: creating and mounting ufs Creating and mounting ufs filesystem on a loopback device on linux - a problem?
Hi, Well following this I had launched a trial on x86_64 running fedora core 4 kernel. THere were problems with the mkufs : /mkufs -O=1 /test/disk-imag gives: =2: bad file system format value and also /mkufs -O=2 /test/disk-imag gives: =2: bad file system format value So maybe there is a problem with x86_64 and ufs / mkufs. Regards, Rami Rosen On 9/26/06, Muli Ben-Yehuda [EMAIL PROTECTED] wrote: On Tue, Sep 26, 2006 at 09:02:44PM +0300, Rami Rosen wrote: Hi Muli, I doubt it (did you try it?). Now I tried it, on Fedora Core 4 (on x86). It works like a charm! feel free to try it ... I did, on Suse 9 SP2 (x86-64) with a 2.6.18 kernel with both -O1 and -O2 and neither worked. Maybe it's x86-64 vs. i386. /me shrugs. Cheers, Muli = 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: creating and mounting ufs Creating and mounting ufs filesystem on a loopback device on linux - a problem?
Oops... I rechecked: on that x86_64 : instead /mkufs -O=1 /test/disk-imag it should have been /mkufs -O 1 /test/disk-imag and now mkufs and mount works ok on that x86_64 machine. Regards, Rami Rosen On 9/27/06, Rami Rosen [EMAIL PROTECTED] wrote: Hi, Well following this I had launched a trial on x86_64 running fedora core 4 kernel. THere were problems with the mkufs : /mkufs -O=1 /test/disk-imag gives: =2: bad file system format value and also ./mkufs -O=2 /test/disk-imag gives: =2: bad file system format value So maybe there is a problem with x86_64 and ufs / mkufs. Regards, Rami Rosen On 9/26/06, Muli Ben-Yehuda [EMAIL PROTECTED] wrote: On Tue, Sep 26, 2006 at 09:02:44PM +0300, Rami Rosen wrote: Hi Muli, I doubt it (did you try it?). Now I tried it, on Fedora Core 4 (on x86). It works like a charm! feel free to try it ... I did, on Suse 9 SP2 (x86-64) with a 2.6.18 kernel with both -O1 and -O2 and neither worked. Maybe it's x86-64 vs. i386. /me shrugs. Cheers, Muli = 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: creating and mounting ufs Creating and mounting ufs filesystem on a loopback device on linux - a problem?
Hi, Try ./mkufs -O 1 /test/disk-image instead of /mkufs /test/disk-image there are 2 file system formats for mkufs : 1 = UFS1, 2 = UFS2 see also : mkufs -help. This should work. Regards, Rami Rosen On 9/26/06, Dan Shimshoni [EMAIL PROTECTED] wrote: Hello linux-il, First, running this series of actions (as many know) creates and mounts an ext3 filesystem on my linux machine: dd if=/dev/zero of=/test/disk-image count=10240 mkfs -t ext3 /test/disk-image mount -o loop -t ext3 /test/disk-image /loopDir ls /loopDir = gives lost+found Now I tried the same with ufs filesystem (which is common in *BSD and solaris os's ; in linux the ufs support is read-only): First, I had built the ufs module and insmoded it. Running cat /proc/filesystems |grep ufs shows that it exists. Second, I had built the mkufs, which is part of ufs-linux. see http://ufs-linux.sourceforge.net/, (I had to apply a patch of Evgeniy Dushistov after some comaplint of Andrew Morton regarding ufs support in linux) see : http://lkml.org/lkml/2006/5/20/48 Than I had tried the following series of actions: dd if=/dev/zero of=/test/disk-image count=10240 = 10240+0 records in = 10240+0 records out cd /work/src/ufs- tools-0.1 ./mkufs /test/disk-image = /test/disk-image: 5.0MB (10240 sectors) block size 16384, fragment size 2048 using 4 cylinder groups of 1.27MB, 81 blks, 192 inodes. = super-block backups (for fsck -b #) at: 160, 2752, 5344, 7936 mount -o loop -t ufs /test/disk-image /loopDir I get: mount: wrong fs type, bad option, bad superblock on /dev/loop1, or too many mounted file systems (could this be the IDE device where you in fact use ide-scsi so that sr0 or sda or so is needed?) why ? Did anyone succeeded with creating and mounting ufs on a loopback device on linux ? Dan = 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: creating and mounting ufs Creating and mounting ufs filesystem on a loopback device on linux - a problem?
Hi Muli, I doubt it (did you try it?). Now I tried it, on Fedora Core 4 (on x86). It works like a charm! feel free to try it ... Aftrer applying the patch and modprbe ufs I ran that series of commands, but mkfus like this : /mkufs -O 1 /test/disk-image Rhere was no error in mount , and running now mount | grep ufs shows it is OK: /test/disk-image on /loopDir type ufs (ro,loop=/dev/loop0) BTW, I tried also mount -r -o loop -t ufs /test/disk-image /loopDir and it did perform mount OK, only the kernel log says Sep 26 20:43:29 localhost kernel: You didn't specify the type of your ufs filesystem Sep 26 20:43:29 localhost kernel: Sep 26 20:43:29 localhost kernel: mount -t ufs -o ufstype=sun|sunx86|44bsd|ufs2|5xbsd|old|hp|nextstep|netxstep-cd|openstep .. Sep 26 20:43:29 localhost kernel: Sep 26 20:43:29 localhost kernel: WARNING Wrong ufstype may corrupt your filesystem, default is ufstype=old When trying the same but with /mkufs -O 2 /test/disk-image I get that error in mount: mount: wrong fs type, bad option, bad superblock on /dev/loop0, missing codepage or other error In some cases useful info is found in syslog - try dmesg | tail or so and in the kernel log : kernel: ufs_read_super: bad magic number As for the reason why UFS1 works and UFS2 not, I will try to hack the code a bit and post my thoughts tomorrow or in two days. I think we should remeber that ufs support in the linux kenel is a bit problematic. We can mount ufs only as a read only filesystem in linux ; this was once the case with NTFS , now it is a bit better , but also when you mount an ntfs from linux , it might not succeed. In their own words: Note: That doesn't mean it always succeeds, it is still experimental and might just as well refuse to complete an operation in order to prevent corruption. see : http://www.linux-ntfs.org/ And looking from the opposite dircetion (namely , an operating system which comes with ufs like solaris/*BSD) : There is a solaris package to mount ext3 and ntfs from solaris ufs; see: http://www.genunix.org/distributions/belenix_site/ It mounts EXT2FS and NTFS partitions, but as read-only partitions. Regards, Rami Rosen On 9/26/06, Muli Ben-Yehuda [EMAIL PROTECTED] wrote: On Tue, Sep 26, 2006 at 04:32:08PM +0300, Rami Rosen wrote: Hi, Try ./mkufs -O 1 /test/disk-image instead of /mkufs /test/disk-image there are 2 file system formats for mkufs : 1 = UFS1, 2 = UFS2 see also : mkufs -help. This should work. I doubt it (did you try it?). It's obvious from the kernel log that fs/ufs gets super block magic number 0, which is wrong any way you look at it. Cheers, Muli = 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: automatic module loading with udev or HAL?
