- 10 Sep, 2020 1 commit
-
-
Daniel Wagner authored
Emit a warning when ->done or ->free are called on an already freed srb. There is a hidden use-after-free bug in the driver which corrupts the srb memory pool which originates from the cleanup callbacks. An extensive search didn't bring any lights on the real problem. The initial fix was to set both pointers to NULL and try to catch invalid accesses. But instead the memory corruption was gone and the driver didn't crash. Since not all calling places check for NULL pointer, add explicitly default handlers. With this we workaround the memory corruption and add a debug help. Link: https://lore.kernel.org/r/20200908081516.8561-2-dwagner@suse.deReviewed-by:
Martin Wilck <mwilck@suse.com> Reviewed-by:
Arun Easi <aeasi@marvell.com> Signed-off-by:
Daniel Wagner <dwagner@suse.de> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
- 09 Sep, 2020 23 commits
-
-
Saurav Kashyap authored
During recovery due to FCoE fn ramrod failure we wait for 2 sec and then call qed->probe. If probe fails then retry max 10 times. Link: https://lore.kernel.org/r/20200907121443.5150-8-jhasan@marvell.comSigned-off-by:
Saurav Kashyap <skashyap@marvell.com> Signed-off-by:
Javed Hasan <jhasan@marvell.com> Signed-off-by:
-
Saurav Kashyap authored
On fan failure, disable the PCI function and initiate recovery for ramrod failure. Link: https://lore.kernel.org/r/20200907121443.5150-7-jhasan@marvell.comSigned-off-by:
-
Saurav Kashyap authored
If SUCCESS is not returned, error handling will escalate. Return SUCCESS similar to other conditions in this function. Link: https://lore.kernel.org/r/20200907121443.5150-6-jhasan@marvell.comSigned-off-by:
-
Javed Hasan authored
Correction in the FDMI attributes required for RHBA and RPA registration. Link: https://lore.kernel.org/r/20200907121443.5150-5-jhasan@marvell.comSigned-off-by:
-
Javed Hasan authored
Firmware expects E_D_TOV field in connection offload parameters as “msec”. Earlier incorrect value (100ms), was leading to abort from driver in the case when data frames for read take more than 100ms from target side, resulting in firmware reporting E_D_TOV expiration. Link: https://lore.kernel.org/r/20200907121443.5150-4-jhasan@marvell.comSigned-off-by:
-
Saurav Kashyap authored
Correct the misleading comment in qedf_initiate_els(). Link: https://lore.kernel.org/r/20200907121443.5150-3-jhasan@marvell.comSigned-off-by:
-
Javed Hasan authored
Change the debug parameter permission to read & write. Gives flexibility to change the debug verbosity dynamically. Link: https://lore.kernel.org/r/20200907121443.5150-2-jhasan@marvell.comSigned-off-by:
-
Stanley Chu authored
Add host reset mechanism to try to recover host-side errors during recovery flow. Link: https://lore.kernel.org/r/20200908064507.30774-5-stanley.chu@mediatek.comSigned-off-by:
Stanley Chu <stanley.chu@mediatek.com> Signed-off-by:
-
Stanley Chu authored
Forcibly leave UniPro low-power mode if UIC commands failed. This makes hba_enable_delay_us as correct (default) value for re-enabling the host. At the same time, change type of parameter "lpm" in function ufs_mtk_unipro_set_pm() to "bool". Link: https://lore.kernel.org/r/20200908064507.30774-4-stanley.chu@mediatek.comSigned-off-by:
-
Stanley Chu authored
Simply add HOST_PA_TACTIVATE quirk back since it was incorrectly removed before. Link: https://lore.kernel.org/r/20200908064507.30774-3-stanley.chu@mediatek.com Fixes: 47d05458 ("scsi: ufs-mediatek: fix HOST_PA_TACTIVATE quirk for Samsung UFS Devices") Signed-off-by:
-
Stanley Chu authored
Some MediaTek platforms does not have to bind MPHY so users shall not see any unnecessary logs. Simply remove logs for this case. Link: https://lore.kernel.org/r/20200908064507.30774-2-stanley.chu@mediatek.com Fixes: fc498301 ("scsi: ufs-mediatek: Allow unbound mphy") Signed-off-by:
-
Manish Rangankar authored
The error recovery is handled by management firmware (MFW) with the help of qed/qedi drivers. Upon detecting errors, driver informs MFW about this event which in turn starts a recovery process. MFW sends ERROR_RECOVERY notification to the driver which performs the required cleanup/recovery from the driver side. Link: https://lore.kernel.org/r/20200908095657.26821-9-mrangankar@marvell.comSigned-off-by:
Manish Rangankar <mrangankar@marvell.com> Signed-off-by:
-
Manish Rangankar authored
Add support to initiate MFW process recovery for all the devices if storage function receives the event first. Also added fix for kernel test robot warning, >> drivers/scsi/qedi/qedi_main.c:1119:6: warning: no previous prototype >> for 'qedi_schedule_hw_err_handler' [-Wmissing-prototypes] Link: https://lore.kernel.org/r/20200908095657.26821-8-mrangankar@marvell.comReported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
-
Nilesh Javali authored
For short time cable pulls, the in-flight I/O to the firmware is never cleaned up, resulting in the behaviour of stale I/O completion causing list_del corruption and soft lockup of the system. On link down event, mark all the connections for recovery, causing cleanup of all the in-flight I/O immediately. Link: https://lore.kernel.org/r/20200908095657.26821-7-mrangankar@marvell.comSigned-off-by:
Nilesh Javali <njavali@marvell.com> Signed-off-by:
-
Manish Rangankar authored
Use snprintf to limit max number of bytes to the buffer. Link: https://lore.kernel.org/r/20200908095657.26821-6-mrangankar@marvell.comSigned-off-by:
-
Nilesh Javali authored
Protect active command list for non-I/O commands like login response, logout response, text response, and recovery cleanup of active list to avoid list corruption. Link: https://lore.kernel.org/r/20200908095657.26821-5-mrangankar@marvell.comSigned-off-by:
-
Nilesh Javali authored
While aborting the I/O, the firmware cleanup task timed out and driver deleted the I/O from active command list. Some time later the firmware sent the cleanup task response and driver again deleted the I/O from active command list causing firmware to send completion for non-existent I/O and list_del corruption of active command list. Add fix to check if I/O is present before deleting it from the active command list to ensure firmware sends valid I/O completion and protect against list_del corruption. Link: https://lore.kernel.org/r/20200908095657.26821-4-mrangankar@marvell.comSigned-off-by:
-
Manish Rangankar authored
In boot from SAN scenario when qedi PCI shutdown handler is called with active iSCSI sessions, sometimes target takes too long time to respond to firmware connection termination request. Instead skip sending termination ramrod and progress with unload path. Link: https://lore.kernel.org/r/20200908095657.26821-3-mrangankar@marvell.comSigned-off-by:
-
Manish Rangankar authored
To avoid unnecessary vector allocation when the number of fast-path queues is less then available msix vectors, use return count from module qed->set_fp_int. Link: https://lore.kernel.org/r/20200908095657.26821-2-mrangankar@marvell.comSigned-off-by:
-
Nícolas F. R. A. Prado authored
Commit 91ebc1fa ("scsi: core: remove Scsi_Cmnd typedef") removed the Scsi_cmnd typedef but it was still mentioned in a paragraph in the "SCSI mid_level - lower_level driver interface" documentation page. Remove this obsolete paragraph. Link: https://lore.kernel.org/r/20200905210211.2286172-1-nfraprado@protonmail.comSuggested-by:
Randy Dunlap <rdunlap@infradead.org> Suggested-by:
Jonathan Corbet <corbet@lwn.net> Acked-by:
Nícolas F. R. A. Prado <nfraprado@protonmail.com> Signed-off-by:
-
Tyrel Datwyler authored
VIOS partitions with SLI-4 enabled Emulex adapters will be capable of driving I/O in parallel through mulitple work queues or channels, and with new hypervisor firmware that supports multiple interrupt sources an ibmvfc NPIV single initiator can be modified to exploit end-to-end channelization in a PowerVM environment. VIOS hosts will also be able to expose fabric perfromance impact notifications (FPIN) via a new asynchronous event to ibmvfc clients that advertise support via IBMVFC_CAN_HANDLE_FPIN in their capabilities flag during NPIV_LOGIN. This patch introduces three new Management Datagrams (MADs) for channelization support negotiation as well as the FPIN asynchronous event and FPIN status flags. Follow up work is required to plumb the ibmvfc client driver to use these new interfaces. Link: https://lore.kernel.org/r/20200904232936.840193-2-tyreld@linux.ibm.comSigned-off-by:
Tyrel Datwyler <tyreld@linux.ibm.com> Signed-off-by:
-
Tyrel Datwyler authored
Update ibmvfc.h structs to use the preferred __packed and __aligned() attribute macros defined in include/linux/compiler_attributes.h in place of __attribute__(). Link: https://lore.kernel.org/r/20200904232936.840193-1-tyreld@linux.ibm.comSigned-off-by:
-
Bao D. Nguyen authored
Setting the Auto-Hibernate Timer to zero is a valid setting which indicates the Auto-Hibernate feature being disabled. Correctly support this setting. In addition, when the timer value is queried from sysfs, read from the host controller's register and return that value instead of using the RAM value. Link: https://lore.kernel.org/r/b141cfcd7998b8933635828b56fbb64f8ad4d175.1598661071.git.nguyenb@codeaurora.orgAcked-by:
Bao D. Nguyen <nguyenb@codeaurora.org> Signed-off-by:
Asutosh Das <asutoshd@codeaurora.org> Signed-off-by:
Can Guo <cang@codeaurora.org> Signed-off-by:
-
- 03 Sep, 2020 16 commits
-
-
John Pittman authored
If virtual_gb is passed while using num_parts, when creating the partitions, virtual_gb is not respected. Set num_sectors using get_sdebug_capacity() to pull virtual_gb if set. Link: https://lore.kernel.org/r/20200902211434.9979-3-jpittman@redhat.comAcked-by:
Douglas Gilbert <dgilbert@interlog.com> Signed-off-by:
John Pittman <jpittman@redhat.com> Signed-off-by:
-
John Pittman authored
Currently when using the num_parts parameter, partitions are aligned and the end sector is one prior to the next start. This creates different sized partitions. Create instead equally sized partitions by trimming the end of each partition to the size of the smallest partition. This aligns better with what one would expect from automatically created partitions and can be helpful with testing things such as raid which often expect legs of the same size. Minimal space is lost as the initial partition starting size is calculated by dividing num_sectors by sdebug_num_parts. Link: https://lore.kernel.org/r/20200902211434.9979-2-jpittman@redhat.comAcked-by:
-
Nilesh Javali authored
Update Marvell owned driver maintainers and add Marvell Upstream email alias to the maintainers list. Link: https://lore.kernel.org/r/20200902073430.11787-1-njavali@marvell.comSigned-off-by:
-
Luo Jiaxing authored
Remove extra blank lines and add spaces around operators. Link: https://lore.kernel.org/r/1598958790-232272-9-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:
-
Xiang Chen authored
Newline is missing from some printk() statements. Add them. Link: https://lore.kernel.org/r/1598958790-232272-8-git-send-email-john.garry@huawei.comSigned-off-by:
Xiang Chen <chenxiang66@hisilicon.com> Signed-off-by:
-
Luo Jiaxing authored
Through the new debugfs interface the user can select fixed code patterns. Add two new interfaces fixed_code and fixed_code1. Link: https://lore.kernel.org/r/1598958790-232272-7-git-send-email-john.garry@huawei.comSigned-off-by:
-
Luo Jiaxing authored
Add BIST support for phy FFE (Feed forward equalizer) setting. The user can configure FFE through the new debugfs interface. FFE is a parameter used for link layer control. It will affect the link quality between the SAS controller and the backplane. In the BIST test, the FFE interface is provided to assist board testers in optimizing link parameters. The modification of the FFE parameter will affect the test after BIST or the normal running of the board. The user should save the initial FFE values and restore them after BIST test is complete. Link: https://lore.kernel.org/r/1598958790-232272-6-git-send-email-john.garry@huawei.comSigned-off-by:
