- 07 Oct, 2020 1 commit
-
-
Luo Jiaxing authored
A call trace is observed when running function level reset with online CPUs less than 16 and MSI auto-affinity enabled. [16538.348038] Call trace: [16538.348422] pci_irq_vector+0x98/0xc0 [16538.348947] disable_host_v3_hw+0x8c/0x288 [hisi_sas_v3_hw] [16538.349706] hisi_sas_reset_prepare_v3_hw+0x60/0x88 [hisi_sas_v3_hw] [16538.350631] pci_dev_save_and_disable+0x38/0x68 [16538.351290] pci_reset_function+0x44/0x88 [16538.351846] reset_store+0x6c/0xb8 [16538.352429] dev_attr_store+0x44/0x60 [16538.353035] sysfs_kf_write+0x58/0x80 [16538.353558] kernfs_fop_write+0x140/0x230 [16538.354175] __vfs_write+0x48/0x80 [16538.354675] vfs_write+0xb8/0x1d8 [16538.355145] ksys_write+0x74/0xf8 [16538.355615] __arm64_sys_write+0x24/0x30 [16538.356240] el0_svc_common.constprop.4+0x80/0x1f0 [16538.356905] do_el0_svc+0x2c/0x38 [16538.357408] el0_svc+0x14/0x40 [16538.357848] el0_sync_handler+0xbc/0x2ec [16538.358388] el0_sync+0x140/0x180 The reason is that if we use pci_alloc_irq_vectors_affinity() to allocate IRQs, the number of CQ IRQs can only be less than or equal to the number of online CPUs, but we use hisi_hba->queue_count (always 16) to iterate during interrupt_disable_v3_hw(). Use hisi_hba->cq_nvecs to replace hisi_hba->queue_count to avoid synchronize IRQ on a CPU which does not exist. Link: https://lore.kernel.org/r/1601649038-25534-2-git-send-email-john.garry@huawei.comSigned-off-by:
Luo Jiaxing <luojiaxing@huawei.com> Signed-off-by:
John Garry <john.garry@huawei.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
- 03 Oct, 2020 19 commits
-
-
Jing Xiangfeng authored
This assignment is meaningless. Remove it. Link: https://lore.kernel.org/r/20200917021906.175933-1-jingxiangfeng@huawei.comSigned-off-by:
Jing Xiangfeng <jingxiangfeng@huawei.com> Signed-off-by:
-
Ye Bin authored
Fixes coccicheck warning: drivers/scsi/lpfc/lpfc_attr.c:5341:5-11: Unneeded variable: "status". Return "- EINVAL" on line 5342 Link: https://lore.kernel.org/r/20200916022859.349089-1-yebin10@huawei.comReported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
Ye Bin <yebin10@huawei.com> Signed-off-by:
-
Liu Shixin authored
Use DEFINE_SEQ_ATTRIBUTE macro to simplify the code. Link: https://lore.kernel.org/r/20200916025030.3992991-1-liushixin2@huawei.comSigned-off-by:
Liu Shixin <liushixin2@huawei.com> Signed-off-by:
-
Ye Bin authored
Fixes coccicheck warning: drivers/scsi/qla4xxx/ql4_init.c:1173:5-11: Unneeded variable: "status". Return "QLA_ERROR" on line 1195 Link: https://lore.kernel.org/r/20200916022749.348923-1-yebin10@huawei.comReported-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Liu Shixin authored
Use module_platform_driver() to eliminate boilerplate code. Link: https://lore.kernel.org/r/20200914065403.3726462-1-liushixin2@huawei.comSigned-off-by:
-
Jing Xiangfeng authored
Return PTR_ERR() from the error handling case instead of 0. Link: https://lore.kernel.org/r/20200910123848.93649-1-jingxiangfeng@huawei.comSigned-off-by:
-
Hannes Reinecke authored
If we fail to issue the iocb in lpfc_gen_req() we need to drop the nodelist reference. Link: https://lore.kernel.org/r/20200910084059.138507-1-hare@suse.deSigned-off-by:
Hannes Reinecke <hare@suse.de> Signed-off-by:
-
Dan Carpenter authored
The be_fill_queue() function can only fail when "eq_vaddress" is NULL and since it's non-NULL here that means the function call can't fail. But imagine if it could, then in that situation we would want to store the "paddr" so that dma memory can be released. Link: https://lore.kernel.org/r/20200928091300.GD377727@mwanda Fixes: bfead3b2 ("[SCSI] be2iscsi: Adding msix and mcc_rings V3") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
-
Tomas Henzl authored
Add a missing resource cleanup in _aac_reset_adapter. Link: https://lore.kernel.org/r/20200926150015.6187-1-thenzl@redhat.comSigned-off-by:
