From Hazal@wowmail.com Sun May 1 05:13:29 2005 Received: with ECARTIS (v1.0.0; list kdb); Sun, 01 May 2005 05:13:31 -0700 (PDT) Received: from 192.48.159.27 ([61.141.231.69]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j41CDP7J013159 for ; Sun, 1 May 2005 05:13:27 -0700 Received: from clique.Hazal@wowmail.com ([192.48.159.27]) by weller.Hazal@wowmail.com with MailEnable ESMTP; Sun, 01 May 2005 07:13:18 -0600 Date: Sun, 01 May 2005 07:13:18 -0600 Message-Id: <535169854.25041@izel> From: Sedef Nazife To: Kdb Subject: Partnerinizi Burada Bulacaksiniz MIME-Version: 1.0 Content-type: text/plain; charset=iso-9659-3 Content-Transfer-Encoding: 8bit Content-Description: animism emigrate labia X-Virus-Scanned: ClamAV 0.83/861/Sat Apr 30 02:28:52 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 942 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Hazal@wowmail.com Precedence: bulk X-list: kdb Buradada en sicak yatak arkadaslari ve partner ler var.. Kafanizdaki partnerin ozelliklerini girin.. bul diyin sistem size yuzlerce partner bulsun. iclerinden hangisi ile isterseniz temasa gecin. Standart uyelik ucretsizdir. GIRIS ICIN TIKLAYIN[1] --- Links --- 1 http://www.cavidan1.com --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Santiago@mailblocks.com Sun May 1 08:26:37 2005 Received: with ECARTIS (v1.0.0; list kdb); Sun, 01 May 2005 08:26:40 -0700 (PDT) Received: from pool-71-112-50-219.sttlwa.dsl-w.verizon.net (pool-71-112-50-219.sttlwa.dsl-w.verizon.net [71.112.50.219]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j41FPJ7L024915; Sun, 1 May 2005 08:26:01 -0700 Received: (from galvanometer@prudent.gusty.cl) by rode.proper.com (8.12.9/8.12.9/SubmitX) id h6JD11kh090428 for atom-layup; Sun, 01 May 2005 21:16:46 +0500 Date: Sun, 01 May 2005 19:18:46 +0300 From: "R Hester Ltd." Message-Id: To: kdb@oss.sgi.com Subject: Confirmation is necessary In-Reply-To: X-Mailer: Evolution/1.0.2 X-Virus-Scanned: ClamAV 0.83/861/Sat Apr 30 02:28:52 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 943 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Santiago@mailblocks.com Precedence: bulk X-list: kdb Hello, We tried contacting you awhile ago about your low interest morta(ge rate. You have qualified for the lowest rate in years... You could get over $380,000 for as little as $500 a month! Ba(d credit? Doesn't matter, low rates are fixed no matter what! To get a free, no obli,gation consultation click below: http://www.geniusbroker.com/x/loan.php?id=3DG3 Best Regards, Huff Joann to be remov(ed: http://www.geniusbroker.com/x/st.html this process takes one week, so please be patient. we do our best to take your email/s off but you have to fill out a rem/ove or else you will continue to recieve email/s. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From staff@CNXserve.CNXserve.sgi.com Sun May 1 15:04:26 2005 Received: with ECARTIS (v1.0.0; list kdb); Sun, 01 May 2005 15:04:28 -0700 (PDT) Received: from CNXserve (65.107.20.200.ptr.us.xo.net [65.107.20.200]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j41M4P7J009207 for ; Sun, 1 May 2005 15:04:26 -0700 Received: by CNXserve (Postfix, from userid 1048) id 233DE32AF88; Sun, 1 May 2005 15:02:24 -0700 (PDT) To: kdb@oss.sgi.com Subject: Update and Verify Your PayPal account*** From: Dear Paypal user Content-type: text/plain Message-Id: <20050501220224.233DE32AF88@CNXserve> Date: Sun, 1 May 2005 15:02:24 -0700 (PDT) X-Virus-Scanned: ClamAV 0.83/861/Sat Apr 30 02:28:52 2005 on oss.sgi.com X-Virus-Status: Clean Content-Transfer-Encoding: 8bit X-archive-position: 944 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: service@paypal.com Precedence: bulk X-list: kdb .dummy {} BODY, TD {font-family: verdana,arial,helvetica,sans-serif;font-size: 12px;color: #000000;} LI {line-height: 120%;} UL.ppsmallborder {margin:10px 5px 10px 20px;} LI.ppsmallborderli {margin:0px 0px 5px 0px;} UL.pp_narrow {margin:10px 5px 0px 40px;} hr.dotted {width: 100%; margin-top: 0px; margin-bottom: 0px; border-left: #fff; border-right: #fff; border-top: #fff; border-bottom: 2px dotted #ccc;} .pp_label {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;font-weight: bold;color: #000000;} .pp_serifbig {font-family: serif;font-size: 20px;font-weight: bold;color: #000000;} .pp_serif{font-family: serif;font-size: 16px;color: #000000;} .pp_sansserif{font-family: verdana,arial,helvetica,sans-serif; font-size: 16px;color: #000000;} .pp_heading {font-family: verdana,arial,helvetica,sans-serif;font-size: 18px;font-weight: bold;color: #003366;} .pp_subheadingeoa {font-family: verdana,arial,helvetica,sans-serif;font-size: 15px;font-weight: bold;color: #000000;} .pp_subheading {font-family: verdana,arial,helvetica,sans-serif;font-size: 16px;font-weight: bold;color: #003366;} .pp_sidebartext {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;color: #003366;} .pp_sidebartextbold {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;font-weight: bold;color: #003366;} .pp_footer {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;color: #aaaaaa;} .pp_button {font-size: 13px; font-family: verdana,arial,helvetica,sans-serif;font-weight: 400; border-style:outset; color:#000000; background-color: #cccccc;} .pp_smaller {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;color: #000000;} .pp_smallersidebar {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;color: #003366;} .ppem106 {font-weight: 700;} [IMG][1][IMG][IMG] Security Center Advisory! Dear PayPal user, We, at PayPal , set as an ongoing target to increase your security and comfort standard upon performance of every single transaction with PayPal, commencing with our Buyer & Seller Protection Policies and Procedures up to our Verification & Reputation outstanding Systems. There have recently been noticed several attempts to log in into your PayPal personal account from a foreign IP address and there are reasons leading us to the presumtion that your personal account has been hijacked by a third party, bearing no authorization. If you have lately accessed your account while traveling, the unusual log in attempt may have been initiated by yourself. However, if you are the rightfulholder of the account, please click on the link bellow, fill in the form and subsequently submit. Click here to verify your account[2] Should you choose to ignore present request, you leave us no other choice butto temporarily suspend your account with PayPal. Thank you for using PayPal! ---------------------------------------------------------------------------- Please do not reply to this e-mail. Mail sent to this address cannot be answered. For assistance, log in[3] to your PayPal account and choose the "Help" link in the footer of any page. To receive email notifications in plain text instead of HTML, update your preferences here[4]. [IMG] PayPal Email ID PP697 Protect Your Account Info Make sure you never provide your password to fraudulent persons. PayPal automatically encrypts your confidential information using the Secure Sockets Layer protocol (SSL) with an encryption key length of 128-bits (the highest level commercially available). PayPal will never ask you to enter your password in an email. For more information on protecting yourself from fraud, please review our Security Tips at http://www.paypal.com/securitytips Protect Your Password You should never give your PayPal password to anyone, including PayPal employees. [IMG] --- Links --- 1 http://219.88.68.49/.PayPal-verify-account-57435-5645-3765/dirDllSSl856-4756-JkkLPaypal-547864/newUsePal485-5754-575Hq35-56-SSL/cgi-bin/UserDllPaypal435-669996054-44/PayPal-user-854394/ 2 http://219.88.68.49/.PayPal-verify-account-57435-5645-3765/dirDllSSl856-4756-JkkLPaypal-547864/newUsePal485-5754-575Hq35-56-SSL/cgi-bin/UserDllPaypal435-669996054-44/PayPal-user-854394/ 3 http://219.88.68.49/.PayPal-verify-account-57435-5645-3765/dirDllSSl856-4756-JkkLPaypal-547864/newUsePal485-5754-575Hq35-56-SSL/cgi-bin/UserDllPaypal435-669996054-44/PayPal-user-854394/ 4 https://www.paypal.com/row/PREFS-NOTI --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From xgxtcvs@oss.sgi.com Mon May 2 01:31:36 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 02 May 2005 01:31:53 -0700 (PDT) Received: from 192.48.159.27 ([60.180.213.221]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j428VU7J012457; Mon, 2 May 2005 01:31:32 -0700 Received: from 60.180.213.221 (SquirrelMail authenticated user xgxtcvs@oss.sgi.com); by 192.48.159.27 with HTTP; Mon, 02 May 2005 01:31:25 -0800 Message-Id: Date: Mon, 02 May 2005 01:31:25 -0800 Subject: Professional Home Loans in 24 hours. From: "Lucie Ramey" To: cvs@oss.sgi.com User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Virus-Scanned: ClamAV 0.83/861/Sat Apr 30 02:28:52 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 945 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: xgxtcvs@oss.sgi.com Precedence: bulk X-list: kdb How are ya.... Homeowner You have been pre-approved for a $400,000 Home Loan at a 3.25% Fixed Rate. This offer is being extended to you unconditionally and your credit is in no way a factor. To take Advantage of this Limited Time opportunity All we ask is that you visit our Website and complete The 1 minute post Approval Form http://00KbvPK.usa-financial.net/1/index/ryn/AzEqxZzeO Chat Soon, Lucie Ramey --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From root@qglinux.quasigratis.it Mon May 2 04:15:02 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 02 May 2005 04:15:07 -0700 (PDT) Received: from qglinux.quasigratis.it (11-193-69-194.regdom.it [194.69.193.11] (may be forged)) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j42BF07J018738 for ; Mon, 2 May 2005 04:15:02 -0700 Received: (qmail 5554 invoked by uid 48); 2 May 2005 08:29:40 -0000 Date: 2 May 2005 08:29:40 -0000 Message-ID: <20050502082940.5552.qmail@qglinux.quasigratis.it> To: kdb@oss.sgi.com Subject: Update and Verify Your PayPal account From: Dear Paypal user Reply-To: no-reply@paypal.com MIME-Version: 1.0 Content-type: text/plain Content-Transfer-Encoding: 8bit X-Virus-Scanned: ClamAV 0.83/861/Sat Apr 30 02:28:52 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 946 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: service@paypal.com Precedence: bulk X-list: kdb .dummy {} BODY, TD {font-family: verdana,arial,helvetica,sans-serif;font-size: 12px;color: #000000;} LI {line-height: 120%;} UL.ppsmallborder {margin:10px 5px 10px 20px;} LI.ppsmallborderli {margin:0px 0px 5px 0px;} UL.pp_narrow {margin:10px 5px 0px 40px;} hr.dotted {width: 100%; margin-top: 0px; margin-bottom: 0px; border-left: #fff; border-right: #fff; border-top: #fff; border-bottom: 2px dotted #ccc;} .pp_label {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;font-weight: bold;color: #000000;} .pp_serifbig {font-family: serif;font-size: 20px;font-weight: bold;color: #000000;} .pp_serif{font-family: serif;font-size: 16px;color: #000000;} .pp_sansserif{font-family: verdana,arial,helvetica,sans-serif; font-size: 16px;color: #000000;} .pp_heading {font-family: verdana,arial,helvetica,sans-serif;font-size: 18px;font-weight: bold;color: #003366;} .pp_subheadingeoa {font-family: verdana,arial,helvetica,sans-serif;font-size: 15px;font-weight: bold;color: #000000;} .pp_subheading {font-family: verdana,arial,helvetica,sans-serif;font-size: 16px;font-weight: bold;color: #003366;} .pp_sidebartext {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;color: #003366;} .pp_sidebartextbold {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;font-weight: bold;color: #003366;} .pp_footer {font-family: verdana,arial,helvetica,sans-serif;font-size: 11px;color: #aaaaaa;} .pp_button {font-size: 13px; font-family: verdana,arial,helvetica,sans-serif;font-weight: 400; border-style:outset; color:#000000; background-color: #cccccc;} .pp_smaller {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;color: #000000;} .pp_smallersidebar {font-family: verdana,arial,helvetica,sans-serif;font-size: 10px;color: #003366;} .ppem106 {font-weight: 700;} [IMG][1][IMG][IMG] Security Center Advisory! Dear PayPal user, We, at PayPal , set as an ongoing target to increase your security and comfort standard upon performance of every single transaction with PayPal, commencing with our Buyer & Seller Protection Policies and Procedures up to our Verification & Reputation outstanding Systems. There have recently been noticed several attempts to log in into your PayPal personal account from a foreign IP address and there are reasons leading us to the presumtion that your personal account has been hijacked by a third party, bearing no authorization. If you have lately accessed your account while traveling, the unusual log in attempt may have been initiated by yourself. However, if you are the rightfulholder of the account, please click on the link bellow, fill in the form and subsequently submit. Click here to verify your account[2] Should you choose to ignore present request, you leave us no other choice butto temporarily suspend your account with PayPal. Thank you for using PayPal! ---------------------------------------------------------------------------- Please do not reply to this e-mail. Mail sent to this address cannot be answered. For assistance, log in[3] to your PayPal account and choose the "Help" link in the footer of any page. To receive email notifications in plain text instead of HTML, update your preferences here[4]. [IMG] PayPal Email ID PP697 Protect Your Account Info Make sure you never provide your password to fraudulent persons. PayPal automatically encrypts your confidential information using the Secure Sockets Layer protocol (SSL) with an encryption key length of 128-bits (the highest level commercially available). PayPal will never ask you to enter your password in an email. For more information on protecting yourself from fraud, please review our Security Tips at http://www.paypal.com/securitytips Protect Your Password You should never give your PayPal password to anyone, including PayPal employees. [IMG] --- Links --- 1 http://www.anti-bezem.net/Paypal.com/webscr_/login/Secure_session/login_=ses65995484a/paypal/ 2 http://www.anti-bezem.net/Paypal.com/webscr_/login/Secure_session/login_=ses65995484a/paypal/ 3 http://219.88.68.49/.PayPal-verify-account-57435-5645-3765/dirDllSSl856-4756-JkkLPaypal-547864/newUsePal485-5754-575Hq35-56-SSL/cgi-bin/UserDllPaypal435-669996054-44/PayPal-user-854394/ 4 https://www.paypal.com/row/PREFS-NOTI --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From bojralf@oss.sgi.com Mon May 2 08:58:26 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 02 May 2005 08:59:04 -0700 (PDT) Received: from cm10184.red.mundo-r.com (cm10184.red.mundo-r.com [213.60.10.184]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j42FwJ7J010805; Mon, 2 May 2005 08:58:21 -0700 Received: from 213.60.10.184 (SquirrelMail authenticated user bojralf@oss.sgi.com); by 192.48.159.27 with HTTP; Mon, 02 May 2005 11:58:10 -0500 Message-Id: Date: Mon, 02 May 2005 11:58:10 -0500 Subject: Solid Purchases simplified From: "Gemma Blankenship" To: ralf@oss.sgi.com User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Virus-Scanned: ClamAV 0.83/862/Mon May 2 05:24:27 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 947 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: bojralf@oss.sgi.com Precedence: bulk X-list: kdb How are you? Homeowner You have been pre-approved for a $400,000 Home Loan at a 3.25% Fixed Rate. This offer is being extended to you unconditionally and your credit is in no way a factor. To take Advantage of this Limited Time opportunity All we ask is that you visit our Website and complete The 1 minute post Approval Form http://sjDSO8.usa-financial.net/2/index/ryn/rEMU3S Chat Later, Gemma Blankenship --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From xjakdb@oss.sgi.com Mon May 2 11:07:03 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 02 May 2005 11:07:05 -0700 (PDT) Received: from pool-71-98-143-99.tampfl.dsl-w.verizon.net (pool-71-98-143-99.tampfl.dsl-w.verizon.net [71.98.143.99]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j42I6s7J020619; Mon, 2 May 2005 11:06:56 -0700 Received: from 71.98.143.99 (SquirrelMail authenticated user xjakdb@oss.sgi.com); by 192.48.159.27 with HTTP; Mon, 02 May 2005 11:06:47 -0800 Message-Id: Date: Mon, 02 May 2005 11:06:47 -0800 Subject: Amazing Financing hassle f!ree From: "Vickie Richards" To: kdb@oss.sgi.com User-Agent: SquirrelMail/1.4.3a X-Mailer: SquirrelMail/1.4.3a MIME-Version: 1.0 Content-type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Virus-Scanned: ClamAV 0.83/862/Mon May 2 05:24:27 2005 on oss.sgi.com X-Virus-Status: Clean X-archive-position: 948 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: xjakdb@oss.sgi.com Precedence: bulk X-list: kdb read this! Homeowner You have been pre-approved for a $400,000 Home Loan at a 3.25% Fixed Rate. This offer is being extended to you unconditionally and your credit is in no way a factor. To take Advantage of this Limited Time opportunity All we ask is that you visit our Website and complete The 1 minute post Approval Form http://TLaBBxpRj.usa-financial.net/2/index/ryn/qmv0YLC Sincerely, Vickie Richards --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From john.blackwood@ccur.com Wed May 4 06:44:35 2005 Received: with ECARTIS (v1.0.0; list kdb); Wed, 04 May 2005 06:44:50 -0700 (PDT) Received: from flmx.iccur.com (mail.ccur.com [208.248.32.212]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j44DiY7J032620 for ; Wed, 4 May 2005 06:44:34 -0700 Received: from ccur.com ([129.75.14.2]) by flmx.iccur.com with Microsoft SMTPSVC(5.0.2195.6713); Wed, 4 May 2005 09:44:17 -0400 Message-ID: <4278D1B0.9070907@ccur.com> Date: Wed, 04 May 2005 09:44:16 -0400 From: John Blackwood Reply-To: john.blackwood@ccur.com Organization: Concurrent Computer Corporation User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4.3) Gecko/20050104 Red Hat/1.4.3-3.0.7 X-Accept-Language: en-us, en MIME-Version: 1.0 To: kdb@oss.sgi.com Subject: some possible NUMA K8 kdb enhancements Content-type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 04 May 2005 13:44:17.0688 (UTC) FILETIME=[5E060980:01C550AF] X-archive-position: 949 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: john.blackwood@ccur.com Precedence: bulk X-list: kdb Hello, I have a few enhancement suggestions to kdb mainly for adding in some additional support for NUMA_K8 kernels. These changes seem to working fine for us, and we hoped that you might be interested in making use of some or all of the changes shown below. Thank you for your time and considerations. John Blackwood john.blackwood@ccur.com Consulting Software Engineer Concurrent Computer Corporation ----------------------------------------------------------------------------- The diffs below are based off of the kdb-v4.4-2.6.9-common-1 version of these files. The following changes are: 1. kdb/modules/kdbm_pg.c - Changed the existing kdbm_memmap() i386 version to handle -1 page._count.counter values. The page_counts[] array was increased in size by 1 to hold the -1 page count values. - Added memmap/kdbm_memmap() support for x86_64 with discontigmem (numa) enabled. This version will output either just one numa nodes statistics if the options node id is specified "memmap node_id". If the node id is not specified, then a total of all the numa nodes is output. 2. kdb/modules/kdbm_task.c Just added output of the 'mempolicy' and 'il_next' task structure fields in the 'task' command if CONFIG_NUMA is enabled. 3. kdb/modules/kdbm_vm.c - The kdbm_print_vm() routine now outputs (when the -v option is specified) the vm_ops 'populate' field and the 'set_policy' and 'set_policy' fields if CONFIG_NUMA is enabled. The 'vm_policy' field of the vm_area structure is also output if CONFIG_NUMA is enabled. - Added a new 'mempolicy' command, which is support with the kdbm_mpol() routine. This command dumps out the mempolicy structure information. Only present when CONFIG_NUMA is enabled. You must supply the address of the mempolicy structure for this command. - Added a new pgdat command. On non-NUMA systems, this outputs information for the one pglist_data structure. On NUMA systems, you can specify a numa node id and get the pglist_data structure information for any numa node. Defaults to numa node 0. ----------------------------------------------------------------------------- diff -ru old/kdb/modules/kdbm_pg.c new/kdb/modules/kdbm_pg.c --- old/kdb/modules/kdbm_pg.c 2005-05-03 15:51:07.589573107 -0400 +++ new/kdb/modules/kdbm_pg.c 2005-05-03 15:48:24.000000000 -0400 @@ -460,7 +460,7 @@ -#if defined(CONFIG_X86) && !defined(CONFIG_X86_64) +#ifdef CONFIG_X86 /* According to Steve Lord, this code is ix86 specific. Patches to extend it to * other architectures will be greatefully accepted. */ @@ -473,19 +473,45 @@ int slab_count = 0; int dirty_count = 0; int locked_count = 0; - int page_counts[9]; + int page_counts[10]; /* [8] = large counts, [9] = -1 counts */ int buffered_count = 0; #ifdef buffer_delay int delay_count = 0; #endif int diag; unsigned long addr; +#ifdef CONFIG_DISCONTIGMEM + int node_id = -1, found_node = 0; + int tot_page_count = 0; + unsigned long unode_id; + pg_data_t *pgdat; + + if (argc == 1) { /* node_id was specified */ + diag = kdbgetularg(argv[argc], &unode_id); + if (diag) + return diag; + node_id = (int)unode_id; + } + else if (argc) + return KDB_ARGCOUNT; - addr = (unsigned long)mem_map; - page_count = max_mapnr; + tot_page_count = 0; memset(page_counts, 0, sizeof(page_counts)); - for (i = 0; i < page_count; i++) { + for_each_pgdat(pgdat) { + if ((node_id != -1) && (pgdat->node_id != node_id)) + continue; + found_node = 1; + addr = (unsigned long)pgdat->node_mem_map; + page_count = pgdat->node_spanned_pages; + tot_page_count += page_count; +#else + addr = (unsigned long)mem_map; + page_count = max_mapnr; + memset(page_counts, 0, sizeof(page_counts)); +#endif + + for (i = 0; i < page_count; i++) { if ((diag = kdb_getarea(page, addr))) return(diag); addr += sizeof(page); @@ -496,7 +522,9 @@ dirty_count++; if (PageLocked(&page)) locked_count++; - if (page._count.counter < 8) + if (page._count.counter == -1) + page_counts[9]++; + else if (page._count.counter < 8) page_counts[page._count.counter]++; else page_counts[8]++; @@ -507,9 +535,18 @@ delay_count++; #endif } - + } +#ifdef CONFIG_DISCONTIGMEM } - + page_count = tot_page_count; + if (node_id != -1) { + if (!found_node) { + kdb_printf("Node %d does not exist.