Hi, One little remark: The use of hotplug is now deprecated from kernel version 2.6.15 on. See for example , http://lwn.net/Articles/166954 or udev README in http://www.kernel.org/git/?p=linux/hotplug/udev.git;a=blob;hb=HEAD;f=README So if you have a kernel higher than 2.6.15 you should use the udev.rules located in /etc/udev/rules.d. I tried it a USB disk on key and it worked for me. The NAME parameter is the name that will be used for the device node see also : http://www.reactivated.net/writing_udev_rules.html and take in account that it is not updated to the last version of udev. (make sure that the hotplug is disbaled by looking at: /proc/sys/kernel/hotplug ) Regards, Rami Rosen. On 8/1/06, Ilya Konstantinov [EMAIL PROTECTED] wrote: hotplug (as found in /sbin/hotplug and /etc/hotplug/) is what's responsible on Linux 2.6 for automatic loading of modules for permanent (e.g. PCI) as well as removable (USB) hardware. The modern fashion is _not_ to load kernel modules just the moment when its their device file are used, but rather when the hardware is inserted. This is just like the Windows 2000's model of PnP: it simplifies things (i.e. does a stat() trigger the autoloading? And how about access()?) and the growth in memory usage is not significant. For you, it means that you should configure 'hotplug' to support your device and load its module as soon as its inserted. In /etc/hotplug/, you have usermap files which map hardware IDs to modules. Most modules specify what hardware they support (you can see what with 'modinfo') but your hardware is supported by a generic module which's not tied to any specific hardware ID, so you have to tie it yourself (and submit a patch upstream so next version 'hotplug' would come preconfigured to load usbserial for your device). Oded Arbel wrote: Problem is, I can't get usbserial to auto load (by HAL?), so I have to modprobe it by hand. Some instruction pages which I utilized to get to where I am now (the card working properly) suggest adding = 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: rtp
Hello, I had some experience with the Robust Audio Tool (RAT) libraries, which are based on RTP ; see: http://www-mice.cs.ucl.ac.uk/multimedia/software/rat/index.html This RAT project was born before VLC , and it works also with video. see : http://www-mice.cs.ucl.ac.uk/multimedia/software/vic/ The RTP there works fine for video. However, VLC , which is basically a streaming solution with some additions , seems to be more suitable and more organized than the vic project. But it seems that you will have to dig a lot in the code of VLC; I had looked a bit inside and it is quite heavy. Regards, Rami Rosen On 7/9/06, Yossy Dreyfus [EMAIL PROTECTED] wrote: I need a library in C for sending video buffers via network to a player. Is it exists? = 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: wireless card 2200GB doesn't start.
Hi, You said: 2. In the INSTALL document of ipw2200 (taken from the generic package) is an explenation about sysfs which should be mounted. It seems to me it is mounted by default even though it is not in mtab (can't see it in df unless you are in /sys and do df .) I assume there is no problem with sysfs. I also don't see it in df. simply run mount |grep sysfs. It should be there. Regards, Rami Rosen On 5/29/06, David Harel [EMAIL PROTECTED] wrote: Thanks for all of you who replied. Unfortunately none of the recommendations worked but if you don't mind I would like to get more of your advice. 1. How can I track which file is attempted to be loaded to the firmware when modprobe is performed. Obviously, strace modprobe ipw2200 does not do the job. 2. In the INSTALL document of ipw2200 (taken from the generic package) is an explenation about sysfs which should be mounted. It seems to me it is mounted by default even though it is not in mtab (can't see it in df unless you are in /sys and do df .) Is this case good enough for loading the firmware? David Harel wrote: Hi all, I did as in: http://gentoo-wiki.com/HARDWARE_Gentoo_on_Fujitsu-Siemens_S7020#Onboard_Wireless and when I do: modprobe ipw2200 I get the following message in dmesg: ipw2200: ipw2200-bss.fw request_firmware failed: Reason -2 ipw2200: Unable to load firmware: -2 ipw2200: failed to register network device ACPI: PCI interrupt for device :06:05.0 disabled ipw2200: probe of :06:05.0 failed with error -5 Any idea? -- Thanks. David Harel, == Home office +972 77 4422234 Fax:+972 77 4422234 Cellular: +972 54 4534502 Snail Mail: Amuka D.N Merom Hagalil 13802 Israel 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] = 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: wireless card 2200GB doesn't start.
Hello, It could be a wrong firmware problem OR wrong path, as was suggested The error request_firmware failed: Reason... stems from the call to request_firmware(), which is the generic kernel method for loading firmware. ( If you want more details, look at the kernel module drivers/base/firmware_class.c). I have a ZyDas 1211 USB Wireless LAN card. The original driver (http://zd1211.ath.cx/) has the firmware inside and does NOT call request_firmware(). But there is also a rewirte driver (a really interesting project in itself: http://zd1211.ath.cx/wiki/DriverRewrite ) which DOES use this same generic request_firmware() method. And of course firmware should be installed so it can work. This rewrite driver works ok for me ; it still does not support all the features(for example, Ad-Hoc mode is not supported yet) and it must be run agains a kernel with softmac support, (I use 2.6.17-rc3) but loading firmware is OK. I made a test now and removed the firmware folder (/lib/firmware/zd1211). The error that I get when insmoding the driver is the SAME -2 error: zd1211 Couldn't load firmware. Error number -2 Still my feeling is that there are chances that the problem is wrong firmware and not wrong path/filename . To eliminate this problem I suggest: 1) look at the kernel log which version of the driver you are using. (when modprobing the driver, the ipw_init() method print the version. The latest one is 1.1.2 if I am not wrong. 2) Verify that you have , in /etc/hotplug/firmware.agent, FIRMWARE_DIR=/lib/firmware. I assume most chances it is probably so. 3) goto http://ipw2200.sourceforge.net/firmware.php look for the firmware which corresopond to the driver version you have; erase /lib/firmware/ipw* and reinstall the firmware files. 4) try modprobe again. 5) In case it does not succeed - how about downloading the lates driver sources + corrseponding firmware and building them? Please tell us the results. Hope this helps, Rami Rosen On 5/27/06, David Harel [EMAIL PROTECTED] wrote: Thanks for your reply.Seems to me the driver is installed correctly (I reinstalled and triedolder versions also). It resides in: /lib/firmware/ipw2200-bss.fw (olderversions have the names somewhat different). I couldn't figure out how modprobe loads the right file and I didn'tfind any trace of it in /etc/modules.conf and in /etc/modules.d/ipw2200so maybe those are not set correctly but I don't know how to set them. modprobe ipw2200 takes some time before it gives me the error as if itis trying to load the firmware. I also found somewhere that it might bean issue of timeout so I increased the time but that didn't help either. Ilya Konstantinov wrote: Just a wild guess here, but Linux error number -2 is ENOENT (No such file or directory). I'd see it as a strong hint the driver wants to load the firmware off a disk file, perhaps supplied through a parameter (see 'modinfo ipw2200'), and fails to find it. David Harel wrote: Hi all, I did as in: http://gentoo-wiki.com/HARDWARE_Gentoo_on_Fujitsu-Siemens_S7020#Onboard_Wireless and when I do: modprobe ipw2200 I get the following message in dmesg: ipw2200: ipw2200-bss.fw request_firmware failed: Reason -2 ipw2200: Unable to load firmware: -2 ipw2200: failed to register network device ACPI: PCI interrupt for device :06: 05.0 disabled ipw2200: probe of :06:05.0 failed with error -5 Any idea?--Thanks.David Harel,== Home office +972 77 4422234Fax:+972 77 4422234Cellular: +972 54 4534502Snail Mail: AmukaD.N Merom Hagalil13802IsraelEmail: [EMAIL PROTECTED]=To unsubscribe, send mail to [EMAIL PROTECTED] withthe word unsubscribe in the message body, e.g., run the commandecho unsubscribe | mail [EMAIL PROTECTED]
Re: Most recommanded webcam for Linux?