Tomas Henzl <thenzl@redhat.com> Signed-off-by:
-
Manish Rangankar authored
On fan failure event from MFW, bring down active connections and unload the firmware context. Link: https://lore.kernel.org/r/20200924070338.8270-1-mrangankar@marvell.comSigned-off-by:
Manish Rangankar <mrangankar@marvell.com> Signed-off-by:
-
John Donnelly authored
Corrects drivers/target/target_core_user.c:688:6: warning: 'page' may be used uninitialized. Link: https://lore.kernel.org/r/20200924001920.43594-1-john.p.donnelly@oracle.com Fixes: 3c58f737 ("scsi: target: tcmu: Optimize use of flush_dcache_page") Cc: Mike Christie <michael.christie@oracle.com> Acked-by:
Mike Christie <michael.christie@oracle.com> Signed-off-by:
John Donnelly <john.p.donnelly@oracle.com> Signed-off-by:
-
Ye Bin authored
Fix the following warnings: [drivers/scsi/fnic/fnic_debugfs.c:123]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'int'. [drivers/scsi/fnic/fnic_debugfs.c:125]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'int'. [drivers/scsi/fnic/fnic_debugfs.c:127]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'int'. Link: https://lore.kernel.org/r/20200930021919.2832860-2-yebin10@huawei.comReported-by:
Karan Tilak Kumar <kartilak@cisco.com> Signed-off-by:
-
Jing Xiangfeng authored
ret is always zero or error so the assignment is redundant. Link: https://lore.kernel.org/r/20200925060754.156599-1-jingxiangfeng@huawei.comAcked-by:
-
Hannes Reinecke authored
The fnic drivers assigns an ioreq structure to each command and severs this assignment once scsi_done() has been called and the command has been completed. When traversing commands to terminate outstanding I/O we should not call scsi_done() on commands which do not have a corresponding ioreq structure; these commands have either never entered the driver or have already been completed. [mkp: fixed unused label warning] Link: https://lore.kernel.org/r/20200515112647.49260-1-hare@suse.deReported-by:
kbuild test robot <lkp@intel.com> Reviewed-by:
Laurence Oberman <loberman@redhat.com> Acked-by:
Satish Kharat <satishkh@cisco.com> Acked-by:
Hannes Reinecke <hare@suse.com> Signed-off-by:
-
- 02 Oct, 2020 7 commits
-
-
Pujin Shi authored
For older versions of gcc, the array = {0}; will cause warnings: drivers/scsi/ufs/ufshcd-crypto.c: In function 'ufshcd_crypto_keyslot_program': drivers/scsi/ufs/ufshcd-crypto.c:62:8: warning: missing braces around initializer [-Wmissing-braces] union ufs_crypto_cfg_entry cfg = { 0 }; ^ drivers/scsi/ufs/ufshcd-crypto.c:62:8: warning: (near initialization for 'cfg.reg_val') [-Wmissing-braces] drivers/scsi/ufs/ufshcd-crypto.c: In function 'ufshcd_clear_keyslot': drivers/scsi/ufs/ufshcd-crypto.c:103:8: warning: missing braces around initializer [-Wmissing-braces] union ufs_crypto_cfg_entry cfg = { 0 }; ^ 2 warnings generated Link: https://lore.kernel.org/r/20201002063538.1250-1-shipujin.t@gmail.com Fixes: 70297a8a ("scsi: ufs: UFS crypto API") Reviewed-by:Eric Biggers <ebiggers@google.com> Signed-off-by:
Pujin Shi <shipujin.t@gmail.com> Signed-off-by:
-
Ye Bin authored
Fix the following warning: [drivers/scsi/qla2xxx/qla_dbg.c:2451]: (warning) %ld in format string (no. 4) requires 'long' but the argument type is 'unsigned long'. Link: https://lore.kernel.org/r/20200930022515.2862532-4-yebin10@huawei.comReported-by:
Himanshu Madhani <himanshu.madhani@oracle.com> Reviewed-by:
Nilesh Javali <njavali@marvell.com> Signed-off-by:
-
Ye Bin authored
Fix the following warnings: [drivers/scsi/qla2xxx/qla_os.c:4882]: (warning) %ld in format string (no. 2) requires 'long' but the argument type is 'unsigned long'. [drivers/scsi/qla2xxx/qla_os.c:5011]: (warning) %ld in format string (no. 1) requires 'long' but the argument type is 'unsigned long'. Link: https://lore.kernel.org/r/20200930022515.2862532-3-yebin10@huawei.comReported-by:
-
Ye Bin authored
