[PATCH AUTOSEL 6.14 01/30] scsi: hisi_sas: Fix I/O errors caused by hardware port ID changes

[PATCH AUTOSEL 6.14 01/30] scsi: hisi_sas: Fix I/O errors caused by hardware port ID changes

[Sasha Levin]

From: Xingui Yang <yangxingui@huawei.com>

[ Upstream commit daff37f00c7506ca322ccfce95d342022f06ec58 ]

The hw port ID of phy may change when inserting disks in batches, causing
the port ID in hisi_sas_port and itct to be inconsistent with the hardware,
resulting in I/O errors. The solution is to set the device state to gone to
intercept I/O sent to the device, and then execute linkreset to discard and
find the disk to re-update its information.

Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20250312095135.3048379-3-yangxingui@huawei.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 3596414d970b2..7a484ad0f9abe 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -935,8 +935,28 @@ static void hisi_sas_phyup_work_common(struct work_struct *work,
 		container_of(work, typeof(*phy), works[event]);
 	struct hisi_hba *hisi_hba = phy->hisi_hba;
 	struct asd_sas_phy *sas_phy = &phy->sas_phy;
+	struct asd_sas_port *sas_port = sas_phy->port;
+	struct hisi_sas_port *port = phy->port;
+	struct device *dev = hisi_hba->dev;
+	struct domain_device *port_dev;
 	int phy_no = sas_phy->id;
 
+	if (!test_bit(HISI_SAS_RESETTING_BIT, &hisi_hba->flags) &&
+	    sas_port && port && (port->id != phy->port_id)) {
+		dev_info(dev, "phy%d's hw port id changed from %d to %llu\n",
+				phy_no, port->id, phy->port_id);
+		port_dev = sas_port->port_dev;
+		if (port_dev && !dev_is_expander(port_dev->dev_type)) {
+			/*
+			 * Set the device state to gone to block
+			 * sending IO to the device.
+			 */
+			set_bit(SAS_DEV_GONE, &port_dev->state);
+			hisi_sas_notify_phy_event(phy, HISI_PHYE_LINK_RESET);
+			return;
+		}
+	}
+
 	phy->wait_phyup_cnt = 0;
 	if (phy->identify.target_port_protocols == SAS_PROTOCOL_SSP)
 		hisi_hba->hw->sl_notify_ssp(hisi_hba, phy_no);
-- 
2.39.5

[PATCH AUTOSEL 6.14 02/30] scsi: ufs: exynos: Ensure pre_link() executes before exynos_ufs_phy_init()

[Sasha Levin]

From: Peter Griffin <peter.griffin@linaro.org>

[ Upstream commit 3d101165e72316775947d71321d97194f03dfef3 ]

Ensure clocks are enabled before configuring unipro. Additionally move
the pre_link() hook before the exynos_ufs_phy_init() calls. This means
the register write sequence more closely resembles the ordering of the
downstream driver.

Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Link: https://lore.kernel.org/r/20250319-exynos-ufs-stability-fixes-v2-1-96722cc2ba1b@linaro.org
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/ufs/host/ufs-exynos.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index 13dd5dfc03eb3..cd750786187cd 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -1049,9 +1049,14 @@ static int exynos_ufs_pre_link(struct ufs_hba *hba)
 	exynos_ufs_config_intr(ufs, DFES_DEF_L4_ERRS, UNIPRO_L4);
 	exynos_ufs_set_unipro_pclk_div(ufs);
 
+	exynos_ufs_setup_clocks(hba, true, PRE_CHANGE);
+
 	/* unipro */
 	exynos_ufs_config_unipro(ufs);
 
+	if (ufs->drv_data->pre_link)
+		ufs->drv_data->pre_link(ufs);
+
 	/* m-phy */
 	exynos_ufs_phy_init(ufs);
 	if (!(ufs->opts & EXYNOS_UFS_OPT_SKIP_CONFIG_PHY_ATTR)) {
@@ -1059,11 +1064,6 @@ static int exynos_ufs_pre_link(struct ufs_hba *hba)
 		exynos_ufs_config_phy_cap_attr(ufs);
 	}
 
-	exynos_ufs_setup_clocks(hba, true, PRE_CHANGE);
-
-	if (ufs->drv_data->pre_link)
-		ufs->drv_data->pre_link(ufs);
-
 	return 0;
 }
 
-- 
2.39.5

[PATCH AUTOSEL 6.14 03/30] scsi: ufs: exynos: Enable PRDT pre-fetching with UFSHCD_CAP_CRYPTO

[Sasha Levin]

From: Peter Griffin <peter.griffin@linaro.org>

[ Upstream commit deac9ad496ec17e1ec06848964ecc635bdaca703 ]

PRDT_PREFETCH_ENABLE[31] bit should be set when desctype field of
fmpsecurity0 register is type2 (double file encryption) or type3
(support for file and disk encryption). Setting this bit enables PRDT
pre-fetching on both TXPRDT and RXPRDT.

Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Link: https://lore.kernel.org/r/20250319-exynos-ufs-stability-fixes-v2-5-96722cc2ba1b@linaro.org
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/ufs/host/ufs-exynos.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index cd750786187cd..b441178089cdd 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -34,7 +34,7 @@
  * Exynos's Vendor specific registers for UFSHCI
  */
 #define HCI_TXPRDT_ENTRY_SIZE	0x00
-#define PRDT_PREFECT_EN		BIT(31)
+#define PRDT_PREFETCH_EN	BIT(31)
 #define HCI_RXPRDT_ENTRY_SIZE	0x04
 #define HCI_1US_TO_CNT_VAL	0x0C
 #define CNT_VAL_1US_MASK	0x3FF
@@ -1087,12 +1087,17 @@ static int exynos_ufs_post_link(struct ufs_hba *hba)
 	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
 	struct phy *generic_phy = ufs->phy;
 	struct exynos_ufs_uic_attr *attr = ufs->drv_data->uic_attr;
+	u32 val = ilog2(DATA_UNIT_SIZE);
 
 	exynos_ufs_establish_connt(ufs);
 	exynos_ufs_fit_aggr_timeout(ufs);
 
 	hci_writel(ufs, 0xa, HCI_DATA_REORDER);
-	hci_writel(ufs, ilog2(DATA_UNIT_SIZE), HCI_TXPRDT_ENTRY_SIZE);
+
+	if (hba->caps & UFSHCD_CAP_CRYPTO)
+		val |= PRDT_PREFETCH_EN;
+	hci_writel(ufs, val, HCI_TXPRDT_ENTRY_SIZE);
+
 	hci_writel(ufs, ilog2(DATA_UNIT_SIZE), HCI_RXPRDT_ENTRY_SIZE);
 	hci_writel(ufs, (1 << hba->nutrs) - 1, HCI_UTRL_NEXUS_TYPE);
 	hci_writel(ufs, (1 << hba->nutmrs) - 1, HCI_UTMRL_NEXUS_TYPE);
-- 
2.39.5

[PATCH AUTOSEL 6.14 04/30] scsi: ufs: exynos: Move phy calls to .exit() callback

[Sasha Levin]

From: Peter Griffin <peter.griffin@linaro.org>

[ Upstream commit 67e4085015c33bf2fb552af1f171c58b81ef0616 ]

ufshcd_pltfrm_remove() calls ufshcd_remove(hba) which in turn calls
ufshcd_hba_exit().

By moving the phy_power_off() and phy_exit() calls to the newly created
.exit callback they get called by ufshcd_variant_hba_exit() before
ufshcd_hba_exit() turns off the regulators. This is also similar flow to
the ufs-qcom driver.

Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Link: https://lore.kernel.org/r/20250319-exynos-ufs-stability-fixes-v2-6-96722cc2ba1b@linaro.org
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/ufs/host/ufs-exynos.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index b441178089cdd..e5c101480f09d 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -1501,6 +1501,14 @@ static int exynos_ufs_init(struct ufs_hba *hba)
 	return ret;
 }
 
+static void exynos_ufs_exit(struct ufs_hba *hba)
+{
+	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
+
+	phy_power_off(ufs->phy);
+	phy_exit(ufs->phy);
+}
+
 static int exynos_ufs_host_reset(struct ufs_hba *hba)
 {
 	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
@@ -1956,6 +1964,7 @@ static int gs101_ufs_pre_pwr_change(struct exynos_ufs *ufs,
 static const struct ufs_hba_variant_ops ufs_hba_exynos_ops = {
 	.name				= "exynos_ufs",
 	.init				= exynos_ufs_init,
+	.exit				= exynos_ufs_exit,
 	.hce_enable_notify		= exynos_ufs_hce_enable_notify,
 	.link_startup_notify		= exynos_ufs_link_startup_notify,
 	.pwr_change_notify		= exynos_ufs_pwr_change_notify,
@@ -1994,13 +2003,7 @@ static int exynos_ufs_probe(struct platform_device *pdev)
 
 static void exynos_ufs_remove(struct platform_device *pdev)
 {
-	struct ufs_hba *hba =  platform_get_drvdata(pdev);
-	struct exynos_ufs *ufs = ufshcd_get_variant(hba);
-
 	ufshcd_pltfrm_remove(pdev);
-
-	phy_power_off(ufs->phy);
-	phy_exit(ufs->phy);
 }
 
 static struct exynos_ufs_uic_attr exynos7_uic_attr = {
-- 
2.39.5

[PATCH AUTOSEL 6.14 05/30] scsi: ufs: exynos: gs101: Put UFS device in reset on .suspend()

[Sasha Levin]

From: Peter Griffin <peter.griffin@linaro.org>

[ Upstream commit cd4c0025069f16fc666c6ffc56c49c9b1154841f ]

GPIO_OUT[0] is connected to the reset pin of embedded UFS device.
Before powering off the phy assert the reset signal.

