[dm-devel] fix the name_to_dev_t mess v2

[dm-devel] fix the name_to_dev_t mess v2

[Christoph Hellwig]

Hi all,

this series tries to sort out accumulated mess around the name_to_dev_t
function.  This function is intended to allow looking up the dev_t of a
block device based on a name string before the root file systems is
mounted and thus the normal path based lookup is available.

Unfortunately a few years ago it managed to get exported and used in
non-init contexts, leading to the something looking like a path name
also beeing lookuped up by a different and potential dangerous
algorithm.

This series does a fair amount of refactoring and finally ends up with
the renamed and improved name_to_dev_t only beeing available for the
early init code again.

The series is against Jens' for-6.5/block tree but probably applies
against current mainline just fine as well.

A git tree is also available here:

    git://git.infradead.org/users/hch/block.git blk-init-cleanup

Gitweb:

    http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/blk-init-cleanup

Changes since v1:
 - really propagate the actual error in dm_get_device
 - improve the documentation in kernel-parameters.txt
 - spelling fixes

Diffstat:

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

fix the name_to_dev_t mess v2

[Christoph Hellwig]

Hi all,

this series tries to sort out accumulated mess around the name_to_dev_t
function.  This function is intended to allow looking up the dev_t of a
block device based on a name string before the root file systems is
mounted and thus the normal path based lookup is available.

Unfortunately a few years ago it managed to get exported and used in
non-init contexts, leading to the something looking like a path name
also beeing lookuped up by a different and potential dangerous
algorithm.

This series does a fair amount of refactoring and finally ends up with
the renamed and improved name_to_dev_t only beeing available for the
early init code again.

The series is against Jens' for-6.5/block tree but probably applies
against current mainline just fine as well.

A git tree is also available here:

    git://git.infradead.org/users/hch/block.git blk-init-cleanup

Gitweb:

    http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/blk-init-cleanup

Changes since v1:
 - really propagate the actual error in dm_get_device
 - improve the documentation in kernel-parameters.txt
 - spelling fixes

Diffstat:

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

fix the name_to_dev_t mess v2

[Christoph Hellwig]

Hi all,

this series tries to sort out accumulated mess around the name_to_dev_t
function.  This function is intended to allow looking up the dev_t of a
block device based on a name string before the root file systems is
mounted and thus the normal path based lookup is available.

Unfortunately a few years ago it managed to get exported and used in
non-init contexts, leading to the something looking like a path name
also beeing lookuped up by a different and potential dangerous
algorithm.

This series does a fair amount of refactoring and finally ends up with
the renamed and improved name_to_dev_t only beeing available for the
early init code again.

The series is against Jens' for-6.5/block tree but probably applies
against current mainline just fine as well.

A git tree is also available here:

    git://git.infradead.org/users/hch/block.git blk-init-cleanup

Gitweb:

    http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/blk-init-cleanup

Changes since v1:
 - really propagate the actual error in dm_get_device
 - improve the documentation in kernel-parameters.txt
 - spelling fixes

Diffstat:

[dm-devel] [PATCH 01/24] driver core: return bool from driver_probe_done

[Christoph Hellwig]

bool is the most sensible return value for a yes/no return.  Also
add __init as this funtion is only called from the early boot code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/base/dd.c             | 6 ++----
 include/linux/device/driver.h | 2 +-
 init/do_mounts.c              | 2 +-
 3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 9c09ca5c4ab68e..878aa7646b37e4 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -751,14 +751,12 @@ static int really_probe_debug(struct device *dev, struct device_driver *drv)
  *
  * Should somehow figure out how to use a semaphore, not an atomic variable...
  */
-int driver_probe_done(void)
+bool __init driver_probe_done(void)
 {
 	int local_probe_count = atomic_read(&probe_count);
 
 	pr_debug("%s: probe_count = %d\n", __func__, local_probe_count);
-	if (local_probe_count)
-		return -EBUSY;
-	return 0;
+	return !local_probe_count;
 }
 
