mirror of
https://github.com/SCST-project/scst.git
synced 2026-05-22 21:21:28 +00:00
34cca9e432
........ r7095 | bvassche | 2017-02-23 18:07:29 -0800 (Thu, 23 Feb 2017) | 2 lines scst_copy_mgr: Convert comments about locking into runtime checks ........ r7097 | bvassche | 2017-02-23 23:07:41 -0800 (Thu, 23 Feb 2017) | 37 lines scst: Fix locking annotations Avoid that the following warnings are triggered: WARNING: CPU: 5 PID: 32343 at scst/src/scst_copy_mgr.c:2741 scst_cm_on_add_acg+0x12d/0x180 [scst] Call Trace: [<ffffffff812a4d23>] dump_stack+0x86/0xc3 [<ffffffff8106412b>] __warn+0xcb/0xf0 [<ffffffff8106421d>] warn_slowpath_null+0x1d/0x20 [<ffffffffa04f9b0d>] scst_cm_on_add_acg+0x12d/0x180 [scst] [<ffffffffa04b8084>] scst_alloc_add_acg+0xc4/0x280 [scst] [<ffffffffa0486b30>] scst_register_target+0x190/0x3a0 [scst] [<ffffffffa054f2de>] scst_cm_init+0xb8/0xdda [scst] [<ffffffffa054eab9>] init_scst+0xa06/0xc3f [scst] [<ffffffff8100043d>] do_one_initcall+0x3d/0x160 [<ffffffff81154ae0>] do_init_module+0x5f/0x1f1 [<ffffffff810efa69>] load_module+0x2719/0x2a50 [<ffffffff810effcc>] SYSC_finit_module+0xbc/0xf0 [<ffffffff810f001e>] SyS_finit_module+0xe/0x10 [<ffffffff8152fe40>] entry_SYSCALL_64_fastpath+0x23/0xc1 WARNING: CPU: 2 PID: 12885 at scst/src/scst_copy_mgr.c:2826 scst_cm_check_access_acg+0x180/0x1e0 [scst] Call Trace: [<ffffffff812a4d23>] dump_stack+0x86/0xc3 [<ffffffff8106412b>] __warn+0xcb/0xf0 [<ffffffff8106421d>] warn_slowpath_null+0x1d/0x20 [<ffffffffa0502b00>] scst_cm_check_access_acg+0x180/0x1e0 [scst] [<ffffffffa0502cba>] scst_cm_check_access+0x15a/0x280 [scst] [<ffffffffa05030cb>] scst_cm_parse_id_tgt_descr+0x2eb/0x590 [scst] [<ffffffffa050a1d8>] scst_cm_parse_descriptors+0x308/0xa60 [scst] [<ffffffffa04d4440>] scst_parse_descriptors+0x60/0x130 [scst] [<ffffffffa04a3c05>] scst_tgt_pre_exec+0x375/0x4a0 [scst] [<ffffffffa04ac2a9>] scst_process_active_cmd+0x439/0x740 [scst] [<ffffffffa04adbb2>] scst_cmd_thread+0x202/0x6a0 [scst] [<ffffffff810883cf>] kthread+0xef/0x110 [<ffffffff8153008f>] ret_from_fork+0x1f/0x40 ........ r7098 | bvassche | 2017-02-24 17:54:58 -0800 (Fri, 24 Feb 2017) | 18 lines scst_copy_mgr: Avoid that scst_cm_on_add_lun() triggers a kernel warning Avoid that the following warning is triggered: WARNING: CPU: 1 PID: 1282 at scst/src/scst_copy_mgr.c:2784 scst_cm_on_add_lun+0x141/0x1b0 [scst] Call Trace: dump_stack+0x86/0xc3 __warn+0xc1/0xe0 warn_slowpath_null+0x18/0x20 scst_cm_on_add_lun+0x141/0x1b0 [scst] scst_acg_add_lun+0x27f/0x370 [scst] __scst_process_luns_mgmt_store+0x589/0x5f0 [scst] scst_luns_mgmt_store_work_fn+0x22/0x30 [scst] scst_process_sysfs_works+0x7b/0x1c0 [scst] sysfs_work_thread_fn+0x11b/0x280 [scst] kthread+0x102/0x140 ret_from_fork+0x31/0x40 ........ r7099 | bvassche | 2017-02-24 17:56:30 -0800 (Fri, 24 Feb 2017) | 21 lines scst_copy_mgr: Avoid that scst_cm_check_access_acg() triggers a kernel warning Avoid that the following kernel warning is triggered: WARNING: CPU: 2 PID: 1761 at scst/src/scst_copy_mgr.c:2835 Call Trace: dump_stack+0x86/0xc3 __warn+0xc1/0xe0 warn_slowpath_null+0x18/0x20 scst_cm_check_access_acg+0x1c2/0x1d0 [scst] scst_cm_check_access+0x155/0x280 [scst] scst_cm_parse_id_tgt_descr+0x2e6/0x590 [scst] scst_cm_parse_descriptors+0x303/0xa60 [scst] scst_parse_descriptors+0x5b/0x130 [scst] scst_tgt_pre_exec+0x370/0x4a0 [scst] scst_process_active_cmd+0x455/0x770 [scst] scst_cmd_thread+0x1fb/0x690 [scst] kthread+0x102/0x140 ret_from_fork+0x31/0x40 ........ r7100 | bvassche | 2017-02-27 14:36:33 -0800 (Mon, 27 Feb 2017) | 4 lines scst_copy_mgr: Warn if LUN lookup fails It really is a bug if LUN lookup fails. ........ r7102 | bvassche | 2017-03-01 16:09:14 -0800 (Wed, 01 Mar 2017) | 5 lines scst: Coding style updates related to whitespace This patch changes whitespace only and nothing else. ........ r7103 | bvassche | 2017-03-01 16:10:24 -0800 (Wed, 01 Mar 2017) | 5 lines scst: Coding style update related to the use of the keyword 'unsigned' Use 'unsigned int' instead of 'unsigned' and 'unsigned long long' instead of 'long long unsigned'. ........ r7105 | bvassche | 2017-03-07 20:43:23 -0800 (Tue, 07 Mar 2017) | 1 line scst_lib.c: Fix indentation ........ r7112 | vlnb | 2017-04-13 16:15:17 -0700 (Thu, 13 Apr 2017) | 9 lines iscsi: fix misleading error logging in config.c This fixes four similar instances of a very misleading and confusing logging statement, which would print a "Wrong value" that had already been corrected by the check function. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7113 | vlnb | 2017-04-13 16:16:57 -0700 (Thu, 13 Apr 2017) | 7 lines iscsi: logging improvements for iscsi_scstd.c Logging improvements for iscsi_scstd.c print more information about failures. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7114 | vlnb | 2017-04-13 16:19:05 -0700 (Thu, 13 Apr 2017) | 5 lines scst_vdisk: logging to distinguish different cases of EINVAL Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7119 | vlnb | 2017-04-13 16:43:55 -0700 (Thu, 13 Apr 2017) | 8 lines scst: remove superfluous assignment in scst_dg_add() The "res" is unconditionally set by the call to scst_dg_sysfs_add(), so setting it to ENOMEM above that is superfluous. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7120 | vlnb | 2017-04-13 16:45:15 -0700 (Thu, 13 Apr 2017) | 9 lines scst_vdisk: remove parentheses from DEF_DIF_FILENAME_TMPL DEF_DIF_FILENAME_TMPL should not have the parentheses; I used it in some context where that became apparent. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7121 | vlnb | 2017-04-13 16:53:35 -0700 (Thu, 13 Apr 2017) | 10 lines scst_vdisk: move declarations inside #ifdef Move some #ifdefs to also cover declarations used only inside the #ifdef. This is to avoid compiler warnings about unused variables. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7122 | vlnb | 2017-04-13 16:58:32 -0700 (Thu, 13 Apr 2017) | 9 lines Fix a few compiler messages issued when some extra warnings are enabled: casting const to non-const uninitialized structure members non-static local function missing enumerated switch-value cases Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7123 | vlnb | 2017-04-13 18:03:25 -0700 (Thu, 13 Apr 2017) | 8 lines Fix a few minor "extra" compiler warnings (mostly "const" issues) Add "XXX" comments in a few places about potential problems seen in SCST code, for future investigation and possible repair. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7124 | vlnb | 2017-04-13 18:16:10 -0700 (Thu, 13 Apr 2017) | 5 lines iscsi-scst: clean up the conftest subdirectory too Reported-by: David Butterfield <dab21774@gmail.com> ........ r7126 | vlnb | 2017-04-13 18:19:06 -0700 (Thu, 13 Apr 2017) | 5 lines backport ACCESS_ONCE() before 2.6.26 Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7127 | bvassche | 2017-04-13 21:21:38 -0700 (Thu, 13 Apr 2017) | 4 lines ib_srpt: Optimize Makefile Cache conftest results and enable parallel invocation of conftests. ........ r7128 | bvassche | 2017-04-13 21:29:38 -0700 (Thu, 13 Apr 2017) | 4 lines iscsi-scst/Makefile: Optimize Makefile Cache conftest results and enable parallel invocation of conftests. ........ r7129 | bvassche | 2017-04-14 13:46:50 -0700 (Fri, 14 Apr 2017) | 1 line ib_srpt: Fix two configuration tests ........ r7130 | bvassche | 2017-04-14 13:47:26 -0700 (Fri, 14 Apr 2017) | 1 line isert-scst: Fix two configuration tests ........ r7131 | bvassche | 2017-04-14 14:38:56 -0700 (Fri, 14 Apr 2017) | 6 lines iscsi-scst/Makefile: Fix detection of whether InfiniBand is enabled Inspecting modules.symbols is fine after a kernel has been installed but not if a kernel tree has not been installed. If a kernel tree has not been installed, check the .config file. ........ r7132 | bvassche | 2017-04-14 21:10:31 -0700 (Fri, 14 Apr 2017) | 1 line iscsi-scst/Makefile: Add support for make -i ........ r7133 | bvassche | 2017-04-14 21:10:44 -0700 (Fri, 14 Apr 2017) | 1 line srpt/Makefile: Add support for make -i ........ r7138 | vlnb | 2017-04-20 17:57:10 -0700 (Thu, 20 Apr 2017) | 11 lines iscsi-scst: change local names of ioctl() and open() to not conflict Change local names of ioctl() and open() to not conflict with C library names. This isn't really a bug in a strictly kernel-resident build, but the change avoids symbol conflicts with libc when building for usermode. Signed-off-by: David Butterfield <dab21774@gmail.com> with some changes ........ r7140 | vlnb | 2017-04-20 18:04:16 -0700 (Thu, 20 Apr 2017) | 9 lines iscsi-scstd: daemon handle EOF (rc == 0) from nl_fd Daemon now handles receipt of EOF (rc == 0) from nl_fd. Probably this never happens in a real kernel build, but it does during shutdown in a usermode build and it seems like it is "generically correct" in either case. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7142 | vlnb | 2017-04-20 18:17:53 -0700 (Thu, 20 Apr 2017) | 10 lines iscsi-scst: reset conn->rx_task in scsi_cmnd_start() for EXTRACHECKS In scsi_cmnd_start() conn->rx_task is intended to be set only during the call to _stage1_done, but there was no code to reset it. It doesn't actually matter in execution, but it was a little confusing to wonder and try to understand why it wasn't reset. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7144 | vlnb | 2017-04-20 18:36:34 -0700 (Thu, 20 Apr 2017) | 16 lines iscsi_scst: conn_lookup() skips any conn already closing This change helped a secondary problem I had under valgrind (due to some other bug) with initiators timing out and reconnecting their sessions faster than the threads were getting cleaned up (for one thing, valgrind only ever runs one thread at a time). It seems logical that the code in conn_lookup() would want to skip any connection that's already known closing in this case, for the same reason it searches the list in reverse. And the dropping connections don't have to drop in order, so searching in reverse doesn't seem sufficient to avoid finding a wrong (stale, closing) connection structure. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7145 | vlnb | 2017-04-21 20:04:34 -0700 (Fri, 21 Apr 2017) | 3 lines Web updates ........ r7146 | vlnb | 2017-04-21 20:19:52 -0700 (Fri, 21 Apr 2017) | 3 lines Copyright updates ........ r7147 | vlnb | 2017-04-21 20:21:07 -0700 (Fri, 21 Apr 2017) | 3 lines Web updates ........ r7148 | vlnb | 2017-04-21 20:33:13 -0700 (Fri, 21 Apr 2017) | 3 lines Web updates ........ r7149 | vlnb | 2017-04-21 20:50:03 -0700 (Fri, 21 Apr 2017) | 3 lines Web updates ........ r7150 | vlnb | 2017-04-21 20:58:24 -0700 (Fri, 21 Apr 2017) | 3 lines Web updates ........ r7151 | vlnb | 2017-04-21 21:15:21 -0700 (Fri, 21 Apr 2017) | 3 lines ql2x00t: howto update ........ r7155 | vlnb | 2017-04-28 18:00:59 -0700 (Fri, 28 Apr 2017) | 8 lines scst: add mutex_lock/unlock around scst_alloc_add_acg() The CONFIG_SCST_PROC build was calling scst_alloc_add_acg() without holding the scst_mutex. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7156 | bvassche | 2017-05-01 13:22:56 -0700 (Mon, 01 May 2017) | 1 line scst_debug: Remove a duplicated #include directive ........ r7157 | bvassche | 2017-05-01 13:26:08 -0700 (Mon, 01 May 2017) | 3 lines Surround #include <linux/version.h> with #ifndef INSIDE_KERNEL_TREE / #endif ........ r7159 | bvassche | 2017-05-01 22:18:05 -0700 (Mon, 01 May 2017) | 1 line scst_sysfs: Fix spelling in a source code comment ........ r7160 | bvassche | 2017-05-01 22:28:29 -0700 (Mon, 01 May 2017) | 1 line qla2x00t: Convert the non-standard %Lx into %llx ........ r7161 | bvassche | 2017-05-02 07:10:22 -0700 (Tue, 02 May 2017) | 1 line scripts: Update kernel download location ........ r7164 | bvassche | 2017-05-02 11:32:56 -0700 (Tue, 02 May 2017) | 1 line scripts/generate-kernel-patch: Stop applying the put_page_callback and exec_req_fifo patches ........ r7165 | bvassche | 2017-05-02 15:32:06 -0700 (Tue, 02 May 2017) | 4 lines Fix spelling of "Ardis Technologies" Reference: http://www.ddpsan.com/support-sales/about-us. ........ r7166 | bvassche | 2017-05-03 19:36:37 -0700 (Wed, 03 May 2017) | 1 line scstadmin: Fix a few bugs in the regression test code ........ r7167 | bvassche | 2017-05-03 19:45:35 -0700 (Wed, 03 May 2017) | 1 line scst_lib: Convert a comment into a runtime check ........ r7172 | vlnb | 2017-05-10 20:57:04 -0700 (Wed, 10 May 2017) | 8 lines scst: scst_finish_cmd() not try to left-shift lba of -1 Change scst_finish_cmd() so that it does not attempt to left-shift lba when it is -1 (the error value). Left-shifting of a negative value is undefined in C. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7174 | vlnb | 2017-05-12 20:30:07 -0700 (Fri, 12 May 2017) | 6 lines scst: add module parameter to control the copy manager auto assignment Reported-By: Guru Prasad <gp22online@gmail.com> Suggested-By: Sebastian Herbszt <herbszt@gmx.de> ........ r7180 | bvassche | 2017-05-13 20:49:39 -0700 (Sat, 13 May 2017) | 5 lines scst_vdisk: Avoid shifting left by a negative number of bits scst_calc_block_shift() can return a negative number. Hence avoid 1 << scst_calc_block_shift(). Detected by Coverity. ........ r7181 | mlx-storage | 2017-05-14 06:45:02 -0700 (Sun, 14 May 2017) | 6 lines isert-scst: fix isert connection teardown flow Synchronizing the rdma_disconnect() with the post drain wr as other ULPs like iSER do. Signed-off-by: Israel Rukshin <israelr@mellanox.com> ........ r7182 | bvassche | 2017-05-14 07:55:00 -0700 (Sun, 14 May 2017) | 1 line ib_srpt: Fix a logging statement ........ r7183 | bvassche | 2017-05-14 11:49:04 -0700 (Sun, 14 May 2017) | 4 lines isert-scst: Enable type checking for isert portal pointers This patch does not change any functionality. ........ r7184 | bvassche | 2017-05-14 11:53:01 -0700 (Sun, 14 May 2017) | 14 lines isert-scst: Avoid that shutdown sporadically hangs Waiting for other threads to release an object using code like "while (object->refcnt > 0) msleep(100)" without holding a reference on 'object' is wrong because the memory object points at may be freed before or while this loop is in progress. Hence introduce a global portal object count and wait on that count instead of waiting for the per-portal reference count to reach zero. The use-after-free was introduced in r6952 ("isert: faster release of isert_scst module"). See also https://sourceforge.net/p/scst/tickets/2/. ........ r7186 | bvassche | 2017-05-15 19:04:47 -0700 (Mon, 15 May 2017) | 1 line isert-scst: Fix a race condition that was introduced accidentally in r7184 ........ r7187 | vlnb | 2017-05-18 19:37:49 -0700 (Thu, 18 May 2017) | 8 lines scst: implement BLOCKIO devices "active" attribute Intended to be able to load on passive side as described in https://sourceforge.net/p/scst/mailman/message/35549446/ Suggested-by: Кирилл Тюшев <kirill.tyushev8@gmail.com> ........ r7188 | vlnb | 2017-05-18 19:58:23 -0700 (Thu, 18 May 2017) | 21 lines qla2x00t: Disable Out-of-order processing by default From: Quinn Tran <quinn.tran@cavium.com> Out of order(OOO) processing requires initiator, switch and target to support OOO. In today¹s environment, none of the switches support OOO. OOO requires extra buffer space which affect performance. By turning ON this feature in QLogic's FW, it delays error recovery because droped frame is treated as out of order frame. We¹re turning OFF this option of speed up error recovery. Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Quinn Tran <quinn.tran@cavium.com> Signed-off-by: Himanshu Madhani <himanshu.madhani@cavium.com> Merge of mainline commit 300af14bdb28157090f0c6f89d244fda940082da EXPERIMENTAL! ........ r7189 | vlnb | 2017-05-19 19:04:16 -0700 (Fri, 19 May 2017) | 9 lines scst: compute max_transfer args as long, not int Compute max_transfer args as long, not int, to avoid overflow when max_sg_cnt=32768 and PAGE_SHIFT=16 (64K PAGE_SIZE). Note it is not only the min() call, but also the << PAGE_SHIFT needs to be long. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7191 | vlnb | 2017-05-19 20:19:54 -0700 (Fri, 19 May 2017) | 6 lines scst: restore locking documentation comments Better to follow the common rule and always have locking expectations documented in addition to runtime checks. Partial revert of r7095. ........ r7192 | vlnb | 2017-05-19 20:29:41 -0700 (Fri, 19 May 2017) | 5 lines scst: minor cleanup Let's keep it simple without unused functionality ........ r7194 | vlnb | 2017-05-22 20:50:43 -0700 (Mon, 22 May 2017) | 91 lines scst: prevent possible memory reclaim deadlock Reported by lockdep: [ 415.813281] ================================= [ 415.817626] [ INFO: inconsistent lock state ] [ 415.821974] 4.10.12-scst-dbg #4 Tainted: G O [ 415.827360] --------------------------------- [ 415.831705] inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage. [ 415.838217] kswapd0/558 [HC0[0]:SC0[0]:HE1:SE1] takes: [ 415.843342] (sgv_pools_mutex){+.+.?.}, at: [<ffffffffa036d300>] __sgv_shrink+0x90/0x1d0 [scst] [ 415.852075] {RECLAIM_FS-ON-W} state was registered at: [ 415.857208] mark_held_locks+0x6c/0x90 [ 415.861034] lockdep_trace_alloc+0x8a/0xd0 [ 415.865210] kmem_cache_alloc_node+0x32/0x2f0 [ 415.869670] sgv_pool_create_node+0x186/0x2e0 [scst] [ 415.874735] sgv_pool_create+0x12/0x20 [scst] [ 415.879194] scst_sgv_pools_init+0xb9/0x3c0 [scst] [ 415.884065] __dev_user_set_opt+0x152/0x5f0 [scst_user] [ 415.889367] do_one_initcall+0x59/0x190 [ 415.893280] do_init_module+0x6a/0x230 [ 415.897106] load_module+0x591/0x830 [ 415.900758] SYSC_init_module+0xd3/0x120 [ 415.904758] SyS_init_module+0xe/0x10 [ 415.908501] entry_SYSCALL_64_fastpath+0x18/0xad [ 415.913193] irq event stamp: 3831137 [ 415.916763] hardirqs last enabled at (3831137): [<ffffffff8106ff37>] __local_bh_enable_ip+0x67/0xd0 [ 415.925878] hardirqs last disabled at (3831135): [<ffffffff8106fefe>] __local_bh_enable_ip+0x2e/0xd0 [ 415.935018] softirqs last enabled at (3831136): [<ffffffffa036c72d>] __sgv_can_be_shrunk+0x9d/0x180 [scst] [ 415.944764] softirqs last disabled at (3831134): [<ffffffffa036c6e3>] __sgv_can_be_shrunk+0x53/0x180 [scst] [ 415.954481] [ 415.954481] other info that might help us debug this: [ 415.960994] Possible unsafe locking scenario: [ 415.960994] [ 415.966899] CPU0 [ 415.969336] ---- [ 415.971774] lock(sgv_pools_mutex); [ 415.975339] <Interrupt> [ 415.977950] lock(sgv_pools_mutex); [ 415.981691] [ 415.981691] *** DEADLOCK *** [ 415.981691] [ 415.987599] 1 lock held by kswapd0/558: [ 415.991424] #0: (shrinker_rwsem){++++..}, at: [<ffffffff811ace5c>] T.1898+0x4c/0x100 [ 415.999329] [ 415.999329] stack backtrace: [ 416.003677] CPU: 2 PID: 558 Comm: kswapd0 Tainted: G O 4.10.12-scst-dbg #4 [ 416.011663] Hardware name: Dell Inc. PowerEdge R710/00NH4P, BIOS 2.2.10 11/09/2010 [ 416.019216] Call Trace: [ 416.021661] dump_stack+0x6b/0xa0 [ 416.024968] print_usage_bug+0x182/0x190 [ 416.028881] ? print_circular_bug+0x110/0x110 [ 416.033226] mark_lock_irq+0xee/0x290 [ 416.036879] mark_lock+0x113/0x1a0 [ 416.040271] mark_irqflags+0xb0/0x180 [ 416.043924] __lock_acquire+0x245/0x610 [ 416.047750] ? __lock_release+0x408/0x470 [ 416.051750] lock_acquire+0xce/0x2b0 [ 416.055343] ? __sgv_shrink+0x90/0x1d0 [scst] [ 416.059689] ? mutex_lock_nested+0x47/0x750 [ 416.063863] mutex_lock_nested+0x82/0x750 [ 416.067888] ? __sgv_shrink+0x90/0x1d0 [scst] [ 416.072236] ? put_lock_stats+0xe/0x30 [ 416.075975] ? __lock_release+0x408/0x470 [ 416.080001] ? __sgv_shrink+0x90/0x1d0 [scst] [ 416.084349] ? debug_smp_processor_id+0x17/0x20 [ 416.088868] ? put_lock_stats+0xe/0x30 [ 416.092633] ? __sgv_can_be_shrunk+0x9d/0x180 [scst] [ 416.097588] ? mark_held_locks+0x6c/0x90 [ 