Skip to contentSkip to conversation

dma-debug: add dumping facility via debugfs

PATCH Superseded Corentin Labbe 4 participants 5 messages Jan 16, 2019 – Jan 22, 2019 DMA mapping helpers

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 dma-debug: add dumping facility via debugfs

While debugging a DMA mapping leak, I needed to access debug_dma_dump_mappings() but easily from user space.

This patch adds a /sys/kernel/debug/dma-api/dump file which contain all current DMA mapping.

---
 kernel/dma/debug.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)
kernel/dma/debug.c +47 -0
index 23cf5361bcf1..9253382f5729 100644
@@ -144,6 +144,7 @@ static struct dentry *num_free_entries_dent __read_mostly;
 static struct dentry *min_free_entries_dent __read_mostly;
 static struct dentry *nr_total_entries_dent __read_mostly;
 static struct dentry *filter_dent           __read_mostly;
+static struct dentry *dump_dent             __read_mostly;
 
 /* per-driver filter related state */
 
@@ -840,6 +841,47 @@ static const struct file_operations filter_fops = {
 	.llseek = default_llseek,
 };
 
+static int dump_read(struct seq_file *seq, void *v)
+{
+	int idx;
+
+	for (idx = 0; idx < HASH_SIZE; idx++) {
+		struct hash_bucket *bucket = &dma_entry_hash[idx];
+		struct dma_debug_entry *entry;
+		unsigned long flags;
+
+		spin_lock_irqsave(&bucket->lock, flags);
+
+		list_for_each_entry(entry, &bucket->list, list) {
+			seq_printf(seq,
+				   "%s %s %s idx %d P=%llx N=%lx D=%llx L=%llx %s %s\n",
+				   dev_name(entry->dev),
+				   dev_driver_string(entry->dev),
+				   type2name[entry->type], idx,
+				   phys_addr(entry), entry->pfn,
+				   entry->dev_addr, entry->size,
+				   dir2name[entry->direction],
+				   maperr2str[entry->map_err_type]);
+		}
+
+		spin_unlock_irqrestore(&bucket->lock, flags);
+	}
+	return 0;
+}
+
+static int dump_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dump_read, inode->i_private);
+}
+
+static const struct file_operations dump_fops = {
+	.owner = THIS_MODULE,
+	.open = dump_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
 static int dma_debug_fs_init(void)
 {
 	dma_debug_dent = debugfs_create_dir("dma-api", NULL);
@@ -894,6 +936,11 @@ static int dma_debug_fs_init(void)
 	if (!filter_dent)
 		goto out_err;
 
+	dump_dent = debugfs_create_file("dump", 0444,
+					dma_debug_dent, NULL, &dump_fops);
+	if (!dump_dent)
+		goto out_err;
+
 	return 0;
 
 out_err:
Signature · 1 line
2.19.2
reviewer
#1 lore raw
Quoted 44 lines · Corentin Labbe

While debugging a DMA mapping leak, I needed to access debug_dma_dump_mappings() but easily from user space.

This patch adds a /sys/kernel/debug/dma-api/dump file which contain all current DMA mapping.

---
  kernel/dma/debug.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++
  1 file changed, 47 insertions(+)
kernel/dma/debug.c +18 -0
index 23cf5361bcf1..9253382f5729 100644
@@ -144,6 +144,7 @@ static struct dentry *num_free_entries_dent __read_mostly;
  static struct dentry *min_free_entries_dent __read_mostly;
  static struct dentry *nr_total_entries_dent __read_mostly;
  static struct dentry *filter_dent           __read_mostly;
+static struct dentry *dump_dent             __read_mostly;
 
  /* per-driver filter related state */
 
@@ -840,6 +841,47 @@ static const struct file_operations filter_fops = {
  	.llseek = default_llseek,
  };
 
+static int dump_read(struct seq_file *seq, void *v)
+{
+	int idx;
+
+	for (idx = 0; idx < HASH_SIZE; idx++) {
+		struct hash_bucket *bucket = &dma_entry_hash[idx];
+		struct dma_debug_entry *entry;
+		unsigned long flags;
+
+		spin_lock_irqsave(&bucket->lock, flags);
+
+		list_for_each_entry(entry, &bucket->list, list) {
+			seq_printf(seq,
+				   "%s %s %s idx %d P=%llx N=%lx D=%llx L=%llx %s %s\n",
+				   dev_name(entry->dev),
+				   dev_driver_string(entry->dev),
+				   type2name[entry->type], idx,

kernel/dma/debug.c static const struct file_operations filter_fops = { · line 870

Quoted 10 lines
kernel/dma/debug.c +10 -0
+				   phys_addr(entry), entry->pfn,
+				   entry->dev_addr, entry->size,
+				   dir2name[entry->direction],
+				   maperr2str[entry->map_err_type]);
+		}
+
+		spin_unlock_irqrestore(&bucket->lock, flags);
+	}
+	return 0;
+}

It's a shame that this is pretty much a duplication of debug_dma_dump_mappings(), but there seems no straightforward way to define one in terms of the other :/

(although given that we'd rather not have that weird external interface anyway, maybe "now you can use debugfs instead" might be justification enough for cleaning it up...)

Quoted 4 lines
kernel/dma/debug.c +4 -0
+
+static int dump_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dump_read, inode->i_private);

kernel/dma/debug.c static const struct file_operations filter_fops = { · line 884

Quoted 10 lines
kernel/dma/debug.c +10 -0
+}
+
+static const struct file_operations dump_fops = {
+	.owner = THIS_MODULE,
+	.open = dump_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+

DEFINE_SHOW_ATTRIBUTE()?

Robin.

Quoted 16 lines
kernel/dma/debug.c +5 -0
  static int dma_debug_fs_init(void)
  {
  	dma_debug_dent = debugfs_create_dir("dma-api", NULL);
@@ -894,6 +936,11 @@ static int dma_debug_fs_init(void)
  	if (!filter_dent)
  		goto out_err;
 
+	dump_dent = debugfs_create_file("dump", 0444,
+					dma_debug_dent, NULL, &dump_fops);
+	if (!dump_dent)
+		goto out_err;
+
  	return 0;
 
  out_err:
#2 lore raw
Quoted 3 lines · Robin Murphy

It's a shame that this is pretty much a duplication of debug_dma_dump_mappings(), but there seems no straightforward way to define one in terms of the other :/

We could always play some macro magic, but I think that is worse than duplicating a little bit of functionality.

Btw, the dev argument to debug_dma_dump_mappings is always NULL and could be removed..

Quoted 3 lines

(although given that we'd rather not have that weird external interface anyway, maybe "now you can use debugfs instead" might be justification enough for cleaning it up...)

One argument against that is the system might be in a shape where you can't easily read a file when it is in that shape. The argument for that is that in many systems the full list of mappings might overwhelm the kernel log.

Adding Joerg, who originally wrote the code in case he has an opinion.

Quoted 20 lines
Quoted 5 lines
kernel/dma/debug.c +4 -0
+
+static int dump_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dump_read, inode->i_private);

kernel/dma/debug.c static const struct file_operations filter_fops = { · line 884

Quoted 10 lines
kernel/dma/debug.c +10 -0
+}
+
+static const struct file_operations dump_fops = {
+	.owner = THIS_MODULE,
+	.open = dump_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+

DEFINE_SHOW_ATTRIBUTE()?

Robin.

Quoted 14 lines
Quoted 4 lines
  static int dma_debug_fs_init(void)
  {
  	dma_debug_dent = debugfs_create_dir("dma-api", NULL);

kernel/dma/debug.c static int dma_debug_fs_init(void) · line 945

Quoted 10 lines
kernel/dma/debug.c +4 -0
@@ -894,6 +936,11 @@ static int dma_debug_fs_init(void)
  	if (!filter_dent)
  		goto out_err;
  +	dump_dent = debugfs_create_file("dump", 0444,
+					dma_debug_dent, NULL, &dump_fops);
+	if (!dump_dent)
+		goto out_err;
+
  	return 0;
    out_err:
 

---end quoted text---

#3 lore raw
Quoted 7 lines · Christoph Hellwig
Quoted 3 lines · Robin Murphy

It's a shame that this is pretty much a duplication of debug_dma_dump_mappings(), but there seems no straightforward way to define one in terms of the other :/

We could always play some macro magic, but I think that is worse than duplicating a little bit of functionality.

I havn't checked in detail, can seq_buf be used somehow to write a function that fits both cases?

Quoted 2 lines

Btw, the dev argument to debug_dma_dump_mappings is always NULL and could be removed..

This function was also written as a debug-helper for driver developers. As such it might not make it into the final driver, but the developer might want to use it to dump the mappings for her device only. So I'd like to keep the parameter, even when it is always NULL for in-kernel uses.

Quoted 8 lines
Quoted 3 lines

(although given that we'd rather not have that weird external interface anyway, maybe "now you can use debugfs instead" might be justification enough for cleaning it up...)

One argument against that is the system might be in a shape where you can't easily read a file when it is in that shape. The argument for that is that in many systems the full list of mappings might overwhelm the kernel log.

For driver developers a file is easier to use, but in case of an unusable system one can at least easily read out parts of the dma-mappings from a crash-dump if it is in the kernel-log. So I think it makes sense to have both.

Regards,

Joerg

#4 lore raw
In reply to Robin Murphy
Quoted 64 lines · Robin Murphy
Quoted 44 lines · Corentin Labbe

While debugging a DMA mapping leak, I needed to access debug_dma_dump_mappings() but easily from user space.

This patch adds a /sys/kernel/debug/dma-api/dump file which contain all current DMA mapping.

---
  kernel/dma/debug.c | 47 ++++++++++++++++++++++++++++++++++++++++++++++
  1 file changed, 47 insertions(+)
kernel/dma/debug.c +18 -0
index 23cf5361bcf1..9253382f5729 100644
@@ -144,6 +144,7 @@ static struct dentry *num_free_entries_dent __read_mostly;
  static struct dentry *min_free_entries_dent __read_mostly;
  static struct dentry *nr_total_entries_dent __read_mostly;
  static struct dentry *filter_dent           __read_mostly;
+static struct dentry *dump_dent             __read_mostly;
 
  /* per-driver filter related state */
 
@@ -840,6 +841,47 @@ static const struct file_operations filter_fops = {
  	.llseek = default_llseek,
  };
 
+static int dump_read(struct seq_file *seq, void *v)
+{
+	int idx;
+
+	for (idx = 0; idx < HASH_SIZE; idx++) {
+		struct hash_bucket *bucket = &dma_entry_hash[idx];
+		struct dma_debug_entry *entry;
+		unsigned long flags;
+
+		spin_lock_irqsave(&bucket->lock, flags);
+
+		list_for_each_entry(entry, &bucket->list, list) {
+			seq_printf(seq,
+				   "%s %s %s idx %d P=%llx N=%lx D=%llx L=%llx %s %s\n",
+				   dev_name(entry->dev),
+				   dev_driver_string(entry->dev),
+				   type2name[entry->type], idx,

kernel/dma/debug.c static const struct file_operations filter_fops = { · line 870

Quoted 10 lines
kernel/dma/debug.c +10 -0
+				   phys_addr(entry), entry->pfn,
+				   entry->dev_addr, entry->size,
+				   dir2name[entry->direction],
+				   maperr2str[entry->map_err_type]);
+		}
+
+		spin_unlock_irqrestore(&bucket->lock, flags);
+	}
+	return 0;
+}

It's a shame that this is pretty much a duplication of debug_dma_dump_mappings(), but there seems no straightforward way to define one in terms of the other :/

(although given that we'd rather not have that weird external interface anyway, maybe "now you can use debugfs instead" might be justification enough for cleaning it up...)

Quoted 17 lines
Quoted 4 lines
kernel/dma/debug.c +4 -0
+
+static int dump_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, dump_read, inode->i_private);

kernel/dma/debug.c static const struct file_operations filter_fops = { · line 884

Quoted 10 lines
kernel/dma/debug.c +10 -0
+}
+
+static const struct file_operations dump_fops = {
+	.owner = THIS_MODULE,
+	.open = dump_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+

DEFINE_SHOW_ATTRIBUTE()?

I will use it

Thanks