 /**
diff --git a/include/linux/device/driver.h b/include/linux/device/driver.h
index c244267a67443e..7738f458995fba 100644
--- a/include/linux/device/driver.h
+++ b/include/linux/device/driver.h
@@ -126,7 +126,7 @@ int __must_check driver_register(struct device_driver *drv);
 void driver_unregister(struct device_driver *drv);
 
 struct device_driver *driver_find(const char *name, const struct bus_type *bus);
-int driver_probe_done(void);
+bool __init driver_probe_done(void);
 void wait_for_device_probe(void);
 void __init wait_for_init_devices_probe(void);
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 811e94daf0a84a..2fe7901b5bcfaf 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -635,7 +635,7 @@ void __init prepare_namespace(void)
 	if ((ROOT_DEV == 0) && root_wait) {
 		printk(KERN_INFO "Waiting for root device %s...\n",
 			saved_root_name);
-		while (driver_probe_done() != 0 ||
+		while (!driver_probe_done() ||
 			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
 			msleep(5);
 		async_synchronize_full();
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 01/24] driver core: return bool from driver_probe_done

[Christoph Hellwig]

bool is the most sensible return value for a yes/no return.  Also
add __init as this funtion is only called from the early boot code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/base/dd.c             | 6 ++----
 include/linux/device/driver.h | 2 +-
 init/do_mounts.c              | 2 +-
 3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 9c09ca5c4ab68e..878aa7646b37e4 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -751,14 +751,12 @@ static int really_probe_debug(struct device *dev, struct device_driver *drv)
  *
  * Should somehow figure out how to use a semaphore, not an atomic variable...
  */
-int driver_probe_done(void)
+bool __init driver_probe_done(void)
 {
 	int local_probe_count = atomic_read(&probe_count);
 
 	pr_debug("%s: probe_count = %d\n", __func__, local_probe_count);
-	if (local_probe_count)
-		return -EBUSY;
-	return 0;
+	return !local_probe_count;
 }
 
 /**
diff --git a/include/linux/device/driver.h b/include/linux/device/driver.h
index c244267a67443e..7738f458995fba 100644
--- a/include/linux/device/driver.h
+++ b/include/linux/device/driver.h
@@ -126,7 +126,7 @@ int __must_check driver_register(struct device_driver *drv);
 void driver_unregister(struct device_driver *drv);
 
 struct device_driver *driver_find(const char *name, const struct bus_type *bus);
-int driver_probe_done(void);
+bool __init driver_probe_done(void);
 void wait_for_device_probe(void);
 void __init wait_for_init_devices_probe(void);
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 811e94daf0a84a..2fe7901b5bcfaf 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -635,7 +635,7 @@ void __init prepare_namespace(void)
 	if ((ROOT_DEV == 0) && root_wait) {
 		printk(KERN_INFO "Waiting for root device %s...\n",
 			saved_root_name);
-		while (driver_probe_done() != 0 ||
+		while (!driver_probe_done() ||
 			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
 			msleep(5);
 		async_synchronize_full();
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 01/24] driver core: return bool from driver_probe_done

[Christoph Hellwig]

bool is the most sensible return value for a yes/no return.  Also
add __init as this funtion is only called from the early boot code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/base/dd.c             | 6 ++----
 include/linux/device/driver.h | 2 +-
 init/do_mounts.c              | 2 +-
 3 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 9c09ca5c4ab68e..878aa7646b37e4 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -751,14 +751,12 @@ static int really_probe_debug(struct device *dev, struct device_driver *drv)
  *
  * Should somehow figure out how to use a semaphore, not an atomic variable...
  */
-int driver_probe_done(void)
+bool __init driver_probe_done(void)
 {
 	int local_probe_count = atomic_read(&probe_count);
 
 	pr_debug("%s: probe_count = %d\n", __func__, local_probe_count);
-	if (local_probe_count)
-		return -EBUSY;
-	return 0;
+	return !local_probe_count;
 }
 
 /**
diff --git a/include/linux/device/driver.h b/include/linux/device/driver.h
index c244267a67443e..7738f458995fba 100644
--- a/include/linux/device/driver.h
+++ b/include/linux/device/driver.h
@@ -126,7 +126,7 @@ int __must_check driver_register(struct device_driver *drv);
 void driver_unregister(struct device_driver *drv);
 
 struct device_driver *driver_find(const char *name, const struct bus_type *bus);
-int driver_probe_done(void);
+bool __init driver_probe_done(void);
 void wait_for_device_probe(void);
 void __init wait_for_init_devices_probe(void);
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 811e94daf0a84a..2fe7901b5bcfaf 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -635,7 +635,7 @@ void __init prepare_namespace(void)
 	if ((ROOT_DEV == 0) && root_wait) {
 		printk(KERN_INFO "Waiting for root device %s...\n",
 			saved_root_name);
-		while (driver_probe_done() != 0 ||
+		while (!driver_probe_done() ||
 			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
 			msleep(5);
 		async_synchronize_full();
-- 
2.39.2

Re: [dm-devel] [PATCH 01/24] driver core: return bool from driver_probe_done

[Jens Axboe]

On Wed, 31 May 2023 14:55:12 +0200, Christoph Hellwig wrote:
> bool is the most sensible return value for a yes/no return.  Also
> add __init as this funtion is only called from the early boot code.
> 
> 

Applied, thanks!

[01/24] driver core: return bool from driver_probe_done
        commit: aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
[02/24] PM: hibernate: factor out a helper to find the resume device
        commit: 02b42d58f3898134b900ff3030561099e38adb32
[03/24] PM: hibernate: remove the global snapshot_test variable
        commit: d6545e687271ab27472eebff770f2de6a5f1a464
[04/24] PM: hibernate: move finding the resume device out of software_resume
        commit: cc89c63e2fe37d476357c82390dfb12edcd41cdd
[05/24] init: remove pointless Root_* values
        commit: f5524c3fadba35c075a5131bad74e3041507a694
[06/24] init: rename mount_block_root to mount_root_generic
        commit: e3102722ffe77094ba9e7e46380792b3dd8a7abd
[07/24] init: refactor mount_root
        commit: a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
[08/24] init: pass root_device_name explicitly
        commit: c8643c72bc42781fc169c6498a3902bec447099e
[09/24] init: don't remove the /dev/ prefix from error messages
        commit: 73231b58b1b496d631fa0ecf9fa7f64f5a07c6e3
[10/24] init: handle ubi/mtd root mounting like all other root types
        commit: 07d63cbb67cdb5e2a7720fdd8579b3be979c2d66
[11/24] init: factor the root_wait logic in prepare_namespace into a helper
        commit: 3701c600a3e735b9fbac6f7a73e4c086090c97ca
[12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t
        commit: c0c1a7dcb6f5db4500e6574294674213bc24940c
[13/24] init: improve the name_to_dev_t interface
        commit: cf056a43121559d3642419917d405c3237ded90a
[14/24] init: clear root_wait on all invalid root= strings
        commit: 079caa35f7863cd9958b4555ae873ea4d352a502
[15/24] block: move the code to do early boot lookup of block devices to block/
        commit: 702f3189e454b3c3c2f3c99dbf30acf41aab707c
[16/24] block: move more code to early-lookup.c
        commit: 7cadcaf1d82618852745e7206fffa2c72c17ce4b
[17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr
        commit: 26110d5afe8117d1b505fe735ac709bdf063f4da
[18/24] dm: open code dm_get_dev_t in dm_init_init
        commit: 49177377e910a8fd5cd1388c966d8fbb51075c3c
[19/24] dm: remove dm_get_dev_t
        commit: d4a28d7defe79006e59293a4b43d518ba8483fb0
[20/24] dm: only call early_lookup_bdev from early boot context
        commit: 7a126d5bf975f082281fb9b45d110cd49b7c3ee4
[21/24] PM: hibernate: don't use early_lookup_bdev in resume_store
        commit: 1e8c813b083c4122dfeaa5c3b11028331026e85d
[22/24] mtd: block2mtd: factor the early block device open logic into a helper
        commit: b2baa57475e3a24bb9ad27bb9047ea3be94627f5
[23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running
        commit: 8d03187ee7328af8e18ef1782289e0b034e75485
[24/24] block: mark early_lookup_bdev as __init
        commit: 2577f53f42947d8ca01666e3444bb7307319ea38

Best regards,
-- 
Jens Axboe



--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 01/24] driver core: return bool from driver_probe_done

[Jens Axboe]

On Wed, 31 May 2023 14:55:12 +0200, Christoph Hellwig wrote:
> bool is the most sensible return value for a yes/no return.  Also
> add __init as this funtion is only called from the early boot code.
> 
> 

Applied, thanks!

