Re: [ewg] [PATCH] IB/core: Add support for Receive Core Affinity
+int ib_create_qp_range(struct ib_pd *pd, struct ib_qp_init_attr *qp_init_attr, + int nqps, int align, struct ib_qp *list[]) It just seems wrong to me to require the caller to specify the alignment restrictions. Isn't this HCA specific? I agree with you about this, but since my previous posts on the issue did not receive too much attention, I did not want to change my implementation before the issue has been discussed. Is IPoIB really going to know whether or not the QP numbers returned by this call are aligned or not? What if I call ib_create_qp_range() with nqps=3 and align=0? I am not sure I understand your argument here: in this case you create 3 consecutive QPs with no other restrictions on the number of of the first. Also, in ib_verbs.h, struct ib_qp_attr now has a struct rca_attr field. I don't see why the struct rca_attr field is needed for ib_modify_qp(). It seems to me that this information should be stored as part of the QP info when creating the N QPs. Why should the verbs caller need to know about this? The values are determined by the HCA when the QPs are created. I totally agree with you here too and I also sent an email about that some time ago. Again, I want to trigger a discussion to close on the API before re-implementing. I hope to get more opinions and then re-work the patch. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] buy now Viagra (Sildenafil) 100mg x 90 pills US $ 159.95
Price for Viagra 50mg x 30 pills US $ 3.00 Per Pill http://akyzag.bay.livefilestore.com/y1poD_EyPOjWZrdLptTpG3hoLWhLQcVNNTnDfdIpSNOrmS17pBgtHVjSBnZ7d9spoQAYnD6ExJWvwzv6OtLDXEvMg/index.html ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] :o)
Nei Ho, ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] Suggestion to improve usibility of the install script
On Sunday 03 August 2008 15:24:54 Vladimir Sokolovsky wrote: Hi Stefan, I did the required change, Please check in the OFED-1.4-20080804-0600 build. Regards, Vladimir Great, works fine for me. Thank you! regards Stefan ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Only 3 month are required for your ehnancing, and she will notice that
Wild nights. 10 reasons to try absolutely new enhancing product. Click here! ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] ofed-1.4 beta
Tziporet, I request that we make sure the following is in before we release beta: - NFSRDMA fastreg patches _and_ NFSRDMA working at least on 2.6.27-rc1. Currently it doesn't install correctly. Also we're waiting for the fastreg patch series from Tom Tucker which should be ready this week. - RDS iWARP bcopy. RDS doesn't currently build in the ofed_2_6_27 branch, and there are a set of patches that apparently never got delivered to Vlad. These should be ready this week as well. I'm asking that we hold off spinning Beta until these are ready. I don't see the point of a beta release if all the anchor functionality isn't in. And from an iWARP perspective, the above two are anchor items. Thanks, Steve. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] [GIT PULL ofed-1.4 ofed_2_6_27] more cxgb3 backports
Vlad, Please pull from git://git.openfabrics.org/~swise/ofed-1.4 ofed_2_6_27 These patches fix up suse and rehat backports. I think cxgb3 should now compile on all the platforms. commit 0f6891ea9533ccb00e25431352f7ea5f2ffcc161 Author: Steve Wise [EMAIL PROTECTED] Date: Mon Aug 4 12:34:40 2008 -0500 RDMA/cxgb3: backports for suse kernels. Signed-off-by: Steve Wise [EMAIL PROTECTED] commit 87bfa88d4daf4a5c2f39e9e6e8fdda3eb338ae0e Author: Steve Wise [EMAIL PROTECTED] Date: Mon Aug 4 12:10:47 2008 -0500 RDMA/cxgb3: backports for redhat kernels. Signed-off-by: Steve Wise [EMAIL PROTECTED] ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] ofed-1.4 ofed_2_6_27 and NFSRDMA
Jeff Becker wrote: Steve Wise wrote: Jeff, I'm running on the ofed-1.4 ofed_2_6_27 branch. Building against a 2.6.27-rc1 running kernel that has nfsrdma configured. I then build/install the ofed drivers with nfsrdma configured in the ofed tree. This _should_ effectively replace the xprtrdma and scvrdma modules from the 2.6.27-rc1 tree with the ofed built versions. It apparently is not doing this. The result is an oops when you try to use nfsrdma because the header files for key rdma structs are different from the 2.6.27-rc1 kernel includes and the ofed-1.4 includes. It appears that the nfsrdma modules are _not_ getting installed by the ofed makefiles. EG I don't see an xprtrdma.ko in /lib/modules/$(uname -r)/updates/kernel/net/sunrpc/xprtrdma. I only see on in /lib/modules/$(uname -r)/kernel/net/sunrpc/xprtrdma. Hi Steve. Thanks for finding this. Actually, I had only tested the base nfsrdma build/install on 2.6.26. I then switched to working on backports, and haven't had a chance to sync up my tree with the 2.6.27-rc1 stuff. I'll do that as soon as I can. Note there are no backports for 2.6.27-rc1. I think this is makefile/bldenv issue. I'm willing to debug this. Also, i'm going to be taking Tom Tucker's fast-reg patches and adding them to ofed-1.4#ofed_2_6_27. I'll try and get all this working on 2.6.27-rc1. How does that sound? Also, did you include the nfsutils package in ofed-1.4? We'll need that for mount.nfs. Thanks! Steve. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] ofed-1.4 beta