-
Luo Jiaxing authored
We use "phy_id" to identify phy in the BIST code but the rest of code always uses "phy_no". Change it for consistency. Link: https://lore.kernel.org/r/1598958790-232272-5-git-send-email-john.garry@huawei.comSigned-off-by:
-
Luo Jiaxing authored
When updating PROG_PHY_LINK_RATE to set linkrate for a phy we used a hard-coded initial value instead of getting the current value from the register. The assumption was that this register would not be modified, but in fact it was partially modified in a new version of hardware. The hard-coded value we used changed the default value of the register to a an incorrect setting and as a result the SAS controller could not change linkrate for the phy. Delete hard-coded value and always read the latest value of register before updating it. Link: https://lore.kernel.org/r/1598958790-232272-4-git-send-email-john.garry@huawei.comSigned-off-by:
-
Luo Jiaxing authored
The macro for OOB phy linkrate is named CFG_PROG_PHY_LINK_RATE_* but that is inaccurate. For clarification, include OOB in macro name. Link: https://lore.kernel.org/r/1598958790-232272-3-git-send-email-john.garry@huawei.comSigned-off-by:
-
Xiang Chen authored
hisi_sas_slot_task_free() attempts to dereference SSP task for non-ATA tasks. If the task is SMP, the code may reference the wrong structure although this may not cause any problems. To avoid this, only access to SSP task when slot->n_elem_dif is not 0 which indicates this is an SSP task. Link: https://lore.kernel.org/r/1598958790-232272-2-git-send-email-john.garry@huawei.comSigned-off-by:
-
James Smart authored
Add 256GBit speed setting to the SCSI FC transport. This speed can be reached via FC trunking techniques. Link: https://lore.kernel.org/r/20200831213518.48409-1-james.smart@broadcom.comReviewed-by:
Himanshu Madhani <himanshu.madhani@oracle.com> Signed-off-by:
James Smart <james.smart@broadcom.com> Signed-off-by:
-
Jason Yan authored
This addresses the following gcc warning with "make W=1": drivers/scsi/fnic/fnic_fcs.c: In function ‘fnic_rq_cmpl_frame_recv’: drivers/scsi/fnic/fnic_fcs.c:840:15: warning: variable ‘eth_hdrs_stripped’ set but not used [-Wunused-but-set-variable] 840 | unsigned int eth_hdrs_stripped; | ^~~~~~~~~~~~~~~~~ Link: https://lore.kernel.org/r/20200831081126.3251288-5-yanaijie@huawei.comReported-by:Hulk Robot <hulkci@huawei.com> Signed-off-by:
Jason Yan <yanaijie@huawei.com> Signed-off-by:
-
Jason Yan authored
This addresses the following gcc warning with "make W=1": drivers/scsi/fnic/fnic_fcs.c: In function ‘fnic_fcoe_send_vlan_req’: drivers/scsi/fnic/fnic_fcs.c:379:6: warning: variable ‘fr_len’ set but not used [-Wunused-but-set-variable] 379 | int fr_len; | ^~~~~~ Link: https://lore.kernel.org/r/20200831081126.3251288-4-yanaijie@huawei.comReported-by: -
Jason Yan authored
This addresses the following gcc warning with "make W=1": drivers/scsi/fnic/fnic_fcs.c: In function ‘is_fnic_fip_flogi_reject’: drivers/scsi/fnic/fnic_fcs.c:317:9: warning: variable ‘els_len’ set but not used [-Wunused-but-set-variable] 317 | size_t els_len = 0; | ^~~~~~~ drivers/scsi/fnic/fnic_fcs.c:312:21: warning: variable ‘els_dtype’ set but not used [-Wunused-but-set-variable] 312 | enum fip_desc_type els_dtype = 0; | ^~~~~~~~~ Link: https://lore.kernel.org/r/20200831081126.3251288-3-yanaijie@huawei.comReported-by: -
Jason Yan authored
This addresses the following gcc warning with "make W=1": drivers/scsi/fnic/fnic_main.c: In function ‘fnic_set_vlan’: drivers/scsi/fnic/fnic_main.c:555:6: warning: variable ‘old_vlan’ set but not used [-Wunused-but-set-variable] 555 | u16 old_vlan; | ^~~~~~~~ Link: https://lore.kernel.org/r/20200831081126.3251288-2-yanaijie@huawei.comReported-by:
-
