linux-2.6-microblaze.git
3 months agodrm/panthor: Use a local iomem base for firmware control registers
Karunika Choo [Mon, 27 Apr 2026 15:59:33 +0000 (16:59 +0100)]
drm/panthor: Use a local iomem base for firmware control registers

Add an MCU_CONTROL-local iomem pointer to struct panthor_fw and use it
for firmware control and status register accesses.

Job interrupt accesses continue to go through the IRQ-local base, while
doorbell writes stay on the device-wide mapping because they live
outside the MCU control window. This keeps firmware register accesses
scoped to the component that owns them.

No functional change intended.

v3:
- Pick up R-bs from Liviu and Steve
v2:
- Pick up Ack from Boris.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-8-karunika.choo@arm.com
3 months agodrm/panthor: Use a local iomem base for PWR registers
Karunika Choo [Mon, 27 Apr 2026 15:59:32 +0000 (16:59 +0100)]
drm/panthor: Use a local iomem base for PWR registers

Add a PWR_CONTROL-local iomem pointer to struct panthor_pwr and switch
power controller register accesses to that base.

Update IRQ-local iomem base to use PWR_CONTROl-local iomem and update
the register definitions so the PWR block can be addressed relative to
its local base. This removes the remaining dependence on the global
device MMIO mapping for PWR register accesses. Update
panthor_gpu_info_init() to also use the correct PWR_CONTROL iomem for
the *_PRESENT registers.

No functional change intended.

v3:
- Clean up definitions for pwr->iomem and pwr->irq.iomem.
- Update PWR_INT_BASE to be relative to pwr->iomem.
v2:
- Update panthor_gpu_info_init() to use block-local iomem pointer.

Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-7-karunika.choo@arm.com
3 months agodrm/panthor: Use a local iomem base for GPU registers
Karunika Choo [Mon, 27 Apr 2026 15:59:31 +0000 (16:59 +0100)]
drm/panthor: Use a local iomem base for GPU registers

Add a GPU_CONTROL-local iomem pointer to struct panthor_gpu and use it
for GPU register accesses.

This limits GPU register accesses to the GPU block instead of using the
device-wide MMIO mapping directly. Interrupt register accesses continue
to use the IRQ-local base provided by the common IRQ helpers. Update
panthor_gpu_info_init() to also use a local iomem offset for GPU
features and capability.

This is a refactoring only and does not change behaviour.

v3:
- Pick up R-bs from Liviu and Steve
v2:
- Update panthor_gpu_info_init() to use block-local iomem pointer.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-6-karunika.choo@arm.com
3 months agodrm/panthor: Store IRQ register base iomem pointer in panthor_irq
Karunika Choo [Mon, 27 Apr 2026 15:59:30 +0000 (16:59 +0100)]
drm/panthor: Store IRQ register base iomem pointer in panthor_irq

Update common IRQ handling code to work from an IRQ-local iomem base
instead of referencing block-specific interrupt register offsets.

Store the interrupt base address iomem pointer in struct panthor_irq and
switch the shared IRQ helpers to use generic INT_* offsets from that
local base. This removes the need for each caller to expose absolute IRQ
register addresses while keeping the common IRQ flow unchanged.

No functional change intended.

v3:
- Clean up definition of pwr->irq.iomem.
v2:
- Change IRQ request function to accept an iomem pointer instead of
  computing it from an offset argument.

Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-5-karunika.choo@arm.com
3 months agodrm/panthor: Replace cross-component register accesses with helpers
Karunika Choo [Mon, 27 Apr 2026 15:59:29 +0000 (16:59 +0100)]
drm/panthor: Replace cross-component register accesses with helpers

Stop reaching into other components' registers directly and route those
operations through the component that owns them.

Move the timestamp/coherency helpers into panthor_gpu, add a doorbell
helper, and update call sites accordingly. This keeps register knowledge
local to each block and avoids spreading cross-component register
accesses across the driver.

This is a preparatory cleanup for using per-component iomem bases.

v3:
- Pick up Ack from Boris and R-bs from Liviu and Steve
v2:
- Fix incorrect spelling of timestamp helpers
- Fix unintended trailing backslash

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-4-karunika.choo@arm.com
3 months agodrm/panthor: Split register definitions by components
Karunika Choo [Mon, 27 Apr 2026 15:59:28 +0000 (16:59 +0100)]
drm/panthor: Split register definitions by components

Split the panthor register definitions into per-component headers for
the GPU, MMU, firmware, power and generic hardware registers.

This makes the register layout easier to follow and prepares the driver
for component-local iomem mappings by grouping definitions with the code
that owns them. The old monolithic panthor_regs.h header can then be
dropped.

No functional change intended.

v3:
- Pick up Ack from Boris and R-bs from Liviu and Steve
v2:
- Merge GPU_ID definitions into panthor_gpu_regs.h
- deleted panthor_hw_regs.h

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-3-karunika.choo@arm.com
3 months agodrm/panthor: Pass an iomem pointer to GPU register access helpers
Karunika Choo [Mon, 27 Apr 2026 15:59:27 +0000 (16:59 +0100)]
drm/panthor: Pass an iomem pointer to GPU register access helpers

Convert the Panthor register access helpers to take an iomem pointer
instead of a panthor_device pointer.

This makes the helpers usable with block-local registers instead of
routing all accesses to go through ptdev->iomem. It is a preparatory
change for splitting the register space by components and for moving
callers away from cross-component register accesses.

No functional change intended.

v3:
- Pick up R-bs from Liviu and Steve
v2:
- Pick up Ack from Boris.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Tested-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Liviu Dudau <liviu.dudau@arm.com>
Link: https://patch.msgid.link/20260427155934.416502-2-karunika.choo@arm.com
3 months agoMAINTAINERS: accel/ivpu: Remove myself and add Andrzej as maintainer
Maciej Falkowski [Wed, 29 Apr 2026 08:39:58 +0000 (10:39 +0200)]
MAINTAINERS: accel/ivpu: Remove myself and add Andrzej as maintainer

As I will be departing from my current position, I will
no longer maintain the driver, and remove myself from
ivpu's maintainers entry. Andrzej will support Karol
in maintaining the driver.

Signed-off-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Reviewed-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Reviewed-by: Andrzej Kacprowski <andrzej.kacprowski@linux.intel.com>
Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Link: https://patch.msgid.link/20260429083958.2148777-1-maciej.falkowski@linux.intel.com
3 months agodrm: renesas: shmobile: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:50 +0000 (13:55 +0200)]
drm: renesas: shmobile: remove now-redundant call to drm_connector_attach_encoder()

shmob_drm_connector_create() can init the connector in two ways, based on
the 'if (sdev->pdata)':

 1. manually in shmob_drm_connector_create(), or
 2. delegating to drm_bridge_connector_init()

Whichever branch is taken, drm_connector_attach_encoder() is called
immediately after to attach the connector to the encoder.

Now drm_bridge_connector_init() calls drm_connector_attach_encoder() on the
connector so it is not needed anymore in case 2 and should be removed, but
it is still needed in case 1. Move drm_connector_attach_encoder() from the
common path to inside shmob_drm_connector_create() in order to get back to
a single drm_connector_attach_encoder() in both cases.

Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-7-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: rgb: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:49 +0000 (13:55 +0200)]
drm/rockchip: rgb: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-6-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/imx: dc: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:48 +0000 (13:55 +0200)]
drm/imx: dc: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-5-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm: zynqmp_kms: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:47 +0000 (13:55 +0200)]
drm: zynqmp_kms: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-4-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/tilcdc: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:46 +0000 (13:55 +0200)]
drm/tilcdc: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-3-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/tidss: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:55:45 +0000 (13:55 +0200)]
drm/tidss: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115550.444930-2-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: lvds: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 11:53:28 +0000 (13:53 +0200)]
drm/rockchip: lvds: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423115334.444750-1-luca.ceresoli@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/mxsfb/lcdif: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:28 +0000 (11:17 +0200)]
drm/mxsfb/lcdif: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-34-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/msm/dsi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:27 +0000 (11:17 +0200)]
drm/msm/dsi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-33-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/msm/mdp4: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:26 +0000 (11:17 +0200)]
drm/msm/mdp4: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-32-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: inno-hdmi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:25 +0000 (11:17 +0200)]
drm/rockchip: inno-hdmi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-31-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: dw_hdmi_qp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:24 +0000 (11:17 +0200)]
drm/rockchip: dw_hdmi_qp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-30-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: dw_dp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:23 +0000 (11:17 +0200)]
drm/rockchip: dw_dp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-29-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: analogix_dp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:22 +0000 (11:17 +0200)]
drm/rockchip: analogix_dp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-28-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm: renesas: rz-du: rzg2l_du_encoder: remove now-redundant call to drm_connector_att...
Luca Ceresoli [Thu, 23 Apr 2026 09:17:21 +0000 (11:17 +0200)]
drm: renesas: rz-du: rzg2l_du_encoder: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Tested-by: Biju Das <biju.das.jz@bp.renesas.com> # RZ/G2L SMARC EVK
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-27-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm: rcar-du: encoder: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:20 +0000 (11:17 +0200)]
drm: rcar-du: encoder: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-26-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/exynos: exynos_dp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:19 +0000 (11:17 +0200)]
drm/exynos: exynos_dp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-25-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm: verisilicon: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:18 +0000 (11:17 +0200)]
drm: verisilicon: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Acked-by: Icenowy Zheng <zhengxingda@iscas.ac.cn>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-24-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/tests: bridge: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:17 +0000 (11:17 +0200)]
drm/tests: bridge: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-23-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/tegra: rgb: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:16 +0000 (11:17 +0200)]
drm/tegra: rgb: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-22-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/tegra: hdmi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:15 +0000 (11:17 +0200)]
drm/tegra: hdmi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-21-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: rk3066_hdmi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:14 +0000 (11:17 +0200)]
drm/rockchip: rk3066_hdmi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-20-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/rockchip: cdn-dp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:13 +0000 (11:17 +0200)]
drm/rockchip: cdn-dp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Liu Ying <victor.liu@nxp.com>
[Luca: fixed typo in commit subject line]
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-19-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/omapdrm: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:12 +0000 (11:17 +0200)]
drm/omapdrm: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-18-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/msm/hdmi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:11 +0000 (11:17 +0200)]
drm/msm/hdmi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-17-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/msm/dp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:10 +0000 (11:17 +0200)]
drm/msm/dp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-16-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/meson: encoder_hdmi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:09 +0000 (11:17 +0200)]
drm/meson: encoder_hdmi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-15-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/meson: encoder_cvbs: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:08 +0000 (11:17 +0200)]
drm/meson: encoder_cvbs: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-14-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/mediatek: mtk_dsi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:07 +0000 (11:17 +0200)]
drm/mediatek: mtk_dsi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-13-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/mediatek: mtk_dpi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:06 +0000 (11:17 +0200)]
drm/mediatek: mtk_dpi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-12-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/kmb/dsi: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:05 +0000 (11:17 +0200)]
drm/kmb/dsi: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-11-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/ingenic: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:04 +0000 (11:17 +0200)]
drm/ingenic: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Acked-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-10-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/imx/lcdc: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:03 +0000 (11:17 +0200)]
drm/imx/lcdc: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-9-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/imx: parallel-display: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:02 +0000 (11:17 +0200)]
drm/imx: parallel-display: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-8-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/imx: ldb: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:01 +0000 (11:17 +0200)]
drm/imx: ldb: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-7-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/imx/dcss: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:17:00 +0000 (11:17 +0200)]
drm/imx/dcss: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Laurentiu Palcu <laurentiu.palcu@oss.nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-6-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/bridge: ti-sn65dsi86: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:16:59 +0000 (11:16 +0200)]
drm/bridge: ti-sn65dsi86: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-5-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/bridge: ite-it6263: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:16:58 +0000 (11:16 +0200)]
drm/bridge: ite-it6263: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-4-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/bridge: adv7511: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:16:57 +0000 (11:16 +0200)]
drm/bridge: adv7511: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-3-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm: adp: remove now-redundant call to drm_connector_attach_encoder()
Luca Ceresoli [Thu, 23 Apr 2026 09:16:56 +0000 (11:16 +0200)]
drm: adp: remove now-redundant call to drm_connector_attach_encoder()

drm_connector_attach_encoder() is now called by
drm_bridge_connector_init().

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-2-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/display: bridge-connector: attach the encoder to the created connector
Luca Ceresoli [Thu, 23 Apr 2026 09:16:55 +0000 (11:16 +0200)]
drm/display: bridge-connector: attach the encoder to the created connector

Currently all users of the bridge-connector must call
drm_connector_attach_encoder() immediately after a successful
drm_bridge_connector_init().

This is an unnecessary burden for users. Move the call to the end of
drm_bridge_connector_init() so all callers can be simplified.

Update documentation accordingly, rewording a bit the whole paragraph which
would otherwise become poorly readable due to the growing list of actions.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://patch.msgid.link/20260423-drm-bridge-connector-attach_encoder-v2-1-2ae6ca69b390@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
3 months agodrm/nouveau: expose VBIOS via debugfs on GSP-RM systems
Timur Tabi [Tue, 28 Apr 2026 20:28:25 +0000 (15:28 -0500)]
drm/nouveau: expose VBIOS via debugfs on GSP-RM systems

When Nouveau boots with GSP-RM, it bypasses several traditional code
paths to allow GSP-RM to handle those features.  In particular,
some VBIOS parsing is skipped, and a side effect is that the VBIOS
is not exposed in the DRM debugfs entries.

Fix this by updating the drm BIOS struct (nvbios) with the VBIOS data
from the nvkm BIOS struct (nvkm_bios).  This happens normally in
NVInitVBIOS(), but that function is skipped when booting with GSP-RM.

Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Link: https://patch.msgid.link/20260428202825.1123719-1-ttabi@nvidia.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
3 months agodrm/bridge: prefer drm_printf_indent() over inline \t
Jani Nikula [Wed, 8 Apr 2026 08:22:11 +0000 (11:22 +0300)]
drm/bridge: prefer drm_printf_indent() over inline \t

We have a helper drm_printf_indent() for tab indenting the prints. It
makes the actual strings more readable, and highlights the indented
parts better in source.

Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Link: https://patch.msgid.link/20260408082211.3040194-2-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
3 months agodrm/atomic: prefer drm_printf_indent() over inline \t
Jani Nikula [Wed, 8 Apr 2026 08:22:10 +0000 (11:22 +0300)]
drm/atomic: prefer drm_printf_indent() over inline \t

We have a helper drm_printf_indent() for tab indenting the prints. It
makes the actual strings more readable, and highlights the indented
parts better in source.

Reviewed-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Link: https://patch.msgid.link/20260408082211.3040194-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
3 months agoaccel/ivpu: Add support for limiting NPU frequency
Andrzej Kacprowski [Wed, 8 Apr 2026 15:01:52 +0000 (17:01 +0200)]
accel/ivpu: Add support for limiting NPU frequency

Add configurable frequency limits to allow users to constrain the NPU
operating frequency range for power and thermal management. This support
requires firmware API version 3.34.0 or newer.

New sysfs interface:

The freq/ subdirectory contains the following attributes:

- hw_min_freq: Minimum frequency supported by hardware (read-only)
- hw_max_freq: Maximum frequency supported by hardware (read-only)
- hw_efficient_freq: Hardware's optimal operating frequency (read-only)
- current_freq: Current NPU frequency in MHz (read-only)
- set_min_freq: Configure minimum operating frequency (50XX+ devices)
- set_max_freq: Configure maximum operating frequency (50XX+ devices)

Legacy attributes npu_max_frequency_mhz and npu_current_frequency_mhz
are maintained for backward compatibility.

Implementation details:

- Frequency configuration is communicated to firmware via JSM messages
- User-specified frequency values are clamped to hardware limits
- Power-efficient frequency (pn_ratio) is adjusted dynamically to stay
  within the configured range
- Frequency configuration is initialized during device boot
- The JSM API header is updated to version 3.34.0 to support the new
  VPU_JSM_MSG_FREQ_CONFIG firmware message

Added description for the sysfs attributes in the Documentation/ABI.

Signed-off-by: Andrzej Kacprowski <andrzej.kacprowski@linux.intel.com>
Reviewed-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Signed-off-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Link: https://patch.msgid.link/20260408150152.2093638-1-andrzej.kacprowski@linux.intel.com
3 months agodrm/nouveau: WQ_PERCPU added to alloc_workqueue users
Marco Crivellari [Fri, 31 Oct 2025 10:20:20 +0000 (11:20 +0100)]
drm/nouveau: WQ_PERCPU added to alloc_workqueue users

Currently if a user enqueue a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.
This lack of consistentcy cannot be addressed without refactoring the API.

alloc_workqueue() treats all queues as per-CPU by default, while unbound
workqueues must opt-in via WQ_UNBOUND.

This default is suboptimal: most workloads benefit from unbound queues,
allowing the scheduler to place worker threads where they’re needed and
reducing noise when CPUs are isolated.

This change adds a new WQ_PERCPU flag to explicitly request
alloc_workqueue() to be per-cpu when WQ_UNBOUND has not been specified.

With the introduction of the WQ_PERCPU flag (equivalent to !WQ_UNBOUND),
any alloc_workqueue() caller that doesn’t explicitly specify WQ_UNBOUND
must now use WQ_PERCPU.

Once migration is complete, WQ_UNBOUND can be removed and unbound will
become the implicit default.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Link: https://patch.msgid.link/20251031102020.95349-3-marco.crivellari@suse.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
3 months agodrm/nouveau: replace use of system_unbound_wq with system_dfl_wq
Marco Crivellari [Fri, 31 Oct 2025 10:20:19 +0000 (11:20 +0100)]
drm/nouveau: replace use of system_unbound_wq with system_dfl_wq

Currently if a user enqueue a work item using schedule_delayed_work() the
used wq is "system_wq" (per-cpu wq) while queue_delayed_work() use
WORK_CPU_UNBOUND (used when a cpu is not specified). The same applies to
schedule_work() that is using system_wq and queue_work(), that makes use
again of WORK_CPU_UNBOUND.

This lack of consistency cannot be addressed without refactoring the API.

system_unbound_wq should be the default workqueue so as not to enforce
locality constraints for random work whenever it's not required.

Adding system_dfl_wq to encourage its use when unbound work should be used.

The old system_unbound_wq will be kept for a few release cycles.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Marco Crivellari <marco.crivellari@suse.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patch.msgid.link/20251031102020.95349-2-marco.crivellari@suse.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
3 months agoaccel/amdxdna: Add configuring low and medium power mode
Nishad Saraf [Fri, 24 Apr 2026 04:08:23 +0000 (21:08 -0700)]
accel/amdxdna: Add configuring low and medium power mode

Add support for POWER_MODE_LOW and POWER_MODE_MEDIUM.

Signed-off-by: Nishad Saraf <nishads@amd.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260424040824.2253607-2-lizhi.hou@amd.com
3 months agoaccel/amdxdna: Set the system efficiency factor to 2
Nishad Saraf [Fri, 24 Apr 2026 04:08:24 +0000 (21:08 -0700)]
accel/amdxdna: Set the system efficiency factor to 2

The system efficiency factor is used for QoS calculation. Change it to 2
to account for the efficiency overhead.

Signed-off-by: Nishad Saraf <nishads@amd.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260424040824.2253607-3-lizhi.hou@amd.com
3 months agoaccel/amdxdna: Set default DPM level based on QoS for temporal-only mode
Lizhi Hou [Fri, 24 Apr 2026 04:08:22 +0000 (21:08 -0700)]
accel/amdxdna: Set default DPM level based on QoS for temporal-only mode

The QoS request provided when creating a hardware context is currently
ignored when operating in temporal-only mode. Change this to use resource
allocation through xrs_allocate_resource(), which sets the default DPM
level according to the QoS request.

When multiple hardware contexts are active, track their required DPM
levels and set the default DPM level to the highest among them.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20260424040824.2253607-1-lizhi.hou@amd.com
3 months agodrm/imagination: Restrict init_fw_trace_mask module param to read only
Brajesh Gupta [Mon, 27 Apr 2026 05:31:38 +0000 (11:01 +0530)]
drm/imagination: Restrict init_fw_trace_mask module param to read only

Param used for setting FW trace mask at module load time. Other debugfs
entry exist to allow update at run time.

Signed-off-by: Brajesh Gupta <brajesh.gupta@imgtec.com>
Reviewed-by: Alessio Belle <alessio.belle@imgtec.com>
Link: https://patch.msgid.link/20260427-ftrace_fix-v3-2-e081530759a8@imgtec.com
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
3 months agoMerge drm/drm-next into drm-misc-next
Thomas Zimmermann [Mon, 27 Apr 2026 08:49:13 +0000 (10:49 +0200)]
Merge drm/drm-next into drm-misc-next

Getting fixes and updates from v7.1-rc1.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
3 months agoLinux 7.1-rc1
Linus Torvalds [Sun, 26 Apr 2026 21:19:00 +0000 (14:19 -0700)]
Linux 7.1-rc1

3 months agoMerge tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux
Linus Torvalds [Sun, 26 Apr 2026 21:03:20 +0000 (14:03 -0700)]
Merge tag 'clk-for-linus' of git://git./linux/kernel/git/clk/linux

Pull clk fix from Stephen Boyd:
 "One more fix for the merge window to avoid a boot hang on
  Raspberry Pi 3B by marking the VEC clk critical so that it
  doesn't get turned off and hang the bus"

* tag 'clk-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux:
  clk: bcm: rpi: Mark VEC clock as CLK_IGNORE_UNUSED

3 months agoMerge tag 'tsm-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/devsec/tsm
Linus Torvalds [Sun, 26 Apr 2026 16:51:29 +0000 (09:51 -0700)]
Merge tag 'tsm-for-7.1' of git://git./linux/kernel/git/devsec/tsm

Pull PCIe TSP update from Dan Williams:
 "A small update for the TSM core. It is arguably a fix and coming in
  late as I have been offline the past few weeks:

   - Drop class_create() for the 'tsm' class"

* tag 'tsm-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/devsec/tsm:
  virt: coco: change tsm_class to a const struct

3 months agodrm/hyperv: use VMBUS_RING_SIZE()
Hamza Mahfooz [Sat, 25 Apr 2026 18:17:19 +0000 (11:17 -0700)]
drm/hyperv: use VMBUS_RING_SIZE()

VMBUS ring buffers must be page aligned. So, use VMBUS_RING_SIZE() to
ensure they are always aligned and large enough to hold all of the
relevant data.

Cc: stable@kernel.vger.org
Fixes: 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic video device")
Signed-off-by: Hamza Mahfooz <hamzamahfooz@linux.microsoft.com>
Reviewed-by: Saurabh Sengar <ssengar@linux.microsoft.com>
Link: https://patch.msgid.link/20260425181719.1538483-2-hamzamahfooz@linux.microsoft.com
3 months agodrm/edid: add CTA Video Format Data Block support
Hamza Mahfooz [Wed, 25 Feb 2026 17:57:08 +0000 (12:57 -0500)]
drm/edid: add CTA Video Format Data Block support

Video Format Data Blocks (VFDBs) contain the necessary information that
needs to be fed to the Optimized Video Timings (OVT) Algorithm.
Also, we require OVT support to cover modes that aren't supported by
earlier standards (e.g. CVT). So, parse all of the relevant VFDB data
and feed it to the OVT Algorithm, to extract all of the missing OVT
modes.

Suggested-by: Karol Herbst <kherbst@redhat.com>
Signed-off-by: Hamza Mahfooz <someguy@effective-light.com>
Link: https://patch.msgid.link/20260225175709.408010-1-someguy@effective-light.com
3 months agoMerge tag 'kbuild-fixes-7.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuil...
Linus Torvalds [Sun, 26 Apr 2026 00:04:15 +0000 (17:04 -0700)]
Merge tag 'kbuild-fixes-7.1-1' of git://git./linux/kernel/git/kbuild/linux

Pull Kbuild fixes from Nicolas Schier:

 - builddeb - avoid recompiles for non-cross-compiles

   Avoid triggering complete rebuilds for non-cross-compile Debian
   package builds by only triggering the rebuild of host tools for
   actual cross-compile builds

 - Never respect CONFIG_WERROR / W=e to fixdep

   Avoid spurious rebuilds of fixdep w/ and w/o -Werror during a single
   kbuild invocation by never respecting CONFIG_WERROR for fixdep

* tag 'kbuild-fixes-7.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux:
  kbuild: Never respect CONFIG_WERROR / W=e to fixdep
  kbuild: builddeb - avoid recompiles for non-cross-compiles

3 months agoMerge tag 'power-utilities-2026.04.25' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 25 Apr 2026 23:58:34 +0000 (16:58 -0700)]
Merge tag 'power-utilities-2026.04.25' of git://git./linux/kernel/git/lenb/linux

Pull power utility updates from Len Brown:
 "x86_energy_perf_policy:
   - Initial SoC Slider support

 turbostat:
   - Display HT siblings in cpu# order
   - Add Module-ID column
   - Print Core-ID and APIC-ID in hex
   - Fix misc bugs"

* tag 'power-utilities-2026.04.25' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux:
  tools/power x86_energy_perf_policy: Version 2026.04.25
  tools/power x86_energy_perf_policy.8: Document SoC Slider Options
  tools/power x86_energy_perf_policy: Enhances SoC Slider related checks
  tools/power turbostat: v2026.04.21
  tools/power turbostat: Process HT siblings in CPU order
  tools/power turbostat: Show module_id column
  tools/power turbostat: Print core_id and apic_id in hex
  tools/power turbostat: Cleanup print helper functions
  tools/power turbostat: Fix --cpu-set 1 regression on HT systems
  tools/power turbostat: Fix --cpu-set 0 regression on HT systems
  tools/power turbostat: Fix unrecognized option '-P'
  tools/power turbostat: Fix AMD RAPL regression on big systems
  tools/power/x86: Add SOC slider and platform profile support

3 months agoMerge tag 'rtc-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux
Linus Torvalds [Sat, 25 Apr 2026 23:39:03 +0000 (16:39 -0700)]
Merge tag 'rtc-7.1' of git://git./linux/kernel/git/abelloni/linux

Pull RTC updates from Alexandre Belloni:
 "Subsystem:
   - add data_race() in rtc_dev_poll()

  Drivers:
   - remove i2c_match_id usage
   - abx80x: Disable alarm feature if no interrupt attached
   - ti-k3: support resuming from IO DDR low power mode"

* tag 'rtc-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/abelloni/linux:
  rtc: abx80x: Disable alarm feature if no interrupt attached
  rtc: ntxec: fix OF node reference imbalance
  rtc: pic32: allow driver to be compiled with COMPILE_TEST
  rtc: ti-k3: Add support to resume from IO DDR low power mode
  rtc: cmos: Use platform_get_irq_optional() in cmos_platform_probe()
  dt-bindings: rtc: add olpc,xo1-rtc to trivial-rtc
  dt-bindings: rtc: sc2731: Add compatible for SC2730
  rtc: add data_race() in rtc_dev_poll()
  rtc: armada38x: zalloc + calloc to single allocation
  dt-bindings: rtc: isl12026: convert to YAML schema
  dt-bindings: rtc: microcrystal,rv3028: Allow to specify vdd-supply
  rtc: max77686: convert to i2c_new_ancillary_device
  dt-bindings: rtc: mpfs-rtc: permit resets
  rtc: rx8025: Remove use of i2c_match_id()
  rtc: rv8803: Remove use of i2c_match_id()
  rtc: rs5c372: Remove use of i2c_match_id()
  rtc: pcf2127: Remove use of i2c_match_id()
  rtc: m41t80: Remove use of i2c_match_id()
  rtc: abx80x: Remove use of i2c_match_id()

3 months agoMerge tag 'for-next-tpm-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 25 Apr 2026 23:20:52 +0000 (16:20 -0700)]
Merge tag 'for-next-tpm-7.1-rc1' of git://git./linux/kernel/git/jarkko/linux-tpmdd

Pull tpm updates from Jarkko Sakkinen:
 "Here are the accumulated fixes for 7.1-rc1 and a single structural
  change worth mentioning separately: Rafael's commit converting tpm_crb
  from ACPI driver to a platform driver"

* tag 'for-next-tpm-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
  tpm: tpm_tis: stop transmit if retries are exhausted
  tpm: tpm_tis: add error logging for data transfer
  tpm: avoid -Wunused-but-set-variable
  tpm: Use kfree_sensitive() to free auth session in tpm_dev_release()
  tpm2-sessions: Fix missing tpm_buf_destroy() in tpm2_read_public()
  tpm: Fix auth session leak in tpm2_get_random() error path
  tpm: i2c: atmel: fix block comment formatting
  tpm_crb: Convert ACPI driver to a platform one
  tpm: Make tcpci_pm_ops variable static const

3 months agoMerge branches 'turbostat' and 'x86_energy_perf_policy' into power-utilities
Len Brown [Sat, 25 Apr 2026 18:26:32 +0000 (14:26 -0400)]
Merge branches 'turbostat' and 'x86_energy_perf_policy' into power-utilities

3 months agotools/power x86_energy_perf_policy: Version 2026.04.25
Len Brown [Sat, 25 Apr 2026 17:26:16 +0000 (13:26 -0400)]
tools/power x86_energy_perf_policy: Version 2026.04.25

Since v2025.11.22:
Initial SoC Slider support
SoC Slider is an SoC-wide power/performance policy setting.
On SoC Slider systems, EPP plays a diminished role.

Whitespace cleanup via: indent -npro -kr -i8 -ts8 -sob -l160 -ss -ncs -cp1

No functional changes

Signed-off-by: Len Brown <len.brown@intel.com>
3 months agotools/power x86_energy_perf_policy.8: Document SoC Slider Options
Len Brown [Sat, 25 Apr 2026 16:10:54 +0000 (12:10 -0400)]
tools/power x86_energy_perf_policy.8: Document SoC Slider Options

x86_energy_perf_policy accesses the SoC Slider via standard
user/kernel APIs to the processor_thermal_soc_slider driver.

Machines that support SoC Slider largely use it instead of EPP,
which may continue to exist in a diminished role, or vanish entirely.

Signed-off-by: Len Brown <len.brown@intel.com>
3 months agotools/power x86_energy_perf_policy: Enhances SoC Slider related checks
Len Brown [Wed, 15 Apr 2026 19:12:29 +0000 (15:12 -0400)]
tools/power x86_energy_perf_policy: Enhances SoC Slider related checks

When processor_thermal_soc_slider is loaded, its slider
and offset modparams are visible.  Check that the driver
actually registered the profile named "SoC Slider" before
reading or writing these modparams.

n.b. This utility allows writing the Slider and Offset modparams
even if the driver policy is not "balanced".  Currently the
processor_thermal_soc_slider consults those modparams
only in "balanced" mode.

Signed-off-by: Len Brown <len.brown@intel.com>
3 months agoclk: bcm: rpi: Mark VEC clock as CLK_IGNORE_UNUSED
Maíra Canal [Fri, 24 Apr 2026 15:34:52 +0000 (16:34 +0100)]
clk: bcm: rpi: Mark VEC clock as CLK_IGNORE_UNUSED

On Raspberry Pi 3B, the VEC clock is used by the VideoCore firmware
display driver, which remains active until the vc4 driver loads and
sends NOTIFY_DISPLAY_DONE. If this clock is disabled during boot, a bus
lockup happens and the firmware becomes unresponsive, causing a complete
system lockup.

Mark the VEC clock with CLK_IGNORE_UNUSED so it survives the unused
clock disablement and remains available until the vc4 driver takes over
display management.

Fixes: 672299736af6 ("clk: bcm: rpi: Manage clock rate in prepare/unprepare callbacks")
Reported-by: Mark Brown <broonie@kernel.org>
Closes: https://lore.kernel.org/r/5f0bec08-f458-4fba-8bf3-06817a100c4c@sirena.org.uk
Signed-off-by: Maíra Canal <mcanal@igalia.com>
Link: https://patch.msgid.link/20260401111416.562279-2-mcanal@igalia.com
Tested-by: Mark Brown <broonie@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
Acked-by: Brian Masney <bmasney@redhat.com> # Active contributor to clk
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
3 months agoMerge tag 'fbdev-for-7.1-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 25 Apr 2026 14:48:33 +0000 (07:48 -0700)]
Merge tag 'fbdev-for-7.1-rc1-2' of git://git./linux/kernel/git/deller/linux-fbdev

Pull fbdev fixes from Helge Deller:

 - request memory region before use (cobalt_lcdfb, clps711x-fb, hgafb)

 - reference cleanups in failure path (offb, savage)

 - a spelling fix (atyfb)

* tag 'fbdev-for-7.1-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/linux-fbdev:
  fbdev: hgafb: Request memory region before ioremap
  fbdev: clps711x-fb: Request memory region for MMIO
  fbdev: cobalt_lcdfb: Request memory region
  fbdev: atyfb: Fix spelling mistake "enfore" -> "enforce"
  fbdev: savage: fix probe-path EDID cleanup leaks
  fbdev: offb: fix PCI device reference leak on probe failure

3 months agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux
Linus Torvalds [Sat, 25 Apr 2026 14:44:26 +0000 (07:44 -0700)]
Merge tag 'for-linus' of git://git./linux/kernel/git/rmk/linux

Pull ARM updates from Russell King:

 - fix a race condition handling PG_dcache_clean

 - further cleanups for the fault handling, allowing RT to be enabled

 - fixing nzones validation in adfs filesystem driver

 - fix for module unwinding

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rmk/linux:
  ARM: 9463/1: Allow to enable RT
  ARM: 9472/1: fix race condition on PG_dcache_clean in __sync_icache_dcache()
  ARM: 9471/1: module: fix unwind section relocation out of range error
  fs/adfs: validate nzones in adfs_validate_bblk()
  ARM: provide individual is_translation_fault() and is_permission_fault()
  ARM: move FSR fault status definitions before fsr_fs()
  ARM: use BIT() and GENMASK() for fault status register fields
  ARM: move is_permission_fault() and is_translation_fault() to fault.h
  ARM: move vmalloc() lazy-page table population
  ARM: ensure interrupts are enabled in __do_user_fault()

3 months agoMerge tag 'trace-ring-buffer-v7.1-3' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 22:17:23 +0000 (15:17 -0700)]
Merge tag 'trace-ring-buffer-v7.1-3' of git://git./linux/kernel/git/trace/linux-trace

Pull ring-buffer fix from Steven Rostedt:

 - Fix accounting of persistent ring buffer rewind

   On boot up, the head page is moved back to the earliest point of the
   saved ring buffer. This is because the ring buffer being read by user
   space on a crash may not save the part it read. Rewinding the head
   page back to the earliest saved position helps keep those events from
   being lost.

   The number of events is also read during boot up and displayed in the
   stats file in the tracefs directory. It's also used for other
   accounting as well. On boot up, the "reader page" is accounted for
   but a rewind may put it back into the buffer and then the reader page
   may be accounted for again.

   Save off the original reader page and skip accounting it when
   scanning the pages in the ring buffer.

* tag 'trace-ring-buffer-v7.1-3' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Do not double count the reader_page

3 months agoMerge tag 'block-7.1-20260424' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe...
Linus Torvalds [Fri, 24 Apr 2026 22:06:55 +0000 (15:06 -0700)]
Merge tag 'block-7.1-20260424' of git://git./linux/kernel/git/axboe/linux

Pull block fixes from Jens Axboe:

 - Series for zloop, fixing a variety of issues

 - t10-pi code cleanup

 - Fix for a merge window regression with the bio memory allocation mask

 - Fix for a merge window regression in ublk, caused by an issue with
   the maple tree iteration code at teardown

 - ublk self tests additions

 - Zoned device pgmap fixes

 - Various little cleanups and fixes

* tag 'block-7.1-20260424' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux: (21 commits)
  Revert "floppy: fix reference leak on platform_device_register() failure"
  ublk: avoid unpinning pages under maple tree spinlock
  ublk: refactor common helper ublk_shmem_remove_ranges()
  ublk: fix maple tree lockdep warning in ublk_buf_cleanup
  selftests: ublk: add ublk auto integrity test
  selftests: ublk: enable test_integrity_02.sh on fio 3.42
  selftests: ublk: remove unused argument to _cleanup
  block: only restrict bio allocation gfp mask asked to block
  block/blk-throttle: Add WQ_PERCPU to alloc_workqueue users
  block: Add WQ_PERCPU to alloc_workqueue users
  block: relax pgmap check in bio_add_page for compatible zone device pages
  block: add pgmap check to biovec_phys_mergeable
  floppy: fix reference leak on platform_device_register() failure
  ublk: use unchecked copy helpers for bio page data
  t10-pi: reduce ref tag code duplication
  zloop: remove irq-safe locking
  zloop: factor out zloop_mark_{full,empty} helpers
  zloop: set RQF_QUIET when completing requests on deleted devices
  zloop: improve the unaligned write pointer warning
  zloop: use vfs_truncate
  ...

3 months agoMerge tag 'io_uring-7.1-20260424' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 24 Apr 2026 22:00:54 +0000 (15:00 -0700)]
Merge tag 'io_uring-7.1-20260424' of git://git./linux/kernel/git/axboe/linux

Pull io_uring fixes from Jens Axboe:

 - Fix for a NOMMU bug with io_uring, where NOMMU doesn't grab page refs
   at mmap time. NOMMU also has entirely broken FOLL_PIN support, yet
   here we are

 - A few fixes covering minor issues introduced in this merge window

 - data race annotation to shut up KCSAN for when io-wq limits are
   applied

 - A nospec addition for direct descriptor file updating. Rest of the
   direct descriptor path already had this, but for some reason the
   update did not. Now they are all the same

 - Various minor defensive changes that claude identified and suggested
   terrible fixes for, turned into actually useful cleanups:

       - Use kvfree() for the imu cache. These can come from kmalloc or
         vmalloc depending on size, but the in-cache ones are capped
         where it's always kmalloc based. Change to kvfree() in the
         cleanup path, making future changes unlikely to mess that up

       - Negative kbuf consumption lengths. Can't happen right now, but
         cqe->res is used directly, which if other codes changes could
         then be an error value

 - Fix for an issue with the futex code, where partial wakes on a
   vectored fuxes would potentially wake the same futex twice, rather
   than move on to the next one. This could confuse an application as it
   would've expected the next futex to have been woken

 - Fix for a bug with ring resizing, where SQEs or CQEs might not have
   been copied correctly if large SQEs or CQEs are used in the ring.
   Application side issue, where SQEs or CQEs might have been lost
   during resize

 - Fix for a bug where EPOLL_URING_WAKE might have been lost, causing a
   multishot poll to not be terminated when it's nested, like it should
   have been

 - Fix for an issue with signed comparison of poll references for the
   slow path

 - Fix for a user struct UAF in the zcrx code

 - Two minor zcrx cleanups

* tag 'io_uring-7.1-20260424' of git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux:
  io_uring: take page references for NOMMU pbuf_ring mmaps
  io_uring/poll: ensure EPOLL_ONESHOT is propagated for EPOLL_URING_WAKE
  io_uring/zcrx: warn on freelist violations
  io_uring/zcrx: clear RQ headers on init
  io_uring/zcrx: fix user_struct uaf
  io_uring/register: fix ring resizing with mixed/large SQEs/CQEs
  io_uring/futex: ensure partial wakes are appropriately dequeued
  io_uring/rw: add defensive hardening for negative kbuf lengths
  io_uring/rsrc: use kvfree() for the imu cache
  io_uring/rsrc: unify nospec indexing for direct descriptors
  io_uring: fix spurious fput in registered ring path
  io_uring: fix iowq_limits data race in tctx node addition
  io_uring/tctx: mark io_wq as exiting before error path teardown
  io_uring/tctx: check for setup tctx->io_wq before teardown
  io_uring/poll: fix signed comparison in io_poll_get_ownership()

3 months agoMerge tag 'nfs-for-7.1-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Linus Torvalds [Fri, 24 Apr 2026 21:20:03 +0000 (14:20 -0700)]
Merge tag 'nfs-for-7.1-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs

Pull NFS client updates from Trond Myklebust:
 "Bugfixes:

   - Fix handling of ENOSPC so that if we have to resend writes, they
     are written synchronously

   - SUNRPC RDMA transport fixes from Chuck

   - Several fixes for delegated timestamps in NFSv4.2

   - Failure to obtain a directory delegation should not cause stat() to
     fail with NFSv4

   - Rename was failing to update timestamps when a directory delegation
     is held on NFSv4

   - Ensure we check rsize/wsize after crossing a NFSv4 filesystem
     boundary

   - NFSv4/pnfs:

      - If the server is down, retry the layout returns on reboot

      - Fallback to MDS could result in a short write being incorrectly
        logged

  Cleanups:

   - Use memcpy_and_pad in decode_fh"

* tag 'nfs-for-7.1-1' of git://git.linux-nfs.org/projects/trondmy/linux-nfs: (21 commits)
  NFS: Fix RCU dereference of cl_xprt in nfs_compare_super_address
  NFS: remove redundant __private attribute from nfs_page_class
  NFSv4.2: fix CLONE/COPY attrs in presence of delegated attributes
  NFS: fix writeback in presence of errors
  nfs: use memcpy_and_pad in decode_fh
  NFSv4.1: Apply session size limits on clone path
  NFSv4: retry GETATTR if GET_DIR_DELEGATION failed
  NFS: fix RENAME attr in presence of directory delegations
  pnfs/flexfiles: validate ds_versions_cnt is non-zero
  NFS/blocklayout: print each device used for SCSI layouts
  xprtrdma: Post receive buffers after RPC completion
  xprtrdma: Scale receive batch size with credit window
  xprtrdma: Replace rpcrdma_mr_seg with xdr_buf cursor
  xprtrdma: Decouple frwr_wp_create from frwr_map
  xprtrdma: Close lost-wakeup race in xprt_rdma_alloc_slot
  xprtrdma: Avoid 250 ms delay on backlog wakeup
  xprtrdma: Close sendctx get/put race that can block a transport
  nfs: update inode ctime after removexattr operation
  nfs: fix utimensat() for atime with delegated timestamps
  NFS: improve "Server wrote zero bytes" error
  ...

3 months agoMerge tag 'ceph-for-7.1-rc1' of https://github.com/ceph/ceph-client
Linus Torvalds [Fri, 24 Apr 2026 20:47:19 +0000 (13:47 -0700)]
Merge tag 'ceph-for-7.1-rc1' of https://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:
 "We have a series from Alex which extends CephFS client metrics with
  support for per-subvolume data I/O performance and latency tracking
  (metadata operations aren't included) and a good variety of fixes and
  cleanups across RBD and CephFS"

* tag 'ceph-for-7.1-rc1' of https://github.com/ceph/ceph-client:
  ceph: add subvolume metrics collection and reporting
  ceph: parse subvolume_id from InodeStat v9 and store in inode
  ceph: handle InodeStat v8 versioned field in reply parsing
  libceph: Fix slab-out-of-bounds access in auth message processing
  rbd: fix null-ptr-deref when device_add_disk() fails
  crush: cleanup in crush_do_rule() method
  ceph: clear s_cap_reconnect when ceph_pagelist_encode_32() fails
  ceph: only d_add() negative dentries when they are unhashed
  libceph: update outdated comment in ceph_sock_write_space()
  libceph: Remove obsolete session key alignment logic
  ceph: fix num_ops off-by-one when crypto allocation fails
  libceph: Prevent potential null-ptr-deref in ceph_handle_auth_reply()

3 months agoMerge tag 'ntfs-for-7.1-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 24 Apr 2026 20:40:25 +0000 (13:40 -0700)]
Merge tag 'ntfs-for-7.1-rc1-part2' of git://git./linux/kernel/git/linkinjeon/ntfs

Pull ntfs updates from Namjae Jeon:

 - Fix potential data leakage by zeroing the portion of the straddle
   block beyond initialized_size when reading non-resident attributes

 - Remove unnecessary zeroing in ntfs_punch_hole() for ranges beyond
   initialized_size, as they are already returned as zeros on read

 - Fix writable check in ntfs_file_mmap_prepare() to correctly handle
   shared mappings using VMA_SHARED_BIT | VMA_MAYWRITE_BIT

 - Use page allocation instead of kmemdup() for IOMAP_INLINE data to
   ensure page-aligned address and avoid BUG trap in
   iomap_inline_data_valid() caused by the page boundary check

 - Add a size check before memory allocation in ntfs_attr_readall() and
   reject overly large attributes

 - Remove unneeded noop_direct_IO from ntfs_aops as it is no longer
   required following the FMODE_CAN_ODIRECT flag

 - Fix seven static analysis warnings reported by Smatch

* tag 'ntfs-for-7.1-rc1-part2' of git://git.kernel.org/pub/scm/linux/kernel/git/linkinjeon/ntfs:
  ntfs: use page allocation for resident attribute inline data
  ntfs: fix mmap_prepare writable check for shared mappings
  ntfs: fix potential 32-bit truncation in ntfs_write_cb()
  ntfs: fix uninitialized variable in ntfs_map_runlist_nolock
  ntfs: delete dead code
  ntfs: add missing error code in ntfs_mft_record_alloc()
  ntfs: fix uninitialized variables in ntfs_ea_set_wsl_inode()
  ntfs: fix uninitialized pointer in ntfs_write_mft_block
  ntfs: fix uninitialized variable in ntfs_write_simple_iomap_begin_non_resident
  ntfs: remove noop_direct_IO from address_space_operations
  ntfs: limit memory allocation in ntfs_attr_readall
  ntfs: not zero out range beyond init in punch_hole
  ntfs: zero out stale data in straddle block beyond initialized_size

3 months agoMerge tag '9p-for-7.1-rc1' of https://github.com/martinetd/linux
Linus Torvalds [Fri, 24 Apr 2026 20:37:26 +0000 (13:37 -0700)]
Merge tag '9p-for-7.1-rc1' of https://github.com/martinetd/linux

Pull 9p updates from Dominique Martinet:

 - 9p access flag fix (cannot change access flag since new mount API implem)

 - some minor cleanup

* tag '9p-for-7.1-rc1' of https://github.com/martinetd/linux:
  9p/trans_xen: replace simple_strto* with kstrtouint
  9p/trans_xen: make cleanup idempotent after dataring alloc errors
  9p: document missing enum values in kernel-doc comments
  9p: fix access mode flags being ORed instead of replaced
  9p: fix memory leak in v9fs_init_fs_context error path

3 months agoMerge tag 'spdx-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 24 Apr 2026 20:30:54 +0000 (13:30 -0700)]
Merge tag 'spdx-7.1-rc1' of git://git./linux/kernel/git/gregkh/spdx

Pull SPDX update from Greg KH:
 "Here is a single SPDX-like change for 7.1-rc1. It explicitly allows
  the use of SPDX-FileCopyrightText which has been used already in many
  files.

  At the same time, update checkpatch to catch any "non allowed" spdx
  identifiers as we don't want to go overboard here.

  This has been in linux-next for a long time with no reported problems"

* tag 'spdx-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/spdx:
  LICENSES: Explicitly allow SPDX-FileCopyrightText

3 months agoMerge tag 'char-misc-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 24 Apr 2026 20:23:50 +0000 (13:23 -0700)]
Merge tag 'char-misc-7.1-rc1' of git://git./linux/kernel/git/gregkh/char-misc

Pull char / misc / IIO / and others driver updates from Greg KH:
 "Here is the char/misc/iio and other smaller driver subsystem updates
  for 7.1-rc1. Lots of stuff in here, all tiny, but relevant for the
  different drivers they touch. Major points in here is:

   - the usual large set of new IIO drivers and updates for that
     subsystem (the large majority of this diffstat)

   - lots of comedi driver updates and bugfixes

   - coresight driver updates

   - interconnect driver updates and additions

   - mei driver updates

   - binder (both rust and C versions) updates and fixes

   - lots of other smaller driver subsystem updates and additions

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'char-misc-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc: (405 commits)
  coresight: tpdm: fix invalid MMIO access issue
  mei: me: add nova lake point H DID
  mei: lb: add late binding version 2
  mei: bus: add mei_cldev_uuid
  w1: ds2490: drop redundant device reference
  bus: mhi: host: pci_generic: Add Telit FE912C04 modem support
  mei: csc: wake device while reading firmware status
  mei: csc: support controller with separate PCI device
  mei: convert PCI error to common errno
  mei: trace: print return value of pci_cfg_read
  mei: me: move trace into firmware status read
  mei: fix idle print specifiers
  mei: me: use PCI_DEVICE_DATA macro
  sonypi: Convert ACPI driver to a platform one
  misc: apds990x: fix all kernel-doc warnings
  most: usb: Use kzalloc_objs for endpoint address array
  hpet: Convert ACPI driver to a platform one
  misc: vmw_vmci: Fix spelling mistakes in comments
  parport: Remove completed item from to-do list
  char: remove unnecessary module_init/exit functions
  ...

3 months agoMerge tag 'spi-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 20:16:36 +0000 (13:16 -0700)]
Merge tag 'spi-fix-v7.1-merge-window' of git://git./linux/kernel/git/broonie/spi

Pull spi fixes from Mark Brown:
 "This is quite a big set of fixes, almost all from Johan Hovold who is
  on an ongoing quest to clean up issues with probe and removal handling
  in drivers.

  There isn't anything too concerning here especially with the
  deregistration stuff which will very rarely get run in production
  systems since this is all platform devices in the SoC on embedded
  hardware, but it's all real issues which should be fixed. There's more
  in flight here.

  We also have a few other minor fixes, one from Felix Gu along the same
  lines as Johan's work and a couple of documentation things"

* tag 'spi-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi: (23 commits)
  spi: fix controller cleanup() documentation
  spi: fix resource leaks on device setup failure
  spi: axiado: clean up probe return value
  spi: axiado: rename probe error labels
  spi: axiado: fix runtime pm imbalance on probe failure
  spi: orion: clean up probe return value
  spi: orion: fix clock imbalance on registration failure
  spi: orion: fix runtime pm leak on unbind
  spi: imx: fix runtime pm leak on probe deferral
  spi: mpc52xx: fix use-after-free on registration failure
  spi: Fix the error description in the `ptp_sts_word_post` comment
  spi: topcliff-pch: fix use-after-free on unbind
  spi: topcliff-pch: fix controller deregistration
  spi: orion: fix controller deregistration
  spi: mxic: fix controller deregistration
  spi: mpc52xx: fix use-after-free on unbind
  spi: mpc52xx: fix controller deregistration
  spi: cadence-quadspi: fix controller deregistration
  spi: cadence: fix controller deregistration
  spi: mtk-snfi: fix memory leak in probe
  ...

3 months agoMerge tag 'regulator-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux...
Linus Torvalds [Fri, 24 Apr 2026 20:06:25 +0000 (13:06 -0700)]
Merge tag 'regulator-fix-v7.1-merge-window' of git://git./linux/kernel/git/broonie/regulator

Pull regulator fix from Mark Brown:
 "Just one trivial cleanup of the user visible prompts in Kconfig here,
  standardising how we describe Qualcomm"

* tag 'regulator-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
  regulator: qcom: Unify user-visible "Qualcomm" name

3 months agoring-buffer: Do not double count the reader_page
Masami Hiramatsu (Google) [Fri, 24 Apr 2026 06:52:10 +0000 (15:52 +0900)]
ring-buffer: Do not double count the reader_page

Since the cpu_buffer->reader_page is updated if there are unwound
pages. After that update, we should skip the page if it is the
original reader_page, because the original reader_page is already
checked.

Cc: stable@vger.kernel.org
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Ian Rogers <irogers@google.com>
Link: https://patch.msgid.link/177701353063.2223789.1471163147644103306.stgit@mhiramat.tok.corp.google.com
Fixes: ca296d32ece3 ("tracing: ring_buffer: Rewind persistent ring buffer on reboot")
Signed-off-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
3 months agoMerge tag 'regmap-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 19:11:26 +0000 (12:11 -0700)]
Merge tag 'regmap-fix-v7.1-merge-window' of git://git./linux/kernel/git/broonie/regmap

Pull regmap fixes from Mark Brown:
 "There's couple of patches here that came in since my pull request:

   - What is effectively a quirk for shoehorning support for a wider
     range of I2C regmaps on weirdly restricted SMBus controllers

   - One minor fix for a memory leak on in error handling in the dummy
     driver used by the KUnit tests"

* tag 'regmap-fix-v7.1-merge-window' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regmap:
  regmap: ram: fix memory leaks in __regmap_init_ram() on error
  regmap-i2c: add SMBus byte/word reg16 bus for adapters lacking I2C_FUNC_I2C

3 months agoMerge tag 'gpio-fixes-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 18:59:46 +0000 (11:59 -0700)]
Merge tag 'gpio-fixes-for-v7.1-rc1' of git://git./linux/kernel/git/brgl/linux

Pull gpio fixes from Bartosz Golaszewski:

 - fix a regression in gpio-rockchip introduced on older chips during
   the merge window when converting to dynamic GPIO base

 - fix AST2700 debounce selector bit definitions in gpio-aspeed

* tag 'gpio-fixes-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: aspeed: fix AST2700 debounce selector bit definitions
  gpio: rockchip: Fix GPIO regression after conversion to dynamic base allocation

3 months agoMerge tag 'sound-fix-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 24 Apr 2026 18:49:20 +0000 (11:49 -0700)]
Merge tag 'sound-fix-7.1-rc1' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "Here are the rest of small updates for 7.1-rc1. All small fixes mostly
  for device-specific issues or regressions.

  Core:
   - Fix a potential data race in fasync handling

  USB-audio:
   - New device support: Line6 POD HD PRO, NexiGo N930W webcam
   - Fixes for Audio Advantage Micro II SPDIF switch and E-MU sample
     rates
   - Limit UAC2 rate parsing to prevent potential overflows

  HD-Audio:
   - Device-specific quirks for HP, Acer, and Honor laptops
   - Fix for TAS2781 SPI device abnormal sound
   - Move Intel firmware loading into probe work to avoid stalling

  ASoC:
   - New support for TI TAS5832
   - Fixes for SoundWire SDCA/DisCo boolean parsing
   - Driver-specific fixes for Intel SOF, ES8311, RT1320, and PXA2xx

  Misc:
   - Fixes for resource leaks and data races in 6fire, caiaq, als4000,
     and pcmtest drivers"

* tag 'sound-fix-7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (41 commits)
  Revert "ALSA: pcmtest: fix reference leak on failed device registration"
  ASoC: tas2781: Add tas5832 support
  ASoC: dt-bindings: ti,tas2781: Add TAS5832 support
  ALSA: usb-audio: Fix Audio Advantage Micro II SPDIF switch
  ALSA: usb-audio: Avoid false E-MU sample-rate notifications
  ASoC: sdw_utils: cs42l43: allow spk component names to be combined
  ASoC: qcom: x1e80100: limit speaker volumes
  ALSA: hda/realtek - Add mute LED support for HP Victus 15-fa2xxx
  ALSA: pcmtest: Fix resource leaks in module init error paths
  ALSA: usb-audio/line6: Add support for POD HD PRO
  ALSA: hda/realtek: Add LED fixup for HP EliteBook 6 G2a Laptops
  ASoC: SDCA: Fix reading of mipi-sdca-control-deferrable
  regmap: sdw-mbq: Allow defers on undeferrable controls
  Revert "ALSA: usb-audio: Add quirk for SmartlinkTechnology M01"
  ALSA: als4000: Fix capture trigger chip->mode race
  ALSA: core: Fix potential data race at fasync handling
  ALSA: hda/tas2781: Fix sound abnormal issue on some SPI device
  ALSA: hda/realtek: add quirk for Acer Nitro 16 AN16-41
  ALSA: caiaq: Fix control_put() result and cache rollback
  ALSA: pcmtest: fix reference leak on failed device registration
  ...

3 months agoMerge tag 'drm-fixes-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Fri, 24 Apr 2026 18:44:52 +0000 (11:44 -0700)]
Merge tag 'drm-fixes-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel

Pull more drm fixes from Dave Airlie:
 "These are the regular fixes that have built up over last couple of
  weeks, all pretty minor and spread all over.

  atomic:
   - raise the vblank timeout to avoid it on virtual drivers
   - fix colorop duplication

  bridge:
   - stm_lvds: state check fix
   - dw-mipi-dsi: bridge reference leak fix

  panel:
   - visionx-rm69299: init fix

  dma-fence:
   - fix sparse warning

  dma-buf:
   - UAF fix

  panthor:
   - mapping fix

  arcgpu:
   - device_node reference leak fix

  nouveau:
   - memory leak in error path fix
   - overflow in reloc path for old hw fix

  hv:
   - Kconfig fix

  v3d:
   - infinite loop fix"

* tag 'drm-fixes-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel:
  drm/nouveau: fix u32 overflow in pushbuf reloc bounds check
  MAINTAINERS: split hisilicon maintenance and add Yongbang Shi for hibmc-drm matainers
  drm/v3d: Reject empty multisync extension to prevent infinite loop
  drm/panel: visionox-rm69299: Make use of prepare_prev_first
  drm/drm_atomic: duplicate colorop states if plane color pipeline in use
  drm/nouveau: fix nvkm_device leak on aperture removal failure
  hv: Select CONFIG_SYSFB only for CONFIG_HYPERV_VMBUS
  dma-fence: Silence sparse warning in dma_fence_describe
  drm/bridge: dw-mipi-dsi: Fix bridge leak when host attach fails
  drm/arcpgu: fix device node leak
  drm/panthor: Fix outdated function documentation
  drm/panthor: Extend VM locked region for remap case to be a superset
  dma-buf: fix UAF in dma_buf_put() tracepoint
  drm/bridge: stm_lvds: Do not fail atomic_check on disabled connector
  drm/atomic: Increase timeout in drm_atomic_helper_wait_for_vblanks()

3 months agoMerge tag 'drm-next-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel
Linus Torvalds [Fri, 24 Apr 2026 18:33:23 +0000 (11:33 -0700)]
Merge tag 'drm-next-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel

Pull drm next fixes from Dave Airlie:
 "This is the first of two fixes for the merge PRs, the other is based
  on 7.0 branch. This mostly AMD fixes, a couple of weeks of backlog
  built up and this weeks. The main complaint I've seen is some boot
  warnings around the FP code handling which this should fix. Otherwise
  a single rcar-du and a single i915 fix.

  amdgpu:
   - SMU 14 fixes
   - Partition fixes
   - SMUIO 15.x fix
   - SR-IOV fixes
   - JPEG fix
   - PSP 15.x fix
   - NBIF fix
   - Devcoredump fixes
   - DPC fix
   - RAS fixes
   - Aldebaran smu fix
   - IP discovery fix
   - SDMA 7.1 fix
   - Runtime pm fix
   - MES 12.1 fix
   - DML2 fixes
   - DCN 4.2 fixes
   - YCbCr fixes
   - Freesync fixes
   - ISM fixes
   - Overlay cursor fix
   - DC FP fixes
   - UserQ locking fixes
   - DC idle state manager fix
   - ASPM fix
   - GPUVM SVM fix
   - DCE 6 fix

  amdkfd:
   - Fix memory clear handling
   - num_of_nodes bounds check fix

  i915:
   - Fix uninitialized variable in the alignment loop [psr]

  rcar-du:
   - fix NULL-ptr crash"

* tag 'drm-next-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel: (75 commits)
  drm/amdkfd: Add upper bound check for num_of_nodes
  drm: rcar-du: Fix crash when no CMM is available
  drm/amd/display: Disable 10-bit truncation and dithering on DCE 6.x
  drm/amdgpu: OR init_pte_flags into invalid leaf PTE updates
  drm/amd: Adjust ASPM support quirk to cover more Intel hosts
  drm/amd/display: Undo accidental fix revert in amdgpu_dm_ism.c
  drm/i915/psr: Init variable to avoid early exit from et alignment loop
  drm/amdgpu: drop userq fence driver refs out of fence process()
  drm/amdgpu/userq: unpin and unref doorbell and wptr outside mutex
  drm/amdgpu/userq: use pm_runtime_resume_and_get and fix err handling
  drm/amdgpu/userq: unmap_helper dont return the queue state
  drm/amdgpu/userq: unmap is to be called before freeing doorbell/wptr bo
  drm/amdgpu/userq: hold root bo lock in caller of input_va_validate
  drm/amdgpu/userq: caller to take reserv lock for vas_list_cleanup
  drm/amdgpu/userq: create_mqd does not need userq_mutex
  drm/amdgpu/userq: dont lock root bo with userq_mutex held
  drm/amdgpu/userq: fix kerneldoc for amdgpu_userq_ensure_ev_fence
  drm/amdgpu/userq: clean the VA mapping list for failed queue creation
  drm/amdgpu/userq: avoid uneccessary locking in amdgpu_userq_create
  drm/amd/display: Fix ISM teardown crash from NULL dc dereference
  ...

3 months agoMerge tag 'locking-urgent-2026-04-24' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 17:14:29 +0000 (10:14 -0700)]
Merge tag 'locking-urgent-2026-04-24' of git://git./linux/kernel/git/tip/tip

Pull locking fixes from Ingo Molnar:

 - Fix ww_mutex regression, which caused hangs/pauses in some DRM drivers

 - Fix rtmutex proxy-rollback bug

* tag 'locking-urgent-2026-04-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/mutex: Fix ww_mutex wait_list operations
  rtmutex: Use waiter::task instead of current in remove_waiter()

3 months agodrm/panel: panasonic-vvx10f034n00: transition to mipi_dsi wrapped functions
Avinal Kumar [Fri, 17 Apr 2026 06:18:42 +0000 (11:48 +0530)]
drm/panel: panasonic-vvx10f034n00: transition to mipi_dsi wrapped functions

Change the panasonic-vvx10f034n00 panel to multi
style functions for improved error handling and
remove redundant error printout.

Signed-off-by: Avinal Kumar <avinal.xlvii@gmail.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patch.msgid.link/20260417061842.66631-3-avinal.xlvii@gmail.com
3 months agodrm/mipi-dsi: add mipi_dsi_shutdown_peripheral_multi
Avinal Kumar [Fri, 17 Apr 2026 06:18:41 +0000 (11:48 +0530)]
drm/mipi-dsi: add mipi_dsi_shutdown_peripheral_multi

Add mipi_dsi_shutdown_peripheral_multi function and mark
mipi_dsi_shutdown_peripheral function as deprecated.