Steve Wise wrote: I'm asking that we hold off spinning Beta until these are ready. I don't see the point of a beta release if all the anchor functionality isn't in. And from an iWARP perspective, the above two are anchor items. I think there are several show-stopper from IB perspective too (e.g. IPoIB deadlock), and also no new MPI package is ready. Do you think you can make it for next week? Tziporet ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] ofed-1.4 beta
Tziporet Koren wrote: Steve Wise wrote: I'm asking that we hold off spinning Beta until these are ready. I don't see the point of a beta release if all the anchor functionality isn't in. And from an iWARP perspective, the above two are anchor items. I think there are several show-stopper from IB perspective too (e.g. IPoIB deadlock), and also no new MPI package is ready. Do you think you can make it for next week? Yes. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Re: [GIT PULL ofed-1.4 ofed_2_6_27] more cxgb3 backports
Steve Wise wrote: Vlad, Please pull from git://git.openfabrics.org/~swise/ofed-1.4 ofed_2_6_27 These patches fix up suse and rehat backports. I think cxgb3 should now compile on all the platforms. commit 0f6891ea9533ccb00e25431352f7ea5f2ffcc161 Author: Steve Wise [EMAIL PROTECTED] Date: Mon Aug 4 12:34:40 2008 -0500 RDMA/cxgb3: backports for suse kernels. Signed-off-by: Steve Wise [EMAIL PROTECTED] commit 87bfa88d4daf4a5c2f39e9e6e8fdda3eb338ae0e Author: Steve Wise [EMAIL PROTECTED] Date: Mon Aug 4 12:10:47 2008 -0500 RDMA/cxgb3: backports for redhat kernels. Signed-off-by: Steve Wise [EMAIL PROTECTED] Hi Steve, cxgb3 compilation failed on: Build failed on 2.6.20 Build failed on 2.6.19 Build failed on 2.6.16 Build failed on 2.6.17 Build failed on 2.6.18.6 Build failed on 2.6.9-55.ELsmp Build failed on 2.6.9-42.ELsmp Build failed on 2.6.9-67.ELsmp gcc -Wp,-MD,/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/.l2t.o.d -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/4.1.1/include -D__KERNEL__ -include include/linux/autoconf.h -include /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/include/linux/autoconf.h -I/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/kernel_addons/backport/2.6.20/include/ -include /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/infiniband/debug/mtrack.h -Werror-implicit-function-declaration -I/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/include -I/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/infiniband/debug -I/usr/local/include/scst -I/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/infiniband/ulp/srpt -I/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3 -Iinclude -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Os -mtune=generic -m64 -mno-red-zone -mcmodel=kernel -pipe -fno-reorder-blocks -Wno-sign-compare -fno-asynchronous-unwind-tables -funit-at-a-time -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -fomit-frame-pointer -g -fno-stack-protector -Wdeclaration-after-statement -Wno-pointer-sign -DMODULE -DKBUILD_STR(s)=#s -DKBUILD_BASENAME=KBUILD_STR(l2t) -DKBUILD_MODNAME=KBUILD_STR(cxgb3) -c -o /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/l2t.o /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/l2t.c /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/l2t.c: In function 't3_l2t_get': /tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/l2t.c:340: error: implicit declaration of function 'vlan_dev_vlan_id' make[3]: *** [/tmp/ofa_1_4_dev_kernel-20080805-1757_linux-2.6.20_check/drivers/net/cxgb3/l2t.o] Error 1 Regards, Vladimir ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Re: [GIT PULL ofed-1.4 ofed_2_6_27] more cxgb3 backports
Hi Steve, cxgb3 compilation failed on: Build failed on 2.6.20 Build failed on 2.6.19 Build failed on 2.6.16 Build failed on 2.6.17 Build failed on 2.6.18.6 Build failed on 2.6.9-55.ELsmp Build failed on 2.6.9-42.ELsmp Build failed on 2.6.9-67.ELsmp I'll fix this asap. I thought I had all the k.o. kernels working... Steve. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] ofed-1.4 ofed_2_6_27 and NFSRDMA
