Skip to contentSkip to conversation

zram: auto add/del devices on demand

PATCH Superseded Timofey Titovets 3 participants 8 messages Jul 17, 2014

Subsystems (3)
Documentation Block layer zram

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Timofey Titovets

author
#0 lore raw

PATCH v2 1/1 zram: auto add/del devices on demand

From: Timofey Titovets <[email protected]>

This add supporting of autochange count of zram devices on demand, like loop devices; This working by following rules:
- Always save minimum devices count specified by num_device (can be
specified while kernel module loading)
- if last device already using add new device;
- if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2:
Delete useless variable 'ret', added documentation for explain new zram behaviour

Please pull from:
https://github.com/Nefelim4ag/linux.git
---
Tested on 3.15.5-2-ARCH, can be applied on any kernel version
after this patch 'zram: add LZ4 compression support' - https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260
---

Documentation/blockdev/zram.txt +11 -1
index 0595c3f..7f5c921 100644
@@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using zram.
 
 1) Load Module:
 	modprobe zram num_devices=4
-	This creates 4 devices: /dev/zram{0,1,2,3}
+	This pre creates 4 devices: /dev/zram{0,1,2,3}
 	(num_devices parameter is optional. Default: 1)
 
+	Kernel dynamically changes number of zram devices by demand
+	(between num_devices and 31)
+	If set disk size(4) for last device
+	kernel automatically adds new zram device
+	If last two devices have zero disk size after reset(8),
+	kernel will destroy last device
+
+	Summing up all features, comes one simple rule:
+	"The last zram device is always free for use"
+
 2) Set max number of compression streams
 	Compression backend may use up to max_comp_streams compression streams,
 	thus allowing up to max_comp_streams concurrent compression operations.
drivers/block/zram/zram_drv.c +51 -4

131 lines not on this page · raw message

#1 lore raw
Quoted 33 lines · Timofey Titovets

This add supporting of autochange count of zram devices on demand, like loop devices;
This working by following rules:
- Always save minimum devices count specified by num_device (can be
specified while kernel module loading)
- if last device already using add new device;
- if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2:
Delete useless variable 'ret', added documentation for explain new zram behaviour

Please pull from:
https://github.com/Nefelim4ag/linux.git
---
Tested on 3.15.5-2-ARCH, can be applied on any kernel version
after this patch 'zram: add LZ4 compression support' - https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260
---

Documentation/blockdev/zram.txt +0 -0
index 0595c3f..7f5c921 100644
@@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using

zram.

1) Load Module:

Documentation/blockdev/zram.txt +1 -1
 	modprobe zram num_devices=4
-	This creates 4 devices: /dev/zram{0,1,2,3}
+	This pre creates 4 devices: /dev/zram{0,1,2,3}
 	(num_devices parameter is optional. Default: 1)

Documentation/blockdev/zram.txt Following shows a typical sequence of steps for using zram. · line 32

Quoted 10 lines
Documentation/blockdev/zram.txt +9 -0
 
+	Kernel dynamically changes number of zram devices by demand
+	(between num_devices and 31)
+	If set disk size(4) for last device
+	kernel automatically adds new zram device
+	If last two devices have zero disk size after reset(8),
+	kernel will destroy last device
+
+	Summing up all features, comes one simple rule:
+	"The last zram device is always free for use"

I can't understand what is the real benefit?

Quoted 70 lines

2) Set max number of compression streams

Documentation/blockdev/zram.txt +0 -0
 	Compression backend may use up to max_comp_streams compression streams,
 	thus allowing up to max_comp_streams concurrent compression operations.
drivers/block/zram/zram_drv.c +7 -0
index 089e72c..9b2fc89 100644
@@ -42,6 +42,10 @@ static const char *default_compressor = "lzo";
 
 /* Module params (documentation at end) */
 static unsigned int num_devices = 1;
+static unsigned int last_created_dev = 1;
+
+static void zram_add_dev(void);
+static void zram_del_dev(void);
 #define ZRAM_ATTR_RO(name)						\
 static ssize_t zram_attr_##name##_show(struct device *d,		\

@@ -168,6 +172,7 @@ static ssize_t comp_algorithm_store(struct device *dev,
 		struct device_attribute *attr, const char *buf, size_t len)
 {
 	struct zram *zram = dev_to_zram(dev);
+
 	down_write(&zram->init_lock);
 	if (init_done(zram)) {
 		up_write(&zram->init_lock);

@@ -239,6 +244,7 @@ static struct zram_meta *zram_meta_alloc(u64 disksize)
 {
 	size_t num_pages;
 	struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL);
+
 	if (!meta)
 		goto out;
 

@@ -374,6 +380,7 @@ static int zram_bvec_read(struct zram *zram, struct
 bio_vec *bvec,
 	struct page *page;
 	unsigned char *user_mem, *uncmem = NULL;
 	struct zram_meta *meta = zram->meta;
+
 	page = bvec->bv_page;
drivers/block/zram/zram_drv.c +3 -0
 	read_lock(&meta->tb_lock);

@@ -607,6 +614,7 @@ static void zram_reset_device(struct zram *zram, bool
 reset_capacity)
 	/* Free all pages that are still in this zram device */
 	for (index = 0; index < zram->disksize >> PAGE_SHIFT; index++) {
 		unsigned long handle = meta->table[index].handle;
+
 		if (!handle)
 			continue;

@@ -668,6 +676,7 @@ static ssize_t disksize_store(struct device *dev,
 	set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
 	revalidate_disk(zram->disk);
 	up_write(&zram->init_lock);
+	zram_add_dev();
 	return len;
 
 out_destroy_comp:

@@ -712,6 +721,7 @@ static ssize_t reset_store(struct device *dev,
 	bdput(bdev);
 
 	zram_reset_device(zram, true);
+	zram_del_dev();
 	return len;
 
 out:

@@ -954,6 +964,41 @@ static void destroy_device(struct zram *zram)
 	blk_cleanup_queue(zram->queue);
 }

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 975

Quoted 10 lines
drivers/block/zram/zram_drv.c +9 -0
 
+/* remove last free device, if last and prelast dev a free */
+static void zram_del_dev(void)
+{
+	if (last_created_dev < num_devices)
+		return;
+
+	if (zram_devices[last_created_dev].disksize == 0 &&
+		zram_devices[last_created_dev-1].disksize == 0
+		) {

racy?