Fix the following warnings: [drivers/scsi/qla2xxx/tcm_qla2xxx.c:884]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'signed int'. [drivers/scsi/qla2xxx/tcm_qla2xxx.c:885]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'signed int'. [drivers/scsi/qla2xxx/tcm_qla2xxx.c:886]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'signed int'. [drivers/scsi/qla2xxx/tcm_qla2xxx.c:887]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'signed int'. [drivers/scsi/qla2xxx/tcm_qla2xxx.c:888]: (warning) %u in format string (no. 1) requires 'unsigned int' but the argument type is 'signed int'. Link: https://lore.kernel.org/r/20200930022515.2862532-2-yebin10@huawei.comReported-by:
-
Mike Christie authored
Some iSCSI targets went with the traditional "export N ports" approach and then allowed the initiator to multipath over them. Other targets went the opposite direction and export a single port, and then software on the target side performs load balancing and failover to other targets via an iSCSI specific feature or IP takover. The problem for the 2nd type of config is we quickly run out of our five retries and get I/O errors. In these setups we want to reduce resource use on the initiator side so we only wanted the one session and no dm-multipath. To handle traditional multipath operations like failover we do IP takover on the target side. So we would have an iSCSI target running on node1. Some monitoring software decides it's dead or the node is overloaded so it starts the iSCSI target on node2. The problem is for the failover case where we might have the equivalent of a dm-multipath temporary all paths down, or we just have to try more than 5 nodes before finding a good one. To handle this type of issue allow the user to configure the disk cmd retries from -1 to the current max of 5. -1 means infinite retries and should be used for setups where some other setting is going to control when to fail. For example iSCSI has the replacement/recovery timeout and fc (some users have used FC with NPIV and done something similar as IP takover) has dev_loss_tmo/fast_io_fail which will eventually expire and fail I/O. Link: https://lore.kernel.org/r/1601566554-26752-3-git-send-email-michael.christie@oracle.comReviewed-by:
Bart Van Assche <bvanassche@acm.org> Signed-off-by:
-
Mike Christie authored
Add infinite retry support to SCSI midlayer by combining common checks for retries into some helper functions, and then checking for the -1/SCSI_CMD_RETRIES_NO_LIMIT. Link: https://lore.kernel.org/r/1601566554-26752-2-git-send-email-michael.christie@oracle.comReviewed-by:
-
Roman Bolshakov authored
trace-cmd report doesn't show events from target subsystem because scsi_command_size() leaks through event format string: [target:target_sequencer_start] function scsi_command_size not defined [target:target_cmd_complete] function scsi_command_size not defined Addition of scsi_command_size() to plugin_scsi.c in trace-cmd doesn't help because an expression is used inside TP_printk(). trace-cmd event parser doesn't understand minus sign inside [ ]: Error: expected ']' but read '-' Rather than duplicating kernel code in plugin_scsi.c, provide a dedicated field for CONTROL byte. Link: https://lore.kernel.org/r/20200929125957.83069-1-r.bolshakov@yadro.comReviewed-by:
Roman Bolshakov <r.bolshakov@yadro.com> Signed-off-by:
-
- 30 Sep, 2020 8 commits
-
-
Nilesh Javali authored
Link: https://lore.kernel.org/r/20200929102152.32278-8-njavali@marvell.comReviewed-by:
-
Arun Easi authored
Driver was using a shorter timeout waiting for PLOGI from the peer in point-to-point configurations. Some devices takes some time (~4 seconds) to initiate the PLOGI. This peer initiating PLOGI is when the peer has a higher P-WWN. Increase the wait time based on N2N R_A_TOV. Link: https://lore.kernel.org/r/20200929102152.32278-7-njavali@marvell.comReviewed-by:
Arun Easi <aeasi@marvell.com> Signed-off-by:
-
Quinn Tran authored