This is added as a gs101 specific suspend hook so as not to have any
unintended consequences for other SoCs supported by this driver.

Signed-off-by: Peter Griffin <peter.griffin@linaro.org>
Link: https://lore.kernel.org/r/20250319-exynos-ufs-stability-fixes-v2-7-96722cc2ba1b@linaro.org
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/ufs/host/ufs-exynos.c | 10 ++++++++++
 drivers/ufs/host/ufs-exynos.h |  1 +
 2 files changed, 11 insertions(+)

diff --git a/drivers/ufs/host/ufs-exynos.c b/drivers/ufs/host/ufs-exynos.c
index e5c101480f09d..c593d6b73c4c7 100644
--- a/drivers/ufs/host/ufs-exynos.c
+++ b/drivers/ufs/host/ufs-exynos.c
@@ -1679,6 +1679,12 @@ static void exynos_ufs_hibern8_notify(struct ufs_hba *hba,
 	}
 }
 
+static int gs101_ufs_suspend(struct exynos_ufs *ufs)
+{
+	hci_writel(ufs, 0 << 0, HCI_GPIO_OUT);
+	return 0;
+}
+
 static int exynos_ufs_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op,
 	enum ufs_notify_change_status status)
 {
@@ -1687,6 +1693,9 @@ static int exynos_ufs_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op,
 	if (status == PRE_CHANGE)
 		return 0;
 
+	if (ufs->drv_data->suspend)
+		ufs->drv_data->suspend(ufs);
+
 	if (!ufshcd_is_link_active(hba))
 		phy_power_off(ufs->phy);
 
@@ -2147,6 +2156,7 @@ static const struct exynos_ufs_drv_data gs101_ufs_drvs = {
 	.pre_link		= gs101_ufs_pre_link,
 	.post_link		= gs101_ufs_post_link,
 	.pre_pwr_change		= gs101_ufs_pre_pwr_change,
+	.suspend		= gs101_ufs_suspend,
 };
 
 static const struct of_device_id exynos_ufs_of_match[] = {
diff --git a/drivers/ufs/host/ufs-exynos.h b/drivers/ufs/host/ufs-exynos.h
index 9670dc138d1e4..b981cfa08b596 100644
--- a/drivers/ufs/host/ufs-exynos.h
+++ b/drivers/ufs/host/ufs-exynos.h
@@ -191,6 +191,7 @@ struct exynos_ufs_drv_data {
 				struct ufs_pa_layer_attr *pwr);
 	int (*pre_hce_enable)(struct exynos_ufs *ufs);
 	int (*post_hce_enable)(struct exynos_ufs *ufs);
+	int (*suspend)(struct exynos_ufs *ufs);
 };
 
 struct ufs_phy_time_cfg {
-- 
2.39.5

[PATCH AUTOSEL 6.14 06/30] scsi: pm80xx: Set phy_attached to zero when device is gone

[Sasha Levin]

From: Igor Pylypiv <ipylypiv@google.com>

[ Upstream commit f7b705c238d1483f0a766e2b20010f176e5c0fb7 ]

When a fatal error occurs, a phy down event may not be received to set
phy->phy_attached to zero.

Signed-off-by: Igor Pylypiv <ipylypiv@google.com>
Signed-off-by: Salomon Dushimirimana <salomondush@google.com>
Link: https://lore.kernel.org/r/20250319230305.3172920-1-salomondush@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/scsi/pm8001/pm8001_sas.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/scsi/pm8001/pm8001_sas.c b/drivers/scsi/pm8001/pm8001_sas.c
index 183ce00aa671e..f7067878b34f3 100644
--- a/drivers/scsi/pm8001/pm8001_sas.c
+++ b/drivers/scsi/pm8001/pm8001_sas.c
@@ -766,6 +766,7 @@ static void pm8001_dev_gone_notify(struct domain_device *dev)
 			spin_lock_irqsave(&pm8001_ha->lock, flags);
 		}
 		PM8001_CHIP_DISP->dereg_dev_req(pm8001_ha, device_id);
+		pm8001_ha->phy[pm8001_dev->attached_phy].phy_attached = 0;
 		pm8001_free_dev(pm8001_dev);
 	} else {
 		pm8001_dbg(pm8001_ha, DISC, "Found dev has gone.\n");
-- 
2.39.5