[ewg] cannot install SCST with SRP support
Hi, I'm trying to install SCST with SRP support. I have followed the instructions on the SRPT_README.txt file: So far I have done the following steps: 1. a. git clone git://git.openfabrics.org/~vu/srpt_inc ~/srpt_inc b. Checking out scst's svn development tree revision 245 svn co https://scst.svn.sourceforge.net/svnroot/scst/trunk/scst -r 245 c. cd scst d. patch -p0 ~/srpt_inc/scst_r245.patch e. make and make install On the SCST's Readme file it says: To compile SCST type 'make scst'. It will build SCST itself and its device handlers. To install them type 'make scst_install'. Why it is different then the instructions of the SRPT_README? 2. I have also installed OFED1.3 with srpt. but I cannot load the following modules: modprobe scsi_tgt FATAL: Module scsi_tgt not found. modprobe scst_disk FATAL: Error inserting scst_disk (/lib/modules/2.6.16.21-0.8-smp/extra/dev_handlers/scst_disk.ko): No such device I'm using a x86_64 SUSE Linux Enterprise Server 10 (2.6.16.21-0.8-smp) machine Thanks, Doron ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Re: [ofa-general] Using --build32 on 32 bits machine (OFED-1.3 GA )
Moshe Kazir wrote: If a user uses ofed.conf file include the line build32=y build of libibverbs fails on rh4 u 5 and sles 10 . The following patch fix this problem (also in the attached file ) Moshe --- a/install.pl2008-03-02 06:23:04.0 -0500 +++ b/install.pl2008-03-03 00:59:53.0 -0500 @@ -1463,8 +1463,11 @@ sub supported32bit ###print RED \n32-bit libraries are not supported on this platform, RESET \n if (not $quiet); ###return 0; ###} + if ( $arch eq x86_64 or $arch eq ppc64 or $arch eq ia64 ) { return 1 } +return 0 +} # Check whether compiler $1 exist sub set_compilers Moshe Katzir | +972-9971-8639 (o) | +972-52-860-6042 (m) Voltaire - The Grid Backbone www.voltaire.com Hi Moshe, Please send the log file of the failure. Regards, Vladimir ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Explore unforgettable gifts for you and your loved ones!
Think about gifts now! The store of these awesome goods is decreasing! (http://Pimmonester.com/)___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] IBV_WR_RDMA_WRITE_WITH_IMM problem
I am attempting to perform an RDMA write which will create WQE's on both the source and sink of the operation. I may be incorrect, but my understanding is that when one performs an RDMA write w/ imm and sets the solicited flag that this will create a WQE in the receivers CQ. So I post a send and set the IBV_WR_RDMA_WRITE_WITH_IMM opcode, and the signaled and solicited flags then poll on both the sender and receiver to get the elements out of the CQ (I have no completion channel). HANG. No WQE appear on either side. Now, if I switch the opcode to a IBV_WR_RDMA_WRITE the send WQE appears in the sender's CQ and, obviously, the receiver CQ is empty (hangs). Can anyone suggest what I am doing wrong? -Matt ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] Re: cannot install SCST with SRP support
Vu Pham wrote: 2. I have also installed OFED1.3 with srpt. but I cannot load the following modules: modprobe scsi_tgt FATAL: Module scsi_tgt not found. This is the error in the SRPT_README.txt - it should say *modprobe scst* instead scsi_tgt Thanks for the answer but when trying to follow example 1: Example 1: working with real back-end scsi disks a. modprobe scsi_tgt (I changed it to modprobe scst) b. modprobe scst_disk c. cat /proc/scsi_tgt/scsi_tgt I got: a. modprobe scst b. modprobe scst_disk FATAL: Error inserting scst_disk (/lib/modules/2.6.16.21-0.8-smp/extra/dev_handlers/scst_disk.ko): No such device c. cat /proc/scsi_tgt/scsi_tgt Device (host:ch:id:lun or name) Device handler 0:0:0:0 none 1:0:0:0 none 2:0:1:0 none 4:0:1:0 none 2:0:1:1 none 4:0:1:1 none 2:0:1:2 none 4:0:1:2 none 2:0:1:3 none 4:0:1:3 none 2:0:1:4 none 4:0:1:4 none 2:0:1:5 none 4:0:1:5 none 2:0:1:6 none 4:0:1:6 none 4:0:1:7 none 2:0:1:7 none 4:0:1:8 none 2:0:1:8 none Doesn't the Device handler need to be dev_disk? -vu ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg Thanks, Doron ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] IBV_WR_RDMA_WRITE_WITH_IMM problem
Hi. Matthew Small wrote: I am attempting to perform an RDMA write which will create WQE's on both the source and sink of the operation. I may be incorrect, but my understanding is that when one performs an RDMA write w/ imm and sets the solicited flag that this will create a WQE in the receivers CQ. Any RDMA write with immediate (even without the solicited flag enabled) will consume a Receive Request in the responder side. So I post a send and set the IBV_WR_RDMA_WRITE_WITH_IMM opcode, and the signaled and solicited flags then poll on both the sender and receiver to get the elements out of the CQ (I have no completion channel). HANG. No WQE appear on either side. Now, if I switch the opcode to a IBV_WR_RDMA_WRITE the send WQE appears in the sender's CQ and, obviously, the receiver CQ is empty (hangs). Can anyone suggest what I am doing wrong? this is trivial, but did you post Receive Request to the responder QP? -Matt ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg