linux.git/drivers/ufs/host, branch v6.6.72

scsi: ufs: qcom: Power off the PHY if it was already powered on in ufs_qcom_power_up_sequence()

2025-01-17T12:36:17+00:00

commit 7bac65687510038390a0a54cbe14fba08d037e46 upstream.

PHY might already be powered on during ufs_qcom_power_up_sequence() in a
couple of cases:

 1. During UFSHCD_QUIRK_REINIT_AFTER_MAX_GEAR_SWITCH quirk

 2. Resuming from spm_lvl = 5 suspend

In those cases, it is necessary to call phy_power_off() and phy_exit() in
ufs_qcom_power_up_sequence() function to power off the PHY before calling
phy_init() and phy_power_on().

Case (1) is doing it via ufs_qcom_reinit_notify() callback, but case (2) is
not handled. So to satisfy both cases, call phy_power_off() and phy_exit()
if the phy_count is non-zero. And with this change, the reinit_notify()
callback is no longer needed.

This fixes the below UFS resume failure with spm_lvl = 5:

ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: ufshcd_host_reset_and_restore: Host init failed -5
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: ufshcd_host_reset_and_restore: Host init failed -5
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: ufshcd_host_reset_and_restore: Host init failed -5
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: ufshcd_host_reset_and_restore: Host init failed -5
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: Enabling the controller failed
ufshcd-qcom 1d84000.ufshc: ufshcd_host_reset_and_restore: Host init failed -5
ufs_device_wlun 0:0:0:49488: ufshcd_wl_resume failed: -5
ufs_device_wlun 0:0:0:49488: PM: dpm_run_callback(): scsi_bus_resume returns -5
ufs_device_wlun 0:0:0:49488: PM: failed to resume async: error -5

Cc: stable@vger.kernel.org # 6.3
Fixes: baf5ddac90dc ("scsi: ufs: ufs-qcom: Add support for reinitializing the UFS device")
Reported-by: Ram Kumar Dwivedi 
Tested-by: Amit Pundir  # on SM8550-HDK
Reviewed-by: Bart Van Assche 
Tested-by: Neil Armstrong  # on SM8550-QRD
Signed-off-by: Manivannan Sadhasivam 
Link: https://lore.kernel.org/r/20241219-ufs-qcom-suspend-fix-v3-1-63c4b95a70b9@linaro.org
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

scsi: ufs: core: Make DMA mask configuration more flexible

2024-12-14T19:00:11+00:00

[ Upstream commit 78bc671bd1501e2f6c571e063301a4fdc5db53b2 ]

Replace UFSHCD_QUIRK_BROKEN_64BIT_ADDRESS with
ufs_hba_variant_ops::set_dma_mask.  Update the Renesas driver
accordingly.  This patch enables supporting other configurations than
32-bit or 64-bit DMA addresses, e.g. 36-bit DMA addresses.

Signed-off-by: Bart Van Assche 
Link: https://lore.kernel.org/r/20241018194753.775074-1-bvanassche@acm.org
Reviewed-by: Avri Altman 
Reviewed-by: Peter Wang 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: exynos: Fix hibern8 notify callbacks

2024-12-09T09:33:06+00:00

commit ceef938bbf8b93ba3a218b4adc244cde94b582aa upstream.

v1 of the patch which introduced the ufshcd_vops_hibern8_notify()
callback used a bool instead of an enum. In v2 this was updated to an
enum based on the review feedback in [1].

ufs-exynos hibernate calls have always been broken upstream as it
follows the v1 bool implementation.

Link: https://patchwork.kernel.org/project/linux-scsi/patch/001f01d23994$719997c0$54ccc740$@samsung.com/ [1]
Fixes: 55f4b1f73631 ("scsi: ufs: ufs-exynos: Add UFS host support for Exynos SoCs")
Signed-off-by: Peter Griffin 
Link: https://lore.kernel.org/r/20241031150033.3440894-13-peter.griffin@linaro.org
Cc: stable@vger.kernel.org
Reviewed-by: Tudor Ambarus 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

scsi: ufs: qcom: Update MODE_MAX cfg_bw value

2024-10-04T14:29:49+00:00

commit 0c40f079f1c808e7e480c795a79009f200366eb1 upstream.

Commit 8db8f6ce556a ("scsi: ufs: qcom: Add missing interconnect bandwidth
values for Gear 5") updated the ufs_qcom_bw_table for Gear 5. However, it
missed updating the cfg_bw value for the max mode.

Hence update the cfg_bw value for the max mode for UFS 4.x devices.

Fixes: 8db8f6ce556a ("scsi: ufs: qcom: Add missing interconnect bandwidth values for Gear 5")
Cc: stable@vger.kernel.org
Signed-off-by: Manish Pandey 
Link: https://lore.kernel.org/r/20240903063709.4335-1-quic_mapa@quicinc.com
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Greg Kroah-Hartman

scsi: ufs: cdns-pltfrm: Perform read back after writing HCLKDIV

2024-06-12T09:11:39+00:00

[ Upstream commit b715c55daf598aac8fa339048e4ca8a0916b332e ]

Currently, HCLKDIV is written to and then completed with an mb().

mb() ensures that the write completes, but completion doesn't mean that it
isn't stored in a buffer somewhere. The recommendation for ensuring this
bit has taken effect on the device is to perform a read back to force it to
make it all the way to the device. This is documented in device-io.rst and
a talk by Will Deacon on this can be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

Let's do that to ensure the bit hits the device. Because the mb()'s purpose
wasn't to add extra ordering (on top of the ordering guaranteed by
writel()/readl()), it can safely be removed.

