April 2018 - Linux-stable-mirror

Patch "scsi: libsas: Use dynamic alloced work to avoid sas event lost" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: libsas: Use dynamic alloced work to avoid sas event lost to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-libsas-use-dynamic-alloced-work-to-avoid-sas-event-lost.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Jason Yan <yanaijie(a)huawei.com> Date: Fri, 8 Dec 2017 17:42:04 +0800 Subject: scsi: libsas: Use dynamic alloced work to avoid sas event lost From: Jason Yan <yanaijie(a)huawei.com> [ Upstream commit 1c393b970e0f4070e4376d45f89a2d19a5c895d0 ] Now libsas hotplug work is static, every sas event type has its own static work, LLDD driver queues the hotplug work into shost->work_q. If LLDD driver burst posts lots hotplug events to libsas, the hotplug events may pending in the workqueue like shost->work_q new work[PORTE_BYTES_DMAED] --> |[PHYE_LOSS_OF_SIGNAL][PORTE_BYTES_DMAED] -> processing |<-------wait worker to process-------->| In this case, a new PORTE_BYTES_DMAED event coming, libsas try to queue it to shost->work_q, but this work is already pending, so it would be lost. Finally, libsas delete the related sas port and sas devices, but LLDD driver expect libsas add the sas port and devices(last sas event). This patch use dynamic allocated work to avoid this issue. Signed-off-by: Yijing Wang <wangyijing(a)huawei.com> CC: John Garry <john.garry(a)huawei.com> CC: Johannes Thumshirn <jthumshirn(a)suse.de> CC: Ewan Milne <emilne(a)redhat.com> CC: Christoph Hellwig <hch(a)lst.de> CC: Tomas Henzl <thenzl(a)redhat.com> CC: Dan Williams <dan.j.williams(a)intel.com> Reviewed-by: Hannes Reinecke <hare(a)suse.com> Signed-off-by: Jason Yan <yanaijie(a)huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/libsas/sas_event.c | 74 ++++++++++++++++++++++++++++--------- drivers/scsi/libsas/sas_init.c | 27 ++++++++++++- drivers/scsi/libsas/sas_internal.h | 6 +++ drivers/scsi/libsas/sas_phy.c | 44 ++++------------------ drivers/scsi/libsas/sas_port.c | 18 ++++----- include/scsi/libsas.h | 17 +++++--- 6 files changed, 115 insertions(+), 71 deletions(-) --- a/drivers/scsi/libsas/sas_event.c +++ b/drivers/scsi/libsas/sas_event.c @@ -29,7 +29,8 @@ int sas_queue_work(struct sas_ha_struct *ha, struct sas_work *sw) { - int rc = 0; + /* it's added to the defer_q when draining so return succeed */ + int rc = 1; if (!test_bit(SAS_HA_REGISTERED, &ha->state)) return 0; @@ -44,19 +45,15 @@ int sas_queue_work(struct sas_ha_struct return rc; } -static int sas_queue_event(int event, unsigned long *pending, - struct sas_work *work, +static int sas_queue_event(int event, struct sas_work *work, struct sas_ha_struct *ha) { - int rc = 0; + unsigned long flags; + int rc; - if (!test_and_set_bit(event, pending)) { - unsigned long flags; - - spin_lock_irqsave(&ha->lock, flags); - rc = sas_queue_work(ha, work); - spin_unlock_irqrestore(&ha->lock, flags); - } + spin_lock_irqsave(&ha->lock, flags); + rc = sas_queue_work(ha, work); + spin_unlock_irqrestore(&ha->lock, flags); return rc; } @@ -66,6 +63,7 @@ void __sas_drain_work(struct sas_ha_stru { struct workqueue_struct *wq = ha->core.shost->work_q; struct sas_work *sw, *_sw; + int ret; set_bit(SAS_HA_DRAINING, &ha->state); /* flush submitters */ @@ -78,7 +76,10 @@ void __sas_drain_work(struct sas_ha_stru clear_bit(SAS_HA_DRAINING, &ha->state); list_for_each_entry_safe(sw, _sw, &ha->defer_q, drain_node) { list_del_init(&sw->drain_node); - sas_queue_work(ha, sw); + ret = sas_queue_work(ha, sw); + if (ret != 1) + sas_free_event(to_asd_sas_event(&sw->work)); + } spin_unlock_irq(&ha->lock); } @@ -119,29 +120,68 @@ void sas_enable_revalidation(struct sas_ if (!test_and_clear_bit(ev, &d->pending)) continue; - sas_queue_event(ev, &d->pending, &d->disc_work[ev].work, ha); + sas_queue_event(ev, &d->disc_work[ev].work, ha); } mutex_unlock(&ha->disco_mutex); } + +static void sas_port_event_worker(struct work_struct *work) +{ + struct asd_sas_event *ev = to_asd_sas_event(work); + + sas_port_event_fns[ev->event](work); + sas_free_event(ev); +} + +static void sas_phy_event_worker(struct work_struct *work) +{ + struct asd_sas_event *ev = to_asd_sas_event(work); + + sas_phy_event_fns[ev->event](work); + sas_free_event(ev); +} + static int sas_notify_port_event(struct asd_sas_phy *phy, enum port_event event) { + struct asd_sas_event *ev; struct sas_ha_struct *ha = phy->ha; + int ret; BUG_ON(event >= PORT_NUM_EVENTS); - return sas_queue_event(event, &phy->port_events_pending, - &phy->port_events[event].work, ha); + ev = sas_alloc_event(phy); + if (!ev) + return -ENOMEM; + + INIT_SAS_EVENT(ev, sas_port_event_worker, phy, event); + + ret = sas_queue_event(event, &ev->work, ha); + if (ret != 1) + sas_free_event(ev); + + return ret; } int sas_notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) { + struct asd_sas_event *ev; struct sas_ha_struct *ha = phy->ha; + int ret; BUG_ON(event >= PHY_NUM_EVENTS); - return sas_queue_event(event, &phy->phy_events_pending, - &phy->phy_events[event].work, ha); + ev = sas_alloc_event(phy); + if (!ev) + return -ENOMEM; + + INIT_SAS_EVENT(ev, sas_phy_event_worker, phy, event); + + ret = sas_queue_event(event, &ev->work, ha); + if (ret != 1) + sas_free_event(ev); + + return ret; } int sas_init_events(struct sas_ha_struct *sas_ha) --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -39,6 +39,7 @@ #include "../scsi_sas_internal.h" static struct kmem_cache *sas_task_cache; +static struct kmem_cache *sas_event_cache; struct sas_task *sas_alloc_task(gfp_t flags) { @@ -364,8 +365,6 @@ void sas_prep_resume_ha(struct sas_ha_st struct asd_sas_phy *phy = ha->sas_phy[i]; memset(phy->attached_sas_addr, 0, SAS_ADDR_SIZE); - phy->port_events_pending = 0; - phy->phy_events_pending = 0; phy->frame_rcvd_size = 0; } } @@ -555,20 +554,42 @@ sas_domain_attach_transport(struct sas_d } EXPORT_SYMBOL_GPL(sas_domain_attach_transport); + +struct asd_sas_event *sas_alloc_event(struct asd_sas_phy *phy) +{ + gfp_t flags = in_interrupt() ? GFP_ATOMIC : GFP_KERNEL; + + return kmem_cache_zalloc(sas_event_cache, flags); +} + +void sas_free_event(struct asd_sas_event *event) +{ + kmem_cache_free(sas_event_cache, event); +} + /* ---------- SAS Class register/unregister ---------- */ static int __init sas_class_init(void) { sas_task_cache = KMEM_CACHE(sas_task, SLAB_HWCACHE_ALIGN); if (!sas_task_cache) - return -ENOMEM; + goto out; + + sas_event_cache = KMEM_CACHE(asd_sas_event, SLAB_HWCACHE_ALIGN); + if (!sas_event_cache) + goto free_task_kmem; return 0; +free_task_kmem: + kmem_cache_destroy(sas_task_cache); +out: + return -ENOMEM; } static void __exit sas_class_exit(void) { kmem_cache_destroy(sas_task_cache); + kmem_cache_destroy(sas_event_cache); } MODULE_AUTHOR("Luben Tuikov <luben_tuikov(a)adaptec.com>"); --- a/drivers/scsi/libsas/sas_internal.h +++ b/drivers/scsi/libsas/sas_internal.h @@ -61,6 +61,9 @@ int sas_show_oob_mode(enum sas_oob_mode int sas_register_phys(struct sas_ha_struct *sas_ha); void sas_unregister_phys(struct sas_ha_struct *sas_ha); +struct asd_sas_event *sas_alloc_event(struct asd_sas_phy *phy); +void sas_free_event(struct asd_sas_event *event); + int sas_register_ports(struct sas_ha_struct *sas_ha); void sas_unregister_ports(struct sas_ha_struct *sas_ha); @@ -99,6 +102,9 @@ void sas_hae_reset(struct work_struct *w void sas_free_device(struct kref *kref); +extern const work_func_t sas_phy_event_fns[PHY_NUM_EVENTS]; +extern const work_func_t sas_port_event_fns[PORT_NUM_EVENTS]; + #ifdef CONFIG_SCSI_SAS_HOST_SMP extern void sas_smp_host_handler(struct bsg_job *job, struct Scsi_Host *shost); #else --- a/drivers/scsi/libsas/sas_phy.c +++ b/drivers/scsi/libsas/sas_phy.c @@ -35,7 +35,6 @@ static void sas_phye_loss_of_signal(stru struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PHYE_LOSS_OF_SIGNAL, &phy->phy_events_pending); phy->error = 0; sas_deform_port(phy, 1); } @@ -45,7 +44,6 @@ static void sas_phye_oob_done(struct wor struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PHYE_OOB_DONE, &phy->phy_events_pending); phy->error = 0; } @@ -58,8 +56,6 @@ static void sas_phye_oob_error(struct wo struct sas_internal *i = to_sas_internal(sas_ha->core.shost->transportt); - clear_bit(PHYE_OOB_ERROR, &phy->phy_events_pending); - sas_deform_port(phy, 1); if (!port && phy->enabled && i->dft->lldd_control_phy) { @@ -88,8 +84,6 @@ static void sas_phye_spinup_hold(struct struct sas_internal *i = to_sas_internal(sas_ha->core.shost->transportt); - clear_bit(PHYE_SPINUP_HOLD, &phy->phy_events_pending); - phy->error = 0; i->dft->lldd_control_phy(phy, PHY_FUNC_RELEASE_SPINUP_HOLD, NULL); } @@ -99,8 +93,6 @@ static void sas_phye_resume_timeout(stru struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PHYE_RESUME_TIMEOUT, &phy->phy_events_pending); - /* phew, lldd got the phy back in the nick of time */ if (!phy->suspended) { dev_info(&phy->phy->dev, "resume