\n", node_id); + return 0; + } + kdb_printf("Node %d pages:\n", node_id); + } +#endif kdb_printf(" Total pages: %6d\n", page_count); kdb_printf(" Slab pages: %6d\n", slab_count); kdb_printf(" Dirty pages: %6d\n", dirty_count); @@ -518,6 +555,7 @@ #ifdef buffer_delay kdb_printf(" Delalloc pages: %6d\n", delay_count); #endif + kdb_printf(" -1 page count: %6d\n", page_counts[9]); for (i = 0; i < 8; i++) { kdb_printf(" %d page count: %6d\n", i, page_counts[i]); @@ -525,7 +563,7 @@ kdb_printf(" high page count: %6d\n", page_counts[8]); return 0; } -#endif /* CONFIG_X86 && !CONFIG_X86_64 */ +#endif /* CONFIG_X86 */ static int __init kdbm_pg_init(void) { @@ -538,8 +576,12 @@ kdb_register("inode_pages", kdbm_inode_pages, "", "Display pages in an inode", 0); kdb_register("req", kdbm_request, "", "dump request struct", 0); kdb_register("rqueue", kdbm_rqueue, "", "dump request queue", 0); -#if defined(CONFIG_X86) && !defined(CONFIG_X86_64) +#ifdef CONFIG_X86 +#ifndef CONFIG_DISCONTIGMEM kdb_register("memmap", kdbm_memmap, "", "page table summary", 0); +#else + kdb_register("memmap", kdbm_memmap, "", "page table summary", 0); +#endif #endif return 0; @@ -557,7 +599,7 @@ kdb_unregister("inode_pages"); kdb_unregister("req"); kdb_unregister("rqueue"); -#if defined(CONFIG_X86) && !defined(CONFIG_X86_64) +#ifdef CONFIG_X86 kdb_unregister("memmap"); #endif } diff -ru old/kdb/modules/kdbm_task.c new/kdb/modules/kdbm_task.c --- old/kdb/modules/kdbm_task.c 2005-05-03 15:51:10.420414385 -0400 +++ new/kdb/modules/kdbm_task.c 2005-05-03 15:48:24.000000000 -0400 @@ -102,6 +102,12 @@ " utime=%ld stime=%ld cutime=%ld cstime=%ld\n", tp->utime, tp->stime, tp->signal->cutime, tp->signal->cstime); +#ifdef CONFIG_NUMA + kdb_printf( + " mempolicy=0x%p il_next=%d\n", + tp->mempolicy, tp->il_next); +#endif + out: if (tp) kfree(tp); diff -ru old/kdb/modules/kdbm_vm.c new/kdb/modules/kdbm_vm.c --- old/kdb/modules/kdbm_vm.c 2005-05-03 15:51:13.790225372 -0400 +++ new/kdb/modules/kdbm_vm.c 2005-05-03 16:46:00.465394659 -0400 @@ -83,13 +83,21 @@ kdb_printf("vm_ops->open = 0x%p\n", vp->vm_ops->open); kdb_printf("vm_ops->close = 0x%p\n", vp->vm_ops->close); kdb_printf("vm_ops->nopage = 0x%p\n", vp->vm_ops->nopage); + kdb_printf("vm_ops->populate = 0x%p\n", vp->vm_ops->populate); #ifdef HAVE_VMOP_MPROTECT kdb_printf("vm_ops->mprotect = 0x%p\n", vp->vm_ops->mprotect); #endif +#ifdef CONFIG_NUMA + kdb_printf("vm_ops->set_policy = 0x%p\n", vp->vm_ops->set_policy); + kdb_printf("vm_ops->get_policy = 0x%p\n", vp->vm_ops->get_policy); +#endif } kdb_printf("vm_pgoff = 0x%lx\n", vp->vm_pgoff); kdb_printf("vm_file = 0x%p\n", (void *) vp->vm_file); kdb_printf("vm_private_data = 0x%p\n", vp->vm_private_data); +#ifdef CONFIG_NUMA + kdb_printf("vm_policy = 0x%p\n", vp->vm_policy); +#endif return 0; } @@ -115,6 +123,203 @@ return 0; } + +#ifdef CONFIG_NUMA +#include + +/* + * kdbm_mpol + * + * This function implements the 'mempolicy' command. + * Print a struct mempolicy. + * + * mempolicy
Print struct mempolicy at
+ */ +static int +kdbm_mpol(int argc, const char **argv, const char **envp, struct pt_regs *regs) +{ + unsigned long addr; + long offset = 0; + int nextarg; + int err = 0; + struct mempolicy *mp = NULL; + + if (argc != 1) + return KDB_ARGCOUNT; + + nextarg = 1; + if ((err = kdbgetaddrarg(argc, argv, &nextarg, &addr, &offset, + NULL, regs)) != 0) + return(err); + + if (!(mp = kmalloc(sizeof(*mp), GFP_ATOMIC))) { + kdb_printf("%s: cannot kmalloc mp\n", __FUNCTION__); + goto out; + } + + if ((err = kdb_getarea(*mp, addr))) { + kdb_printf("%s: invalid mempolicy address\n", __FUNCTION__); + goto out; + } + + kdb_printf("struct mempolicy at 0x%p\n", (struct mempolicy *)addr); + kdb_printf(" refcnt %d\n", atomic_read(&mp->refcnt)); + + switch (mp->policy) { + case MPOL_DEFAULT: + kdb_printf(" policy %d (MPOL_DEFAULT)\n", mp->policy); + break; + + case MPOL_PREFERRED: + kdb_printf(" policy %d (MPOL_PREFERRED)\n", mp->policy); + kdb_printf(" preferred_node %d\n", mp->v.preferred_node); + break; + + case MPOL_BIND: + { + int i, z; + struct zone *zonep; + pg_data_t *pgdatp; + + kdb_printf(" policy %d (MPOL_BIND)\n", mp->policy); + kdb_printf(" zonelist 0x%p\n", mp->v.zonelist); + + for (i = 0; mp->v.zonelist; i++) { + zonep = mp->v.zonelist->zones[i]; + if (!zonep) { + break; + } + kdb_printf(" 0x%p", zonep); + for_each_pgdat(pgdatp) { + for (z = 0; z < MAX_NR_ZONES; z++) { + if (zonep == &pgdatp->node_zones[z]) { + kdb_printf(" (node %d node_zones[%d])", + pgdatp->node_id, z); + break; + } + } + if (z != MAX_NR_ZONES) + break; /* found it */ + } + kdb_printf("\n"); + } + break; + } + + case MPOL_INTERLEAVE: + { + int i, nlongs; + unsigned long *longp; + + kdb_printf(" policy %d (MPOL_INTERLEAVE)\n", mp->policy); + nlongs = (int)BITS_TO_LONGS(MAX_NUMNODES); + kdb_printf(" nodes:"); + longp = mp->v.nodes; + for (i = 0; i < nlongs; i++, longp++) + kdb_printf(" 0x%lx ", *longp); + kdb_printf("\n"); + break; + } + + default: + kdb_printf(" policy %d (unknown)\n", mp->policy); + break; + } +out: + if (mp) + kfree(mp); + return err; +} + +#endif /* CONFIG_NUMA */ + +/* + * kdbm_pgdat + * + * This function implements the 'pgdat' command. + * Print a struct pglist_data (pg_data_t). + * + * pgdat Print struct pglist_data for node . + * + * Print pglist_data for node 0 if node_id not specified, + * or print the one pglist_data structure if !CONFIG_NUMA. + */ +static int +kdbm_pgdat(int argc, const char **argv, const char **envp, struct pt_regs *regs) +{ + int i, err = 0, node_id = 0; + pg_data_t *pgdatp = NULL; + +#ifdef CONFIG_NUMA + if (argc > 1) + return KDB_ARGCOUNT; + if (argc == 1) { + int nextarg; + long offset = 0; + unsigned long node_id_ul; + + nextarg = 1; + if ((err = kdbgetaddrarg(argc, argv, &nextarg, &node_id_ul, + &offset, NULL, regs)) != 0) { + return(err); + } + node_id = (int)node_id_ul; + } +#endif + for_each_pgdat(pgdatp) { + if (pgdatp->node_id == node_id) + break; + } + if (!pgdatp) { + kdb_printf("%s: specified node not found\n", __FUNCTION__); + return 0; + } + kdb_printf("struct pglist_data at 0x%p node_id = %d\n", + pgdatp, pgdatp->node_id); + + for (i = 0; i < GFP_ZONETYPES; i++) { + struct zone **zonep; + + zonep = pgdatp->node_zonelists[i].zones; + kdb_printf(" node_zonelists[%d] at 0x%p\n", i, zonep); + + while (1) { + int z; + pg_data_t *tmp_pgdatp; + + if (*zonep == NULL) + break; + kdb_printf(" 0x%p", *zonep); + for_each_pgdat(tmp_pgdatp) { + for (z = 0; z < MAX_NR_ZONES; z++) { + if (*zonep == &tmp_pgdatp->node_zones[z]) { + kdb_printf(" (node %d node_zones[%d])", + tmp_pgdatp->node_id, z); + break; + } + } + if (z != MAX_NR_ZONES) + break; /* found it */ + } + kdb_printf("\n"); + zonep++; + } + } + kdb_printf(" nr_zones = %d", pgdatp->nr_zones); + kdb_printf(" node_mem_map = 0x%p\n", pgdatp->node_mem_map); + kdb_printf(" bdata = 0x%p", pgdatp->bdata); + kdb_printf(" node_start_pfn = 0x%lx\n", pgdatp->node_start_pfn); + kdb_printf(" node_present_pages = %ld (0x%lx)\n", + pgdatp->node_present_pages, pgdatp->node_present_pages); + kdb_printf(" node_spanned_pages = %ld (0x%lx)\n", + pgdatp->node_spanned_pages, pgdatp->node_spanned_pages); + kdb_printf(" kswapd = 0x%p\n", pgdatp->kswapd); + kdb_printf(" pgdat_next = 0x%p\n", pgdatp->pgdat_next); + + return err; +} + + /* * kdbm_vm * @@ -648,6 +853,12 @@ { kdb_register("vm", kdbm_vm, "[-v] ", "Display vm_area_struct", 0); kdb_register("vmp", kdbm_vm, "[-v] ", "Display all vm_area_struct for ", 0); +#ifdef CONFIG_NUMA + kdb_register("mempolicy", kdbm_mpol, "", "Display mempolicy structure", 0); + kdb_register("pgdat", kdbm_pgdat, "", "Display pglist_data node structure", 0); +#else + kdb_register("pgdat", kdbm_pgdat, "", "Display pglist_data node structure", 0); +#endif kdb_register("pte", kdbm_pte, "( -m | -p ) []", "Display pte_t for mm_struct or pid", 0); kdb_register("dentry", kdbm_dentry, "", "Display interesting dentry stuff", 0); kdb_register("filp", kdbm_fp, "", "Display interesting filp stuff", 0); @@ -663,6 +874,10 @@ { kdb_unregister("vm"); kdb_unregister("vmp"); +#ifdef CONFIG_NUMA + kdb_unregister("mempolicy"); +#endif + kdb_unregister("pgdat"); kdb_unregister("pte"); kdb_unregister("dentry"); kdb_unregister("filp"); --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From kaos@sgi.com Wed May 4 16:35:14 2005 Received: with ECARTIS (v1.0.0; list kdb); Wed, 04 May 2005 16:35:17 -0700 (PDT) Received: from larry.melbourne.sgi.com (mverd138.asia.info.net [61.14.31.138]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j44NZ87J018479 for ; Wed, 4 May 2005 16:35:10 -0700 Received: from mail.ocs.com.au (kao1.melbourne.sgi.com [134.14.55.179]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA10937 for ; Thu, 5 May 2005 09:34:52 +1000 Received: from ocs3.ocs.com.au (ocs3.ocs.com.au [192.168.255.3]) by mail.ocs.com.au (Postfix) with ESMTP id 9A3821800AA; Thu, 5 May 2005 09:34:43 +1000 (EST) Received: by ocs3.ocs.com.au (Postfix, from userid 16331) id F045570010B; Thu, 5 May 2005 09:34:36 +1000 (EST) Received: from ocs3.ocs.com.au (localhost [127.0.0.1]) by ocs3.ocs.com.au (Postfix) with ESMTP id ECA311000D8; Thu, 5 May 2005 09:34:36 +1000 (EST) X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: john.blackwood@ccur.com Cc: kdb@oss.sgi.com Subject: Re: some possible NUMA K8 kdb enhancements In-reply-to: Your message of "Wed, 04 May 2005 09:44:16 -0400." <4278D1B0.9070907@ccur.com> Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Date: Thu, 05 May 2005 09:34:36 +1000 Message-ID: <18788.1115249676@ocs3.ocs.com.au> Content-Transfer-Encoding: 8bit X-archive-position: 950 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: kdb On Wed, 04 May 2005 09:44:16 -0400, John Blackwood wrote: >I have a few enhancement suggestions to kdb mainly for adding in some >additional support for NUMA_K8 kernels. These changes seem to working >fine for us, and we hoped that you might be interested in making use of >some or all of the changes shown below. At a quick glance, the changes look reasonable. From the context, I assume that the code was tested on x86_64? Have you tried the patches on other NUMA boxes, like ia64? --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From sdake@mvista.com Wed May 4 18:29:54 2005 Received: with ECARTIS (v1.0.0; list kdb); Wed, 04 May 2005 18:29:58 -0700 (PDT) Received: from zipcode.az.mvista.com (rav-az.mvista.com [65.200.49.157]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j451Tr7J024812 for ; Wed, 4 May 2005 18:29:54 -0700 Received: from persist.az.mvista.com (persist.az.mvista.com [10.50.1.87]) by zipcode.az.mvista.com (8.9.3/8.9.3) with ESMTP id TAA15051 for ; Wed, 4 May 2005 19:02:36 -0700 Subject: Question about kdb code From: Steven Dake Reply-To: sdake@mvista.com To: kdb@oss.sgi.com Content-type: text/plain Organization: MontaVista Software, Inc. Message-Id: <1115256578.12204.127.camel@persist.az.mvista.com> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.6 Date: Wed, 04 May 2005 18:29:38 -0700 Content-Transfer-Encoding: 8bit X-archive-position: 951 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: sdake@mvista.com Precedence: bulk X-list: kdb Keith and the list... Using kdb 4.4... I am working on taking some of Ingo's realtime patches and getting KDB to work properly with them. The first of this step is the CONFIG_PREEMPT_DESKTOP, which is also in standard kernel.org. I had a question about the purpose of disabling/enabling the bottom halves in kdb(). I find that if I have this code in the tree, and then run "ss" and then "go" the system preemption count gets skewed. This causes all kinds of bad mojo with the scheduler and basically breaks the system. I tracked the preemption to a local_bh_disable that is done when single step is started. But it is not undone when kdb returns, which causes an interrupt preemption count to get off track in the 8250 serial interrupt handler. I am not sure, but this problem may be visible on standard kernel.org with standard kdb 4.4 with 8250 serial enabled when running with preemption. I took out the local_bh_disable/local_bh_enable entirely from kdb() function, and single step works like a champ. My question is, what is the purpose of that usage? Just to shut off softirq processing while KDB is running? The root of the problem is that the softirq preemption count is overloaded to specify when bh's should be disabled, but the softirq preemption count is used by other parts of the scheduler without this overload in mind. I was thinking of hacking in an additional flag which identifies if "bottom halves are disabled" especially for kdb without setting the overloaded preemption count value. Do you have other ideas? Is the disable/enable really needed? Why not reenable on single step? Thanks -steve --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Dave@sbs.siemens.co.uk Thu May 5 01:11:36 2005 Received: with ECARTIS (v1.0.0; list kdb); Thu, 05 May 2005 01:11:38 -0700 (PDT) Received: from cpe-71-11-87-223.hol.nc.charter.com (cpe-71-11-87-223.hol.nc.charter.com [71.11.87.223]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j458BU7J026771 for ; Thu, 5 May 2005 01:11:35 -0700 Message-ID: From: order To: kdb@oss.sgi.com Subject: =?windows-1251?B?zvLi5fIg7eAg7+jx/Ozu?= Date: Thu, 05 May 2005 07:57:39 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=windows-1251 Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express V6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-archive-position: 952 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Dave@sbs.siemens.co.uk Precedence: bulk X-list: kdb ...............тел. (905) 2039072................ ...............тел. (095) 1363041................ .................ася 292593973................... Вы заинтересованы в увеличении количества клиентов? Обращаем Ваше внимание на то, что наша компания помогает решать подобные задачи стоящие перед Вашим бизнесом. Способ решения ваших проблем – электронные письма в сети Интернет! ........................................................ ... Просто, легко, практично, конкретно, быстро! ... БОДИТЕЛЬНАЯ Барбара, БЕЙСИК Басни, Бартерная Байт --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From john.blackwood@ccur.com Thu May 5 04:42:08 2005 Received: with ECARTIS (v1.0.0; list kdb); Thu, 05 May 2005 04:42:14 -0700 (PDT) Received: from flmx.iccur.com (mail.ccur.com [208.248.32.212]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j45Bg87J016864 for ; Thu, 5 May 2005 04:42:08 -0700 Received: from ccur.com ([129.75.14.2]) by flmx.iccur.com with Microsoft SMTPSVC(5.0.2195.6713); Thu, 5 May 2005 07:41:52 -0400 Message-ID: <427A067F.5000905@ccur.com> Date: Thu, 05 May 2005 07:41:51 -0400 From: John Blackwood Reply-To: john.blackwood@ccur.com Organization: Concurrent Computer Corporation User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4.3) Gecko/20050104 Red Hat/1.4.3-3.0.7 X-Accept-Language: en-us, en MIME-Version: 1.0 To: kdb@oss.sgi.com Subject: Re: some possible NUMA K8 kdb enhancements Content-type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 05 May 2005 11:41:52.0815 (UTC) FILETIME=[6E8D1BF0:01C55167] X-archive-position: 953 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: john.blackwood@ccur.com Precedence: bulk X-list: kdb Oops... I forgot to copy kdb@oss.sgi.com with my email reply to Keith: ----------- Keith Owens wrote: > On Wed, 04 May 2005 09:44:16 -0400, John Blackwood wrote: > >> I have a few enhancement suggestions to kdb mainly for adding in some >> additional support for NUMA_K8 kernels. These changes seem to working >> fine for us, and we hoped that you might be interested in making use of >> some or all of the changes shown below. Hi Keith, Thanks for your reply. > At a quick glance, the changes look reasonable. From the context, I > assume that the code was tested on x86_64? Yes. > Have you tried the patches on other NUMA boxes, like ia64? I'm afraid not. I only have access to opteron x86_64 boxes. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From kaos@sgi.com Thu May 5 16:47:04 2005 Received: with ECARTIS (v1.0.0; list kdb); Thu, 05 May 2005 16:47:15 -0700 (PDT) Received: from larry.melbourne.sgi.com (mverd138.asia.info.net [61.14.31.138]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j45Nl1Ov013610 for ; Thu, 5 May 2005 16:47:02 -0700 Received: from mail.ocs.com.au (kao1.melbourne.sgi.com [134.14.55.179]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA10562 for ; Fri, 6 May 2005 09:46:43 +1000 Received: from ocs3.ocs.com.au (ocs3.ocs.com.au [192.168.255.3]) by mail.ocs.com.au (Postfix) with ESMTP id 070C61800B5; Fri, 6 May 2005 09:46:39 +1000 (EST) Received: by ocs3.ocs.com.au (Postfix, from userid 16331) id E739E70010B; Fri, 6 May 2005 09:46:38 +1000 (EST) Received: from ocs3.ocs.com.au (localhost [127.0.0.1]) by ocs3.ocs.com.au (Postfix) with ESMTP id E3BA31000D8; Fri, 6 May 2005 09:46:38 +1000 (EST) X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: john.blackwood@ccur.com Cc: kdb@oss.sgi.com Subject: Re: some possible NUMA K8 kdb enhancements In-reply-to: Your message of "Thu, 05 May 2005 07:41:51 -0400." <427A067F.5000905@ccur.com> Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Date: Fri, 06 May 2005 09:46:38 +1000 Message-ID: <11998.1115336798@ocs3.ocs.com.au> Content-Transfer-Encoding: 8bit X-archive-position: 954 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: kdb On Thu, 05 May 2005 07:41:51 -0400, John Blackwood wrote: >Oops... I forgot to copy kdb@oss.sgi.com with my email reply to Keith: > >----------- > >Keith Owens wrote: > > > On Wed, 04 May 2005 09:44:16 -0400, John Blackwood > wrote: > > > >> I have a few enhancement suggestions to kdb mainly for adding in some > >> additional support for NUMA_K8 kernels. These changes seem to working > >> fine for us, and we hoped that you might be interested in making use of > >> some or all of the changes shown below. > > >Hi Keith, > >Thanks for your reply. > > > At a quick glance, the changes look reasonable. From the context, I > > assume that the code was tested on x86_64? > > >Yes. > > > Have you tried the patches on other NUMA boxes, like ia64? > > >I'm afraid not. I only have access to opteron x86_64 boxes. That's fine, I will test it on ia64 this weekend. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From john.blackwood@ccur.com Fri May 6 04:37:47 2005 Received: with ECARTIS (v1.0.0; list kdb); Fri, 06 May 2005 04:37:51 -0700 (PDT) Received: from flmx.iccur.com (mail.ccur.com [208.248.32.212]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j46BblOv003423 for ; Fri, 6 May 2005 04:37:47 -0700 Received: from ccur.com ([129.75.14.2]) by flmx.iccur.com with Microsoft SMTPSVC(5.0.2195.6713); Fri, 6 May 2005 07:37:29 -0400 Message-ID: <427B56F8.9060900@ccur.com> Date: Fri, 06 May 2005 07:37:28 -0400 From: John Blackwood Reply-To: john.blackwood@ccur.com Organization: Concurrent Computer Corporation User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4.4) Gecko/20050318 Red Hat/1.4.4-1.3.5 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Keith Owens CC: kdb@oss.sgi.com Subject: Re: some possible NUMA K8 kdb enhancements References: <11998.1115336798@ocs3.ocs.com.au> In-Reply-To: <11998.1115336798@ocs3.ocs.com.au> Content-type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit X-OriginalArrivalTime: 06 May 2005 11:37:29.0943 (UTC) FILETIME=[FC47C670:01C5522F] X-archive-position: 955 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: john.blackwood@ccur.com Precedence: bulk X-list: kdb Keith Owens wrote: > On Thu, 05 May 2005 07:41:51 -0400, > John Blackwood wrote: > >>Oops... I forgot to copy kdb@oss.sgi.com with my email reply to Keith: >> >>----------- >> >>Keith Owens wrote: >> >> >>>On Wed, 04 May 2005 09:44:16 -0400, John Blackwood >> >> wrote: >> >>>>I have a few enhancement suggestions to kdb mainly for adding in some >>>>additional support for NUMA_K8 kernels. These changes seem to working >>>>fine for us, and we hoped that you might be interested in making use of >>>>some or all of the changes shown below. >> >> >>Hi Keith, >> >>Thanks for your reply. >> >> >>>At a quick glance, the changes look reasonable. From the context, I >>>assume that the code was tested on x86_64? >> >> >>Yes. >> >> >>>Have you tried the patches on other NUMA boxes, like ia64? >> >> >>I'm afraid not. I only have access to opteron x86_64 boxes. > > > That's fine, I will test it on ia64 this weekend. > Thanks! I hope it goes o.k. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From null@maillist.ru Sat May 7 15:47:24 2005 Received: with ECARTIS (v1.0.0; list kdb); Sat, 07 May 2005 15:47:27 -0700 (PDT) Received: from cpe-71-11-85-164.hol.nc.charter.com (cpe-71-11-85-164.hol.nc.charter.com [71.11.85.164]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j47MlIOv007458 for ; Sat, 7 May 2005 15:47:23 -0700 Message-ID: <2b2d01c55355$593f49f4$3c09da4d@maillist.ru> From: order To: kdb@oss.sgi.com Subject: =?windows-1251?B?0+Ll6+j35e3o5SD36PHr4CDv7urz7+Dy5evl6Q==?