Quoted 4 lines
drivers/block/zram/zram_drv.c +4 -0
+		destroy_device(&zram_devices[last_created_dev]);
+		last_created_dev--;
+		pr_info("Deleted zram%u device\n", last_created_dev);
+	}

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 989

Quoted 10 lines
drivers/block/zram/zram_drv.c +10 -0
+}
+
+/* Auto add empty zram device, if last device in use */
+static void zram_add_dev(void)
+{
+	if (last_created_dev+1 > max_num_devices) {
+		pr_warn("Can't add zram%u, max device number reached\n",
+			num_devices);
+		return;
+	}

racy?

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 995

Quoted 5 lines
drivers/block/zram/zram_drv.c +5 -0
+	if (&zram_devices[last_created_dev].disksize > 0) {
+		last_created_dev++;
+		if (create_device(&zram_devices[last_created_dev], last_created_dev)) {
+			destroy_device(&zram_devices[last_created_dev]);
+			last_created_dev--;

racy?

	-ss
Quoted 51 lines
drivers/block/zram/zram_drv.c +12 -4
+			return;
+		}
+		pr_info("Created zram%u device\n", last_created_dev);
+	}
+}
+
 static int __init zram_init(void)
 {
 	int ret, dev_id;

@@ -972,18 +1017,20 @@ static int __init zram_init(void)
 		goto out;
 	}
 
-	/* Allocate the device array and initialize each one */
-	zram_devices = kzalloc(num_devices * sizeof(struct zram), GFP_KERNEL);
+	/* Allocate the device array */
+	zram_devices = kcalloc(max_num_devices, sizeof(struct zram), GFP_KERNEL);
 	if (!zram_devices) {
 		ret = -ENOMEM;
 		goto unregister;
 	}
 
+	/* Initialise zram{0..num_devices} */
 	for (dev_id = 0; dev_id < num_devices; dev_id++) {
 		ret = create_device(&zram_devices[dev_id], dev_id);
 		if (ret)
 			goto free_devices;
 	}
+	last_created_dev = num_devices-1;
 
 	pr_info("Created %u device(s) ...\n", num_devices);
 

@@ -1004,7 +1051,7 @@ static void __exit zram_exit(void)
 	int i;
 	struct zram *zram;
 
-	for (i = 0; i < num_devices; i++) {
+	for (i = 0; i < last_created_dev+1; i++) {
 		zram = &zram_devices[i];
 
 		destroy_device(zram);

@@ -1025,7 +1072,7 @@ module_init(zram_init);
 module_exit(zram_exit);
 
 module_param(num_devices, uint, 0);
-MODULE_PARM_DESC(num_devices, "Number of zram devices");
+MODULE_PARM_DESC(num_devices, "Number of pre created  zram devices");
 
 MODULE_LICENSE("Dual BSD/GPL");
 MODULE_AUTHOR("Nitin Gupta <[email protected]>");
author
#2 lore raw

PATCH v2 1/1 zram: auto add/del devices on demand

Quoted 204 lines · Sergey Senozhatsky
Quoted 33 lines · Timofey Titovets

This add supporting of autochange count of zram devices on demand, like loop devices;
This working by following rules:
- Always save minimum devices count specified by num_device (can be
specified while kernel module loading)
- if last device already using add new device;
- if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2:
Delete useless variable 'ret', added documentation for explain new zram behaviour

Please pull from:
https://github.com/Nefelim4ag/linux.git
---
Tested on 3.15.5-2-ARCH, can be applied on any kernel version
after this patch 'zram: add LZ4 compression support' - https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260
---

Documentation/blockdev/zram.txt +0 -0
index 0595c3f..7f5c921 100644
@@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using

zram.

1) Load Module:
modprobe zram num_devices=4

Documentation/blockdev/zram.txt +1 -1
-	This creates 4 devices: /dev/zram{0,1,2,3}
+	This pre creates 4 devices: /dev/zram{0,1,2,3}
  	(num_devices parameter is optional. Default: 1)

Documentation/blockdev/zram.txt Following shows a typical sequence of steps for using zram. · line 32

Quoted 10 lines
Documentation/blockdev/zram.txt +9 -0
 
+	Kernel dynamically changes number of zram devices by demand
+	(between num_devices and 31)
+	If set disk size(4) for last device
+	kernel automatically adds new zram device
+	If last two devices have zero disk size after reset(8),
+	kernel will destroy last device
+
+	Summing up all features, comes one simple rule:
+	"The last zram device is always free for use"

I can't understand what is the real benefit?

Quoted 70 lines

2) Set max number of compression streams
Compression backend may use up to max_comp_streams compression streams, thus allowing up to max_comp_streams concurrent compression operations.

drivers/block/zram/zram_drv.c +7 -0
index 089e72c..9b2fc89 100644
@@ -42,6 +42,10 @@ static const char *default_compressor = "lzo";
 
  /* Module params (documentation at end) */
  static unsigned int num_devices = 1;
+static unsigned int last_created_dev = 1;
+
+static void zram_add_dev(void);
+static void zram_del_dev(void);
  #define ZRAM_ATTR_RO(name)						\
  static ssize_t zram_attr_##name##_show(struct device *d,		\

@@ -168,6 +172,7 @@ static ssize_t comp_algorithm_store(struct device *dev,
  		struct device_attribute *attr, const char *buf, size_t len)
  {
  	struct zram *zram = dev_to_zram(dev);
+
  	down_write(&zram->init_lock);
  	if (init_done(zram)) {
  		up_write(&zram->init_lock);

@@ -239,6 +244,7 @@ static struct zram_meta *zram_meta_alloc(u64 disksize)
  {
  	size_t num_pages;
  	struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL);
+
  	if (!meta)
  		goto out;
 

@@ -374,6 +380,7 @@ static int zram_bvec_read(struct zram *zram, struct
 bio_vec *bvec,
  	struct page *page;
  	unsigned char *user_mem, *uncmem = NULL;
  	struct zram_meta *meta = zram->meta;
+
  	page = bvec->bv_page;
 

  	read_lock(&meta->tb_lock);
drivers/block/zram/zram_drv.c +3 -0
@@ -607,6 +614,7 @@ static void zram_reset_device(struct zram *zram, bool
 reset_capacity)
  	/* Free all pages that are still in this zram device */
  	for (index = 0; index < zram->disksize >> PAGE_SHIFT; index++) {
  		unsigned long handle = meta->table[index].handle;
+
  		if (!handle)
  			continue;

@@ -668,6 +676,7 @@ static ssize_t disksize_store(struct device *dev,
  	set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);
  	revalidate_disk(zram->disk);
  	up_write(&zram->init_lock);
+	zram_add_dev();
  	return len;
 
  out_destroy_comp:

@@ -712,6 +721,7 @@ static ssize_t reset_store(struct device *dev,
  	bdput(bdev);
 
  	zram_reset_device(zram, true);
+	zram_del_dev();
  	return len;
 
  out:

@@ -954,6 +964,41 @@ static void destroy_device(struct zram *zram)
  	blk_cleanup_queue(zram->queue);
  }

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 975

Quoted 10 lines
drivers/block/zram/zram_drv.c +9 -0
 
+/* remove last free device, if last and prelast dev a free */
+static void zram_del_dev(void)
+{
+	if (last_created_dev < num_devices)
+		return;
+
+	if (zram_devices[last_created_dev].disksize == 0 &&
+		zram_devices[last_created_dev-1].disksize == 0
+		) {

racy?

Quoted 4 lines
drivers/block/zram/zram_drv.c +4 -0
+		destroy_device(&zram_devices[last_created_dev]);
+		last_created_dev--;
+		pr_info("Deleted zram%u device\n", last_created_dev);
+	}

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 989

Quoted 10 lines
drivers/block/zram/zram_drv.c +10 -0
+}
+
+/* Auto add empty zram device, if last device in use */
+static void zram_add_dev(void)
+{
+	if (last_created_dev+1 > max_num_devices) {
+		pr_warn("Can't add zram%u, max device number reached\n",
+			num_devices);
+		return;
+	}

racy?

drivers/block/zram/zram_drv.c static void destroy_device(struct zram *zram) · line 995

Quoted 5 lines
drivers/block/zram/zram_drv.c +5 -0
+	if (&zram_devices[last_created_dev].disksize > 0) {
+		last_created_dev++;
+		if (create_device(&zram_devices[last_created_dev], last_created_dev)) {
+			destroy_device(&zram_devices[last_created_dev]);
+			last_created_dev--;

racy?

	-ss
Quoted 44 lines
drivers/block/zram/zram_drv.c +11 -3
+			return;
+		}
+		pr_info("Created zram%u device\n", last_created_dev);
+	}
+}
+
  static int __init zram_init(void)
  {
  	int ret, dev_id;

@@ -972,18 +1017,20 @@ static int __init zram_init(void)
  		goto out;
  	}
 
-	/* Allocate the device array and initialize each one */
-	zram_devices = kzalloc(num_devices * sizeof(struct zram), GFP_KERNEL);
+	/* Allocate the device array */
+	zram_devices = kcalloc(max_num_devices, sizeof(struct zram), GFP_KERNEL);
  	if (!zram_devices) {
  		ret = -ENOMEM;
  		goto unregister;
  	}
 
+	/* Initialise zram{0..num_devices} */
  	for (dev_id = 0; dev_id < num_devices; dev_id++) {
  		ret = create_device(&zram_devices[dev_id], dev_id);
  		if (ret)
  			goto free_devices;
  	}
+	last_created_dev = num_devices-1;
 
  	pr_info("Created %u device(s) ...\n", num_devices);
 

@@ -1004,7 +1051,7 @@ static void __exit zram_exit(void)
  	int i;
  	struct zram *zram;
 
-	for (i = 0; i < num_devices; i++) {
+	for (i = 0; i < last_created_dev+1; i++) {
  		zram = &zram_devices[i];
 
  		destroy_device(zram);

@@ -1025,7 +1072,7 @@ module_init(zram_init);
  module_exit(zram_exit);
Quoted 7 lines

drivers/block/zram/zram_drv.c module_init(zram_init); · line 1078

Quoted 7 lines
drivers/block/zram/zram_drv.c +1 -1
  module_param(num_devices, uint, 0);
-MODULE_PARM_DESC(num_devices, "Number of zram devices");
+MODULE_PARM_DESC(num_devices, "Number of pre created  zram devices");
 
  MODULE_LICENSE("Dual BSD/GPL");
  MODULE_AUTHOR("Nitin Gupta <[email protected]>");

I can't understand what do you mean on "racy?"
if you mind random empty lines in code, I've just checked code by checkpatch.pl and fixed warnings.

I'll try to explain, why I wrote a patch:
zram, as a loop device, is a virtual device, not physical, and i think that adding new free devices automatically (like loop) is useful and expected behavior (for virtual device).

By default, zram creates only one device, and if I need to use two devices (for example), my actions lead to module’s reloading:
1. umount busy device
2. lose my data. Yes, I can backup it, and restore, but this is also useless, if I can do it more beautiful
3. unload module
4. load module again with new num devices parameter or make it parametre "permanent" by creating file in modprobe.d

This is handwork and such behavior is uncomfortable.

if number of devices is dynamically changeable, I can skip this actions and save my time and energy instead of wasting on these activities.

#5 lore raw
In reply to Sergey Senozhatsky

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 13 lines · Sergey Senozhatsky
Quoted 12 lines · Timofey Titovets

This add supporting of autochange count of zram devices on demand, like loop devices; This working by following rules: - Always save minimum devices count specified by num_device (can be specified while kernel module loading) - if last device already using add new device; - if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2: Delete useless variable 'ret', added documentation for explain new zram behaviour

Quoted 6 lines
Quoted 6 lines

Please pull from: https://github.com/Nefelim4ag/linux.git --- Tested on 3.15.5-2-ARCH, can be applied on any kernel version after this patch 'zram: add LZ4 compression support' -
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260

- ---

Quoted 86 lines
Quoted 12 lines
(quoted diff) +0 -0

b/Documentation/blockdev/zram.txt index 0595c3f..7f5c921 100644 --- a/Documentation/blockdev/zram.txt +++ b/Documentation/blockdev/zram.txt @@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using zram.

1) Load Module: modprobe zram num_devices=4 - This creates 4
devices: /dev/zram{0,1,2,3} + This pre creates 4 devices:
/dev/zram{0,1,2,3} (num_devices parameter is optional. Default:
1)

(quoted diff) +1 -0
+	Kernel dynamically changes number of zram devices by demand +
Quoted 5 lines

(between num_devices and 31) + If set disk size(4) for last
device + kernel automatically adds new zram device + If last two
devices have zero disk size after reset(8), + kernel will destroy
last device + + Summing up all features, comes one simple rule: +
"The last zram device is always free for use"

I can't understand what is the real benefit?

