6.12.64-rc1 review

@@ -191,11 +191,11 @@
 #define SSC_STATUS_PLL_LOCK_MASK	0x800
 #define PCIE_BRCM_MAX_MEMC		3
 
-#define IDX_ADDR(pcie)			((pcie)->reg_offsets[EXT_CFG_INDEX])
-#define DATA_ADDR(pcie)			((pcie)->reg_offsets[EXT_CFG_DATA])
-#define PCIE_RGR1_SW_INIT_1(pcie)	((pcie)->reg_offsets[RGR1_SW_INIT_1])
-#define HARD_DEBUG(pcie)		((pcie)->reg_offsets[PCIE_HARD_DEBUG])
-#define INTR2_CPU_BASE(pcie)		((pcie)->reg_offsets[PCIE_INTR2_CPU_BASE])
+#define IDX_ADDR(pcie)			((pcie)->cfg->offsets[EXT_CFG_INDEX])
+#define DATA_ADDR(pcie)			((pcie)->cfg->offsets[EXT_CFG_DATA])
+#define PCIE_RGR1_SW_INIT_1(pcie)	((pcie)->cfg->offsets[RGR1_SW_INIT_1])
+#define HARD_DEBUG(pcie)		((pcie)->cfg->offsets[PCIE_HARD_DEBUG])
+#define INTR2_CPU_BASE(pcie)		((pcie)->cfg->offsets[PCIE_INTR2_CPU_BASE])
 
 /* Rescal registers */
 #define PCIE_DVT_PMU_PCIE_PHY_CTRL				0xc700

@@ -276,8 +276,6 @@ struct brcm_pcie {
 	int			gen;
 	u64			msi_target_addr;
 	struct brcm_msi		*msi;
-	const int		*reg_offsets;
-	enum pcie_soc_base	soc_base;
 	struct reset_control	*rescal;
 	struct reset_control	*perst_reset;
 	struct reset_control	*bridge_reset;

@@ -285,17 +283,14 @@ struct brcm_pcie {
 	int			num_memc;
 	u64			memc_size[PCIE_BRCM_MAX_MEMC];
 	u32			hw_rev;
-	int			(*perst_set)(struct brcm_pcie *pcie, u32 val);
-	int			(*bridge_sw_init_set)(struct brcm_pcie *pcie, u32 val);
 	struct subdev_regulators *sr;
 	bool			ep_wakeup_capable;
-	bool			has_phy;
-	u8			num_inbound_wins;
+	const struct pcie_cfg_data	*cfg;
 };
 
 static inline bool is_bmips(const struct brcm_pcie *pcie)
 {
-	return pcie->soc_base == BCM7435 || pcie->soc_base == BCM7425;
+	return pcie->cfg->soc_base == BCM7435 || pcie->cfg->soc_base == BCM7425;
 }
 
 /*

@@ -855,7 +850,7 @@ static int brcm_pcie_get_inbound_wins(st
 	 * security considerations, and is not implemented in our modern
 	 * SoCs.
 	 */
-	if (pcie->soc_base != BCM7712)
+	if (pcie->cfg->soc_base != BCM7712)
 		add_inbound_win(b++, &n, 0, 0, 0);
 
 	resource_list_for_each_entry(entry, &bridge->dma_ranges) {

@@ -872,10 +867,10 @@ static int brcm_pcie_get_inbound_wins(st
 		 * That being said, each BARs size must still be a power of
 		 * two.
 		 */
-		if (pcie->soc_base == BCM7712)
+		if (pcie->cfg->soc_base == BCM7712)
 			add_inbound_win(b++, &n, size, cpu_start, pcie_start);
 
-		if (n > pcie->num_inbound_wins)
+		if (n > pcie->cfg->num_inbound_wins)
 			break;
 	}
 

@@ -889,7 +884,7 @@ static int brcm_pcie_get_inbound_wins(st
 	 * that enables multiple memory controllers.  As such, it can return
 	 * now w/o doing special configuration.
 	 */
-	if (pcie->soc_base == BCM7712)
+	if (pcie->cfg->soc_base == BCM7712)
 		return n;
 
 	ret = of_property_read_variable_u64_array(pcie->np, "brcm,scb-sizes", pcie->memc_size, 1,

@@ -1012,7 +1007,7 @@ static void set_inbound_win_registers(st
 		 * 7712:
 		 *     All of their BARs need to be set.
 		 */
-		if (pcie->soc_base == BCM7712) {
+		if (pcie->cfg->soc_base == BCM7712) {
 			/* BUS remap register settings */
 			reg_offset = brcm_ubus_reg_offset(i);
 			tmp = lower_32_bits(cpu_addr) & ~0xfff;

@@ -1036,15 +1031,15 @@ static int brcm_pcie_setup(struct brcm_p
 	int memc, ret;
 
 	/* Reset the bridge */
-	ret = pcie->bridge_sw_init_set(pcie, 1);
+	ret = pcie->cfg->bridge_sw_init_set(pcie, 1);
 	if (ret)
 		return ret;
 
 	/* Ensure that PERST# is asserted; some bootloaders may deassert it. */
-	if (pcie->soc_base == BCM2711) {
-		ret = pcie->perst_set(pcie, 1);
+	if (pcie->cfg->soc_base == BCM2711) {
+		ret = pcie->cfg->perst_set(pcie, 1);
 		if (ret) {
-			pcie->bridge_sw_init_set(pcie, 0);
+			pcie->cfg->bridge_sw_init_set(pcie, 0);
 			return ret;
 		}
 	}

@@ -1052,7 +1047,7 @@ static int brcm_pcie_setup(struct brcm_p
 	usleep_range(100, 200);
 
 	/* Take the bridge out of reset */
-	ret = pcie->bridge_sw_init_set(pcie, 0);
+	ret = pcie->cfg->bridge_sw_init_set(pcie, 0);
 	if (ret)
 		return ret;
 