Hi, I also have a positive experience with logitech 4000 Pro, both on Linux and Windows. Another feature which was not mentioned here of this webcam is it's high sensitivity; when lights are off you can get good video. Regarding streaming solutions: I once made tests with vlc , see: http://www.videolan.org and generally it was OK , but I cannot compare with other solutions as I did not try anything else. (It is open source and works both in linux and windows ). What I tried with VLC was http streaming (there are other types of streaming). Regarding the market in the last two years: I am following what happens and I agree that not much was changed in recent 2 years (or even more). This might change because now much higher upload bitrate is more and more common abroad (also here), so transmitting video in resolutions higher the CIF (CIF is 352x288) becomes more realistic Also the new H264 codec which is more and more entering the market, can contibute to better quality. - Though I looked for it, I did not encounte a full list of webcams for video4linux 1 or video4linux 2 (there were some partial lists). Regards, Rami Rosen On 10/25/05, Amos Shapira [EMAIL PROTECTED] wrote: On 10/25/05, Hetz Ben Hamo[EMAIL PROTECTED] wrote: It really depends which quickcam you got..Oops you are right. I forgot to mention this but I'm not near the camera to give its exact details - it's one of the cheapo cams which were availablealready three years ago. I think it's just called quickcam pro. I have Logitech Quickcam 4000 - and it's one of the best, and fully supported under Linux (although the face tracking feature is not supported).Is this the Pro or is there a non-Pro version? I've seen the Pro mentionedmany times in relation to Linux. Thanks for the recommandation. If you want more then 640x480 resolution with good frame rate, it will cost you - a lot more. Go search for USB 2.0 cams and you'll find that 90% of them are actually working with USB 1.1 bandwidth limitation andThat's what I found too - there are either USB 1.1 cameras or network webcamswhich provide AVI movies or auto-updated still frames when connected directlyover HTTP or somesuch, very expensive, apparently many of them come with motors to remotely control their position, but also with nice high-resolution(for instance, see http://tinyurl.com/9363c and pick CITY from the menu onthe left, notice the Luna-park under the bridge, then pick Luna Park from the menu). up to 640x480 and up to 30 FPS (I say up to, because most of them will give you 640x480 with 15FPS max before they start pulling some dirty tricks and heavy compression, which seriously degrates the video quality). The sad truth is that webcam manufacturers follow the software companies like Microsoft with MSN Messenger, Yahoo (Yahoo Messenger), and AOL (ICQ, AIM) - and none of them really gives you resolution selection with their Instant Messaging services so all of them are pretty much stuck with 320x240 (or 352x288) resolution. No demand - no new features. It's pretty much stuck at this situation for 2 years now if I recall correctly.Thanks - that seems to explain the (disappointing) state of affairs asI found it.Maybe the most telling example is that I found the same model of camera Ibought three years ago still prominantly displayed on the shelfs today. Still - can't I take advantage of the higher resolution available with some ofthe cameras? I mean - if you look at zap.co.il you can find cameras with640x480 resolution: http://www.zap.co.il/models.asp?Sog=C-WebcamDB66204=66205Another confusing matter is the state of Video 4 Linux - should I track V4l2or v4l (1)? Is there a comprehansive list of supported cameras (as opposed to supported cheaps)?And one last bit - does anyone here have experience in streaming webcam outputbetween Linux and Windows XP? Thanks, HetzThank you.--Amos To unsubscribe, send mail to [EMAIL PROTECTED] withthe word unsubscribe in the message body, e.g., run the commandecho unsubscribe | mail [EMAIL PROTECTED]
Re: Xen vs. VMware (was: Re: VMware Internals Expert)
Hi, Developing processors with Virtualization support is an attractive idea; in this way there will no need to patch Linux for using Xen and this will enable using the other OS. It will sure give the Xen project , which is evolving rapidly, a siginificant boost. However, adding a layer in a processor of course will add, apart from the market cost, also a performance cost. Xen as it is today acts at nearly to native performance on x86 (having about 3% overhead). Windows XP indeed runs on Xen under VT-enabled processors: see http://www.xensource.com/news/pr082305.html. I assume that the performance cost of Xen under such processors will be higher than 3% (and I hope it will be not much higher). However I did not see yet anywhere an official performance metrics of Xen on the new Intel VT-Enabled processors or AMD Pacifica processors. If anybody had seen such document I will be glad if he/she will share this info with us. (maybe AMD even have some advantage when using a tagged TLB,but this is really hard to tell). Regards, Rami Rosen On 10/10/05, Amos Shapira [EMAIL PROTECTED] wrote: On 10/10/05, Muli Ben-Yehuda [EMAIL PROTECTED] wrote: What Ira said. And while you're at it, Xen is much more of a VMWare replacement than qemu.[snipped] It's closed source; contact VMWare. Now if you were asking about, oh for example, Xen... that would be a completely different answer.Speaking of which - assuming the use of VMware is to host Windows under Linux - isn't it a bit too early to consider Xen as a replacement?I mean - up to version 3.0 of Xen and the CPU generation which willsupport virtualization in hardware (not released yet, as far as I'm aware) it doesn't sound like a practical laternative to VMware, is it?See http://www.linux-mag.com/content/view/2264/ for a currentstate of the Xen nation. BTW - has anyone any idea how much would the newvirtualization-supporting CPU's cost? Would they be in the high-endserver range or can we expect them to be priced for middle-rangedesktops as well? Cheers,--AmosTo unsubscribe, send mail to [EMAIL PROTECTED] withthe word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Xen vs. VMware (was: Re: VMware Internals Expert)
Hi, Para-virtualizing the OS provides a performance improvement. If you've already done the work, why give it up? It seems to me that there will always be people who would be more satisfied to deal with unpatched kernel , especially in an highly evolving and changing project like Xen. And it seems to me that there will always be poeple who would be satisfied with the ability to run also Windows XP under Xen. >From these two reasons I think using Virtualization Processors may give Xen a priority boost. (regardless of the fact that para virtaulization develoment will continue). An obvious solution is to load para-virtualized drivers into the unmodified OS after it has booted up. It will be interesting to see performance metrics of such a solution. Regards. Rami RosenOn 10/10/05, Muli Ben-Yehuda [EMAIL PROTECTED] wrote: On Mon, Oct 10, 2005 at 01:04:48PM +0200, Rami Rosen wrote: Hi, Developing processors with Virtualization support is an attractive idea; in this way there will no need to patch Linux for using Xen and this will enable using the other OS. It will sure give the Xen project , which is evolving rapidly, a siginificant boost.I don't quite agree. The Xen plan, as I understand it, is to continue supporting both para-virtualized and fully virtualizedOS's. Para-virtualizing the OS provides a performance improvement. Ifyou've already done the work, why give it up? I assume that the performance cost of Xen under such processors will be higher than 3% (and I hope it will be not much higher).The main problem with it at the moment is the IO emulation, which isfairly slow. An obvious solution is to load para-virtualized drivers into the unmodified OS after it has booted up.Cheers,Muli--Muli Ben-Yehudahttp://www.mulix.org | http://mulix.livejournal.com/
Re: Connecting from Linux to MSN Messenger?
