dmaengine fixes-2 for v5.11

Some late fixes for dmaengine:
  - Core: fix channel device_node deletion
  - Driver fixes for:
    - dw: revert of runtime pm enabling
    - idxd: device state fix, interrupt completion and list corruption
    - ti: resource leak
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCAAdFiEE+vs47OPLdNbVcHzyfBQHDyUjg0cFAmAjrkQACgkQfBQHDyUj
 g0foZA/+Iqpi6fU0Dth4bdoJa5HO63a62G5nrhofF/vH681GMaazNj46byol3vuA
 Gc0/EZ2UtIkEY29ix0XaHkksQrsqn/Q5E4QK+5u9x32DHf3jvtbOblOSIBCdr//3
 i+uc/K90ot4ERtNvwiPxQGWjS7rF+6BvHItRDxOaiele0Uvf18/VGn2x7fH5vNeK
 GqtZK47E11y5UhqpJAiwcgNAhKXC6I6s/tP0pidyWuXWeqVm+usr6Pun9YExMJQm
 N+kiR8eJoh5F0N9KAg3rOppxf4iEblvgh2vfMgcNC63GdeWB2x1OMgizAXjE136K
 HAvcp/3rQf76tUhjZkr/YZaNB7wCqzCRRcgQ/xyhSJt24yswfv9NFGVHd2ltkfx9
 Yp+rl8ZC0dSvdGR3ECF9z98MzRbBPgu+TCW/50/Hh42Va0FJZbXyY45hpfR9qPe2
 hiXwQkJ8IKH7C8BpDKA8vMlJc4xhbNsYW0GaSyoAUzhaStwTHcKNB4+5Xeia55e3
 RR2OPJXl+y3jywcO15fmFdNIRsSvRVGYioFH0NzneaVVIlbQk5hRqADNMWelnwiA
 DJc21v7yurHeCh3lefn5Aml10n986S1b7XNPA7Ls+2FMmJeIt2vrKqvmKLhHVANY
 bvSKEXda2pAvb3zw2fCcCuPq6KUdJAfDrB0oorIlRBM3IkY+B5Y=
 =j/zV
 -----END PGP SIGNATURE-----

Merge tag 'dmaengine-fix2-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine

Pull dmaengine fixes from Vinod Koul:
 "Some late fixes for dmaengine:

  Core:
   - fix channel device_node deletion

  Driver fixes:
   - dw: revert of runtime pm enabling
   - idxd: device state fix, interrupt completion and list corruption
   - ti: resource leak

* tag 'dmaengine-fix2-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/dmaengine:
  dmaengine dw: Revert "dmaengine: dw: Enable runtime PM"
  dmaengine: idxd: check device state before issue command
  dmaengine: ti: k3-udma: Fix a resource leak in an error handling path
  dmaengine: move channel device_node deletion to driver
  dmaengine: idxd: fix misc interrupt completion
  dmaengine: idxd: Fix list corruption in description completion

drivers/dma/dmaengine.c

@@ -1110,7 +1110,6 @@ static void __dma_async_device_channel_unregister(struct dma_device *device,
 		  "%s called while %d clients hold a reference\n",
 		  __func__, chan->client_count);
 	mutex_lock(&dma_list_mutex);
-	list_del(&chan->device_node);
 	device->chancnt--;
 	chan->dev->chan = NULL;
 	mutex_unlock(&dma_list_mutex);

drivers/dma/dw/core.c

@@ -982,11 +982,8 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 
 	dev_vdbg(chan2dev(chan), "%s\n", __func__);
 
-	pm_runtime_get_sync(dw->dma.dev);
-
 	/* ASSERT:  channel is idle */
 	if (dma_readl(dw, CH_EN) & dwc->mask) {
-		pm_runtime_put_sync_suspend(dw->dma.dev);
 		dev_dbg(chan2dev(chan), "DMA channel not idle?\n");
 		return -EIO;
 	}
@@ -1003,7 +1000,6 @@ static int dwc_alloc_chan_resources(struct dma_chan *chan)
 	 * We need controller-specific data to set up slave transfers.
 	 */
 	if (chan->private && !dw_dma_filter(chan, chan->private)) {
-		pm_runtime_put_sync_suspend(dw->dma.dev);
 		dev_warn(chan2dev(chan), "Wrong controller-specific data\n");
 		return -EINVAL;
 	}
@@ -1047,8 +1043,6 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 	if (!dw->in_use)
 		do_dw_dma_off(dw);
 
-	pm_runtime_put_sync_suspend(dw->dma.dev);
-
 	dev_vdbg(chan2dev(chan), "%s: done\n", __func__);
 }
 

drivers/dma/idxd/device.c

@@ -398,17 +398,31 @@ static inline bool idxd_is_enabled(struct idxd_device *idxd)
 	return false;
 }
 
+static inline bool idxd_device_is_halted(struct idxd_device *idxd)
+{
+	union gensts_reg gensts;
+
+	gensts.bits = ioread32(idxd->reg_base + IDXD_GENSTATS_OFFSET);
+
+	return (gensts.state == IDXD_DEVICE_STATE_HALT);
+}
+
 /*
  * This is function is only used for reset during probe and will
  * poll for completion. Once the device is setup with interrupts,
  * all commands will be done via interrupt completion.
  */
-void idxd_device_init_reset(struct idxd_device *idxd)
+int idxd_device_init_reset(struct idxd_device *idxd)
 {
 	struct device *dev = &idxd->pdev->dev;
 	union idxd_command_reg cmd;
 	unsigned long flags;
 
+	if (idxd_device_is_halted(idxd)) {
+		dev_warn(&idxd->pdev->dev, "Device is HALTED!\n");
+		return -ENXIO;
+	}
+
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.cmd = IDXD_CMD_RESET_DEVICE;
 	dev_dbg(dev, "%s: sending reset for init.\n", __func__);
@@ -419,6 +433,7 @@ void idxd_device_init_reset(struct idxd_device *idxd)
 	       IDXD_CMDSTS_ACTIVE)
 		cpu_relax();
 	spin_unlock_irqrestore(&idxd->dev_lock, flags);
+	return 0;
 }
 
 static void idxd_cmd_exec(struct idxd_device *idxd, int cmd_code, u32 operand,
@@ -428,6 +443,12 @@ static void idxd_cmd_exec(struct idxd_device *idxd, int cmd_code, u32 operand,
 	DECLARE_COMPLETION_ONSTACK(done);
 	unsigned long flags;
 
+	if (idxd_device_is_halted(idxd)) {
+		dev_warn(&idxd->pdev->dev, "Device is HALTED!\n");
+		*status = IDXD_CMDSTS_HW_ERR;
+		return;
+	}
+
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.cmd = cmd_code;
 	cmd.operand = operand;

drivers/dma/idxd/dma.c

@@ -205,5 +205,8 @@ int idxd_register_dma_channel(struct idxd_wq *wq)
 
 void idxd_unregister_dma_channel(struct idxd_wq *wq)
 {
-	dma_async_device_channel_unregister(&wq->idxd->dma_dev, &wq->dma_chan);
+	struct dma_chan *chan = &wq->dma_chan;
+
+	dma_async_device_channel_unregister(&wq->idxd->dma_dev, chan);
+	list_del(&chan->device_node);
 }

drivers/dma/idxd/idxd.h

@@ -326,7 +326,7 @@ void idxd_mask_msix_vector(struct idxd_device *idxd, int vec_id);
 void idxd_unmask_msix_vector(struct idxd_device *idxd, int vec_id);
 
 /* device control */
-void idxd_device_init_reset(struct idxd_device *idxd);
+int idxd_device_init_reset(struct idxd_device *idxd);
 int idxd_device_enable(struct idxd_device *idxd);
 int idxd_device_disable(struct idxd_device *idxd);
 void idxd_device_reset(struct idxd_device *idxd);

drivers/dma/idxd/init.c

@@ -335,7 +335,10 @@ static int idxd_probe(struct idxd_device *idxd)
 	int rc;
 
 	dev_dbg(dev, "%s entered and resetting device\n", __func__);
-	idxd_device_init_reset(idxd);
+	rc = idxd_device_init_reset(idxd);
+	if (rc < 0)
+		return rc;
+
 	dev_dbg(dev, "IDXD reset complete\n");
 
 	if (IS_ENABLED(CONFIG_INTEL_IDXD_SVM)) {

drivers/dma/idxd/irq.c

@@ -111,19 +111,14 @@ irqreturn_t idxd_irq_handler(int vec, void *data)
 	return IRQ_WAKE_THREAD;
 }
 
-irqreturn_t idxd_misc_thread(int vec, void *data)
+static int process_misc_interrupts(struct idxd_device *idxd, u32 cause)
 {
-	struct idxd_irq_entry *irq_entry = data;
-	struct idxd_device *idxd = irq_entry->idxd;
 	struct device *dev = &idxd->pdev->dev;
 	union gensts_reg gensts;
-	u32 cause, val = 0;
+	u32 val = 0;
 	int i;
 	bool err = false;
 
-	cause = ioread32(idxd->reg_base + IDXD_INTCAUSE_OFFSET);
-	iowrite32(cause, idxd->reg_base + IDXD_INTCAUSE_OFFSET);
-
 	if (cause & IDXD_INTC_ERR) {
 		spin_lock_bh(&idxd->dev_lock);
 		for (i = 0; i < 4; i++)
@@ -181,7 +176,7 @@ irqreturn_t idxd_misc_thread(int vec, void *data)
 			      val);
 
 	if (!err)
-		goto out;
+		return 0;
 
 	/*
 	 * This case should rarely happen and typically is due to software
@@ -211,37 +206,58 @@ irqreturn_t idxd_misc_thread(int vec, void *data)
 				gensts.reset_type == IDXD_DEVICE_RESET_FLR ?
 				"FLR" : "system reset");
 			spin_unlock_bh(&idxd->dev_lock);
+			return -ENXIO;
 		}
 	}
 
- out:
+	return 0;
+}
+
+irqreturn_t idxd_misc_thread(int vec, void *data)
+{
+	struct idxd_irq_entry *irq_entry = data;
+	struct idxd_device *idxd = irq_entry->idxd;
+	int rc;
+	u32 cause;
+
+	cause = ioread32(idxd->reg_base + IDXD_INTCAUSE_OFFSET);
+	if (cause)
+		iowrite32(cause, idxd->reg_base + IDXD_INTCAUSE_OFFSET);
+
+	while (cause) {
+		rc = process_misc_interrupts(idxd, cause);
+		if (rc < 0)
+			break;
+		cause = ioread32(idxd->reg_base + IDXD_INTCAUSE_OFFSET);
+		if (cause)
+			iowrite32(cause, idxd->reg_base + IDXD_INTCAUSE_OFFSET);
+	}
+
 	idxd_unmask_msix_vector(idxd, irq_entry->id);
 	return IRQ_HANDLED;
 }
 
-static bool process_fault(struct idxd_desc *desc, u64 fault_addr)
+static inline bool match_fault(struct idxd_desc *desc, u64 fault_addr)
 {
 	/*
 	 * Completion address can be bad as well. Check fault address match for descriptor
 	 * and completion address.
 	 */
-	if ((u64)desc->hw == fault_addr ||
+	if ((u64)desc->hw == fault_addr || (u64)desc->completion == fault_addr) {
-	    (u64)desc->completion == fault_addr) {
+		struct idxd_device *idxd = desc->wq->idxd;
-		idxd_dma_complete_txd(desc, IDXD_COMPLETE_DEV_FAIL);
+		struct device *dev = &idxd->pdev->dev;
+
+		dev_warn(dev, "desc with fault address: %#llx\n", fault_addr);
 		return true;
 	}
 
 	return false;
 }
 
-static bool complete_desc(struct idxd_desc *desc)
+static inline void complete_desc(struct idxd_desc *desc, enum idxd_complete_type reason)
 {
-	if (desc->completion->status) {
+	idxd_dma_complete_txd(desc, reason);
-		idxd_dma_complete_txd(desc, IDXD_COMPLETE_NORMAL);
+	idxd_free_desc(desc->wq, desc);
-		return true;
-	}
-
-	return false;
 }
 
 static int irq_process_pending_llist(struct idxd_irq_entry *irq_entry,
@@ -251,25 +267,25 @@ static int irq_process_pending_llist(struct idxd_irq_entry *irq_entry,
 	struct idxd_desc *desc, *t;
 	struct llist_node *head;
 	int queued = 0;
-	bool completed = false;
 	unsigned long flags;
+	enum idxd_complete_type reason;
 
 	*processed = 0;
 	head = llist_del_all(&irq_entry->pending_llist);
 	if (!head)
 		goto out;
 
-	llist_for_each_entry_safe(desc, t, head, llnode) {
+	if (wtype == IRQ_WORK_NORMAL)
-		if (wtype == IRQ_WORK_NORMAL)
+		reason = IDXD_COMPLETE_NORMAL;
-			completed = complete_desc(desc);
+	else
-		else if (wtype == IRQ_WORK_PROCESS_FAULT)
+		reason = IDXD_COMPLETE_DEV_FAIL;
-			completed = process_fault(desc, data);
 
-		if (completed) {
+	llist_for_each_entry_safe(desc, t, head, llnode) {
-			idxd_free_desc(desc->wq, desc);
+		if (desc->completion->status) {
+			if ((desc->completion->status & DSA_COMP_STATUS_MASK) != DSA_COMP_SUCCESS)
+				match_fault(desc, data);
+			complete_desc(desc, reason);
 			(*processed)++;
-			if (wtype == IRQ_WORK_PROCESS_FAULT)
-				break;
 		} else {
 			spin_lock_irqsave(&irq_entry->list_lock, flags);
 			list_add_tail(&desc->list,
@@ -287,42 +303,46 @@ static int irq_process_work_list(struct idxd_irq_entry *irq_entry,
 				 enum irq_work_type wtype,
 				 int *processed, u64 data)
 {
-	struct list_head *node, *next;
 	int queued = 0;
-	bool completed = false;
 	unsigned long flags;
+	LIST_HEAD(flist);
+	struct idxd_desc *desc, *n;
+	enum idxd_complete_type reason;
 
 	*processed = 0;
+	if (wtype == IRQ_WORK_NORMAL)
+		reason = IDXD_COMPLETE_NORMAL;
+	else
+		reason = IDXD_COMPLETE_DEV_FAIL;
+
+	/*
+	 * This lock protects list corruption from access of list outside of the irq handler
+	 * thread.
+	 */
 	spin_lock_irqsave(&irq_entry->list_lock, flags);
-	if (list_empty(&irq_entry->work_list))
+	if (list_empty(&irq_entry->work_list)) {
-		goto out;
-
-	list_for_each_safe(node, next, &irq_entry->work_list) {
-		struct idxd_desc *desc =
-			container_of(node, struct idxd_desc, list);
-
 		spin_unlock_irqrestore(&irq_entry->list_lock, flags);
-		if (wtype == IRQ_WORK_NORMAL)
+		return 0;
-			completed = complete_desc(desc);
+	}
-		else if (wtype == IRQ_WORK_PROCESS_FAULT)
-			completed = process_fault(desc, data);
 
-		if (completed) {
+	list_for_each_entry_safe(desc, n, &irq_entry->work_list, list) {
-			spin_lock_irqsave(&irq_entry->list_lock, flags);
+		if (desc->completion->status) {
 			list_del(&desc->list);
-			spin_unlock_irqrestore(&irq_entry->list_lock, flags);
-			idxd_free_desc(desc->wq, desc);
 			(*processed)++;
-			if (wtype == IRQ_WORK_PROCESS_FAULT)
+			list_add_tail(&desc->list, &flist);
-				return queued;
 		} else {
 			queued++;
 		}
-		spin_lock_irqsave(&irq_entry->list_lock, flags);
 	}
 
- out:
 	spin_unlock_irqrestore(&irq_entry->list_lock, flags);
+
+	list_for_each_entry(desc, &flist, list) {
+		if ((desc->completion->status & DSA_COMP_STATUS_MASK) != DSA_COMP_SUCCESS)
+			match_fault(desc, data);
+		complete_desc(desc, reason);
+	}
+
 	return queued;
 }
 

drivers/dma/ti/k3-udma.c

@@ -2401,7 +2401,8 @@ static int bcdma_alloc_chan_resources(struct dma_chan *chan)
 			dev_err(ud->ddev.dev,
 				"Descriptor pool allocation failed\n");
 			uc->use_dma_pool = false;
-			return -ENOMEM;
+			ret = -ENOMEM;
+			goto err_res_free;
 		}
 
 		uc->use_dma_pool = true;