Fixes: d90996dae8e4 ("scsi: ufs: Add UFS platform driver for Cadence UFS")
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Andrew Halaney 
Link: https://lore.kernel.org/r/20240329-ufs-reset-ensure-effect-before-delay-v5-6-181252004586@redhat.com
Reviewed-by: Bart Van Assche 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: qcom: Perform read back after writing CGC enable

2024-06-12T09:11:39+00:00

[ Upstream commit d9488511b3ac7eb48a91bc5eded7027525525e03 ]

Currently, the CGC enable bit is written and then an mb() is used to ensure
that completes before continuing.

mb() ensures that the write completes, but completion doesn't mean that it
isn't stored in a buffer somewhere. The recommendation for ensuring this
bit has taken effect on the device is to perform a read back to force it to
make it all the way to the device. This is documented in device-io.rst and
a talk by Will Deacon on this can be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

Let's do that to ensure the bit hits the device. Because the mb()'s purpose
wasn't to add extra ordering (on top of the ordering guaranteed by
writel()/readl()), it can safely be removed.

Reviewed-by: Manivannan Sadhasivam 
Reviewed-by: Can Guo 
Fixes: 81c0fc51b7a7 ("ufs-qcom: add support for Qualcomm Technologies Inc platforms")
Signed-off-by: Andrew Halaney 
Link: https://lore.kernel.org/r/20240329-ufs-reset-ensure-effect-before-delay-v5-5-181252004586@redhat.com
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: qcom: Perform read back after writing unipro mode

2024-06-12T09:11:39+00:00

[ Upstream commit 823150ecf04f958213cf3bf162187cd1a91c885c ]

Currently, the QUNIPRO_SEL bit is written to and then an mb() is used to
ensure that completes before continuing.

mb() ensures that the write completes, but completion doesn't mean that it
isn't stored in a buffer somewhere. The recommendation for ensuring this
bit has taken effect on the device is to perform a read back to force it to
make it all the way to the device. This is documented in device-io.rst and
a talk by Will Deacon on this can be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

But, there's really no reason to even ensure completion before
continuing. The only requirement here is that this write is ordered to this
endpoint (which readl()/writel() guarantees already). For that reason the
mb() can be dropped altogether without anything forcing completion.

Fixes: f06fcc7155dc ("scsi: ufs-qcom: add QUniPro hardware support and power optimizations")
Signed-off-by: Andrew Halaney 
Link: https://lore.kernel.org/r/20240329-ufs-reset-ensure-effect-before-delay-v5-4-181252004586@redhat.com
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: qcom: Perform read back after writing REG_UFS_SYS1CLK_1US

2024-06-12T09:11:39+00:00

[ Upstream commit a862fafa263aea0f427d51aca6ff7fd9eeaaa8bd ]

Currently after writing to REG_UFS_SYS1CLK_1US a mb() is used to ensure
that write has gone through to the device.

mb() ensures that the write completes, but completion doesn't mean that it
isn't stored in a buffer somewhere. The recommendation for ensuring this
bit has taken effect on the device is to perform a read back to force it to
make it all the way to the device. This is documented in device-io.rst and
a talk by Will Deacon on this can be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

Let's do that to ensure the bit hits the device. Because the mb()'s purpose
wasn't to add extra ordering (on top of the ordering guaranteed by
writel()/readl()), it can safely be removed.

Fixes: f06fcc7155dc ("scsi: ufs-qcom: add QUniPro hardware support and power optimizations")
Reviewed-by: Can Guo 
Signed-off-by: Andrew Halaney 
Link: https://lore.kernel.org/r/20240329-ufs-reset-ensure-effect-before-delay-v5-2-181252004586@redhat.com
Reviewed-by: Manivannan Sadhasivam 
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: qcom: Perform read back after writing reset bit

2024-06-12T09:11:39+00:00

[ Upstream commit c4d28e06b0c94636f6e35d003fa9ebac0a94e1ae ]

Currently, the reset bit for the UFS provided reset controller (used by its
phy) is written to, and then a mb() happens to try and ensure that hit the
device. Immediately afterwards a usleep_range() occurs.

mb() ensures that the write completes, but completion doesn't mean that it
isn't stored in a buffer somewhere. The recommendation for ensuring this
bit has taken effect on the device is to perform a read back to force it to
make it all the way to the device. This is documented in device-io.rst and
a talk by Will Deacon on this can be seen over here:

    https://youtu.be/i6DayghhA8Q?si=MiyxB5cKJXSaoc01&t=1678

Let's do that to ensure the bit hits the device. By doing so and
guaranteeing the ordering against the immediately following usleep_range(),
the mb() can safely be removed.

Fixes: 81c0fc51b7a7 ("ufs-qcom: add support for Qualcomm Technologies Inc platforms")
Reviewed-by: Manivannan Sadhasivam 
Reviewed-by: Can Guo 
Signed-off-by: Andrew Halaney 
Link: https://lore.kernel.org/r/20240329-ufs-reset-ensure-effect-before-delay-v5-1-181252004586@redhat.com
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin

scsi: ufs: qcom: Add missing interconnect bandwidth values for Gear 5

2024-04-27T15:11:31+00:00

[ Upstream commit 8db8f6ce556af60ca9a9fd5e826d369ded70fcc7 ]

These entries are necessary to scale the interconnect bandwidth while
operating in Gear 5.

Cc: Amit Pundir 
Fixes: 03ce80a1bb86 ("scsi: ufs: qcom: Add support for scaling interconnects")
Tested-by: Amit Pundir 
Signed-off-by: Manivannan Sadhasivam 
Link: https://lore.kernel.org/r/20240403-ufs-icc-fix-v2-1-958412a5eb45@linaro.org
Signed-off-by: Martin K. Petersen 
Signed-off-by: Sasha Levin