Hi, I did connect to an MSN Messenger with gaim (http://gaim.sourceforge.net/) on RH9; it was quite a time ago ; I don't remember the exact details of configuration (it could be that I downloaded some plugin) but it was not difficult (when you add a new accoount you should select from a list of protocols : ICQ,MSN messenger , Jabber and others). Regards, Rami Rosen On 9/13/05, Omer Zak [EMAIL PROTECTED] wrote: Did anyone succeed in connecting to MSN Messenger from Linux (DebianSarge distribution)?If yes - what was the recipe used to accomplish this end?- Package/s- Manual configuration required- Full registration procedure assuming that you are not trying to reuse an existing accountThanks, --- Omer--MS-Windows is the Pal-Kal of the PC world.My own blog is at http://www.livejournal.com/users/tddpirate/My opinions, as expressed in this E-mail message, are mine alone. They do not represent the official policy of any organization with whichI may be affiliated in any way.WARNING TO SPAMMERS:at http://www.zak.co.il/spamwarning.html =To unsubscribe, send mail to [EMAIL PROTECTED] withthe word unsubscribe in the message body, e.g., run the commandecho unsubscribe | mail [EMAIL PROTECTED]
Re: Kernel mode disk I/O. Take 2. (Was: File I/O within kernel threads?)
Hi,
Maybe you should verify you don't have an alignment problem when using O_DIRECT.
When you try , from user space , to use O_DIRECT for reading
from a file, you must use page granularity for achieving page alignemnent.
according to the man (2) read , regarding O_DIRECT flag:
...
Transfer sizes, and the alignment of user buffer and
file offset must all be multiples of the logical block size
of
the file system.
...
If you look at the following code exmaple , a little app from user space,
you will see that it will not work probably if you try to use ptr1 for reading, which is (probably)
not page aligned. ptr2 is changed to the next page boundary, so it will
do the job.
To be more accurate I shouldn't have use 4096 but get the page
size by a syscall (there are cases when it is 8k); and maybe there is something
which can be written better which demonstrates the alignment issue regarding
O_DIRECT, but this seems to illustrate the point.
Also the size parameter of the read() must be a multiple of a page size, as far as
I understand.
// readFile.c
#include sys/types.h
#include sys/stat.h
#include unistd.h
#include asm/types.h
#include asm/fcntl.h
int main()
{
ssize_t size;
char* ptr1;
char* ptr2;
int fd = open(/work/input.txt,O_DIRECT);
ptr1 = (void*)malloc(4096+4096-1);
if (((int)ptr1 % 4096)!=0)
printf(ptr1 % 4096)!=0)\n);
else
printf(ptr1 % 4096)==0)\n);
ptr2 = (void*)((int)ptr1 - (int)ptr1 % 4096 + 4096);
printf(%p %p\n,ptr1,ptr2);
if (((int)ptr2 % 4096)!=0)
printf(ptr2 % 4096)!=0)\n);
else
printf(ptr2 % 4096)==0)\n);
if (fd 0)
{
while (1)
{
size = read(fd,ptr2,4096);
if (size = 0)
{
printf(\n);
break;
}
printf(%s,ptr2);
}
}
}
Regards,
Rami Rosen
On 8/30/05, Gilboa Davara [EMAIL PROTECTED] wrote:
On Tue, 2005-08-30 at 14:40 +0300, guy keren wrote: On Tue, 30 Aug 2005, Gilboa Davara wrote: Anyway, if you don't know how to use O_DIRECT from inside the kernel, why don't you take a look at the part in teh kernel that implments
O_DIRECT from user space and do the same thing? Already on it; However, this code is fairly complex. It involves the file management, vmm, etc. It's very easy to get lost in there... :/
If I can save the next 3 days of ctags drilling (or at least get some pointers as for where to look) I'll be one happy man. you can save the next 3 days by deciding this is a very good idea to learn
how this code works, without taking short-cuts, since you'll need this knowledge later on when you'll have bugs, when you'll need to make it behave a little differently, etc... Let me correct myself.
I'm not looking for a code sample, I am looking for info.Book, URL, etc.Seems that as you go deeper inside the kernel, the less 'printed'documentation exists.Gilboa=
To unsubscribe, send mail to [EMAIL PROTECTED] withthe word unsubscribe in the message body, e.g., run the commandecho unsubscribe | mail
[EMAIL PROTECTED]
Re: NAPI performance ; tg3 and broadcom driver
Hello, I asked about the tg3 and broadcom drivers at Broadcom support; according to them there is almost no difference in quality between the tg3 and broadcom drivers. The broadcom drivers have some additional features, like BASP (Broadcom Advanced Server Program Driver) support. In the near future I will launch some tests to verify this (and notify you). Regards, Rami Rosen On 8/1/05, Marc A. Volovic [EMAIL PROTECTED] wrote: Gilad Ben-Yossef wrote: Did you enable NAPI? Both no and yes ;-)... Tried it both ways. M 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: NAPI performance ; tg3 and broadcom driver
Hi, I've seen several patches to tg3 from broadcom engineers, so hopefully tg3 supports everything as well. Well ,I must admt when I posted the original message, I had looked at the tg3.c only in 2.4.20-8 kernel (since I inteneded then to test only in 2.4.20.* kenels) and I saw in the beginning of the file only David S. Miller (of RH) and Jeff Garzik as authors , and no mention of Broadcom invlovemrnt. Following the answers I got I had looke at a newer kernel ; (2.6.12-rc3) and there wasn't a changes detail description there (as sometimes do occur), but the following lines were added: * Copyright (C) 2004 Sun Microsystems Inc. * Copyright (C) 2005 Broadcom Corporation. * * Firmware is: * Copyright (C) 2000-2003 Broadcom Corporation. So I assume that they Broadcom added something (starting somewhere in etween these 2 versions) without specifying exactly what. Best thing will probably test also with a new kernel, with both drivers ,tg3.c and the bcm5700, in all scenarios. Regards, Rami Rosen On 8/1/05, Gilad Ben-Yossef [EMAIL PROTECTED] wrote: Muli Ben-Yehuda wrote: On Mon, Aug 01, 2005 at 02:32:28PM +0300, Gilad Ben-Yossef wrote: The only problem is that the Broadcom people update their BCM version with newer versions of the chipset form time to time. I've seen several patches to tg3 from broadcom engineers, so hopefully tg3 supports everything as well. hmpf... maybe they just decided to support both, or maybe, seeing as both drivers are GPLed the broadcom blokes are simple doing the Right Thing (tm). Donno. Gilad -- Gilad Ben-Yossef [EMAIL PROTECTED] Codefidence. A name you can trust(tm) Web: http://codefidence.com | SIP: [EMAIL PROTECTED] IL: +972.9.8650475 ext. 201 | Fax:+972.9.8850643 US: +1.360.2275194 ext. 201 | Cel: +972.52.8260388 I am Jack's Overwritten Stack Pointer -- Hackers Club, the movie 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]
NAPI performance ; tg3 and broadcom driver
Hi, NAPI (New API) is a technique to improve network performance on Linux. It is not so new (relatively) - first howto is from 16/2/2002. In a really very brief descriptiom , it uses polling intsead of interrupts in some scenarios. This polling is done for receiving packets (the network card must be able to disbale interrupts). Transmitting packets is done as usual, by asserting interrupts. Polling is usually discouraged in linux device drivers , but there are cases (like when the interrupt rate is very high) in which this technique can improve performance. My question is: 1) does anybody have an experience with this technique ? what is the threshold (of k packets for a second) from which it became efficinet to use NAPI over usual non-NAPI solutions ? (I am talking about Xeon processor ~2.4 Ghz , but also data on other prcoessors can help) 2) Specically regarding NAPI and Broadcom cards: On many distrubutins, the Broadcom BCM5700 family of Network Drivers uses the tg3 driver. ( Tigon3 ethernet driver, in /linux/drivers/net/tg3.c) On broadcom website there is a driver for linux that they wrote for Linux. http://www.broadcom.com/drivers/downloaddrivers.php (the BCM57xx Drivers). Is there any advantage of using the Broadcom drivers to using the tg3 driver ? (and does the kernel version - 2.4 or 2.6 - has any importance in this respect?) Regards, Rami Rosen 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: File I/O within kernel threads?
Hello, There is an articel in LinuxJournal from this April issue by Greg KH which explains how to read and how to write a file from kernel module. He reads the file using set_fs(KERNEL_DS) and write to the file using fget(fd) and vfs_write(). It is clearly not recommended to do such a thing , as he admitted in this article. Moreover, the article itself is titled Driving Me Nuts - Things You Never Should Do in the Kernel. see: http://www.linuxjournal.com/article/8110 Regards, Rami Rosen From: Gilboa Davara [EMAIL PROTECTED] To: IL List linux-il@linux.org.il Subject: Re: File I/O within kernel threads? Date: Thu, 28 Jul 2005 17:24:41 +0300 Gilad, First, thanks for the answer. On Thu, 2005-07-28 at 16:44 +0300, Gilad Ben-Yossef wrote: A kernel thread is indeed proccess context, but since you brought it up already and assuming we're talking 2.6, the correct way to do is use work queues (and hence schedule_work) instead of schedule_task. This will give you the option to start your own work queue (and hence your own kernel thread) thus not having to worry about starving other tasks. And while we're on the subject, you might want to take a look into using a work queue instead of a kernel thread all together - it's a very nice interface to do these sort of things. To learn more about Work queues see here (page 23): http://lwn.net/images/pdf/LDD3/ch07.pdf Funny... I forgot all about work queues. Thanks! It'll save me the need to create way-too-many-kernel-threads. On another different and related note, if you're consider doing IO from kernel sapce (which is a terribly bad as you've pointed out already but I wonder if you know just *how* difficult to debug and racy it is) The rest of the kernel code is already there. Let me put this way. The reset button is my new best friend :-/ I assume that I must guard all the file-ops-read/write file operation myself to prevent ugly race conditions, right? for fear of perfomance, a saner approach that may or may not fit your bill is using RelayFS: http://sourcefrog.net/weblog/software/linux-kernel/relayfs.html Interesting... I remember reading about this in KernelTrap. I doubt that i'll help me. I need concurrent R/W from within the kernel space. Hope this helps, Gilad = 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] _ Don't just search. Find. Check out the new MSN Search! http://search.msn.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: Lightweight graphical environment for Linux
Hello , You may consider taking a look at the Nano-X Window System (previously called Microwindows). see http://www.microwindows.org I had tested it about 2 years ago on an embedded project (it was with PowerPC , but this framework is intended also for x86 and other processors). The project was shut down quite early after it started so I can't say much on this environment. This system takes less resources than X ; it writes directly to the display hardware (but it can also work with framebuffer). I do not know about language support in Nano-X Window Regards, Rami Rosen From: Shachar Shemesh [EMAIL PROTECTED] To: Linux-IL linux-il@linux.org.il Subject: Lightweight graphical environment for Linux Date: Mon, 06 Jun 2005 09:47:22 +0300 Hi all, I'm looking for a light weight graphical environment to write a Linux program in. I'm currently aware of X (free, org), and that they have a light weight version (though I don't remember what or where, links would be welcome), and of svgalib. Is there anything besides those two? Also, is there anything that will give me actual widgets? I don't need anything fancy. Simple buttons and a progress bar would more than do. I'll just define what I mean by light. I need it to require close to zero configuration, and to load quickly. The dotted screen X puts up when it loads is something I would rather avoid. The environment is a semi embedded situation. It's not embedded platform, but I have practical reasons I would like to keep memory and space footprints to a minimum. Ideas welcome, Shachar -- Shachar Shemesh Lingnu Open Source Consulting ltd. Have you backed up today's work? http://www.lingnu.com/backup.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] _ Don't just search. Find. Check out the new MSN Search! http://search.msn.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: VoIP Phone
Hi, We had tested 2 types of phones : one from Cisco and one from Pingtel. Generally the Cisco was OK in terms of configuration and mgmt and voice quality (very low echo if it all). The Pingtel was quite disappointing; However, these phones are more than 2 years in the market so I suppose that there are now newer models ; (I can find out the exact models we have and tell you if it's is significat to you). How about using SW Sip phone ? I saw that XTen released an X-Lite version for Linux on May 16. see: http://www.xten.com/index.php?menu=productssmenu=xlitessmenu=download Regards, Rami Rosen From: Noam Rathaus [EMAIL PROTECTED] To: Linux-IL linux-il@linux.org.il Subject: VoIP Phone Date: Thu, 2 Jun 2005 12:16:58 +0300 Hi, We have started deployment of VoIP in our offices and we have come across an issue where our Grandstream (BT101) phones appear to be semi-ok, but very poor in their voice quality (echo cancellation, etc), configuration options, management, etc. I was wondering if anyone has experience with other Hard-VoIP phones? -- Noam Rathaus CTO Beyond Security Ltd. http://www.beyondsecurity.com http://www.securiteam.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] _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: gedit
Hi, You said: BTW does anyone know how to type hebrew in jedit? I assume it's a type and you mean gedit ; in this I cannot help. On the other hand , in JEdit you simply switch the keyboard to hebrew and type. it works. If using JEdit is a relevant option , this can solve your problem. Dan From: Aaron [EMAIL PROTECTED] To: linux-il linux-il@linux.org.il Subject: gedit Date: Wed, 1 Jun 2005 10:50:25 +0300 Hi all, I have been using lilypond with hebrew and vim does work with hebrew for me, but I must fiddle each time to get it to work. Vim has a plugins for lilypond which is a plus. someone suggested using gedit which types hebrew out of the box. I installed it and indeed it just works. But I wanted a few buzzers a whistles to go with it. I found out that for syntax highlighting gedit uses gtksourceview language files so I dutifully open a latex version and customized it to lilypond. But I am not sure what the file is capable of doing, only syntax highlighting? If so what was the keyword list good for? only highlighting? I would like to duplicate the functions of emacs,vim or jedit, especially auto completion of common terms. BTW does anyone know how to type hebrew in jedit? I remember that there is on this list a gedit developer or at least someone who know its insides. The question is how do I create a plugin for lilypond, and what is it capable of doing. Thanks Aaron = 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] _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: gedit
Does'nt changing the keyboard to hebrew work with gedit ? I do not use gedit. From: Aaron [EMAIL PROTECTED] To: linux-il linux-il@linux.org.il Subject: gedit Date: Wed, 1 Jun 2005 10:50:25 +0300 Hi all, I have been using lilypond with hebrew and vim does work with hebrew for me, but I must fiddle each time to get it to work. Vim has a plugins for lilypond which is a plus. someone suggested using gedit which types hebrew out of the box. I installed it and indeed it just works. But I wanted a few buzzers a whistles to go with it. I found out that for syntax highlighting gedit uses gtksourceview language files so I dutifully open a latex version and customized it to lilypond. But I am not sure what the file is capable of doing, only syntax highlighting? If so what was the keyword list good for? only highlighting? I would like to duplicate the functions of emacs,vim or jedit, especially auto completion of common terms. BTW does anyone know how to type hebrew in jedit? I remember that there is on this list a gedit developer or at least someone who know its insides. The question is how do I create a plugin for lilypond, and what is it capable of doing. Thanks Aaron = 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] _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: motion (was: Re: WebCam Server)