416.101500] ? __local_bh_enable_ip+0x67/0xd0 [ 416.105873] __sgv_shrink+0x90/0x1d0 [scst] [ 416.110047] ? __local_bh_enable_ip+0x67/0xd0 [ 416.114420] sgv_scan_shrink+0x5d/0x120 [scst] [ 416.118853] do_shrink_slab+0x19b/0x5f0 [ 416.122678] ? T.1898+0x4c/0x100 [ 416.125899] ? down_read_trylock+0x60/0x80 [ 416.129985] T.1898+0xb4/0x100 [ 416.133031] shrink_node+0x62/0xa0 [ 416.136424] balance_pgdat+0x1c8/0x2c0 [ 416.140164] kswapd+0xa0/0x2b0 [ 416.143211] ? preempt_count_sub+0x9f/0xe0 [ 416.147409] ? schedule+0x4e/0xc0 [ 416.150715] ? balance_pgdat+0x2c0/0x2c0 [ 416.154627] kthread+0x111/0x150 [ 416.157849] ? call_usermodehelper_exec_work+0xa0/0xa0 [ 416.162974] ? __kthread_init_worker+0x80/0x80 [ 416.167407] ? call_usermodehelper_exec_work+0xa0/0xa0 [ 416.172534] ? call_usermodehelper_exec_work+0xa0/0xa0 [ 416.177660] ret_from_fork+0x2e/0x40 ........ r7195 | vlnb | 2017-05-24 19:11:06 -0700 (Wed, 24 May 2017) | 8 lines iscsi-scst: cleanup There's no need to end-align the padding. Plus, it's not optimal to access front misaligned memory Reported-by: David Butterfield <dab21774@gmail.com> ........ r7196 | vlnb | 2017-05-25 19:45:47 -0700 (Thu, 25 May 2017) | 5 lines scst_vdisk: add "active" attridute to the list of BLOCKIO add device attributes This is forgotten piece to r7187 ........ r7197 | vlnb | 2017-05-30 21:01:00 -0700 (Tue, 30 May 2017) | 3 lines scst: a better version of possible memory reclaim deadlock fix r7194 ........ r7198 | vlnb | 2017-05-31 20:46:06 -0700 (Wed, 31 May 2017) | 5 lines scst: cleanup of dead code Reported-By: David Butterfield <dab21774@gmail.com> ........ r7199 | vlnb | 2017-05-31 20:49:26 -0700 (Wed, 31 May 2017) | 5 lines scst: scst_free_acg() not dereference NULL acg->tgt for proc interface Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7200 | vlnb | 2017-05-31 20:53:08 -0700 (Wed, 31 May 2017) | 11 lines iscsi-scst: fix a couple of bad-function-cast warnings This is a "marginal" warning, but there are only six of them in the SCST code and this fixes two of those. session.c:105:4: warning: cast from function call of type void * to non-matching type _Bool [-Wbad-function-cast] target.c:668:14: warning: cast from function call of type void * to non-matching type _Bool [-Wbad-function-cast] Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7201 | vlnb | 2017-05-31 21:04:45 -0700 (Wed, 31 May 2017) | 13 lines scst: remove a few unnecessary shadow declarations Remove a few "shadow" declarations that appear unnecessary and probably unintended. (These are cases where a declaration in an inner scope hides another declaration of the same name in an outer scope.) Not all shadow declarations are superfluous -- so this particular compiler warning is one that cannot be simply heeded while half asleep. But these particular ones I removed appear to be superfluous. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7202 | vlnb | 2017-05-31 21:08:02 -0700 (Wed, 31 May 2017) | 8 lines usr: fix -Wmissing-prototypes warnings Fix all twelve [-Wmissing-prototypes] warnings by adding "static" to their declarations. Signed-off-by: David Butterfield <dab21774@gmail.com> ........ r7204 | vlnb | 2017-06-02 21:02:22 -0700 (Fri, 02 Jun 2017) | 5 lines Fix dereference beyond last sg element Reported-By: David Butterfield <dab21774@gmail.com> ........ r7205 | vlnb | 2017-06-05 21:20:09 -0700 (Mon, 05 Jun 2017) | 10 lines scst_vdisk: 'numa_node_id' signed attribute value fix The "numa_node_id" attribute reside in the scst_vdisk_dev struct as signed integer and may well be negative values (eg, -1), so move fetching it above the unsigned attributes and use kstrtoll() to retrieve the values. Signed-off-by: Marc A. Smith <marc.smith@parodyne.com> ........ r7206 | vlnb | 2017-06-06 18:13:58 -0700 (Tue, 06 Jun 2017) | 3 lines scst: update ALUA documentation (Active/Non-Optimized) ........ r7207 | vlnb | 2017-06-06 18:23:59 -0700 (Tue, 06 Jun 2017) | 3 lines scst: minor cleanup for previous commit ........ git-svn-id: http://svn.code.sf.net/p/scst/svn/branches/3.2.x@7210 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Overview ======== This is the SCST development repository. It contains not a single project SCST as one can think, but a number of them, which are divided as the following: 1. SCST core in scst/ subdirectory 2. Administration utility for SCST core scstadmin in scstadmin/ 3. Target drivers in own subdirectories qla2x00t/, iscsi-scst/, etc. 4. User space programs in usr/ subdirectory, like fileio_tgt. 5. Some various docs in doc/ subdirectory. Those subprojects are in most cases independent from each other, although some of them depend from the SCST core. They put in the single repository only to simplify their development, they are released independently. Thus, use "make all" only if you really need to build everything. Otherwise build only what you need, like for iSCSI-SCST: make scst scst_install iscsi iscsi_install For more information about each subproject see their README files. QLogic target driver ==================== QLogic target driver qla2x00t is the old driver, forked from qla2xxx several years ago. It is very stable, well tested and actively used in many production setups. The ultimate goal is to have the mainstream (git) QLogic target driver to be the main and the only QLogic target driver, but, unfortunately, this driver not yet reached level of quality and maturity of qla2x00t. We with QLogic are working toward it. You can find the latest version of the git driver in git://git.qlogic.com/scst-qla2xxx.git. It is maintained by QLogic, hence located in the QLogic's git. To integrate it into the SCST build tree you need: 1. Clone the QLogic git tree somewhere in your system. 2. Create in the SCST root, i.e. this directory, a subdirectory with name qla2x00t_git 3. Optional for the in-tree build: symlink drivers/scsi/qla2xxx subdirectory in the cloned git tree to the qla2x00t_git subdirectory Thats all. Now common and QLA specific root Makefile targets "magically" start working with the new driver. For detail instructions how to setup QLA target build environment see its README or HOWTO. Run either "make all" to build the SCST core and all target drivers with in-tree QLogic drivers, or the following command with standalone QLogic drivers: BUILD_2X_MODULE=y CONFIG_SCSI_QLA_FC=y CONFIG_SCSI_QLA2XXX_TARGET=y make all In presence of qla2x00t_git subdirectory you can still build the qla2x00t using qla_old* root Makefile targets. Vladislav Bolkhovitin <vst@vlnb.net>, http://scst.sourceforge.net