[01/24] driver core: return bool from driver_probe_done
        commit: aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
[02/24] PM: hibernate: factor out a helper to find the resume device
        commit: 02b42d58f3898134b900ff3030561099e38adb32
[03/24] PM: hibernate: remove the global snapshot_test variable
        commit: d6545e687271ab27472eebff770f2de6a5f1a464
[04/24] PM: hibernate: move finding the resume device out of software_resume
        commit: cc89c63e2fe37d476357c82390dfb12edcd41cdd
[05/24] init: remove pointless Root_* values
        commit: f5524c3fadba35c075a5131bad74e3041507a694
[06/24] init: rename mount_block_root to mount_root_generic
        commit: e3102722ffe77094ba9e7e46380792b3dd8a7abd
[07/24] init: refactor mount_root
        commit: a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
[08/24] init: pass root_device_name explicitly
        commit: c8643c72bc42781fc169c6498a3902bec447099e
[09/24] init: don't remove the /dev/ prefix from error messages
        commit: 73231b58b1b496d631fa0ecf9fa7f64f5a07c6e3
[10/24] init: handle ubi/mtd root mounting like all other root types
        commit: 07d63cbb67cdb5e2a7720fdd8579b3be979c2d66
[11/24] init: factor the root_wait logic in prepare_namespace into a helper
        commit: 3701c600a3e735b9fbac6f7a73e4c086090c97ca
[12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t
        commit: c0c1a7dcb6f5db4500e6574294674213bc24940c
[13/24] init: improve the name_to_dev_t interface
        commit: cf056a43121559d3642419917d405c3237ded90a
[14/24] init: clear root_wait on all invalid root= strings
        commit: 079caa35f7863cd9958b4555ae873ea4d352a502
[15/24] block: move the code to do early boot lookup of block devices to block/
        commit: 702f3189e454b3c3c2f3c99dbf30acf41aab707c
[16/24] block: move more code to early-lookup.c
        commit: 7cadcaf1d82618852745e7206fffa2c72c17ce4b
[17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr
        commit: 26110d5afe8117d1b505fe735ac709bdf063f4da
[18/24] dm: open code dm_get_dev_t in dm_init_init
        commit: 49177377e910a8fd5cd1388c966d8fbb51075c3c
[19/24] dm: remove dm_get_dev_t
        commit: d4a28d7defe79006e59293a4b43d518ba8483fb0
[20/24] dm: only call early_lookup_bdev from early boot context
        commit: 7a126d5bf975f082281fb9b45d110cd49b7c3ee4
[21/24] PM: hibernate: don't use early_lookup_bdev in resume_store
        commit: 1e8c813b083c4122dfeaa5c3b11028331026e85d
[22/24] mtd: block2mtd: factor the early block device open logic into a helper
        commit: b2baa57475e3a24bb9ad27bb9047ea3be94627f5
[23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running
        commit: 8d03187ee7328af8e18ef1782289e0b034e75485
[24/24] block: mark early_lookup_bdev as __init
        commit: 2577f53f42947d8ca01666e3444bb7307319ea38

Best regards,
-- 
Jens Axboe




______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 01/24] driver core: return bool from driver_probe_done

[Jens Axboe]

On Wed, 31 May 2023 14:55:12 +0200, Christoph Hellwig wrote:
> bool is the most sensible return value for a yes/no return.  Also
> add __init as this funtion is only called from the early boot code.
> 
> 

Applied, thanks!

[01/24] driver core: return bool from driver_probe_done
        commit: aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
[02/24] PM: hibernate: factor out a helper to find the resume device
        commit: 02b42d58f3898134b900ff3030561099e38adb32
[03/24] PM: hibernate: remove the global snapshot_test variable
        commit: d6545e687271ab27472eebff770f2de6a5f1a464
[04/24] PM: hibernate: move finding the resume device out of software_resume
        commit: cc89c63e2fe37d476357c82390dfb12edcd41cdd
[05/24] init: remove pointless Root_* values
        commit: f5524c3fadba35c075a5131bad74e3041507a694
[06/24] init: rename mount_block_root to mount_root_generic
        commit: e3102722ffe77094ba9e7e46380792b3dd8a7abd
[07/24] init: refactor mount_root
        commit: a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
[08/24] init: pass root_device_name explicitly
        commit: c8643c72bc42781fc169c6498a3902bec447099e
[09/24] init: don't remove the /dev/ prefix from error messages
        commit: 73231b58b1b496d631fa0ecf9fa7f64f5a07c6e3
[10/24] init: handle ubi/mtd root mounting like all other root types
        commit: 07d63cbb67cdb5e2a7720fdd8579b3be979c2d66
[11/24] init: factor the root_wait logic in prepare_namespace into a helper
        commit: 3701c600a3e735b9fbac6f7a73e4c086090c97ca
[12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t
        commit: c0c1a7dcb6f5db4500e6574294674213bc24940c
[13/24] init: improve the name_to_dev_t interface
        commit: cf056a43121559d3642419917d405c3237ded90a
[14/24] init: clear root_wait on all invalid root= strings
        commit: 079caa35f7863cd9958b4555ae873ea4d352a502
[15/24] block: move the code to do early boot lookup of block devices to block/
        commit: 702f3189e454b3c3c2f3c99dbf30acf41aab707c
[16/24] block: move more code to early-lookup.c
        commit: 7cadcaf1d82618852745e7206fffa2c72c17ce4b
[17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr
        commit: 26110d5afe8117d1b505fe735ac709bdf063f4da
[18/24] dm: open code dm_get_dev_t in dm_init_init
        commit: 49177377e910a8fd5cd1388c966d8fbb51075c3c
[19/24] dm: remove dm_get_dev_t
        commit: d4a28d7defe79006e59293a4b43d518ba8483fb0
[20/24] dm: only call early_lookup_bdev from early boot context
        commit: 7a126d5bf975f082281fb9b45d110cd49b7c3ee4
[21/24] PM: hibernate: don't use early_lookup_bdev in resume_store
        commit: 1e8c813b083c4122dfeaa5c3b11028331026e85d
[22/24] mtd: block2mtd: factor the early block device open logic into a helper
        commit: b2baa57475e3a24bb9ad27bb9047ea3be94627f5
[23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running
        commit: 8d03187ee7328af8e18ef1782289e0b034e75485
[24/24] block: mark early_lookup_bdev as __init
        commit: 2577f53f42947d8ca01666e3444bb7307319ea38

