- Jun 17, 2023
-
-
Stanley Chu authored
In MCQ mode, when a device command uses a hardware queue shared with other commands, a race condition may occur in the following scenario: 1. A device command is completed in CQx with CQE entry "e". 2. The interrupt handler copies the "cqe" pointer to "hba->dev_cmd.cqe" and completes "hba->dev_cmd.complete". 3. The "ufshcd_wait_for_dev_cmd()" function is awakened and retrieves the OCS value from "hba->dev_cmd.cqe". However, there is a possibility that the CQE entry "e" will be overwritten by newly completed commands in CQx, resulting in an incorrect OCS value being received by "ufshcd_wait_for_dev_cmd()". To avoid this race condition, the OCS value should be immediately copied to the struct "lrb" of the device command. Then "ufshcd_wait_for_dev_cmd()" can retrieve the OCS value from the struct "lrb". Fixes: 57b1c0ef ("scsi: ufs: core: mcq: Add support to allocate multiple queues") Suggested-by:Can Guo <quic_cang@quicinc.com> Signed-off-by:
Stanley Chu <stanley.chu@mediatek.com> Link: https://lore.kernel.org/r/20230610021553.1213-2-powen.kao@mediatek.com Tested-by:
Po-Wen Kao <powen.kao@mediatek.com> Reviewed-by:
Bart Van Assche <bvanassche@acm.org> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
Krzysztof Kozlowski authored
Clean up bindings dropping unneeded quotes. Once all these are fixed, checking for this can be enabled in yamllint. Signed-off-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://lore.kernel.org/r/20230609140651.64488-1-krzysztof.kozlowski@linaro.org Acked-by:
Alim Akhtar <alim.akhtar@samsung.com> Signed-off-by:
-
Damien Le Moal authored
The introduction of the macro IOPRIO_PRIO_LEVEL() in commit eca20409 ("scsi: block: ioprio: Clean up interface definition") results in an iopriority level to always be masked using the macro IOPRIO_LEVEL_MASK, and thus to the kernel always seeing an acceptable value for an I/O priority level when checked in ioprio_check_cap(). Before this patch, this function would return an error for some (but not all) invalid values for a level valid range of [0..7]. Restore and improve the detection of invalid priority levels by introducing the inline function ioprio_value() to check an ioprio class, level and hint value before combining these fields into a single value to be used with ioprio_set() or AIOs. If an invalid value for the class, level or hint of an ioprio is detected, ioprio_value() returns an ioprio using the class IOPRIO_CLASS_INVALID, indicating an invalid value and causing ioprio_check_cap() to return -EINVAL. Fixes: 6c913257 ("scsi: block: I...
-
- Jun 08, 2023
-
-
Christophe JAILLET authored
'new_fcports' is unused, so drop it. Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Link: https://lore.kernel.org/r/49bb77624c9edc8d9bf8fe71d0c8a4cd7e582175.1685854354.git.christophe.jaillet@wanadoo.fr Signed-off-by:
-
Gustavo A. R. Silva authored
One-element arrays as fake flex arrays are deprecated and we are moving towards adopting C99 flexible-array members, instead. So, replace one-element array declaration in struct ct_sns_gpnft_rsp, which is ultimately being used inside a union: drivers/scsi/qla2xxx/qla_def.h: 3240 struct ct_sns_gpnft_pkt { 3241 union { 3242 struct ct_sns_req req; 3243 struct ct_sns_gpnft_rsp rsp; 3244 } p; 3245 }; Refactor the rest of the code, accordingly. This issue was found with the help of Coccinelle. Link: https://github.com/KSPP/linux/issues/245 Link: https://github.com/KSPP/linux/issues/193 Reviewed-by:Kees Cook <keescook@chromium.org> Signed-off-by:
Gustavo A. R. Silva <gustavoars@kernel.org> Link: https://lore.kernel.org/r/ZH+/rZ1R1cBjIxjS@work Signed-off-by:
-
Uwe Kleine-König authored
The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is (mostly) ignored and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new() which already returns void. hisi_sas_remove() returned zero unconditionally so this was changed to return void. Then it has the right prototype to be used directly as remove callback for the two hisi_sas drivers. Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20230518202043.261739-1-u.kleine-koenig@pengutronix.de Signed-off-by:
-
Gustavo A. R. Silva authored
Prevent any potential integer wrapping issue, and avoid a -Wstringop-overflow warning by using the check_mul_overflow() helper. drivers/scsi/lpfc/lpfc.h: 837:#define LPFC_RAS_MIN_BUFF_POST_SIZE (256 * 1024) drivers/scsi/lpfc/lpfc_debugfs.c: 2266 size = LPFC_RAS_MIN_BUFF_POST_SIZE * phba->cfg_ras_fwlog_buffsize; this can wrap to negative if cfg_ras_fwlog_buffsize is large enough. And even when in practice this is not possible (due to phba->cfg_ras_fwlog_buffsize never being larger than 4[1]), the compiler is legitimately warning us about potentially buggy code. Fix the following warning seen under GCC-13: In function ‘lpfc_debugfs_ras_log_data’, inlined from ‘lpfc_debugfs_ras_log_open’ at drivers/scsi/lpfc/lpfc_debugfs.c:2271:15: drivers/scsi/lpfc/lpfc_debugfs.c:2210:25: warning: ‘memcpy’ specified bound between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=] 2210 | memcpy(buffer + copied, dmabuf->virt, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2211 | size - copied - 1); | ~~~~~~~~~~~~~~~~~~ Link: https://github.com/KSPP/linux/issues/305 Link: https://lore.kernel.org/linux-hardening/CABPRKS8zyzrbsWt4B5fp7kMowAZFiMLKg5kW26uELpg1cDKY3A@mail.gmail.com/ [1] Co-developed-by:Justin Tee <justin.tee@broadcom.com> Signed-off-by:
-
Justin Tee authored
Prefer struct_size() over open-coded versions of idiom: sizeof(struct-with-flex-array) + sizeof(typeof-flex-array-elements) * count where count is the max number of items the flexible array is supposed to contain. Link: https://github.com/KSPP/linux/issues/160 Co-developed-by:
-
Stanley Chu authored
Currently, ufshcd_mcq_poll_cqe_nolock() is only called by ufshcd_mcq_poll_cqe_lock() with the addition of a spinlock wrapper for ufshcd_mcq_poll_cqe_nolock(). Combining these two functions into one results in cleaner code. Reviewed-by:
Bao D. Nguyen <quic_nguyenb@quicinc.com> Signed-off-by:
Keoseong Park <keosung.park@samsung.com> Acked-by:
Bean Huo <beanhuo@micron.com> Reviewed-by:
-
zhanghui authored
When qdepth is not power of 2, not every bit of the mask is 1, so in sq_tail_slot some bits will be cleared unexpectedly. Signed-off-by:
zhanghui <zhanghui31@xiaomi.com> Link: https://lore.kernel.org/r/20230601124613.1446-1-zhanghui31@xiaomi.com Reviewed-by:
-
Bart Van Assche authored
ufshcd_add_command_trace() traces SCSI commands. Remove a ufshcd_add_command_trace() call from a code path that is not related to SCSI commands. Signed-off-by:
Avri Altman <avri.altman@wdc.com> Reviewed-by:
-
- Jun 01, 2023
-
-
Martin K. Petersen authored
Bart Van Assche <bvanassche@acm.org> says: Please consider these four UFS host controller driver patches for the next merge window. Link: https://lore.kernel.org/r/20230524203659.1394307-1-bvanassche@acm.org Signed-off-by:
-
Bart Van Assche authored
All UFS host drivers call ufshcd_shutdown(). Hence, instead of calling ufshcd_shutdown() from the host driver .shutdown() callback, inline that function into ufshcd_wl_shutdown(). Reviewed-by:
Adrian Hunter <adrian.hunter@intel.com> Signed-off-by:
-
Bart Van Assche authored
Move the definition of ufshcd_wl_shutdown() to make the next patch in this series easier to review. Reviewed-by:
-
Bart Van Assche authored
ufshcd_queuecommand() may be called two times in a row for a SCSI command before it is completed. Hence make the following changes: - In the functions that submit a command, do not check the old value of lrbp->cmd nor clear lrbp->cmd in error paths. - In ufshcd_release_scsi_cmd(), do not clear lrbp->cmd. See also scsi_send_eh_cmnd(). This commit prevents that the following appears if a command times out: WARNING: at drivers/ufs/core/ufshcd.c:2965 ufshcd_queuecommand+0x6f8/0x9a8 Call trace: ufshcd_queuecommand+0x6f8/0x9a8 scsi_send_eh_cmnd+0x2c0/0x960 scsi_eh_test_devices+0x100/0x314 scsi_eh_ready_devs+0xd90/0x114c scsi_error_handler+0x2b4/0xb70 kthread+0x16c/0x1e0 Fixes: 5a0b0cb9 ("[SCSI] ufs: Add support for sending NOP OUT UPIU") Signed-off-by:
-
Bart Van Assche authored
One UFS vendor asked to increase the UFS timeout from 1 s to 3 s. Another UFS vendor asked to increase the UFS timeout from 1 s to 10 s. Hence this patch that increases the UFS timeout to 10 s. This patch can cause the total timeout to exceed 20 s, the Android shutdown timeout. This is fine since the loop around ufshcd_execute_start_stop() exists to deal with unit attentions and because unit attentions are reported quickly. Fixes: dcd5b763 ("scsi: ufs: Reduce the START STOP UNIT timeout") Fixes: 8f2c9642 ("scsi: ufs: core: Reduce the power mode change timeout") Acked-by:
-
Sathya Prakash authored
Copy the sense data to internal driver buffer when the firmware completes any SCSI I/O command sent through admin queue with sense data for further use. Fixes: 506bc1a0 ("scsi: mpi3mr: Add support for MPT commands") Cc: <stable@vger.kernel.org> Signed-off-by:
Sathya Prakash <sathya.prakash@broadcom.com> Signed-off-by:
Sumit Saxena <sumit.saxena@broadcom.com> Link: https://lore.kernel.org/r/20230531184025.3803-1-sumit.saxena@broadcom.com Signed-off-by:
-
Martin K. Petersen authored
Bao D. Nguyen <quic_nguyenb@quicinc.com> says: This patch series enables support for ufshcd_abort() and error handler in MCQ mode. Link: https://lore.kernel.org/r/cover.1685396241.git.quic_nguyenb@quicinc.com Signed-off-by:
-
Bao D. Nguyen authored
Add support for error handling for MCQ mode. Suggested-by:
-
Bao D. Nguyen authored
In preparation for adding MCQ error handler support, update the MCQ code to use the ufshcd_mcq_poll_cqe_lock() in interrupt context instead of using ufshcd_mcq_poll_cqe_nolock(). This is to keep synchronization between MCQ interrupt and error handler contexts because both need to access the MCQ hardware in separate contexts. Signed-off-by:
-
Bao D. Nguyen authored
Add ufshcd_mcq_abort() to support UFS abort in MCQ mode. Signed-off-by:
-
Bao D. Nguyen authored
Update ufshcd_clear_cmd() to clean up the MCQ resources similar to the function ufshcd_utrl_clear() does for SDB mode. Update ufshcd_try_to_abort_task() to support MCQ mode so that this function can be invoked in either mcq or SDB mode. Signed-off-by:
-
Bao D. Nguyen authored
Add supporting functions to handle UFS abort in MCQ mode. Signed-off-by:
-
Bao D. Nguyen authored
In the ufshcd_clear_cmds(), the 2nd parameter would be the bit mask of the command to be cleared in the transfer request door bell register. This bit mask mechanism does not scale well in MCQ mode when the queue depth becomes much greater than 64. Change the 2nd parameter to the function to be the task_tag number of the corresponding bit to be cleared in the door bell register. By doing so, MCQ mode with a large queue depth can reuse this function. Since the behavior of this function is changed from handling multiple commands into a single command, rename ufshcd_clear_cmds() into ufshcd_clear_cmd(). Signed-off-by:
-
Bao D. Nguyen authored
The UTP command descriptor base address is a 57-bit field in the UTP transfer request descriptor. Combine the two 32-bit command_desc_base_addr_lo/hi fields into a 64-bit for better handling of this field. Signed-off-by:
-
Christophe JAILLET authored
Use vzalloc() instead of hand writing it with vmalloc()+memset(). This is less verbose. This also fixes some style issues :) Signed-off-by:
-
Changyuan Lyu authored
Add fatal error checking for the pm8001_phy_control() and pm8001_lu_reset() functions. Signed-off-by:
Changyuan Lyu <changyuanl@google.com> Signed-off-by:
Pranav Prasad <pranavpp@google.com> Link: https://lore.kernel.org/r/20230526235155.433243-1-pranavpp@google.com Signed-off-by:
-
Niklas Schnelle authored
In a future patch HAS_IOPORT=n will result in inb()/outb() and friends not being declared. We thus need to add HAS_IOPORT as dependency for those drivers using them. Co-developed-by:
Arnd Bergmann <arnd@kernel.org> Signed-off-by:
Niklas Schnelle <schnelle@linux.ibm.com> Link: https://lore.kernel.org/r/20230522105049.1467313-32-schnelle@linux.ibm.com Signed-off-by:
-
Niklas Schnelle authored
In a future patch HAS_IOPORT=n will result in inb()/outb() and friends not being declared. We thus need to add HAS_IOPORT as dependency for those drivers using them. Co-developed-by:
-
Justin Tee authored
The kernel test robot reported sparse warnings regarding the improper usage of beXX_to_cpu() macros. Change the flagged FDMI and VMID member variables to __beXX and redo the beXX_to_cpu() macros appropriately. Signed-off-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202305261159.lTW5NYrv-lkp@intel.com/ Closes: https://lore.kernel.org/oe-kbuild-all/202305260751.NWFvhLY5-lkp@intel.com/ Signed-off-by:
-
Martin K. Petersen authored
Justin Tee <justintee8345@gmail.com> says: Update lpfc to revision 14.2.0.13 This patch set contains discovery bug fixes, firmware logging improvements, clean up of CQ handling, and statistics collection enhancements. The patches were cut against Martin's 6.5/scsi-queue tree. Link: https://lore.kernel.org/r/20230523183206.7728-1-justintee8345@gmail.com Signed-off-by:
-
Justin Tee authored
Update copyrights to 2023 for files modified in the 14.2.0.13 patch set. Signed-off-by:
-
Justin Tee authored
Update lpfc version to 14.2.0.13 Signed-off-by:
-
Justin Tee authored
Various improvements are made for collecting congestion statistics: - Pre-existing logic is replaced with use of an hrtimer for increased reporting accuracy. - Congestion timestamp information is reorganized into a single struct. - Common statistic collection logic is refactored into a helper routine. Signed-off-by:
-
Justin Tee authored
There is mishandling of SLI-4 CQE status values larger than what is allowed by the LPFC_IOCB_STATUS_MASK of 4 bits. The LPFC_IOCB_STATUS_MASK is a leftover SLI-3 construct and serves no purpose in SLI-4 path. Remove the LPFC_IOCB_STATUS_MASK and clean up general CQE status handling in SLI-4 completion paths. Signed-off-by:
-
Justin Tee authored
A firmware upgrade does not necessitate dumping of phba->dbg_log[] to kmsg via LOG_TRACE_EVENT. A simple KERN_NOTICE log message should suffice to notify the user of successful or unsuccessful firmware upgrade. As such, firmware upgrade log messages are updated to use KERN_NOTICE instead of LOG_TRACE_EVENT. Additionally, in order to notify the user of reset type for instantiating newly downloaded firmware, lpfc_log_msg's default KERN_LEVEL is updated to 5 or KERN_NOTICE. Signed-off-by:
-
Justin Tee authored
When NPIV ports are zoned to devices that support both initiator and target mode, a remote device's initiated PRLI results in unintended final kref clean up of the device's ndlp structure. This disrupts NPIV ports' discovery for target devices that support both initiator and target mode. Modify the NPIV lpfc_drop_node clause such that we allow the ndlp to live so long as it was in NLP_STE_PLOGI_ISSUE, NLP_STE_REG_LOGIN_ISSUE, or NLP_STE_PRLI_ISSUE nlp_state. This allows lpfc's issued PRLI completion routine to determine if the final kref clean up should execute rather than a remote device's issued PRLI. Fixes: db651ec2 ("scsi: lpfc: Correct used_rpi count when devloss tmo fires with no recovery") Signed-off-by:
-
Justin Tee authored
Pre-existing device loss recovery logic via the NLP_IN_RECOV_POST_DEV_LOSS flag only handled Fabric Port Login, Fabric Controller, Management, and Name Server addresses. Fabric domain controllers fall under the same category for usage of the NLP_IN_RECOV_POST_DEV_LOSS flag. Add a default case statement to mark an ndlp for device loss recovery. Signed-off-by:
Martin Wilck <mwilck@suse.com> Signed-off-by:
-
Justin Tee authored
In dev_loss_tmo callback routine, we early return if the ndlp is in a state of rediscovery. This occurs when a target proactively PLOGIs or PRLIs after an RSCN before the dev_loss_tmo callback routine is scheduled to run. Move clear of the NLP_IN_DEV_LOSS flag before the ndlp state check in such cases. Signed-off-by:
-
Justin Tee authored
Due to a target port D_ID swap, it is possible for the lpfc_register_remote_port() routine to touch post mortem fc_rport memory when trying to access fc_rport->dd_data. The D_ID swap causes a simultaneous call to lpfc_unregister_remote_port(), where fc_remote_port_delete() reclaims fc_rport memory. Remove the fc_rport->dd_data->pnode NULL assignment because the following line reassigns ndlp->rport with an fc_rport object from fc_remote_port_add() anyways. The pnode nullification is superfluous. Signed-off-by:
-