Steve Wise wrote: Jeff Becker wrote: Steve Wise wrote: Jeff, I'm running on the ofed-1.4 ofed_2_6_27 branch. Building against a 2.6.27-rc1 running kernel that has nfsrdma configured. I then build/install the ofed drivers with nfsrdma configured in the ofed tree. This _should_ effectively replace the xprtrdma and scvrdma modules from the 2.6.27-rc1 tree with the ofed built versions. It apparently is not doing this. The result is an oops when you try to use nfsrdma because the header files for key rdma structs are different from the 2.6.27-rc1 kernel includes and the ofed-1.4 includes. It appears that the nfsrdma modules are _not_ getting installed by the ofed makefiles. EG I don't see an xprtrdma.ko in /lib/modules/$(uname -r)/updates/kernel/net/sunrpc/xprtrdma. I only see on in /lib/modules/$(uname -r)/kernel/net/sunrpc/xprtrdma. Hi Steve. Thanks for finding this. Actually, I had only tested the base nfsrdma build/install on 2.6.26. I then switched to working on backports, and haven't had a chance to sync up my tree with the 2.6.27-rc1 stuff. I'll do that as soon as I can. Note there are no backports for 2.6.27-rc1. I think this is makefile/bldenv issue. I'm willing to debug this. I thought backports are only included for kernels older than the version (2.6.27-rc1 in this case) on which the OFED release is based? Also, i'm going to be taking Tom Tucker's fast-reg patches and adding them to ofed-1.4#ofed_2_6_27. I'll try and get all this working on 2.6.27-rc1. How does that sound? Great. Also, did you include the nfsutils package in ofed-1.4? We'll need that for mount.nfs. I haven't done that yet, as I've been focusing on the kernel stuff. Also, I was thinking that it might be OK for folks to go grab that independently of OFED. But on second thought (and after seeing your suggestion) I agree. I'll see if I can pull together an RPM. Thanks. -jeff Thanks! Steve. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] iscsi backport for OFED1.4
Hi Doron, Please prepare iSCSI backport patches for OFED-1.4 (based on 2.6.27). git://git.openfabrics.org/ofed_1_4/linux-2.6.git ofed_2_6_27 Note, I removed all existing iSCSI backports (they can't be applied for 2.6.27-rc1). There is a copy of these backports under kernel_patches/attic/backports. Regards, Vladimir ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] [PATCH] IB/core: Add support for Receive Core Affinity
On Tue, 2008-08-05 at 09:10 +0300, Eli Cohen wrote: +int ib_create_qp_range(struct ib_pd *pd, struct ib_qp_init_attr *qp_init_attr, +int nqps, int align, struct ib_qp *list[]) It just seems wrong to me to require the caller to specify the alignment restrictions. Isn't this HCA specific? I agree with you about this, but since my previous posts on the issue did not receive too much attention, I did not want to change my implementation before the issue has been discussed. Is IPoIB really going to know whether or not the QP numbers returned by this call are aligned or not? What if I call ib_create_qp_range() with nqps=3 and align=0? I am not sure I understand your argument here: in this case you create 3 consecutive QPs with no other restrictions on the number of of the first. My point is that I am guessing that the ib_modify_qp() will return an error unless the QPs are created with the right alignment but the caller has no way of knowing what the right alignment value is and it shouldn't need to know since the ib_create_qp_range() could have an argument or flag which says the created QPs should be usable for receive affinity. The HCA driver can then use whatever alignment is needed and mark the QP struct as being part of a group. The ib_modify_qp() probably only needs a flag to say enable receive affinity. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] Interesting RX Offers
Splendid Medical Satisfactions http://everyhingwanted.spaces.live.com/default.aspx Most importantly, someone struggles Victoria K., Boston ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] gitweb add request
Hi, I understand that an admin needs to take some action to get a user's git trees to show up on openfabrics, could an admin please do that for user agrover? (BTW for those who haven't heard: I'm taking over RDS maint/devel from Olaf. Hi, nice to meetcha. :) Thanks -- Andy ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] gitweb add request
Andy Grover wrote: Hi, I understand that an admin needs to take some action to get a user's git trees to show up on openfabrics, could an admin please do that for user agrover? (BTW for those who haven't heard: I'm taking over RDS maint/devel from Olaf. Hi, nice to meetcha. :) Thanks -- Andy ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg Hey Andy, What are the paths to your git trees that you want published? I think the admin needs that. BTW: EWG: Who is admin these days? Steve. ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
[ewg] is it you? karina here
Hi, i am here sitting in the internet caffe. Found your email and decided to write. I am 25 y.o.girl. I have a picture if you want. No need to reply here as this is not may email. Write me at [EMAIL PROTECTED] ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Re: [ewg] gitweb add request
Steve Wise wrote: What are the paths to your git trees that you want published? I think the admin needs that. BTW: EWG: Who is admin these days? Really? You can't just do everything in ~/scm? Ok then: /home/agrover ~/scm/ofed_1_3/linux-2.6.git ~/scm/ofed_1_3/rds-tools.git ~/scm/ofed_1_4/linux-2.6.git ~/scm/ofed_1_4/rds-tools.git Regards -- Andy ___ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg