From: Jason Yan <yanaijie@huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID9GQH -------------------------------- Upstream: No sas_rediscover() returns error code if discover failed for a expander phy. And sas_ex_revalidate_domain() only returns the last phy's error code. So when sas_revalidate_domain() prints the return value of the discover process, we do not know if the revalidation for every phy is successful or not. We just know the last bcast phy revalidation succeeded or not. No need to return a error code for sas_ex_revalidate_domain() and sas_rediscover(), and just print the debug log for each bcast phy directly in sas_rediscover(). Signed-off-by: Jason Yan <yanaijie@huawei.com> CC: John Garry <john.garry@huawei.com> CC: Johannes Thumshirn <jthumshirn@suse.de> CC: Ewan Milne <emilne@redhat.com> CC: Christoph Hellwig <hch@lst.de> CC: Tomas Henzl <thenzl@redhat.com> CC: Dan Williams <dan.j.williams@intel.com> CC: Hannes Reinecke <hare@suse.com> Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> --- drivers/scsi/libsas/sas_discover.c | 7 +++---- drivers/scsi/libsas/sas_expander.c | 11 ++++++----- include/scsi/libsas.h | 2 +- 3 files changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c index f2ad4c5eadc3..4932f63b69bb 100644 --- a/drivers/scsi/libsas/sas_discover.c +++ b/drivers/scsi/libsas/sas_discover.c @@ -518,7 +518,6 @@ static void sas_discover_domain(struct work_struct *work) static void sas_revalidate_domain(struct work_struct *work) { - int res = 0; struct sas_discovery_event *ev = to_sas_discovery_event(work); struct asd_sas_port *port = ev->port; struct sas_ha_struct *ha = port->ha; @@ -538,10 +537,10 @@ static void sas_revalidate_domain(struct work_struct *work) task_pid_nr(current)); if (ddev && dev_is_expander(ddev->dev_type)) - res = sas_ex_revalidate_domain(ddev); + sas_ex_revalidate_domain(ddev); - pr_debug("done REVALIDATING DOMAIN on port %d, pid:%d, res 0x%x\n", - port->id, task_pid_nr(current), res); + pr_debug("done REVALIDATING DOMAIN on port %d, pid:%d\n", + port->id, task_pid_nr(current)); out: mutex_unlock(&ha->disco_mutex); diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index b27d39f56d1a..1b6230c646a1 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -2112,7 +2112,7 @@ static int sas_rediscover_dev(struct domain_device *dev, int phy_id, * first phy,for other phys in this port, we add it to the port to * forming the wide-port. */ -static int sas_rediscover(struct domain_device *dev, const int phy_id) +static void sas_rediscover(struct domain_device *dev, const int phy_id) { struct expander_device *ex = &dev->ex_dev; struct ex_phy *changed_phy = &ex->ex_phy[phy_id]; @@ -2138,7 +2138,9 @@ static int sas_rediscover(struct domain_device *dev, const int phy_id) res = sas_rediscover_dev(dev, phy_id, last, i); } else res = sas_discover_new(dev, phy_id); - return res; + + pr_debug("ex %016llx phy%d discover returned 0x%x\n", + SAS_ADDR(dev->sas_addr), phy_id, res); } /** @@ -2150,7 +2152,7 @@ static int sas_rediscover(struct domain_device *dev, const int phy_id) * Discover process only interrogates devices in order to discover the * domain. */ -int sas_ex_revalidate_domain(struct domain_device *port_dev) +void sas_ex_revalidate_domain(struct domain_device *port_dev) { int res; struct domain_device *dev = NULL; @@ -2165,11 +2167,10 @@ int sas_ex_revalidate_domain(struct domain_device *port_dev) res = sas_find_bcast_phy(dev, &phy_id, i, true); if (phy_id == -1) break; - res = sas_rediscover(dev, phy_id); + sas_rediscover(dev, phy_id); i = phy_id + 1; } while (i < ex->num_phys); } - return res; } void sas_smp_handler(struct bsg_job *job, struct Scsi_Host *shost, diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index 99228996991a..d97624e77fb1 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -684,7 +684,7 @@ int sas_discover_root_expander(struct domain_device *); void sas_init_ex_attr(void); -int sas_ex_revalidate_domain(struct domain_device *); +void sas_ex_revalidate_domain(struct domain_device *); void sas_unregister_domain_devices(struct asd_sas_port *port, int gone); void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *); -- 2.46.1