Signed-off-by: Avinal Kumar <avinal.xlvii@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://patch.msgid.link/20260417061842.66631-2-avinal.xlvii@gmail.com
3 months agoMerge tag 'x86-urgent-2026-04-24' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 24 Apr 2026 17:05:42 +0000 (10:05 -0700)]
Merge tag 'x86-urgent-2026-04-24' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Ingo Molnar:

 - Prevent deadlock during shstk sigreturn (Rick Edgecombe)

 - Disable FRED when PTI is forced on (Dave Hansen)

 - Revert a CPA INVLPGB optimization that did not properly handle
   discontiguous virtual addresses (Dave Hansen)

* tag 'x86-urgent-2026-04-24' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Revert INVLPGB optimization for set_memory code
  x86/cpu: Disable FRED when PTI is forced on
  x86/shstk: Prevent deadlock during shstk sigreturn

3 months agoMerge tag 'riscv-for-linus-7.1-mw1' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 24 Apr 2026 17:00:37 +0000 (10:00 -0700)]
Merge tag 'riscv-for-linus-7.1-mw1' of git://git./linux/kernel/git/riscv/linux

Pull RISC-V updates from Paul Walmsley:
 "There is one significant change outside arch/riscv in this pull
  request: the addition of a set of KUnit tests for strlen(), strnlen(),
  and strrchr().

  Otherwise, the most notable changes are to add some RISC-V-specific
  string function implementations, to remove XIP kernel support, to add
  hardware error exception handling, and to optimize our runtime
  unaligned access speed testing.

  A few comments on the motivation for removing XIP support. It's been
  broken in the RISC-V kernel for months. The code is not easy to
  maintain. Furthermore, for XIP support to truly be useful for RISC-V,
  we think that compile-time feature switches would need to be added for
  many of the RISC-V ISA features and microarchitectural properties that
  are currently implemented with runtime patching. No one has stepped
  forward to take responsibility for that work, so many of us think it's
  best to remove it until clear use cases and champions emerge.

  Summary:

   - Add Kunit correctness testing and microbenchmarks for strlen(),
     strnlen(), and strrchr()

   - Add RISC-V-specific strnlen(), strchr(), strrchr() implementations

   - Add hardware error exception handling

   - Clean up and optimize our unaligned access probe code

   - Enable HAVE_IOREMAP_PROT to be able to use generic_access_phys()

   - Remove XIP kernel support

   - Warn when addresses outside the vmemmap range are passed to
     vmemmap_populate()

   - Update the ACPI FADT revision check to warn if it's not at least
     ACPI v6.6, which is when key RISC-V-specific tables were added to
     the specification

   - Increase COMMAND_LINE_SIZE to 2048 to match ARM64, x86, PowerPC,
     etc.

   - Make kaslr_offset() a static inline function, since there's no need
     for it to show up in the symbol table

   - Add KASLR offset and SATP to the VMCOREINFO ELF notes to improve
     kdump support

   - Add Makefile cleanup rule for vdso_cfi copied source files, and add
     a .gitignore for the build artifacts in that directory

   - Remove some redundant ifdefs that check Kconfig macros

   - Add missing SPDX license tag to the CFI selftest

   - Simplify UTS_MACHINE assignment in the RISC-V Makefile

   - Clarify some unclear comments and remove some superfluous comments

   - Fix various English typos across the RISC-V codebase"

* tag 'riscv-for-linus-7.1-mw1' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux: (31 commits)
  riscv: Remove support for XIP kernel
  riscv: Reuse compare_unaligned_access() in check_vector_unaligned_access()
  riscv: Split out compare_unaligned_access()
  riscv: Reuse measure_cycles() in check_vector_unaligned_access()
  riscv: Split out measure_cycles() for reuse
  riscv: Clean up & optimize unaligned scalar access probe
  riscv: lib: add strrchr() implementation
  riscv: lib: add strchr() implementation
  riscv: lib: add strnlen() implementation
  lib/string_kunit: extend benchmarks to strnlen() and chr searches
  lib/string_kunit: add performance benchmark for strlen()
  lib/string_kunit: add correctness test for strrchr()
  lib/string_kunit: add correctness test for strnlen()
  lib/string_kunit: add correctness test for strlen()
  riscv: vdso_cfi: Add .gitignore for build artifacts
  riscv: vdso_cfi: Add clean rule for copied sources
  riscv: enable HAVE_IOREMAP_PROT
  riscv: mm: WARN_ON() for bad addresses in vmemmap_populate()
  riscv: acpi: update FADT revision check to 6.6
  riscv: add hardware error trap handler support
  ...

3 months agoMerge tag 'loongarch-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai...
Linus Torvalds [Fri, 24 Apr 2026 16:54:45 +0000 (09:54 -0700)]
Merge tag 'loongarch-7.1' of git://git./linux/kernel/git/chenhuacai/linux-loongson

Pull LoongArch updates from Huacai Chen:

 - Adjust build infrastructure for 32BIT/64BIT

 - Add HIGHMEM (PKMAP and FIX_KMAP) support

 - Show and handle CPU vulnerabilites correctly

 - Batch the icache maintenance for jump_label

 - Add more atomic instructions support for BPF JIT

 - Add more features (e.g. fsession) support for BPF trampoline

 - Some bug fixes and other small changes

* tag 'loongarch-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson: (21 commits)
  selftests/bpf: Enable CAN_USE_LOAD_ACQ_STORE_REL for LoongArch
  LoongArch: BPF: Add fsession support for trampolines
  LoongArch: BPF: Introduce emit_store_stack_imm64() helper
  LoongArch: BPF: Support up to 12 function arguments for trampoline
  LoongArch: BPF: Support small struct arguments for trampoline
  LoongArch: BPF: Open code and remove invoke_bpf_mod_ret()
  LoongArch: BPF: Support load-acquire and store-release instructions
  LoongArch: BPF: Support 8 and 16 bit read-modify-write instructions
  LoongArch: BPF: Add the default case in emit_atomic() and rename it
  LoongArch: Define instruction formats for AM{SWAP/ADD}.{B/H} and DBAR
  LoongArch: Batch the icache maintenance for jump_label
  LoongArch: Add flush_icache_all()/local_flush_icache_all()
  LoongArch: Add spectre boundry for syscall dispatch table
  LoongArch: Show CPU vulnerabilites correctly
  LoongArch: Make arch_irq_work_has_interrupt() true only if IPI HW exist
  LoongArch: Use get_random_canary() for stack canary init
  LoongArch: Improve the logging of disabling KASLR
  LoongArch: Align FPU register state to 32 bytes
  LoongArch: Handle CONFIG_32BIT in syscall_get_arch()
  LoongArch: Add HIGHMEM (PKMAP and FIX_KMAP) support
  ...

3 months agoMerge tag 'net-deletions' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev...
Linus Torvalds [Fri, 24 Apr 2026 16:41:58 +0000 (09:41 -0700)]
Merge tag 'net-deletions' of git://git./linux/kernel/git/netdev/net-next

Pull networking deletions from Jakub Kicinski:
 "Delete some obsolete networking code

  Old code like amateur radio and NFC have long been a burden to core
  networking developers. syzbot loves to find bugs in BKL-era code, and
  noobs try to fix them.

  If we want to have a fighting chance of surviving the LLM-pocalypse
  this code needs to find a dedicated owner or get deleted. We've talked
  about these deletions multiple times in the past and every time
  someone wanted the code to stay. It is never very clear to me how many
  of those people actually use the code vs are just nostalgic to see it
  go. Amateur radio did have occasional users (or so I think) but most
  users switched to user space implementations since its all super slow
  stuff. Nobody stepped up to maintain the kernel code.

  We were lucky enough to find someone who wants to help with NFC so
  we're giving that a chance. Let's try to put the rest of this code
  behind us"

* tag 'net-deletions' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next:
  drivers: net: 8390: wd80x3: Remove this driver
  drivers: net: 8390: ultra: Remove this driver
  drivers: net: 8390: AX88190: Remove this driver
  drivers: net: fujitsu: fmvj18x: Remove this driver
  drivers: net: smsc: smc91c92: Remove this driver
  drivers: net: smsc: smc9194: Remove this driver
  drivers: net: amd: nmclan: Remove this driver
  drivers: net: amd: lance: Remove this driver
  drivers: net: 3com: 3c589: Remove this driver
  drivers: net: 3com: 3c574: Remove this driver
  drivers: net: 3com: 3c515: Remove this driver
  drivers: net: 3com: 3c509: Remove this driver
  net: packetengines: remove obsolete yellowfin driver and vendor dir
  net: packetengines: remove obsolete hamachi driver
  net: remove unused ATM protocols and legacy ATM device drivers
  net: remove ax25 and amateur radio (hamradio) subsystem
  net: remove ISDN subsystem and Bluetooth CMTP
  caif: remove CAIF NETWORK LAYER