timeout cancelled\n"); @@ -119,39 +111,12 @@ int sas_register_phys(struct sas_ha_stru { int i; - static const work_func_t sas_phy_event_fns[PHY_NUM_EVENTS] = { - [PHYE_LOSS_OF_SIGNAL] = sas_phye_loss_of_signal, - [PHYE_OOB_DONE] = sas_phye_oob_done, - [PHYE_OOB_ERROR] = sas_phye_oob_error, - [PHYE_SPINUP_HOLD] = sas_phye_spinup_hold, - [PHYE_RESUME_TIMEOUT] = sas_phye_resume_timeout, - - }; - - static const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = { - [PORTE_BYTES_DMAED] = sas_porte_bytes_dmaed, - [PORTE_BROADCAST_RCVD] = sas_porte_broadcast_rcvd, - [PORTE_LINK_RESET_ERR] = sas_porte_link_reset_err, - [PORTE_TIMER_EVENT] = sas_porte_timer_event, - [PORTE_HARD_RESET] = sas_porte_hard_reset, - }; - /* Now register the phys. */ for (i = 0; i < sas_ha->num_phys; i++) { - int k; struct asd_sas_phy *phy = sas_ha->sas_phy[i]; phy->error = 0; INIT_LIST_HEAD(&phy->port_phy_el); - for (k = 0; k < PORT_NUM_EVENTS; k++) { - INIT_SAS_WORK(&phy->port_events[k].work, sas_port_event_fns[k]); - phy->port_events[k].phy = phy; - } - - for (k = 0; k < PHY_NUM_EVENTS; k++) { - INIT_SAS_WORK(&phy->phy_events[k].work, sas_phy_event_fns[k]); - phy->phy_events[k].phy = phy; - } phy->port = NULL; phy->ha = sas_ha; @@ -179,3 +144,12 @@ int sas_register_phys(struct sas_ha_stru return 0; } + +const work_func_t sas_phy_event_fns[PHY_NUM_EVENTS] = { + [PHYE_LOSS_OF_SIGNAL] = sas_phye_loss_of_signal, + [PHYE_OOB_DONE] = sas_phye_oob_done, + [PHYE_OOB_ERROR] = sas_phye_oob_error, + [PHYE_SPINUP_HOLD] = sas_phye_spinup_hold, + [PHYE_RESUME_TIMEOUT] = sas_phye_resume_timeout, + +}; --- a/drivers/scsi/libsas/sas_port.c +++ b/drivers/scsi/libsas/sas_port.c @@ -261,8 +261,6 @@ void sas_porte_bytes_dmaed(struct work_s struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PORTE_BYTES_DMAED, &phy->port_events_pending); - sas_form_port(phy); } @@ -273,8 +271,6 @@ void sas_porte_broadcast_rcvd(struct wor unsigned long flags; u32 prim; - clear_bit(PORTE_BROADCAST_RCVD, &phy->port_events_pending); - spin_lock_irqsave(&phy->sas_prim_lock, flags); prim = phy->sas_prim; spin_unlock_irqrestore(&phy->sas_prim_lock, flags); @@ -288,8 +284,6 @@ void sas_porte_link_reset_err(struct wor struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PORTE_LINK_RESET_ERR, &phy->port_events_pending); - sas_deform_port(phy, 1); } @@ -298,8 +292,6 @@ void sas_porte_timer_event(struct work_s struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PORTE_TIMER_EVENT, &phy->port_events_pending); - sas_deform_port(phy, 1); } @@ -308,8 +300,6 @@ void sas_porte_hard_reset(struct work_st struct asd_sas_event *ev = to_asd_sas_event(work); struct asd_sas_phy *phy = ev->phy; - clear_bit(PORTE_HARD_RESET, &phy->port_events_pending); - sas_deform_port(phy, 1); } @@ -353,3 +343,11 @@ void sas_unregister_ports(struct sas_ha_ sas_deform_port(sas_ha->sas_phy[i], 0); } + +const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = { + [PORTE_BYTES_DMAED] = sas_porte_bytes_dmaed, + [PORTE_BROADCAST_RCVD] = sas_porte_broadcast_rcvd, + [PORTE_LINK_RESET_ERR] = sas_porte_link_reset_err, + [PORTE_TIMER_EVENT] = sas_porte_timer_event, + [PORTE_HARD_RESET] = sas_porte_hard_reset, +}; --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -292,6 +292,7 @@ struct asd_sas_port { struct asd_sas_event { struct sas_work work; struct asd_sas_phy *phy; + int event; }; static inline struct asd_sas_event *to_asd_sas_event(struct work_struct *work) @@ -301,17 +302,21 @@ static inline struct asd_sas_event *to_a return ev; } +static inline void INIT_SAS_EVENT(struct asd_sas_event *ev, + void (*fn)(struct work_struct *), + struct asd_sas_phy *phy, int event) +{ + INIT_SAS_WORK(&ev->work, fn); + ev->phy = phy; + ev->event = event; +} + + /* The phy pretty much is controlled by the LLDD. * The class only reads those fields. */ struct asd_sas_phy { /* private: */ - struct asd_sas_event port_events[PORT_NUM_EVENTS]; - struct asd_sas_event phy_events[PHY_NUM_EVENTS]; - - unsigned long port_events_pending; - unsigned long phy_events_pending; - int error; int suspended; Patches currently in stable-queue which might be from yanaijie(a)huawei.com are queue-4.15/scsi-libsas-initialize-sas_phy-status-according-to-response-of-discover.patch queue-4.15/scsi-libsas-fix-error-when-getting-phy-events.patch queue-4.15/scsi-libsas-fix-memory-leak-in-sas_smp_get_phy_events.patch queue-4.15/scsi-libsas-use-dynamic-alloced-work-to-avoid-sas-event-lost.patch