@@ -1072,9 +1067,9 @@ static int brcm_pcie_setup(struct brcm_p
 	 */
 	if (is_bmips(pcie))
 		burst = 0x1; /* 256 bytes */
-	else if (pcie->soc_base == BCM2711)
+	else if (pcie->cfg->soc_base == BCM2711)
 		burst = 0x0; /* 128 bytes */
-	else if (pcie->soc_base == BCM7278)
+	else if (pcie->cfg->soc_base == BCM7278)
 		burst = 0x3; /* 512 bytes */
 	else
 		burst = 0x2; /* 512 bytes */

@@ -1199,7 +1194,7 @@ static void brcm_extend_rbus_timeout(str
 	u32 timeout_us = 4000000; /* 4 seconds, our setting for L1SS */
 
 	/* 7712 does not have this (RGR1) timer */
-	if (pcie->soc_base == BCM7712)
+	if (pcie->cfg->soc_base == BCM7712)
 		return;
 
 	/* Each unit in timeout register is 1/216,000,000 seconds */

@@ -1281,7 +1276,7 @@ static int brcm_pcie_start_link(struct b
 		brcm_pcie_set_gen(pcie, pcie->gen);
 
 	/* Unassert the fundamental reset */
-	ret = pcie->perst_set(pcie, 0);
+	ret = pcie->cfg->perst_set(pcie, 0);
 	if (ret)
 		return ret;
 

@@ -1465,12 +1460,12 @@ static int brcm_phy_cntl(struct brcm_pci
 
 static inline int brcm_phy_start(struct brcm_pcie *pcie)
 {
-	return pcie->has_phy ? brcm_phy_cntl(pcie, 1) : 0;
+	return pcie->cfg->has_phy ? brcm_phy_cntl(pcie, 1) : 0;
 }
 
 static inline int brcm_phy_stop(struct brcm_pcie *pcie)
 {
-	return pcie->has_phy ? brcm_phy_cntl(pcie, 0) : 0;
+	return pcie->cfg->has_phy ? brcm_phy_cntl(pcie, 0) : 0;
 }
 
 static int brcm_pcie_turn_off(struct brcm_pcie *pcie)

@@ -1481,7 +1476,7 @@ static int brcm_pcie_turn_off(struct brc
 	if (brcm_pcie_link_up(pcie))
 		brcm_pcie_enter_l23(pcie);
 	/* Assert fundamental reset */
-	ret = pcie->perst_set(pcie, 1);
+	ret = pcie->cfg->perst_set(pcie, 1);
 	if (ret)
 		return ret;
 

@@ -1496,7 +1491,7 @@ static int brcm_pcie_turn_off(struct brc
 	writel(tmp, base + HARD_DEBUG(pcie));
 
 	/* Shutdown PCIe bridge */
-	ret = pcie->bridge_sw_init_set(pcie, 1);
+	ret = pcie->cfg->bridge_sw_init_set(pcie, 1);
 
 	return ret;
 }

@@ -1584,7 +1579,7 @@ static int brcm_pcie_resume_noirq(struct
 		goto err_reset;
 
 	/* Take bridge out of reset so we can access the SERDES reg */
-	pcie->bridge_sw_init_set(pcie, 0);
+	pcie->cfg->bridge_sw_init_set(pcie, 0);
 
 	/* SERDES_IDDQ = 0 */
 	tmp = readl(base + HARD_DEBUG(pcie));

@@ -1805,12 +1800,7 @@ static int brcm_pcie_probe(struct platfo
 	pcie = pci_host_bridge_priv(bridge);
 	pcie->dev = &pdev->dev;
 	pcie->np = np;
-	pcie->reg_offsets = data->offsets;
-	pcie->soc_base = data->soc_base;
-	pcie->perst_set = data->perst_set;
-	pcie->bridge_sw_init_set = data->bridge_sw_init_set;
-	pcie->has_phy = data->has_phy;
-	pcie->num_inbound_wins = data->num_inbound_wins;
+	pcie->cfg = data;
 
 	pcie->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(pcie->base))

@@ -1845,7 +1835,7 @@ static int brcm_pcie_probe(struct platfo
 	if (ret)
 		return dev_err_probe(&pdev->dev, ret, "could not enable clock\n");
 
-	pcie->bridge_sw_init_set(pcie, 0);
+	pcie->cfg->bridge_sw_init_set(pcie, 0);
 
 	if (pcie->swinit_reset) {
 		ret = reset_control_assert(pcie->swinit_reset);

@@ -1884,7 +1874,8 @@ static int brcm_pcie_probe(struct platfo
 		goto fail;
 
 	pcie->hw_rev = readl(pcie->base + PCIE_MISC_REVISION);
-	if (pcie->soc_base == BCM4908 && pcie->hw_rev >= BRCM_PCIE_HW_REV_3_20) {
+	if (pcie->cfg->soc_base == BCM4908 &&
+	    pcie->hw_rev >= BRCM_PCIE_HW_REV_3_20) {
 		dev_err(pcie->dev, "hardware revision with unsupported PERST# setup\n");
 		ret = -ENODEV;
 		goto fail;

@@ -1904,7 +1895,8 @@ static int brcm_pcie_probe(struct platfo
 		}
 	}
 
-	bridge->ops = pcie->soc_base == BCM7425 ? &brcm7425_pcie_ops : &brcm_pcie_ops;
+	bridge->ops = pcie->cfg->soc_base == BCM7425 ?
+				&brcm7425_pcie_ops : &brcm_pcie_ops;
 	bridge->sysdata = pcie;
 
 	platform_set_drvdata(pdev, pcie);

@@ -46,6 +46,7 @@
 #define  PCIE_RC_CFG_PRIV1_ID_VAL3_CLASS_CODE_MASK	0xffffff
 
 #define PCIE_RC_CFG_PRIV1_LINK_CAPABILITY			0x04dc
+#define  PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_MAX_LINK_WIDTH_MASK	0x1f0
 #define  PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_ASPM_SUPPORT_MASK	0xc00
 
 #define PCIE_RC_CFG_PRIV1_ROOT_CAP			0x4f8

@@ -55,6 +56,9 @@
 #define PCIE_RC_DL_MDIO_WR_DATA				0x1104
 #define PCIE_RC_DL_MDIO_RD_DATA				0x1108
 
+#define PCIE_RC_PL_REG_PHY_CTL_1			0x1804
+#define  PCIE_RC_PL_REG_PHY_CTL_1_REG_P2_POWERDOWN_ENA_NOSYNC_MASK	0x8
+
 #define PCIE_MISC_MISC_CTRL				0x4008
 #define  PCIE_MISC_MISC_CTRL_PCIE_RCB_64B_MODE_MASK	0x80
 #define  PCIE_MISC_MISC_CTRL_PCIE_RCB_MPS_MODE_MASK	0x400

@@ -1025,7 +1029,7 @@ static int brcm_pcie_setup(struct brcm_p
 	void __iomem *base = pcie->base;
 	struct pci_host_bridge *bridge;
 	struct resource_entry *entry;
-	u32 tmp, burst, aspm_support;
+	u32 tmp, burst, aspm_support, num_lanes, num_lanes_cap;
 	u8 num_out_wins = 0;
 	int num_inbound_wins = 0;
 	int memc, ret;

@@ -1133,6 +1137,27 @@ static int brcm_pcie_setup(struct brcm_p
 		PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_ASPM_SUPPORT_MASK);
 	writel(tmp, base + PCIE_RC_CFG_PRIV1_LINK_CAPABILITY);
 
+	/* 'tmp' still holds the contents of PRIV1_LINK_CAPABILITY */
+	num_lanes_cap = u32_get_bits(tmp, PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_MAX_LINK_WIDTH_MASK);
+	num_lanes = 0;
+
+	/*
+	 * Use hardware negotiated Max Link Width value by default.  If the
+	 * "num-lanes" DT property is present, assume that the chip's default
+	 * link width capability information is incorrect/undesired and use the
+	 * specified value instead.
+	 */
+	if (!of_property_read_u32(pcie->np, "num-lanes", &num_lanes) &&
+	    num_lanes && num_lanes <= 4 && num_lanes_cap != num_lanes) {
+		u32p_replace_bits(&tmp, num_lanes,
+			PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_MAX_LINK_WIDTH_MASK);
+		writel(tmp, base + PCIE_RC_CFG_PRIV1_LINK_CAPABILITY);
+		tmp = readl(base + PCIE_RC_PL_REG_PHY_CTL_1);
+		u32p_replace_bits(&tmp, 1,
+			PCIE_RC_PL_REG_PHY_CTL_1_REG_P2_POWERDOWN_ENA_NOSYNC_MASK);
+		writel(tmp, base + PCIE_RC_PL_REG_PHY_CTL_1);
+	}
+
 	/*
 	 * For config space accesses on the RC, show the right class for
 	 * a PCIe-PCIe bridge (the default setting is to be EP mode).

@@ -47,7 +47,6 @@
 
 #define PCIE_RC_CFG_PRIV1_LINK_CAPABILITY			0x04dc
 #define  PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_MAX_LINK_WIDTH_MASK	0x1f0
-#define  PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_ASPM_SUPPORT_MASK	0xc00
 
 #define PCIE_RC_CFG_PRIV1_ROOT_CAP			0x4f8
 #define  PCIE_RC_CFG_PRIV1_ROOT_CAP_L1SS_MODE_MASK	0xf8

@@ -1029,7 +1028,7 @@ static int brcm_pcie_setup(struct brcm_p
 	void __iomem *base = pcie->base;
 	struct pci_host_bridge *bridge;
 	struct resource_entry *entry;
-	u32 tmp, burst, aspm_support, num_lanes, num_lanes_cap;
+	u32 tmp, burst, num_lanes, num_lanes_cap;
 	u8 num_out_wins = 0;
 	int num_inbound_wins = 0;
 	int memc, ret;

@@ -1129,12 +1128,9 @@ static int brcm_pcie_setup(struct brcm_p
 
 
 	/* Don't advertise L0s capability if 'aspm-no-l0s' */
-	aspm_support = PCIE_LINK_STATE_L1;
-	if (!of_property_read_bool(pcie->np, "aspm-no-l0s"))
-		aspm_support |= PCIE_LINK_STATE_L0S;
 	tmp = readl(base + PCIE_RC_CFG_PRIV1_LINK_CAPABILITY);
-	u32p_replace_bits(&tmp, aspm_support,
-		PCIE_RC_CFG_PRIV1_LINK_CAPABILITY_ASPM_SUPPORT_MASK);
+	if (of_property_read_bool(pcie->np, "aspm-no-l0s"))
+		tmp &= ~PCI_EXP_LNKCAP_ASPM_L0S;
 	writel(tmp, base + PCIE_RC_CFG_PRIV1_LINK_CAPABILITY);
 
 	/* 'tmp' still holds the contents of PRIV1_LINK_CAPABILITY */

@@ -93,12 +93,6 @@ size_t balloon_page_list_dequeue(struct
 		if (!trylock_page(page))
 			continue;
 
-		if (IS_ENABLED(CONFIG_BALLOON_COMPACTION) &&
-		    PageIsolated(page)) {
-			/* raced with isolation */
-			unlock_page(page);
-			continue;
-		}
 		balloon_page_delete(page);
 		__count_vm_event(BALLOON_DEFLATE);
 		list_add(&page->lru, pages);

@@ -532,7 +532,6 @@ static int cmm_migratepage(struct balloo
 
 	spin_lock_irqsave(&b_dev_info->pages_lock, flags);
 	balloon_page_insert(b_dev_info, newpage);
-	balloon_page_delete(page);
 	b_dev_info->isolated_pages--;
 	spin_unlock_irqrestore(&b_dev_info->pages_lock, flags);
 

@@ -542,6 +541,7 @@ static int cmm_migratepage(struct balloo
 	 */
 	plpar_page_set_active(page);
 
+	balloon_page_finalize(page);
 	/* balloon page list reference */
 	put_page(page);
 

