Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
I guess I was all wet...the latest code works with the name local_sa... Sorry for the noise... Steve. On Sat, 2007-01-20 at 09:44 -0600, Steve WIse wrote: On Sat, 2007-01-20 at 09:27 -0600, Steve WIse wrote: BTW: local_sa is still too long. I reduced it to lsa and it worked. I cannot explain this. The BUG_ON() is for (strlen(name) 10). local_sa is 8...9 if you count the NUL at the end. But I hit that BUG_ON() for rhel4u4 and sles9sp3. I reduced the name to lsa and it loads fine. On Thu, 2007-01-18 at 14:09 -0800, Sean Hefty wrote: Can you see if the call to create_singlethread_workqueue() passes in a string longer than 8 characters? If so, try reducing the size of the string. I've committed a change to the local_sa.c file this morning to avoid this sort of backport issue, if this is indeed what it is. - Sean Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH- 20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
BTW: local_sa is still too long. I reduced it to lsa and it worked. On Thu, 2007-01-18 at 14:09 -0800, Sean Hefty wrote: Can you see if the call to create_singlethread_workqueue() passes in a string longer than 8 characters? If so, try reducing the size of the string. I've committed a change to the local_sa.c file this morning to avoid this sort of backport issue, if this is indeed what it is. - Sean Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH- 20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
On Sat, 2007-01-20 at 09:27 -0600, Steve WIse wrote: BTW: local_sa is still too long. I reduced it to lsa and it worked. I cannot explain this. The BUG_ON() is for (strlen(name) 10). local_sa is 8...9 if you count the NUL at the end. But I hit that BUG_ON() for rhel4u4 and sles9sp3. I reduced the name to lsa and it loads fine. On Thu, 2007-01-18 at 14:09 -0800, Sean Hefty wrote: Can you see if the call to create_singlethread_workqueue() passes in a string longer than 8 characters? If so, try reducing the size of the string. I've committed a change to the local_sa.c file this morning to avoid this sort of backport issue, if this is indeed what it is. - Sean Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH- 20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
[openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH-20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
It hit this: struct workqueue_struct *__create_workqueue(const char *name, int singlethread) { int cpu, destroy = 0; struct workqueue_struct *wq; struct task_struct *p; BUG_ON(strlen(name) 10); On Thu, 2007-01-18 at 16:01 -0600, Steve Wise wrote: Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH-20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
Re: [openib-general] ib_local_sa workqueue creation crash with ofed 1.2 + sles9sp3
Can you see if the call to create_singlethread_workqueue() passes in a string longer than 8 characters? If so, try reducing the size of the string. I've committed a change to the local_sa.c file this morning to avoid this sort of backport issue, if this is indeed what it is. - Sean Loading ib_local_sa crashes on sles9sp3 with a BUG() trap. Is this a known issue? --- [cut here ] - [please bite here ] - Kernel BUG at workqueue:308 invalid operand: [1] SMP CPU 0 Pid: 7309, comm: modprobe Tainted: GF U (2.6.5-7.244-smp SLES9_SP3_BRANCH- 20051212183225) RIP: 0010:[8014f9c1] 8014f9c1{__create_workqueue+33} RSP: :01003f4f7d38 EFLAGS: 00010202 RAX: 000b RBX: a02b4e69 RCX: 0bb8 RDX: 0bb8 RSI: 0001 RDI: a02b4e69 RBP: 0001 R08: 003d R09: R10: 000493e0 R11: 0001 R12: 803e3fc0 R13: 000a R14: 803e3fc0 R15: FS: 002a95894b00() GS:8057cc00() knlGS: CS: 0010 DS: ES: CR0: 8005003b CR2: 00539ccf CR3: 00101000 CR4: 06e0 Process modprobe (pid: 7309, threadinfo 01003f4f6000, task 01001fcdc9e0) Stack: 01003fe3f350 803e4020 a02b7180 803e3fc0 000a 803e3fc0 a02b2300 a02bd08f 803e4020 8015d4dd Call Trace:a02bd08f{:ib_local_sa:sa_db_init+143} 8015d4dd{sys_init_module+6589} 80166d5b{generic_file_read+187} a02bd000{:ib_local_sa:sa_db_init+0} 8018da34{vfs_read+244} 8018dc8d{sys_read+157} 80110694{system_call+124} Code: 0f 0b 17 66 38 80 ff ff ff ff 34 01 66 66 90 48 8b 3d 61 50 RIP 8014f9c1{__create_workqueue+33} RSP 01003f4f7d38 ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general ___ openib-general mailing list openib-general@openib.org http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general