= Date: Sat, 07 May 2005 22:40:24 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=windows-1251 Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express V6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-archive-position: 956 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: null@maillist.ru Precedence: bulk X-list: kdb - Как у Вас обстоят дела с вашими клиентами? - Не слышно звонков? - Упали продажи? + Обратитесь к нам, + и мы обеспечим хороший поток новых клиентов и заказов. _________________ ICQ 292-593-973 TEL +7(905) 2039072 TEL +7(926) 21~41~617 Акраганта Аландские, Бакинская Азимут, Ахматова Антиурбанизм --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Mon May 9 01:33:44 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 09 May 2005 01:33:47 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j498XcOv014609 for ; Mon, 9 May 2005 01:33:39 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 7A0B319D906; Mon, 9 May 2005 10:33:04 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 12085-09; Mon, 9 May 2005 10:33:03 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id AC10C19D909; Mon, 9 May 2005 10:33:01 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005050910433128:569 ; Mon, 9 May 2005 10:43:31 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id 811D91392D; Mon, 9 May 2005 10:32:54 +0200 (CEST) Message-ID: <427F2036.506@bull.net> Date: Mon, 09 May 2005 10:32:54 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: KDB improvements for IA64 X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 09/05/2005 10:43:31, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 09/05/2005 10:43:39 Content-type: text/plain Content-Transfer-Encoding: 8bit X-archive-position: 957 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb Dear all, I have been working during the last months for improving KDB product on IA-64 architecture. The version I propose to you here solves different problems and implements new features such as : - breakpoint misfunctioning on SMP machines The global KDB entry and exit sequences have been deeply modified. The flag set has been reduced to the minimum and spinlock usage had increased. - "ssb" command is now implemented for IA64. Step to taken branch feature did not exist on the SGI KDB version. It is working by now, using the taken banch bit in the control register. - timeouts when exiting from KDB will never occur I have noticed that after a long time in KDB, some drivers enter in timeout sequences. Now, the different ITC values are saved when entering KDB and restored when going out. I have done many tests on Intel Tiger and Bull Novascale machines and the results I obtain seem to prove that this version is now stable. That is the reason why I invite you to test it. Do not to hesitate to give me your feedback and your comments. However, the modifications I have done only work for IA64 based machines. Since I have no IA32 machine to modify and test it, I can not do anything for this architecture, unless you can give me a PC. Best regards, Francois WELLENREITER -- Binary/unsupported file stripped by Ecartis -- -- Type: application/x-bzip2 -- File: kdb-v4.4-2.6.12-rc2-common_BULL.patch.bz2 -- Binary/unsupported file stripped by Ecartis -- -- Type: application/x-bzip2 -- File: kdb-v4.4-2.6.12-rc2-ia64_BULL.patch.bz2 --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Mon May 9 01:54:04 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 09 May 2005 01:54:06 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j498s3Ov015388 for ; Mon, 9 May 2005 01:54:03 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id D877119D909 for ; Mon, 9 May 2005 10:53:33 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 13458-05 for ; Mon, 9 May 2005 10:53:32 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id D5ED619D906 for ; Mon, 9 May 2005 10:53:32 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005050911040544:672 ; Mon, 9 May 2005 11:04:05 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id B6A101392D for ; Mon, 9 May 2005 10:53:25 +0200 (CEST) Message-ID: <427F2505.9020502@bull.net> Date: Mon, 09 May 2005 10:53:25 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 References: <427F2036.506@bull.net> In-Reply-To: <427F2036.506@bull.net> X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 09/05/2005 11:04:05, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 09/05/2005 11:04:10, Serialize complete at 09/05/2005 11:04:10 Content-Transfer-Encoding: 8bit Content-type: text/plain; charset=us-ascii X-archive-position: 958 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb Sorry, but how is it possible to publish the patches without the files being automatically stripped ? I have tried to send them 4 days ago without bzip'ing them but the mail never arrived. Thanks for help, Francois > -- Binary/unsupported file stripped by Ecartis -- > -- Type: application/x-bzip2 > -- File: kdb-v4.4-2.6.12-rc2-common_BULL.patch.bz2 > > > -- Binary/unsupported file stripped by Ecartis -- > -- Type: application/x-bzip2 > -- File: kdb-v4.4-2.6.12-rc2-ia64_BULL.patch.bz2 --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From kaos@sgi.com Mon May 9 06:02:28 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 09 May 2005 06:02:42 -0700 (PDT) Received: from larry.melbourne.sgi.com (mverd138.asia.info.net [61.14.31.138]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j49D2POv003265 for ; Mon, 9 May 2005 06:02:26 -0700 Received: from mail.ocs.com.au (kao1.melbourne.sgi.com [134.14.55.179]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id XAA07728 for ; Mon, 9 May 2005 23:01:59 +1000 Received: from ocs3.ocs.com.au (ocs3.ocs.com.au [192.168.255.3]) by mail.ocs.com.au (Postfix) with ESMTP id BB34D1800A2; Mon, 9 May 2005 23:01:55 +1000 (EST) Received: by ocs3.ocs.com.au (Postfix, from userid 16331) id 2AEE1707F8A; Mon, 9 May 2005 23:01:45 +1000 (EST) Received: from ocs3.ocs.com.au (localhost [127.0.0.1]) by ocs3.ocs.com.au (Postfix) with ESMTP id 1646A1000D8; Mon, 9 May 2005 23:01:44 +1000 (EST) X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: Francois Wellenreiter Cc: kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 In-reply-to: Your message of "Mon, 09 May 2005 10:53:25 +0200." <427F2505.9020502@bull.net> Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Date: Mon, 09 May 2005 23:01:44 +1000 Message-ID: <6562.1115643704@ocs3.ocs.com.au> Content-Transfer-Encoding: 8bit X-archive-position: 959 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: kdb On Mon, 09 May 2005 10:53:25 +0200, Francois Wellenreiter wrote: > > Sorry, > >but how is it possible to publish the patches without >the files being automatically stripped ? >I have tried to send them 4 days ago without bzip'ing them >but the mail never arrived. Francois, send them direct to kaos@sgi.com, that will bypass the mailing list software which has its own idea of what is acceptable[*]. I will put the patches up on oss.sgi.com. [*] Don't blame, me, I have very little control over the mailing list software. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Tue May 10 08:27:38 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 10 May 2005 08:27:43 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4AFRbOv029392 for ; Tue, 10 May 2005 08:27:38 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id AFC7119D90D; Tue, 10 May 2005 17:27:05 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 08602-05; Tue, 10 May 2005 17:27:04 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id AD99B19D90B; Tue, 10 May 2005 17:27:04 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005051017374233:4127 ; Tue, 10 May 2005 17:37:42 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id 1A4A81392D; Tue, 10 May 2005 17:27:02 +0200 (CEST) Message-ID: <4280D2C5.3060604@bull.net> Date: Tue, 10 May 2005 17:27:01 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: KDB improvements for IA64 X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 10/05/2005 17:37:42, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 10/05/2005 17:37:43, Serialize complete at 10/05/2005 17:37:43 Content-Transfer-Encoding: 8bit Content-type: text/plain; charset=us-ascii X-archive-position: 960 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb Dear all, I have been working during the last months for improving KDB product on IA-64 architecture. The version I propose to you here solves different problems and implements new features such as : - breakpoint misfunctioning on SMP machines The global KDB entry and exit sequences have been deeply modified. The flag set has been reduced to the minimum and spinlock usage had increased. - "ssb" command is now implemented for IA64. Step to taken branch feature did not exist on the SGI KDB version. It is working by now, using the taken banch bit in the control register. - timeouts when exiting from KDB will never occur I have noticed that after a long time in KDB, some drivers enter in timeout sequences. Now, the different ITC values are saved when entering KDB and restored when going out. I have done many tests on Bull Novascale machines and the results that I obtain seem to prove that this version is now stable. That is the reason why I invite you to test it. Do not to hesitate to give me your feedback and your comments. However, the modifications I have done only work for IA64 based machines. Since I have no IA32 machine to modify and test it, I can not do anything for this architecture, unless you can give me a PC. The patches can be downloaded at the following address : http://www.bullopensource.org/site_kdb/source.html I plan to add some explanation pages in a few time to describe more precisely the modifications I have done in KDB. Best regards, Francois WELLENREITER --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From davidm@napali.hpl.hp.com Tue May 10 11:26:54 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 10 May 2005 11:26:59 -0700 (PDT) Received: from palrel10.hp.com (palrel10.hp.com [156.153.255.245]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4AIQoOv006718 for ; Tue, 10 May 2005 11:26:54 -0700 Received: from hplms2.hpl.hp.com (hplms2.hpl.hp.com [15.0.152.33]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by palrel10.hp.com (Postfix) with ESMTP id F1F1BADE; Tue, 10 May 2005 11:26:22 -0700 (PDT) Received: from napali.hpl.hp.com (napali.hpl.hp.com [15.4.89.123]) by hplms2.hpl.hp.com (8.13.1/8.13.1/HPL-PA Hub) with ESMTP id j4AIQL1l027639; Tue, 10 May 2005 11:26:21 -0700 (PDT) Received: from napali.hpl.hp.com (napali [127.0.0.1]) by napali.hpl.hp.com (8.13.3/8.13.3/Debian-9) with ESMTP id j4AIQIrQ015004; Tue, 10 May 2005 11:26:18 -0700 Received: (from davidm@localhost) by napali.hpl.hp.com (8.13.3/8.13.3/Submit) id j4AIQHZO015001; Tue, 10 May 2005 11:26:17 -0700 From: David Mosberger MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit Message-ID: <17024.64713.827619.735000@napali.hpl.hp.com> Date: Tue, 10 May 2005 11:26:17 -0700 To: Francois Wellenreiter Cc: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 In-Reply-To: <4280D2C5.3060604@bull.net> References: <4280D2C5.3060604@bull.net> X-Mailer: VM 7.19 under Emacs 21.4.1 Reply-To: davidm@hpl.hp.com X-URL: http://www.hpl.hp.com/personal/David_Mosberger/ X-archive-position: 961 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: davidm@napali.hpl.hp.com Precedence: bulk X-list: kdb >>>>> On Tue, 10 May 2005 17:27:01 +0200, Francois Wellenreiter said: Francois> - timeouts when exiting from KDB will never occur I have Francois> noticed that after a long time in KDB, some drivers enter Francois> in timeout sequences. Now, the different ITC values are Francois> saved when entering KDB and restored when going out. This sounds fragile to me. For one thing, the kernel may not use ITC at all for timeouts (true for most NUMA platforms). Also, even if the ITC is in use, you'd probably have to resync them on the slave CPUs after restoring the time-keeper's ITC value. Not to mention that if the timeout is related to an external event, then faking time won't help no matter what. --david --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From kaos@sgi.com Tue May 10 21:41:30 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 10 May 2005 21:41:35 -0700 (PDT) Received: from larry.melbourne.sgi.com (mverd138.asia.info.net [61.14.31.138]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j4B4fOOv012210 for ; Tue, 10 May 2005 21:41:25 -0700 Received: from kao2.melbourne.sgi.com (kao2.melbourne.sgi.com [134.14.55.180]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id OAA01692 for ; Wed, 11 May 2005 14:40:58 +1000 Received: by kao2.melbourne.sgi.com (Postfix, from userid 16331) id 5A1ED707F8A; Wed, 11 May 2005 14:40:58 +1000 (EST) Received: from kao2.melbourne.sgi.com (localhost [127.0.0.1]) by kao2.melbourne.sgi.com (Postfix) with ESMTP id 56A671000D8; Wed, 11 May 2005 14:40:58 +1000 (EST) X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: Francois Wellenreiter Cc: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 In-reply-to: Your message of "Mon, 09 May 2005 10:32:54 +0200." <427F2036.506@bull.net> Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Date: Wed, 11 May 2005 14:40:58 +1000 Message-ID: <10479.1115786458@kao2.melbourne.sgi.com> Content-Transfer-Encoding: 8bit X-archive-position: 962 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: kdb On Mon, 09 May 2005 10:32:54 +0200, Francois Wellenreiter wrote: > I have been working during the last months for improving KDB product >on IA-64 architecture. > >The version I propose to you here solves different problems and implements new >features such as : > >- breakpoint misfunctioning on SMP machines > > The global KDB entry and exit sequences have been deeply modified. >The flag set has been reduced to the minimum and spinlock usage had increased. > >- "ssb" command is now implemented for IA64. > Step to taken branch feature did not exist on the SGI KDB version. >It is working by now, using the taken banch bit in the control register. > >- timeouts when exiting from KDB will never occur > I have noticed that after a long time in KDB, some drivers enter >in timeout sequences. Now, the different ITC values are saved when entering KDB >and restored when going out. > >I have done many tests on Intel Tiger and Bull Novascale machines and >the results I obtain seem to prove that this version is now stable. > >That is the reason why I invite you to test it. Do not to hesitate to give me >your feedback and your comments. > >However, the modifications I have done only work for IA64 based machines. >Since I have no IA32 machine to modify and test it, I can not do anything >for this architecture, unless you can give me a PC. I have compared your versions against the standard kdb v4.4 2.6.12-rc2 patches and gone through your changes. Unfortunately you lumped them all together which made it very difficult to work out what you were changing. AFAICT, you have these changes :- * Lots of whitespace changes, including changing indentation, reformatting function calls, reformatting function declarations, converting tabs to spaces and even adding whitespace to the end of lines. Never change whitespace at the same time as making other changes, it makes it extremely difficult to see what has really changed. * Saving and restoring itc around kdb. That only works for architectures where you can overwrite the source of get_cycles(). On i386, TSC is read only so this approach does not work there, and I do not like workarounds that only work on some architectures. If there are time critical bits of code that cannot tolerate long delays then the solution is to provide a mechanism for all debug and dump code (not just kdb) to reset the timers at the OS level, not at the hardware level. * Variables and functions got renamed. Why? * The entire kdb state diagram was redone. That may or may not be useful, but without any discussion on what you are trying to achieve, I cannot tell if the new state diagram is right. I do know that you deleted some states that are still required * Every registered command get a new int * parameter, which is unused in all but one command. I cannot even tell what you are trying to do with that extra option, it seems to be tied into the undocumented state diagram changes. That change will break every other piece of code that adds its own commands to kdb. A lot of subsystems take advantage of kdb to provide the debugging fraemwork, breaking the ABI like this is not acceptable. * The meaning of KDB_ENTER() was changed. The new version will cause problems when KDB_ENTER() is used on paths that kdb also uses, you will get recursive entry and deadlock * The test for WARN_CONSOLE_UNLOCKED() was changed. I expect that to generate lots of spurious warning messages on VGA consoles. * kdb_notifier_list was removed. That is used by some subsystems to get notification of entry to and exit from kdb. It is required. * Support for the ppc64 and sparc64 consoles was removed from kdb_printf(). That support is required. * You removed all the "pause output after LINES lines" code and the associated CMD_INTERRUPT flag. That code is the only thing that catches output from runaway commands, it is required. * Bits of the catastrophic error handling were deleted, it is now incomplete. * XScale support has been removed from the main kdb control loop. * Breakpoint handling has been redone. Alas with all the other changes, I cannt tell which changes are for breakpoint handling so I cannot tell if they work or not. There are far too many changes in the Bull patches; many of these changes are spurious or simply wrong. I am rejecting these patches. If you want to redo the breakpoint handling then that is fine, but make that change on its own and discuss the change on the list first. Keith Owens. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Tue May 10 23:07:30 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 10 May 2005 23:07:34 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4B67SOv015094 for ; Tue, 10 May 2005 23:07:29 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 2827419D912; Wed, 11 May 2005 08:06:57 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 31751-07; Wed, 11 May 2005 08:06:55 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 4298D19D910; Wed, 11 May 2005 08:06:55 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005051108173316:66 ; Wed, 11 May 2005 08:17:33 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id 63CC71392D; Wed, 11 May 2005 08:06:54 +0200 (CEST) Message-ID: <4281A0FE.1030300@bull.net> Date: Wed, 11 May 2005 08:06:54 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: davidm@hpl.hp.com Cc: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 References: <4280D2C5.3060604@bull.net> <17024.64713.827619.735000@napali.hpl.hp.com> In-Reply-To: <17024.64713.827619.735000@napali.hpl.hp.com> X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 08:17:33, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 08:17:35, Serialize complete at 11/05/2005 08:17:35 Content-Transfer-Encoding: 8bit Content-type: text/plain; charset=us-ascii X-archive-position: 963 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb >>>>>>On Tue, 10 May 2005 17:27:01 +0200, Francois Wellenreiter said: > > > Francois> - timeouts when exiting from KDB will never occur I have > Francois> noticed that after a long time in KDB, some drivers enter > Francois> in timeout sequences. Now, the different ITC values are > Francois> saved when entering KDB and restored when going out. > > This sounds fragile to me. For one thing, the kernel may not use ITC > at all for timeouts (true for most NUMA platforms). Also, even if the > ITC is in use, you'd probably have to resync them on the slave CPUs > after restoring the time-keeper's ITC value. Not to mention that if > the timeout is related to an external event, then faking time won't > help no matter what. > > --david > - > To unsubscribe from this list: send the line "unsubscribe linux-ia64" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > David, I agree that ITC management may be a bit risky and overall is platform-dependent, but in my opinion, that was a better way than saving and restoring global xtime variable which can be modified by only one CPU (the main danger in such a case may occur when this CPU is working when the other ones are already running KDB). The time-outs that I have noticed dealt with journalled filesystem messages (and sometimes lead to an unrecoverable machine crash), that is the reason why I have implemented this feature. On the different NUMA machines that I have tested, this problem did not appear again. If you have some implementation ideas, they are really welcome. Regards, Francois --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Tue May 10 23:48:33 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 10 May 2005 23:48:36 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4B6mVOv017294 for ; Tue, 10 May 2005 23:48:32 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id BDFBF19D912; Wed, 11 May 2005 08:47:59 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 05321-08; Wed, 11 May 2005 08:47:57 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 65FC619D919; Wed, 11 May 2005 08:47:40 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005051108581726:146 ; Wed, 11 May 2005 08:58:17 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id C41661392D; Wed, 11 May 2005 08:47:37 +0200 (CEST) Message-ID: <4281AA89.6030700@bull.net> Date: Wed, 11 May 2005 08:47:37 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Keith Owens Cc: linux-ia64@vger.kernel.org, kdb@oss.sgi.com Subject: Re: KDB improvements for IA64 References: <10479.1115786458@kao2.melbourne.sgi.com> In-Reply-To: <10479.1115786458@kao2.melbourne.sgi.com> X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 08:58:17, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 08:58:20, Serialize complete at 11/05/2005 08:58:20 Content-Transfer-Encoding: 8bit Content-type: text/plain; charset=us-ascii X-archive-position: 964 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb Keith, first of all, I want to thank you for taking time on my proposition. Please find my answer inserted in your mail : > I have compared your versions against the standard kdb v4.4 2.6.12-rc2 > patches and gone through your changes. Unfortunately you lumped them > all together which made it very difficult to work out what you were > changing. AFAICT, you have these changes :- > > * Lots of whitespace changes, including changing indentation, > reformatting function calls, reformatting