Skip to contentSkip to conversation

drm/amd/display: Do not set drr on pipe commit

PATCH In review Sasha Levin 2 participants 59 messages Jul 24, 2023

Subsystems (13)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/58 drm/amd/display: Do not set drr on pipe commit

From: Wesley Chalmers <[email protected]>

[ Upstream commit e101bf95ea87ccc03ac2f48dfc0757c6364ff3c7 ]

[WHY]
Writing to DRR registers such as OTG_V_TOTAL_MIN on the same frame as a pipe commit can cause underflow.

[HOW]
Move DMUB p-state delegate into optimze_bandwidth; enabling FAMS sets optimized_required.

This change expects that Freesync requests are blocked when optimized_required is true.

---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c | 7 +++++++
 2 files changed, 13 insertions(+)
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +6 -0
index 5403e9399a465..6ce10fd4bb1a4 100644
@@ -2113,6 +2113,12 @@ void dcn20_optimize_bandwidth(
 	if (hubbub->funcs->program_compbuf_size)
 		hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true);
 
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+		dc_dmub_srv_p_state_delegate(dc,
+			true, context);
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+	}
+
 	dc->clk_mgr->funcs->update_clocks(
 			dc->clk_mgr,
 			context,
drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +7 -0
index 32121db2851e6..f923224e85fc6 100644
@@ -985,11 +985,18 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 void dcn30_prepare_bandwidth(struct dc *dc,
 			     struct dc_state *context)
 {
+	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
+		dc->optimized_required = true;
+		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
+	}
+
 	if (dc->clk_mgr->dc_mode_softmax_enabled)
 		if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 &&
 				context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000)
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
+
+	dc_dmub_srv_p_state_delegate(dc, false, context);
 }
 
Signature · 1 line
2.39.2
author maintainer
#1 lore raw

PATCH 2/58 drm/amd/display: Update DTBCLK for DCN32

From: Alvin Lee <[email protected]>

[ Upstream commit 128c1ca0303fe764a4cde5f761e72810d9e40b6e ]

[Why&How]
- Implement interface to program DTBCLK DTO’s according to reference DTBCLK returned by PMFW
- This is required because DTO programming requires exact DTBCLK reference freq or it could result in underflow

---
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c  | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)
drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +27 -0
index 8d9444db092ab..eea103908b09f 100644
@@ -233,6 +233,32 @@ void dcn32_init_clocks(struct clk_mgr *clk_mgr_base)
 	DC_FP_END();
 }
 
+static void dcn32_update_clocks_update_dtb_dto(struct clk_mgr_internal *clk_mgr,
+			struct dc_state *context,
+			int ref_dtbclk_khz)
+{
+	struct dccg *dccg = clk_mgr->dccg;
+	uint32_t tg_mask = 0;
+	int i;
+
+	for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) {
+		struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+		struct dtbclk_dto_params dto_params = {0};
+
+		/* use mask to program DTO once per tg */
+		if (pipe_ctx->stream_res.tg &&
+				!(tg_mask & (1 << pipe_ctx->stream_res.tg->inst))) {
+			tg_mask |= (1 << pipe_ctx->stream_res.tg->inst);
+
+			dto_params.otg_inst = pipe_ctx->stream_res.tg->inst;
+			dto_params.ref_dtbclk_khz = ref_dtbclk_khz;
+
+			dccg->funcs->set_dtbclk_dto(clk_mgr->dccg, &dto_params);
+			//dccg->funcs->set_audio_dtbclk_dto(clk_mgr->dccg, &dto_params);
+		}
+	}
+}
+
 /* Since DPPCLK request to PMFW needs to be exact (due to DPP DTO programming),
  * update DPPCLK to be the exact frequency that will be set after the DPPCLK
  * divider is updated. This will prevent rounding issues that could cause DPP
@@ -570,6 +596,7 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base,
 		/* DCCG requires KHz precision for DTBCLK */
 		clk_mgr_base->clks.ref_dtbclk_khz =
 				dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DTBCLK, khz_to_mhz_ceil(new_clocks->ref_dtbclk_khz));
+		dcn32_update_clocks_update_dtb_dto(clk_mgr, context, clk_mgr_base->clks.ref_dtbclk_khz);
 	}
 
 	if (dc->config.forced_clocks == false || (force_reset && safe_to_lower)) {
Signature · 1 line
2.39.2
author maintainer
#2 lore raw

PATCH 3/58 drm/scheduler: set entity to NULL in drm_sched_entity_pop_job()

In reply to Sasha Levin

From: Danilo Krummrich <[email protected]>

[ Upstream commit 96c7c2f4d5bd94b15fe63448c087f01607b56f4a ]

It already happend a few times that patches slipped through which implemented access to an entity through a job that was already removed from the entities queue. Since jobs and entities might have different lifecycles, this can potentially cause UAF bugs.

In order to make it obvious that a jobs entity pointer shouldn't be accessed after drm_sched_entity_pop_job() was called successfully, set the jobs entity pointer to NULL once the job is removed from the entity queue.

Moreover, debugging a potential NULL pointer dereference is way easier than potentially corrupted memory through a UAF.

---
 drivers/gpu/drm/scheduler/sched_entity.c | 6 ++++++
 drivers/gpu/drm/scheduler/sched_main.c   | 4 ++++
 2 files changed, 10 insertions(+)
drivers/gpu/drm/scheduler/sched_entity.c +6 -0
index e0a8890a62e23..3e2a31d8190eb 100644
@@ -448,6 +448,12 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 			drm_sched_rq_update_fifo(entity, next->submit_ts);
 	}
 
+	/* Jobs and entities might have different lifecycles. Since we're
+	 * removing the job from the entities queue, set the jobs entity pointer
+	 * to NULL to prevent any future access of the entity through this job.
+	 */
+	sched_job->entity = NULL;
+
 	return sched_job;
 }
 
drivers/gpu/drm/scheduler/sched_main.c +4 -0
index aea5a90ff98b9..cdd67676c3d1b 100644
@@ -42,6 +42,10 @@
  *    the hardware.
  *
  * The jobs in a entity are always scheduled in the order that they were pushed.
+ *
+ * Note that once a job was taken from the entities queue and pushed to the
+ * hardware, i.e. the pending queue, the entity must not be referenced anymore
+ * through the jobs entity pointer.
  */
 
 #include <linux/kthread.h>
Signature · 1 line
2.39.2
author maintainer
#4 lore raw

PATCH 5/58 drm/radeon: Fix integer overflow in radeon_cs_parser_init

In reply to Sasha Levin

From: hackyzh002 <[email protected]>

[ Upstream commit f828b681d0cd566f86351c0b913e6cb6ed8c7b9c ]

The type of size is unsigned, if size is 0x40000000, there will be an integer overflow, size will be zero after size *= sizeof(uint32_t), will cause uninitialized memory to be referenced later

---
 drivers/gpu/drm/radeon/radeon_cs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/radeon/radeon_cs.c +2 -1
index 46a27ebf4588a..a6700d7278bf3 100644
@@ -270,7 +270,8 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data)
 {
 	struct drm_radeon_cs *cs = data;
 	uint64_t *chunk_array_ptr;
-	unsigned size, i;
+	u64 size;
+	unsigned i;
 	u32 ring = RADEON_CS_RING_GFX;
 	s32 priority = 0;
 
Signature · 1 line
2.39.2
author maintainer
#3 lore raw

PATCH 4/58 drm/amdgpu: fix calltrace warning in amddrm_buddy_fini

In reply to Sasha Levin

From: Longlong Yao <[email protected]>

[ Upstream commit 01382501509871d0799bab6bd412c228486af5bf ]

The following call trace is observed when removing the amdgpu driver, which is caused by that BOs allocated for psp are not freed until removing.

[61811.450562] RIP: 0010:amddrm_buddy_fini.cold+0x29/0x47 [amddrm_buddy]
[61811.450577] Call Trace:
[61811.450577]  <TASK>
[61811.450579]  amdgpu_vram_mgr_fini+0x135/0x1c0 [amdgpu]
[61811.450728]  amdgpu_ttm_fini+0x207/0x290 [amdgpu]
[61811.450870]  amdgpu_bo_fini+0x27/0xa0 [amdgpu]
[61811.451012]  gmc_v9_0_sw_fini+0x4a/0x60 [amdgpu]
[61811.451166]  amdgpu_device_fini_sw+0x117/0x520 [amdgpu]
[61811.451306]  amdgpu_driver_release_kms+0x16/0x30 [amdgpu]
[61811.451447]  devm_drm_dev_init_release+0x4d/0x80 [drm]
[61811.451466]  devm_action_release+0x15/0x20
[61811.451469]  release_nodes+0x40/0xb0
[61811.451471]  devres_release_all+0x9b/0xd0
[61811.451473]  __device_release_driver+0x1bb/0x2a0
[61811.451476]  driver_detach+0xf3/0x140
[61811.451479]  bus_remove_driver+0x6c/0xf0
[61811.451481]  driver_unregister+0x31/0x60
[61811.451483]  pci_unregister_driver+0x40/0x90
[61811.451486]  amdgpu_exit+0x15/0x447 [amdgpu]

For smu v13_0_2, if the GPU supports xgmi, refer to

commit f5c7e7797060 ("drm/amdgpu: Adjust removal control flow for smu v13_0_2"),

it will run gpu recover in AMDGPU_RESET_FOR_DEVICE_REMOVE mode when removing, which makes all devices in hive list have hw reset but no resume except the basic ip blocks, then other ip blocks will not call .hw_fini according to ip_block.status.hw.

Since psp_free_shared_bufs just includes some software operations, so move it to psp_sw_fini.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +2 -2
index a150b7a4b4aae..fc58a51cb0c00 100644
@@ -520,6 +520,8 @@ static int psp_sw_fini(void *handle)
 	kfree(cmd);
 	cmd = NULL;
 
+	psp_free_shared_bufs(psp);
+
 	if (psp->km_ring.ring_mem)
 		amdgpu_bo_free_kernel(&adev->firmware.rbuf,
 				      &psp->km_ring.ring_mem_mc_addr,
@@ -2655,8 +2657,6 @@ static int psp_hw_fini(void *handle)
 
 	psp_ring_destroy(psp, PSP_RING_TYPE__KM);
 
-	psp_free_shared_bufs(psp);
-
 	return 0;
 }
 
Signature · 1 line
2.39.2
author maintainer
#5 lore raw

PATCH 6/58 drm/amdgpu: Fix integer overflow in amdgpu_cs_pass1

In reply to Sasha Levin

From: hackyzh002 <[email protected]>

[ Upstream commit 87c2213e85bd81e4a9a4d0880c256568794ae388 ]

The type of size is unsigned int, if size is 0x40000000, there will be an integer overflow, size will be zero after size *= sizeof(uint32_t), will cause uninitialized memory to be referenced later.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +1 -1
index 2eb2c66843a88..d3b7148a4e30f 100644
@@ -192,7 +192,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 	uint64_t *chunk_array_user;
 	uint64_t *chunk_array;
 	uint32_t uf_offset = 0;
-	unsigned int size;
+	size_t size;
 	int ret;
 	int i;
 
Signature · 1 line
2.39.2
author maintainer
#6 lore raw

PATCH 7/58 drm/amdgpu: fix memory leak in mes self test

In reply to Sasha Levin

From: Jack Xiao <[email protected]>

[ Upstream commit 31d7c3a4fc3d312a0646990767647925d5bde540 ]

The fences associated with mes queue have to be freed up during amdgpu_ring_fini.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +2 -0
index 49de3a3eebc78..de04606c2061e 100644
@@ -361,6 +361,8 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 		amdgpu_bo_free_kernel(&ring->ring_obj,
 				      &ring->gpu_addr,
 				      (void **)&ring->ring);
+	} else {
+		kfree(ring->fence_drv.fences);
 	}
 
 	dma_fence_put(ring->vmid_wait);
Signature · 1 line
2.39.2
author maintainer
#7 lore raw

PATCH 8/58 Revert "drm/amd/display: disable SubVP + DRR to prevent underflow"

In reply to Sasha Levin

From: Aurabindo Pillai <[email protected]>

[ Upstream commit f38129bb081758176dd78304faaee95007fb8838 ]

This reverts commit 80c6d6804f31451848a3956a70c2bcb1f07cfcb0. The orignal commit was intended as a workaround to prevent underflow and flickering when using one normal monitor and the other high refresh rate monitor (> 120Hz).

This patch is being reverted in favour of a software solution to enable SubVP+DRR

---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c    | 5 -----
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c | 4 ----
 drivers/gpu/drm/amd/include/amd_shared.h             | 1 -
 3 files changed, 10 deletions(-)
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +0 -5
index 7acd73e5004fb..80756c7b92a54 100644
@@ -1646,11 +1646,6 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 	if (amdgpu_dc_feature_mask & DC_DISABLE_LTTPR_DP2_0)
 		init_data.flags.allow_lttpr_non_transparent_mode.bits.DP2_0 = true;
 
-	/* Disable SubVP + DRR config by default */
-	init_data.flags.disable_subvp_drr = true;
-	if (amdgpu_dc_feature_mask & DC_ENABLE_SUBVP_DRR)
-		init_data.flags.disable_subvp_drr = false;
-
 	init_data.flags.seamless_boot_edp_requested = false;
 
 	if (check_seamless_boot_capability(adev)) {
drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +0 -4
index 0c4c3208def17..d7eca24bd09d2 100644
@@ -880,10 +880,6 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context, struc
 	int16_t stretched_drr_us = 0;
 	int16_t drr_stretched_vblank_us = 0;
 	int16_t max_vblank_mallregion = 0;
-	const struct dc_config *config = &dc->config;
-
-	if (config->disable_subvp_drr)
-		return false;
 
 	// Find SubVP pipe
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
drivers/gpu/drm/amd/include/amd_shared.h +0 -1
index e4a22c68517d1..f175e65b853a0 100644
@@ -240,7 +240,6 @@ enum DC_FEATURE_MASK {
 	DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default
 	DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default
 	DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default
-	DC_ENABLE_SUBVP_DRR = (1 << 9), // 0x200, disabled by default
 };
 
 enum DC_DEBUG_MASK {
Signature · 1 line
2.39.2
author maintainer
#13 lore raw

PATCH 14/58 iommu: Make __iommu_group_set_domain() handle error unwind

In reply to Sasha Levin

From: Jason Gunthorpe <[email protected]>

[ Upstream commit dcf40ed3a20d727be054c4a20db47b32cb5036d4 ]

Let's try to have a consistent and clear strategy for error handling during domain attach failures.

There are two broad categories, the first is callers doing destruction and trying to set the domain back to a previously good domain. These cases cannot handle failure during destruction flows and must succeed, or at least avoid a UAF on the current group->domain which is likely about to be freed.

Many of the drivers are well behaved here and will not hit the WARN_ON's or a UAF, but some are doing hypercalls/etc that can fail unpredictably and don't meet the expectations.

The second case is attaching a domain for the first time in a failable context, failure should restore the attachment back to group->domain using the above unfailable operation.

Have __iommu_group_set_domain_internal() execute a common algorithm that tries to achieve this, and in the worst case, would leave a device "detached" or assigned to a global blocking domain. This relies on some existing common driver behaviors where attach failure will also do detatch and true IOMMU_DOMAIN_BLOCK implementations that are not allowed to ever fail.

Name the first case with __iommu_group_set_domain_nofail() to make it clear.

Pull all the error handling and WARN_ON generation into __iommu_group_set_domain_internal().

Avoid the obfuscating use of __iommu_group_for_each_dev() and be more careful about what should happen during failures by only touching devices we've already touched.

---
 drivers/iommu/iommu.c | 137 ++++++++++++++++++++++++++++++++++--------
 1 file changed, 112 insertions(+), 25 deletions(-)
drivers/iommu/iommu.c +112 -25
index f1dcfa3f1a1b4..873e66ab1e2d7 100644
@@ -97,8 +97,26 @@ static int __iommu_attach_device(struct iommu_domain *domain,
 				 struct device *dev);
 static int __iommu_attach_group(struct iommu_domain *domain,
 				struct iommu_group *group);
+
+enum {
+	IOMMU_SET_DOMAIN_MUST_SUCCEED = 1 << 0,
+};
+
+static int __iommu_group_set_domain_internal(struct iommu_group *group,
+					     struct iommu_domain *new_domain,
+					     unsigned int flags);
 static int __iommu_group_set_domain(struct iommu_group *group,
-				    struct iommu_domain *new_domain);
+				    struct iommu_domain *new_domain)
+{
+	return __iommu_group_set_domain_internal(group, new_domain, 0);
+}
+static void __iommu_group_set_domain_nofail(struct iommu_group *group,
+					    struct iommu_domain *new_domain)
+{
+	WARN_ON(__iommu_group_set_domain_internal(
+		group, new_domain, IOMMU_SET_DOMAIN_MUST_SUCCEED));
+}
+
 static int iommu_create_device_direct_mappings(struct iommu_group *group,
 					       struct device *dev);
 static struct iommu_group *iommu_group_get_for_dev(struct device *dev);
@@ -2028,15 +2046,13 @@ EXPORT_SYMBOL_GPL(iommu_domain_free);
 static void __iommu_group_set_core_domain(struct iommu_group *group)
 {
 	struct iommu_domain *new_domain;
-	int ret;
 
 	if (group->owner)
 		new_domain = group->blocking_domain;
 	else
 		new_domain = group->default_domain;
 
-	ret = __iommu_group_set_domain(group, new_domain);
-	WARN(ret, "iommu driver failed to attach the default/blocking domain");
+	__iommu_group_set_domain_nofail(group, new_domain);
 }
 
 static int __iommu_attach_device(struct iommu_domain *domain,
@@ -2221,21 +2237,55 @@ int iommu_attach_group(struct iommu_domain *domain, struct iommu_group *group)
 }
 EXPORT_SYMBOL_GPL(iommu_attach_group);
 
-static int iommu_group_do_set_platform_dma(struct device *dev, void *data)
+static int __iommu_device_set_domain(struct iommu_group *group,
+				     struct device *dev,
+				     struct iommu_domain *new_domain,
+				     unsigned int flags)
 {
-	const struct iommu_ops *ops = dev_iommu_ops(dev);
-
-	if (!WARN_ON(!ops->set_platform_dma_ops))
-		ops->set_platform_dma_ops(dev);
+	int ret;
 
+	ret = __iommu_attach_device(new_domain, dev);
+	if (ret) {
+		/*
+		 * If we have a blocking domain then try to attach that in hopes
+		 * of avoiding a UAF. Modern drivers should implement blocking
+		 * domains as global statics that cannot fail.
+		 */
+		if ((flags & IOMMU_SET_DOMAIN_MUST_SUCCEED) &&
+		    group->blocking_domain &&
+		    group->blocking_domain != new_domain)
+			__iommu_attach_device(group->blocking_domain, dev);
+		return ret;
+	}
 	return 0;
 }
 
-static int __iommu_group_set_domain(struct iommu_group *group,
-				    struct iommu_domain *new_domain)
+/*
+ * If 0 is returned the group's domain is new_domain. If an error is returned
+ * then the group's domain will be set back to the existing domain unless
+ * IOMMU_SET_DOMAIN_MUST_SUCCEED, otherwise an error is returned and the group's
+ * domains is left inconsistent. This is a driver bug to fail attach with a
+ * previously good domain. We try to avoid a kernel UAF because of this.
+ *
+ * IOMMU groups are really the natural working unit of the IOMMU, but the IOMMU
+ * API works on domains and devices.  Bridge that gap by iterating over the
+ * devices in a group.  Ideally we'd have a single device which represents the
+ * requestor ID of the group, but we also allow IOMMU drivers to create policy
+ * defined minimum sets, where the physical hardware may be able to distiguish
+ * members, but we wish to group them at a higher level (ex. untrusted
+ * multi-function PCI devices).  Thus we attach each device.
+ */
+static int __iommu_group_set_domain_internal(struct iommu_group *group,
+					     struct iommu_domain *new_domain,
+					     unsigned int flags)
 {
+	struct group_device *last_gdev;
+	struct group_device *gdev;
+	int result;
 	int ret;
 
+	lockdep_assert_held(&group->mutex);
+
 	if (group->domain == new_domain)
 		return 0;
 
@@ -2245,8 +2295,12 @@ static int __iommu_group_set_domain(struct iommu_group *group,
 	 * platform specific behavior.
 	 */
 	if (!new_domain) {
-		__iommu_group_for_each_dev(group, NULL,
-					   iommu_group_do_set_platform_dma);
+		for_each_group_device(group, gdev) {
+			const struct iommu_ops *ops = dev_iommu_ops(gdev->dev);
+
+			if (!WARN_ON(!ops->set_platform_dma_ops))
+				ops->set_platform_dma_ops(gdev->dev);
+		}
 		group->domain = NULL;
 		return 0;
 	}
@@ -2256,16 +2310,52 @@ static int __iommu_group_set_domain(struct iommu_group *group,
 	 * domain. This switch does not have to be atomic and DMA can be
 	 * discarded during the transition. DMA must only be able to access
 	 * either new_domain or group->domain, never something else.
-	 *
-	 * Note that this is called in error unwind paths, attaching to a
-	 * domain that has already been attached cannot fail.
 	 */
-	ret = __iommu_group_for_each_dev(group, new_domain,
-					 iommu_group_do_attach_device);
-	if (ret)
-		return ret;
+	result = 0;
+	for_each_group_device(group, gdev) {
+		ret = __iommu_device_set_domain(group, gdev->dev, new_domain,
+						flags);
+		if (ret) {
+			result = ret;
+			/*
+			 * Keep trying the other devices in the group. If a
+			 * driver fails attach to an otherwise good domain, and
+			 * does not support blocking domains, it should at least
+			 * drop its reference on the current domain so we don't
+			 * UAF.
+			 */
+			if (flags & IOMMU_SET_DOMAIN_MUST_SUCCEED)
+				continue;
+			goto err_revert;
+		}
+	}
 	group->domain = new_domain;
-	return 0;
+	return result;
+
+err_revert:
+	/*
+	 * This is called in error unwind paths. A well behaved driver should
+	 * always allow us to attach to a domain that was already attached.
+	 */
+	last_gdev = gdev;
+	for_each_group_device(group, gdev) {
+		const struct iommu_ops *ops = dev_iommu_ops(gdev->dev);
+
+		/*
+		 * If set_platform_dma_ops is not present a NULL domain can
+		 * happen only for first probe, in which case we leave
+		 * group->domain as NULL and let release clean everything up.
+		 */
+		if (group->domain)
+			WARN_ON(__iommu_device_set_domain(
+				group, gdev->dev, group->domain,
+				IOMMU_SET_DOMAIN_MUST_SUCCEED));
+		else if (ops->set_platform_dma_ops)
+			ops->set_platform_dma_ops(gdev->dev);
+		if (gdev == last_gdev)
+			break;
+	}
+	return ret;
 }
 
 void iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group)
@@ -3182,16 +3272,13 @@ EXPORT_SYMBOL_GPL(iommu_device_claim_dma_owner);
 
 static void __iommu_release_dma_ownership(struct iommu_group *group)
 {
-	int ret;
-
 	if (WARN_ON(!group->owner_cnt || !group->owner ||
 		    !xa_empty(&group->pasid_array)))
 		return;
 
 	group->owner_cnt = 0;
 	group->owner = NULL;
-	ret = __iommu_group_set_domain(group, group->default_domain);
-	WARN(ret, "iommu driver failed to attach the default domain");
+	__iommu_group_set_domain_nofail(group, group->default_domain);
 }
 
 /**
Signature · 1 line
2.39.2
author maintainer
#15 lore raw

PATCH 16/58 drm/stm: ltdc: fix late dereference check

In reply to Sasha Levin

From: Raphael Gallais-Pou <[email protected]>

[ Upstream commit 898a9e3f56db9860ab091d4bf41b6caa99aafc3d ]

In ltdc_crtc_set_crc_source(), struct drm_crtc was dereferenced in a container_of() before the pointer check. This could cause a kernel panic.

Fix this smatch warning:
drivers/gpu/drm/stm/ltdc.c:1124 ltdc_crtc_set_crc_source() warn: variable dereferenced before check 'crtc' (see line 1119)

---
 drivers/gpu/drm/stm/ltdc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/gpu/drm/stm/ltdc.c +3 -1
index 03c6becda795c..b8be4c1db4235 100644
@@ -1145,7 +1145,7 @@ static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc)
 
 static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 {
-	struct ltdc_device *ldev = crtc_to_ltdc(crtc);
+	struct ltdc_device *ldev;
 	int ret;
 
 	DRM_DEBUG_DRIVER("\n");
@@ -1153,6 +1153,8 @@ static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 	if (!crtc)
 		return -ENODEV;
 
+	ldev = crtc_to_ltdc(crtc);
+
 	if (source && strcmp(source, "auto") == 0) {
 		ldev->crc_active = true;
 		ret = regmap_set_bits(ldev->regmap, LTDC_GCR, GCR_CRCEN);
Signature · 1 line
2.39.2
author maintainer
#17 lore raw

PATCH 18/58 drm: rcar-du: remove R-Car H3 ES1.* workarounds

In reply to Sasha Levin

From: Wolfram Sang <[email protected]>

[ Upstream commit 2da4b728f994a1f9189a8066b0be90b615768764 ]

R-Car H3 ES1.* was only available to an internal development group and needed a lot of quirks and workarounds. These become a maintenance burden now, so our development group decided to remove upstream support for this SoC and prevent booting it. Public users only have ES2 onwards.

---
 drivers/gpu/drm/rcar-du/rcar_du_crtc.c | 37 ++------------------
 drivers/gpu/drm/rcar-du/rcar_du_drv.c  | 48 --------------------------
 drivers/gpu/drm/rcar-du/rcar_du_drv.h  |  2 --
 drivers/gpu/drm/rcar-du/rcar_du_regs.h |  3 +-
 4 files changed, 4 insertions(+), 86 deletions(-)
drivers/gpu/drm/rcar-du/rcar_du_crtc.c +3 -34
index d6d29be6b4f48..7e175dbfd8924 100644
@@ -223,20 +223,6 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 		 * DU channels that have a display PLL can't use the internal
 		 * system clock, and have no internal clock divider.
 		 */
-
-		/*
-		 * The H3 ES1.x exhibits dot clock duty cycle stability issues.
-		 * We can work around them by configuring the DPLL to twice the
-		 * desired frequency, coupled with a /2 post-divider. Restrict
-		 * the workaround to H3 ES1.x as ES2.0 and all other SoCs have
-		 * no post-divider when a display PLL is present (as shown by
-		 * the workaround breaking HDMI output on M3-W during testing).
-		 */
-		if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY) {
-			target *= 2;
-			div = 1;
-		}
-
 		extclk = clk_get_rate(rcrtc->extclock);
 		rcar_du_dpll_divider(rcrtc, &dpll, extclk, target);
 
@@ -245,30 +231,13 @@ static void rcar_du_crtc_set_display_timing(struct rcar_du_crtc *rcrtc)
 		       | DPLLCR_N(dpll.n) | DPLLCR_M(dpll.m)
 		       | DPLLCR_STBY;
 
-		if (rcrtc->index == 1) {
+		if (rcrtc->index == 1)
 			dpllcr |= DPLLCR_PLCS1
 			       |  DPLLCR_INCS_DOTCLKIN1;
-		} else {
-			dpllcr |= DPLLCR_PLCS0_PLL
+		else
+			dpllcr |= DPLLCR_PLCS0
 			       |  DPLLCR_INCS_DOTCLKIN0;
 
-			/*
-			 * On ES2.x we have a single mux controlled via bit 21,
-			 * which selects between DCLKIN source (bit 21 = 0) and
-			 * a PLL source (bit 21 = 1), where the PLL is always
-			 * PLL1.
-			 *
-			 * On ES1.x we have an additional mux, controlled
-			 * via bit 20, for choosing between PLL0 (bit 20 = 0)
-			 * and PLL1 (bit 20 = 1). We always want to use PLL1,
-			 * so on ES1.x, in addition to setting bit 21, we need
-			 * to set the bit 20.
-			 */
-
-			if (rcdu->info->quirks & RCAR_DU_QUIRK_H3_ES1_PLL)
-				dpllcr |= DPLLCR_PLCS0_H3ES1X_PLL1;
-		}
-
 		rcar_du_group_write(rcrtc->group, DPLLCR, dpllcr);
 
 		escr = ESCR_DCLKSEL_DCLKIN | div;
drivers/gpu/drm/rcar-du/rcar_du_drv.c +0 -48

78 lines not on this page · raw message

drivers/gpu/drm/rcar-du/rcar_du_drv.h +0 -2
index acc3673fefe18..5cfa2bb7ad93d 100644
@@ -34,8 +34,6 @@ struct rcar_du_device;
 #define RCAR_DU_FEATURE_NO_BLENDING	BIT(5)	/* PnMR.SPIM does not have ALP nor EOR bits */
 
 #define RCAR_DU_QUIRK_ALIGN_128B	BIT(0)	/* Align pitches to 128 bytes */
-#define RCAR_DU_QUIRK_H3_ES1_PCLK_STABILITY BIT(1)	/* H3 ES1 has pclk stability issue */
-#define RCAR_DU_QUIRK_H3_ES1_PLL	BIT(2)	/* H3 ES1 PLL setup differs from non-ES1 */
 
 enum rcar_du_output {
 	RCAR_DU_OUTPUT_DPAD0,
drivers/gpu/drm/rcar-du/rcar_du_regs.h +1 -2
index 6c750fab6ebb7..391de6661d8bc 100644
@@ -283,8 +283,7 @@
 #define DPLLCR			0x20044
 #define DPLLCR_CODE		(0x95 << 24)
 #define DPLLCR_PLCS1		(1 << 23)
-#define DPLLCR_PLCS0_PLL	(1 << 21)
-#define DPLLCR_PLCS0_H3ES1X_PLL1	(1 << 20)
+#define DPLLCR_PLCS0		(1 << 21)
 #define DPLLCR_CLKE		(1 << 18)
 #define DPLLCR_FDPLL(n)		((n) << 12)
 #define DPLLCR_N(n)		((n) << 5)
Signature · 1 line
2.39.2
author maintainer
#11 lore raw

PATCH 12/58 PCI: tegra194: Fix possible array out of bounds access

In reply to Sasha Levin

From: Sumit Gupta <[email protected]>

[ Upstream commit 205b3d02d57ce6dce96f6d2b9c230f56a9bf9817 ]

Add check to fix the possible array out of bounds violation by making speed equal to GEN1_CORE_CLK_FREQ when its value is more than the size of "pcie_gen_freq" array. This array has size of four but possible speed (CLS) values are from "0 to 0xF". So, "speed - 1" values are "-1 to 0xE".

---
 drivers/pci/controller/dwc/pcie-tegra194.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)
drivers/pci/controller/dwc/pcie-tegra194.c +11 -2
index 09825b4a075e5..e6eec85480ca9 100644
@@ -223,6 +223,7 @@
 #define EP_STATE_ENABLED	1
 
 static const unsigned int pcie_gen_freq[] = {
+	GEN1_CORE_CLK_FREQ,	/* PCI_EXP_LNKSTA_CLS == 0; undefined */
 	GEN1_CORE_CLK_FREQ,
 	GEN2_CORE_CLK_FREQ,
 	GEN3_CORE_CLK_FREQ,
@@ -459,7 +460,11 @@ static irqreturn_t tegra_pcie_ep_irq_thread(int irq, void *arg)
 
 	speed = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA) &
 		PCI_EXP_LNKSTA_CLS;
-	clk_set_rate(pcie->core_clk, pcie_gen_freq[speed - 1]);
+
+	if (speed >= ARRAY_SIZE(pcie_gen_freq))
+		speed = 0;
+
+	clk_set_rate(pcie->core_clk, pcie_gen_freq[speed]);
 
 	if (pcie->of_data->has_ltr_req_fix)
 		return IRQ_HANDLED;
@@ -1020,7 +1025,11 @@ static int tegra_pcie_dw_start_link(struct dw_pcie *pci)
 
 	speed = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA) &
 		PCI_EXP_LNKSTA_CLS;
-	clk_set_rate(pcie->core_clk, pcie_gen_freq[speed - 1]);
+
+	if (speed >= ARRAY_SIZE(pcie_gen_freq))
+		speed = 0;
+
+	clk_set_rate(pcie->core_clk, pcie_gen_freq[speed]);
 
 	tegra_pcie_enable_interrupts(pp);
 
Signature · 1 line
2.39.2
author maintainer
#8 lore raw

PATCH 9/58 ALSA: emu10k1: roll up loops in DSP setup code for Audigy

In reply to Sasha Levin

From: Oswald Buddenhagen <[email protected]>

[ Upstream commit 8cabf83c7aa54530e699be56249fb44f9505c4f3 ]

There is no apparent reason for the massive code duplication.

---
 sound/pci/emu10k1/emufx.c | 112 +++-----------------------------------
 1 file changed, 9 insertions(+), 103 deletions(-)
sound/pci/emu10k1/emufx.c +9 -103
index 3f64ccab0e632..fba19a854f27a 100644
@@ -1559,14 +1559,8 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 	gpr += 2;
 
 	/* Master volume (will be renamed later) */
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+0+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+0+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+1+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+1+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+2+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+2+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+3+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+3+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+4+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+4+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+5+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+5+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+6+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+6+SND_EMU10K1_PLAYBACK_CHANNELS));
-	A_OP(icode, &ptr, iMAC0, A_GPR(playback+7+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+7+SND_EMU10K1_PLAYBACK_CHANNELS));
+	for (z = 0; z < 8; z++)
+		A_OP(icode, &ptr, iMAC0, A_GPR(playback+z+SND_EMU10K1_PLAYBACK_CHANNELS), A_C_00000000, A_GPR(gpr), A_GPR(playback+z+SND_EMU10K1_PLAYBACK_CHANNELS));
 	snd_emu10k1_init_mono_control(&controls[nctl++], "Wave Master Playback Volume", gpr, 0);
 	gpr += 2;
 
@@ -1653,102 +1647,14 @@ A_OP(icode, &ptr, iMAC0, A_GPR(var), A_GPR(var), A_GPR(vol), A_EXTIN(input))
 			dev_dbg(emu->card->dev, "emufx.c: gpr=0x%x, tmp=0x%x\n",
 			       gpr, tmp);
 			*/
-			/* For the EMU1010: How to get 32bit values from the DSP. High 16bits into L, low 16bits into R. */
-			/* A_P16VIN(0) is delayed by one sample,
-			 * so all other A_P16VIN channels will need to also be delayed
-			 */
-			/* Left ADC in. 1 of 2 */
 			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_P16VIN(0x0), A_FXBUS2(0) );
-			/* Right ADC in 1 of 2 */
-			gpr_map[gpr++] = 0x00000000;
-			/* Delaying by one sample: instead of copying the input
-			 * value A_P16VIN to output A_FXBUS2 as in the first channel,
-			 * we use an auxiliary register, delaying the value by one
-			 * sample
-			 */
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(2) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x1), A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(4) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x2), A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(6) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x3), A_C_00000000, A_C_00000000);
-			/* For 96kHz mode */
-			/* Left ADC in. 2 of 2 */
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(0x8) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x4), A_C_00000000, A_C_00000000);
-			/* Right ADC in 2 of 2 */
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(0xa) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x5), A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(0xc) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x6), A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr - 1), A_FXBUS2(0xe) );
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x7), A_C_00000000, A_C_00000000);
-			/* Pavel Hofman - we still have voices, A_FXBUS2s, and
-			 * A_P16VINs available -
-			 * let's add 8 more capture channels - total of 16
-			 */
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x10));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x8),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x12));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0x9),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x14));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xa),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x16));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xb),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x18));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xc),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x1a));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xd),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x1c));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xe),
-			     A_C_00000000, A_C_00000000);
-			gpr_map[gpr++] = 0x00000000;
-			snd_emu10k1_audigy_dsp_convert_32_to_2x16(icode, &ptr, tmp,
-								  bit_shifter16,
-								  A_GPR(gpr - 1),
-								  A_FXBUS2(0x1e));
-			A_OP(icode, &ptr, iACC3, A_GPR(gpr - 1), A_P16VIN(0xf),
-			     A_C_00000000, A_C_00000000);
+			/* A_P16VIN(0) is delayed by one sample, so all other A_P16VIN channels
+			 * will need to also be delayed; we use an auxiliary register for that. */
+			for (z = 1; z < 0x10; z++) {
+				snd_emu10k1_audigy_dsp_convert_32_to_2x16( icode, &ptr, tmp, bit_shifter16, A_GPR(gpr), A_FXBUS2(z * 2) );
+				A_OP(icode, &ptr, iACC3, A_GPR(gpr), A_P16VIN(z), A_C_00000000, A_C_00000000);
+				gpr_map[gpr++] = 0x00000000;
+			}
 		}
 
 #if 0
Signature · 1 line
2.39.2
author maintainer
#19 lore raw

PATCH 20/58 RDMA/mana_ib: Use v2 version of cfg_rx_steer_req to enable RX coalescing

In reply to Sasha Levin

From: Long Li <[email protected]>

[ Upstream commit 2145328515c8fa9b8a9f7889250bc6c032f2a0e6 ]

With RX coalescing, one CQE entry can be used to indicate multiple packets on the receive queue. This saves processing time and PCI bandwidth over the CQ.

The MANA Ethernet driver also uses the v2 version of the protocol. It doesn't use RX coalescing and its behavior is not changed.

---
 drivers/infiniband/hw/mana/qp.c               | 5 ++++-
 drivers/net/ethernet/microsoft/mana/mana_en.c | 5 ++++-
 include/net/mana/mana.h                       | 4 +++-
 3 files changed, 11 insertions(+), 3 deletions(-)
drivers/infiniband/hw/mana/qp.c +4 -1
index 54b61930a7fdb..4b3b5b274e849 100644
@@ -13,7 +13,7 @@ static int mana_ib_cfg_vport_steering(struct mana_ib_dev *dev,
 				      u8 *rx_hash_key)
 {
 	struct mana_port_context *mpc = netdev_priv(ndev);
-	struct mana_cfg_rx_steer_req *req = NULL;
+	struct mana_cfg_rx_steer_req_v2 *req;
 	struct mana_cfg_rx_steer_resp resp = {};
 	mana_handle_t *req_indir_tab;
 	struct gdma_context *gc;
@@ -33,6 +33,8 @@ static int mana_ib_cfg_vport_steering(struct mana_ib_dev *dev,
 	mana_gd_init_req_hdr(&req->hdr, MANA_CONFIG_VPORT_RX, req_buf_size,
 			     sizeof(resp));
 
+	req->hdr.req.msg_version = GDMA_MESSAGE_V2;
+
 	req->vport = mpc->port_handle;
 	req->rx_enable = 1;
 	req->update_default_rxobj = 1;
@@ -46,6 +48,7 @@ static int mana_ib_cfg_vport_steering(struct mana_ib_dev *dev,
 	req->num_indir_entries = MANA_INDIRECT_TABLE_SIZE;
 	req->indir_tab_offset = sizeof(*req);
 	req->update_indir_tab = true;
+	req->cqe_coalescing_enable = 1;
 
 	req_indir_tab = (mana_handle_t *)(req + 1);
 	/* The ind table passed to the hardware must have
drivers/net/ethernet/microsoft/mana/mana_en.c +4 -1
index d907727c7b7a5..7f4e861e398e6 100644
@@ -972,7 +972,7 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 				   bool update_tab)
 {
 	u16 num_entries = MANA_INDIRECT_TABLE_SIZE;
-	struct mana_cfg_rx_steer_req *req = NULL;
+	struct mana_cfg_rx_steer_req_v2 *req;
 	struct mana_cfg_rx_steer_resp resp = {};
 	struct net_device *ndev = apc->ndev;
 	mana_handle_t *req_indir_tab;
@@ -987,6 +987,8 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 	mana_gd_init_req_hdr(&req->hdr, MANA_CONFIG_VPORT_RX, req_buf_size,
 			     sizeof(resp));
 
+	req->hdr.req.msg_version = GDMA_MESSAGE_V2;
+
 	req->vport = apc->port_handle;
 	req->num_indir_entries = num_entries;
 	req->indir_tab_offset = sizeof(*req);
@@ -996,6 +998,7 @@ static int mana_cfg_vport_steering(struct mana_port_context *apc,
 	req->update_hashkey = update_key;
 	req->update_indir_tab = update_tab;
 	req->default_rxobj = apc->default_rxobj;
+	req->cqe_coalescing_enable = 0;
 
 	if (update_key)
 		memcpy(&req->hashkey, apc->hashkey, MANA_HASH_KEY_SIZE);
include/net/mana/mana.h +3 -1
index 9eef199728454..024ad8ddb27e5 100644
@@ -579,7 +579,7 @@ struct mana_fence_rq_resp {
 }; /* HW DATA */
 
 /* Configure vPort Rx Steering */
-struct mana_cfg_rx_steer_req {
+struct mana_cfg_rx_steer_req_v2 {
 	struct gdma_req_hdr hdr;
 	mana_handle_t vport;
 	u16 num_indir_entries;
@@ -592,6 +592,8 @@ struct mana_cfg_rx_steer_req {
 	u8 reserved;
 	mana_handle_t default_rxobj;
 	u8 hashkey[MANA_HASH_KEY_SIZE];
+	u8 cqe_coalescing_enable;
+	u8 reserved2[7];
 }; /* HW DATA */
 
 struct mana_cfg_rx_steer_resp {
Signature · 1 line
2.39.2
author maintainer
#12 lore raw

PATCH 13/58 ASoC: cs35l56: Move DSP part string generation so that it is done only once

In reply to Sasha Levin

From: Simon Trimmer <[email protected]>

[ Upstream commit 608f1b0dbddec6b2fd766c10bcce2b651995e936 ]

Each time we go through dsp_work() it does a devm_kasprintf() to allocate memory to hold the part name string. It's not strictly a memory leak because devm will free it all if the driver is removed. But we keep allocating more and more memory to hold the same string.

Move the allocation so that it is performed after the version and secured state information is gathered and handle allocation errors.

---
 sound/soc/codecs/cs35l56.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
sound/soc/codecs/cs35l56.c +6 -6
index e0d2b9bb23262..f3fee448d759e 100644
@@ -834,12 +834,6 @@ static void cs35l56_dsp_work(struct work_struct *work)
 	if (!cs35l56->init_done)
 		return;
 
-	cs35l56->dsp.part = devm_kasprintf(cs35l56->dev, GFP_KERNEL, "cs35l56%s-%02x",
-					   cs35l56->secured ? "s" : "", cs35l56->rev);
-
-	if (!cs35l56->dsp.part)
-		return;
-
 	pm_runtime_get_sync(cs35l56->dev);
 
 	/*
@@ -1505,6 +1499,12 @@ int cs35l56_init(struct cs35l56_private *cs35l56)
 	dev_info(cs35l56->dev, "Cirrus Logic CS35L56%s Rev %02X OTP%d\n",
 		 cs35l56->secured ? "s" : "", cs35l56->rev, otpid);
 
+	/* Populate the DSP information with the revision and security state */
+	cs35l56->dsp.part = devm_kasprintf(cs35l56->dev, GFP_KERNEL, "cs35l56%s-%02x",
+					   cs35l56->secured ? "s" : "", cs35l56->rev);
+	if (!cs35l56->dsp.part)
+		return -ENOMEM;
+
 	/* Wake source and *_BLOCKED interrupts default to unmasked, so mask them */
 	regmap_write(cs35l56->regmap, CS35L56_IRQ1_MASK_20, 0xffffffff);
 	regmap_update_bits(cs35l56->regmap, CS35L56_IRQ1_MASK_1,
Signature · 1 line
2.39.2
author maintainer
#16 lore raw

PATCH 17/58 arm64: dts: qcom: ipq5332: add QFPROM node

In reply to Sasha Levin

From: Kathiravan T <[email protected]>

[ Upstream commit 2f34a2aa4c88f4882e3c5df8c9b78f8bbd3f564f ]

IPQ5332 has efuse region to determine the various HW quirks. Lets add the initial support and the individual fuses will be added as they are required.

---
 arch/arm64/boot/dts/qcom/ipq5332.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)
arch/arm64/boot/dts/qcom/ipq5332.dtsi +7 -0
index af4d97143bcf5..c2d6cc65a323a 100644
@@ -135,6 +135,13 @@ soc@0 {
 		#size-cells = <1>;
 		ranges = <0 0 0 0xffffffff>;
 
+		qfprom: efuse@a4000 {
+			compatible = "qcom,ipq5332-qfprom", "qcom,qfprom";
+			reg = <0x000a4000 0x721>;
+			#address-cells = <1>;
+			#size-cells = <1>;
+		};
+
 		rng: rng@e3000 {
 			compatible = "qcom,prng-ee";
 			reg = <0x000e3000 0x1000>;
Signature · 1 line
2.39.2
author maintainer
#18 lore raw

PATCH 19/58 ASoC: amd: vangogh: Add check for acp config flags in vangogh platform

In reply to Sasha Levin

From: Venkata Prasad Potturu <[email protected]>

[ Upstream commit e89f45edb747ed88e97a5771dd6d3dd1eb517873 ]

We have SOF and generic ACP support enabled for Vangogh platform on some machines. Since we have same PCI id used for probing, add check for machine configuration flag to avoid conflict with newer pci drivers. Such machine flag has been initialized via dmi match on few Vangogh based machines. If no flag is specified probe and register older platform device.

---
 sound/soc/amd/vangogh/acp5x.h     | 2 ++
 sound/soc/amd/vangogh/pci-acp5x.c | 7 ++++++-
 2 files changed, 8 insertions(+), 1 deletion(-)
sound/soc/amd/vangogh/acp5x.h +2 -0
index bd9f1c5684d17..ac1936a8c43ff 100644
@@ -147,6 +147,8 @@ static inline void acp_writel(u32 val, void __iomem *base_addr)
 	writel(val, base_addr - ACP5x_PHY_BASE_ADDRESS);
 }
 
+int snd_amd_acp_find_config(struct pci_dev *pci);
+
 static inline u64 acp_get_byte_count(struct i2s_stream_instance *rtd,
 				     int direction)
 {
sound/soc/amd/vangogh/pci-acp5x.c +6 -1
index e0df17c88e8e0..c4634a8a17cdc 100644
@@ -125,10 +125,15 @@ static int snd_acp5x_probe(struct pci_dev *pci,
 {
 	struct acp5x_dev_data *adata;
 	struct platform_device_info pdevinfo[ACP5x_DEVS];
-	unsigned int irqflags;
+	unsigned int irqflags, flag;
 	int ret, i;
 	u32 addr, val;
 
+	/* Return if acp config flag is defined */
+	flag = snd_amd_acp_find_config(pci);
+	if (flag)
+		return -ENODEV;
+
 	irqflags = IRQF_SHARED;
 	if (pci->revision != 0x50)
 		return -ENODEV;
Signature · 1 line
2.39.2
author maintainer
#20 lore raw

PATCH 21/58 ARM: dts: imx6dl: prtrvt, prtvt7, prti6q, prtwd2: fix USB related warnings

In reply to Sasha Levin

From: Oleksij Rempel <[email protected]>

[ Upstream commit 1d14bd943fa2bbdfda1efbcc080b298fed5f1803 ]

Fix USB-related warnings in prtrvt, prtvt7, prti6q and prtwd2 device trees by disabling unused usbphynop1 and usbphynop2 USB PHYs and providing proper configuration for the over-current detection. This fixes the following warnings with the current kernel:
usb_phy_generic usbphynop1: dummy supplies not allowed for exclusive requests usb_phy_generic usbphynop2: dummy supplies not allowed for exclusive requests imx_usb 2184200.usb: No over current polarity defined

By the way, fix over-current detection on usbotg port for prtvt7, prti6q and prtwd2 boards. Only prtrvt do not have OC on USB OTG port.

---
 arch/arm/boot/dts/imx6dl-prtrvt.dts   |  4 ++++
 arch/arm/boot/dts/imx6qdl-prti6q.dtsi | 11 ++++++++++-
 2 files changed, 14 insertions(+), 1 deletion(-)
arch/arm/boot/dts/imx6dl-prtrvt.dts +4 -0
index 56bb1ca56a2df..36b031236e475 100644
@@ -124,6 +124,10 @@ &usbh1 {
 	status = "disabled";
 };
 
+&usbotg {
+	disable-over-current;
+};
+
 &vpu {
 	status = "disabled";
 };
arch/arm/boot/dts/imx6qdl-prti6q.dtsi +10 -1
index f0db0d4471f40..36f84f4da6b0d 100644
@@ -69,6 +69,7 @@ &usbh1 {
 	vbus-supply = <&reg_usb_h1_vbus>;
 	phy_type = "utmi";
 	dr_mode = "host";
+	disable-over-current;
 	status = "okay";
 };
 
@@ -78,10 +79,18 @@ &usbotg {
 	pinctrl-0 = <&pinctrl_usbotg>;
 	phy_type = "utmi";
 	dr_mode = "host";
-	disable-over-current;
+	over-current-active-low;
 	status = "okay";
 };
 
+&usbphynop1 {
+	status = "disabled";
+};
+
+&usbphynop2 {
+	status = "disabled";
+};
+
 &usdhc1 {
 	pinctrl-names = "default";
 	pinctrl-0 = <&pinctrl_usdhc1>;
Signature · 1 line
2.39.2
author maintainer
#9 lore raw

PATCH 10/58 ASoC: Intel: sof_sdw: add quirk for MTL RVP

In reply to Sasha Levin

From: Bard Liao <[email protected]>

[ Upstream commit 289e1df00e49a229a1c924c059242e759a552f01 ]

We should use RT711_JD2_100K for on board rt711.

---
 sound/soc/intel/boards/sof_sdw.c | 8 ++++++++
 1 file changed, 8 insertions(+)
sound/soc/intel/boards/sof_sdw.c +8 -0
index 144f082c63fda..db99f7fefdfc7 100644
@@ -415,6 +415,14 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 		},
 		.driver_data = (void *)(RT711_JD1 | SOF_SDW_TGL_HDMI),
 	},
+	{
+		.callback = sof_sdw_quirk_cb,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Intel Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Meteor Lake Client Platform"),
+		},
+		.driver_data = (void *)(RT711_JD2_100K),
+	},
 	{}
 };
 
Signature · 1 line
2.39.2
author maintainer
#10 lore raw

PATCH 11/58 ASoC: Intel: sof_sdw: add quirk for LNL RVP

In reply to Sasha Levin

From: Peter Ujfalusi <[email protected]>

[ Upstream commit dfe25fea968dc4884e12d471c8263f0f611b380a ]

We should use RT711_JD2_100K for on board rt711

---
 sound/soc/intel/boards/sof_sdw.c | 9 +++++++++
 1 file changed, 9 insertions(+)
sound/soc/intel/boards/sof_sdw.c +9 -0
index db99f7fefdfc7..342d4f23a8dc6 100644
@@ -423,6 +423,15 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 		},
 		.driver_data = (void *)(RT711_JD2_100K),
 	},
+	/* LunarLake devices */
+	{
+		.callback = sof_sdw_quirk_cb,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Intel Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Lunar Lake Client Platform"),
+		},
+		.driver_data = (void *)(RT711_JD2_100K),
+	},
 	{}
 };
 
Signature · 1 line
2.39.2
author maintainer
#21 lore raw

PATCH 22/58 ASoC: Intel: sof_sdw_rt_sdca_jack_common: test SOF_JACK_JDSRC in _exit

In reply to Sasha Levin

From: Bard Liao <[email protected]>

[ Upstream commit 526a1876fc48e2d0c0ea8ad63b58bdb2cc13047f ]

if (!SOF_RT711_JDSRC(sof_sdw_quirk)) is tested in rt711_sdca_add_codec_ device_props(), and we don't add software node to the device if jack source is not set. We need to do the same test in sof_sdw_rt711_sdca_exit(), and avoid removing software node if jack source is not set.

---
 sound/soc/intel/boards/sof_sdw_rt711_sdca.c | 3 +++
 1 file changed, 3 insertions(+)
sound/soc/intel/boards/sof_sdw_rt711_sdca.c +3 -0
index 7f16304d025be..cf8b9793fe0e5 100644
@@ -143,6 +143,9 @@ int sof_sdw_rt711_sdca_exit(struct snd_soc_card *card, struct snd_soc_dai_link *
 	if (!ctx->headset_codec_dev)
 		return 0;
 
+	if (!SOF_RT711_JDSRC(sof_sdw_quirk))
+		return 0;
+
 	device_remove_software_node(ctx->headset_codec_dev);
 	put_device(ctx->headset_codec_dev);
 
Signature · 1 line
2.39.2
author maintainer
#14 lore raw

PATCH 15/58 ASoC: SOF: amd: Add pci revision id check

In reply to Sasha Levin

From: Venkata Prasad Potturu <[email protected]>

[ Upstream commit 1d4a84632b90d88316986b05bcdfe715399a33db ]

Add pci revision id check for renoir and rembrandt platforms.

---
 sound/soc/sof/amd/acp.h     | 3 +++
 sound/soc/sof/amd/pci-rmb.c | 3 +++
 sound/soc/sof/amd/pci-rn.c  | 3 +++
 3 files changed, 9 insertions(+)
sound/soc/sof/amd/acp.h +3 -0
index 1c535cc6c3a95..dc624f727aa37 100644
@@ -55,6 +55,9 @@
 
 #define ACP_DSP_TO_HOST_IRQ			0x04
 
+#define ACP_RN_PCI_ID				0x01
+#define ACP_RMB_PCI_ID				0x6F
+
 #define HOST_BRIDGE_CZN				0x1630
 #define HOST_BRIDGE_RMB				0x14B5
 #define ACP_SHA_STAT				0x8000
sound/soc/sof/amd/pci-rmb.c +3 -0
index eaf70ea6e556e..58b3092425f1a 100644
@@ -65,6 +65,9 @@ static int acp_pci_rmb_probe(struct pci_dev *pci, const struct pci_device_id *pc
 {
 	unsigned int flag;
 
+	if (pci->revision != ACP_RMB_PCI_ID)
+		return -ENODEV;
+
 	flag = snd_amd_acp_find_config(pci);
 	if (flag != FLAG_AMD_SOF && flag != FLAG_AMD_SOF_ONLY_DMIC)
 		return -ENODEV;
sound/soc/sof/amd/pci-rn.c +3 -0
index 4809cb644152b..7409e21ce5aa7 100644
@@ -65,6 +65,9 @@ static int acp_pci_rn_probe(struct pci_dev *pci, const struct pci_device_id *pci
 {
 	unsigned int flag;
 
+	if (pci->revision != ACP_RN_PCI_ID)
+		return -ENODEV;
+
 	flag = snd_amd_acp_find_config(pci);
 	if (flag != FLAG_AMD_SOF && flag != FLAG_AMD_SOF_ONLY_DMIC)
 		return -ENODEV;
Signature · 1 line
2.39.2
author maintainer
#23 lore raw

PATCH 24/58 ASoC: Intel: sof_sdw: Add support for Rex soundwire

In reply to Sasha Levin

From: Uday M Bhat <[email protected]>

[ Upstream commit 164e5dc17525181c05563f0a06796f1a363801d5 ]

Add rex entry in the soundwire quirk table

---
 sound/soc/intel/boards/sof_sdw.c | 8 ++++++++
 1 file changed, 8 insertions(+)
sound/soc/intel/boards/sof_sdw.c +8 -0
index 1238a13744e2e..f0de85edb1de6 100644
@@ -433,6 +433,14 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 		},
 		.driver_data = (void *)(RT711_JD2_100K),
 	},
+	{
+		.callback = sof_sdw_quirk_cb,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Google"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Rex"),
+		},
+		.driver_data = (void *)(SOF_SDW_PCH_DMIC),
+	},
 	/* LunarLake devices */
 	{
 		.callback = sof_sdw_quirk_cb,
Signature · 1 line
2.39.2
author maintainer
#26 lore raw

PATCH 27/58 quota: fix warning in dqgrab()

In reply to Sasha Levin

From: Ye Bin <[email protected]>

[ Upstream commit d6a95db3c7ad160bc16b89e36449705309b52bcb ]

There's issue as follows when do fault injection:

WARNING: CPU: 1 PID: 14870 at include/linux/quotaops.h:51 dquot_disable+0x13b7/0x18c0
Modules linked in:
CPU: 1 PID: 14870 Comm: fsconfig Not tainted 6.3.0-next-20230505-00006-g5107a9c821af-dirty #541
RIP: 0010:dquot_disable+0x13b7/0x18c0
RSP: 0018:ffffc9000acc79e0 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff88825e41b980
RDX: 0000000000000000 RSI: ffff88825e41b980 RDI: 0000000000000002
RBP: ffff888179f68000 R08: ffffffff82087ca7 R09: 0000000000000000
R10: 0000000000000001 R11: ffffed102f3ed026 R12: ffff888179f68130
R13: ffff888179f68110 R14: dffffc0000000000 R15: ffff888179f68118
FS:  00007f450a073740(0000) GS:ffff88882fc00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffe96f2efd8 CR3: 000000025c8ad000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 dquot_load_quota_sb+0xd53/0x1060
 dquot_resume+0x172/0x230
 ext4_reconfigure+0x1dc6/0x27b0
 reconfigure_super+0x515/0xa90
 __x64_sys_fsconfig+0xb19/0xd20
 do_syscall_64+0x39/0xb0
 entry_SYSCALL_64_after_hwframe+0x63/0xcd

Above issue may happens as follows:
ProcessA ProcessB ProcessC
sys_fsconfig
vfs_fsconfig_locked
reconfigure_super

     ext4_remount
      dquot_suspend -> suspend all type quota

                 sys_fsconfig
                  vfs_fsconfig_locked
                    reconfigure_super
                     ext4_remount
                      dquot_resume
                       ret = dquot_load_quota_sb
                        add_dquot_ref
                                           do_open  -> open file O_RDWR
                                            vfs_open
                                             do_dentry_open
                                              get_write_access
                                               atomic_inc_unless_negative(&inode->i_writecount)
                                              ext4_file_open
                                               dquot_file_open
                                                dquot_initialize
                                                  __dquot_initialize
                                                   dqget
						    atomic_inc(&dquot->dq_count);

                          __dquot_initialize
                           __dquot_initialize
                            dqget
                             if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags))
                               ext4_acquire_dquot
			        -> Return error DQ_ACTIVE_B flag isn't set
                         dquot_disable
			  invalidate_dquots
			   if (atomic_read(&dquot->dq_count))
	                    dqgrab
			     WARN_ON_ONCE(!test_bit(DQ_ACTIVE_B, &dquot->dq_flags))
	                      -> Trigger warning

In the above scenario, 'dquot->dq_flags' has no DQ_ACTIVE_B is normal when dqgrab().
To solve above issue just replace the dqgrab() use in invalidate_dquots() with atomic_inc(&dquot->dq_count).

Message-Id: <[email protected]>

---
 fs/quota/dquot.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/quota/dquot.c +1 -1
index 6beceed34e089..e3e4f40476579 100644
@@ -555,7 +555,7 @@ static void invalidate_dquots(struct super_block *sb, int type)
 			continue;
 		/* Wait for dquot users */
 		if (atomic_read(&dquot->dq_count)) {
-			dqgrab(dquot);
+			atomic_inc(&dquot->dq_count);
 			spin_unlock(&dq_list_lock);
 			/*
 			 * Once dqput() wakes us up, we know it's time to free
Signature · 1 line
2.39.2
author maintainer
#24 lore raw

PATCH 25/58 iopoll: Call cpu_relax() in busy loops

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit b407460ee99033503993ac7437d593451fcdfe44 ]

It is considered good practice to call cpu_relax() in busy loops, see Documentation/process/volatile-considered-harmful.rst. This can not only lower CPU power consumption or yield to a hyperthreaded twin processor, but also allows an architecture to mitigate hardware issues (e.g. ARM Erratum 754327 for Cortex-A9 prior to r2p0) in the architecture-specific cpu_relax() implementation.

In addition, cpu_relax() is also a compiler barrier. It is not immediately obvious that the @op argument "function" will result in an actual function call (e.g. in case of inlining).

Where a function call is a C sequence point, this is lost on inlining. Therefore, with agressive enough optimization it might be possible for the compiler to hoist the:

        (val) = op(args);

"load" out of the loop because it doesn't see the value changing. The addition of cpu_relax() would inhibit this.

As the iopoll helpers lack calls to cpu_relax(), people are sometimes reluctant to use them, and may fall back to open-coded polling loops (including cpu_relax() calls) instead.

Fix this by adding calls to cpu_relax() to the iopoll helpers:
- For the non-atomic case, it is sufficient to call cpu_relax() in case of a zero sleep-between-reads value, as a call to usleep_range() is a safe barrier otherwise. However, it doesn't hurt to add the call regardless, for simplicity, and for similarity with the atomic case below.
- For the atomic case, cpu_relax() must be called regardless of the sleep-between-reads value, as there is no guarantee all architecture-specific implementations of udelay() handle this.

---
 include/linux/iopoll.h | 2 ++
 1 file changed, 2 insertions(+)
include/linux/iopoll.h +2 -0
index 2c8860e406bd8..0417360a6db9b 100644
@@ -53,6 +53,7 @@
 		} \
 		if (__sleep_us) \
 			usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
+		cpu_relax(); \
 	} \
 	(cond) ? 0 : -ETIMEDOUT; \
 })
@@ -95,6 +96,7 @@
 		} \
 		if (__delay_us) \
 			udelay(__delay_us); \
+		cpu_relax(); \
 	} \
 	(cond) ? 0 : -ETIMEDOUT; \
 })
Signature · 1 line
2.39.2
author maintainer
#22 lore raw

PATCH 23/58 ASoC: Intel: sof_sdw: add quick for Dell SKU 0BDA

In reply to Sasha Levin

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit 3daf02819ac3fd8d7605804a00213cf123ac880d ]

The SKU numbering isn't quite consistent with the existing RaptorLake SKUs but the PCI ID is definitively RaptorLake.

---
 sound/soc/intel/boards/sof_sdw.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
sound/soc/intel/boards/sof_sdw.c +10 -0
index 342d4f23a8dc6..1238a13744e2e 100644
@@ -367,6 +367,16 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 					RT711_JD2),
 	},
 	/* RaptorLake devices */
+	{
+		.callback = sof_sdw_quirk_cb,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc"),
+			DMI_EXACT_MATCH(DMI_PRODUCT_SKU, "0BDA")
+		},
+		.driver_data = (void *)(SOF_SDW_TGL_HDMI |
+					RT711_JD2 |
+					SOF_SDW_FOUR_SPK),
+	},
 	{
 		.callback = sof_sdw_quirk_cb,
 		.matches = {
Signature · 1 line
2.39.2
author maintainer
#25 lore raw

PATCH 26/58 quota: Properly disable quotas when add_dquot_ref() fails

In reply to Sasha Levin

From: Jan Kara <[email protected]>

[ Upstream commit 6a4e3363792e30177cc3965697e34ddcea8b900b ]

When add_dquot_ref() fails (usually due to IO error or ENOMEM), we want to disable quotas we are trying to enable. However dquot_disable() call was passed just the flags we are enabling so in case flags == DQUOT_USAGE_ENABLED dquot_disable() call will just fail with EINVAL instead of properly disabling quotas. Fix the problem by always passing DQUOT_LIMITS_ENABLED | DQUOT_USAGE_ENABLED to dquot_disable() in this case.

Message-Id: <[email protected]>

---
 fs/quota/dquot.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/quota/dquot.c +2 -1
index ffd40dc3e4e99..6beceed34e089 100644
@@ -2420,7 +2420,8 @@ int dquot_load_quota_sb(struct super_block *sb, int type, int format_id,
 
 	error = add_dquot_ref(sb, type);
 	if (error)
-		dquot_disable(sb, type, flags);
+		dquot_disable(sb, type,
+			      DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);
 
 	return error;
 out_fmt:
Signature · 1 line
2.39.2
author maintainer
#27 lore raw

PATCH 28/58 ALSA: hda: Add Loongson LS7A HD-Audio support

In reply to Sasha Levin

From: Yanteng Si <[email protected]>

[ Upstream commit 28bd137a3c8e105587ba8c55b68ef43b519b270f ]

Add the new PCI ID 0x0014 0x7a07 and the new PCI ID 0x0014 0x7a37 Loongson HDA controller.

---
 include/linux/pci_ids.h    | 3 +++
 sound/hda/hdac_device.c    | 1 +
 sound/pci/hda/hda_intel.c  | 7 +++++++
 sound/pci/hda/patch_hdmi.c | 1 +
 4 files changed, 12 insertions(+)
include/linux/pci_ids.h +3 -0
index 95f33dadb2be2..c0c4ca8e28510 100644
@@ -158,6 +158,9 @@
 
 #define PCI_VENDOR_ID_LOONGSON		0x0014
 
+#define PCI_DEVICE_ID_LOONGSON_HDA      0x7a07
+#define PCI_DEVICE_ID_LOONGSON_HDMI     0x7a37
+
 #define PCI_VENDOR_ID_TTTECH		0x0357
 #define PCI_DEVICE_ID_TTTECH_MC322	0x000a
 
sound/hda/hdac_device.c +1 -0
index 6c043fbd606f1..bbf7bcdb449a8 100644
@@ -645,6 +645,7 @@ struct hda_vendor_id {
 };
 
 static const struct hda_vendor_id hda_vendor_ids[] = {
+	{ 0x0014, "Loongson" },
 	{ 0x1002, "ATI" },
 	{ 0x1013, "Cirrus Logic" },
 	{ 0x1057, "Motorola" },
sound/pci/hda/hda_intel.c +7 -0
index 3226691ac923c..9c353dc7740c4 100644
@@ -237,6 +237,7 @@ enum {
 	AZX_DRIVER_CTHDA,
 	AZX_DRIVER_CMEDIA,
 	AZX_DRIVER_ZHAOXIN,
+	AZX_DRIVER_LOONGSON,
 	AZX_DRIVER_GENERIC,
 	AZX_NUM_DRIVERS, /* keep this as last entry */
 };
@@ -360,6 +361,7 @@ static const char * const driver_short_names[] = {
 	[AZX_DRIVER_CTHDA] = "HDA Creative",
 	[AZX_DRIVER_CMEDIA] = "HDA C-Media",
 	[AZX_DRIVER_ZHAOXIN] = "HDA Zhaoxin",
+	[AZX_DRIVER_LOONGSON] = "HDA Loongson",
 	[AZX_DRIVER_GENERIC] = "HD-Audio Generic",
 };
 
@@ -2809,6 +2811,11 @@ static const struct pci_device_id azx_ids[] = {
 	  .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI },
 	/* Zhaoxin */
 	{ PCI_DEVICE(0x1d17, 0x3288), .driver_data = AZX_DRIVER_ZHAOXIN },
+	/* Loongson HDAudio*/
+	{PCI_DEVICE(PCI_VENDOR_ID_LOONGSON, PCI_DEVICE_ID_LOONGSON_HDA),
+	  .driver_data = AZX_DRIVER_LOONGSON },
+	{PCI_DEVICE(PCI_VENDOR_ID_LOONGSON, PCI_DEVICE_ID_LOONGSON_HDMI),
+	  .driver_data = AZX_DRIVER_LOONGSON },
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, azx_ids);
sound/pci/hda/patch_hdmi.c +1 -0
index 5c0b1a09fd57c..260d3e64f6589 100644
@@ -4505,6 +4505,7 @@ static int patch_gf_hdmi(struct hda_codec *codec)
  * patch entries
  */
 static const struct hda_device_id snd_hda_id_hdmi[] = {
+HDA_CODEC_ENTRY(0x00147a47, "Loongson HDMI",	patch_generic_hdmi),
 HDA_CODEC_ENTRY(0x1002793c, "RS600 HDMI",	patch_atihdmi),
 HDA_CODEC_ENTRY(0x10027919, "RS600 HDMI",	patch_atihdmi),
 HDA_CODEC_ENTRY(0x1002791a, "RS690/780 HDMI",	patch_atihdmi),
Signature · 1 line
2.39.2
author maintainer
#30 lore raw

PATCH 30/58 dma-remap: use kvmalloc_array/kvfree for larger dma memory remap

In reply to Sasha Levin

From: gaoxu <[email protected]>

[ Upstream commit 51ff97d54f02b4444dfc42e380ac4c058e12d5dd ]

If dma_direct_alloc() alloc memory in size of 64MB, the inner function dma_common_contiguous_remap() will allocate 128KB memory by invoking the function kmalloc_array(). and the kmalloc_array seems to fail to try to allocate 128KB mem.

Call trace:
[14977.928623] qcrosvm: page allocation failure: order:5, mode:0x40cc0
[14977.928638] dump_backtrace.cfi_jt+0x0/0x8
[14977.928647] dump_stack_lvl+0x80/0xb8
[14977.928652] warn_alloc+0x164/0x200
[14977.928657] __alloc_pages_slowpath+0x9f0/0xb4c
[14977.928660] __alloc_pages+0x21c/0x39c
[14977.928662] kmalloc_order+0x48/0x108
[14977.928666] kmalloc_order_trace+0x34/0x154
[14977.928668] __kmalloc+0x548/0x7e4
[14977.928673] dma_direct_alloc+0x11c/0x4f8
[14977.928678] dma_alloc_attrs+0xf4/0x138
[14977.928680] gh_vm_ioctl_set_fw_name+0x3c4/0x610 [gunyah]
[14977.928698] gh_vm_ioctl+0x90/0x14c [gunyah]
[14977.928705] __arm64_sys_ioctl+0x184/0x210

work around by doing kvmalloc_array instead.

---
 kernel/dma/remap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/dma/remap.c +2 -2
index b4526668072e7..27596f3b4aef3 100644
@@ -43,13 +43,13 @@ void *dma_common_contiguous_remap(struct page *page, size_t size,
 	void *vaddr;
 	int i;
 
-	pages = kmalloc_array(count, sizeof(struct page *), GFP_KERNEL);
+	pages = kvmalloc_array(count, sizeof(struct page *), GFP_KERNEL);
 	if (!pages)
 		return NULL;
 	for (i = 0; i < count; i++)
 		pages[i] = nth_page(page, i);
 	vaddr = vmap(pages, count, VM_DMA_COHERENT, prot);
-	kfree(pages);
+	kvfree(pages);
 
 	return vaddr;
 }
Signature · 1 line
2.39.2
author maintainer
#31 lore raw

PATCH 31/58 accel/habanalabs: add pci health check during heartbeat

In reply to Sasha Levin

From: Ofir Bitton <[email protected]>

[ Upstream commit d8b9cea584661b30305cf341bf9f675dc0a25471 ]

Currently upon a heartbeat failure, we don't know if the failure is due to firmware hang or due to a bad PCI link. Hence, we are reading a PCI config space register with a known value (vendor ID) so we will know which of the two possibilities caused the heartbeat failure.

---
 drivers/accel/habanalabs/common/device.c         | 15 ++++++++++++++-
 drivers/accel/habanalabs/common/habanalabs.h     |  2 ++
 drivers/accel/habanalabs/common/habanalabs_drv.c |  2 --
 3 files changed, 16 insertions(+), 3 deletions(-)
drivers/accel/habanalabs/common/device.c +14 -1
index fabfc501ef543..a39dd346a1678 100644
@@ -981,6 +981,18 @@ static void device_early_fini(struct hl_device *hdev)
 		hdev->asic_funcs->early_fini(hdev);
 }
 
+static bool is_pci_link_healthy(struct hl_device *hdev)
+{
+	u16 vendor_id;
+
+	if (!hdev->pdev)
+		return false;
+
+	pci_read_config_word(hdev->pdev, PCI_VENDOR_ID, &vendor_id);
+
+	return (vendor_id == PCI_VENDOR_ID_HABANALABS);
+}
+
 static void hl_device_heartbeat(struct work_struct *work)
 {
 	struct hl_device *hdev = container_of(work, struct hl_device,
@@ -995,7 +1007,8 @@ static void hl_device_heartbeat(struct work_struct *work)
 		goto reschedule;
 
 	if (hl_device_operational(hdev, NULL))
-		dev_err(hdev->dev, "Device heartbeat failed!\n");
+		dev_err(hdev->dev, "Device heartbeat failed! PCI link is %s\n",
+			is_pci_link_healthy(hdev) ? "healthy" : "broken");
 
 	info.err_type = HL_INFO_FW_HEARTBEAT_ERR;
 	info.event_mask = &event_mask;
drivers/accel/habanalabs/common/habanalabs.h +2 -0
index eaae69a9f8178..7f5d1b6e3fb08 100644
@@ -36,6 +36,8 @@
 struct hl_device;
 struct hl_fpriv;
 
+#define PCI_VENDOR_ID_HABANALABS	0x1da3
+
 /* Use upper bits of mmap offset to store habana driver specific information.
  * bits[63:59] - Encode mmap type
  * bits[45:0]  - mmap offset value
drivers/accel/habanalabs/common/habanalabs_drv.c +0 -2
index d9df64e75f33a..1ec97da3dddb8 100644
@@ -54,8 +54,6 @@ module_param(boot_error_status_mask, ulong, 0444);
 MODULE_PARM_DESC(boot_error_status_mask,
 	"Mask of the error status during device CPU boot (If bitX is cleared then error X is masked. Default all 1's)");
 
-#define PCI_VENDOR_ID_HABANALABS	0x1da3
-
 #define PCI_IDS_GOYA			0x0001
 #define PCI_IDS_GAUDI			0x1000
 #define PCI_IDS_GAUDI_SEC		0x1010
Signature · 1 line
2.39.2
author maintainer
#32 lore raw

PATCH 32/58 accel/habanalabs: fix mem leak in capture user mappings

In reply to Sasha Levin

From: Moti Haimovski <[email protected]>

[ Upstream commit 314a7ffd7c196b27eedd50cb7553029e17789b55 ]

This commit fixes a memory leak caused when clearing the user_mappings info when a new context is opened immediately after user_mapping is captured and a hard reset is performed.

---
 drivers/accel/habanalabs/common/habanalabs_drv.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/accel/habanalabs/common/habanalabs_drv.c +2 -0
index 1ec97da3dddb8..70fb2df9a93b8 100644
@@ -13,6 +13,7 @@
 
 #include <linux/pci.h>
 #include <linux/module.h>
+#include <linux/vmalloc.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/habanalabs.h>
@@ -218,6 +219,7 @@ int hl_device_open(struct inode *inode, struct file *filp)
 
 	hl_debugfs_add_file(hpriv);
 
+	vfree(hdev->captured_err_info.page_fault_info.user_mappings);
 	memset(&hdev->captured_err_info, 0, sizeof(hdev->captured_err_info));
 	atomic_set(&hdev->captured_err_info.cs_timeout.write_enable, 1);
 	hdev->captured_err_info.undef_opcode.write_enable = true;
Signature · 1 line
2.39.2
author maintainer
#29 lore raw

PATCH 29/58 ASoC: SOF: Intel: fix SoundWire/HDaudio mutual exclusion

In reply to Sasha Levin

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit f751b99255cacd9ffe8c4bbf99767ad670cee1f7 ]

The functionality described in Commit 61bef9e68dca ("ASoC: SOF: Intel: hda: enforce exclusion between HDaudio and SoundWire") does not seem to be properly implemented with two issues that need to
be corrected.

a) The test used is incorrect when DisplayAudio codecs are not supported.

b) Conversely when only Display Audio codecs can be found, we do want to start the SoundWire links, if any. That will help add the relevant topologies and machine descriptors, and identify cases where the SoundWire information in ACPI needs to be modified with a quirk.

---
 sound/soc/sof/intel/hda.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
sound/soc/sof/intel/hda.c +11 -1
index 3153e21f100ab..3853582e32e12 100644
@@ -1343,12 +1343,22 @@ static void hda_generic_machine_select(struct snd_sof_dev *sdev,
 			hda_mach->mach_params.dmic_num = dmic_num;
 			pdata->tplg_filename = tplg_filename;
 
-			if (codec_num == 2) {
+			if (codec_num == 2 ||
+			    (codec_num == 1 && !HDA_IDISP_CODEC(bus->codec_mask))) {
 				/*
 				 * Prevent SoundWire links from starting when an external
 				 * HDaudio codec is used
 				 */
 				hda_mach->mach_params.link_mask = 0;
+			} else {
+				/*
+				 * Allow SoundWire links to start when no external HDaudio codec
+				 * was detected. This will not create a SoundWire card but
+				 * will help detect if any SoundWire codec reports as ATTACHED.
+				 */
+				struct sof_intel_hda_dev *hdev = sdev->pdata->hw_pdata;
+
+				hda_mach->mach_params.link_mask = hdev->info.link_mask;
 			}
 
 			*mach = hda_mach;
Signature · 1 line
2.39.2
author maintainer
#37 lore raw

PATCH 37/58 iommu/amd: Introduce Disable IRTE Caching Support

In reply to Sasha Levin

From: Suravee Suthikulpanit <[email protected]>

[ Upstream commit 66419036f68a838c00cbccacd6cb2e99da6e5710 ]

An Interrupt Remapping Table (IRT) stores interrupt remapping configuration for each device. In a normal operation, the AMD IOMMU caches the table to optimize subsequent data accesses. This requires the IOMMU driver to invalidate IRT whenever it updates the table. The invalidation process includes issuing an INVALIDATE_INTERRUPT_TABLE command following by a COMPLETION_WAIT command.

However, there are cases in which the IRT is updated at a high rate. For example, for IOMMU AVIC, the IRTE[IsRun] bit is updated on every vcpu scheduling (i.e. amd_iommu_update_ga()). On system with large amount of vcpus and VFIO PCI pass-through devices, the invalidation process could potentially become a performance bottleneck.

Introducing a new kernel boot option:

amd_iommu=irtcachedis

which disables IRTE caching by setting the IRTCachedis bit in each IOMMU Control register, and bypass the IRT invalidation process.

---
 .../admin-guide/kernel-parameters.txt         |  1 +
 drivers/iommu/amd/amd_iommu_types.h           |  4 +++
 drivers/iommu/amd/init.c                      | 36 +++++++++++++++++++
 3 files changed, 41 insertions(+)
Documentation/admin-guide/kernel-parameters.txt +1 -0
index 9e5bab29685ff..986ac2b73ea22 100644
@@ -323,6 +323,7 @@
 				       option with care.
 			pgtbl_v1     - Use v1 page table for DMA-API (Default).
 			pgtbl_v2     - Use v2 page table for DMA-API.
+			irtcachedis  - Disable Interrupt Remapping Table (IRT) caching.
 
 	amd_iommu_dump=	[HW,X86-64]
 			Enable AMD IOMMU driver option to dump the ACPI table
drivers/iommu/amd/amd_iommu_types.h +4 -0
index 2ddbda3a43746..5a224e244be8a 100644
@@ -174,6 +174,7 @@
 #define CONTROL_GAINT_EN	29
 #define CONTROL_XT_EN		50
 #define CONTROL_INTCAPXT_EN	51
+#define CONTROL_IRTCACHEDIS	59
 #define CONTROL_SNPAVIC_EN	61
 
 #define CTRL_INV_TO_MASK	(7 << CONTROL_INV_TIMEOUT)
@@ -716,6 +717,9 @@ struct amd_iommu {
 	/* if one, we need to send a completion wait command */
 	bool need_sync;
 
+	/* true if disable irte caching */
+	bool irtcachedis_enabled;
+
 	/* Handle for IOMMU core code */
 	struct iommu_device iommu;
 
drivers/iommu/amd/init.c +36 -0
index c2d80a4e5fb06..02846299af0ef 100644
@@ -162,6 +162,7 @@ static int amd_iommu_xt_mode = IRQ_REMAP_XAPIC_MODE;
 static bool amd_iommu_detected;
 static bool amd_iommu_disabled __initdata;
 static bool amd_iommu_force_enable __initdata;
+static bool amd_iommu_irtcachedis;
 static int amd_iommu_target_ivhd_type;
 
 /* Global EFR and EFR2 registers */
@@ -484,6 +485,9 @@ static void iommu_disable(struct amd_iommu *iommu)
 
 	/* Disable IOMMU hardware itself */
 	iommu_feature_disable(iommu, CONTROL_IOMMU_EN);
+
+	/* Clear IRTE cache disabling bit */
+	iommu_feature_disable(iommu, CONTROL_IRTCACHEDIS);
 }
 
 /*
@@ -2710,6 +2714,33 @@ static void iommu_enable_ga(struct amd_iommu *iommu)
 #endif
 }
 
+static void iommu_disable_irtcachedis(struct amd_iommu *iommu)
+{
+	iommu_feature_disable(iommu, CONTROL_IRTCACHEDIS);
+}
+
+static void iommu_enable_irtcachedis(struct amd_iommu *iommu)
+{
+	u64 ctrl;
+
+	if (!amd_iommu_irtcachedis)
+		return;
+
+	/*
+	 * Note:
+	 * The support for IRTCacheDis feature is dertermined by
+	 * checking if the bit is writable.
+	 */
+	iommu_feature_enable(iommu, CONTROL_IRTCACHEDIS);
+	ctrl = readq(iommu->mmio_base +  MMIO_CONTROL_OFFSET);
+	ctrl &= (1ULL << CONTROL_IRTCACHEDIS);
+	if (ctrl)
+		iommu->irtcachedis_enabled = true;
+	pr_info("iommu%d (%#06x) : IRT cache is %s\n",
+		iommu->index, iommu->devid,
+		iommu->irtcachedis_enabled ? "disabled" : "enabled");
+}
+
 static void early_enable_iommu(struct amd_iommu *iommu)
 {
 	iommu_disable(iommu);
@@ -2720,6 +2751,7 @@ static void early_enable_iommu(struct amd_iommu *iommu)
 	iommu_set_exclusion_range(iommu);
 	iommu_enable_ga(iommu);
 	iommu_enable_xt(iommu);
+	iommu_enable_irtcachedis(iommu);
 	iommu_enable(iommu);
 	iommu_flush_all_caches(iommu);
 }
@@ -2770,10 +2802,12 @@ static void early_enable_iommus(void)
 		for_each_iommu(iommu) {
 			iommu_disable_command_buffer(iommu);
 			iommu_disable_event_buffer(iommu);
+			iommu_disable_irtcachedis(iommu);
 			iommu_enable_command_buffer(iommu);
 			iommu_enable_event_buffer(iommu);
 			iommu_enable_ga(iommu);
 			iommu_enable_xt(iommu);
+			iommu_enable_irtcachedis(iommu);
 			iommu_set_device_table(iommu);
 			iommu_flush_all_caches(iommu);
 		}
@@ -3426,6 +3460,8 @@ static int __init parse_amd_iommu_options(char *str)
 			amd_iommu_pgtable = AMD_IOMMU_V1;
 		} else if (strncmp(str, "pgtbl_v2", 8) == 0) {
 			amd_iommu_pgtable = AMD_IOMMU_V2;
+		} else if (strncmp(str, "irtcachedis", 11) == 0) {
+			amd_iommu_irtcachedis = true;
 		} else {
 			pr_notice("Unknown option - '%s'\n", str);
 		}
Signature · 1 line
2.39.2
author maintainer
#38 lore raw

PATCH 38/58 drm/amdgpu: install stub fence into potential unused fence pointers

In reply to Sasha Levin

From: Lang Yu <[email protected]>

[ Upstream commit 187916e6ed9d0c3b3abc27429f7a5f8c936bd1f0 ]

When using cpu to update page tables, vm update fences are unused. Install stub fence into these fence pointers instead of NULL to avoid NULL dereference when calling dma_fence_wait() on them.

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +4 -2
index 3c0310576b3bf..b6bd667df6763 100644
@@ -1358,6 +1358,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	amdgpu_vm_bo_base_init(&bo_va->base, vm, bo);
 
 	bo_va->ref_count = 1;
+	bo_va->last_pt_update = dma_fence_get_stub();
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 
@@ -2067,7 +2068,8 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		vm->update_funcs = &amdgpu_vm_cpu_funcs;
 	else
 		vm->update_funcs = &amdgpu_vm_sdma_funcs;
-	vm->last_update = NULL;
+
+	vm->last_update = dma_fence_get_stub();
 	vm->last_unlocked = dma_fence_get_stub();
 	vm->last_tlb_flush = dma_fence_get_stub();
 
@@ -2192,7 +2194,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 		goto unreserve_bo;
 
 	dma_fence_put(vm->last_update);
-	vm->last_update = NULL;
+	vm->last_update = dma_fence_get_stub();
 	vm->is_compute_context = true;
 
 	/* Free the shadow bo for compute VM */
Signature · 1 line
2.39.2
author maintainer
#39 lore raw

PATCH 39/58 drm/amd/display: Remove v_startup workaround for dcn3+

In reply to Sasha Levin

From: Daniel Miess <[email protected]>

[ Upstream commit 3a31e8b89b7240d9a17ace8a1ed050bdcb560f9e ]

[Why]
Calls to dcn20_adjust_freesync_v_startup are no longer needed as of dcn3+ and can cause underflow in some cases

[How]
Move calls to dcn20_adjust_freesync_v_startup up into validate_bandwidth for dcn2.x

---
 .../drm/amd/display/dc/dml/dcn20/dcn20_fpu.c  | 24 +++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)
drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +20 -4
index f1c1a4b5fcac3..2044c8fe35688 100644
@@ -1098,10 +1098,6 @@ void dcn20_calculate_dlg_params(struct dc *dc,
 		context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz =
 						pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000;
 		context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest;
-		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
-			dcn20_adjust_freesync_v_startup(
-				&context->res_ctx.pipe_ctx[i].stream->timing,
-				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
 
 		pipe_idx++;
 	}
@@ -1915,6 +1911,7 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co
 	int vlevel = 0;
 	int pipe_split_from[MAX_PIPES];
 	int pipe_cnt = 0;
+	int i = 0;
 	display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC);
 	DC_LOGGER_INIT(dc->ctx->logger);
 
@@ -1938,6 +1935,15 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co
 	dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate);
 	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
 
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
+			dcn20_adjust_freesync_v_startup(
+				&context->res_ctx.pipe_ctx[i].stream->timing,
+				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
+	}
+
 	BW_VAL_TRACE_END_WATERMARKS();
 
 	goto validate_out;
@@ -2210,6 +2216,7 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc,
 	int vlevel = 0;
 	int pipe_split_from[MAX_PIPES];
 	int pipe_cnt = 0;
+	int i = 0;
 	display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC);
 	DC_LOGGER_INIT(dc->ctx->logger);
 
@@ -2238,6 +2245,15 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc,
 	dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate);
 	dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel);
 
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (!context->res_ctx.pipe_ctx[i].stream)
+			continue;
+		if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid)
+			dcn20_adjust_freesync_v_startup(
+				&context->res_ctx.pipe_ctx[i].stream->timing,
+				&context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start);
+	}
+
 	BW_VAL_TRACE_END_WATERMARKS();
 
 	goto validate_out;
Signature · 1 line
2.39.2
author maintainer
#36 lore raw

PATCH 36/58 HID: logitech-hidpp: Add USB and Bluetooth IDs for the Logitech G915 TKL Keyboard

In reply to Sasha Levin

From: stuarthayhurst <[email protected]>

[ Upstream commit 48aea8b445c422a372cf15915101035a47105421 ]

Adds the USB and Bluetooth IDs for the Logitech G915 TKL keyboard, for device detection For this device, this provides battery reporting on top of hid-generic

---
 drivers/hid/hid-logitech-hidpp.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/hid/hid-logitech-hidpp.c +4 -0
index 5e1a412fd28fa..2ff007244d5e1 100644
@@ -4608,6 +4608,8 @@ static const struct hid_device_id hidpp_devices[] = {
 	  HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, 0xC086) },
 	{ /* Logitech G903 Hero Gaming Mouse over USB */
 	  HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, 0xC091) },
+	{ /* Logitech G915 TKL Keyboard over USB */
+	  HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, 0xC343) },
 	{ /* Logitech G920 Wheel over USB */
 	  HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_G920_WHEEL),
 		.driver_data = HIDPP_QUIRK_CLASS_G920 | HIDPP_QUIRK_FORCE_OUTPUT_REPORTS},
@@ -4630,6 +4632,8 @@ static const struct hid_device_id hidpp_devices[] = {
 	{ /* MX5500 keyboard over Bluetooth */
 	  HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb30b),
 	  .driver_data = HIDPP_QUIRK_HIDPP_CONSUMER_VENDOR_KEYS },
+	{ /* Logitech G915 TKL keyboard over Bluetooth */
+	  HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb35f) },
 	{ /* M-RCQ142 V470 Cordless Laser Mouse over Bluetooth */
 	  HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_LOGITECH, 0xb008) },
 	{ /* MX Master mouse over Bluetooth */
Signature · 1 line
2.39.2
author maintainer
#40 lore raw

PATCH 40/58 drm/amd/display: Trigger DIO FIFO resync on commit streams

In reply to Sasha Levin

From: Saaem Rizvi <[email protected]>

[ Upstream commit 3e8d74cb128fb1a4d56270ffbecea6056c55739a ]

[WHY]
Currently, there is an intermittent issue where a screen can either go blank or be corrupted.

[HOW]
To resolve the issue we trigger the ramping logic for DIO FIFO so that it goes back up to the correct speed.

---
 .../display/dc/dce110/dce110_hw_sequencer.c   |  3 +++
 .../gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h |  4 ++-
 .../drm/amd/display/dc/dcn314/dcn314_dccg.c   | 11 ++++++++
 .../drm/amd/display/dc/dcn314/dcn314_dccg.h   |  5 +++-
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.c  | 27 +++++++++++++++++++
 .../drm/amd/display/dc/dcn314/dcn314_hwseq.h  |  2 ++
 .../drm/amd/display/dc/dcn314/dcn314_init.c   |  1 +
 drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h  |  3 +++
 .../amd/display/dc/inc/hw_sequencer_private.h |  2 ++
 9 files changed, 56 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +3 -0
index 8d2460d06bced..1a0be40d125c6 100644
@@ -2291,6 +2291,9 @@ enum dc_status dce110_apply_ctx_to_hw(
 
 		if (DC_OK != status)
 			return status;
+
+		if (hws->funcs.resync_fifo_dccg_dio)
+			hws->funcs.resync_fifo_dccg_dio(hws, dc, context);
 	}
 
 	if (dc->fbc_compressor)
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dccg.h +3 -1
index 7bdc146f7cb59..c8602bcfa393a 100644
@@ -208,7 +208,9 @@
 #define DCCG314_REG_FIELD_LIST(type) \
 	type DSCCLK3_DTO_PHASE;\
 	type DSCCLK3_DTO_MODULO;\
-	type DSCCLK3_DTO_ENABLE;
+	type DSCCLK3_DTO_ENABLE;\
+	type DENTIST_DISPCLK_RDIVIDER;\
+	type DENTIST_DISPCLK_WDIVIDER;
 
 #define DCCG32_REG_FIELD_LIST(type) \
 	type DPSTREAMCLK0_EN;\
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c +11 -0
index de7bfba2c1798..e0e7d32bb1a0e 100644
@@ -45,6 +45,16 @@
 #define DC_LOGGER \
 	dccg->ctx->logger
 
+static void dccg314_trigger_dio_fifo_resync(
+	struct dccg *dccg)
+{
+	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
+	uint32_t dispclk_rdivider_value = 0;
+
+	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value);
+	REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value);
+}
+
 static void dccg314_get_pixel_rate_div(
 		struct dccg *dccg,
 		uint32_t otg_inst,
@@ -357,6 +367,7 @@ static const struct dccg_funcs dccg314_funcs = {
 	.disable_dsc = dccg31_disable_dscclk,
 	.enable_dsc = dccg31_enable_dscclk,
 	.set_pixel_rate_div = dccg314_set_pixel_rate_div,
+	.trigger_dio_fifo_resync = dccg314_trigger_dio_fifo_resync,
 	.set_valid_pixel_rate = dccg314_set_valid_pixel_rate,
 };
 
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.h +4 -1
index 90687a9e8fddd..8e07d3151f915 100644
@@ -192,7 +192,10 @@
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYBSYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYCSYMCLK_GATE_DISABLE, mask_sh),\
 	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYDSYMCLK_GATE_DISABLE, mask_sh),\
-	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh)
+	DCCG_SF(DCCG_GATE_DISABLE_CNTL2, PHYESYMCLK_GATE_DISABLE, mask_sh),\
+	DCCG_SF(HDMISTREAMCLK0_DTO_PARAM, HDMISTREAMCLK0_DTO_MODULO, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, mask_sh),\
+	DCCG_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh)
 
 struct dccg *dccg314_create(
 	struct dc_context *ctx,
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.c +27 -0
index cc3fe9cac5b53..df5c5117bc9b2 100644
@@ -390,6 +390,33 @@ void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx)
 				pix_per_cycle);
 }
 
+void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context)
+{
+	uint8_t i;
+	struct pipe_ctx *pipe = NULL;
+	bool otg_disabled[MAX_PIPES] = {false};
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		pipe = &dc->current_state->res_ctx.pipe_ctx[i];
+
+		if (pipe->top_pipe || pipe->prev_odm_pipe)
+			continue;
+
+		if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) {
+			pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg);
+			reset_sync_context_for_pipe(dc, context, i);
+			otg_disabled[i] = true;
+		}
+	}
+
+	hws->ctx->dc->res_pool->dccg->funcs->trigger_dio_fifo_resync(hws->ctx->dc->res_pool->dccg);
+
+	for (i = 0; i < dc->res_pool->pipe_count; i++) {
+		if (otg_disabled[i])
+			pipe->stream_res.tg->funcs->enable_crtc(pipe->stream_res.tg);
+	}
+}
+
 void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on)
 {
 	if (!hws->ctx->dc->debug.root_clock_optimization.bits.dpp)
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_hwseq.h +2 -0
index 6d0b62503caa6..559d71002e8a4 100644
@@ -41,6 +41,8 @@ unsigned int dcn314_calculate_dccg_k1_k2_values(struct pipe_ctx *pipe_ctx, unsig
 
 void dcn314_set_pixels_per_cycle(struct pipe_ctx *pipe_ctx);
 
+void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context);
+
 void dcn314_hubp_pg_control(struct dce_hwseq *hws, unsigned int hubp_inst, bool power_on);
 
 void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool clock_on);
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c +1 -0
index a588f46b166f4..6b8586852c0eb 100644
@@ -151,6 +151,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = {
 	.setup_hpo_hw_control = dcn31_setup_hpo_hw_control,
 	.calculate_dccg_k1_k2_values = dcn314_calculate_dccg_k1_k2_values,
 	.set_pixels_per_cycle = dcn314_set_pixels_per_cycle,
+	.resync_fifo_dccg_dio = dcn314_resync_fifo_dccg_dio,
 };
 
 void dcn314_hw_sequencer_construct(struct dc *dc)
drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +3 -0
index ad6acd1b34e1d..0b700b3d7d972 100644
@@ -159,6 +159,9 @@ struct dccg_funcs {
 			int otg_inst,
 			int pixclk_khz);
 
+	void (*trigger_dio_fifo_resync)(
+			struct dccg *dccg);
+
 	void (*dpp_root_clock_control)(
 			struct dccg *dccg,
 			unsigned int dpp_inst,
drivers/gpu/drm/amd/display/dc/inc/hw_sequencer_private.h +2 -0
index 4513544559be2..4ca4192c1e127 100644
@@ -160,6 +160,8 @@ struct hwseq_private_funcs {
 			unsigned int *k1_div,
 			unsigned int *k2_div);
 	void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx);
+	void (*resync_fifo_dccg_dio)(struct dce_hwseq *hws, struct dc *dc,
+			struct dc_state *context);
 	bool (*is_dp_dig_pixel_rate_div_policy)(struct pipe_ctx *pipe_ctx);
 #endif
 };
Signature · 1 line
2.39.2
author maintainer
#41 lore raw

PATCH 41/58 drm/amd/display: Apply 60us prefetch for DCFCLK <= 300Mhz

In reply to Sasha Levin

From: Alvin Lee <[email protected]>

[ Upstream commit 7e60ab4eb3e4ba2adac46d737fdbbc5732bebd58 ]

[Description]
- Previously we wanted to apply extra 60us of prefetch for min DCFCLK (200Mhz), but DCFCLK can be calculated to be 201Mhz which underflows also without the extra prefetch
- Instead, apply the the extra 60us prefetch for any DCFCLK freq <= 300Mhz

---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c    | 4 ++--
 .../gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h    | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.c +2 -2
index d75248b6cae99..9a5150e96017a 100644
@@ -811,7 +811,7 @@ static void DISPCLKDPPCLKDCFCLKDeepSleepPrefetchParametersWatermarksAndPerforman
 					v->SwathHeightC[k],
 					TWait,
 					(v->DRAMSpeedPerState[mode_lib->vba.VoltageLevel] <= MEM_STROBE_FREQ_MHZ ||
-						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= MIN_DCFCLK_FREQ_MHZ) ?
+						v->DCFCLKPerState[mode_lib->vba.VoltageLevel] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 							mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 					/* Output */
 					&v->DSTXAfterScaler[k],
@@ -3311,7 +3311,7 @@ void dml32_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l
 							v->swath_width_chroma_ub_this_state[k],
 							v->SwathHeightYThisState[k],
 							v->SwathHeightCThisState[k], v->TWait,
-							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= MIN_DCFCLK_FREQ_MHZ) ?
+							(v->DRAMSpeedPerState[i] <= MEM_STROBE_FREQ_MHZ || v->DCFCLKState[i][j] <= DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ) ?
 									mode_lib->vba.ip.min_prefetch_in_strobe_us : 0,
 
 							/* Output */
drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_32.h +1 -1
index d98e36a9a09cc..c4745d63039bb 100644
@@ -53,7 +53,7 @@
 #define BPP_BLENDED_PIPE 0xffffffff
 
 #define MEM_STROBE_FREQ_MHZ 1600
-#define MIN_DCFCLK_FREQ_MHZ 200
+#define DCFCLK_FREQ_EXTRA_PREFETCH_REQ_MHZ 300
 #define MEM_STROBE_MAX_DELIVERY_TIME_US 60.0
 
 struct display_mode_lib;
Signature · 1 line
2.39.2
author maintainer
#42 lore raw

PATCH 42/58 Revert "drm/amd/display: Do not set drr on pipe commit"

In reply to Sasha Levin

From: Michel Dänzer <[email protected]>

[ Upstream commit 360930985ec9f394c82ba0b235403b4a366d1560 ]

This reverts commit e101bf95ea87ccc03ac2f48dfc0757c6364ff3c7.

Caused a regression:

Samsung Odyssey Neo G9, running at 5120x1440@240/VRR, connected to Navi 21 via DisplayPort, blanks and the GPU hangs while starting the Steam game Assetto Corsa Competizione (via Proton 7.0).

Example dmesg excerpt:

amdgpu 0000:0c:00.0: [drm] ERROR [CRTC:82:crtc-0] flip_done timed out NMI watchdog: Watchdog detected hard LOCKUP on cpu 6
[...]

 RIP: 0010:amdgpu_device_rreg.part.0+0x2f/0xf0 [amdgpu]
 Code: 41 54 44 8d 24 b5 00 00 00 00 55 89 f5 53 48 89 fb 4c 3b a7 60 0b 00 00 73 6a 83 e2 02 74 29 4c 03 a3 68 0b 00 00 45 8b 24 24 <48> 8b 43 08 0f b7 70 3e 66 90 44 89 e0 5b 5d 41 5c 31 d2 31 c9 31
 RSP: 0000:ffffb39a119dfb88 EFLAGS: 00000086
 RAX: ffffffffc0eb96a0 RBX: ffff9e7963dc0000 RCX: 0000000000007fff
 RDX: 0000000000000000 RSI: 0000000000004ff6 RDI: ffff9e7963dc0000
 RBP: 0000000000004ff6 R08: ffffb39a119dfc40 R09: 0000000000000010
 R10: ffffb39a119dfc40 R11: ffffb39a119dfc44 R12: 00000000000e05ae
 R13: 0000000000000000 R14: ffff9e7963dc0010 R15: 0000000000000000
 FS:  000000001012f6c0(0000) GS:ffff9e805eb80000(0000) knlGS:000000007fd40000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00000000461ca000 CR3: 00000002a8a20000 CR4: 0000000000350ee0
 Call Trace:
  <TASK>
  dm_read_reg_func+0x37/0xc0 [amdgpu]
  generic_reg_get2+0x22/0x60 [amdgpu]
  optc1_get_crtc_scanoutpos+0x6a/0xc0 [amdgpu]
  dc_stream_get_scanoutpos+0x74/0x90 [amdgpu]
  dm_crtc_get_scanoutpos+0x82/0xf0 [amdgpu]
  amdgpu_display_get_crtc_scanoutpos+0x91/0x190 [amdgpu]
  ? dm_read_reg_func+0x37/0xc0 [amdgpu]
  amdgpu_get_vblank_counter_kms+0xb4/0x1a0 [amdgpu]
  dm_pflip_high_irq+0x213/0x2f0 [amdgpu]
  amdgpu_dm_irq_handler+0x8a/0x200 [amdgpu]
  amdgpu_irq_dispatch+0xd4/0x220 [amdgpu]
  amdgpu_ih_process+0x7f/0x110 [amdgpu]
  amdgpu_irq_handler+0x1f/0x70 [amdgpu]
  __handle_irq_event_percpu+0x46/0x1b0
  handle_irq_event+0x34/0x80
  handle_edge_irq+0x9f/0x240
  __common_interrupt+0x66/0x110
  common_interrupt+0x5c/0xd0
  asm_common_interrupt+0x22/0x40
---
 drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 6 ------
 drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c | 7 -------
 2 files changed, 13 deletions(-)
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +0 -6
index 6ce10fd4bb1a4..5403e9399a465 100644
@@ -2113,12 +2113,6 @@ void dcn20_optimize_bandwidth(
 	if (hubbub->funcs->program_compbuf_size)
 		hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true);
 
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
-		dc_dmub_srv_p_state_delegate(dc,
-			true, context);
-		context->bw_ctx.bw.dcn.clk.p_state_change_support = true;
-	}
-
 	dc->clk_mgr->funcs->update_clocks(
 			dc->clk_mgr,
 			context,
drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +0 -7
index f923224e85fc6..32121db2851e6 100644
@@ -985,18 +985,11 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc,
 void dcn30_prepare_bandwidth(struct dc *dc,
 			     struct dc_state *context)
 {
-	if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) {
-		dc->optimized_required = true;
-		context->bw_ctx.bw.dcn.clk.p_state_change_support = false;
-	}
-
 	if (dc->clk_mgr->dc_mode_softmax_enabled)
 		if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 &&
 				context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000)
 			dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz);
 
 	dcn20_prepare_bandwidth(dc, context);
-
-	dc_dmub_srv_p_state_delegate(dc, false, context);
 }
 
Signature · 1 line
2.39.2
author maintainer
#44 lore raw

PATCH 44/58 drm/amd/pm: Fill metrics data for SMUv13.0.6

In reply to Sasha Levin

From: Lijo Lazar <[email protected]>

[ Upstream commit 1718e973e3d23b653cd77994073a9deda3875689 ]

Populate metrics data table for SMU v13.0.6. Add PCIe link speed/width information also.

---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 107 +++++++++++-------
 1 file changed, 66 insertions(+), 41 deletions(-)
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +66 -41
index ea8f3d6fb98b3..de63e7fd1d7bf 100644
@@ -80,7 +80,10 @@
 /* possible frequency drift (1Mhz) */
 #define EPSILON 1
 
-#define smnPCIE_ESM_CTRL 0x111003D0
+#define smnPCIE_ESM_CTRL 0x193D0
+#define smnPCIE_LC_LINK_WIDTH_CNTL 0x1ab40288
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK 0x00000070L
+#define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT 0x4
 
 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
@@ -201,6 +204,7 @@ struct PPTable_t {
 };
 
 #define SMUQ10_TO_UINT(x) ((x) >> 10)
+#define SMUQ16_TO_UINT(x) ((x) >> 16)
 
 struct smu_v13_0_6_dpm_map {
 	enum smu_clk_type clk_type;
@@ -1823,6 +1827,15 @@ static void smu_v13_0_6_log_thermal_throttling_event(struct smu_context *smu)
 						   smu_v13_0_6_throttler_map));
 }
 
+static int
+smu_v13_0_6_get_current_pcie_link_width_level(struct smu_context *smu)
+{
+	struct amdgpu_device *adev = smu->adev;
+
+	return REG_GET_FIELD(RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL),
+			     PCIE_LC_LINK_WIDTH_CNTL, LC_LINK_WIDTH_RD);
+}
+
 static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
@@ -1841,8 +1854,12 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 	struct smu_table_context *smu_table = &smu->smu_table;
 	struct gpu_metrics_v1_3 *gpu_metrics =
 		(struct gpu_metrics_v1_3 *)smu_table->gpu_metrics_table;
+	struct amdgpu_device *adev = smu->adev;
+	int ret = 0, inst0, xcc0;
 	MetricsTable_t *metrics;
-	int i, ret = 0;
+
+	inst0 = adev->sdma.instance[0].aid_id;
+	xcc0 = GET_INST(GC, 0);
 
 	metrics = kzalloc(sizeof(MetricsTable_t), GFP_KERNEL);
 	ret = smu_v13_0_6_get_metrics_table(smu, metrics, true);
@@ -1851,51 +1868,59 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 
 	smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 3);
 
-	/* TODO: Decide on how to fill in zero value fields */
-	gpu_metrics->temperature_edge = 0;
-	gpu_metrics->temperature_hotspot = 0;
-	gpu_metrics->temperature_mem = 0;
-	gpu_metrics->temperature_vrgfx = 0;
-	gpu_metrics->temperature_vrsoc = 0;
-	gpu_metrics->temperature_vrmem = 0;
-
-	gpu_metrics->average_gfx_activity = 0;
-	gpu_metrics->average_umc_activity = 0;
-	gpu_metrics->average_mm_activity = 0;
-
-	gpu_metrics->average_socket_power = 0;
-	gpu_metrics->energy_accumulator = 0;
-
-	gpu_metrics->average_gfxclk_frequency = 0;
-	gpu_metrics->average_socclk_frequency = 0;
-	gpu_metrics->average_uclk_frequency = 0;
-	gpu_metrics->average_vclk0_frequency = 0;
-	gpu_metrics->average_dclk0_frequency = 0;
-
-	gpu_metrics->current_gfxclk = 0;
-	gpu_metrics->current_socclk = 0;
-	gpu_metrics->current_uclk = 0;
-	gpu_metrics->current_vclk0 = 0;
-	gpu_metrics->current_dclk0 = 0;
-
+	gpu_metrics->temperature_hotspot =
+		SMUQ10_TO_UINT(metrics->MaxSocketTemperature);
+	/* Individual HBM stack temperature is not reported */
+	gpu_metrics->temperature_mem =
+		SMUQ10_TO_UINT(metrics->MaxHbmTemperature);
+	/* Reports max temperature of all voltage rails */
+	gpu_metrics->temperature_vrsoc =
+		SMUQ10_TO_UINT(metrics->MaxVrTemperature);
+
+	gpu_metrics->average_gfx_activity =
+		SMUQ10_TO_UINT(metrics->SocketGfxBusy);
+	gpu_metrics->average_umc_activity =
+		SMUQ10_TO_UINT(metrics->DramBandwidthUtilization);
+
+	gpu_metrics->average_socket_power =
+		SMUQ10_TO_UINT(metrics->SocketPower);
+	gpu_metrics->energy_accumulator =
+		SMUQ16_TO_UINT(metrics->SocketEnergyAcc);
+
+	gpu_metrics->current_gfxclk =
+		SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc0]);
+	gpu_metrics->current_socclk =
+		SMUQ10_TO_UINT(metrics->SocclkFrequency[inst0]);
+	gpu_metrics->current_uclk = SMUQ10_TO_UINT(metrics->UclkFrequency);
+	gpu_metrics->current_vclk0 =
+		SMUQ10_TO_UINT(metrics->VclkFrequency[inst0]);
+	gpu_metrics->current_dclk0 =
+		SMUQ10_TO_UINT(metrics->DclkFrequency[inst0]);
+
+	gpu_metrics->average_gfxclk_frequency = gpu_metrics->current_gfxclk;
+	gpu_metrics->average_socclk_frequency = gpu_metrics->current_socclk;
+	gpu_metrics->average_uclk_frequency = gpu_metrics->current_uclk;
+	gpu_metrics->average_vclk0_frequency = gpu_metrics->current_vclk0;
+	gpu_metrics->average_dclk0_frequency = gpu_metrics->current_dclk0;
+
+	/* Throttle status is not reported through metrics now */
 	gpu_metrics->throttle_status = 0;
-	gpu_metrics->indep_throttle_status = smu_cmn_get_indep_throttler_status(
-		gpu_metrics->throttle_status, smu_v13_0_6_throttler_map);
-
-	gpu_metrics->current_fan_speed = 0;
 
-	gpu_metrics->pcie_link_width = 0;
-	gpu_metrics->pcie_link_speed = smu_v13_0_6_get_current_pcie_link_speed(smu);
+	if (!(adev->flags & AMD_IS_APU)) {
+		gpu_metrics->pcie_link_width =
+			smu_v13_0_6_get_current_pcie_link_width_level(smu);
+		gpu_metrics->pcie_link_speed =
+			smu_v13_0_6_get_current_pcie_link_speed(smu);
+	}
 
 	gpu_metrics->system_clock_counter = ktime_get_boottime_ns();
 
-	gpu_metrics->gfx_activity_acc = 0;
-	gpu_metrics->mem_activity_acc = 0;
-
-	for (i = 0; i < NUM_HBM_INSTANCES; i++)
-		gpu_metrics->temperature_hbm[i] = 0;
+	gpu_metrics->gfx_activity_acc =
+		SMUQ10_TO_UINT(metrics->SocketGfxBusyAcc);
+	gpu_metrics->mem_activity_acc =
+		SMUQ10_TO_UINT(metrics->DramBandwidthUtilizationAcc);
 
-	gpu_metrics->firmware_timestamp = 0;
+	gpu_metrics->firmware_timestamp = metrics->Timestamp;
 
 	*table = (void *)gpu_metrics;
 	kfree(metrics);
Signature · 1 line
2.39.2
author maintainer
#43 lore raw

PATCH 43/58 HID: add quirk for 03f0:464a HP Elite Presenter Mouse

In reply to Sasha Levin

From: Marco Morandini <[email protected]>

[ Upstream commit 0db117359e47750d8bd310d19f13e1c4ef7fc26a ]

HP Elite Presenter Mouse HID Record Descriptor shows two mouses (Repord ID 0x1 and 0x2), one keypad (Report ID 0x5), two Consumer Controls (Report IDs 0x6 and 0x3). Previous to this commit it registers one mouse, one keypad and one Consumer Control, and it was usable only as a digitl laser pointer (one of the two mouses). This patch defines the 464a USB device ID and enables the HID_QUIRK_MULTI_INPUT quirk for it, allowing to use the device both as a mouse and a digital laser pointer.

---
 drivers/hid/hid-ids.h    | 1 +
 drivers/hid/hid-quirks.c | 1 +
 2 files changed, 2 insertions(+)
drivers/hid/hid-ids.h +1 -0
index 5d29abac2300e..55a436a6dde98 100644
@@ -620,6 +620,7 @@
 #define USB_DEVICE_ID_UGCI_FIGHTING	0x0030
 
 #define USB_VENDOR_ID_HP		0x03f0
+#define USB_PRODUCT_ID_HP_ELITE_PRESENTER_MOUSE_464A		0x464a
 #define USB_PRODUCT_ID_HP_LOGITECH_OEM_USB_OPTICAL_MOUSE_0A4A	0x0a4a
 #define USB_PRODUCT_ID_HP_LOGITECH_OEM_USB_OPTICAL_MOUSE_0B4A	0x0b4a
 #define USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE		0x134a
drivers/hid/hid-quirks.c +1 -0
index 804fc03600cc9..3983b4f282f8f 100644
@@ -96,6 +96,7 @@ static const struct hid_device_id hid_quirks[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD_A096), HID_QUIRK_NO_INIT_REPORTS },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HOLTEK_ALT, USB_DEVICE_ID_HOLTEK_ALT_KEYBOARD_A293), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_LOGITECH_OEM_USB_OPTICAL_MOUSE_0A4A), HID_QUIRK_ALWAYS_POLL },
+	{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_ELITE_PRESENTER_MOUSE_464A), HID_QUIRK_MULTI_INPUT },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_LOGITECH_OEM_USB_OPTICAL_MOUSE_0B4A), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_094A), HID_QUIRK_ALWAYS_POLL },
Signature · 1 line
2.39.2
author maintainer
#35 lore raw

PATCH 35/58 HID: i2c-hid: goodix: Add support for "goodix,no-reset-during-suspend" property

In reply to Sasha Levin

From: Fei Shao <[email protected]>

[ Upstream commit 7607f12ba735f04e0ef8718dabadf3a765c9a477 ]

In the beginning, commit 18eeef46d359 ("HID: i2c-hid: goodix: Tie the reset line to true state of the regulator") introduced a change to tie the reset line of the Goodix touchscreen to the state of the regulator to fix a power leakage issue in suspend.

After some time, the change was deemed unnecessary and was reverted in commit 557e05fa9fdd ("HID: i2c-hid: goodix: Stop tying the reset line to the regulator") due to difficulties in managing regulator notifiers for designs like Evoker, which provides a second power rail to touchscreen.

However, the revert caused a power regression on another Chromebook device Steelix in the field, which has a dedicated always-on regulator for touchscreen and was covered by the workaround in the first commit.

To address both cases, this patch adds the support for the new "goodix,no-reset-during-suspend" property in the driver:
- When set to true, the driver does not assert the reset GPIO during power-down.
Instead, the GPIO will be asserted during power-up to ensure the touchscreen always has a clean start and consistent behavior after resuming.
This is for designs with a dedicated always-on regulator.
- When set to false or unset, the driver uses the original control flow and asserts GPIO and disables regulators normally.
This is for the two-regulator and shared-regulator designs.

---
 drivers/hid/i2c-hid/i2c-hid-of-goodix.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)
drivers/hid/i2c-hid/i2c-hid-of-goodix.c +15 -1
index 0060e3dcd775d..db4639db98407 100644
@@ -28,6 +28,7 @@ struct i2c_hid_of_goodix {
 	struct regulator *vdd;
 	struct regulator *vddio;
 	struct gpio_desc *reset_gpio;
+	bool no_reset_during_suspend;
 	const struct goodix_i2c_hid_timing_data *timings;
 };
 
@@ -37,6 +38,14 @@ static int goodix_i2c_hid_power_up(struct i2chid_ops *ops)
 		container_of(ops, struct i2c_hid_of_goodix, ops);
 	int ret;
 
+	/*
+	 * We assert reset GPIO here (instead of during power-down) to ensure
+	 * the device will have a clean state after powering up, just like the
+	 * normal scenarios will have.
+	 */
+	if (ihid_goodix->no_reset_during_suspend)
+		gpiod_set_value_cansleep(ihid_goodix->reset_gpio, 1);
+
 	ret = regulator_enable(ihid_goodix->vdd);
 	if (ret)
 		return ret;
@@ -60,7 +69,9 @@ static void goodix_i2c_hid_power_down(struct i2chid_ops *ops)
 	struct i2c_hid_of_goodix *ihid_goodix =
 		container_of(ops, struct i2c_hid_of_goodix, ops);
 
-	gpiod_set_value_cansleep(ihid_goodix->reset_gpio, 1);
+	if (!ihid_goodix->no_reset_during_suspend)
+		gpiod_set_value_cansleep(ihid_goodix->reset_gpio, 1);
+
 	regulator_disable(ihid_goodix->vddio);
 	regulator_disable(ihid_goodix->vdd);
 }
@@ -91,6 +102,9 @@ static int i2c_hid_of_goodix_probe(struct i2c_client *client)
 	if (IS_ERR(ihid_goodix->vddio))
 		return PTR_ERR(ihid_goodix->vddio);
 
+	ihid_goodix->no_reset_during_suspend =
+		of_property_read_bool(client->dev.of_node, "goodix,no-reset-during-suspend");
+
 	ihid_goodix->timings = device_get_match_data(&client->dev);
 
 	return i2c_hid_core_probe(client, &ihid_goodix->ops, 0x0001, 0);
Signature · 1 line
2.39.2
author maintainer
#45 lore raw

PATCH 45/58 drm/amdgpu: unmap and remove csa_va properly

In reply to Sasha Levin

From: Lang Yu <[email protected]>

[ Upstream commit 5daff15cd013422bc6d1efcfe82b586800025384 ]

Root PD BO should be reserved before unmap and remove a bo_va from VM otherwise lockdep will complain.

v2: check fpriv->csa_va is not NULL instead of amdgpu_mcbp (christian)

[14616.936827] WARNING: CPU: 6 PID: 1711 at drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:1762 amdgpu_vm_bo_del+0x399/0x3f0 [amdgpu]
[14616.937096] Call Trace:
[14616.937097]  <TASK>
[14616.937102]  amdgpu_driver_postclose_kms+0x249/0x2f0 [amdgpu]
[14616.937187]  drm_file_free+0x1d6/0x300 [drm]
[14616.937207]  drm_close_helper.isra.0+0x62/0x70 [drm]
[14616.937220]  drm_release+0x5e/0x100 [drm]
[14616.937234]  __fput+0x9f/0x280
[14616.937239]  ____fput+0xe/0x20
[14616.937241]  task_work_run+0x61/0x90
[14616.937246]  exit_to_user_mode_prepare+0x215/0x220
[14616.937251]  syscall_exit_to_user_mode+0x2a/0x60
[14616.937254]  do_syscall_64+0x48/0x90
[14616.937257]  entry_SYSCALL_64_after_hwframe+0x63/0xcd
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 38 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h |  3 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 10 +++----
 3 files changed, 46 insertions(+), 5 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +38 -0
index c6d4d41c4393e..23d054526e7c7 100644
@@ -106,3 +106,41 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	ttm_eu_backoff_reservation(&ticket, &list);
 	return 0;
 }
+
+int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			    struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va,
+			    uint64_t csa_addr)
+{
+	struct ww_acquire_ctx ticket;
+	struct list_head list;
+	struct amdgpu_bo_list_entry pd;
+	struct ttm_validate_buffer csa_tv;
+	int r;
+
+	INIT_LIST_HEAD(&list);
+	INIT_LIST_HEAD(&csa_tv.head);
+	csa_tv.bo = &bo->tbo;
+	csa_tv.num_shared = 1;
+
+	list_add(&csa_tv.head, &list);
+	amdgpu_vm_get_pd_bo(vm, &list, &pd);
+
+	r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL);
+	if (r) {
+		DRM_ERROR("failed to reserve CSA,PD BOs: err=%d\n", r);
+		return r;
+	}
+
+	r = amdgpu_vm_bo_unmap(adev, bo_va, csa_addr);
+	if (r) {
+		DRM_ERROR("failed to do bo_unmap on static CSA, err=%d\n", r);
+		ttm_eu_backoff_reservation(&ticket, &list);
+		return r;
+	}
+
+	amdgpu_vm_bo_del(adev, bo_va);
+
+	ttm_eu_backoff_reservation(&ticket, &list);
+
+	return 0;
+}
drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h +3 -0
index 524b4437a0217..7dfc1f2012ebf 100644
@@ -34,6 +34,9 @@ int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo
 int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			  struct amdgpu_bo *bo, struct amdgpu_bo_va **bo_va,
 			  uint64_t csa_addr, uint32_t size);
+int amdgpu_unmap_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			    struct amdgpu_bo *bo, struct amdgpu_bo_va *bo_va,
+			    uint64_t csa_addr);
 void amdgpu_free_static_csa(struct amdgpu_bo **bo);
 
 #endif
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +5 -5
index 0efb38539d70c..724e80c192973 100644
@@ -1284,12 +1284,12 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
 	if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCE) != NULL)
 		amdgpu_vce_free_handles(adev, file_priv);
 
-	if (amdgpu_mcbp) {
-		/* TODO: how to handle reserve failure */
-		BUG_ON(amdgpu_bo_reserve(adev->virt.csa_obj, true));
-		amdgpu_vm_bo_del(adev, fpriv->csa_va);
+	if (fpriv->csa_va) {
+		uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK;
+
+		WARN_ON(amdgpu_unmap_static_csa(adev, &fpriv->vm, adev->virt.csa_obj,
+						fpriv->csa_va, csa_addr));
 		fpriv->csa_va = NULL;
-		amdgpu_bo_unreserve(adev->virt.csa_obj);
 	}
 
 	pasid = fpriv->vm.pasid;
Signature · 1 line
2.39.2
author maintainer
#48 lore raw

PATCH 48/58 drm/amd/display: Skip DPP DTO update if root clock is gated

In reply to Sasha Levin

From: Nicholas Kazlauskas <[email protected]>

[ Upstream commit 30f90f3c1c2c63c2fa44f61233737d27b72637c2 ]

[Why]
Hardware implements root clock gating by utilizing the DPP DTO registers with a special case of DTO enabled, phase = 0, modulo = 1. This conflicts with our policy to always update the DPPDTO for cases where it's expected to be disabled.

The pipes unexpectedly enter a higher power state than expected because of this programming flow.

[How]
Guard the upper layers of HWSS against this hardware quirk with programming the register with an internal state flag in DCCG.

While technically acting as global state for the DCCG, HWSS shouldn't be expected to understand the hardware quirk for having DTO disabled causing more power than DTO enabled with this specific setting.

This also prevents sequencing errors from occuring in the future if we have to program DPP DTO in multiple locations.

---
 drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c   | 8 ++++++++
 drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c | 5 +++++
 drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h        | 1 +
 3 files changed, 14 insertions(+)
drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +8 -0
index 4c2fdfea162f5..65c1d754e2d6b 100644
@@ -47,6 +47,14 @@ void dccg31_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+	if (dccg->dpp_clock_gated[dpp_inst]) {
+		/*
+		 * Do not update the DPPCLK DTO if the clock is stopped.
+		 * It is treated the same as if the pipe itself were in PG.
+		 */
+		return;
+	}
+
 	if (dccg->ref_dppclk && req_dppclk) {
 		int ref_dppclk = dccg->ref_dppclk;
 		int modulo, phase;
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c +5 -0
index e0e7d32bb1a0e..cf23d7bc560a6 100644
@@ -332,6 +332,9 @@ static void dccg314_dpp_root_clock_control(
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
+	if (dccg->dpp_clock_gated[dpp_inst] == clock_on)
+		return;
+
 	if (clock_on) {
 		/* turn off the DTO and leave phase/modulo at max */
 		REG_UPDATE(DPPCLK_DTO_CTRL, DPPCLK_DTO_ENABLE[dpp_inst], 0);
@@ -345,6 +348,8 @@ static void dccg314_dpp_root_clock_control(
 			  DPPCLK0_DTO_PHASE, 0,
 			  DPPCLK0_DTO_MODULO, 1);
 	}
+
+	dccg->dpp_clock_gated[dpp_inst] = !clock_on;
 }
 
 static const struct dccg_funcs dccg314_funcs = {
drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +1 -0
index 0b700b3d7d972..8dc804bbe98bd 100644
@@ -68,6 +68,7 @@ struct dccg {
 	const struct dccg_funcs *funcs;
 	int pipe_dppclk_khz[MAX_PIPES];
 	int ref_dppclk;
+	bool dpp_clock_gated[MAX_PIPES];
 	//int dtbclk_khz[MAX_PIPES];/* TODO needs to be removed */
 	//int audio_dtbclk_khz;/* TODO needs to be removed */
 	//int ref_dtbclk_khz;/* TODO needs to be removed */
Signature · 1 line
2.39.2
author maintainer
#33 lore raw

PATCH 33/58 hwmon: (aquacomputer_d5next) Add support for Aquacomputer Leakshield

In reply to Sasha Levin

From: Aleksa Savic <[email protected]>

[ Upstream commit b3d3be6c4cd1908b9ffdb3d347de232a6c34a0a9 ]

Extend aquacomputer_d5next driver to expose various hardware sensors of the Aquacomputer Leakshield leak prevention system, which communicates through a proprietary USB HID protocol. Implemented by Noah Bergbauer [1].

Two temperature sensors are exposed, along with pressure (current, min, max and target), reservoir volume (total and filled), pump speed and flow. Pump speed and flow values are user provided and allow the Leakshield to optimize its operation. Writing them to the device is subject of future patches.

[1] https://github.com/aleksamagicka/aquacomputer_d5next-hwmon/pull/41

Originally-from: Noah Bergbauer <[email protected]>

---
 Documentation/hwmon/aquacomputer_d5next.rst |   9 ++
 drivers/hwmon/aquacomputer_d5next.c         | 109 +++++++++++++++++++-
 2 files changed, 114 insertions(+), 4 deletions(-)
Documentation/hwmon/aquacomputer_d5next.rst +9 -0
index 14b37851af0ca..94dc2d93d1808 100644
@@ -12,6 +12,7 @@ Supported devices:
 * Aquacomputer Octo fan controller
 * Aquacomputer Quadro fan controller
 * Aquacomputer High Flow Next sensor
+* Aquacomputer Leakshield leak prevention system
 * Aquacomputer Aquastream XT watercooling pump
 * Aquacomputer Aquastream Ultimate watercooling pump
 * Aquacomputer Poweradjust 3 fan controller
@@ -57,6 +58,11 @@ The High Flow Next exposes +5V voltages, water quality, conductivity and flow re
 A temperature sensor can be connected to it, in which case it provides its reading
 and an estimation of the dissipated/absorbed power in the liquid cooling loop.
 
+The Leakshield exposes two temperature sensors and coolant pressure (current, min, max and
+target readings). It also exposes the estimated reservoir volume and how much of it is
+filled with coolant. Pump RPM and flow can be set to enhance on-device calculations,
+but this is not yet implemented here.
+
 The Aquastream XT pump exposes temperature readings for the coolant, external sensor
 and fan IC. It also exposes pump and fan speeds (in RPM), voltages, as well as pump
 current.
@@ -83,6 +89,9 @@ Sysfs entries
 temp[1-20]_input Physical/virtual temperature sensors (in millidegrees Celsius)
 temp[1-8]_offset Temperature sensor correction offset (in millidegrees Celsius)
 fan[1-8]_input   Pump/fan speed (in RPM) / Flow speed (in dL/h)
+fan1_min         Minimal fan speed (in RPM)
+fan1_max         Maximal fan speed (in RPM)
+fan1_target      Target fan speed (in RPM)
 fan5_pulses      Quadro flow sensor pulses
 power[1-8]_input Pump/fan power (in micro Watts)
 in[0-7]_input    Pump/fan voltage (in milli Volts)
drivers/hwmon/aquacomputer_d5next.c +105 -4

195 lines not on this page · raw message

Signature · 1 line
2.39.2
author maintainer
#34 lore raw

PATCH 34/58 dt-bindings: input: goodix: Add "goodix,no-reset-during-suspend" property

In reply to Sasha Levin

From: Fei Shao <[email protected]>

[ Upstream commit 359ed24a0dd3802e703ec8071dc3b6ed446de5f0 ]

We observed that on Chromebook device Steelix, if Goodix GT7375P touchscreen is powered in suspend (because, for example, it connects to an always-on regulator) and with the reset GPIO asserted, it will introduce about 14mW power leakage.

To address that, we add this property to skip reset during suspend. If it's set, the driver will stop asserting the reset GPIO during power-down. Refer to the comments in the driver for details.

---
 .../devicetree/bindings/input/goodix,gt7375p.yaml        | 9 +++++++++
 1 file changed, 9 insertions(+)
Documentation/devicetree/bindings/input/goodix,gt7375p.yaml +9 -0
index ce18d7dadae23..1edad1da1196d 100644
@@ -43,6 +43,15 @@ properties:
       itself as long as it allows the main board to make signals compatible
       with what the touchscreen is expecting for its IO rails.
 
+  goodix,no-reset-during-suspend:
+    description:
+      Set this to true to enforce the driver to not assert the reset GPIO
+      during suspend.
+      Due to potential touchscreen hardware flaw, back-powering could happen in
+      suspend if the power supply is on and with active-low reset GPIO asserted.
+      This property is used to avoid the back-powering issue.
+    type: boolean
+
 required:
   - compatible
   - reg
Signature · 1 line
2.39.2
author maintainer
#49 lore raw

PATCH 49/58 drm/amd/display: Enable dcn314 DPP RCO

In reply to Sasha Levin

From: Daniel Miess <[email protected]>

[ Upstream commit 17fbdbda9cc87ff5a013898de506212d25323ed7 ]

[Why and How]
Add back debug bits enabling RCO for dcn314 as underflow associated with this change has been resolved

---
 .../drm/amd/display/dc/dcn314/dcn314_resource.c  | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c +16 -0
index abeeede38fb39..049fafd3d908c 100644
@@ -921,6 +921,22 @@ static const struct dc_debug_options debug_defaults_drv = {
 			.afmt = true,
 		}
 	},
+
+	.root_clock_optimization = {
+			.bits = {
+					.dpp = true,
+					.dsc = false,
+					.hdmistream = false,
+					.hdmichar = false,
+					.dpstream = false,
+					.symclk32_se = false,
+					.symclk32_le = false,
+					.symclk_fe = false,
+					.physymclk = false,
+					.dpiasymclk = false,
+			}
+	},
+
 	.seamless_boot_odm_combine = true
 };
 
Signature · 1 line
2.39.2
author maintainer
#52 lore raw

PATCH 52/58 HID: intel-ish-hid: ipc: Add Arrow Lake PCI device ID

In reply to Sasha Levin

From: Even Xu <[email protected]>

[ Upstream commit 4982126e3029cd59fbd1da0d9cc0365a0585fe64 ]

Add device ID of Arrow Lake-H into ishtp support list.

---
 drivers/hid/intel-ish-hid/ipc/hw-ish.h  | 1 +
 drivers/hid/intel-ish-hid/ipc/pci-ish.c | 1 +
 2 files changed, 2 insertions(+)
drivers/hid/intel-ish-hid/ipc/hw-ish.h +1 -0
index fc108f19a64c3..e99f3a3c65e15 100644
@@ -33,6 +33,7 @@
 #define ADL_N_DEVICE_ID		0x54FC
 #define RPL_S_DEVICE_ID		0x7A78
 #define MTL_P_DEVICE_ID		0x7E45
+#define ARL_H_DEVICE_ID		0x7745
 
 #define	REVISION_ID_CHT_A0	0x6
 #define	REVISION_ID_CHT_Ax_SI	0x0
drivers/hid/intel-ish-hid/ipc/pci-ish.c +1 -0
index 7120b30ac51d0..55cb25038e632 100644
@@ -44,6 +44,7 @@ static const struct pci_device_id ish_pci_tbl[] = {
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, ADL_N_DEVICE_ID)},
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, RPL_S_DEVICE_ID)},
 	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, MTL_P_DEVICE_ID)},
+	{PCI_DEVICE(PCI_VENDOR_ID_INTEL, ARL_H_DEVICE_ID)},
 	{0, }
 };
 MODULE_DEVICE_TABLE(pci, ish_pci_tbl);
Signature · 1 line
2.39.2
author maintainer
#53 lore raw

PATCH 53/58 udf: Fix uninitialized array access for some pathnames

In reply to Sasha Levin

From: Jan Kara <[email protected]>

[ Upstream commit 028f6055c912588e6f72722d89c30b401bbcf013 ]

For filenames that begin with . and are between 2 and 5 characters long, UDF charset conversion code would read uninitialized memory in the output buffer. The only practical impact is that the name may be prepended a "unification hash" when it is not actually needed but still it is good to fix this.

---
 fs/udf/unicode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/udf/unicode.c +1 -1
index 622569007b530..2142cbd1dde24 100644
@@ -247,7 +247,7 @@ static int udf_name_from_CS0(struct super_block *sb,
 	}
 
 	if (translate) {
-		if (str_o_len <= 2 && str_o[0] == '.' &&
+		if (str_o_len > 0 && str_o_len <= 2 && str_o[0] == '.' &&
 		    (str_o_len == 1 || str_o[1] == '.'))
 			needsCRC = 1;
 		if (needsCRC) {
Signature · 1 line
2.39.2
author maintainer
#54 lore raw

PATCH 54/58 ALSA: hda/realtek: Add quirks for ROG ALLY CS35l41 audio

In reply to Sasha Levin

From: Matthew Anderson <[email protected]>

[ Upstream commit 724418b84e6248cd27599607b7e5fac365b8e3f5 ]

This requires a patched ACPI table or a firmware from ASUS to work because the system does not come with the _DSD field for the CSC3551.

---
 sound/pci/hda/patch_realtek.c | 46 +++++++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)
sound/pci/hda/patch_realtek.c +46 -0
index dabfdecece264..9310014c70b81 100644
@@ -7120,6 +7120,10 @@ enum {
 	ALC294_FIXUP_ASUS_DUAL_SPK,
 	ALC285_FIXUP_THINKPAD_X1_GEN7,
 	ALC285_FIXUP_THINKPAD_HEADSET_JACK,
+	ALC294_FIXUP_ASUS_ALLY,
+	ALC294_FIXUP_ASUS_ALLY_PINS,
+	ALC294_FIXUP_ASUS_ALLY_VERBS,
+	ALC294_FIXUP_ASUS_ALLY_SPEAKER,
 	ALC294_FIXUP_ASUS_HPE,
 	ALC294_FIXUP_ASUS_COEF_1B,
 	ALC294_FIXUP_ASUS_GX502_HP,
@@ -8432,6 +8436,47 @@ static const struct hda_fixup alc269_fixups[] = {
 		.chained = true,
 		.chain_id = ALC294_FIXUP_SPK2_TO_DAC1
 	},
+	[ALC294_FIXUP_ASUS_ALLY] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = cs35l41_fixup_i2c_two,
+		.chained = true,
+		.chain_id = ALC294_FIXUP_ASUS_ALLY_PINS
+	},
+	[ALC294_FIXUP_ASUS_ALLY_PINS] = {
+		.type = HDA_FIXUP_PINS,
+		.v.pins = (const struct hda_pintbl[]) {
+			{ 0x19, 0x03a11050 },
+			{ 0x1a, 0x03a11c30 },
+			{ 0x21, 0x03211420 },
+			{ }
+		},
+		.chained = true,
+		.chain_id = ALC294_FIXUP_ASUS_ALLY_VERBS
+	},
+	[ALC294_FIXUP_ASUS_ALLY_VERBS] = {
+		.type = HDA_FIXUP_VERBS,
+		.v.verbs = (const struct hda_verb[]) {
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x45 },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x5089 },
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x46 },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x0004 },
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x47 },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0xa47a },
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x49 },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x0049},
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x4a },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x201b },
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x6b },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x4278},
+			{ }
+		},
+		.chained = true,
+		.chain_id = ALC294_FIXUP_ASUS_ALLY_SPEAKER
+	},
+	[ALC294_FIXUP_ASUS_ALLY_SPEAKER] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc285_fixup_speaker2_to_dac1,
+	},
 	[ALC285_FIXUP_THINKPAD_X1_GEN7] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc285_fixup_thinkpad_x1_gen7,
@@ -9535,6 +9580,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1043, 0x16e3, "ASUS UX50", ALC269_FIXUP_STEREO_DMIC),
 	SND_PCI_QUIRK(0x1043, 0x1740, "ASUS UX430UA", ALC295_FIXUP_ASUS_DACS),
 	SND_PCI_QUIRK(0x1043, 0x17d1, "ASUS UX431FL", ALC294_FIXUP_ASUS_DUAL_SPK),
+	SND_PCI_QUIRK(0x1043, 0x17f3, "ROG Ally RC71L_RC71L", ALC294_FIXUP_ASUS_ALLY),
 	SND_PCI_QUIRK(0x1043, 0x1881, "ASUS Zephyrus S/M", ALC294_FIXUP_ASUS_GX502_PINS),
 	SND_PCI_QUIRK(0x1043, 0x18b1, "Asus MJ401TA", ALC256_FIXUP_ASUS_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1043, 0x18f1, "Asus FX505DT", ALC256_FIXUP_ASUS_HEADSET_MIC),
Signature · 1 line
2.39.2
author maintainer
#50 lore raw

PATCH 50/58 ASoC: SOF: core: Free the firmware trace before calling snd_sof_shutdown()

In reply to Sasha Levin

From: Peter Ujfalusi <[email protected]>

[ Upstream commit d389dcb3a48cec4f03c16434c0bf98a4c635372a ]

The shutdown is called on reboot/shutdown of the machine.
At this point the firmware tracing cannot be used anymore but in case of IPC3 it is using and keeping a DMA channel active (dtrace).

For Tiger Lake platforms we have a quirk in place to fix rare reboot issues when a DMA was active before rebooting the system.
If the tracing is enabled this quirk will be always used and a print appears on the kernel log which might be misleading or not even correct.

Release the fw tracing before executing the shutdown to make sure that this known DMA user is cleared away.

---
 sound/soc/sof/core.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
sound/soc/sof/core.c +3 -1
index 9a9d82220fd0d..30db685cc5f4b 100644
@@ -504,8 +504,10 @@ int snd_sof_device_shutdown(struct device *dev)
 	if (IS_ENABLED(CONFIG_SND_SOC_SOF_PROBE_WORK_QUEUE))
 		cancel_work_sync(&sdev->probe_work);
 
-	if (sdev->fw_state == SOF_FW_BOOT_COMPLETE)
+	if (sdev->fw_state == SOF_FW_BOOT_COMPLETE) {
+		sof_fw_trace_free(sdev);
 		return snd_sof_shutdown(sdev);
+	}
 
 	return 0;
 }
Signature · 1 line
2.39.2
author maintainer
#55 lore raw

PATCH 55/58 fs: jfs: Fix UBSAN: array-index-out-of-bounds in dbAllocDmapLev

In reply to Sasha Levin

From: Yogesh <[email protected]>

[ Upstream commit 4e302336d5ca1767a06beee7596a72d3bdc8d983 ]

Syzkaller reported the following issue:

UBSAN: array-index-out-of-bounds in fs/jfs/jfs_dmap.c:1965:6
index -84 is out of range for type 's8[341]' (aka 'signed char[341]')
CPU: 1 PID: 4995 Comm: syz-executor146 Not tainted 6.4.0-rc6-syzkaller-00037-gb6dad5178cea #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/27/2023
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
 ubsan_epilogue lib/ubsan.c:217 [inline]
 __ubsan_handle_out_of_bounds+0x11c/0x150 lib/ubsan.c:348
 dbAllocDmapLev+0x3e5/0x430 fs/jfs/jfs_dmap.c:1965
 dbAllocCtl+0x113/0x920 fs/jfs/jfs_dmap.c:1809
 dbAllocAG+0x28f/0x10b0 fs/jfs/jfs_dmap.c:1350
 dbAlloc+0x658/0xca0 fs/jfs/jfs_dmap.c:874
 dtSplitUp fs/jfs/jfs_dtree.c:974 [inline]
 dtInsert+0xda7/0x6b00 fs/jfs/jfs_dtree.c:863
 jfs_create+0x7b6/0xbb0 fs/jfs/namei.c:137
 lookup_open fs/namei.c:3492 [inline]
 open_last_lookups fs/namei.c:3560 [inline]
 path_openat+0x13df/0x3170 fs/namei.c:3788
 do_filp_open+0x234/0x490 fs/namei.c:3818
 do_sys_openat2+0x13f/0x500 fs/open.c:1356
 do_sys_open fs/open.c:1372 [inline]
 __do_sys_openat fs/open.c:1388 [inline]
 __se_sys_openat fs/open.c:1383 [inline]
 __x64_sys_openat+0x247/0x290 fs/open.c:1383
 do_syscall_x64 arch/x86/entry/common.c:50 [inline]
 do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
 entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7f1f4e33f7e9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 51 14 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffc21129578 EFLAGS: 00000246 ORIG_RAX: 0000000000000101
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f1f4e33f7e9
RDX: 000000000000275a RSI: 0000000020000040 RDI: 00000000ffffff9c
RBP: 00007f1f4e2ff080 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f1f4e2ff110
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
 </TASK>

The bug occurs when the dbAllocDmapLev()function attempts to access dp->tree.stree[leafidx + LEAFIND] while the leafidx value is negative.

To rectify this, the patch introduces a safeguard within the dbAllocDmapLev() function. A check has been added to verify if leafidx is negative. If it is, the function immediately returns an I/O error, preventing any further execution that could potentially cause harm.

Tested via syzbot.

---
 fs/jfs/jfs_dmap.c | 3 +++
 1 file changed, 3 insertions(+)
fs/jfs/jfs_dmap.c +3 -0
index a3eb1e8269477..455300d38d315 100644
@@ -1953,6 +1953,9 @@ dbAllocDmapLev(struct bmap * bmp,
 	if (dbFindLeaf((dmtree_t *) & dp->tree, l2nb, &leafidx))
 		return -ENOSPC;
 
+	if (leafidx < 0)
+		return -EIO;
+
 	/* determine the block number within the file system corresponding
 	 * to the leaf at which free space was found.
 	 */
Signature · 1 line
2.39.2
author maintainer
#56 lore raw

PATCH 56/58 MIPS: dec: prom: Address -Warray-bounds warning

In reply to Sasha Levin

From: "Gustavo A. R. Silva" <[email protected]>

[ Upstream commit 7b191b9b55df2a844bd32d1d380f47a7df1c2896 ]

Zero-length arrays are deprecated, and we are replacing them with flexible array members instead. So, replace zero-length array with flexible-array member in struct memmap.

Address the following warning found after building (with GCC-13) mips64
with decstation_64_defconfig:
In function 'rex_setup_memory_region',
inlined from 'prom_meminit' at arch/mips/dec/prom/memory.c:91:3:
arch/mips/dec/prom/memory.c:72:31: error: array subscript i is outside array bounds of 'unsigned char[0]' [-Werror=array-bounds=]
72 | if (bm->bitmap[i] == 0xff)

      |                     ~~~~~~~~~~^~~

In file included from arch/mips/dec/prom/memory.c:16:
./arch/mips/include/asm/dec/prom.h: In function 'prom_meminit':
./arch/mips/include/asm/dec/prom.h:73:23: note: while referencing 'bitmap'
73 | unsigned char bitmap[0];

This helps with the ongoing efforts to globally enable -Warray-bounds.

This results in no differences in binary output.

---
 arch/mips/include/asm/dec/prom.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/include/asm/dec/prom.h +1 -1
index 1e1247add1cf8..908e96e3a3117 100644
@@ -70,7 +70,7 @@ static inline bool prom_is_rex(u32 magic)
  */
 typedef struct {
 	int pagesize;
-	unsigned char bitmap[0];
+	unsigned char bitmap[];
 } memmap;
 
 
Signature · 1 line
2.39.2
author maintainer
#58 lore raw

PATCH 58/58 FS: JFS: Check for read-only mounted filesystem in txBegin

In reply to Sasha Levin

From: Immad Mir <[email protected]>

[ Upstream commit 95e2b352c03b0a86c5717ba1d24ea20969abcacc ]

This patch adds a check for read-only mounted filesystem in txBegin before starting a transaction potentially saving from NULL pointer deref.

---
 fs/jfs/jfs_txnmgr.c | 5 +++++
 1 file changed, 5 insertions(+)
fs/jfs/jfs_txnmgr.c +5 -0
index ffd4feece0785..ce4b4760fcb1d 100644
@@ -354,6 +354,11 @@ tid_t txBegin(struct super_block *sb, int flag)
 	jfs_info("txBegin: flag = 0x%x", flag);
 	log = JFS_SBI(sb)->log;
 
+	if (!log) {
+		jfs_error(sb, "read-only filesystem\n");
+		return 0;
+	}
+
 	TXN_LOCK();
 
 	INCREMENT(TxStat.txBegin);
Signature · 1 line
2.39.2
author maintainer
#46 lore raw

PATCH 46/58 RDMA/mlx5: Return the firmware result upon destroying QP/RQ

In reply to Sasha Levin

From: Patrisious Haddad <[email protected]>

[ Upstream commit 22664c06e997087fe37f9ba208008c948571214a ]

Previously when destroying a QP/RQ, the result of the firmware destruction function was ignored and upper layers weren't informed about the failure.
Which in turn could lead to various problems since when upper layer isn't aware of the failure it continues its operation thinking that the related QP/RQ was successfully destroyed while it actually wasn't, which could lead to the below kernel WARN.

Currently, we return the correct firmware destruction status to upper layers which in case of the RQ would be mlx5_ib_destroy_wq() which was already capable of handling RQ destruction failure or in case of a QP to destroy_qp_common(), which now would actually warn upon qp destruction failure.

WARNING: CPU: 3 PID: 995 at drivers/infiniband/core/rdma_core.c:940 uverbs_destroy_ufile_hw+0xcb/0xe0 [ib_uverbs]
Modules linked in: xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink xt_addrtype iptable_nat nf_nat br_netfilter rpcrdma rdma_ucm ib_iser libiscsi scsi_transport_iscsi rdma_cm ib_umad ib_ipoib iw_cm ib_cm mlx5_ib ib_uverbs ib_core overlay mlx5_core fuse
CPU: 3 PID: 995 Comm: python3 Not tainted 5.16.0-rc5+ #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
RIP: 0010:uverbs_destroy_ufile_hw+0xcb/0xe0 [ib_uverbs]
Code: 41 5c 41 5d 41 5e e9 44 34 f0 e0 48 89 df e8 4c 77 ff ff 49 8b 86 10 01 00 00 48 85 c0 74 a1 4c 89 e7 ff d0 eb 9a 0f 0b eb c1 <0f> 0b be 04 00 00 00 48 89 df e8 b6 f6 ff ff e9 75 ff ff ff 90 0f
RSP: 0018:ffff8881533e3e78 EFLAGS: 00010287
RAX: ffff88811b2cf3e0 RBX: ffff888106209700 RCX: 0000000000000000
RDX: ffff888106209780 RSI: ffff8881533e3d30 RDI: ffff888109b101a0
RBP: 0000000000000001 R08: ffff888127cb381c R09: 0de9890000000009
R10: ffff888127cb3800 R11: 0000000000000000 R12: ffff888106209780
R13: ffff888106209750 R14: ffff888100f20660 R15: 0000000000000000
FS:  00007f8be353b740(0000) GS:ffff88852c980000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f8bd5b117c0 CR3: 000000012cd8a004 CR4: 0000000000370ea0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 ib_uverbs_close+0x1a/0x90 [ib_uverbs]
 __fput+0x82/0x230
 task_work_run+0x59/0x90
 exit_to_user_mode_prepare+0x138/0x140
 syscall_exit_to_user_mode+0x1d/0x50
 ? __x64_sys_close+0xe/0x40
 do_syscall_64+0x4a/0x90
 entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f8be3ae0abb
Code: 03 00 00 00 0f 05 48 3d 00 f0 ff ff 77 41 c3 48 83 ec 18 89 7c 24 0c e8 83 43 f9 ff 8b 7c 24 0c 41 89 c0 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 89 44 24 0c e8 c1 43 f9 ff 8b 44
RSP: 002b:00007ffdb51909c0 EFLAGS: 00000293 ORIG_RAX: 0000000000000003
RAX: 0000000000000000 RBX: 0000557bb7f7c020 RCX: 00007f8be3ae0abb
RDX: 0000557bb7c74010 RSI: 0000557bb7f14ca0 RDI: 0000000000000005
RBP: 0000557bb7fbd598 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000293 R12: 0000557bb7fbd5b8
R13: 0000557bb7fbd5a8 R14: 0000000000001000 R15: 0000557bb7f7c020
 </TASK>
---
 drivers/infiniband/hw/mlx5/qpc.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/infiniband/hw/mlx5/qpc.c +4 -6
index bae0334d6e7f1..aec011557b4a7 100644
@@ -298,8 +298,7 @@ int mlx5_core_destroy_qp(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp)
 	MLX5_SET(destroy_qp_in, in, opcode, MLX5_CMD_OP_DESTROY_QP);
 	MLX5_SET(destroy_qp_in, in, qpn, qp->qpn);
 	MLX5_SET(destroy_qp_in, in, uid, qp->uid);
-	mlx5_cmd_exec_in(dev->mdev, destroy_qp, in);
-	return 0;
+	return mlx5_cmd_exec_in(dev->mdev, destroy_qp, in);
 }
 
 int mlx5_core_set_delay_drop(struct mlx5_ib_dev *dev,
@@ -551,14 +550,14 @@ int mlx5_core_xrcd_dealloc(struct mlx5_ib_dev *dev, u32 xrcdn)
 	return mlx5_cmd_exec_in(dev->mdev, dealloc_xrcd, in);
 }
 
-static void destroy_rq_tracked(struct mlx5_ib_dev *dev, u32 rqn, u16 uid)
+static int destroy_rq_tracked(struct mlx5_ib_dev *dev, u32 rqn, u16 uid)
 {
 	u32 in[MLX5_ST_SZ_DW(destroy_rq_in)] = {};
 
 	MLX5_SET(destroy_rq_in, in, opcode, MLX5_CMD_OP_DESTROY_RQ);
 	MLX5_SET(destroy_rq_in, in, rqn, rqn);
 	MLX5_SET(destroy_rq_in, in, uid, uid);
-	mlx5_cmd_exec_in(dev->mdev, destroy_rq, in);
+	return mlx5_cmd_exec_in(dev->mdev, destroy_rq, in);
 }
 
 int mlx5_core_create_rq_tracked(struct mlx5_ib_dev *dev, u32 *in, int inlen,
@@ -589,8 +588,7 @@ int mlx5_core_destroy_rq_tracked(struct mlx5_ib_dev *dev,
 				 struct mlx5_core_qp *rq)
 {
 	destroy_resource_common(dev, rq);
-	destroy_rq_tracked(dev, rq->qpn, rq->uid);
-	return 0;
+	return destroy_rq_tracked(dev, rq->qpn, rq->uid);
 }
 
 static void destroy_sq_tracked(struct mlx5_ib_dev *dev, u32 sqn, u16 uid)
Signature · 1 line
2.39.2
author maintainer
#51 lore raw

PATCH 51/58 ovl: check type and offset of struct vfsmount in ovl_entry

In reply to Sasha Levin

From: Christian Brauner <[email protected]>

[ Upstream commit f723edb8a532cd26e1ff0a2b271d73762d48f762 ]

Porting overlayfs to the new amount api I started experiencing random crashes that couldn't be explained easily. So after much debugging and reasoning it became clear that struct ovl_entry requires the point to struct vfsmount to be the first member and of type struct vfsmount.

During the port I added a new member at the beginning of struct ovl_entry which broke all over the place in the form of random crashes and cache corruptions. While there's a comment in ovl_free_fs() to the effect of "Hack! Reuse ofs->layers as a vfsmount array before freeing it" there's no such comment on struct ovl_entry which makes this easy to trip over.

Add a comment and two static asserts for both the offset and the type of pointer in struct ovl_entry.

---
 fs/overlayfs/ovl_entry.h | 9 +++++++++
 1 file changed, 9 insertions(+)
fs/overlayfs/ovl_entry.h +9 -0
index fd11fe6d6d45f..6b9f7917fc1bb 100644
@@ -32,6 +32,7 @@ struct ovl_sb {
 };
 
 struct ovl_layer {
+	/* ovl_free_fs() relies on @mnt being the first member! */
 	struct vfsmount *mnt;
 	/* Trap in ovl inode cache */
 	struct inode *trap;
@@ -42,6 +43,14 @@ struct ovl_layer {
 	int fsid;
 };
 
+/*
+ * ovl_free_fs() relies on @mnt being the first member when unmounting
+ * the private mounts created for each layer. Let's check both the
+ * offset and type.
+ */
+static_assert(offsetof(struct ovl_layer, mnt) == 0);
+static_assert(__same_type(typeof_member(struct ovl_layer, mnt), struct vfsmount *));
+
 struct ovl_path {
 	const struct ovl_layer *layer;
 	struct dentry *dentry;
Signature · 1 line
2.39.2
author maintainer
#57 lore raw

PATCH 57/58 FS: JFS: Fix null-ptr-deref Read in txBegin

In reply to Sasha Levin

From: Immad Mir <[email protected]>

[ Upstream commit 47cfdc338d674d38f4b2f22b7612cc6a2763ba27 ]

Syzkaller reported an issue where txBegin may be called on a superblock in a read-only mounted filesystem which leads to NULL pointer deref. This could be solved by checking if the filesystem is read-only before calling txBegin, and returning with appropiate error code.

---
 fs/jfs/namei.c | 5 +++++
 1 file changed, 5 insertions(+)
fs/jfs/namei.c +5 -0
index b29d68b5eec53..f370c76051205 100644
@@ -799,6 +799,11 @@ static int jfs_link(struct dentry *old_dentry,
 	if (rc)
 		goto out;
 
+	if (isReadOnly(ip)) {
+		jfs_error(ip->i_sb, "read-only filesystem\n");
+		return -EROFS;
+	}
+
 	tid = txBegin(ip->i_sb, 0);
 
 	mutex_lock_nested(&JFS_IP(dir)->commit_mutex, COMMIT_MUTEX_PARENT);
Signature · 1 line
2.39.2
author maintainer
#47 lore raw

PATCH 47/58 RDMA/bnxt_re: consider timeout of destroy ah as success.

In reply to Sasha Levin

From: Kashyap Desai <[email protected]>

[ Upstream commit bb8c93618fb0b8567d309f1aebc6df0cd31da1a2 ]

If destroy_ah is timed out, it is likely to be destroyed by firmware but it is taking longer time due to temporary slowness
in processing the rcfw command. In worst case, there might be AH resource leak in firmware.

Sending timeout return value can dump warning message from ib_core which can be avoided if we map timeout of destroy_ah as success.

---
 drivers/infiniband/hw/bnxt_re/bnxt_re.h  |  2 ++
 drivers/infiniband/hw/bnxt_re/ib_verbs.c | 16 ++++++++++++----
 drivers/infiniband/hw/bnxt_re/qplib_sp.c |  8 +++++---
 drivers/infiniband/hw/bnxt_re/qplib_sp.h |  4 ++--
 4 files changed, 21 insertions(+), 9 deletions(-)
drivers/infiniband/hw/bnxt_re/bnxt_re.h +2 -0
index 2c95e6f3d47ac..eef3ef3fabb42 100644
@@ -179,6 +179,8 @@ struct bnxt_re_dev {
 #define BNXT_RE_ROCEV2_IPV4_PACKET	2
 #define BNXT_RE_ROCEV2_IPV6_PACKET	3
 
+#define BNXT_RE_CHECK_RC(x) ((x) && ((x) != -ETIMEDOUT))
+
 static inline struct device *rdev_to_dev(struct bnxt_re_dev *rdev)
 {
 	if (rdev)
drivers/infiniband/hw/bnxt_re/ib_verbs.c +12 -4
index 952811c40c54b..6a086c42f85f6 100644
@@ -614,12 +614,20 @@ int bnxt_re_destroy_ah(struct ib_ah *ib_ah, u32 flags)
 {
 	struct bnxt_re_ah *ah = container_of(ib_ah, struct bnxt_re_ah, ib_ah);
 	struct bnxt_re_dev *rdev = ah->rdev;
+	bool block = true;
+	int rc = 0;
 
-	bnxt_qplib_destroy_ah(&rdev->qplib_res, &ah->qplib_ah,
-			      !(flags & RDMA_DESTROY_AH_SLEEPABLE));
+	block = !(flags & RDMA_DESTROY_AH_SLEEPABLE);
+	rc = bnxt_qplib_destroy_ah(&rdev->qplib_res, &ah->qplib_ah, block);
+	if (BNXT_RE_CHECK_RC(rc)) {
+		if (rc == -ETIMEDOUT)
+			rc = 0;
+		else
+			goto fail;
+	}
 	atomic_dec(&rdev->ah_count);
-
-	return 0;
+fail:
+	return rc;
 }
 
 static u8 bnxt_re_stack_to_dev_nw_type(enum rdma_network_type ntype)
drivers/infiniband/hw/bnxt_re/qplib_sp.c +5 -3
index b967a17a44beb..10919532bca29 100644
@@ -468,13 +468,14 @@ int bnxt_qplib_create_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
 	return 0;
 }
 
-void bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
-			   bool block)
+int bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
+			  bool block)
 {
 	struct bnxt_qplib_rcfw *rcfw = res->rcfw;
 	struct creq_destroy_ah_resp resp = {};
 	struct bnxt_qplib_cmdqmsg msg = {};
 	struct cmdq_destroy_ah req = {};
+	int rc;
 
 	/* Clean up the AH table in the device */
 	bnxt_qplib_rcfw_cmd_prep((struct cmdq_base *)&req,
@@ -485,7 +486,8 @@ void bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
 
 	bnxt_qplib_fill_cmdqmsg(&msg, &req, &resp, NULL, sizeof(req),
 				sizeof(resp), block);
-	bnxt_qplib_rcfw_send_message(rcfw, &msg);
+	rc = bnxt_qplib_rcfw_send_message(rcfw, &msg);
+	return rc;
 }
 
 /* MRW */
drivers/infiniband/hw/bnxt_re/qplib_sp.h +2 -2
index 5de874659cdfa..4061616048e85 100644
@@ -327,8 +327,8 @@ int bnxt_qplib_set_func_resources(struct bnxt_qplib_res *res,
 				  struct bnxt_qplib_ctx *ctx);
 int bnxt_qplib_create_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
 			 bool block);
-void bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
-			   bool block);
+int bnxt_qplib_destroy_ah(struct bnxt_qplib_res *res, struct bnxt_qplib_ah *ah,
+			  bool block);
 int bnxt_qplib_alloc_mrw(struct bnxt_qplib_res *res,
 			 struct bnxt_qplib_mrw *mrw);
 int bnxt_qplib_dereg_mrw(struct bnxt_qplib_res *res, struct bnxt_qplib_mrw *mrw,
Signature · 1 line
2.39.2
#28 lore raw

Re: [PATCH AUTOSEL 6.4 28/58] ALSA: hda: Add Loongson LS7A HD-Audio support

In reply to Sasha Levin

Hi Sasha,

Quoted 10 lines · Sasha Levin

From: Yanteng Si <[email protected]>

[ Upstream commit 28bd137a3c8e105587ba8c55b68ef43b519b270f ]

Add the new PCI ID 0x0014 0x7a07 and the new PCI ID 0x0014 0x7a37 Loongson HDA controller.

Loongson HDA can't work if AUTOSEL only ports this one patch, because 6.4 also needs the other three patches inside this thread. Also, 6.1, 5.15, and 5.10 have the same problem.

Give me a shout if you need anything.

Thanks,
Yanteng

Quoted 79 lines
---
  include/linux/pci_ids.h    | 3 +++
  sound/hda/hdac_device.c    | 1 +
  sound/pci/hda/hda_intel.c  | 7 +++++++
  sound/pci/hda/patch_hdmi.c | 1 +
  4 files changed, 12 insertions(+)
include/linux/pci_ids.h +3 -0
index 95f33dadb2be2..c0c4ca8e28510 100644
@@ -158,6 +158,9 @@
 
  #define PCI_VENDOR_ID_LOONGSON		0x0014
 
+#define PCI_DEVICE_ID_LOONGSON_HDA      0x7a07
+#define PCI_DEVICE_ID_LOONGSON_HDMI     0x7a37
+
  #define PCI_VENDOR_ID_TTTECH		0x0357
  #define PCI_DEVICE_ID_TTTECH_MC322	0x000a
 
sound/hda/hdac_device.c +1 -0
index 6c043fbd606f1..bbf7bcdb449a8 100644
@@ -645,6 +645,7 @@ struct hda_vendor_id {
  };
 
  static const struct hda_vendor_id hda_vendor_ids[] = {
+	{ 0x0014, "Loongson" },
  	{ 0x1002, "ATI" },
  	{ 0x1013, "Cirrus Logic" },
  	{ 0x1057, "Motorola" },
sound/pci/hda/hda_intel.c +7 -0
index 3226691ac923c..9c353dc7740c4 100644
@@ -237,6 +237,7 @@ enum {
  	AZX_DRIVER_CTHDA,
  	AZX_DRIVER_CMEDIA,
  	AZX_DRIVER_ZHAOXIN,
+	AZX_DRIVER_LOONGSON,
  	AZX_DRIVER_GENERIC,
  	AZX_NUM_DRIVERS, /* keep this as last entry */
  };
@@ -360,6 +361,7 @@ static const char * const driver_short_names[] = {
  	[AZX_DRIVER_CTHDA] = "HDA Creative",
  	[AZX_DRIVER_CMEDIA] = "HDA C-Media",
  	[AZX_DRIVER_ZHAOXIN] = "HDA Zhaoxin",
+	[AZX_DRIVER_LOONGSON] = "HDA Loongson",
  	[AZX_DRIVER_GENERIC] = "HD-Audio Generic",
  };
 
@@ -2809,6 +2811,11 @@ static const struct pci_device_id azx_ids[] = {
  	  .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI },
  	/* Zhaoxin */
  	{ PCI_DEVICE(0x1d17, 0x3288), .driver_data = AZX_DRIVER_ZHAOXIN },
+	/* Loongson HDAudio*/
+	{PCI_DEVICE(PCI_VENDOR_ID_LOONGSON, PCI_DEVICE_ID_LOONGSON_HDA),
+	  .driver_data = AZX_DRIVER_LOONGSON },
+	{PCI_DEVICE(PCI_VENDOR_ID_LOONGSON, PCI_DEVICE_ID_LOONGSON_HDMI),
+	  .driver_data = AZX_DRIVER_LOONGSON },
  	{ 0, }
  };
  MODULE_DEVICE_TABLE(pci, azx_ids);
sound/pci/hda/patch_hdmi.c +1 -0
index 5c0b1a09fd57c..260d3e64f6589 100644
@@ -4505,6 +4505,7 @@ static int patch_gf_hdmi(struct hda_codec *codec)
   * patch entries
   */
  static const struct hda_device_id snd_hda_id_hdmi[] = {
+HDA_CODEC_ENTRY(0x00147a47, "Loongson HDMI",	patch_generic_hdmi),
  HDA_CODEC_ENTRY(0x1002793c, "RS600 HDMI",	patch_atihdmi),
  HDA_CODEC_ENTRY(0x10027919, "RS600 HDMI",	patch_atihdmi),
  HDA_CODEC_ENTRY(0x1002791a, "RS690/780 HDMI",	patch_atihdmi),