Hello, Hello, Guy Keren said: in the dark, movement causes only a few pixels to change 1) A lot depends of course on the type of the WebCam. I had tested Logitech 4000 Pro and it gave very good results in darkness; At first I did not belive the quality of what I saw. (If I am not wrong it is relatively more expensive than the average). 2) Specifically regarding what Amichai Rotman asked this point is probably less important (since kindergardens work at daylight) ; But he mentioned that he wants to use a couple of USB WebCams , and also that he want to use it on old PIII machine. If he insists on more than one WebCam , this can pose a problem in terms of Bandwidth/Power. It also depends on how many free USB ports he has (and sometimes connecting a USB hub won't solve the problem because of power/bw considerations; though sometimes it does work, as you probably know). Also the type of USB Host Controller can have some significance in some cameras (the most common USB Host Controllers on Intel x386 based PCs are OHCI and UHCI). You can get details on USB devices connected to your Linux (like MxPwr and MxPS , which are max power and max packet size ) by cat /proc/bus/usb/devices. You can also get it visually if you install USBView from http://www.kroah.com/linux-usb This is a gtk based application which GregKH wrote for viewing USB devices. USB webcams usually use high speed isochronous (ISO) transfers. However , it seems to me that the best thing to do, instead of delving into the complexities of linux USB layer and starting meaurmenets (and it is quite complex) is simply connect more than one webcam and run some tests. There is another aspect here - the audio; some webcams have a built in USB microphone. (Like logitech 4000 , logitech quickcam messenger, some Creative Labs models, and many more). It is important to disable the USB audio (assuming that it is not needed) in these tests; it takes quite a bandwidth. (You can achieve it simply by removing the USB audio driver running rmmod audio.) If have some experience in video4linux and Linux USB and if anyone interested in more details on Linux USB , he is invited to read an article I wrote titled : An Overview of Linux USB which was published in LinuxJournal last month : http://www.linuxjournal.com/article/8093 (Though this article deals generally with USB layer , it has some relevance to using Video4Linux 1 devices). good luck, Regards, Rami Rosen From: guy keren [EMAIL PROTECTED] To: Yoni Levy [EMAIL PROTECTED] CC: Amichai Rotman [EMAIL PROTECTED], linux-il@linux.org.il Subject: motion (was: Re: WebCam Server) Date: Sun, 29 May 2005 03:40:46 +0300 (IDT) On Sat, 28 May 2005, Yoni Levy wrote: Hello, there is an application called motion which might help you, it records a video or captures a picture when it detects motion . You can place a camera near the entrance with motion running, and it will take a picture of anyone entering ... and it indeed work(ed) when i tried it (about 2 years ago), to spy on what my dog was doing when i wasn't home. i found one problem: the way that 'motion' is defined, is by the number of pixels of difference between two consecutive images. this made it hard to use with light canges (in the dak, movement causes only a few pixels to change, so it requires a low threshold. in the light, movement causes much more pixels to change. it was also not possible to define where in the pictures to make these calculations - it takes the entire picture for this calculation, whether you like it or not. thus, you need to do some calibrating for your lighting and movement conditions, and then it works quite well. it also only records once picture every X seconds, so it does not fill your hard drive too quickly. good luck, -- guy For world domination - press 1, or dial 0, and please hold, for the creator. -- nob o. dy = 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] Regards, Rami Rosen MxPwr From: guy keren [EMAIL PROTECTED] To: Yoni Levy [EMAIL PROTECTED] CC: Amichai Rotman [EMAIL PROTECTED], linux-il@linux.org.il Subject: motion (was: Re: WebCam Server) Date: Sun, 29 May 2005 03:40:46 +0300 (IDT) On Sat, 28 May 2005, Yoni Levy wrote: Hello, there is an application called motion which might help you, it records a video or captures a picture when it detects motion . You can place a camera near the entrance with motion running, and it will take a picture of anyone entering ... and it indeed work(ed) when i tried it (about 2 years ago), to spy on what my dog was doing when i wasn't home. i found one problem: the way that 'motion' is defined, is by the number of pixels of difference between two consecutive images. this made it hard to use with light canges (in the dak
The return of pwc drivers to linux
Hello, I last August there was a thread in this forum regarding that the linux kernel will not include anymore pwc drivers (for Philips and many other USB webcams). see for example : August Bad news for philips webcam users in http://www.mail-archive.com/linux-il@cs.huji.ac.il/msg35857.html The reason for this discontinuation decision was that this driver included a binary-only module,the decompressor (called pwcx); And due to this fact , Greg Kroah Hartman, the USB maintainer, decided to remove that hook from PWC. As a result, the driver developer (Nemosoft Unv.) declared that he halts his support. see: http://www.smcc.demon.nl/webcam I had git-pulled in the beginning of this week the latest kernel version (linux-2.6.12-rc3). I had a surprise: I saw under /linux-2.6.12-rc3/drivers/usb/media (the folder where the webcam video drivers reside) a folder named pwc. In some files in this pwc fodler there you will find the comment: The decompression routines have been implemented by reverse-engineering the Nemosoft binary pwcx module. Caveat emptor. Googling a bit found the next interesting item in kerneltrap : ( http://kerneltrap.org/node/5066 ) , where you can find the Nemosoft Unv. (the pwc driver) declared The return of PWC in a mail to Greg Kroah. I do not want to deal here (or start a discussoin) on the legality of this reverse-engineering or the technical aspects of how it was done; There can be really various opinions on this ; Just wanted to let you know. Regards, Rami Rosen _ Don't just search. Find. Check out the new MSN Search! http://search.msn.click-url.com/go/onm00200636ave/direct/01/ = 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: The return of pwc drivers to linux
Hello, A little after it Nemosoft declared halting the support for the PWC driver,somebody named Luc Saillard became a new maintainer for the PWC driver; and later Alan Cox made a limited version of the PWC as a patch. But as far as I know, the PWC did came back to the official kernel tree only lately, starting 2.6.12-rc1. ; I am not talking about ( limited?) patches to the tree. BTW: I had looked at a linux-2.6.11 kernel (unpatched) tree (from two months ago,2/3/05); I did not not find the pwc drivers there under drivers/usb/media or anywhere else). Regards, Rami Rosen From: Amos Shapira [EMAIL PROTECTED] Reply-To: Amos Shapira [EMAIL PROTECTED] To: linux-il@linux.org.il Subject: Re: The return of pwc drivers to linux Date: Tue, 3 May 2005 22:21:01 +1000 On 5/3/05, rami rosen [EMAIL PROTECTED] wrote: can be really various opinions on this ; Just wanted to let you know. Thanks. Apparently you just lost track as it unfolded. The return of the driver happened very soon after it was pulled, and nobody questioned the legality of what has happened (IANAL). --Amos 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] _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: [Job Offer] Lingnu Open Source Consulting is looking for a Linux admin
Hi, And if you call it Asterix, you'll probably get sewed: Isn't it (actually) sued? Cheers, Rami Rosen From: Tzafrir Cohen [EMAIL PROTECTED] To: Linux-IL linux-il@linux.org.il Subject: Re: [Job Offer] Lingnu Open Source Consulting is looking for a Linux admin Date: Mon, 18 Apr 2005 20:43:08 +0300 On Mon, Apr 18, 2005 at 07:45:20PM +0300, Gilad Ben-Yossef wrote: Shachar Shemesh wrote: advantage: Apache, Samba, Mail systems, Asterix, FTP, MySQL or It's Asterisk, actually. And if you call it Asterix, you'll probably get sewed: http://tuxmobil.org/mobilix_asterix.html :-) -- Tzafrir Cohen | [EMAIL PROTECTED] | VIM is http://tzafrir.org.il | | a Mutt's [EMAIL PROTECTED] | | best ICQ# 16849755 | | friend = 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] _ FREE pop-up blocking with the new MSN Toolbar - get it now! http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/ = 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: distro with best USB device support
Hello, Can you be more specific - which USB device (or devices) do you intend to use? I also don't want to start an (endless) distro wars and I really do not know if there is a substantial difference in USB from one distro to another. But what I can tell you from my experience with really a large variety of USB webcams on RedHat 9 and Fedora 3 that it was OK (also from the hot plugging aspect). There were a few drivers which compiled not immediately but after slight changes (like logitech quickcam). In newer version of some of the drivers this was fixed. Also USB mouse was OK and USB dongel (I tried M-System diskOnKey.) Mounting of the diskOnKey is done simply by one mount command. I don't have an experience with USB hard drive ; but since I looked at this topic a bit and from what I saw it is not a big issue to configure a system to boot from diskOnKey. (If I remeber well you shoud prepare a proper initrd for support with SCSI emulation becaus the USB disk should use it at boot,but this seems not to be difficult). Hope it helpw, Rami Rosen From: Geoffrey S. Mendelson [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] To: Israel Linux Mailing list linux-il@linux.org.il Subject: distro with best USB device support Date: Tue, 5 Apr 2005 09:36:31 +0300 Hi, I don't want to start a distro war, but I am looking for the distro (x86 32 bit) that has the best USB device support. It does not have to be leading edge, or easy to install, or have a great graphic front end or installer. All I am really concerned about is installing it and playing with a USB device. I want whatever gives me the best chance of it working. TIA, Geoff. -- Geoffrey S. Mendelson, Jerusalem, Israel [EMAIL PROTECTED] N3OWJ/4X1GM IL Voice: 972-2-679-6896 IL Fax: 972-2-648-1443 U.S. Voice: 1-215-821-1838 VoN Skype: mendelsonfamily = 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] _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: LDD3
Hello, You said in your first post that you could not found LDD3 mentioned it in OReilly /Amazon WebSite. Well, I looked at OReilly WebSite this morning and this has chnaged. see http://www.oreilly.com/catalog/linuxdrive3/ (with links for online purchase). regrads, Rami Rosen From: Leonid Podolny [EMAIL PROTECTED] To: rami rosen [EMAIL PROTECTED] Subject: Re: LDD3 Date: Sat, 29 Jan 2005 23:28:18 +0200 -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 rami rosen wrote: | Notice that according to the thread , a third author joined them (Greg | Kroach Hartman) , he is known for his contribution to Linux Kernel (he | is the maintainer of USB and PCI and others) ; and he is also known | for many articles he wrote (in lwn.net,linux journal and more). | That's great. I have managed to find it at amazon.com (by an ISBN) and they list only one author -- J. Corbet. I thought, he will be the one to do the work of adaptation of the book to 2.6. Now I wonder what will be the fastest way to lay my greedy hands on it. :) - -- - ~ Leonid Podolny | /\ ~ | \ / ASCII Ribbon Campaign ~ leonidp(at)gmail.com |x Against HTML Mail ~ +972-54-5696948 | / \ - PGP fingerprint: 51B2 F1DB 485E 2C48 2E17 94D1 7EC4 E524 B156 B9F0 PGP key:http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0xB156B9F0 - -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.6 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFB+//yfsTlJLFWufARAnguAJ9jT1xXzzMEv1UctnHSmYh8a09r9QCfSgIX u0PI9WZvD1LOIx2LJA44Fpo= =XbVF -END PGP SIGNATURE- _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: LDD3
Hello, I have the habbit of reading the lkml fom time to time. So,in a thread on somewhat a different topic, from the beginning of this month, (3/1/05) I saw a message from one of the authors (Jonathan Corbet). According to him at will be due at LinuxWorld in Boston as other mentioned here; Notice that according to the thread , a third author joined them (Greg Kroach Hartman) , he is known for his contribution to Linux Kernel (he is the maintainer of USB and PCI and others) ; and he is also known for many articles he wrote (in lwn.net,linux journal and more). see: http://www.uwsg.iu.edu/hypermail/linux/kernel/0501.0/0286.html the LinuxWorld in Boson is starting in 2 weeks from now, in 14/2/05. I hope they will made it to that date; I am sure that you are not the only one who is expecting this book. see http://www.linuxworldexpo.com/live/12/events/12BOS05A Regards, Rami Rosen _ Express yourself instantly with MSN Messenger! Download today it's FREE! http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/ = 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: Quickcam 4000 pro
Hello, From: Cyril Scetbon [EMAIL PROTECTED] To: rami rosen [EMAIL PROTECTED] CC: [EMAIL PROTECTED] Subject: Re: Quickcam 4000 pro Date: Fri, 10 Sep 2004 07:55:40 +0200 rami rosen wrote: Hello, I had tried the logitech 4000 pro on Linux with the pwc and pwcx driver. You get good results for a CIF image (352x288). The quality is much higher comparing to logitech quickcam. I did not tried it against NetMeeting. The SW I tried was mainly gnomemeeting. Regarding the Audio : I had tried the built in microphone in this camera. When you use a headset - the quality is reasonable. When you try to use spekaers on both sides - there is an echo which becomes stronger during time (until it is simply unbearable shrike). yeah, I noticed it :-( that's why I don't use the webcam speaker Is it necessary to use skype for the sound combinate with gnomemeeting for the video or gnomemeeting manages both well ? - Your question , If I understand you right, is regarding using GnomeMeeting for management of Audio and Video session and combining using Skype. - As I wrote , if you use a headset on both sides the GnomeMeeting is OK for both. - If you want to use speakers on both sides, and to use simultanously GnomeMeeting only for Video on both sides and Skype for Audio on both sides , this can be an interesting idea. - I did not try it yet. (I really intend to give it a try) ; I did find in the GnomeMeeting GUI that Video receive/transmit can be disabled (there are check boxes for it in preferences) but I did not saw such check boxes for Audio. I suppose it is not really a problem ; if it will disturb Skype transmit/receive audio (because trying to use the same audio device which GnomeMeeting uses), it seems to me that you can configure GnomeMeeting to use faked audio devices (in Preferences - Devices-Audio Devices). (fore example, /dev/dsp2) Anyhow, the Skype nework signalling layer is p2p. GnomeMeeting nework signalling