Best regards,
-- 
Jens Axboe

[dm-devel] [PATCH 02/24] PM: hibernate: factor out a helper to find the resume device

[Christoph Hellwig]

Split the logic to find the resume device out software_resume and into
a separate helper to start unwindig the convoluted goto logic.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 72 +++++++++++++++++++++-------------------
 1 file changed, 37 insertions(+), 35 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 30d1274f03f625..07279506366255 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -910,6 +910,41 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
+static int find_resume_device(void)
+{
+	if (!strlen(resume_file))
+		return -ENOENT;
+
+	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
+
+	if (resume_delay) {
+		pr_info("Waiting %dsec before reading resume device ...\n",
+			resume_delay);
+		ssleep(resume_delay);
+	}
+
+	/* Check if the device is there */
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (swsusp_resume_device)
+		return 0;
+
+	/*
+	 * Some device discovery might still be in progress; we need to wait for
+	 * this to finish.
+	 */
+	wait_for_device_probe();
+	if (resume_wait) {
+		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+			msleep(10);
+		async_synchronize_full();
+	}
+
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (!swsusp_resume_device)
+		return -ENODEV;
+	return 0;
+}
+
 /**
  * software_resume - Resume from a saved hibernation image.
  *
@@ -949,45 +984,12 @@ static int software_resume(void)
 
 	snapshot_test = false;
 
-	if (swsusp_resume_device)
-		goto Check_image;
-
-	if (!strlen(resume_file)) {
-		error = -ENOENT;
-		goto Unlock;
-	}
-
-	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
-
-	if (resume_delay) {
-		pr_info("Waiting %dsec before reading resume device ...\n",
-			resume_delay);
-		ssleep(resume_delay);
-	}
-
-	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
 	if (!swsusp_resume_device) {
-		/*
-		 * Some device discovery might still be in progress; we need
-		 * to wait for this to finish.
-		 */
-		wait_for_device_probe();
-
-		if (resume_wait) {
-			while ((swsusp_resume_device = name_to_dev_t(resume_file)) == 0)
-				msleep(10);
-			async_synchronize_full();
-		}
-
-		swsusp_resume_device = name_to_dev_t(resume_file);
-		if (!swsusp_resume_device) {
-			error = -ENODEV;
+		error = find_resume_device();
+		if (error)
 			goto Unlock;
-		}
 	}
 
- Check_image:
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 02/24] PM: hibernate: factor out a helper to find the resume device

[Christoph Hellwig]

