- 01 Dec, 2020 8 commits
-
-
Chaitanya Kulkarni authored
Right now nvme_alloc_request() allocates a request from block layer based on the value of the qid. When qid set to NVME_QID_ANY it used blk_mq_alloc_request() else blk_mq_alloc_request_hctx(). The function nvme_alloc_request() is called from different context, The only place where it uses non NVME_QID_ANY value is for fabrics connect commands :- nvme_submit_sync_cmd() NVME_QID_ANY nvme_features() NVME_QID_ANY nvme_sec_submit() NVME_QID_ANY nvmf_reg_read32() NVME_QID_ANY nvmf_reg_read64() NVME_QID_ANY nvmf_reg_write32() NVME_QID_ANY nvmf_connect_admin_queue() NVME_QID_ANY nvme_submit_user_cmd() NVME_QID_ANY nvme_alloc_request() nvme_keep_alive() NVME_QID_ANY nvme_alloc_request() nvme_timeout() NVME_QID_ANY nvme_alloc_request() nvme_delete_queue() NVME_QID_ANY nvme_alloc_request() nvmet_passthru_execute_cmd() NVME_QID_ANY nvme_alloc_request() nvmf_connect_io_queue() QID __nvme_submit_sync_cmd() nvme_alloc_request() With passthru nvme_alloc_request() now falls into the I/O fast path such that blk_mq_alloc_request_hctx() is never gets called and that adds additional branch check in fast path. Split the nvme_alloc_request() into nvme_alloc_request() and nvme_alloc_request_qid(). Replace each call of the nvme_alloc_request() with NVME_QID_ANY param with a call to newly added nvme_alloc_request() without NVME_QID_ANY. Replace a call to nvme_alloc_request() with QID param with a call to newly added nvme_alloc_request() and nvme_alloc_request_qid() based on the qid value set in the __nvme_submit_sync_cmd(). Signed-off-by:
Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> Reviewed-by:
Logan Gunthorpe <logang@deltatee.com> Signed-off-by:
Christoph Hellwig <hch@lst.de>
-
Chaitanya Kulkarni authored
This is a preparation patch to have minimal block layer request bio append functionality in the context of the NVMeOF Passthru driver which falls in the fast path and doesn't need calls from blk_rq_append_bio(). Signed-off-by:
-
Chaitanya Kulkarni authored
NVMeOF controller in the passsthru mode is capable of handling wide set of I/O commands including vender specific passhtru io comands. The vendor specific I/O commands are used to read the large drive logs and can take longer than default NVMe commands, i.e. for passthru requests the timeout value may differ from the passthru controller's default timeout values (nvme-core:io_timeout). Add a configfs attribute so that user can set the io timeout values. In case if this configfs value is not set nvme_alloc_request() will set the NVME_IO_TIMEOUT value when request queuedata is NULL. Signed-off-by:
-
Chaitanya Kulkarni authored
NVMeOF controller in the passsthru mode is capable of handling wide set of admin commands including vender specific passhtru admin comands. The vendor specific admin commands are used to read the large drive logs and can take longer than default NVMe commands, i.e. for passthru requests the timeout value may differ from the passthru controller's default timeout values (nvme-core:admin_timeout). Add a configfs attribute so that user can set the admin timeout values. In case if this configfs value is not set nvme_alloc_request() will set the ADMIN_TIMEOUT value when request queuedata is NULL. Signed-off-by:
-
Chaitanya Kulkarni authored
This is purely a clenaup patch, add prefix NVME to the ADMIN_TIMEOUT to make consistent with NVME_IO_TIMEOUT. Signed-off-by:
-
Chaitanya Kulkarni authored
The function nvme_alloc_request() is called from different context (I/O and Admin queue) where callers do not consider the I/O timeout when called from I/O queue context. Update nvme_alloc_request() to set the default I/O and Admin timeout value based on whether the queuedata is set or not. Signed-off-by:
Sagi Grimberg <sagi@grimberg.me> Signed-off-by:
-
Baolin Wang authored
Use the request's '->mq_hctx->queue_num' directly to simplify the nvme_req_qid() function. Signed-off-by:
Baolin Wang <baolin.wang@linux.alibaba.com> Signed-off-by:
-
James Smart authored
Add sysfs attribute to specify parameters for dropping a command. The attribute takes a string of: <opcode>:<starting a what instance>:<number of times> Opcode is formatted as lower 8 bits are opcode. If a fabrics opcode, a bit above bits 7:0 will be set. Once set, each sqe is looked at. If the opcode matches the running instance count is updated. If the instance count is in the range of where to drop (based on starting and # of times), then drop the command by not passing it to the target layer. Signed-off-by:James Smart <james.smart@broadcom.com>
-
- 30 Nov, 2020 8 commits
-
-
Jens Axboe authored
Merge branch 'md-next' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md into for-5.11/drivers Pull MD changes from Song: "Summary: 1. Fix race condition in md_ioctl(), by Dae R. Jeong; 2. Initialize read_slot properly for raid10, by Kevin Vigor; 3. Code cleanup, by Pankaj Gupta; 4. md-cluster resync/reshape fix, by Zhao Heming." * 'md-next' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md: md/cluster: fix deadlock when node is doing resync job md/cluster: block reshape with remote resync job md: use current request time as base for ktime comparisons md: add comments in md_flush_request() md: improve variable names in md_flush_request() md/raid10: initialize r10_bio->read_slot before use. md: fix a warning caused by a race between concurrent md_ioctl()s
-
Zhao Heming authored
md-cluster uses MD_CLUSTER_SEND_LOCK to make node can exclusively send msg. During sending msg, node can concurrently receive msg from another node. When node does resync job, grab token_lockres:EX may trigger a deadlock: ``` nodeA nodeB -------------------- -------------------- a. send METADATA_UPDATED held token_lockres:EX b. md_do_sync resync_info_update send RESYNCING + set MD_CLUSTER_SEND_LOCK + wait for holding token_lockres:EX c. mdadm /dev/md0 --remove /dev/sdg + held reconfig_mutex + send REMOVE + wait_event(MD_CLUSTER_SEND_LOCK) d. recv_daemon //METADATA_UPDATED from A process_metadata_update + (mddev_trylock(mddev) || MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD) //this time, both return false forever ``` Explaination: a. A send METADATA_UPDATED This will block another node to send msg b. B does sync jobs, which will send RESYNCING at intervals. This will be block for holding token_lockres:EX lock. c. B do "mdadm --remove", which will send REMOVE. This will be blocked by step <b>: MD_CLUSTER_SEND_LOCK is 1. d. B recv METADATA_UPDATED msg, which send from A in step <a>. This will be blocked by step <c>: holding mddev lock, it makes wait_event can't hold mddev lock. (btw, MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD keep ZERO in this scenario.) There is a similar deadlock in commit 0ba95977 ("md-cluster: use sync way to handle METADATA_UPDATED msg") In that commit, step c is "update sb". This patch step c is "mdadm --remove". For fixing this issue, we can refer the solution of function: metadata_update_start. Which does the same grab lock_token action. lock_comm can use the same steps to avoid deadlock. By moving MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD from lock_token to lock_comm. It enlarge a little bit window of MD_CLUSTER_HOLDING_MUTEX_FOR_RECVD, but it is safe & can break deadlock. Repro steps (I only triggered 3 times with hundreds tests): two nodes share 3 iSCSI luns: sdg/sdh/sdi. Each lun size is 1GB. ``` ssh root@node2 "mdadm -S --scan" mdadm -S --scan for i in {g,h,i};do dd if=/dev/zero of=/dev/sd$i oflag=direct bs=1M \ count=20; done mdadm -C /dev/md0 -b clustered -e 1.2 -n 2 -l mirror /dev/sdg /dev/sdh \ --bitmap-chunk=1M ssh root@node2 "mdadm -A /dev/md0 /dev/sdg /dev/sdh" sleep 5 mkfs.xfs /dev/md0 mdadm --manage --add /dev/md0 /dev/sdi mdadm --wait /dev/md0 mdadm --grow --raid-devices=3 /dev/md0 mdadm /dev/md0 --fail /dev/sdg mdadm /dev/md0 --remove /dev/sdg mdadm --grow --raid-devices=2 /dev/md0 ``` test script will hung when executing "mdadm --remove". ``` # dump stacks by "echo t > /proc/sysrq-trigger" md0_cluster_rec D 0 5329 2 0x80004000 Call Trace: __schedule+0x1f6/0x560 ? _cond_resched+0x2d/0x40 ? schedule+0x4a/0xb0 ? process_metadata_update.isra.0+0xdb/0x140 [md_cluster] ? wait_woken+0x80/0x80 ? process_recvd_msg+0x113/0x1d0 [md_cluster] ? recv_daemon+0x9e/0x120 [md_cluster] ? md_thread+0x94/0x160 [md_mod] ? wait_woken+0x80/0x80 ? md_congested+0x30/0x30 [md_mod] ? kthread+0x115/0x140 ? __kthread_bind_mask+0x60/0x60 ? ret_from_fork+0x1f/0x40 mdadm D 0 5423 1 0x00004004 Call Trace: __schedule+0x1f6/0x560 ? __schedule+0x1fe/0x560 ? schedule+0x4a/0xb0 ? lock_comm.isra.0+0x7b/0xb0 [md_cluster] ? wait_woken+0x80/0x80 ? remove_disk+0x4f/0x90 [md_cluster] ? hot_remove_disk+0xb1/0x1b0 [md_mod] ? md_ioctl+0x50c/0xba0 [md_mod] ? wait_woken+0x80/0x80 ? blkdev_ioctl+0xa2/0x2a0 ? block_ioctl+0x39/0x40 ? ksys_ioctl+0x82/0xc0 ? __x64_sys_ioctl+0x16/0x20 ? do_syscall_64+0x5f/0x150 ? entry_SYSCALL_64_after_hwframe+0x44/0xa9 md0_resync D 0 5425 2 0x80004000 Call Trace: __schedule+0x1f6/0x560 ? schedule+0x4a/0xb0 ? dlm_lock_sync+0xa1/0xd0 [md_cluster] ? wait_woken+0x80/0x80 ? lock_token+0x2d/0x90 [md_cluster] ? resync_info_update+0x95/0x100 [md_cluster] ? raid1_sync_request+0x7d3/0xa40 [raid1] ? md_do_sync.cold+0x737/0xc8f [md_mod] ? md_thread+0x94/0x160 [md_mod] ? md_congested+0x30/0x30 [md_mod] ? kthread+0x115/0x140 ? __kthread_bind_mask+0x60/0x60 ? ret_from_fork+0x1f/0x40 ``` At last, thanks for Xiao's solution. Cc: stable@vger.kernel.org Signed-off-by:Zhao Heming <heming.zhao@suse.com> Suggested-by:
Xiao Ni <xni@redhat.com> Reviewed-by:
Song Liu <songliubraving@fb.com>
-
Zhao Heming authored
Reshape request should be blocked with ongoing resync job. In cluster env, a node can start resync job even if the resync cmd isn't executed on it, e.g., user executes "mdadm --grow" on node A, sometimes node B will start resync job. However, current update_raid_disks() only check local recovery status, which is incomplete. As a result, we see user will execute "mdadm --grow" successfully on local, while the remote node deny to do reshape job when it doing resync job. The inconsistent handling cause array enter unexpected status. If user doesn't observe this issue and continue executing mdadm cmd, the array doesn't work at last. Fix this issue by blocking reshape request. When node executes "--grow" and detects ongoing resync, it should stop and report error to user. The following script reproduces the issue with ~100% probability. (two nodes share 3 iSCSI luns: sdg/sdh/sdi. Each lun size is 1GB) ``` # on node1, node2 is the remote node. ssh root@node2 "mdadm -S --scan" mdadm -S --scan for i in {g,h,i};do dd if=/dev/zero of=/dev/sd$i oflag=direct bs=1M \ count=20; done mdadm -C /dev/md0 -b clustered -e 1.2 -n 2 -l mirror /dev/sdg /dev/sdh ssh root@node2 "mdadm -A /dev/md0 /dev/sdg /dev/sdh" sleep 5 mdadm --manage --add /dev/md0 /dev/sdi mdadm --wait /dev/md0 mdadm --grow --raid-devices=3 /dev/md0 mdadm /dev/md0 --fail /dev/sdg mdadm /dev/md0 --remove /dev/sdg mdadm --grow --raid-devices=2 /dev/md0 ``` Cc: stable@vger.kernel.org Signed-off-by: -
Pankaj Gupta authored
Request coalescing logic uses 'prev_flush_start' as base to compare the current request start time. 'prev_flush_start' is updated in other context. This patch changes this by using ktime comparison base to 'req_start' for better readability of code. Signed-off-by:
Pankaj Gupta <pankaj.gupta@cloud.ionos.com> Signed-off-by:
-
Pankaj Gupta authored
Request coalescing logic is dependent on flush time update in other context. This patch adds comments to understand the code flow better. Signed-off-by:
-
Pankaj Gupta authored
This patch improves readability by using better variable names in flush request coalescing logic. Signed-off-by:
Paul Menzel <pmenzel@molgen.mpg.de> Signed-off-by:
-
Kevin Vigor authored
In __make_request() a new r10bio is allocated and passed to raid10_read_request(). The read_slot member of the bio is not initialized, and the raid10_read_request() uses it to index an array. This leads to occasional panics. Fix by initializing the field to invalid value and checking for valid value in raid10_read_request(). Cc: stable@vger.kernel.org Signed-off-by:
Kevin Vigor <kvigor@gmail.com> Signed-off-by:
-
Dae R. Jeong authored
Syzkaller reports a warning as belows. WARNING: CPU: 0 PID: 9647 at drivers/md/md.c:7169 ... Call Trace: ... RIP: 0010:md_ioctl+0x4017/0x5980 drivers/md/md.c:7169 RSP: 0018:ffff888096027950 EFLAGS: 00010293 RAX: ffff88809322c380 RBX: 0000000000000932 RCX: ffffffff84e266f2 RDX: 0000000000000000 RSI: ffffffff84e299f7 RDI: 0000000000000007 RBP: ffff888096027bc0 R08: ffff88809322c380 R09: ffffed101341a482 R10: ffff888096027940 R11: ffff88809a0d240f R12: 0000000000000932 R13: ffff8880a2c14100 R14: ffff88809a0d2268 R15: ffff88809a0d2408 __blkdev_driver_ioctl block/ioctl.c:304 [inline] blkdev_ioctl+0xece/0x1c10 block/ioctl.c:606 block_ioctl+0xee/0x130 fs/block_dev.c:1930 vfs_ioctl fs/ioctl.c:46 [inline] file_ioctl fs/ioctl.c:509 [inline] do_vfs_ioctl+0xd5f/0x1380 fs/ioctl.c:696 ksys_ioctl+0xab/0xd0 fs/ioctl.c:713 __do_sys_ioctl fs/ioctl.c:720 [inline] __se_sys_ioctl fs/ioctl.c:718 [inline] __x64_sys_ioctl+0x73/0xb0 fs/ioctl.c:718 do_syscall_64+0xfd/0x680 arch/x86/entry/common.c:301 entry_SYSCALL_64_after_hwframe+0x49/0xbe This is caused by a race between two concurrenct md_ioctl()s closing the array. CPU1 (md_ioctl()) CPU2 (md_ioctl()) ------ ------ set_bit(MD_CLOSING, &mddev->flags); did_set_md_closing = true; WARN_ON_ONCE(test_bit(MD_CLOSING, &mddev->flags)); if(did_set_md_closing) clear_bit(MD_CLOSING, &mddev->flags); Fix the warning by returning immediately if the MD_CLOSING bit is set in &mddev->flags which indicates that the array is being closed. Fixes: 065e519e ("md: MD_CLOSING needs to be cleared after called md_set_readonly or do_md_stop") Reported-by: syzbot+1e46a0864c1a6e9bd3d8@syzkaller.appspotmail.com Cc: stable@vger.kernel.org Signed-off-by:Dae R. Jeong <dae.r.jeong@kaist.ac.kr> Signed-off-by:
-
- 16 Nov, 2020 24 commits
-
-
Jan Höppner authored
If the Fibre Channel Endpoint-Security status of a path changes, a corresponding path event is received from the CIO layer. Process this event by re-reading the FCES information. As the information is retrieved for all paths on a single CU in one call, the internal status can also be updated for all paths and no processing per path is necessary. Signed-off-by:
Jan Höppner <hoeppner@linux.ibm.com> Signed-off-by:
Stefan Haberland <sth@linux.ibm.com> Reviewed-by:
Cornelia Huck <cohuck@redhat.com> Signed-off-by:
Jens Axboe <axboe@kernel.dk>
-
Jan Höppner authored
As more path events need to be handled for ECKD the current path verification infrastructure can be reused. Rename all path verifcation code to fit the more broadly based task of path event handling and put the path verification in a new separate function. Signed-off-by:
-
Jan Höppner authored
Add a new sysfs attribute (fc_security) per device and per operational channel path. The information of the current FC Endpoint Security state is received through the CIO layer. The state of the FC Endpoint Security can be either "Unsupported", "Authentication", or "Encryption". For example: $ cat /sys/bus/ccw/devices/0.0.c600/fc_security Encryption If any of the operational paths is in a state different from all others, the device sysfs attribute will display the additional state "Inconsistent". The sysfs attributes per paths are organised in a new directory called "paths_info" with subdirectories for each path. /sys/bus/ccw/devices/0.0.c600/paths_info/ ├── 0.38 │ └── fc_security ├── 0.39 │ └── fc_security ├── 0.3a │ └── fc_security └── 0.3b └── fc_security Signed-off-by: -
Jan Höppner authored
During online processing and setting up a DASD device, the configuration data for operational paths is read and validated two times (dasd_eckd_read_conf()). The first time to provide information that are necessary for the LCU setup. A second time after the LCU setup as a device might report different configuration data then. When the configuration setup for each operational path is being validated, an initial call to dasd_eckd_clear_conf_data() is issued. This call wipes all previously available configuration data and path information for each path. However, the operational path mask is not updated during this process. As a result, the stored operational path mask might no longer correspond to the operational paths mask reported by the CIO layer, as several paths might be gone between the two dasd_eckd_read_conf() calls. This inconsistency leads to more severe issues in later path handling changes. Fix this by removing the channel paths from the operational path mask during the dasd_eckd_clear_conf_data() call. Signed-off-by:
-
Jan Höppner authored
Currently, the configuration data for a path is retrieved during a path verification and used only temporarily. If a path is newly added to the I/O setup after a boot, no configuration data will be stored for this particular path. However, this data is required for later use and should be present for a valid I/O path anyway. Store this data during the path verification so that newly added paths can provide all information necessary. [sth@linux.ibm.com: fix conf_data memleak] Signed-off-by:
-
Jan Höppner authored
For storing retrieved path information both the if and else block in dasd_eckd_read_conf() use the same code. To avoid duplicate code this should be done after the if/else block. To further increase readability, move the code to a new function, dasd_eckd_store_conf_data(). Signed-off-by:
-
Jan Höppner authored
The discipline argument in dasd_generic_probe() isn't used and there is no history how it was used in the past. Remove it. Signed-off-by:
-
Vineeth Vijayan authored
Fibre Channel Endpoint-Security event is received as an sei:nt0 type in the CIO layer. This information needs to be shared with the CCW device drivers using the path_events callback. Co-developed-by:
Sebastian Ott <sebott@linux.ibm.com> Signed-off-by:
Vineeth Vijayan <vneethv@linux.ibm.com> Signed-off-by:
Peter Oberparleiter <oberpar@linux.ibm.com> Acked-by:
Vasily Gorbik <gor@linux.ibm.com> Signed-off-by:
-
Vineeth Vijayan authored
Add an interface in the CIO layer to retrieve the information about the Endpoint-Security Mode (ESM) of the specified CU. The ESM values are defined as 0-None, 1-Authenticated or 2, 3-Encrypted. [vneethv@linux.ibm.com: cleaned-up and modified description] Signed-off-by:
-
Sebastian Ott authored
Add a new sysfs attribute 'esc' per chpid. This new attribute exports the Endpoint-Security-Capability byte of channel-path description block, which could be 0-None, 1-Authentication, 2 and 3-Encryption. For example: $ cat /sys/devices/css0/chp0.34/esc 0 [vneethv@linux.ibm.com: cleaned-up & modified description] Signed-off-by:
-
Christoph Hellwig authored
Switch the comment to talk about __register_blkdev instead of register_blkdev and document the new probe parameter. Fixes: 3da1a61e7046 ("block: add an optional probe callback to major_names") Reported-by:Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
-
Christoph Hellwig authored
Now that bdev_map is only used for finding gendisks, we can use a simple xarray instead of the regions tracking structure for it. Signed-off-by:
Hannes Reinecke <hare@suse.de> Reviewed-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Christoph Hellwig authored
Use separate gendisks (which share a tag_set) for the different operating modes instead of redirecting the gendisk lookup using a probe callback. This avoids potential problems with aliased block_device instances and will eventually allow for removing the blk_register_region framework. Signed-off-by:
-
Christoph Hellwig authored
reindent the driver using Lident as the code style was far away from normal Linux code. Signed-off-by:
-
Christoph Hellwig authored
The Atari floppy driver usually autodetects the media when used with the ormal /dev/fd? devices, which also are the only nodes created by udev. But it also supports various aliases that force a given media format. That is currently supported using the blk_register_region framework which finds the floppy gendisk even for a 'mismatched' dev_t. The problem with this (besides the code complexity) is that it creates multiple struct block_device instances for the whole device of a single gendisk, which can lead to interesting issues in code not aware of that fact. To fix this just create a separate gendisk for each of the aliases if they are accessed. Signed-off-by:
-
Christoph Hellwig authored
Use separate gendisks (which share a tag_set) for the native Amgiga vs the MS-DOS mode instead of redirecting the gendisk lookup using a probe callback. This avoids potential problems with aliased block_device instances and will eventually allow for removing the blk_register_region framework. Signed-off-by:
-
Christoph Hellwig authored
The floppy driver usually autodetects the media when used with the normal /dev/fd? devices, which also are the only nodes created by udev. But it also supports various aliases that force a given media format. That is currently supported using the blk_register_region framework which finds the floppy gendisk even for a 'mismatched' dev_t. The problem with this (besides the code complexity) is that it creates multiple struct block_device instances for the whole device of a single gendisk, which can lead to interesting issues in code not aware of that fact. To fix this just create a separate gendisk for each of the aliases if they are accessed. Signed-off-by:
-
Christoph Hellwig authored
ide is the last user of the blk_register_region framework except for the tracking of allocated gendisk. Switch to __register_blkdev, even if that doesn't allow us to trivially find out which command set to probe for. That means we now always request all modules when a user tries to access an unclaimed ide device node, but except for a few potentially loaded modules for a fringe use case of a deprecated and soon to be removed driver that doesn't make a difference. Signed-off-by:
-
Christoph Hellwig authored
Use the simpler mechanism attached to major_name to allocate a md device when a currently unregistered minor is accessed. Signed-off-by:
Song Liu <song@kernel.org> Signed-off-by:
-
Christoph Hellwig authored
Use the simpler mechanism attached to major_name to allocate a brd device when a currently unregistered minor is accessed. Signed-off-by:
-
Christoph Hellwig authored
Use the simpler mechanism attached to major_name to allocate a brd device when a currently unregistered minor is accessed. Signed-off-by:
-
Christoph Hellwig authored
Switch from using blk_register_region to the probe callback passed to __register_blkdev to disable the request_module call for an unclaimed dev_t in the SD majors. Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Christoph Hellwig authored
The swim driver (unlike various other floppy drivers) doesn't have magic device nodes for certain modes, and already registers a gendisk for each of the floppies supported by a device. Thus the region registered is a no-op and can be removed. Signed-off-by:
-
Christoph Hellwig authored
There is no need to ever register the fake gendisk used for ide-tape. Signed-off-by:
-