On unload, session cleanup prematurely gave the signal for driver unload path to advance. Link: https://lore.kernel.org/r/20200929102152.32278-6-njavali@marvell.com Fixes: 726b8548 ("qla2xxx: Add framework for async fabric discovery") Cc: stable@vger.kernel.org Reviewed-by:
Quinn Tran <qutran@marvell.com> Signed-off-by:
-
Arun Easi authored
Normally, the MPI firmware is reset when an MPI dump is collected. If an unsaved MPI dump exists in the driver, though, an alternate mechanism is used. This mechanism, which was not fully correct, is not recommended and instead an MPI dump template walk is suggested to perform the MPI reset. To allow for the MPI dump template walk, extra space is reserved in the MPI dump buffer which gets used only when there is already an MPI dump in place. Link: https://lore.kernel.org/r/20200929102152.32278-5-njavali@marvell.com Fixes: cbb01c2f ("scsi: qla2xxx: Fix MPI failure AEN (8200) handling") Cc: stable@vger.kernel.org Reviewed-by:
-
Arun Easi authored
When printing the message: "MPI Heartbeat stop. MPI reset is not needed.." ..the wrong register was checked leading to always printing that MPI reset is not needed, even when it is needed. Fix the MPI reset message. Link: https://lore.kernel.org/r/20200929102152.32278-4-njavali@marvell.com Fixes: cbb01c2f ("scsi: qla2xxx: Fix MPI failure AEN (8200) handling") Cc: stable@vger.kernel.org Reviewed-by:
-
Quinn Tran authored
Current code uses wrong mailbox option to extract bbc from firmware. This field is nested inside of PLOGI payload. Extract bbc from PLOGI template payload. Link: https://lore.kernel.org/r/20200929102152.32278-3-njavali@marvell.comReviewed-by:
-
Saurav Kashyap authored
Since the version string has been modified, sscanf() returns 4 instead of 6. Link: https://lore.kernel.org/r/20200929102152.32278-2-njavali@marvell.comReviewed-by:
Saurav Kashyap <skashyap@marvell.com> Signed-off-by:
-
Bean Huo authored
Use devm_platform_ioremap_resource_byname() to simplify the code. Link: https://lore.kernel.org/r/20200916084017.14086-1-huobean@gmail.comAcked-by:
Alim Akhtar <alim.akhtar@samsung.com> Signed-off-by:
Bean Huo <beanhuo@micron.com> Signed-off-by:
-
- 23 Sep, 2020 5 commits
-
-
Stanley Chu authored
Add "mediatek,mt8192-ufshci" compatible string to for MediaTek UFS host controller present on MT8192 chipsets. Link: https://lore.kernel.org/r/20200914050052.3974-3-stanley.chu@mediatek.comAcked-by:
Rob Herring <robh@kernel.org> Signed-off-by:
Stanley Chu <stanley.chu@mediatek.com> Signed-off-by:
-
Stanley Chu authored
Some MediaTek UFS platforms support high-performance mode that inline encryption engine can be boosted while UFS is not clock-gated. The high-performance mode will be enabled if all below conditions are well-declaired in device tree, - Proper platform-specific compatible string which enables the host capability "UFS_MTK_CAP_BOOST_CRYPT_ENGINE". - "dvfsrc-vcore" node is available in this platform. - Required minimum vcore voltage for high-performance mode. - Clock mux and clock parents of inline encryption engine for both "low-power mode" and "high-performance mode". Link: https://lore.kernel.org/r/20200914050052.3974-2-stanley.chu@mediatek.comSigned-off-by:
-
Liu Shixin authored
Simplify the return expression. Link: https://lore.kernel.org/r/20200921134558.3478922-1-liushixin2@huawei.comReviewed-by:
Jason Yan <yanaijie@huawei.com> Signed-off-by:
-
Jing Xiangfeng authored
No point in initializing ret with -ENOMEM. Link: https://lore.kernel.org/r/20200918090747.44645-1-jingxiangfeng@huawei.comAcked-by:
Finn Thain <fthain@telegraphics.com.au> Signed-off-by:
-
Qinglang Miao authored
The mutex bnx2i_dev_lock is initialized statically. It is unnecessary to initialize by mutex_init(). Link: https://lore.kernel.org/r/20200916062133.191000-1-miaoqinglang@huawei.comAcked-by:
Qinglang Miao <miaoqinglang@huawei.com> Signed-off-by:
-
