6 files changed, 106 insertions, 43 deletions

diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
index cca472109135..d1b55fe62817 100644
--- a/drivers/misc/cxl/context.c
+++ b/drivers/misc/cxl/context.c
@@ -34,7 +34,8 @@ struct cxl_context *cxl_context_alloc(void)
 /*
  * Initialises a CXL context.
  */
-int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
+int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master,
+		     struct address_space *mapping)
 {
 	int i;
 
@@ -42,6 +43,8 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
 	ctx->afu = afu;
 	ctx->master = master;
 	ctx->pid = NULL; /* Set in start work ioctl */
+	mutex_init(&ctx->mapping_lock);
+	ctx->mapping = mapping;
 
 	/*
 	 * Allocate the segment table before we put it in the IDR so that we
@@ -82,12 +85,12 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
 	 * Allocating IDR! We better make sure everything's setup that
 	 * dereferences from it.
 	 */
+	mutex_lock(&afu->contexts_lock);
 	idr_preload(GFP_KERNEL);
-	spin_lock(&afu->contexts_lock);
 	i = idr_alloc(&ctx->afu->contexts_idr, ctx, 0,
 		      ctx->afu->num_procs, GFP_NOWAIT);
-	spin_unlock(&afu->contexts_lock);
 	idr_preload_end();
+	mutex_unlock(&afu->contexts_lock);
 	if (i < 0)
 		return i;
 
@@ -97,6 +100,46 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master)
 	return 0;
 }
 
+static int cxl_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	struct cxl_context *ctx = vma->vm_file->private_data;
+	unsigned long address = (unsigned long)vmf->virtual_address;
+	u64 area, offset;
+
+	offset = vmf->pgoff << PAGE_SHIFT;
+
+	pr_devel("%s: pe: %i address: 0x%lx offset: 0x%llx\n",
+			__func__, ctx->pe, address, offset);
+
+	if (ctx->afu->current_mode == CXL_MODE_DEDICATED) {
+		area = ctx->afu->psn_phys;
+		if (offset > ctx->afu->adapter->ps_size)
+			return VM_FAULT_SIGBUS;
+	} else {
+		area = ctx->psn_phys;
+		if (offset > ctx->psn_size)
+			return VM_FAULT_SIGBUS;
+	}
+
+	mutex_lock(&ctx->status_mutex);
+
+	if (ctx->status != STARTED) {
+		mutex_unlock(&ctx->status_mutex);
+		pr_devel("%s: Context not started, failing problem state access\n", __func__);
+		return VM_FAULT_SIGBUS;
+	}
+
+	vm_insert_pfn(vma, address, (area + offset) >> PAGE_SHIFT);
+
+	mutex_unlock(&ctx->status_mutex);
+
+	return VM_FAULT_NOPAGE;
+}
+
+static const struct vm_operations_struct cxl_mmap_vmops = {
+	.fault = cxl_mmap_fault,
+};
+
 /*
  * Map a per-context mmio space into the given vma.
  */
@@ -105,26 +148,25 @@ int cxl_context_iomap(struct cxl_context *ctx, struct vm_area_struct *vma)
 	u64 len = vma->vm_end - vma->vm_start;
 	len = min(len, ctx->psn_size);
 
-	if (ctx->afu->current_mode == CXL_MODE_DEDICATED) {
-		vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
-		return vm_iomap_memory(vma, ctx->afu->psn_phys, ctx->afu->adapter->ps_size);
-	}
+	if (ctx->afu->current_mode != CXL_MODE_DEDICATED) {
+		/* make sure there is a valid per process space for this AFU */
+		if ((ctx->master && !ctx->afu->psa) || (!ctx->afu->pp_psa)) {
+			pr_devel("AFU doesn't support mmio space\n");
+			return -EINVAL;
+		}
 
-	/* make sure there is a valid per process space for this AFU */
-	if ((ctx->master && !ctx->afu->psa) || (!ctx->afu->pp_psa)) {
-		pr_devel("AFU doesn't support mmio space\n");
-		return -EINVAL;
+		/* Can't mmap until the AFU is enabled */
+		if (!ctx->afu->enabled)
+			return -EBUSY;
 	}
 
-	/* Can't mmap until the AFU is enabled */
-	if (!ctx->afu->enabled)
-		return -EBUSY;
-
 	pr_devel("%s: mmio physical: %llx pe: %i master:%i\n", __func__,
 		 ctx->psn_phys, ctx->pe , ctx->master);
 
+	vma->vm_flags |= VM_IO | VM_PFNMAP;
 	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
-	return vm_iomap_memory(vma, ctx->psn_phys, len);
+	vma->vm_ops = &cxl_mmap_vmops;
+	return 0;
 }
 
 /*
@@ -168,21 +210,33 @@ void cxl_context_detach_all(struct cxl_afu *afu)
 	struct cxl_context *ctx;
 	int tmp;
 
-	rcu_read_lock();
-	idr_for_each_entry(&afu->contexts_idr, ctx, tmp)
+	mutex_lock(&afu->contexts_lock);
+	idr_for_each_entry(&afu->contexts_idr, ctx, tmp) {
 		/*
 		 * Anything done in here needs to be setup before the IDR is
 		 * created and torn down after the IDR removed
 		 */
 		__detach_context(ctx);
-	rcu_read_unlock();
+
+		/*
+		 * We are force detaching - remove any active PSA mappings so
+		 * userspace cannot interfere with the card if it comes back.
+		 * Easiest way to exercise this is to unbind and rebind the
+		 * driver via sysfs while it is in use.
+		 */
+		mutex_lock(&ctx->mapping_lock);
+		if (ctx->mapping)
+			unmap_mapping_range(ctx->mapping, 0, 0, 1);
+		mutex_unlock(&ctx->mapping_lock);
+	}
+	mutex_unlock(&afu->contexts_lock);
 }
 
 void cxl_context_free(struct cxl_context *ctx)
 {
-	spin_lock(&ctx->afu->contexts_lock);
+	mutex_lock(&ctx->afu->contexts_lock);
 	idr_remove(&ctx->afu->contexts_idr, ctx->pe);
-	spin_unlock(&ctx->afu->contexts_lock);
+	mutex_unlock(&ctx->afu->contexts_lock);
 	synchronize_rcu();
 
 	free_page((u64)ctx->sstp);
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index b5b6bda44a00..28078f8894a5 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -351,7 +351,7 @@ struct cxl_afu {
 	struct device *chardev_s, *chardev_m, *chardev_d;
 	struct idr contexts_idr;
 	struct dentry *debugfs;
-	spinlock_t contexts_lock;
+	struct mutex contexts_lock;
 	struct mutex spa_mutex;
 	spinlock_t afu_cntl_lock;
 
@@ -398,6 +398,10 @@ struct cxl_context {
 	phys_addr_t psn_phys;
 	u64 psn_size;
 
+	/* Used to unmap any mmaps when force detaching */
+	struct address_space *mapping;
+	struct mutex mapping_lock;
+
 	spinlock_t sste_lock; /* Protects segment table entries */
 	struct cxl_sste *sstp;
 	u64 sstp0, sstp1;
@@ -599,7 +603,8 @@ int cxl_alloc_sst(struct cxl_context *ctx);
 void init_cxl_native(void);
 
 struct cxl_context *cxl_context_alloc(void);
-int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master);
+int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master,
+		     struct address_space *mapping);
 void cxl_context_free(struct cxl_context *ctx);
 int cxl_context_iomap(struct cxl_context *ctx, struct vm_area_struct *vma);
 
diff --git a/drivers/misc/cxl/file.c b/drivers/misc/cxl/file.c
index 378b099e7c0b..b15d8113877c 100644
--- a/drivers/misc/cxl/file.c
+++ b/drivers/misc/cxl/file.c
@@ -77,7 +77,7 @@ static int __afu_open(struct inode *inode, struct file *file, bool master)
 		goto err_put_afu;
 	}
 
-	if ((rc = cxl_context_init(ctx, afu, master)))
+	if ((rc = cxl_context_init(ctx, afu, master, inode->i_mapping)))
 		goto err_put_afu;
 
 	pr_devel("afu_open pe: %i\n", ctx->pe);
@@ -113,6 +113,10 @@ static int afu_release(struct inode *inode, struct file *file)
 		 __func__, ctx->pe);
 	cxl_context_detach(ctx);
 
+	mutex_lock(&ctx->mapping_lock);
+	ctx->mapping = NULL;
+	mutex_unlock(&ctx->mapping_lock);
+
 	put_device(&ctx->afu->dev);
 
 	/*
@@ -136,18 +140,20 @@ static long afu_ioctl_start_work(struct cxl_context *ctx,
 
 	pr_devel("%s: pe: %i\n", __func__, ctx->pe);
 
-	mutex_lock(&ctx->status_mutex);
-	if (ctx->status != OPENED) {
-		rc = -EIO;
-		goto out;
-	}
-
+	/* Do this outside the status_mutex to avoid a circular dependency with
+	 * the locking in cxl_mmap_fault() */
 	if (copy_from_user(&work, uwork,
 			   sizeof(struct cxl_ioctl_start_work))) {
 		rc = -EFAULT;
 		goto out;
 	}
 
+	mutex_lock(&ctx->status_mutex);
+	if (ctx->status != OPENED) {
+		rc = -EIO;
+		goto out;
+	}
+
 	/*
 	 * if any of the reserved fields are set or any of the unused
 	 * flags are set it's invalid
diff --git a/drivers/misc/cxl/native.c b/drivers/misc/cxl/native.c
index 9a5a442269a8..f2b37b41a0da 100644
--- a/drivers/misc/cxl/native.c
+++ b/drivers/misc/cxl/native.c
@@ -277,6 +277,7 @@ static int do_process_element_cmd(struct cxl_context *ctx,
 				  u64 cmd, u64 pe_state)
 {
 	u64 state;
+	unsigned long timeout = jiffies + (HZ * CXL_TIMEOUT);
 
 	WARN_ON(!ctx->afu->enabled);
 
@@ -286,6 +287,10 @@ static int do_process_element_cmd(struct cxl_context *ctx,
 	smp_mb();
 	cxl_p1n_write(ctx->afu, CXL_PSL_LLCMD_An, cmd | ctx->pe);
 	while (1) {
+		if (time_after_eq(jiffies, timeout)) {
+			dev_warn(&ctx->afu->dev, "WARNING: Process Element Command timed out!\n");
+			return -EBUSY;
+		}
 		state = be64_to_cpup(ctx->afu->sw_command_status);
 		if (state == ~0ULL) {
 			pr_err("cxl: Error adding process element to AFU\n");
@@ -610,13 +615,6 @@ static inline int detach_process_native_dedicated(struct cxl_context *ctx)
 	return 0;
 }
 
-/*
- * TODO: handle case when this is called inside a rcu_read_lock() which may
- * happen when we unbind the driver (ie. cxl_context_detach_all()) .  Terminate
- * & remove use a mutex lock and schedule which will not good with lock held.
- * May need to write do_process_element_cmd() that handles outstanding page
- * faults synchronously.
- */
 static inline int detach_process_native_afu_directed(struct cxl_context *ctx)
 {
 	if (!ctx->pe_inserted)
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 10c98ab7f46e..0f2cc9f8b4db 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -502,7 +502,7 @@ static struct cxl_afu *cxl_alloc_afu(struct cxl *adapter, int slice)
 	afu->dev.release = cxl_release_afu;
 	afu->slice = slice;
 	idr_init(&afu->contexts_idr);
-	spin_lock_init(&afu->contexts_lock);
+	mutex_init(&afu->contexts_lock);
 	spin_lock_init(&afu->afu_cntl_lock);
 	mutex_init(&afu->spa_mutex);
 
diff --git a/drivers/misc/cxl/sysfs.c b/drivers/misc/cxl/sysfs.c
index ce7ec06d87d1..461bdbd5d483 100644
--- a/drivers/misc/cxl/sysfs.c
+++ b/drivers/misc/cxl/sysfs.c
@@ -121,7 +121,7 @@ static ssize_t reset_store_afu(struct device *device,
 	int rc;
 
 	/* Not safe to reset if it is currently in use */
-	spin_lock(&afu->contexts_lock);
+	mutex_lock(&afu->contexts_lock);
 	if (!idr_is_empty(&afu->contexts_idr)) {
 		rc = -EBUSY;
 		goto err;
@@ -132,7 +132,7 @@ static ssize_t reset_store_afu(struct device *device,
 
 	rc = count;
 err:
-	spin_unlock(&afu->contexts_lock);
+	mutex_unlock(&afu->contexts_lock);
 	return rc;
 }
 
@@ -247,7 +247,7 @@ static ssize_t mode_store(struct device *device, struct device_attribute *attr,
 	int rc = -EBUSY;
 
 	/* can't change this if we have a user */
-	spin_lock(&afu->contexts_lock);
+	mutex_lock(&afu->contexts_lock);
 	if (!idr_is_empty(&afu->contexts_idr))
 		goto err;
 
@@ -271,7 +271,7 @@ static ssize_t mode_store(struct device *device, struct device_attribute *attr,
 	afu->current_mode = 0;
 	afu->num_procs = 0;
 
-	spin_unlock(&afu->contexts_lock);
+	mutex_unlock(&afu->contexts_lock);
 
 	if ((rc = _cxl_afu_deactivate_mode(afu, old_mode)))
 		return rc;
@@ -280,7 +280,7 @@ static ssize_t mode_store(struct device *device, struct device_attribute *attr,
 
 	return count;
 err:
-	spin_unlock(&afu->contexts_lock);
+	mutex_unlock(&afu->contexts_lock);
 	return rc;
 }