layer is the good old H323 (I heard about some intentions to replace it in the future , in the GnomeMeeting project, with the more modern (and less cumersome) SIP (Session Initiation Protocol) , which is commonly used nowadays in VoIP. But this can take quite a time and does not change what is written in next paragraph: If your question was regarding calling Skype from GnomeMeeting (or vice versa) it seems that to due to this difference in the signalling protocol this is not easy ro implement (NetMeeting , BTW , is based on H323 , and due to this , interconnection between NetMeeting and GnomeMeeting is possible and implemented - though I did not test it ). and , btw: To be more accurate, GnomeMeeting has an Echo Cancellatrion feature if you use special hw (QuickNet card) - you can see it also in Preferences - Devices-Audio Devices and in their documentation. I don't know the details (it could be that this is true only when working against PSTN /special telephony equipment , etc.). It's a pitty that there is no free/open source video conference Linux app with Echo Cancellation (at least I didn't hear of such a project). Regards, Rami Rosen (BTW, this is a known phenomenon which happened to me also in NetMeeting but it seems to me that on Skype the problem is really reduced; If I am not wrong they use sw echo chacellation mechanisms of IPGlobal (which is commercial) and as I know there is no OpenSource VideoConference projects with Echo Chancellation ; I had looked in GnomeMeeting newsgroup and I saw that this probelm is with no real solution (except the advice to lower the speakers, which not really helps). And BTW, probably the sensor is very sensitive - (It is a SONY sensor l at least that is waht you can see in the kernel log) - the quality of this WebCam in almost total darkness in unbelievable. I tried it on 2.4 mhz and also on 700 Mhz Intel machines and it seems enough horsepower for the codecs. I also tried it with opensource ophone 1.4.1 http://www.openh323.org/code.html but the results are not as good as gnomemeeting. Hope this helps, Rami Rosen _ The new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=features/junkmail = 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: Quickcam 4000 pro
Hello, I had tried the logitech 4000 pro on Linux with the pwc and pwcx driver. You get good results for a CIF image (352x288). The quality is much higher comparing to logitech quickcam. I did not tried it against NetMeeting. The SW I tried was mainly gnomemeeting. Regarding the Audio : I had tried the built in microphone in this camera. When you use a headset - the quality is reasonable. When you try to use spekaers on both sides - there is an echo which becomes stronger during time (until it is simply unbearable shrike). (BTW, this is a known phenomenon which happened to me also in NetMeeting but it seems to me that on Skype the problem is really reduced; If I am not wrong they use sw echo chacellation mechanisms of IPGlobal (which is commercial) and as I know there is no OpenSource VideoConference projects with Echo Chancellation ; I had looked in GnomeMeeting newsgroup and I saw that this probelm is with no real solution (except the advice to lower the speakers, which not really helps). And BTW, probably the sensor is very sensitive - (It is a SONY sensor l at least that is waht you can see in the kernel log) - the quality of this WebCam in almost total darkness in unbelievable. I tried it on 2.4 mhz and also on 700 Mhz Intel machines and it seems enough horsepower for the codecs. I also tried it with opensource ophone 1.4.1 http://www.openh323.org/code.html but the results are not as good as gnomemeeting. Hope this helps, Rami Rosen From: Cyril Scetbon [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Quickcam 4000 pro Date: Thu, 09 Sep 2004 08:19:57 +0200 Hi all, I would like to get feedback about using this webcam under linux. Which soft was used ? Have you succeeded in communicating with windows users? (gnomemeeting, others ?) In using the integrated micro (feedback from others webcam are welcomed) ? 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] _ Add photos to your e-mail with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail = 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: Bad news for philips webcam users :-(
Hello, This is indeed bad news ; this driver proved itself along the years, and I hope some arrangement will be found enabling us to see it's comeback. I want to add that this driver also supports other platforms besides x86 (I used it sucessfully with MIPS and PPC ) and it supports many types cameras besides Philips (like Creative Labs ). regards, Rami Rosen From: Cyril Scetbon [EMAIL PROTECTED] To: Linux-IL [EMAIL PROTECTED] Subject: Bad news for philips webcam users :-( Date: Thu, 26 Aug 2004 23:58:49 +0200 Pwc is no longer supported in the linux kernel due to disagreement between the pwc maintainer and the kernel maintainers. http://www.smcc.demon.nl/webcam/ A bad news for users like me who have just bought this webcam a few months ago :-( = 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] _ Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail = 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: motherboard identification
Hello, on RedHat linux you can use dmidecode and look for Board Information Block. dmidecode is part of kernel-utils. It will give you Vendor and Product and Version. You can google further with the info you get if you want the chipset ID. I also do not see the mobo ID in the kernel log ; it appear on the screen in boot (you can pause to see it). regards, Rami From: Efraim Yawitz [EMAIL PROTECTED] To: linux-il list [EMAIL PROTECTED] Subject: motherboard identification Date: Sun, 22 Aug 2004 15:14:07 +0300 (IDT) Isn't there any way to get the motherboard ID from Linux? I've searched around the web and only find DOS or Windows utilities. Thanks, Ephraim = 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 new MSN 8: advanced junk mail protection and 2 months FREE* http://join.msn.com/?page=features/junkmail = 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: retrieving network interface information
Hello Gershon , I assume these interfaces have a different IP addresses. netstat -a shows the IP addresses on which you have opened sockets ; so it seems to me you can deduce from this to which interface each socket belongs. Is it good enough for you ? Or must you do it a program , using ioctl() calls? regards, Rami Rosen _ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail = 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]
usb_set_interface() and alternate settings
Hello, This question is for LINUX USB gurus: ov511 and pwc are both linux kernel usb drivers for 2 families of webcams : ov511 chip based webcams and philips-based webcams. Both are part of the linux kernel (under drivers/usb) (this is only a general description;for details see: http://alpha.dyndns.org/ov511/ http://www.smcc.demon.nl/webcam/). ov511 has a module parameter named packetsize, which enables changing the maximum Packet Size. Looking a little at ov511.c shows that it does it by eventually calling usb_set_interface with a different alternate setting. My question: did anybody tried to use this parameter ? I am trying to do the same in pwc-if.c ; (even though it does not have that module paramter ; but it has a call to usb_set_interface(). It succeeds in usb_set_interface, but hangs afterwads. Could it be that the pwc driver does not support a different alternate setting other than the default one (even though it's usb descriptor (under /proc/bus/usb/devices) shows a lot of other alternate settings). regards, Rami Rosen _ The new MSN 8: smart spam protection and 2 months FREE* http://join.msn.com/?page=features/junkmail = 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]