Quoted 49 lines

2) Set max number of compression streams Compression backend may use up to max_comp_streams compression streams, thus allowing up to max_comp_streams concurrent compression operations. diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 089e72c..9b2fc89 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -42,6 +42,10 @@ static const char *default_compressor = "lzo";

/* Module params (documentation at end) */ static unsigned int num_devices = 1; +static unsigned int last_created_dev = 1; + +static void zram_add_dev(void); +static void
zram_del_dev(void); #define ZRAM_ATTR_RO(name) \ static
ssize_t zram_attr_##name##_show(struct device *d, \ @@ -168,6
+172,7 @@ static ssize_t comp_algorithm_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t len) { struct zram *zram = dev_to_zram(dev); + down_write(&zram->init_lock); if (init_done(zram)) { up_write(&zram->init_lock); @@ -239,6 +244,7 @@ static struct zram_meta *zram_meta_alloc(u64 disksize) { size_t num_pages; struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL); + if (!meta) goto out;

(quoted diff) +0 -0
@@ -374,6 +380,7 @@ static int zram_bvec_read(struct zram *zram,
 struct bio_vec *bvec, struct page *page; unsigned char *user_mem,
 *uncmem = NULL; struct zram_meta *meta = zram->meta; + page =
 bvec->bv_page;

read_lock(&meta->tb_lock); @@ -607,6 +614,7 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity) /* Free all pages that are still in this zram device */ for (index = 0; index < zram->disksize >> PAGE_SHIFT; index++) { unsigned long handle = meta->table[index].handle; + if (!handle) continue;

(quoted diff) +0 -0
@@ -668,6 +676,7 @@ static ssize_t disksize_store(struct device
 *dev, set_capacity(zram->disk, zram->disksize >> SECTOR_SHIFT);

revalidate_disk(zram->disk); up_write(&zram->init_lock); + zram_add_dev(); return len;

out_destroy_comp: @@ -712,6 +721,7 @@ static ssize_t reset_store(struct device *dev, bdput(bdev);

zram_reset_device(zram, true); + zram_del_dev(); return len;

out: @@ -954,6 +964,41 @@ static void destroy_device(struct zram *zram) blk_cleanup_queue(zram->queue); }

(quoted diff) +2 -0
+/* remove last free device, if last and prelast dev a free */ 
+static void zram_del_dev(void) +{ +	if (last_created_dev <
Quoted 3 lines

num_devices) + return; + + if
(zram_devices[last_created_dev].disksize == 0 && +
zram_devices[last_created_dev-1].disksize == 0 + ) {

racy?

Quoted 7 lines
(quoted diff) +1 -0
+		destroy_device(&zram_devices[last_created_dev]); +

last_created_dev--; + pr_info("Deleted zram%u device\n",
last_created_dev); + } +} + +/* Auto add empty zram device, if
last device in use */ +static void zram_add_dev(void) +{ + if
(last_created_dev+1 > max_num_devices) { + pr_warn("Can't add
zram%u, max device number reached\n", + num_devices); +
return; + }

racy?

Quoted 8 lines
Quoted 6 lines
(quoted diff) +1 -0
+	if (&zram_devices[last_created_dev].disksize > 0) { +

last_created_dev++; + if
(create_device(&zram_devices[last_created_dev], last_created_dev)) { + destroy_device(&zram_devices[last_created_dev]); + last_created_dev--;

racy?

Looks like it:

$ cat conctest.sh
#! /bin/sh

modprobe zram

while true; do

    for i in `seq 1 10`; do
	echo 10485760 > /sys/block/zram0/disksize&
	echo 1 > /sys/block/zram0/reset&
    done
done
$ sudo ./conctest.sh
[   51.535387] ------------[ cut here ]------------
[   51.536305] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.537908] kernfs: can not remove 'max_comp_streams', no directory
[   51.539089] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.550844] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
     3.16.0-rc5-memacct-next-20140715+ #146
[   51.552606] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.553691]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.555412]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320242
[   51.557811]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.559580] Call Trace:
[   51.560139]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.561161]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.562333]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.563445]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.564432]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.565627]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.566753]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.567870]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.569089]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.570197]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.571247]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.572201]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.573283]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.574143]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.574976]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.576213]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.577422] ---[ end trace cdb79035dad379f3 ]---
[   51.578401] ------------[ cut here ]------------
[   51.579354] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.580966] kernfs: can not remove 'comp_algorithm', no directory
[   51.582108] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.597177] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
     3.16.0-rc5-memacct-next-20140715+ #146
[   51.598549] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.599661]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.601519]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320233
[   51.603362]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.604913] Call Trace:
[   51.605444]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.606537]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.607507]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.608582]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.609423]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.610436]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.611408]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.612431]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.613214]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.613928]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.614634]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.615387]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.616246]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.619107]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.620063]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.622071]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.623256] ---[ end trace cdb79035dad379f4 ]---

And lots more of similar messages.

Jerome

Quoted 32 lines

-ss

Quoted 27 lines
(quoted diff) +1 -0
+			return; +		} +		pr_info("Created zram%u device\n",

last_created_dev); + } +} + static int __init zram_init(void) {
int ret, dev_id; @@ -972,18 +1017,20 @@ static int __init zram_init(void) goto out; }

(quoted diff) +0 -1
-	/* Allocate the device array and initialize each one */ -

zram_devices = kzalloc(num_devices * sizeof(struct zram),
GFP_KERNEL); + /* Allocate the device array */ + zram_devices =
kcalloc(max_num_devices, sizeof(struct zram), GFP_KERNEL); if (!zram_devices) { ret = -ENOMEM; goto unregister; }

(quoted diff) +1 -0
+	/* Initialise zram{0..num_devices} */ for (dev_id = 0; dev_id <

num_devices; dev_id++) { ret = create_device(&zram_devices[dev_id], dev_id); if (ret) goto
free_devices; } + last_created_dev = num_devices-1;

pr_info("Created %u device(s) ...\n", num_devices);

drivers/block/zram/zram_drv.c +0 -1
@@ -1004,7 +1051,7 @@ static void __exit zram_exit(void) int i; 
 struct zram *zram;
 
-	for (i = 0; i < num_devices; i++) { +	for (i = 0; i <
 last_created_dev+1; i++) { zram = &zram_devices[i];

destroy_device(zram); @@ -1025,7 +1072,7 @@
module_init(zram_init); module_exit(zram_exit);

Quoted 8 lines
Quoted 8 lines

module_param(num_devices, uint, 0);

(quoted diff) +1 -1
-MODULE_PARM_DESC(num_devices, "Number of zram devices"); 
+MODULE_PARM_DESC(num_devices, "Number of pre created  zram

devices");

MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Nitin Gupta <[email protected]>");

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJTx9rXAAoJEHTzHJCtsuoC2UAH/1k+IhcKZjcj8526VTnUdYNX SEJuVbiDU/8qeTLevw15+JLrYVeaoRIincjPVU+J2d0eiobHjdDcx6oSZ/d8ryb3 hKy3r209Thii0LJiqiHL9+8NBGZNldgoirz74PWYUoB+XlSo9mA3OcX1+TDJWgSu SytEUCX+viaclfVwCU+zU10X/Pd8m+AxWXf8zKdIcpE44iGvw+vDx/G8wV2N5ATy 0D/cLZsYMAkNXMap0eCS61oq4ZDdEAi13+563fKpinKqcK0sV41LMBiWbGR0rFm3 BAbZVHAlT7YPnpohlVUqEsxgp38sLNpeBKNi9JR9B4vlMEmsLBuW4FWG9/ibCA4= =VOGL
-----END PGP SIGNATURE-----

author
#6 lore raw
Quoted 100 lines · Jerome Marchand

Looks like it:

$ cat conctest.sh
#! /bin/sh

modprobe zram

while true; do

     for i in `seq 1 10`; do
	echo 10485760 > /sys/block/zram0/disksize&
	echo 1 > /sys/block/zram0/reset&
     done
done
$ sudo ./conctest.sh
[   51.535387] ------------[ cut here ]------------
[   51.536305] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.537908] kernfs: can not remove 'max_comp_streams', no directory
[   51.539089] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.550844] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
      3.16.0-rc5-memacct-next-20140715+ #146
[   51.552606] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.553691]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.555412]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320242
[   51.557811]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.559580] Call Trace:
[   51.560139]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.561161]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.562333]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.563445]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.564432]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.565627]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.566753]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.567870]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.569089]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.570197]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.571247]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.572201]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.573283]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.574143]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.574976]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.576213]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.577422] ---[ end trace cdb79035dad379f3 ]---
[   51.578401] ------------[ cut here ]------------
[   51.579354] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.580966] kernfs: can not remove 'comp_algorithm', no directory
[   51.582108] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.597177] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
      3.16.0-rc5-memacct-next-20140715+ #146
[   51.598549] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.599661]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.601519]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320233
[   51.603362]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.604913] Call Trace:
[   51.605444]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.606537]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.607507]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.608582]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.609423]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.610436]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.611408]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.612431]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.613214]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.613928]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.614634]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.615387]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.616246]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.619107]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.620063]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.622071]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.623256] ---[ end trace cdb79035dad379f4 ]---
Quoted 4 lines

And lots more of similar messages.

Jerome

Thanks for catching, i try fix it and resend patch.

#3 lore raw
In reply to Timofey Titovets

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 16 lines · Timofey Titovets
Quoted 13 lines · Sergey Senozhatsky
Quoted 12 lines · Timofey Titovets

This add supporting of autochange count of zram devices on demand, like loop devices; This working by following rules: - Always save minimum devices count specified by num_device (can be specified while kernel module loading) - if last device already using add new device; - if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2: Delete useless variable 'ret', added documentation for explain new zram behaviour

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

Please pull from: https://github.com/Nefelim4ag/linux.git --- Tested on 3.15.5-2-ARCH, can be applied on any kernel version after this patch 'zram: add LZ4 compression support' -
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260

- ---

Quoted 135 lines
Quoted 135 lines
Quoted 11 lines
(quoted diff) +0 -0

b/Documentation/blockdev/zram.txt index 0595c3f..7f5c921 100644 --- a/Documentation/blockdev/zram.txt +++ b/Documentation/blockdev/zram.txt @@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using zram.

1) Load Module: modprobe zram num_devices=4 - This creates 4
devices: /dev/zram{0,1,2,3} + This pre creates 4 devices:
/dev/zram{0,1,2,3} (num_devices parameter is optional. Default:
1)

Quoted 7 lines

+ Kernel dynamically changes number of zram devices by
demand + (between num_devices and 31) + If set disk
size(4) for last device + kernel automatically adds new zram
device + If last two devices have zero disk size after
reset(8), + kernel will destroy last device + + Summing
up all features, comes one simple rule: + "The last zram
device is always free for use"

I can't understand what is the real benefit?

Quoted 51 lines

2) Set max number of compression streams Compression backend may use up to max_comp_streams compression streams, thus allowing up to max_comp_streams concurrent compression operations. diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 089e72c..9b2fc89 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -42,6 +42,10 @@ static const char *default_compressor = "lzo";

/* Module params (documentation at end) */ static unsigned int num_devices = 1; +static unsigned int last_created_dev = 1; + +static void zram_add_dev(void); +static void zram_del_dev(void); #define ZRAM_ATTR_RO(name)
\ static ssize_t zram_attr_##name##_show(struct device *d, \ @@ -168,6 +172,7 @@ static ssize_t comp_algorithm_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t len) { struct zram *zram = dev_to_zram(dev); + down_write(&zram->init_lock); if (init_done(zram)) { up_write(&zram->init_lock); @@ -239,6 +244,7 @@ static struct zram_meta *zram_meta_alloc(u64 disksize) { size_t num_pages; struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL); + if (!meta) goto out;

(quoted diff) +0 -0
@@ -374,6 +380,7 @@ static int zram_bvec_read(struct zram
 *zram, struct bio_vec *bvec, struct page *page; unsigned char
 *user_mem, *uncmem = NULL; struct zram_meta *meta =
 zram->meta; + page = bvec->bv_page;

read_lock(&meta->tb_lock); @@ -607,6 +614,7 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity) /* Free all pages that are still in this zram device */ for (index = 0; index < zram->disksize >> PAGE_SHIFT; index++) { unsigned long handle = meta->table[index].handle; + if (!handle) continue;

(quoted diff) +0 -0
@@ -668,6 +676,7 @@ static ssize_t disksize_store(struct device
 *dev, set_capacity(zram->disk, zram->disksize >>
 SECTOR_SHIFT); revalidate_disk(zram->disk); 
 up_write(&zram->init_lock); +    zram_add_dev(); return len;

out_destroy_comp: @@ -712,6 +721,7 @@ static ssize_t reset_store(struct device *dev, bdput(bdev);

zram_reset_device(zram, true); + zram_del_dev(); return
len;

out: @@ -954,6 +964,41 @@ static void destroy_device(struct zram *zram) blk_cleanup_queue(zram->queue); }

(quoted diff) +2 -0
+/* remove last free device, if last and prelast dev a free */ 
+static void zram_del_dev(void) +{ +    if (last_created_dev <
Quoted 3 lines

num_devices) + return; + + if
(zram_devices[last_created_dev].disksize == 0 && +
zram_devices[last_created_dev-1].disksize == 0 + ) {

racy?