function declarations, > converting tabs to spaces and even adding whitespace to the end of > lines. Never change whitespace at the same time as making other > changes, it makes it extremely difficult to see what has really > changed. Sorry, I didn't take time to configure xemacs correctly. I will try to remove the lines where it has been changed. It may impact one or two functions. > * Saving and restoring itc around kdb. That only works for > architectures where you can overwrite the source of get_cycles(). On > i386, TSC is read only so this approach does not work there, and I do > not like workarounds that only work on some architectures. If there > are time critical bits of code that cannot tolerate long delays then > the solution is to provide a mechanism for all debug and dump code > (not just kdb) to reset the timers at the OS level, not at the > hardware level. Please, do read my answer to David Mosberger about this point. > * Variables and functions got renamed. Why? Because I estimated that in my code their role have changed (due to a state diagram modification). Then, for example : - kdb_initial_cpu to kdb_monarch_cpu, because its role is to be the CPU which runs kdb_local. - kdb_setstep has been split into kdb_setsinglestep and kdb_settakenbranch (to manage "ss" and "ssb" features) > * The entire kdb state diagram was redone. That may or may not be > useful, but without any discussion on what you are trying to achieve, > I cannot tell if the new state diagram is right. I do know that you > deleted some states that are still required Which ones ? In fact, I have done a deep modification in your code because I estimated that there were to many flags describing some useless KDB states and overall which are read and modified without any spinlock protection (which is dangerous on SMP and NUMA machines). > * Every registered command get a new int * parameter, which is unused > in all but one command. I cannot even tell what you are trying to do > with that extra option, it seems to be tied into the undocumented > state diagram changes. That change will break every other piece of > code that adds its own commands to kdb. A lot of subsystems take > advantage of kdb to provide the debugging fraemwork, breaking the ABI > like this is not acceptable. This new parameter is here to allow to pass KDB parameters instead of using global variables. For example, in your code, you have created a global variable which role is only to pass the CPU number chsoen by kdb_cpu function. I think that this way is not elegant at all (but that might just be a different school point of view) and I prefer to pass arguments into a function. > * The meaning of KDB_ENTER() was changed. The new version will cause > problems when KDB_ENTER() is used on paths that kdb also uses, you > will get recursive entry and deadlock You touched another point of my modifications : my KDB version is not reentrant! This is the main reason why your code deadlocked when setting a breakpoint on a really frequently called piece of code (if you do not believe me, just set a breakpoint in default_idle function and try to "go" again). > * The test for WARN_CONSOLE_UNLOCKED() was changed. I expect that to > generate lots of spurious warning messages on VGA consoles. No, kdb_event flag has been completely removed and this version does not encounter this kind of behaviour. > * kdb_notifier_list was removed. That is used by some subsystems to > get notification of entry to and exit from kdb. It is required. OK, I will add it again. It has only been removed because no module used this feature. > * Support for the ppc64 and sparc64 consoles was removed from > kdb_printf(). That support is required. More generally, the modifications that I have done are completely IA-64 oriented. I have no time and no machine to develop and test it on different platforms. This is what I have told you in my first mail. > * You removed all the "pause output after LINES lines" code and the > associated CMD_INTERRUPT flag. That code is the only thing that > catches output from runaway commands, it is required. Yes, I removed this because on the very first versions, I encountered some deadlock conditions. I can set it again. > * Bits of the catastrophic error handling were deleted, it is now > incomplete. Where ? I do not see where I removed CATASTROPHIC bit in my code ? > * XScale support has been removed from the main kdb control loop. Sorry, but I do not know XScale, what is is exactly ? > * Breakpoint handling has been redone. Alas with all the other > changes, I cannt tell which changes are for breakpoint handling so I > cannot tell if they work or not. The changes required for breakpoints were so deep that I had to rework the entire state diagram, KDB entry and exit conditions and some flags that were redundant or useless for me. > There are far too many changes in the Bull patches; many of these > changes are spurious or simply wrong. Maybe, but they work well. > I am rejecting these patches. > If you want to redo the breakpoint handling then that is fine, but make > that change on its own and discuss the change on the list first. > > Keith Owens. Keith, I tried to modify just a few SGI version to let breakpoints work correctly on SMP machines. But the existing state diagram was so complex and uncorrectly spinlock protected that I decided to rework it. Whatever, if you have some time (I know that you are really busy by other developments), I would prefer to discuss on that point first with you and the people would are interested by increasing the robustness of KDB. Best regards, Francois --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Francois.Wellenreiter@bull.net Wed May 11 04:19:21 2005 Received: with ECARTIS (v1.0.0; list kdb); Wed, 11 May 2005 04:19:26 -0700 (PDT) Received: from ecfrec.frec.bull.fr (ecfrec.frec.bull.fr [129.183.4.8]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4BBJJOv002502 for ; Wed, 11 May 2005 04:19:20 -0700 Received: from localhost (localhost [127.0.0.1]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 1736C19D90B; Wed, 11 May 2005 13:18:46 +0200 (CEST) Received: from ecfrec.frec.bull.fr ([127.0.0.1]) by localhost (ecfrec.frec.bull.fr [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 08822-04; Wed, 11 May 2005 13:18:44 +0200 (CEST) Received: from ecn002.frec.bull.fr (ecn002.frec.bull.fr [129.183.4.6]) by ecfrec.frec.bull.fr (Postfix) with ESMTP id 89EEF19D909; Wed, 11 May 2005 13:18:43 +0200 (CEST) Received: from openx1.frec.bull.fr ([129.183.10.3]) by ecn002.frec.bull.fr (Lotus Domino Release 5.0.12) with ESMTP id 2005051113292157:781 ; Wed, 11 May 2005 13:29:21 +0200 Received: from [127.0.0.1] (localhost [127.0.0.1]) by openx1.frec.bull.fr (Postfix) with ESMTP id 722141392D; Wed, 11 May 2005 13:18:41 +0200 (CEST) Message-ID: <4281EA11.3090701@bull.net> Date: Wed, 11 May 2005 13:18:41 +0200 From: Francois Wellenreiter User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3) Gecko/20040913 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Keith Owens , linux-ia64@vger.kernel.org, kdb@oss.sgi.com, davidm@hpl.hp.com Subject: Re: KDB improvements for IA64 References: <10479.1115786458@kao2.melbourne.sgi.com> <4281AA89.6030700@bull.net> In-Reply-To: <4281AA89.6030700@bull.net> X-MIMETrack: Itemize by SMTP Server on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 13:29:21, Serialize by Router on ECN002/FR/BULL(Release 5.0.12 |February 13, 2003) at 11/05/2005 13:29:24, Serialize complete at 11/05/2005 13:29:24 Content-Transfer-Encoding: 8bit Content-type: text/plain; charset=us-ascii X-archive-position: 965 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Francois.Wellenreiter@bull.net Precedence: bulk X-list: kdb My latest mail seems not to have been dispatched in the mailing-lists. Let's try again ! About 70% of my mails are not received on vger.kernel.org, does someone have the same problem ? _______________________ Keith, first of all, I want to thank you for taking time on my proposition. Please find my answer inserted in your mail : > I have compared your versions against the standard kdb v4.4 2.6.12-rc2 > patches and gone through your changes. Unfortunately you lumped them > all together which made it very difficult to work out what you were > changing. AFAICT, you have these changes :- > > * Lots of whitespace changes, including changing indentation, > reformatting function calls, reformatting function declarations, > converting tabs to spaces and even adding whitespace to the end of > lines. Never change whitespace at the same time as making other > changes, it makes it extremely difficult to see what has really > changed. Sorry, I didn't take time to configure xemacs correctly. I will try to remove the lines where it has been changed. It may impact one or two functions. > * Saving and restoring itc around kdb. That only works for > architectures where you can overwrite the source of get_cycles(). On > i386, TSC is read only so this approach does not work there, and I do > not like workarounds that only work on some architectures. If there > are time critical bits of code that cannot tolerate long delays then > the solution is to provide a mechanism for all debug and dump code > (not just kdb) to reset the timers at the OS level, not at the > hardware level. Please, do read my answer to David Mosberger about this point. > * Variables and functions got renamed. Why? Because I estimated that in my code their role have changed (due to a state diagram modification). Then, for example : - kdb_initial_cpu to kdb_monarch_cpu, because its role is to be the CPU which runs kdb_local. - kdb_setstep has been split into kdb_setsinglestep and kdb_settakenbranch (to manage "ss" and "ssb" features) > * The entire kdb state diagram was redone. That may or may not be > useful, but without any discussion on what you are trying to achieve, > I cannot tell if the new state diagram is right. I do know that you > deleted some states that are still required Which ones ? In fact, I have done a deep modification in your code because I estimated that there were to many flags describing some useless KDB states and overall which are read and modified without any spinlock protection (which is dangerous on SMP and NUMA machines). > * Every registered command get a new int * parameter, which is unused > in all but one command. I cannot even tell what you are trying to do > with that extra option, it seems to be tied into the undocumented > state diagram changes. That change will break every other piece of > code that adds its own commands to kdb. A lot of subsystems take > advantage of kdb to provide the debugging fraemwork, breaking the ABI > like this is not acceptable. This new parameter is here to allow to pass KDB parameters instead of using global variables. For example, in your code, you have created a global variable which role is only to pass the CPU number chsoen by kdb_cpu function. I think that this way is not elegant at all (but that might just be a different school point of view) and I prefer to pass arguments into a function. > * The meaning of KDB_ENTER() was changed. The new version will cause > problems when KDB_ENTER() is used on paths that kdb also uses, you > will get recursive entry and deadlock You touched another point of my modifications : my KDB version is not reentrant! This is the main reason why your code deadlocked when setting a breakpoint on a really frequently called piece of code (if you do not believe me, just set a breakpoint in default_idle function and try to "go" again). > * The test for WARN_CONSOLE_UNLOCKED() was changed. I expect that to > generate lots of spurious warning messages on VGA consoles. No, kdb_event flag has been completely removed and this version does not encounter this kind of behaviour. > * kdb_notifier_list was removed. That is used by some subsystems to > get notification of entry to and exit from kdb. It is required. OK, I will add it again. It has only been removed because no module used this feature. > * Support for the ppc64 and sparc64 consoles was removed from > kdb_printf(). That support is required. More generally, the modifications that I have done are completely IA-64 oriented. I have no time and no machine to develop and test it on different platforms. This is what I have told you in my first mail. > * You removed all the "pause output after LINES lines" code and the > associated CMD_INTERRUPT flag. That code is the only thing that > catches output from runaway commands, it is required. Yes, I removed this because on the very first versions, I encountered some deadlock conditions. I can set it again. > * Bits of the catastrophic error handling were deleted, it is now > incomplete. Where ? I do not see where I removed CATASTROPHIC bit in my code ? > * XScale support has been removed from the main kdb control loop. Sorry, but I do not know XScale, what is is exactly ? > * Breakpoint handling has been redone. Alas with all the other > changes, I cannt tell which changes are for breakpoint handling so I > cannot tell if they work or not. The changes required for breakpoints were so deep that I had to rework the entire state diagram, KDB entry and exit conditions and some flags that were redundant or useless for me. > There are far too many changes in the Bull patches; many of these > changes are spurious or simply wrong. Maybe, but they work well. > I am rejecting these patches. > If you want to redo the breakpoint handling then that is fine, but make > that change on its own and discuss the change on the list first. > > Keith Owens. Keith, I tried to modify just a few SGI version to let breakpoints work correctly on SMP machines. But the existing state diagram was so complex and uncorrectly spinlock protected that I decided to rework it. Whatever, if you have some time (I know that you are really busy by other developments), I would prefer to discuss on that point first with you and the people would are interested by increasing the robustness of KDB. Best regards, Francois --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From Bonnie@prosig.demon.co.uk Fri May 13 03:27:36 2005 Received: with ECARTIS (v1.0.0; list kdb); Fri, 13 May 2005 03:27:38 -0700 (PDT) Received: from 203.88.152.155 ([203.88.152.155]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j4DARNOv019457 for ; Fri, 13 May 2005 03:27:33 -0700 Message-ID: From: order To: kdb@oss.sgi.com Subject: =?windows-1251?B?1e7y6PLlIPfy7uH7IOogwuDsIO/w6Pjr6CDt7uL75SAg7+7q8+/g8uXr6D8=?= Date: Fri, 13 May 2005 10:11:08 +0000 MIME-Version: 1.0 Content-type: text/plain; charset=windows-1251 Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express V6.00.2900.2180 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-archive-position: 966 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: Bonnie@prosig.demon.co.uk Precedence: bulk X-list: kdb Электронные рассылки - это, прежде всего донесение рекламы до высокодоходной, активной, инновационной части целевой аудитории. Заинтересовало? ICQ 292593973 TEL +7(905) 203…90…72 Бежала Аквинским, Американском Активен, БОРЬБЫ Арсенальной --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From info@findthedirt5.info Wed May 18 01:36:37 2005 Received: with ECARTIS (v1.0.0; list kdb); Wed, 18 May 2005 01:36:42 -0700 (PDT) Received: from findthedirt5.info (adsl-71-129-127-190.dsl.pltn13.pacbell.net [71.129.127.190]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4I8aaF3007272 for ; Wed, 18 May 2005 01:36:37 -0700 Received: from ethernet ([127.0.0.1]) by findthedirt5.info with Microsoft SMTPSVC(5.0.2195.6713); Wed, 18 May 2005 01:46:09 -0700 Date: Wed, 18 May 2005 01:46:09 -0800 Subject: A user is trying to meet people who know you. To: kdb@oss.sgi.com From: Dirt6 Message-ID: X-OriginalArrivalTime: 18 May 2005 08:46:09.0703 (UTC) FILETIME=[09BCA370:01C55B86] X-archive-position: 967 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: dirt6@findthedirt1.info Precedence: bulk X-list: kdb Research Alert Urgent - A user at our website is sharing opinions and experiences about you at our website. >>>>>>>>>>>>>>>>>>>>>>>>>>>>> To see what the user posted about you access this link - http://3.dirty2.info/lx.php?a=search&b=5&c=kdb@oss.sgi.com ^^^^^^^^^^^^^^^^^^^^^^^^^^ Urgent - If you prefer not to be informed when members post about you at our website in the future, simply add your email address to our Do Not Email List here - http://7.dirty7.info/lx.php?a=donotemail&b=kdb@oss.sgi.com ----------------------------------------------- Sincerely, Support Department --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From vasu.ss@gmail.com Mon May 30 09:02:52 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 30 May 2005 09:02:55 -0700 (PDT) Received: from wproxy.gmail.com (wproxy.gmail.com [64.233.184.204]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id j4UG2pXq007481 for ; Mon, 30 May 2005 09:02:52 -0700 Received: by wproxy.gmail.com with SMTP id 68so2907372wra for ; Mon, 30 May 2005 09:01:58 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=XKal3yym2DlL2w1WHe33LB8Bm5SIXyQQpsleJ+D/9dhIpwEKTtNtNOddeoAKBHR2ZzgnTGKOoOuWVc1YhqzYjLveem1a626ka77sUAsNX3ZZoecsNCRjir2RHHBoCQvM3Eh4cvA42ElsJipqTNGiht8AA0hp/ocXA7NC9q6FWMI= Received: by 10.54.91.3 with SMTP id o3mr5555396wrb; Mon, 30 May 2005 09:01:58 -0700 (PDT) Received: by 10.54.34.18 with HTTP; Mon, 30 May 2005 09:01:58 -0700 (PDT) Message-ID: <2acab652050530090121a088d0@mail.gmail.com> Date: Mon, 30 May 2005 21:31:58 +0530 From: Vasu S Reply-To: Vasu S To: kdb@oss.sgi.com Subject: kdb installation Mime-Version: 1.0 Content-type: text/plain; charset=ISO-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id j4UG2pXq007481 X-archive-position: 968 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: vasu.ss@gmail.com Precedence: bulk X-list: kdb Hi all, I am using fedora 2.6.5 kernel. I tried to install kdb by applying the patches on the oss.cgi.com but could not do it. As there is no support for gdb on fedora distribution, I am thinking of building a custom kernel with kdb support from 2.6.5 vanilla sources available at kernel.org. I want to run this kernel on my fedora box. Will running a vanilla kernel on a fedora box cause any problems? Thanks for paying attention. Vasu. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From kaos@sgi.com Mon May 30 17:23:44 2005 Received: with ECARTIS (v1.0.0; list kdb); Mon, 30 May 2005 17:23:48 -0700 (PDT) Received: from larry.melbourne.sgi.com (mverd138.asia.info.net [61.14.31.138]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j4V0NfXq008387 for ; Mon, 30 May 2005 17:23:42 -0700 Received: from mail.ocs.com.au (kao1.melbourne.sgi.com [134.14.55.179]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA08081 for ; Tue, 31 May 2005 10:22:45 +1000 Received: from ocs3.ocs.com.au (ocs3.ocs.com.au [192.168.255.3]) by mail.ocs.com.au (Postfix) with ESMTP id 9C32918009D; Tue, 31 May 2005 10:22:36 +1000 (EST) Received: by ocs3.ocs.com.au (Postfix, from userid 16331) id 1B1A2707FAA; Tue, 31 May 2005 10:22:30 +1000 (EST) Received: from ocs3.ocs.com.au (localhost [127.0.0.1]) by ocs3.ocs.com.au (Postfix) with ESMTP id 17A541000DA; Tue, 31 May 2005 10:22:30 +1000 (EST) X-Mailer: exmh version 2.6.3_20040314 03/14/2004 with nmh-1.0.4 From: Keith Owens To: Vasu S Cc: kdb@oss.sgi.com Subject: Re: kdb installation In-reply-to: Your message of "Mon, 30 May 2005 21:31:58 +0530." <2acab652050530090121a088d0@mail.gmail.com> Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Date: Tue, 31 May 2005 10:22:30 +1000 Message-ID: <18874.1117498950@ocs3.ocs.com.au> Content-Transfer-Encoding: 8bit X-archive-position: 969 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: kaos@sgi.com Precedence: bulk X-list: kdb On Mon, 30 May 2005 21:31:58 +0530, Vasu S wrote: >I am using fedora 2.6.5 kernel. I tried to install kdb by applying the >patches on the oss.cgi.com but could not do it. RedHat do not want kdb, so SGI has no interest in supporting kdb on RedHat distributions. >As there is no support for gdb on fedora distribution, I am thinking >of building a custom kernel with kdb support from 2.6.5 vanilla >sources available at kernel.org. I want to run this kernel on my >fedora box. Will running a vanilla kernel on a fedora box cause any >problems? None at all. Since you are building your own kernel, use the latest vanilla kernel (2.6.12-rc5) and the corresponding kdb patches. That kernel+kdb combination works fine with the user space packages from FC3. To stop up2date trying to replace your custom kernel, set this line in /etc/sysconfig/rhn/up2date. pkgSkipList=kernel; --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe. From retranslate@kinsellaphoto.com Tue May 31 07:20:45 2005 Received: with ECARTIS (v1.0.0; list kdb); Tue, 31 May 2005 07:20:50 -0700 (PDT) Received: from pool-71-110-145-5.lsanca.dsl-w.verizon.net (pool-71-110-145-5.lsanca.dsl-w.verizon.net [71.110.145.5]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with SMTP id j4VEKiXq011416 for ; Tue, 31 May 2005 07:20:44 -0700 Received: from [67.49.91.219] (port=4028 helo=[Mauritius]) by pool-71-110-145-5.lsanca.dsl-w.verizon.net with esmtp id 21551117285marvelousness38948 for kdb@oss.sgi.com; Tue, 31 May 2005 07:19:49 -0700 Mime-Version: 1.0 (Apple Message framework v728) Content-Transfer-Encoding: 8bit Message-Id: <711773732.171370190@pool-71-110-145-5.lsanca.dsl-w.verizon.net> Content-type: text/plain; charset=US-ASCII To: kdb@oss.sgi.com From: Sophie Subject: 75% Off for All New Software. Date: Tue, 31 May 2005 07:19:48 -0700 X-Mailer: Apple Mail (2.728) X-archive-position: 970 X-ecartis-version: Ecartis v1.0.0 Sender: kdb-bounce@oss.sgi.com Errors-to: kdb-bounce@oss.sgi.com X-original-sender: retranslate@kinsellaphoto.com Precedence: bulk X-list: kdb Get a head start on a new computer career http://trhvmt.4t81734f1emt15m.visional72d3.com The sole advantage of power is that you can do more good. Part of being sane, is being a little bit crazy. --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.