6 years, 9 months

1
0
0 0

Patch "scsi: megaraid_sas: Error handling for invalid ldcount provided by firmware in RAID map" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: megaraid_sas: Error handling for invalid ldcount provided by firmware in RAID map to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-megaraid_sas-error-handling-for-invalid-ldcount-provided-by-firmware-in-raid-map.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Shivasharan S <shivasharan.srikanteshwara(a)broadcom.com> Date: Fri, 5 Jan 2018 05:27:40 -0800 Subject: scsi: megaraid_sas: Error handling for invalid ldcount provided by firmware in RAID map From: Shivasharan S <shivasharan.srikanteshwara(a)broadcom.com> [ Upstream commit 7ada701d0d5e5c6d357e157a72b841db3e8d03f4 ] Currently driver does not validate ldcount provided by firmware. If the value is invalid, fail RAID map validation accordingly. This issue is rare to hit in field and is fixed as part of code review. Signed-off-by: Sumit Saxena <sumit.saxena(a)broadcom.com> Signed-off-by: Shivasharan S <shivasharan.srikanteshwara(a)broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/megaraid/megaraid_sas_fp.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) --- a/drivers/scsi/megaraid/megaraid_sas_fp.c +++ b/drivers/scsi/megaraid/megaraid_sas_fp.c @@ -168,7 +168,7 @@ static struct MR_LD_SPAN *MR_LdSpanPtrGe /* * This function will Populate Driver Map using firmware raid map */ -void MR_PopulateDrvRaidMap(struct megasas_instance *instance) +static int MR_PopulateDrvRaidMap(struct megasas_instance *instance) { struct fusion_context *fusion = instance->ctrl_context; struct MR_FW_RAID_MAP_ALL *fw_map_old = NULL; @@ -259,7 +259,7 @@ void MR_PopulateDrvRaidMap(struct megasa ld_count = (u16)le16_to_cpu(fw_map_ext->ldCount); if (ld_count > MAX_LOGICAL_DRIVES_EXT) { dev_dbg(&instance->pdev->dev, "megaraid_sas: LD count exposed in RAID map in not valid\n"); - return; + return 1; } pDrvRaidMap->ldCount = (__le16)cpu_to_le16(ld_count); @@ -285,6 +285,12 @@ void MR_PopulateDrvRaidMap(struct megasa fusion->ld_map[(instance->map_id & 1)]; pFwRaidMap = &fw_map_old->raidMap; ld_count = (u16)le32_to_cpu(pFwRaidMap->ldCount); + if (ld_count > MAX_LOGICAL_DRIVES) { + dev_dbg(&instance->pdev->dev, + "LD count exposed in RAID map in not valid\n"); + return 1; + } + pDrvRaidMap->totalSize = pFwRaidMap->totalSize; pDrvRaidMap->ldCount = (__le16)cpu_to_le16(ld_count); pDrvRaidMap->fpPdIoTimeoutSec = pFwRaidMap->fpPdIoTimeoutSec; @@ -300,6 +306,8 @@ void MR_PopulateDrvRaidMap(struct megasa sizeof(struct MR_DEV_HANDLE_INFO) * MAX_RAIDMAP_PHYSICAL_DEVICES); } + + return 0; } /* @@ -317,8 +325,8 @@ u8 MR_ValidateMapInfo(struct megasas_ins u16 ld; u32 expected_size; - - MR_PopulateDrvRaidMap(instance); + if (MR_PopulateDrvRaidMap(instance)) + return 0; fusion = instance->ctrl_context; drv_map = fusion->ld_drv_map[(instance->map_id & 1)]; Patches currently in stable-queue which might be from shivasharan.srikanteshwara(a)broadcom.com are queue-4.15/scsi-megaraid_sas-error-handling-for-invalid-ldcount-provided-by-firmware-in-raid-map.patch queue-4.15/scsi-megaraid_sas-unload-flag-should-be-set-after-scsi_remove_host-is-called.patch

6 years, 9 months

1
0
0 0

Patch "scsi: libsas: initialize sas_phy status according to response of DISCOVER" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: libsas: initialize sas_phy status according to response of DISCOVER to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-libsas-initialize-sas_phy-status-according-to-response-of-discover.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: chenxiang <chenxiang66(a)hisilicon.com> Date: Thu, 4 Jan 2018 21:04:33 +0800 Subject: scsi: libsas: initialize sas_phy status according to response of DISCOVER From: chenxiang <chenxiang66(a)hisilicon.com> [ Upstream commit affc67788fe5dfffad5cda3d461db5cf2b2ff2b0 ] The status of SAS PHY is in sas_phy->enabled. There is an issue that the status of a remote SAS PHY may be initialized incorrectly: if disable remote SAS PHY through sysfs interface (such as echo 0 > /sys/class/sas_phy/phy-1:0:0/enable), then reboot the system, and we will find the status of remote SAS PHY which is disabled before is 1 (cat /sys/class/sas_phy/phy-1:0:0/enable). But actually the status of remote SAS PHY is disabled and the device attached is not found. In SAS protocol, NEGOTIATED LOGICAL LINK RATE field of DISCOVER response is 0x1 when remote SAS PHY is disabled. So initialize sas_phy->enabled according to the value of NEGOTIATED LOGICAL LINK RATE field. Signed-off-by: chenxiang <chenxiang66(a)hisilicon.com> Reviewed-by: John Garry <john.garry(a)huawei.com> Signed-off-by: Jason Yan <yanaijie(a)huawei.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Hannes Reinecke <hare(a)suse.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/libsas/sas_expander.c | 1 + 1 file changed, 1 insertion(+) --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -293,6 +293,7 @@ static void sas_set_ex_phy(struct domain phy->phy->minimum_linkrate = dr->pmin_linkrate; phy->phy->maximum_linkrate = dr->pmax_linkrate; phy->phy->negotiated_linkrate = phy->linkrate; + phy->phy->enabled = (phy->linkrate != SAS_PHY_DISABLED); skip: if (new_phy) Patches currently in stable-queue which might be from chenxiang66(a)hisilicon.com are queue-4.15/scsi-libsas-initialize-sas_phy-status-according-to-response-of-discover.patch queue-4.15/scsi-libsas-fix-error-when-getting-phy-events.patch queue-4.15/scsi-libsas-fix-memory-leak-in-sas_smp_get_phy_events.patch

6 years, 9 months

1
0
0 0

Patch "scsi: libsas: fix error when getting phy events" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: libsas: fix error when getting phy events to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-libsas-fix-error-when-getting-phy-events.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Jason Yan <yanaijie(a)huawei.com> Date: Thu, 4 Jan 2018 21:04:32 +0800 Subject: scsi: libsas: fix error when getting phy events From: Jason Yan <yanaijie(a)huawei.com> [ Upstream commit 2b23d9509fd7174b362482cf5f3b5f9a2265bc33 ] The intend purpose here was to goto out if smp_execute_task() returned error. Obviously something got screwed up. We will never get these link error statistics below: ~:/sys/class/sas_phy/phy-1:0:12 # cat invalid_dword_count 0 ~:/sys/class/sas_phy/phy-1:0:12 # cat running_disparity_error_count 0 ~:/sys/class/sas_phy/phy-1:0:12 # cat loss_of_dword_sync_count 0 ~:/sys/class/sas_phy/phy-1:0:12 # cat phy_reset_problem_count 0 Obviously we should goto error handler if smp_execute_task() returns non-zero. Fixes: 2908d778ab3e ("[SCSI] aic94xx: new driver") Signed-off-by: Jason Yan <yanaijie(a)huawei.com> CC: John Garry <john.garry(a)huawei.com> CC: chenqilin <chenqilin2(a)huawei.com> CC: chenxiang <chenxiang66(a)hisilicon.com> Reviewed-by: Hannes Reinecke <hare(a)suse.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/libsas/sas_expander.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -686,7 +686,7 @@ int sas_smp_get_phy_events(struct sas_ph res = smp_execute_task(dev, req, RPEL_REQ_SIZE, resp, RPEL_RESP_SIZE); - if (!res) + if (res) goto out; phy->invalid_dword_count = scsi_to_u32(&resp[12]); Patches currently in stable-queue which might be from yanaijie(a)huawei.com are queue-4.15/scsi-libsas-initialize-sas_phy-status-according-to-response-of-discover.patch queue-4.15/scsi-libsas-fix-error-when-getting-phy-events.patch queue-4.15/scsi-libsas-fix-memory-leak-in-sas_smp_get_phy_events.patch queue-4.15/scsi-libsas-use-dynamic-alloced-work-to-avoid-sas-event-lost.patch

6 years, 9 months

1
0
0 0

Patch "scsi: libsas: fix memory leak in sas_smp_get_phy_events()" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: libsas: fix memory leak in sas_smp_get_phy_events() to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-libsas-fix-memory-leak-in-sas_smp_get_phy_events.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Jason Yan <yanaijie(a)huawei.com> Date: Thu, 4 Jan 2018 21:04:31 +0800 Subject: scsi: libsas: fix memory leak in sas_smp_get_phy_events() From: Jason Yan <yanaijie(a)huawei.com> [ Upstream commit 4a491b1ab11ca0556d2fda1ff1301e862a2d44c4 ] We've got a memory leak with the following producer: while true; do cat /sys/class/sas_phy/phy-1:0:12/invalid_dword_count >/dev/null; done The buffer req is allocated and not freed after we return. Fix it. Fixes: 2908d778ab3e ("[SCSI] aic94xx: new driver") Signed-off-by: Jason Yan <yanaijie(a)huawei.com> CC: John Garry <john.garry(a)huawei.com> CC: chenqilin <chenqilin2(a)huawei.com> CC: chenxiang <chenxiang66(a)hisilicon.com> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Hannes Reinecke <hare(a)suse.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/libsas/sas_expander.c | 1 + 1 file changed, 1 insertion(+) --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -695,6 +695,7 @@ int sas_smp_get_phy_events(struct sas_ph phy->phy_reset_problem_count = scsi_to_u32(&resp[24]); out: + kfree(req); kfree(resp); return res; Patches currently in stable-queue which might be from yanaijie(a)huawei.com are queue-4.15/scsi-libsas-initialize-sas_phy-status-according-to-response-of-discover.patch queue-4.15/scsi-libsas-fix-error-when-getting-phy-events.patch queue-4.15/scsi-libsas-fix-memory-leak-in-sas_smp_get_phy_events.patch queue-4.15/scsi-libsas-use-dynamic-alloced-work-to-avoid-sas-event-lost.patch

6 years, 9 months

1
0
0 0

Patch "scsi: libiscsi: Allow sd_shutdown on bad transport" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled scsi: libiscsi: Allow sd_shutdown on bad transport to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: scsi-libiscsi-allow-sd_shutdown-on-bad-transport.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Rafael David Tinoco <rafael.tinoco(a)canonical.com> Date: Thu, 7 Dec 2017 19:59:13 -0200 Subject: scsi: libiscsi: Allow sd_shutdown on bad transport From: Rafael David Tinoco <rafael.tinoco(a)canonical.com> [ Upstream commit d754941225a7dbc61f6dd2173fa9498049f9a7ee ] If, for any reason, userland shuts down iscsi transport interfaces before proper logouts - like when logging in to LUNs manually, without logging out on server shutdown, or when automated scripts can't umount/logout from logged LUNs - kernel will hang forever on its sd_sync_cache() logic, after issuing the SYNCHRONIZE_CACHE cmd to all still existent paths. PID: 1 TASK: ffff8801a69b8000 CPU: 1 COMMAND: "systemd-shutdow" #0 [ffff8801a69c3a30] __schedule at ffffffff8183e9ee #1 [ffff8801a69c3a80] schedule at ffffffff8183f0d5 #2 [ffff8801a69c3a98] schedule_timeout at ffffffff81842199 #3 [ffff8801a69c3b40] io_schedule_timeout at ffffffff8183e604 #4 [ffff8801a69c3b70] wait_for_completion_io_timeout at ffffffff8183fc6c #5 [ffff8801a69c3bd0] blk_execute_rq at ffffffff813cfe10 #6 [ffff8801a69c3c88] scsi_execute at ffffffff815c3fc7 #7 [ffff8801a69c3cc8] scsi_execute_req_flags at ffffffff815c60fe #8 [ffff8801a69c3d30] sd_sync_cache at ffffffff815d37d7 #9 [ffff8801a69c3da8] sd_shutdown at ffffffff815d3c3c This happens because iscsi_eh_cmd_timed_out(), the transport layer timeout helper, would tell the queue timeout function (scsi_times_out) to reset the request timer over and over, until the session state is back to logged in state. Unfortunately, during server shutdown, this might never happen again. Other option would be "not to handle" the issue in the transport layer. That would trigger the error handler logic, which would also need the session state to be logged in again. Best option, for such case, is to tell upper layers that the command was handled during the transport layer error handler helper, marking it as DID_NO_CONNECT, which will allow completion and inform about the problem. After the session was marked as ISCSI_STATE_FAILED, due to the first timeout during the server shutdown phase, all subsequent cmds will fail to be queued, allowing upper logic to fail faster. Signed-off-by: Rafael David Tinoco <rafael.tinoco(a)canonical.com> Reviewed-by: Lee Duncan <lduncan(a)suse.com> Signed-off-by: Martin K. Petersen <martin.petersen(a)oracle.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/scsi/libiscsi.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -1696,6 +1696,15 @@ int iscsi_queuecommand(struct Scsi_Host */ switch (session->state) { case ISCSI_STATE_FAILED: + /* + * cmds should fail during shutdown, if the session + * state is bad, allowing completion to happen + */ + if (unlikely(system_state != SYSTEM_RUNNING)) { + reason = FAILURE_SESSION_FAILED; + sc->result = DID_NO_CONNECT << 16; + break; + } case ISCSI_STATE_IN_RECOVERY: reason = FAILURE_SESSION_IN_RECOVERY; sc->result = DID_IMM_RETRY << 16; @@ -1979,6 +1988,19 @@ enum blk_eh_timer_return iscsi_eh_cmd_ti if (session->state != ISCSI_STATE_LOGGED_IN) { /* + * During shutdown, if session is prematurely disconnected, + * recovery won't happen and there will be hung cmds. Not + * handling cmds would trigger EH, also bad in this case. + * Instead, handle cmd, allow completion to happen and let + * upper layer to deal with the result. + */ + if (unlikely(system_state != SYSTEM_RUNNING)) { + sc->result = DID_NO_CONNECT << 16; + ISCSI_DBG_EH(session, "sc on shutdown, handled\n"); + rc = BLK_EH_HANDLED; + goto done; + } + /* * We are probably in the middle of iscsi recovery so let * that complete and handle the error. */ @@ -2082,7 +2104,7 @@ done: task->last_timeout = jiffies; spin_unlock(&session->frwd_lock); ISCSI_DBG_EH(session, "return %s\n", rc == BLK_EH_RESET_TIMER ? - "timer reset" : "nh"); + "timer reset" : "shutdown or nh"); return rc; } EXPORT_SYMBOL_GPL(iscsi_eh_cmd_timed_out); Patches currently in stable-queue which might be from rafael.tinoco(a)canonical.com are queue-4.15/scsi-libiscsi-allow-sd_shutdown-on-bad-transport.patch

6 years, 9 months

1
0
0 0

Patch "rds; Reset rs->rs_bound_addr in rds_add_bound() failure path" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled rds; Reset rs->rs_bound_addr in rds_add_bound() failure path to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rds-reset-rs-rs_bound_addr-in-rds_add_bound-failure-path.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Sowmini Varadhan <sowmini.varadhan(a)oracle.com> Date: Fri, 22 Dec 2017 09:38:59 -0800 Subject: rds; Reset rs->rs_bound_addr in rds_add_bound() failure path From: Sowmini Varadhan <sowmini.varadhan(a)oracle.com> [ Upstream commit 7ae0c649c47f1c5d2db8cee6dd75855970af1669 ] If the rds_sock is not added to the bind_hash_table, we must reset rs_bound_addr so that rds_remove_bound will not trip on this rds_sock. rds_add_bound() does a rds_sock_put() in this failure path, so failing to reset rs_bound_addr will result in a socket refcount bug, and will trigger a WARN_ON with the stack shown below when the application subsequently tries to close the PF_RDS socket. WARNING: CPU: 20 PID: 19499 at net/rds/af_rds.c:496 \ rds_sock_destruct+0x15/0x30 [rds] : __sk_destruct+0x21/0x190 rds_remove_bound.part.13+0xb6/0x140 [rds] rds_release+0x71/0x120 [rds] sock_release+0x1a/0x70 sock_close+0xe/0x20 __fput+0xd5/0x210 task_work_run+0x82/0xa0 do_exit+0x2ce/0xb30 ? syscall_trace_enter+0x1cc/0x2b0 do_group_exit+0x39/0xa0 SyS_exit_group+0x10/0x10 do_syscall_64+0x61/0x1a0 Signed-off-by: Sowmini Varadhan <sowmini.varadhan(a)oracle.com> Acked-by: Santosh Shilimkar <santosh.shilimkar(a)oracle.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- net/rds/bind.c | 1 + 1 file changed, 1 insertion(+) --- a/net/rds/bind.c +++ b/net/rds/bind.c @@ -114,6 +114,7 @@ static int rds_add_bound(struct rds_sock rs, &addr, (int)ntohs(*port)); break; } else { + rs->rs_bound_addr = 0; rds_sock_put(rs); ret = -ENOMEM; break; Patches currently in stable-queue which might be from sowmini.varadhan(a)oracle.com are queue-4.15/selftests-net-fix-bugs-in-address-and-port-initialization.patch queue-4.15/rds-reset-rs-rs_bound_addr-in-rds_add_bound-failure-path.patch

6 years, 9 months

1
0
0 0

Patch "rt2x00: do not pause queue unconditionally on error path" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled rt2x00: do not pause queue unconditionally on error path to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rt2x00-do-not-pause-queue-unconditionally-on-error-path.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: Stanislaw Gruszka <sgruszka(a)redhat.com> Date: Tue, 19 Dec 2017 12:33:56 +0100 Subject: rt2x00: do not pause queue unconditionally on error path From: Stanislaw Gruszka <sgruszka(a)redhat.com> [ Upstream commit 6dd80efd75ce7c2dbd9f117cf585ee2b33a42ee1 ] Pausing queue without checking threshold is racy with txdone path. Moreover we do not need pause queue on any error, but only if queue is full - in case when we send RTS frame ( other cases of almost full queue are already handled in rt2x00queue_write_tx_frame() ). Patch fixes of theoretically possible problem of pausing empty queue. Signed-off-by: Stanislaw Gruszka <sgruszka(a)redhat.com> Tested-by: Enrico Mioso <mrkiko.rs(a)gmail.com> Signed-off-by: Kalle Valo <kvalo(a)codeaurora.org> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/net/wireless/ralink/rt2x00/rt2x00mac.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) --- a/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00mac.c @@ -142,15 +142,25 @@ void rt2x00mac_tx(struct ieee80211_hw *h if (!rt2x00dev->ops->hw->set_rts_threshold && (tx_info->control.rates[0].flags & (IEEE80211_TX_RC_USE_RTS_CTS | IEEE80211_TX_RC_USE_CTS_PROTECT))) { - if (rt2x00queue_available(queue) <= 1) - goto exit_fail; + if (rt2x00queue_available(queue) <= 1) { + /* + * Recheck for full queue under lock to avoid race + * conditions with rt2x00lib_txdone(). + */ + spin_lock(&queue->tx_lock); + if (rt2x00queue_threshold(queue)) + rt2x00queue_pause_queue(queue); + spin_unlock(&queue->tx_lock); + + goto exit_free_skb; + } if (rt2x00mac_tx_rts_cts(rt2x00dev, queue, skb)) - goto exit_fail; + goto exit_free_skb; } if (unlikely(rt2x00queue_write_tx_frame(queue, skb, control->sta, false))) - goto exit_fail; + goto exit_free_skb; /* * Pausing queue has to be serialized with rt2x00lib_txdone(). Note @@ -164,10 +174,6 @@ void rt2x00mac_tx(struct ieee80211_hw *h return; - exit_fail: - spin_lock(&queue->tx_lock); - rt2x00queue_pause_queue(queue); - spin_unlock(&queue->tx_lock); exit_free_skb: ieee80211_free_txskb(hw, skb); } Patches currently in stable-queue which might be from sgruszka(a)redhat.com are queue-4.15/rt2x00-do-not-pause-queue-unconditionally-on-error-path.patch

6 years, 9 months

1
0
0 0

Patch "RDMA/hns: Update the usage of sr_max and rr_max field" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled RDMA/hns: Update the usage of sr_max and rr_max field to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: rdma-hns-update-the-usage-of-sr_max-and-rr_max-field.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: oulijun <oulijun(a)huawei.com> Date: Wed, 3 Jan 2018 10:44:04 +0800 Subject: RDMA/hns: Update the usage of sr_max and rr_max field From: oulijun <oulijun(a)huawei.com> [ Upstream commit 4f3f7a704b3bff9e4eb322ab3c989b505f7562eb ] This patch fixes the usage with sr_max filed and rr_max of qp context when modify qp. Its modifications include: 1. Adjust location of filling sr_max filed of qpc 2. Only assign the number of responder resource if IB_QP_MAX_DEST_RD_ATOMIC bit is set 3. Only assign the number of outstanding resource if IB_QP_MAX_QP_RD_ATOMIC 4. Fix the assgin algorithms for the field of sr_max and rr_max of qp context Signed-off-by: Lijun Ou <oulijun(a)huawei.com> Signed-off-by: Yixian Liu <liuyixian(a)huawei.com> Signed-off-by: Wei Hu (Xavier) <xavier.huwei(a)huawei.com> Reviewed-by: Leon Romanovsky <leonro(a)mellanox.com> Signed-off-by: Jason Gunthorpe <jgg(a)mellanox.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-) --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -2463,11 +2463,14 @@ static int modify_qp_init_to_rtr(struct roce_set_bit(qpc_mask->byte_28_at_fl, V2_QPC_BYTE_28_LBI_S, 0); } - roce_set_field(context->byte_140_raq, V2_QPC_BYTE_140_RR_MAX_M, - V2_QPC_BYTE_140_RR_MAX_S, - ilog2((unsigned int)attr->max_dest_rd_atomic)); - roce_set_field(qpc_mask->byte_140_raq, V2_QPC_BYTE_140_RR_MAX_M, - V2_QPC_BYTE_140_RR_MAX_S, 0); + if ((attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) && + attr->max_dest_rd_atomic) { + roce_set_field(context->byte_140_raq, V2_QPC_BYTE_140_RR_MAX_M, + V2_QPC_BYTE_140_RR_MAX_S, + fls(attr->max_dest_rd_atomic - 1)); + roce_set_field(qpc_mask->byte_140_raq, V2_QPC_BYTE_140_RR_MAX_M, + V2_QPC_BYTE_140_RR_MAX_S, 0); + } roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, V2_QPC_BYTE_56_DQPN_S, attr->dest_qp_num); @@ -2557,12 +2560,6 @@ static int modify_qp_init_to_rtr(struct V2_QPC_BYTE_168_LP_SGEN_INI_M, V2_QPC_BYTE_168_LP_SGEN_INI_S, 0); - roce_set_field(context->byte_208_irrl, V2_QPC_BYTE_208_SR_MAX_M, - V2_QPC_BYTE_208_SR_MAX_S, - ilog2((unsigned int)attr->max_rd_atomic)); - roce_set_field(qpc_mask->byte_208_irrl, V2_QPC_BYTE_208_SR_MAX_M, - V2_QPC_BYTE_208_SR_MAX_S, 0); - roce_set_field(context->byte_28_at_fl, V2_QPC_BYTE_28_SL_M, V2_QPC_BYTE_28_SL_S, rdma_ah_get_sl(&attr->ah_attr)); roce_set_field(qpc_mask->byte_28_at_fl, V2_QPC_BYTE_28_SL_M, @@ -2766,6 +2763,14 @@ static int modify_qp_rtr_to_rts(struct i roce_set_field(qpc_mask->byte_196_sq_psn, V2_QPC_BYTE_196_SQ_MAX_PSN_M, V2_QPC_BYTE_196_SQ_MAX_PSN_S, 0); + if ((attr_mask & IB_QP_MAX_QP_RD_ATOMIC) && attr->max_rd_atomic) { + roce_set_field(context->byte_208_irrl, V2_QPC_BYTE_208_SR_MAX_M, + V2_QPC_BYTE_208_SR_MAX_S, + fls(attr->max_rd_atomic - 1)); + roce_set_field(qpc_mask->byte_208_irrl, + V2_QPC_BYTE_208_SR_MAX_M, + V2_QPC_BYTE_208_SR_MAX_S, 0); + } return 0; } Patches currently in stable-queue which might be from oulijun(a)huawei.com are queue-4.15/rdma-hns-update-the-usage-of-sr_max-and-rr_max-field.patch

6 years, 9 months

1
0
0 0

Patch "powernv-cpufreq: Add helper to extract pstate from PMSR" has been added to the 4.15-stable tree