Quoted 7 lines

+ destroy_device(&zram_devices[last_created_dev]); +
last_created_dev--; + pr_info("Deleted zram%u device\n",
last_created_dev); + } +} + +/* Auto add empty zram device,
if last device in use */ +static void zram_add_dev(void) +{ + if (last_created_dev+1 > max_num_devices) { + pr_warn("Can't add zram%u, max device number reached\n", +
num_devices); + return; + }

racy?

Quoted 6 lines

+ if (&zram_devices[last_created_dev].disksize > 0) { +
last_created_dev++; + if
(create_device(&zram_devices[last_created_dev], last_created_dev)) { + destroy_device(&zram_devices[last_created_dev]); + last_created_dev--;

racy?

-ss

Quoted 33 lines

+ return; + } + pr_info("Created
zram%u device\n", last_created_dev); + } +} + static int
__init zram_init(void) { int ret, dev_id; @@ -972,18 +1017,20 @@ static int __init zram_init(void) goto out; }

- /* Allocate the device array and initialize each one */ -
zram_devices = kzalloc(num_devices * sizeof(struct zram),
GFP_KERNEL); + /* Allocate the device array */ +
zram_devices = kcalloc(max_num_devices, sizeof(struct zram), GFP_KERNEL); if (!zram_devices) { ret = -ENOMEM; goto unregister; }

+ /* Initialise zram{0..num_devices} */ for (dev_id = 0;
dev_id < num_devices; dev_id++) { ret = create_device(&zram_devices[dev_id], dev_id); if (ret) goto
free_devices; } + last_created_dev = num_devices-1;

pr_info("Created %u device(s) ...\n", num_devices);

(quoted diff) +0 -1
@@ -1004,7 +1051,7 @@ static void __exit zram_exit(void) int
 i; struct zram *zram;
 
-    for (i = 0; i < num_devices; i++) { +    for (i = 0; i <
 last_created_dev+1; i++) { zram = &zram_devices[i];

destroy_device(zram); @@ -1025,7 +1072,7 @@
module_init(zram_init); module_exit(zram_exit);

module_param(num_devices, uint, 0);

(quoted diff) +1 -1
-MODULE_PARM_DESC(num_devices, "Number of zram devices"); 
+MODULE_PARM_DESC(num_devices, "Number of pre created  zram

devices");

Quoted 5 lines
Quoted 3 lines
Quoted 3 lines

MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Nitin Gupta <[email protected]>");

I can't understand what do you mean on "racy?"

Racy as in "race condition".

Quoted 7 lines

if you mind random empty lines in code, I've just checked code by checkpatch.pl and fixed warnings.

I'll try to explain, why I wrote a patch: zram, as a loop device, is a virtual device, not physical, and i think that adding new free devices automatically (like loop) is useful and expected behavior (for virtual device).

I don't know much about loop devices. How are they dynamically added/deleted? Is that the same mechanism as your patch implement? That may be a matter of taste, but I personally find that behaviour rather awkward.

Jerome

Quoted 10 lines

By default, zram creates only one device, and if I need to use two devices (for example), my actions lead to module’s reloading: 1. umount busy device 2. lose my data. Yes, I can backup it, and restore, but this is also useless, if I can do it more beautiful 3. unload module 4. load module again with new num devices parameter or make it parametre "permanent" by creating file in modprobe.d

This is handwork and such behavior is uncomfortable.

Quoted 7 lines

if number of devices is dynamically changeable, I can skip this actions and save my time and energy instead of wasting on these activities. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at
http://vger.kernel.org/majordomo-info.html Please read the FAQ at
http://www.tux.org/lkml/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJTx+s4AAoJEHTzHJCtsuoClfkIAKYfuhWQWSvSD9XPQUD0R7HI kW6865EWpNZjWR5WVYng48dvYwPV+2YGKJE1a+FyIwa06jjDO+Gm/FnOcnUhUcDz NFjTSBa1CIbh1hXU5uiI6u3BNw66BDjaprVe6x1tcW96y7T7Po+IWdf5+N7GMCtf 9H14KOCd20X5FsvPz70UkkpkLQZK1UdzmQ+1KYPb+bC3ACgTsuxs9Wxs1Tzn+tP/ DDvbZRd4FAcGZsC88U2gyuNAhoVYXVqzokukeV7hRlsSCFibwEOQzoZl/xljyICK Xv+uuMupmAVWv9CrlYexqT/HFoDOhAs7uRM2sgza39apUdQfkGg0vIw+9GV7+Q0= =Bxhk
-----END PGP SIGNATURE-----

author
#4 lore raw
Quoted 179 lines · Jerome Marchand

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 16 lines · Timofey Titovets
Quoted 13 lines · Sergey Senozhatsky
Quoted 12 lines · Timofey Titovets

This add supporting of autochange count of zram devices on demand, like loop devices; This working by following rules: - Always save minimum devices count specified by num_device (can be specified while kernel module loading) - if last device already using add new device; - if last and prelast devices is free - delete last zram device;

Signed-off-by Timofey Titovets <[email protected]>

From v1 -> v2: Delete useless variable 'ret', added
documentation for explain new zram behaviour

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

Please pull from: https://github.com/Nefelim4ag/linux.git --- Tested on 3.15.5-2-ARCH, can be applied on any kernel version after this patch 'zram: add LZ4 compression support' -
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=6e76668e415adf799839f0ab205142ad7002d260

- ---

Quoted 135 lines
Quoted 135 lines
Quoted 11 lines
(quoted diff) +0 -0

b/Documentation/blockdev/zram.txt index 0595c3f..7f5c921 100644 --- a/Documentation/blockdev/zram.txt +++ b/Documentation/blockdev/zram.txt @@ -18,9 +18,19 @@ Following shows a typical sequence of steps for using zram.

1) Load Module: modprobe zram num_devices=4 - This creates 4
devices: /dev/zram{0,1,2,3} + This pre creates 4 devices:
/dev/zram{0,1,2,3} (num_devices parameter is optional. Default:
1)

Quoted 7 lines

+ Kernel dynamically changes number of zram devices by
demand + (between num_devices and 31) + If set disk
size(4) for last device + kernel automatically adds new zram
device + If last two devices have zero disk size after
reset(8), + kernel will destroy last device + + Summing
up all features, comes one simple rule: + "The last zram
device is always free for use"

I can't understand what is the real benefit?

Quoted 51 lines

2) Set max number of compression streams Compression backend may use up to max_comp_streams compression streams, thus allowing up to max_comp_streams concurrent compression operations. diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 089e72c..9b2fc89 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -42,6 +42,10 @@ static const char *default_compressor = "lzo";

/* Module params (documentation at end) */ static unsigned int num_devices = 1; +static unsigned int last_created_dev = 1; + +static void zram_add_dev(void); +static void zram_del_dev(void); #define ZRAM_ATTR_RO(name)
\ static ssize_t zram_attr_##name##_show(struct device *d, \ @@ -168,6 +172,7 @@ static ssize_t comp_algorithm_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t len) { struct zram *zram = dev_to_zram(dev); + down_write(&zram->init_lock); if (init_done(zram)) { up_write(&zram->init_lock); @@ -239,6 +244,7 @@ static struct zram_meta *zram_meta_alloc(u64 disksize) { size_t num_pages; struct zram_meta *meta = kmalloc(sizeof(*meta), GFP_KERNEL); + if (!meta) goto out;

(quoted diff) +0 -0
@@ -374,6 +380,7 @@ static int zram_bvec_read(struct zram
 *zram, struct bio_vec *bvec, struct page *page; unsigned char
 *user_mem, *uncmem = NULL; struct zram_meta *meta =
 zram->meta; + page = bvec->bv_page;

read_lock(&meta->tb_lock); @@ -607,6 +614,7 @@ static void zram_reset_device(struct zram *zram, bool reset_capacity) /* Free all pages that are still in this zram device */ for (index = 0; index < zram->disksize >> PAGE_SHIFT; index++) { unsigned long handle = meta->table[index].handle; + if (!handle) continue;

(quoted diff) +0 -0
@@ -668,6 +676,7 @@ static ssize_t disksize_store(struct device
 *dev, set_capacity(zram->disk, zram->disksize >>
 SECTOR_SHIFT); revalidate_disk(zram->disk);
 up_write(&zram->init_lock); +    zram_add_dev(); return len;

out_destroy_comp: @@ -712,6 +721,7 @@ static ssize_t reset_store(struct device *dev, bdput(bdev);

zram_reset_device(zram, true); + zram_del_dev(); return
len;

out: @@ -954,6 +964,41 @@ static void destroy_device(struct zram *zram) blk_cleanup_queue(zram->queue); }

(quoted diff) +2 -0
+/* remove last free device, if last and prelast dev a free */
+static void zram_del_dev(void) +{ +    if (last_created_dev <
Quoted 3 lines

num_devices) + return; + + if
(zram_devices[last_created_dev].disksize == 0 && +
zram_devices[last_created_dev-1].disksize == 0 + ) {

racy?

Quoted 7 lines

+ destroy_device(&zram_devices[last_created_dev]); +
last_created_dev--; + pr_info("Deleted zram%u device\n",
last_created_dev); + } +} + +/* Auto add empty zram device,
if last device in use */ +static void zram_add_dev(void) +{ + if (last_created_dev+1 > max_num_devices) { + pr_warn("Can't add zram%u, max device number reached\n", +
num_devices); + return; + }

racy?

Quoted 6 lines

+ if (&zram_devices[last_created_dev].disksize > 0) { +
last_created_dev++; + if
(create_device(&zram_devices[last_created_dev], last_created_dev)) { + destroy_device(&zram_devices[last_created_dev]); + last_created_dev--;

racy?

-ss

Quoted 33 lines

+ return; + } + pr_info("Created
zram%u device\n", last_created_dev); + } +} + static int
__init zram_init(void) { int ret, dev_id; @@ -972,18 +1017,20 @@ static int __init zram_init(void) goto out; }

- /* Allocate the device array and initialize each one */ -
zram_devices = kzalloc(num_devices * sizeof(struct zram),
GFP_KERNEL); + /* Allocate the device array */ +
zram_devices = kcalloc(max_num_devices, sizeof(struct zram), GFP_KERNEL); if (!zram_devices) { ret = -ENOMEM; goto unregister; }

+ /* Initialise zram{0..num_devices} */ for (dev_id = 0;
dev_id < num_devices; dev_id++) { ret = create_device(&zram_devices[dev_id], dev_id); if (ret) goto
free_devices; } + last_created_dev = num_devices-1;

pr_info("Created %u device(s) ...\n", num_devices);

(quoted diff) +0 -1
@@ -1004,7 +1051,7 @@ static void __exit zram_exit(void) int
 i; struct zram *zram;
 
-    for (i = 0; i < num_devices; i++) { +    for (i = 0; i <
 last_created_dev+1; i++) { zram = &zram_devices[i];

destroy_device(zram); @@ -1025,7 +1072,7 @@
module_init(zram_init); module_exit(zram_exit);

module_param(num_devices, uint, 0);

(quoted diff) +1 -1
-MODULE_PARM_DESC(num_devices, "Number of zram devices");
+MODULE_PARM_DESC(num_devices, "Number of pre created  zram

devices");

Quoted 5 lines
Quoted 3 lines
Quoted 3 lines

MODULE_LICENSE("Dual BSD/GPL"); MODULE_AUTHOR("Nitin Gupta <[email protected]>");

I can't understand what do you mean on "racy?"

Racy as in "race condition".

Quoted 7 lines

if you mind random empty lines in code, I've just checked code by checkpatch.pl and fixed warnings.

I'll try to explain, why I wrote a patch: zram, as a loop device, is a virtual device, not physical, and i think that adding new free devices automatically (like loop) is useful and expected behavior (for virtual device).

Quoted 7 lines

I don't know much about loop devices. How are they dynamically added/deleted? Is that the same mechanism as your patch implement? That may be a matter of taste, but I personally find that behaviour rather awkward.

Jerome

loop device has self logic, as i understand it's working through loop-control and losetup.
I will try to understand clearer how loop devices are working and rewrite my patch.
I think if in final i'll write clear, not buggy patch, it can be applied on kernel, but now my code is crap and i must rewrite it =).

Anyway thanks for reviewing.

Timofey

Quoted 31 lines
Quoted 10 lines

By default, zram creates only one device, and if I need to use two devices (for example), my actions lead to module’s reloading: 1. umount busy device 2. lose my data. Yes, I can backup it, and restore, but this is also useless, if I can do it more beautiful 3. unload module 4. load module again with new num devices parameter or make it parametre "permanent" by creating file in modprobe.d

This is handwork and such behavior is uncomfortable.

Quoted 7 lines

if number of devices is dynamically changeable, I can skip this actions and save my time and energy instead of wasting on these activities. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at
http://vger.kernel.org/majordomo-info.html Please read the FAQ at
http://www.tux.org/lkml/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQEcBAEBAgAGBQJTx+s4AAoJEHTzHJCtsuoClfkIAKYfuhWQWSvSD9XPQUD0R7HI kW6865EWpNZjWR5WVYng48dvYwPV+2YGKJE1a+FyIwa06jjDO+Gm/FnOcnUhUcDz NFjTSBa1CIbh1hXU5uiI6u3BNw66BDjaprVe6x1tcW96y7T7Po+IWdf5+N7GMCtf 9H14KOCd20X5FsvPz70UkkpkLQZK1UdzmQ+1KYPb+bC3ACgTsuxs9Wxs1Tzn+tP/ DDvbZRd4FAcGZsC88U2gyuNAhoVYXVqzokukeV7hRlsSCFibwEOQzoZl/xljyICK Xv+uuMupmAVWv9CrlYexqT/HFoDOhAs7uRM2sgza39apUdQfkGg0vIw+9GV7+Q0= =Bxhk
-----END PGP SIGNATURE-----

author
#7 lore raw
In reply to Timofey Titovets
Quoted 102 lines · Timofey Titovets
Quoted 100 lines · Jerome Marchand

Looks like it:

$ cat conctest.sh
#! /bin/sh

modprobe zram

while true; do

     for i in `seq 1 10`; do
    echo 10485760 > /sys/block/zram0/disksize&
    echo 1 > /sys/block/zram0/reset&
     done
done
$ sudo ./conctest.sh
[   51.535387] ------------[ cut here ]------------
[   51.536305] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.537908] kernfs: can not remove 'max_comp_streams', no directory
[   51.539089] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.550844] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
      3.16.0-rc5-memacct-next-20140715+ #146
[   51.552606] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.553691]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.555412]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320242
[   51.557811]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.559580] Call Trace:
[   51.560139]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.561161]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.562333]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.563445]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.564432]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.565627]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.566753]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.567870]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.569089]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.570197]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.571247]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.572201]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.573283]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.574143]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.574976]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.576213]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.577422] ---[ end trace cdb79035dad379f3 ]---
[   51.578401] ------------[ cut here ]------------
[   51.579354] WARNING: CPU: 0 PID: 840 at fs/kernfs/dir.c:1209
kernfs_remove_by_name_ns+0x95/0xa0()
[   51.580966] kernfs: can not remove 'comp_algorithm', no directory
[   51.582108] Modules linked in: zram lz4_compress xt_conntrack

cfg80211 rfkill ebtable_nat ebtable_broute bridge stp llc ebtable_filter ebtables ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw ppdev crct10dif_pclmul crc32_pclmul crc32c_intel joydev ghash_clmulni_intel microcode serio_raw pcspkr virtio_balloon virtio_console parport_pc pvpanic virtio_net parport floppy i2c_piix4 nfsd auth_rpcgss nfs_acl lockd sunrpc qxl drm_kms_helper ttm virtio_blk drm virtio_pci virtio_ring virtio ata_generic pata_acpi

[   51.597177] CPU: 0 PID: 840 Comm: conc_init.sh Tainted: G      D W
      3.16.0-rc5-memacct-next-20140715+ #146
[   51.598549] Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
[   51.599661]  0000000000000009 ffff880078343d20 ffffffff8172d344
ffff880078343d68
[   51.601519]  ffff880078343d58 ffffffff8108ad1d 0000000000000000
ffffffffa0320233
[   51.603362]  ffffffffa0321058 0000000000000002 ffff880078343f50
ffff880078343db8
[   51.604913] Call Trace:
[   51.605444]  [<ffffffff8172d344>] dump_stack+0x45/0x56
[   51.606537]  [<ffffffff8108ad1d>] warn_slowpath_common+0x7d/0xa0
[   51.607507]  [<ffffffff8108ad8c>] warn_slowpath_fmt+0x4c/0x50
[   51.608582]  [<ffffffff81218196>] ? evict+0x106/0x170
[   51.609423]  [<ffffffff81275695>] kernfs_remove_by_name_ns+0x95/0xa0
[   51.610436]  [<ffffffff81277ee9>] remove_files.isra.1+0x39/0x70
[   51.611408]  [<ffffffff812781f4>] sysfs_remove_group+0x44/0xa0
[   51.612431]  [<ffffffffa031e058>] destroy_device.isra.17+0x28/0x50
[zram]
[   51.613214]  [<ffffffffa031e5d0>] reset_store+0x100/0x140 [zram]
[   51.613928]  [<ffffffff8146c3e8>] dev_attr_store+0x18/0x30
[   51.614634]  [<ffffffff81276fd4>] sysfs_kf_write+0x44/0x60
[   51.615387]  [<ffffffff812768d7>] kernfs_fop_write+0xe7/0x170
[   51.616246]  [<ffffffff811fbcf7>] vfs_write+0xb7/0x1f0
[   51.619107]  [<ffffffff811fc8a9>] SyS_write+0x49/0xb0
[   51.620063]  [<ffffffff8112e724>] ? __audit_syscall_entry+0xb4/0x110
[   51.622071]  [<ffffffff817362a9>] system_call_fastpath+0x16/0x1b
[   51.623256] ---[ end trace cdb79035dad379f4 ]---
Quoted 6 lines
Quoted 4 lines

And lots more of similar messages.

Jerome

Thanks for catching, i try fix it and resend patch.

I check loop device logic, and loop device not destroy automatically, only creating on demand.

If i delete automatically destroy logic function from my patch, for working like loop
(also loop create new device only if all device in use) and without destroy function my patch pass your concurrency test. Because we can't catch case when device created and destroyed in same time.

It have meaning to rework patch, for save only auto add logic and make it working like loop (if all device in use -> add new)?

I also have until to setup zram devices (written on C++), like losetup for loop devices, and i think if patch for kernel will be adopted. I can port my util from C++ to C and try add to util-linux, for having generic solution to setup zram devices, instead of several tens of utilities for all distros).

I just think what all in linux can have generic, simple, fast and useful solution, and if i can make something more useful, i must do it.

Thanks,
Timofey.