From: Jason Yan <yanaijie@huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID9GQH -------------------------------- Upstream: No The ata device do not have a real sas address. If a ata device is replaced with another one, the sas address is the same. Now libsas treat this scenario as flutter and do not delete the old one and discover the new one. This will cause the data read from or write to the wrong device. And also when hotplugging a sata device, libsas entered to the flutter case and sometimes found the phy attached address is abnormal. The log is like this: sas: ex 500e004aaaaaaa1f phy6 originated BROADCAST(CHANGE) sas: ex 500e004aaaaaaa1f phy06:U:0 attached: 0000000000000000 (no device) sas: ex 500e004aaaaaaa1f phy 0x6 broadcast flutter Fix this issue by checking the phy attached address and the ata device's class and id if they are the same as the origin. The ata class and id is readed in ata EH process. When ata EH is scheduled, revalidate will be deferred and a new bcast will be raised. Tested-by: Chen Liangfei <chenliangfei1@huawei.com> Signed-off-by: Jason Yan <yanaijie@huawei.com> Reviewed-by: John Garry <john.garry@huawei.com> CC: chenxiang <chenxiang66@hisilicon.com> CC: John Garry <john.garry@huawei.com> CC: Johannes Thumshirn <jthumshirn@suse.de> CC: Ewan Milne <emilne@redhat.com> CC: Christoph Hellwig <hch@lst.de> CC: Tomas Henzl <thenzl@redhat.com> CC: Dan Williams <dan.j.williams@intel.com> CC: Tejun Heo <tj@kernel.org> CC: Hannes Reinecke <hare@suse.com> Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> --- drivers/ata/libata-core.c | 3 +- drivers/scsi/libsas/sas_ata.c | 18 ++++++++ drivers/scsi/libsas/sas_expander.c | 67 +++++++++++++++++++++++++----- include/linux/libata.h | 2 + include/scsi/libsas.h | 1 + 5 files changed, 80 insertions(+), 11 deletions(-) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 107c28ec23b8..11c91d7231aa 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -3628,7 +3628,7 @@ EXPORT_SYMBOL_GPL(ata_std_postreset); * RETURNS: * 1 if @dev matches @new_class and @new_id, 0 otherwise. */ -static int ata_dev_same_device(struct ata_device *dev, unsigned int new_class, +int ata_dev_same_device(struct ata_device *dev, unsigned int new_class, const u16 *new_id) { const u16 *old_id = dev->id; @@ -3660,6 +3660,7 @@ static int ata_dev_same_device(struct ata_device *dev, unsigned int new_class, return 1; } +EXPORT_SYMBOL_GPL(ata_dev_same_device); /** * ata_dev_reread_id - Re-read IDENTIFY data diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index e2445d5d41ca..88cb015fcc70 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -651,6 +651,22 @@ void sas_ata_task_abort(struct sas_task *task) complete(waiting); } +static void sas_ata_store_id(struct domain_device *dev) +{ + struct ata_device *ata_dev = sas_to_ata_dev(dev); + unsigned char model[ATA_ID_PROD_LEN + 1]; + unsigned char serial[ATA_ID_SERNO_LEN + 1]; + + /* store the ata device's class and id */ + memcpy(dev->sata_dev.id, ata_dev->id, ATA_ID_WORDS); + dev->sata_dev.class = ata_dev->class; + + ata_id_c_string(ata_dev->id, model, ATA_ID_PROD, sizeof(model)); + ata_id_c_string(ata_dev->id, serial, ATA_ID_SERNO, sizeof(serial)); + + sas_ata_printk(KERN_INFO, dev, "model:%s serial:%s\n", model, serial); +} + static int sas_get_ata_command_set(struct domain_device *dev) { struct dev_to_host_fis *fis = @@ -689,6 +705,8 @@ void sas_probe_sata(struct asd_sas_port *port) */ if (!ata_dev_enabled(sas_to_ata_dev(dev))) sas_fail_probe(dev, __func__, -ENODEV); + else + sas_ata_store_id(dev); } } diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index a7fd97ae01c3..320778178aca 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -2019,6 +2019,61 @@ static bool dev_type_flutter(enum sas_device_type new, enum sas_device_type old) return false; } +/* + * we think the device is fluttering so just read the phy state and update + * some information of the device, but if some important things changed + * such as the sas address, or the linkrate, or the ata devices id and class, + * we have to unregister the device and re-probe it. + */ +static bool sas_process_flutter(struct domain_device *dev, struct ex_phy *phy, + int phy_id, u8 *sas_addr) +{ + struct domain_device *ata_dev = sas_ex_to_ata(dev, phy_id); + enum sas_linkrate linkrate = phy->linkrate; + char *action = ""; + + sas_ex_phy_discover(dev, phy_id); + + if (ata_dev && phy->attached_dev_type == SAS_SATA_PENDING) + action = ", needs recovery"; + pr_debug("ex %016llx phy%d broadcast flutter%s\n", + SAS_ADDR(dev->sas_addr), phy_id, action); + + if (linkrate != phy->linkrate) { + pr_debug("ex %016llx phy%d linkrate changed from %d to %d\n", + SAS_ADDR(dev->sas_addr), phy_id, + linkrate, phy->linkrate); + return false; + } + + /* the phy attached address will be updated by sas_ex_phy_discover() + * and sometimes become abnormal + */ + if (SAS_ADDR(phy->attached_sas_addr) != SAS_ADDR(sas_addr) || + SAS_ADDR(phy->attached_sas_addr) == 0) { + /* if attached_sas_addr become abnormal, we must set the + * original address back so that the device can be unregistered + */ + memcpy(phy->attached_sas_addr, sas_addr, SAS_ADDR_SIZE); + pr_debug("phy address(%016llx) abnormal, origin:%016llx\n", + SAS_ADDR(phy->attached_sas_addr), + SAS_ADDR(sas_addr)); + return false; + } + + if (ata_dev) { + struct ata_device *adev = sas_to_ata_dev(ata_dev); + unsigned int class = ata_dev->sata_dev.class; + u16 *id = ata_dev->sata_dev.id; + + /* to see if the disk is replaced with another one */ + if (!ata_dev_same_device(adev, class, id)) + return false; + } + + return true; +} + static int sas_unregister(struct domain_device *dev, int phy_id, bool last, bool *retry, int sibling) { @@ -2074,16 +2129,8 @@ static int sas_unregister(struct domain_device *dev, int phy_id, bool last, goto out_free_resp; } else if (SAS_ADDR(sas_addr) == SAS_ADDR(phy->attached_sas_addr) && dev_type_flutter(type, phy->attached_dev_type)) { - struct domain_device *ata_dev = sas_ex_to_ata(dev, phy_id); - char *action = ""; - - sas_ex_phy_discover(dev, phy_id); - - if (ata_dev && phy->attached_dev_type == SAS_SATA_PENDING) - action = ", needs recovery"; - pr_debug("ex %016llx phy%02d broadcast flutter%s\n", - SAS_ADDR(dev->sas_addr), phy_id, action); - goto out_free_resp; + if (sas_process_flutter(dev, phy, phy_id, sas_addr)) + goto out_free_resp; } /* we always have to delete the old device when we went here */ diff --git a/include/linux/libata.h b/include/linux/libata.h index 002cf4d1dcf4..c704335346ca 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -1110,6 +1110,8 @@ extern int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *dev, unsigned int cmd, void __user *arg); extern bool ata_link_online(struct ata_link *link); extern bool ata_link_offline(struct ata_link *link); +extern int ata_dev_same_device(struct ata_device *dev, unsigned int new_class, + const u16 *new_id); #ifdef CONFIG_PM extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg); extern void ata_host_resume(struct ata_host *host); diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index b7bb9d493800..799574dbb737 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -147,6 +147,7 @@ struct sata_device { struct ata_host *ata_host; struct smp_resp rps_resp ____cacheline_aligned; /* report_phy_sata_resp */ u8 fis[ATA_RESP_FIS_SIZE]; + u16 id[ATA_ID_WORDS]; }; struct ssp_device { -- 2.46.1

From: Jason Yan <yanaijie@huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID9GQH -------------------------------- Upstream: No The work flow of revalidation now is scanning expander phy by the sequence of the phy and check if the phy have changed. This will leads to an issue of swapping two sas disks on one expander. Assume we have two sas disks, connected with expander phy10 and phy11: phy10: 5000cca04eb1001d port-0:0:10 phy11: 5000cca04eb043ad port-0:0:11 Swap these two disks, and imaging the following scenario: revalidation 1: -->phy10: 0 --> delete phy10 domain device -->phy11: 5000cca04eb043ad (no change) revalidation done revalidation 2: -->step 1, check phy10: -->phy10: 5000cca04eb043ad --> add to wide port(port-0:0:11) (phy11 address is still 5000cca04eb043ad now) -->step 2, check phy11: -->phy11: 0 --> phy11 address is 0 now, but it's part of wide port(port-0:0:11), the domain device will not be deleted. revalidation done revalidation 3: -->phy10, 5000cca04eb043ad (no change) -->phy11: 5000cca04eb1001d --> try to add port-0:0:11 but failed, port-0:0:11 already exist, trigger a warning as follows revalidation done [14790.189699] sysfs: cannot create duplicate filename '/devices/pci0000:74/0000:74:02.0/host0/port-0:0/expander-0:0/port-0:0:11' [14790.201081] CPU: 25 PID: 5031 Comm: kworker/u192:3 Not tainted 4.16.0-rc1-191134-g138f084-dirty #228 [14790.210199] Hardware name: Huawei D06/D06, BIOS Hisilicon D06 EC UEFI Nemo 2.0 RC0 - B303 05/16/2018 [14790.219323] Workqueue: 0000:74:02.0_disco_q sas_revalidate_domain [14790.225404] Call trace: [14790.227842] dump_backtrace+0x0/0x18c [14790.231492] show_stack+0x14/0x1c [14790.234798] dump_stack+0x88/0xac [14790.238101] sysfs_warn_dup+0x64/0x7c [14790.241751] sysfs_create_dir_ns+0x90/0xa0 [14790.245835] kobject_add_internal+0xa0/0x284 [14790.250092] kobject_add+0xb8/0x11c [14790.253570] device_add+0xe8/0x598 [14790.256960] sas_port_add+0x24/0x50 [14790.260436] sas_ex_discover_devices+0xb10/0xc30 [14790.265040] sas_ex_revalidate_domain+0x1d8/0x518 [14790.269731] sas_revalidate_domain+0x12c/0x154 [14790.274163] process_one_work+0x128/0x2b0 [14790.278160] worker_thread+0x14c/0x408 [14790.281897] kthread+0xfc/0x128 [14790.285026] ret_from_fork+0x10/0x18 [14790.288598] ------------[ cut here ]------------ At last, the disk 5000cca04eb1001d is lost. The basic idea of fix this issue is to let the revalidation first scan all phys, and then unregisterring devices. Only when no devices need to be unregisterred, go to the next step to discover new devices. If there are devices need unregister, unregister those devices and tell the revalidation event processor to retry again. The next revalidation will process the discovering of the new devices. Tested-by: Chen Liangfei <chenliangfei1@huawei.com> Signed-off-by: Jason Yan <yanaijie@huawei.com> CC: Xiaofei Tan <tanxiaofei@huawei.com> CC: chenxiang <chenxiang66@hisilicon.com> CC: John Garry <john.garry@huawei.com> CC: Johannes Thumshirn <jthumshirn@suse.de> CC: Ewan Milne <emilne@redhat.com> CC: Christoph Hellwig <hch@lst.de> CC: Tomas Henzl <thenzl@redhat.com> CC: Dan Williams <dan.j.williams@intel.com> CC: Hannes Reinecke <hare@suse.com> Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> --- drivers/scsi/libsas/sas_expander.c | 153 +++++++++++++++++++---------- 1 file changed, 102 insertions(+), 51 deletions(-) diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index 74cc362a11ad..f08da82e699b 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -2101,22 +2101,18 @@ static bool sas_process_flutter(struct domain_device *dev, struct ex_phy *phy, return true; } -static int sas_unregister(struct domain_device *dev, int phy_id, bool last, - bool *retry, int sibling) +static int sas_ex_unregister(struct domain_device *dev, int phy_id, bool last, + bool *retry) { struct expander_device *ex = &dev->ex_dev; struct ex_phy *phy = &ex->ex_phy[phy_id]; enum sas_device_type type = SAS_PHY_UNUSED; struct smp_disc_resp *disc_resp; u8 sas_addr[SAS_ADDR_SIZE]; - char msg[80] = ""; int res; - if (!last) - sprintf(msg, ", part of a wide port with phy%02d", sibling); - - pr_debug("ex %016llx unregistering phy%02d%s\n", - SAS_ADDR(dev->sas_addr), phy_id, msg); + pr_debug("ex %016llx unregistering phy%02d\n", + SAS_ADDR(dev->sas_addr), phy_id); memset(sas_addr, 0, SAS_ADDR_SIZE); disc_resp = alloc_smp_resp(DISCOVER_RESP_SIZE); @@ -2176,21 +2172,7 @@ static int sas_unregister(struct domain_device *dev, int phy_id, bool last, return res; } -/** - * sas_rediscover - revalidate the domain. - * @dev:domain device to be detect. - * @phy_id: the phy id will be detected. - * - * NOTE: this process _must_ quit (return) as soon as any connection - * errors are encountered. Connection recovery is done elsewhere. - * Discover process only interrogates devices in order to discover the - * domain.For plugging out, we un-register the device only when it is - * the last phy in the port, for other phys in this port, we just delete it - * from the port.For inserting, we do discovery when it is the - * first phy,for other phys in this port, we add it to the port to - * forming the wide-port. - */ -static void sas_rediscover(struct domain_device *dev, const int phy_id, +static void sas_ex_unregister_device(struct domain_device *dev, const int phy_id, bool *retry) { struct expander_device *ex = &dev->ex_dev; @@ -2199,29 +2181,70 @@ static void sas_rediscover(struct domain_device *dev, const int phy_id, int i; bool last = true; /* is this the last phy of the port */ - pr_debug("ex %016llx phy%02d originated BROADCAST(CHANGE)\n", - SAS_ADDR(dev->sas_addr), phy_id); - - if (SAS_ADDR(changed_phy->attached_sas_addr) != 0) { - for (i = 0; i < ex->num_phys; i++) { - struct ex_phy *phy = &ex->ex_phy[i]; + for (i = 0; i < ex->num_phys; i++) { + struct ex_phy *phy = &ex->ex_phy[i]; - if (i == phy_id) - continue; - if (SAS_ADDR(phy->attached_sas_addr) == - SAS_ADDR(changed_phy->attached_sas_addr)) { - last = false; - break; - } + if (i == phy_id) + continue; + if (SAS_ADDR(phy->attached_sas_addr) == + SAS_ADDR(changed_phy->attached_sas_addr)) { + pr_debug("phy%02d part of wide port with phy%02d\n", + phy_id, i); + last = false; + break; } - res = sas_unregister(dev, phy_id, last, retry, i); - } else - res = sas_discover_new(dev, phy_id); + } + res = sas_ex_unregister(dev, phy_id, last, retry); pr_debug("ex %016llx phy%d discover returned 0x%x\n", SAS_ADDR(dev->sas_addr), phy_id, res); } +static int sas_ex_try_unregister(struct domain_device *dev, u8 *changed_phy, + int nr, bool *retry) +{ + struct expander_device *ex = &dev->ex_dev; + int unregistered = 0; + struct ex_phy *phy; + int i; + + for (i = 0; i < nr; i++) { + pr_debug("ex %016llx phy%d originated BROADCAST(CHANGE)\n", + SAS_ADDR(dev->sas_addr), changed_phy[i]); + + phy = &ex->ex_phy[changed_phy[i]]; + + if (SAS_ADDR(phy->attached_sas_addr) == 0) + continue; + + sas_ex_unregister_device(dev, changed_phy[i], retry); + changed_phy[i] = 0xff; + unregistered++; + } + return unregistered; +} + +static void sas_ex_register(struct domain_device *dev, u8 *changed_phy, + int nr) +{ + struct expander_device *ex = &dev->ex_dev; + struct ex_phy *phy; + int res = 0; + int i; + + for (i = 0; i < nr; i++) { + if (changed_phy[i] == 0xff) + continue; + + phy = &ex->ex_phy[changed_phy[i]]; + + res = sas_discover_new(dev, changed_phy[i]); + + pr_debug("ex %016llx phy%d register returned 0x%x\n", + SAS_ADDR(dev->sas_addr), changed_phy[i], res); + } +} + /** * sas_ex_revalidate_domain - revalidate the domain * @port_dev: port domain device. @@ -2236,21 +2259,49 @@ void sas_ex_revalidate_domain(struct domain_device *port_dev, bool *retry) { int res; struct domain_device *dev = NULL; + u8 changed_phy[MAX_EXPANDER_PHYS]; + struct expander_device *ex; + int unregistered = 0; + int phy_id; + int nr = 0; + int i = 0; res = sas_find_bcast_dev(port_dev, &dev); - if (res == 0 && dev) { - struct expander_device *ex = &dev->ex_dev; - int i = 0, phy_id; - - do { - phy_id = -1; - res = sas_find_bcast_phy(dev, &phy_id, i, true); - if (phy_id == -1) - break; - sas_rediscover(dev, phy_id, retry); - i = phy_id + 1; - } while (i < ex->num_phys); + if (res || !dev) + return; + + memset(changed_phy, 0xff, MAX_EXPANDER_PHYS); + ex = &dev->ex_dev; + + do { + phy_id = -1; + res = sas_find_bcast_phy(dev, &phy_id, i, true); + if (phy_id == -1) + break; + changed_phy[nr++] = phy_id; + i = phy_id + 1; + } while (i < dev->ex_dev.num_phys); + + if (nr == 0) + return; + + unregistered = sas_ex_try_unregister(dev, changed_phy, nr, retry); + + if (unregistered > 0) { + struct ex_phy *phy; + + for (i = 0; i < nr; i++) { + if (changed_phy[i] == 0xff) + continue; + phy = &ex->ex_phy[changed_phy[i]]; + phy->phy_change_count = -1; + } + ex->ex_change_count = -1; + *retry = true; + return; } + + sas_ex_register(dev, changed_phy, nr); } void sas_smp_handler(struct bsg_job *job, struct Scsi_Host *shost, -- 2.46.1

From: Luo Jiaxing <luojiaxing@huawei.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID9GQH -------------------------------- Upstream: No We found out that abnormal scenario handle for flutter can not cover all abnormal scenario,so we change the order to let it handle eralier. Signed-off-by: Luo Jiaxing <luojiaxing@huawei.com> DTS:DTS2019041908574 Description: NA Team:HISI_SW Feature or Bugfix:Bugfix Change-Id: I69ea02d622ae2873d8bace66da7ad0fee9cf90fd Signed-off-by: l00437090 <l00437090@huawei.com> Reviewed-on: http://10.90.31.173:8080/6839 Tested-by: public TuringEE <turingee@huawei.com> Reviewed-by: chenxiang 00284940 <chenxiang66@hisilicon.com> Reviewed-by: public TuringEE <turingee@huawei.com> Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> --- drivers/scsi/libsas/sas_expander.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index f1ec8d5d3573..35b0b9e6f3f2 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -2066,13 +2066,6 @@ static bool sas_process_flutter(struct domain_device *dev, struct ex_phy *phy, pr_debug("ex %016llx phy%d broadcast flutter%s\n", SAS_ADDR(dev->sas_addr), phy_id, action); - if (linkrate != phy->linkrate) { - pr_debug("ex %016llx phy%d linkrate changed from %d to %d\n", - SAS_ADDR(dev->sas_addr), phy_id, - linkrate, phy->linkrate); - return false; - } - /* the phy attached address will be updated by sas_ex_phy_discover() * and sometimes become abnormal */ @@ -2088,6 +2081,13 @@ static bool sas_process_flutter(struct domain_device *dev, struct ex_phy *phy, return false; } + if (linkrate != phy->linkrate) { + pr_debug("ex %016llx phy%d linkrate changed from %d to %d\n", + SAS_ADDR(dev->sas_addr), phy_id, + linkrate, phy->linkrate); + return false; + } + if (ata_dev) { struct ata_device *adev = sas_to_ata_dev(ata_dev); unsigned int class = ata_dev->sata_dev.class; -- 2.46.1

From: Xiang Chen <chenxiang66@hisilicon.com> hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID9GQH -------------------------------- Upstream: No delete sas port out of the disco mutex Signed-off-by: Jason Yan <yanaijie@huawei.com> DTS:DTS2019032908508 Description:libsas: delete sas port out of the disco mutex Team:HISI_SW Feature or Bugfix:Bugfix Change-Id: I620775c266273577ea3a7e373904c292870a0f15 Signed-off-by: c00284940 <c00284940@huawei.com> Reviewed-on: http://10.90.31.173:8080/5908 Tested-by: public TuringEE <turingee@huawei.com> Reviewed-by: tanxiaofei 00396350 <tanxiaofei@huawei.com> Reviewed-by: public TuringEE <turingee@huawei.com> Signed-off-by: x00470154 <x00470154@huawei.com> Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com> --- drivers/scsi/libsas/sas_expander.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index 1945def5d78d..48f1ef4ca524 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -931,7 +931,7 @@ static struct domain_device *sas_ex_discover_end_dev( spin_unlock_irq(&parent->port->dev_list_lock); out_free: dev_printk(KERN_INFO, &phy->port->dev, "port deleted due to failed discover\n"); - sas_port_delete(phy->port); + list_add_tail(&phy->port->del_list, &parent->port->sas_port_del_list); out_err: phy->port = NULL; sas_put_device(child); -- 2.46.1