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled powernv-cpufreq: Add helper to extract pstate from PMSR to the 4.15-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=sum… The filename of the patch is: powernv-cpufreq-add-helper-to-extract-pstate-from-pmsr.patch and it can be found in the queue-4.15 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable(a)vger.kernel.org> know about it. >From foo@baz Mon Apr 9 10:16:32 CEST 2018 From: "Gautham R. Shenoy" <ego(a)linux.vnet.ibm.com> Date: Wed, 13 Dec 2017 12:27:39 +0530 Subject: powernv-cpufreq: Add helper to extract pstate from PMSR From: "Gautham R. Shenoy" <ego(a)linux.vnet.ibm.com> [ Upstream commit ee1f4a7dafa997816ff3de96155c6f3edc21c1e6 ] On POWERNV platform, the fields for pstates in the Power Management Status Register (PMSR) and the Power Management Control Register (PMCR) are 8-bits wide. On POWER8 the pstates are negatively numbered while on POWER9 they are positively numbered. The device-tree exports pstates as 32-bit entries. The device-tree implementation sign-extends the 8-bit pstate values to obtain the corresponding 32-bit entry. Eg: On POWER8, a pstate value 0x82 [-126] is represented in the device-tree as 0xfffffff82 while on POWER9, the same value 0x82 [130] is represented in the device-tree as 0x00000082. The powernv-cpufreq driver implementation represents pstates using the integer type. In multiple places in the driver, the code interprets the pstates extracted from the PMSR as a signed byte and assigns it to a integer variable to get the sign-extention. On POWER9 platforms which have greater than 128 pstates, this results in the driver performing incorrect sign-extention, and thereby treating a legitimate pstate (say 130) as an invalid pstates (since it is interpreted as -126). This patch fixes the issue by implementing a helper function to extract Pstates from PMSR register, and correctly sign-extend it to be consistent with the values provided by the device-tree. Signed-off-by: Gautham R. Shenoy <ego(a)linux.vnet.ibm.com> Acked-by: Balbir Singh <bsingharora(a)gmail.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com> Signed-off-by: Sasha Levin <alexander.levin(a)microsoft.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> --- drivers/cpufreq/powernv-cpufreq.c | 37 +++++++++++++++++++++++-------------- 1 file changed, 23 insertions(+), 14 deletions(-) --- a/drivers/cpufreq/powernv-cpufreq.c +++ b/drivers/cpufreq/powernv-cpufreq.c @@ -41,11 +41,9 @@ #define POWERNV_MAX_PSTATES 256 #define PMSR_PSAFE_ENABLE (1UL << 30) #define PMSR_SPR_EM_DISABLE (1UL << 31) -#define PMSR_MAX(x) ((x >> 32) & 0xFF) +#define MAX_PSTATE_SHIFT 32 #define LPSTATE_SHIFT 48 #define GPSTATE_SHIFT 56 -#define GET_LPSTATE(x) (((x) >> LPSTATE_SHIFT) & 0xFF) -#define GET_GPSTATE(x) (((x) >> GPSTATE_SHIFT) & 0xFF) #define MAX_RAMP_DOWN_TIME 5120 /* @@ -94,6 +92,7 @@ struct global_pstate_info { }; static struct cpufreq_frequency_table powernv_freqs[POWERNV_MAX_PSTATES+1]; +u32 pstate_sign_prefix; static bool rebooting, throttled, occ_reset; static const char * const throttle_reason[] = { @@ -148,6 +147,20 @@ static struct powernv_pstate_info { bool wof_enabled; } powernv_pstate_info; +static inline int extract_pstate(u64 pmsr_val, unsigned int shift) +{ + int ret = ((pmsr_val >> shift) & 0xFF); + + if (!ret) + return ret; + + return (pstate_sign_prefix | ret); +} + +#define extract_local_pstate(x) extract_pstate(x, LPSTATE_SHIFT) +#define extract_global_pstate(x) extract_pstate(x, GPSTATE_SHIFT) +#define extract_max_pstate(x) extract_pstate(x, MAX_PSTATE_SHIFT) + /* Use following macros for conversions between pstate_id and index */ static inline int idx_to_pstate(unsigned int i) { @@ -278,6 +291,9 @@ next: powernv_pstate_info.nr_pstates = nr_pstates; pr_debug("NR PStates %d\n", nr_pstates); + + pstate_sign_prefix = pstate_min & ~0xFF; + for (i = 0; i < nr_pstates; i++) { u32 id = be32_to_cpu(pstate_ids[i]); u32 freq = be32_to_cpu(pstate_freqs[i]); @@ -438,17 +454,10 @@ struct powernv_smp_call_data { static void powernv_read_cpu_freq(void *arg) { unsigned long pmspr_val; - s8 local_pstate_id; struct powernv_smp_call_data *freq_data = arg; pmspr_val = get_pmspr(SPRN_PMSR); - - /* - * The local pstate id corresponds bits 48..55 in the PMSR. - * Note: Watch out for the sign! - */ - local_pstate_id = (pmspr_val >> 48) & 0xFF; - freq_data->pstate_id = local_pstate_id; + freq_data->pstate_id = extract_local_pstate(pmspr_val); freq_data->freq = pstate_id_to_freq(freq_data->pstate_id); pr_debug("cpu %d pmsr %016lX pstate_id %d frequency %d kHz\n", @@ -522,7 +531,7 @@ static void powernv_cpufreq_throttle_che chip = this_cpu_read(chip_info); /* Check for Pmax Capping */ - pmsr_pmax = (s8)PMSR_MAX(pmsr); + pmsr_pmax = extract_max_pstate(pmsr); pmsr_pmax_idx = pstate_to_idx(pmsr_pmax); if (pmsr_pmax_idx != powernv_pstate_info.max) { if (chip->throttled) @@ -645,8 +654,8 @@ void gpstate_timer_handler(struct timer_ * value. Hence, read from PMCR to get correct data. */ val = get_pmspr(SPRN_PMCR); - freq_data.gpstate_id = (s8)GET_GPSTATE(val); - freq_data.pstate_id = (s8)GET_LPSTATE(val); + freq_data.gpstate_id = extract_global_pstate(val); + freq_data.pstate_id = extract_local_pstate(val); if (freq_data.gpstate_id == freq_data.pstate_id) { reset_gpstates(policy); spin_unlock(&gpstates->gpstate_lock); Patches currently in stable-queue which might be from ego(a)linux.vnet.ibm.com are queue-4.15/powernv-cpufreq-add-helper-to-extract-pstate-from-pmsr.patch

6 years, 9 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror April 2018