- 27 May, 2020 40 commits
-
-
Coly Li authored
In order to avoid the experimental async registration interface to be treated as new kernel ABI for common users, this patch makes it as an experimental kernel configure BCACHE_ASYNC_REGISTRAION. This interface is for extreme large cached data situation, to make sure the bcache device can always created without the udev timeout issue. For normal users the async or sync registration does not make difference. In future when we decide to use the asynchronous registration as default behavior, this experimental interface may be removed. Signed-off-by:
Coly Li <colyli@suse.de> Signed-off-by:
Jens Axboe <axboe@kernel.dk>
-
Coly Li authored
When there is a lot of data cached on cache device, the bcach internal btree can take a very long to validate during the backing device and cache device registration. In my test, it may takes 55+ minutes to check all the internal btree nodes. The problem is that the registration is invoked by udev rules and the udevd has 180 seconds timeout by default. If the btree node checking time is longer than udevd timeout, the registering process will be killed by udevd with SIGKILL. If the registering process has pending sigal, creating kthread for bcache will fail and the device registration will fail. The result is, for bcache device which cached a lot of data on cache device, the bcache device node like /dev/bcache<N> won't create always due to the very long btree checking time. A solution to avoid the udevd 180 seconds timeout is to register devices in an asynchronous way. Which is, after writing cache or backing device path into /sys/fs/bcache/register_async, the kernel code will create a kworker and move all the btree node checking (for cache device) or dirty data counting (for cached device) in the kwork context. Then the kworder is scheduled on system_wq and the registration code just returned to user space udev rule task. By this asynchronous way, the udev task for bcache rule will complete in seconds, no matter how long time spent in the kworker context, it won't be killed by udevd for a timeout. After all the checking and counting are done asynchronously in the kworker, the bcache device will eventually be created successfully. This patch does the above chagne and add a register sysfs file /sys/fs/bcache/register_async. Writing the registering device path into this sysfs file will do the asynchronous registration. The register_async interface is for very rare condition and won't be used for common users. In future I plan to make the asynchronous registration as default behavior, which depends on feedback for this patch. Signed-off-by:
-
Coly Li authored
The problematic code piece in bcache_device_free() is, 785 static void bcache_device_free(struct bcache_device *d) 786 { 787 struct gendisk *disk = d->disk; [snipped] 799 if (disk) { 800 if (disk->flags & GENHD_FL_UP) 801 del_gendisk(disk); 802 803 if (disk->queue) 804 blk_cleanup_queue(disk->queue); 805 806 ida_simple_remove(&bcache_device_idx, 807 first_minor_to_idx(disk->first_minor)); 808 put_disk(disk); 809 } [snipped] 816 } At line 808, put_disk(disk) may encounter kobject refcount of 'disk' being underflow. Here is how to reproduce the issue, - Attche the backing device to a cache device and do random write to make the cache being dirty. - Stop the bcache device while the cache device has dirty data of the backing device. - Only register the backing device back, NOT register cache device. - The bcache device node /dev/bcache0 won't show up, because backing device waits for the cache device shows up for the missing dirty data. - Now echo 1 into /sys/fs/bcache/pendings_cleanup, to stop the pending backing device. - After the pending backing device stopped, use 'dmesg' to check kernel message, a use-after-free warning from KASA reported the refcount of kobject linked to the 'disk' is underflow. The dropping refcount at line 808 in the above code piece is added by add_disk(d->disk) in bch_cached_dev_run(). But in the above condition the cache device is not registered, bch_cached_dev_run() has no chance to be called and the refcount is not added. The put_disk() for a non- added refcount of gendisk kobject triggers a underflow warning. This patch checks whether GENHD_FL_UP is set in disk->flags, if it is not set then the bcache device was not added, don't call put_disk() and the the underflow issue can be avoided. Signed-off-by: -
Joe Perches authored
Remove the trailing newline from the define of pr_fmt and add newlines to the uses. Miscellanea: o Convert bch_bkey_dump from multiple uses of pr_err to pr_cont as the earlier conversion was inappropriate done causing multiple lines to be emitted where only a single output line was desired o Use vsprintf extension %pV in bch_cache_set_error to avoid multiple line output where only a single line output was desired o Coalesce formats Fixes: 6ae63e35 ("bcache: replace printk() by pr_*() routines") Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Colin Ian King authored
Variables i and n are being assigned but are never used. They are redundant and can be removed. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
git://git.infradead.org/nvmeJens Axboe authored
Pull NVMe updates from Christoph: "The second large batch of nvme updates: - t10 protection information support for nvme-rdma and nvmet-rdma (Israel Rukshin and Max Gurtovoy) - target side AEN improvements (Chaitanya Kulkarni) - various fixes and minor improvements all over, icluding the nvme part of the lpfc driver" * 'nvme-5.8' of git://git.infradead.org/nvme: (38 commits) lpfc: Fix return value in __lpfc_nvme_ls_abort lpfc: fix axchg pointer reference after free and double frees lpfc: Fix pointer checks and comments in LS receive refactoring nvme: set dma alignment to qword nvmet: cleanups the loop in nvmet_async_events_process nvmet: fix memory leak when removing namespaces and controllers concurrently nvmet-rdma: add metadata/T10-PI support nvmet: add metadata support for block devices nvmet: add metadata/T10-PI support nvme: add Metadata Capabilities enumerations nvmet: rename nvmet_check_data_len to nvmet_check_transfer_len nvmet: rename nvmet_rw_len to nvmet_rw_data_len nvmet: add metadata characteristics for a namespace nvme-rdma: add metadata/T10-PI support nvme-rdma: introduce nvme_rdma_sgl structure nvme: introduce NVME_INLINE_METADATA_SG_CNT nvme: enforce extended LBA format for fabrics metadata nvme: introduce max_integrity_segments ctrl attribute nvme: make nvme_ns_has_pi accessible to transports nvme: introduce NVME_NS_METADATA_SUPPORTED flag ...
-
James Smart authored
A static checker reported the following issue: drivers/scsi/lpfc/lpfc_nvmet.c:1366 lpfc_nvmet_ls_abort() warn: 'ret' can be either negative or positive The comment indicates a non-zero value indicates error in the form of -Exxx, but the code is returning "1". Fix the code to return -EINVAL to be compliant to comment. Fixes: e96a22b0 ("lpfc: Refactor Send LS Abort support") Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by:
James Smart <jsmart2021@gmail.com> Reviewed-by:
Christoph Hellwig <hch@lst.de>
-
James Smart authored
The axchg structure is a structure allocated early in the lpfc_nvme_unsol_ls_handler() to represent the newly received exchange. Upon error, the out_fail path in the routine unconditionally frees the pointer, yet subsequently passes the pointer to the abort routine. Additionally, the abort routine, lpfc_nvme_unsol_ls_issue_abort(), also has a failure path that will attempt to delete the pointer on error. Fix these errors by: - Removing the unconditional free so that it stays valid if passed to the abort routine. - Revise the abort routine to not free the pointer. Instead, return a success/failure status. Note: if success, the later completion of the abort frees the structure. - Back in the unsol_ls_handler() error path, if the abort routine was skipped (thus no possible reference) or the abort routine returned error, free the pointer. Fixes: 3a8070c5 ("lpfc: Refactor NVME LS receive handling") Reported-by:
-
James Smart authored
Additional testing encountered null pointers that weren't fully qualified in lpfc_nvmet_xmt_ls_abort_cmp() and lpfc_nvmet_unsol_issue_abort(). The same error was detected and reported by static checker reporting: drivers/scsi/lpfc/lpfc_sli.c:2905 lpfc_nvme_unsol_ls_handler() error: we previously assumed 'phba->targetport' could be null (see line 2837) Fix by making phba->nvmet_support and phba->targetport validity checks in lpfc_nvmet_xmt_ls_abort_cmp() and lpfc_nvmet_unsol_issue_abort(). Fixes: 3a8070c5 (“lpfc: Refactor NVME LS receive handling”) Reported-by:Paul Ely <paul.ely@broadcom.com> Signed-off-by:
-
Keith Busch authored
The default dma alignment mask is 511, which is much larger than any nvme controller requires. NVMe controllers accept qword aligned DMA addresses, so set the request_queue constraints to that. This can help avoid bounce buffers on user passthrough commands. Signed-off-by:
Keith Busch <kbusch@kernel.org> Signed-off-by:
-
David Milburn authored
Based-on-a-patch-by:
Christoph Hellwig <hch@infradead.org> Tested-by:
Yi Zhang <yi.zhang@redhat.com> Signed-off-by:
David Milburn <dmilburn@redhat.com> Signed-off-by:
-
Sagi Grimberg authored
When removing a namespace, we add an NS_CHANGE async event, however if the controller admin queue is removed after the event was added but not yet processed, we won't free the aens, resulting in the below memory leak [1]. Fix that by moving nvmet_async_event_free to the final controller release after it is detached from subsys->ctrls ensuring no async events are added, and modify it to simply remove all pending aens. -- $ cat /sys/kernel/debug/kmemleak unreferenced object 0xffff888c1af2c000 (size 32): comm "nvmetcli", pid 5164, jiffies 4295220864 (age 6829.924s) hex dump (first 32 bytes): 28 01 82 3b 8b 88 ff ff 28 01 82 3b 8b 88 ff ff (..;....(..;.... 02 00 04 65 76 65 6e 74 5f 66 69 6c 65 00 00 00 ...event_file... backtrace: [<00000000217ae580>] nvmet_add_async_event+0x57/0x290 [nvmet] [<0000000012aa2ea9>] nvmet_ns_changed+0x206/0x300 [nvmet] [<00000000bb3fd52e>] nvmet_ns_disable+0x367/0x4f0 [nvmet] [<00000000e91ca9ec>] nvmet_ns_free+0x15/0x180 [nvmet] [<00000000a15deb52>] config_item_release+0xf1/0x1c0 [<000000007e148432>] configfs_rmdir+0x555/0x7c0 [<00000000f4506ea6>] vfs_rmdir+0x142/0x3c0 [<0000000000acaaf0>] do_rmdir+0x2b2/0x340 [<0000000034d1aa52>] do_syscall_64+0xa5/0x4d0 [<00000000211f13bc>] entry_SYSCALL_64_after_hwframe+0x6a/0xdf Fixes: a07b4970 ("nvmet: add a generic NVMe target") Reported-by:Sagi Grimberg <sagi@grimberg.me> Tested-by:
-
Israel Rukshin authored
For capable HCAs (e.g. ConnectX-5/ConnectX-6) this will allow end-to-end protection information passthrough and validation for NVMe over RDMA transport. Metadata support was implemented over the new RDMA signature verbs API. Signed-off-by:
Israel Rukshin <israelr@mellanox.com> Signed-off-by:
Max Gurtovoy <maxg@mellanox.com> Signed-off-by:
-
Israel Rukshin authored
Allocate the metadata SGL buffers and set metadata fields for the request. Then create a block IO request for the metadata from the protection SG list. Signed-off-by:
-
Israel Rukshin authored
Expose the namespace metadata format when PI is enabled. The user needs to enable the capability per subsystem and per port. The other metadata properties are taken from the namespace/bdev. Usage example: echo 1 > /config/nvmet/subsystems/${NAME}/attr_pi_enable echo 1 > /config/nvmet/ports/${PORT_NUM}/param_pi_enable Signed-off-by:James Smart <james.smart@broadcom.com> Signed-off-by:
-
Israel Rukshin authored
The enumerations will be used to expose the namespace metadata format by the target. Suggested-by:
Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by:
-
Israel Rukshin authored
The function doesn't check only the data length, because the transfer length includes also the metadata length in some cases. This is preparation for adding metadata (T10-PI) support. Signed-off-by:
-
Israel Rukshin authored
The function doesn't add the metadata length (only data length is calculated). This is preparation for adding metadata (T10-PI) support. Signed-off-by:
-
Israel Rukshin authored
Fill those namespace fields from the block device format for adding metadata (T10-PI) over fabric support with block devices. Signed-off-by:
-
Max Gurtovoy authored
For capable HCAs (e.g. ConnectX-5/ConnectX-6) this will allow end-to-end protection information passthrough and validation for NVMe over RDMA transport. Metadata offload support was implemented over the new RDMA signature verbs API and it is enabled for capable controllers. Signed-off-by:
-
Israel Rukshin authored
Remove first_sgl pointer from struct nvme_rdma_request and use pointer arithmetic instead. The inline scatterlist, if exists, will be located right after the nvme_rdma_request. This patch is needed as a preparation for adding PI support. Signed-off-by:
-
Israel Rukshin authored
SGL size of metadata is usually small. Thus, 1 inline sg should cover most cases. The macro will be used for pre-allocate a single SGL entry for metadata. The preallocation of small inline SGLs depends on SG_CHAIN capability so if the ARCH doesn't support SG_CHAIN, use the runtime allocation for the SGL. This patch is a preparation for adding metadata (T10-PI) over fabric support. Signed-off-by:
-
Max Gurtovoy authored
An extended LBA is a larger LBA that is created when metadata associated with the LBA is transferred contiguously with the LBA data (AKA interleaved). The metadata may be either transferred as part of the LBA (creating an extended LBA) or it may be transferred as a separate contiguous buffer of data. According to the NVMeoF spec, a fabrics ctrl supports only an Extended LBA format. Fail revalidation in case we have a spec violation. Also add a flag that will imply on capable transports and controllers as part of a preparation for allowing end-to-end protection information for fabric controllers. Suggested-by:
-
Max Gurtovoy authored
This patch doesn't change any logic, and is needed as a preparation for adding PI support for fabrics drivers that will use an extended LBA format for metadata and will support more than 1 integrity segment. Signed-off-by:
-
James Smart authored
Move the nvme_ns_has_pi() inline from core.c to the nvme.h header. This allows use by the transports. Signed-off-by:
-
Max Gurtovoy authored
This is a preparation for adding support for metadata in fabric controllers. New flag will imply that NVMe namespace supports getting metadata that was originally generated by host's block layer. Signed-off-by:
-
Max Gurtovoy authored
Replace the specific ext boolean (that implies on extended LBA format) with a feature in the new namespace features flag. This is a preparation for adding more namespace features (such as metadata specific features). Signed-off-by:
-
Max Gurtovoy authored
This will reduce the amount of ifdefs inside the source code for various drivers and also will reduce the amount of stub functions that were created for the !CONFIG_BLK_DEV_INTEGRITY case. Suggested-by:
-
Chaitanya Kulkarni authored
Add a new attribute "revalidate_size" for the namespace which allows user to revalidate and generate the AEN if needed. This attribute is needed so that we can install userspace rules with systemd service based on inotify/fsnotify/uevent. The registered callback for such a service will end up writing to this attribute to generate AEN if needed. Signed-off-by:
Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com> Reviewed-by:
Sagi Grimberg <sagi@grimbeg.me> Signed-off-by:
-
Chaitanya Kulkarni authored
The newly added function nvmet_ns_revalidate() does update the ns size in the identify namespace in-core target data structure when host issues id-ns command. This can lead to host having inconsistencies between size of the namespace present in the id-ns command result and size of the corresponding block device until host scans the namespaces explicitly. To avoid this scenario generate AEN if old size is not same as the new one in nvmet_ns_revalidate(). This will allow automatic AEN generation when host calls id-ns command and also allows target to install userspace rules so that it can trigger nvmet_ns_revalidate() (using configfs interface with the help of next patch) resulting in appropriate AEN generation when underlying namespace size change is detected. Signed-off-by:
-
Chaitanya Kulkarni authored
This patch adds a wrapper helper to indicate size change in the bdev & file-backed namespace when revalidating ns. This helper is needed in order to minimize code repetition in the next patch for configfs.c and existing admin-cmd.c. Signed-off-by:
-
Chaitanya Kulkarni authored
This adds a new tracepoint for the target to trace async event. This is helpful in debugging and comparing host and target side async events especially when host is connected to different targets on different machines and now that we rely on userspace components to generate AEN. Signed-off-by:
-
Dan Carpenter authored
The nvme_put_ctrl() is implemented earlier as an inline function so this declaration isn't required. Signed-off-by:
-
Gustavo A. R. Silva authored
The current codebase makes use of the zero-length array language extension to the C90 standard, but the preferred mechanism to declare variable-length types such as these ones is a flexible array member[1][2], introduced in C99: struct foo { int stuff; struct boo array[]; }; By making use of the mechanism above, we will get a compiler warning in case the flexible array does not occur last in the structure, which will help us prevent some kind of undefined behavior bugs from being inadvertently introduced[3] to the codebase from now on. Also, notice that, dynamic memory allocations won't be affected by this change: "Flexible array members have incomplete type, and so the sizeof operator may not be applied. As a quirk of the original implementation of zero-length arrays, sizeof evaluates to zero."[1] sizeof(flexible-array-member) triggers a warning because flexible array members have incomplete type[1]. There are some instances of code in which the sizeof operator is being incorrectly/erroneously applied to zero-length arrays and the result is zero. Such instances may be hiding some bugs. So, this work (flexible-array member conversions) will also help to get completely rid of those sorts of issues. This issue was found with the help of Coccinelle. [1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html [2] https://github.com/KSPP/linux/issues/21 [3] commit 76497732 ("cxgb3/l2t: Fix undefined behaviour") Signed-off-by:Gustavo A. R. Silva <gustavoars@kernel.org> Signed-off-by:
-
Damien Le Moal authored
Currently, a namespace io_opt queue limit is set by default to the physical sector size of the namespace and to the the write optimal size (NOWS) when the namespace reports optimal IO sizes. This causes problems with block limits stacking in blk_stack_limits() when a namespace block device is combined with an HDD which generally do not report any optimal transfer size (io_opt limit is 0). The code: /* Optimal I/O a multiple of the physical block size? */ if (t->io_opt & (t->physical_block_size - 1)) { t->io_opt = 0; t->misaligned = 1; ret = -1; } in blk_stack_limits() results in an error return for this function when the combined devices have different but compatible physical sector sizes (e.g. 512B sector SSD with 4KB sector disks). Fix this by not setting the optimal IO size queue limit if the namespace does not report an optimal write size value. Signed-off-by:Damien Le Moal <damien.lemoal@wdc.com> Reviewed-by:
Bart van Assche <bvanassche@acm.org> Reviewed-by:
Hannes Reinecke <hare@suse.de> Signed-off-by:
-
Wu Bo authored
Disable streams again if getting the stream params fails. Signed-off-by:
Wu Bo <wubo40@huawei.com> Signed-off-by:
-
Martin George authored
The nvme-fc devloss_tmo is computed as the min of either the ctrl_loss_tmo (max_retries * reconnect_delay) or the remote port's devloss_tmo. But what gets printed as the nvme-fc devloss_tmo in nvme_fc_reconnect_or_delete() is always the remote port's devloss_tmo value. So correct this by printing the min value instead. Signed-off-by:
Martin George <marting@netapp.com> Reviewed-by:
-
Weiping Zhang authored
Check module parameter write/poll_queues before using it to catch too large values. Reproducer: modprobe -r nvme modprobe nvme write_queues=`nproc` echo $((`nproc`+1)) > /sys/module/nvme/parameters/write_queues echo 1 > /sys/block/nvme0n1/device/reset_controller [ 657.069000] ------------[ cut here ]------------ [ 657.069022] WARNING: CPU: 10 PID: 1163 at kernel/irq/affinity.c:390 irq_create_affinity_masks+0x47c/0x4a0 [ 657.069056] dm_region_hash dm_log dm_mod [ 657.069059] CPU: 10 PID: 1163 Comm: kworker/u193:9 Kdump: loaded Tainted: G W 5.6.0+ #8 [ 657.069060] Hardware name: Inspur SA5212M5/YZMB-00882-104, BIOS 4.0.9 08/27/2019 [ 657.069064] Workqueue: nvme-reset-wq nvme_reset_work [nvme] [ 657.069066] RIP: 0010:irq_create_affinity_masks+0x47c/0x4a0 [ 657.069067] Code: fe ff ff 48 c7 c0 b0 89 14 95 48 89 46 20 e9 e9 fb ff ff 31 c0 e9 90 fc ff ff 0f 0b 48 c7 44 24 08 00 00 00 00 e9 e9 fc ff ff <0f> 0b e9 87 fe ff ff 48 8b 7c 24 28 e8 33 a0 80 00 e9 b6 fc ff ff [ 657.069068] RSP: 0018:ffffb505ce1ffc78 EFLAGS: 00010202 [ 657.069069] RAX: 0000000000000060 RBX: ffff9b97921fe5c0 RCX: 0000000000000000 [ 657.069069] RDX: ffff9b67bad80000 RSI: 00000000ffffffa0 RDI: 0000000000000000 [ 657.069070] RBP: 0000000000000000 R08: 0000000000000000 R09: ffff9b97921fe718 [ 657.069070] R10: ffff9b97921fe710 R11: 0000000000000001 R12: 0000000000000064 [ 657.069070] R13: 0000000000000060 R14: 0000000000000000 R15: 0000000000000001 [ 657.069071] FS: 0000000000000000(0000) GS:ffff9b67c0880000(0000) knlGS:0000000000000000 [ 657.069072] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 657.069072] CR2: 0000559eac6fc238 CR3: 000000057860a002 CR4: 00000000007606e0 [ 657.069073] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 657.069073] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 657.069073] PKRU: 55555554 [ 657.069074] Call Trace: [ 657.069080] __pci_enable_msix_range+0x233/0x5a0 [ 657.069085] ? kernfs_put+0xec/0x190 [ 657.069086] pci_alloc_irq_vectors_affinity+0xbb/0x130 [ 657.069089] nvme_reset_work+0x6e6/0xeab [nvme] [ 657.069093] ? __switch_to_asm+0x34/0x70 [ 657.069094] ? __switch_to_asm+0x40/0x70 [ 657.069095] ? nvme_irq_check+0x30/0x30 [nvme] [ 657.069098] process_one_work+0x1a7/0x370 [ 657.069101] worker_thread+0x1c9/0x380 [ 657.069102] ? max_active_store+0x80/0x80 [ 657.069103] kthread+0x112/0x130 [ 657.069104] ? __kthread_parkme+0x70/0x70 [ 657.069105] ret_from_fork+0x35/0x40 [ 657.069106] ---[ end trace f4f06b7d24513d06 ]--- [ 657.077110] nvme nvme0: 95/1/0 default/read/poll queues Signed-off-by:
Weiping Zhang <zhangweiping@didiglobal.com> Signed-off-by:
-
Sagi Grimberg authored
Have routines handle errors and just bail out of the poll loop. This simplifies the code and will help as we may enhance the poll loop logic and these are somewhat in the way. Signed-off-by:
-
Sagi Grimberg authored
when trying to send the pdu data digest, we should set this flag. Reported-by:
-