@@ -1778,8 +1778,7 @@ static int vmballoon_migratepage(struct
 	 * @pages_lock . We keep holding @comm_lock since we will need it in a
 	 * second.
 	 */
-	balloon_page_delete(page);
-
+	balloon_page_finalize(page);
 	put_page(page);
 
 	/* Inflate */

@@ -866,15 +866,13 @@ static int virtballoon_migratepage(struc
 	tell_host(vb, vb->inflate_vq);
 
 	/* balloon's page migration 2nd step -- deflate "page" */
-	spin_lock_irqsave(&vb_dev_info->pages_lock, flags);
-	balloon_page_delete(page);
-	spin_unlock_irqrestore(&vb_dev_info->pages_lock, flags);
 	vb->num_pfns = VIRTIO_BALLOON_PAGES_PER_PAGE;
 	set_page_pfns(vb, vb->pfns, page);
 	tell_host(vb, vb->deflate_vq);
 
 	mutex_unlock(&vb->balloon_lock);
 
+	balloon_page_finalize(page);
 	put_page(page); /* balloon reference */
 
 	return MIGRATEPAGE_SUCCESS;

61 lines not on this page · raw message

@@ -93,7 +93,8 @@ size_t balloon_page_list_dequeue(struct
 		if (!trylock_page(page))
 			continue;
 
-		balloon_page_delete(page);
+		list_del(&page->lru);
+		balloon_page_finalize(page);
 		__count_vm_event(BALLOON_DEFLATE);
 		list_add(&page->lru, pages);
 		unlock_page(page);

@@ -532,6 +532,7 @@ static int cmm_migratepage(struct balloo
 
 	spin_lock_irqsave(&b_dev_info->pages_lock, flags);
 	balloon_page_insert(b_dev_info, newpage);
+	__count_vm_event(BALLOON_MIGRATE);
 	b_dev_info->isolated_pages--;
 	spin_unlock_irqrestore(&b_dev_info->pages_lock, flags);
 

@@ -47,30 +47,32 @@ static void mtk_vcodec_vpu_reset_dec_han
 {
 	struct mtk_vcodec_dec_dev *dev = priv;
 	struct mtk_vcodec_dec_ctx *ctx;
+	unsigned long flags;
 
 	dev_err(&dev->plat_dev->dev, "Watchdog timeout!!");
 
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_for_each_entry(ctx, &dev->ctx_list, list) {
 		ctx->state = MTK_STATE_ABORT;
 		mtk_v4l2_vdec_dbg(0, ctx, "[%d] Change to state MTK_STATE_ABORT", ctx->id);
 	}
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 }
 
 static void mtk_vcodec_vpu_reset_enc_handler(void *priv)
 {
 	struct mtk_vcodec_enc_dev *dev = priv;
 	struct mtk_vcodec_enc_ctx *ctx;
+	unsigned long flags;
 
 	dev_err(&dev->plat_dev->dev, "Watchdog timeout!!");
 
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_for_each_entry(ctx, &dev->ctx_list, list) {
 		ctx->state = MTK_STATE_ABORT;
 		mtk_v4l2_vdec_dbg(0, ctx, "[%d] Change to state MTK_STATE_ABORT", ctx->id);
 	}
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 }
 
 static const struct mtk_vcodec_fw_ops mtk_vcodec_vpu_msg = {

@@ -198,6 +198,7 @@ static int fops_vcodec_open(struct file
 	struct mtk_vcodec_dec_ctx *ctx = NULL;
 	int ret = 0, i, hw_count;
 	struct vb2_queue *src_vq;
+	unsigned long flags;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)

@@ -268,9 +269,9 @@ static int fops_vcodec_open(struct file
 
 	ctx->dev->vdec_pdata->init_vdec_params(ctx);
 
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_add(&ctx->list, &dev->ctx_list);
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 	mtk_vcodec_dbgfs_create(ctx);
 
 	mutex_unlock(&dev->dev_mutex);

@@ -295,6 +296,7 @@ static int fops_vcodec_release(struct fi
 {
 	struct mtk_vcodec_dec_dev *dev = video_drvdata(file);
 	struct mtk_vcodec_dec_ctx *ctx = fh_to_dec_ctx(file->private_data);
+	unsigned long flags;
 
 	mtk_v4l2_vdec_dbg(0, ctx, "[%d] decoder", ctx->id);
 	mutex_lock(&dev->dev_mutex);

@@ -313,9 +315,9 @@ static int fops_vcodec_release(struct fi
 	v4l2_ctrl_handler_free(&ctx->ctrl_hdl);
 
 	mtk_vcodec_dbgfs_remove(dev, ctx->id);
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_del_init(&ctx->list);
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 	kfree(ctx);
 	mutex_unlock(&dev->dev_mutex);
 	return 0;

@@ -408,7 +410,7 @@ static int mtk_vcodec_probe(struct platf
 	for (i = 0; i < MTK_VDEC_HW_MAX; i++)
 		mutex_init(&dev->dec_mutex[i]);
 	mutex_init(&dev->dev_mutex);
-	mutex_init(&dev->dev_ctx_lock);
+	spin_lock_init(&dev->dev_ctx_lock);
 	spin_lock_init(&dev->irqlock);
 
 	snprintf(dev->v4l2_dev.name, sizeof(dev->v4l2_dev.name), "%s",

@@ -283,7 +283,7 @@ struct mtk_vcodec_dec_dev {
 	/* decoder hardware mutex lock */
 	struct mutex dec_mutex[MTK_VDEC_HW_MAX];
 	struct mutex dev_mutex;
-	struct mutex dev_ctx_lock;
+	spinlock_t dev_ctx_lock;
 	struct workqueue_struct *decode_workqueue;
 
 	spinlock_t irqlock;

@@ -75,16 +75,17 @@ static void handle_get_param_msg_ack(con
 static bool vpu_dec_check_ap_inst(struct mtk_vcodec_dec_dev *dec_dev, struct vdec_vpu_inst *vpu)
 {
 	struct mtk_vcodec_dec_ctx *ctx;
+	unsigned long flags;
 	int ret = false;
 
-	mutex_lock(&dec_dev->dev_ctx_lock);
+	spin_lock_irqsave(&dec_dev->dev_ctx_lock, flags);
 	list_for_each_entry(ctx, &dec_dev->ctx_list, list) {
 		if (!IS_ERR_OR_NULL(ctx) && ctx->vpu_inst == vpu) {
 			ret = true;
 			break;
 		}
 	}
-	mutex_unlock(&dec_dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dec_dev->dev_ctx_lock, flags);
 
 	return ret;
 }

@@ -117,6 +117,7 @@ static int fops_vcodec_open(struct file
 	struct mtk_vcodec_enc_ctx *ctx = NULL;
 	int ret = 0;
 	struct vb2_queue *src_vq;
+	unsigned long flags;
 
 	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)

@@ -177,9 +178,9 @@ static int fops_vcodec_open(struct file
 	mtk_v4l2_venc_dbg(2, ctx, "Create instance [%d]@%p m2m_ctx=%p ",
 			  ctx->id, ctx, ctx->m2m_ctx);
 
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_add(&ctx->list, &dev->ctx_list);
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 
 	mutex_unlock(&dev->dev_mutex);
 	mtk_v4l2_venc_dbg(0, ctx, "%s encoder [%d]", dev_name(&dev->plat_dev->dev),

@@ -204,6 +205,7 @@ static int fops_vcodec_release(struct fi
 {
 	struct mtk_vcodec_enc_dev *dev = video_drvdata(file);
 	struct mtk_vcodec_enc_ctx *ctx = fh_to_enc_ctx(file->private_data);
+	unsigned long flags;
 
 	mtk_v4l2_venc_dbg(1, ctx, "[%d] encoder", ctx->id);
 	mutex_lock(&dev->dev_mutex);

@@ -214,9 +216,9 @@ static int fops_vcodec_release(struct fi
 	v4l2_fh_exit(&ctx->fh);
 	v4l2_ctrl_handler_free(&ctx->ctrl_hdl);
 
-	mutex_lock(&dev->dev_ctx_lock);
+	spin_lock_irqsave(&dev->dev_ctx_lock, flags);
 	list_del_init(&ctx->list);
-	mutex_unlock(&dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&dev->dev_ctx_lock, flags);
 	kfree(ctx);
 	mutex_unlock(&dev->dev_mutex);
 	return 0;

@@ -298,7 +300,7 @@ static int mtk_vcodec_probe(struct platf
 
 	mutex_init(&dev->enc_mutex);
 	mutex_init(&dev->dev_mutex);
-	mutex_init(&dev->dev_ctx_lock);
+	spin_lock_init(&dev->dev_ctx_lock);
 	spin_lock_init(&dev->irqlock);
 
 	snprintf(dev->v4l2_dev.name, sizeof(dev->v4l2_dev.name), "%s",

@@ -206,7 +206,7 @@ struct mtk_vcodec_enc_dev {
 	/* encoder hardware mutex lock */
 	struct mutex enc_mutex;
 	struct mutex dev_mutex;
-	struct mutex dev_ctx_lock;
+	spinlock_t dev_ctx_lock;
 	struct workqueue_struct *encode_workqueue;
 
 	int enc_irq;

@@ -45,16 +45,17 @@ static void handle_enc_encode_msg(struct
 static bool vpu_enc_check_ap_inst(struct mtk_vcodec_enc_dev *enc_dev, struct venc_vpu_inst *vpu)
 {
 	struct mtk_vcodec_enc_ctx *ctx;
+	unsigned long flags;
 	int ret = false;
 
-	mutex_lock(&enc_dev->dev_ctx_lock);
+	spin_lock_irqsave(&enc_dev->dev_ctx_lock, flags);
 	list_for_each_entry(ctx, &enc_dev->ctx_list, list) {
 		if (!IS_ERR_OR_NULL(ctx) && ctx->vpu_inst == vpu) {
 			ret = true;
 			break;
 		}
 	}
-	mutex_unlock(&enc_dev->dev_ctx_lock);
+	spin_unlock_irqrestore(&enc_dev->dev_ctx_lock, flags);
 
 	return ret;
 }

@@ -25,6 +25,10 @@
 #include "vpu_imx8q.h"
 #include "vpu_malone.h"
 
+static bool low_latency;
+module_param(low_latency, bool, 0644);
+MODULE_PARM_DESC(low_latency, "Set low latency frame flush mode: 0 (disable) or 1 (enable)");
+
 #define CMD_SIZE			25600
 #define MSG_SIZE			25600
 #define CODEC_SIZE			0x1000

@@ -1562,7 +1566,15 @@ static int vpu_malone_input_frame_data(s
 
 	vpu_malone_update_wptr(str_buf, wptr);
 
-	if (disp_imm && !vpu_vb_is_codecconfig(vbuf)) {
+	/*
+	 * Enable the low latency flush mode if display delay is set to 0
+	 * or the low latency frame flush mode if it is set to 1.
+	 * The low latency flush mode requires some padding data to be appended to each frame,
+	 * but there must not be any padding data between the sequence header and the frame.
+	 * This module is currently only supported for the H264 and HEVC formats,
+	 * for other formats, vpu_malone_add_scode() will return 0.
+	 */
+	if ((disp_imm || low_latency) && !vpu_vb_is_codecconfig(vbuf)) {
 		ret = vpu_malone_add_scode(inst->core->iface,
 					   inst->id,
 					   &inst->stream_buffer,

@@ -24,6 +24,11 @@
 #include "vpu_msgs.h"
 #include "vpu_helpers.h"
 
+static char *vpu_type_name(u32 type)
+{
+	return V4L2_TYPE_IS_OUTPUT(type) ? "output" : "capture";
+}
+
 void vpu_inst_lock(struct vpu_inst *inst)
 {
 	mutex_lock(&inst->lock);

@@ -42,7 +47,7 @@ dma_addr_t vpu_get_vb_phy_addr(struct vb
 			vb->planes[plane_no].data_offset;
 }
 
-unsigned int vpu_get_vb_length(struct vb2_buffer *vb, u32 plane_no)
+static unsigned int vpu_get_vb_length(struct vb2_buffer *vb, u32 plane_no)
 {
 	if (plane_no >= vb->num_planes)
 		return 0;

@@ -81,7 +86,7 @@ void vpu_v4l2_set_error(struct vpu_inst
 	vpu_inst_unlock(inst);
 }
 
-int vpu_notify_eos(struct vpu_inst *inst)
+static int vpu_notify_eos(struct vpu_inst *inst)
 {
 	static const struct v4l2_event ev = {
 		.id = 0,

@@ -562,7 +567,8 @@ static void vpu_vb2_buf_finish(struct vb
 		call_void_vop(inst, on_queue_empty, q->type);
 }
 
-void vpu_vb2_buffers_return(struct vpu_inst *inst, unsigned int type, enum vb2_buffer_state state)
+static void vpu_vb2_buffers_return(struct vpu_inst *inst, unsigned int type,
+				   enum vb2_buffer_state state)
 {
 	struct vb2_v4l2_buffer *buf;
 

@@ -26,15 +26,12 @@ void vpu_skip_frame(struct vpu_inst *ins
 struct vb2_v4l2_buffer *vpu_find_buf_by_sequence(struct vpu_inst *inst, u32 type, u32 sequence);
 struct vb2_v4l2_buffer *vpu_find_buf_by_idx(struct vpu_inst *inst, u32 type, u32 idx);
 void vpu_v4l2_set_error(struct vpu_inst *inst);
-int vpu_notify_eos(struct vpu_inst *inst);
 int vpu_notify_source_change(struct vpu_inst *inst);
 int vpu_set_last_buffer_dequeued(struct vpu_inst *inst, bool eos);
-void vpu_vb2_buffers_return(struct vpu_inst *inst, unsigned int type, enum vb2_buffer_state state);
 int vpu_get_num_buffers(struct vpu_inst *inst, u32 type);
 bool vpu_is_source_empty(struct vpu_inst *inst);
 
 dma_addr_t vpu_get_vb_phy_addr(struct vb2_buffer *vb, u32 plane_no);
-unsigned int vpu_get_vb_length(struct vb2_buffer *vb, u32 plane_no);
 static inline struct vpu_format *vpu_get_format(struct vpu_inst *inst, u32 type)
 {
 	if (V4L2_TYPE_IS_OUTPUT(type))

@@ -43,11 +40,6 @@ static inline struct vpu_format *vpu_get
 		return &inst->cap_format;
 }
 
-static inline char *vpu_type_name(u32 type)
-{
-	return V4L2_TYPE_IS_OUTPUT(type) ? "output" : "capture";
-}
-
 static inline int vpu_vb_is_codecconfig(struct vb2_v4l2_buffer *vbuf)
 {
 #ifdef V4L2_BUF_FLAG_CODECCONFIG

@@ -1315,22 +1315,18 @@ static int vpu_malone_insert_scode_vc1_g
 {
 	if (!scode->inst->total_input_count)
 		return 0;
-	if (vpu_vb_is_codecconfig(to_vb2_v4l2_buffer(scode->vb)))
-		scode->need_data = 0;
 	return 0;
 }
 
 static int vpu_malone_insert_scode_vc1_g_pic(struct malone_scode_t *scode)
 {
-	struct vb2_v4l2_buffer *vbuf;
 	u8 nal_hdr[MALONE_VC1_NAL_HEADER_LEN];
 	u32 *data = NULL;
 	int ret;
 
-	vbuf = to_vb2_v4l2_buffer(scode->vb);
 	data = vb2_plane_vaddr(scode->vb, 0);
 
-	if (scode->inst->total_input_count == 0 || vpu_vb_is_codecconfig(vbuf))
+	if (scode->inst->total_input_count == 0)
 		return 0;
 	if (MALONE_VC1_CONTAIN_NAL(*data))
 		return 0;

@@ -1351,8 +1347,6 @@ static int vpu_malone_insert_scode_vc1_l
 	int size = 0;
 	u8 rcv_seqhdr[MALONE_VC1_RCV_SEQ_HEADER_LEN];
 
-	if (vpu_vb_is_codecconfig(to_vb2_v4l2_buffer(scode->vb)))
-		scode->need_data = 0;
 	if (scode->inst->total_input_count)
 		return 0;
 	scode->need_data = 0;

@@ -1538,7 +1532,7 @@ static int vpu_malone_input_frame_data(s
 	scode.vb = vb;
 	scode.wptr = wptr;
 	scode.need_data = 1;
-	if (vbuf->sequence == 0 || vpu_vb_is_codecconfig(vbuf))
+	if (vbuf->sequence == 0)
 		ret = vpu_malone_insert_scode(&scode, SCODE_SEQUENCE);
 
 	if (ret < 0)

@@ -1574,7 +1568,7 @@ static int vpu_malone_input_frame_data(s
 	 * This module is currently only supported for the H264 and HEVC formats,
 	 * for other formats, vpu_malone_add_scode() will return 0.
 	 */
-	if ((disp_imm || low_latency) && !vpu_vb_is_codecconfig(vbuf)) {
+	if (disp_imm || low_latency) {
 		ret = vpu_malone_add_scode(inst->core->iface,
 					   inst->id,
 					   &inst->stream_buffer,

@@ -1621,7 +1615,6 @@ int vpu_malone_input_frame(struct vpu_sh
 			   struct vpu_inst *inst, struct vb2_buffer *vb)
 {
 	struct vpu_dec_ctrl *hc = shared->priv;
-	struct vb2_v4l2_buffer *vbuf;
 	struct vpu_malone_str_buffer __iomem *str_buf = hc->str_buf[inst->id];
 	u32 disp_imm = hc->codec_param[inst->id].disp_imm;
 	u32 size;

@@ -1635,16 +1628,6 @@ int vpu_malone_input_frame(struct vpu_sh
 		return ret;
 	size = ret;
 
-	/*
-	 * if buffer only contain codec data, and the timestamp is invalid,
-	 * don't put the invalid timestamp to resync
-	 * merge the data to next frame
-	 */
-	vbuf = to_vb2_v4l2_buffer(vb);
-	if (vpu_vb_is_codecconfig(vbuf)) {
-		inst->extra_size += size;
-		return 0;
-	}
 	if (inst->extra_size) {
 		size += inst->extra_size;
 		inst->extra_size = 0;

@@ -349,16 +349,6 @@ struct vb2_v4l2_buffer *vpu_next_src_buf
 	if (!src_buf || vpu_get_buffer_state(src_buf) == VPU_BUF_STATE_IDLE)
 		return NULL;
 
-	while (vpu_vb_is_codecconfig(src_buf)) {
-		v4l2_m2m_src_buf_remove(inst->fh.m2m_ctx);
-		vpu_set_buffer_state(src_buf, VPU_BUF_STATE_IDLE);
-		v4l2_m2m_buf_done(src_buf, VB2_BUF_STATE_DONE);
-
-		src_buf = v4l2_m2m_next_src_buf(inst->fh.m2m_ctx);
-		if (!src_buf || vpu_get_buffer_state(src_buf) == VPU_BUF_STATE_IDLE)
-			return NULL;
-	}
-
 	return src_buf;
 }
 

@@ -39,14 +39,4 @@ static inline struct vpu_format *vpu_get
 	else
 		return &inst->cap_format;
 }
-
-static inline int vpu_vb_is_codecconfig(struct vb2_v4l2_buffer *vbuf)
-{
-#ifdef V4L2_BUF_FLAG_CODECCONFIG
-	return (vbuf->flags & V4L2_BUF_FLAG_CODECCONFIG) ? 1 : 0;
-#else
-	return 0;
-#endif
-}
-
 #endif

@@ -482,7 +482,7 @@ nvgrace_gpu_map_and_read(struct nvgrace_
 		ret = vfio_pci_core_do_io_rw(&nvdev->core_device, false,
 					     nvdev->resmem.ioaddr,
 					     buf, offset, mem_count,
-					     0, 0, false);
+					     0, 0, false, VFIO_PCI_IO_WIDTH_8);
 	}
 
 	return ret;

@@ -600,7 +600,7 @@ nvgrace_gpu_map_and_write(struct nvgrace
 		ret = vfio_pci_core_do_io_rw(&nvdev->core_device, false,
 					     nvdev->resmem.ioaddr,
 					     (char __user *)buf, pos, mem_count,
-					     0, 0, true);
+					     0, 0, true, VFIO_PCI_IO_WIDTH_8);
 	}
 
 	return ret;

@@ -141,7 +141,8 @@ VFIO_IORDWR(64)
 ssize_t vfio_pci_core_do_io_rw(struct vfio_pci_core_device *vdev, bool test_mem,
 			       void __iomem *io, char __user *buf,
 			       loff_t off, size_t count, size_t x_start,
-			       size_t x_end, bool iswrite)
+			       size_t x_end, bool iswrite,
+			       enum vfio_pci_io_width max_width)
 {
 	ssize_t done = 0;
 	int ret;

@@ -157,7 +158,7 @@ ssize_t vfio_pci_core_do_io_rw(struct vf
 			fillable = 0;
 
 #if defined(ioread64) && defined(iowrite64)
-		if (fillable >= 8 && !(off % 8)) {
+		if (fillable >= 8 && !(off % 8) && max_width >= 8) {
 			ret = vfio_pci_iordwr64(vdev, iswrite, test_mem,
 						io, buf, off, &filled);
 			if (ret)

@@ -165,13 +166,13 @@ ssize_t vfio_pci_core_do_io_rw(struct vf
 
 		} else
 #endif
-		if (fillable >= 4 && !(off % 4)) {
+		if (fillable >= 4 && !(off % 4) && max_width >= 4) {
 			ret = vfio_pci_iordwr32(vdev, iswrite, test_mem,
 						io, buf, off, &filled);
 			if (ret)
 				return ret;
 
-		} else if (fillable >= 2 && !(off % 2)) {
+		} else if (fillable >= 2 && !(off % 2) && max_width >= 2) {
 			ret = vfio_pci_iordwr16(vdev, iswrite, test_mem,
 						io, buf, off, &filled);
 			if (ret)

@@ -242,6 +243,7 @@ ssize_t vfio_pci_bar_rw(struct vfio_pci_
 	void __iomem *io;
 	struct resource *res = &vdev->pdev->resource[bar];
 	ssize_t done;
+	enum vfio_pci_io_width max_width = VFIO_PCI_IO_WIDTH_8;
 
 	if (pci_resource_start(pdev, bar))
 		end = pci_resource_len(pdev, bar);

@@ -268,6 +270,16 @@ ssize_t vfio_pci_bar_rw(struct vfio_pci_
 			goto out;
 		}
 		x_end = end;
+
+		/*
+		 * Certain devices (e.g. Intel X710) don't support qword
+		 * access to the ROM bar. Otherwise PCI AER errors might be
+		 * triggered.
+		 *
+		 * Disable qword access to the ROM bar universally, which
+		 * worked reliably for years before qword access is enabled.
+		 */
+		max_width = VFIO_PCI_IO_WIDTH_4;
 	} else {
 		int ret = vfio_pci_core_setup_barmap(vdev, bar);
 		if (ret) {

@@ -284,7 +296,7 @@ ssize_t vfio_pci_bar_rw(struct vfio_pci_
 	}
 
 	done = vfio_pci_core_do_io_rw(vdev, res->flags & IORESOURCE_MEM, io, buf, pos,
-				      count, x_start, x_end, iswrite);
+				      count, x_start, x_end, iswrite, max_width);
 
 	if (done >= 0)
 		*ppos += done;

@@ -353,7 +365,7 @@ ssize_t vfio_pci_vga_rw(struct vfio_pci_
 	 * to the memory enable bit in the command register.
 	 */
 	done = vfio_pci_core_do_io_rw(vdev, false, iomem, buf, off, count,
-				      0, 0, iswrite);
+				      0, 0, iswrite, VFIO_PCI_IO_WIDTH_8);
 
 	vga_put(vdev->pdev, rsrc);
 

@@ -102,6 +102,13 @@ struct vfio_pci_core_device {
 	struct rw_semaphore	memory_lock;
 };
 
+enum vfio_pci_io_width {
+	VFIO_PCI_IO_WIDTH_1 = 1,
+	VFIO_PCI_IO_WIDTH_2 = 2,
+	VFIO_PCI_IO_WIDTH_4 = 4,
+	VFIO_PCI_IO_WIDTH_8 = 8,
+};
+
 /* Will be exported for vfio pci drivers usage */
 int vfio_pci_core_register_dev_region(struct vfio_pci_core_device *vdev,
 				      unsigned int type, unsigned int subtype,

@@ -137,7 +144,8 @@ pci_ers_result_t vfio_pci_core_aer_err_d
 ssize_t vfio_pci_core_do_io_rw(struct vfio_pci_core_device *vdev, bool test_mem,
 			       void __iomem *io, char __user *buf,
 			       loff_t off, size_t count, size_t x_start,
-			       size_t x_end, bool iswrite);
+			       size_t x_end, bool iswrite,
+			       enum vfio_pci_io_width max_width);
 bool vfio_pci_core_range_intersect_range(loff_t buf_start, size_t buf_cnt,
 					 loff_t reg_start, size_t reg_cnt,
 					 loff_t *buf_offset,

@@ -278,15 +278,35 @@ static void damon_test_split_regions_of(
 	struct damon_target *t;
 	struct damon_region *r;
 
+	if (!c)
+		kunit_skip(test, "ctx alloc fail");
 	t = damon_new_target();
+	if (!t) {
+		damon_destroy_ctx(c);
+		kunit_skip(test, "target alloc fail");
+	}
 	r = damon_new_region(0, 22);
+	if (!r) {
+		damon_destroy_ctx(c);
+		damon_free_target(t);
+		kunit_skip(test, "region alloc fail");
+	}
 	damon_add_region(r, t);
 	damon_split_regions_of(t, 2);
 	KUNIT_EXPECT_LE(test, damon_nr_regions(t), 2u);
 	damon_free_target(t);
 
 	t = damon_new_target();
+	if (!t) {
+		damon_destroy_ctx(c);
+		kunit_skip(test, "second target alloc fail");
+	}
 	r = damon_new_region(0, 220);
+	if (!r) {
+		damon_destroy_ctx(c);
+		damon_free_target(t);
+		kunit_skip(test, "second region alloc fail");
+	}
 	damon_add_region(r, t);
 	damon_split_regions_of(t, 4);
 	KUNIT_EXPECT_LE(test, damon_nr_regions(t), 4u);

@@ -505,6 +505,8 @@ static void damos_test_new_filter(struct
 	struct damos_filter *filter;
 
 	filter = damos_new_filter(DAMOS_FILTER_TYPE_ANON, true);
+	if (!filter)
+		kunit_skip(test, "filter alloc fail");
 	KUNIT_EXPECT_EQ(test, filter->type, DAMOS_FILTER_TYPE_ANON);
 	KUNIT_EXPECT_EQ(test, filter->matching, true);
 	KUNIT_EXPECT_PTR_EQ(test, filter->list.prev, &filter->list);

@@ -136,8 +136,14 @@ static void damon_do_test_apply_three_re
 	int i;
 
 	t = damon_new_target();
+	if (!t)
+		kunit_skip(test, "target alloc fail");
 	for (i = 0; i < nr_regions / 2; i++) {
 		r = damon_new_region(regions[i * 2], regions[i * 2 + 1]);
+		if (!r) {
+			damon_destroy_target(t);
+			kunit_skip(test, "region alloc fail");
+		}
 		damon_add_region(r, t);
 	}
 

@@ -674,12 +674,12 @@ __iomap_dio_rw(struct kiocb *iocb, struc
 			}
 			goto out_free_dio;
 		}
+	}
 
-		if (!wait_for_completion && !inode->i_sb->s_dio_done_wq) {
-			ret = sb_init_dio_done_wq(inode->i_sb);
-			if (ret < 0)
-				goto out_free_dio;
-		}
+	if (!wait_for_completion && !inode->i_sb->s_dio_done_wq) {
+		ret = sb_init_dio_done_wq(inode->i_sb);
+		if (ret < 0)
+			goto out_free_dio;
 	}
 
 	inode_dio_begin(inode);

@@ -746,17 +746,20 @@ static void blk_zone_reset_all_bio_endio
 	unsigned long flags;
 	unsigned int i;
 
-	/* Update the condition of all zone write plugs. */
-	rcu_read_lock();
-	for (i = 0; i < disk_zone_wplugs_hash_size(disk); i++) {
-		hlist_for_each_entry_rcu(zwplug, &disk->zone_wplugs_hash[i],
-					 node) {
-			spin_lock_irqsave(&zwplug->lock, flags);
-			disk_zone_wplug_set_wp_offset(disk, zwplug, 0);
-			spin_unlock_irqrestore(&zwplug->lock, flags);
+	if (atomic_read(&disk->nr_zone_wplugs)) {
+		/* Update the condition of all zone write plugs. */
+		rcu_read_lock();
+		for (i = 0; i < disk_zone_wplugs_hash_size(disk); i++) {
+			hlist_for_each_entry_rcu(zwplug,
+						 &disk->zone_wplugs_hash[i],
+						 node) {
+				spin_lock_irqsave(&zwplug->lock, flags);
+				disk_zone_wplug_set_wp_offset(disk, zwplug, 0);
+				spin_unlock_irqrestore(&zwplug->lock, flags);
+			}
 		}
+		rcu_read_unlock();
 	}
-	rcu_read_unlock();
 }
 
 static void blk_zone_finish_bio_endio(struct bio *bio)