Split the logic to find the resume device out software_resume and into
a separate helper to start unwindig the convoluted goto logic.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 72 +++++++++++++++++++++-------------------
 1 file changed, 37 insertions(+), 35 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 30d1274f03f625..07279506366255 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -910,6 +910,41 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
+static int find_resume_device(void)
+{
+	if (!strlen(resume_file))
+		return -ENOENT;
+
+	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
+
+	if (resume_delay) {
+		pr_info("Waiting %dsec before reading resume device ...\n",
+			resume_delay);
+		ssleep(resume_delay);
+	}
+
+	/* Check if the device is there */
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (swsusp_resume_device)
+		return 0;
+
+	/*
+	 * Some device discovery might still be in progress; we need to wait for
+	 * this to finish.
+	 */
+	wait_for_device_probe();
+	if (resume_wait) {
+		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+			msleep(10);
+		async_synchronize_full();
+	}
+
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (!swsusp_resume_device)
+		return -ENODEV;
+	return 0;
+}
+
 /**
  * software_resume - Resume from a saved hibernation image.
  *
@@ -949,45 +984,12 @@ static int software_resume(void)
 
 	snapshot_test = false;
 
-	if (swsusp_resume_device)
-		goto Check_image;
-
-	if (!strlen(resume_file)) {
-		error = -ENOENT;
-		goto Unlock;
-	}
-
-	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
-
-	if (resume_delay) {
-		pr_info("Waiting %dsec before reading resume device ...\n",
-			resume_delay);
-		ssleep(resume_delay);
-	}
-
-	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
 	if (!swsusp_resume_device) {
-		/*
-		 * Some device discovery might still be in progress; we need
-		 * to wait for this to finish.
-		 */
-		wait_for_device_probe();
-
-		if (resume_wait) {
-			while ((swsusp_resume_device = name_to_dev_t(resume_file)) == 0)
-				msleep(10);
-			async_synchronize_full();
-		}
-
-		swsusp_resume_device = name_to_dev_t(resume_file);
-		if (!swsusp_resume_device) {
-			error = -ENODEV;
+		error = find_resume_device();
+		if (error)
 			goto Unlock;
-		}
 	}
 
- Check_image:
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 02/24] PM: hibernate: factor out a helper to find the resume device

[Christoph Hellwig]

Split the logic to find the resume device out software_resume and into
a separate helper to start unwindig the convoluted goto logic.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 72 +++++++++++++++++++++-------------------
 1 file changed, 37 insertions(+), 35 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 30d1274f03f625..07279506366255 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -910,6 +910,41 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
+static int find_resume_device(void)
+{
+	if (!strlen(resume_file))
+		return -ENOENT;
+
+	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
+
+	if (resume_delay) {
+		pr_info("Waiting %dsec before reading resume device ...\n",
+			resume_delay);
+		ssleep(resume_delay);
+	}
+
+	/* Check if the device is there */
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (swsusp_resume_device)
+		return 0;
+
+	/*
+	 * Some device discovery might still be in progress; we need to wait for
+	 * this to finish.
+	 */
+	wait_for_device_probe();
+	if (resume_wait) {
+		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+			msleep(10);
+		async_synchronize_full();
+	}
+
+	swsusp_resume_device = name_to_dev_t(resume_file);
+	if (!swsusp_resume_device)
+		return -ENODEV;
+	return 0;
+}
+
 /**
  * software_resume - Resume from a saved hibernation image.
  *
@@ -949,45 +984,12 @@ static int software_resume(void)
 
 	snapshot_test = false;
 
-	if (swsusp_resume_device)
-		goto Check_image;
-
-	if (!strlen(resume_file)) {
-		error = -ENOENT;
-		goto Unlock;
-	}
-
-	pm_pr_dbg("Checking hibernation image partition %s\n", resume_file);
-
-	if (resume_delay) {
-		pr_info("Waiting %dsec before reading resume device ...\n",
-			resume_delay);
-		ssleep(resume_delay);
-	}
-
-	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
 	if (!swsusp_resume_device) {
-		/*
-		 * Some device discovery might still be in progress; we need
-		 * to wait for this to finish.
-		 */
-		wait_for_device_probe();
-
-		if (resume_wait) {
-			while ((swsusp_resume_device = name_to_dev_t(resume_file)) == 0)
-				msleep(10);
-			async_synchronize_full();
-		}
-
-		swsusp_resume_device = name_to_dev_t(resume_file);
-		if (!swsusp_resume_device) {
-			error = -ENODEV;
+		error = find_resume_device();
+		if (error)
 			goto Unlock;
-		}
 	}
 
- Check_image:
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
-- 
2.39.2

[dm-devel] [PATCH 03/24] PM: hibernate: remove the global snapshot_test variable

[Christoph Hellwig]

Passing call dependent variable in global variables is a huge
antipattern.  Fix it up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 17 ++++++-----------
 kernel/power/power.h     |  3 +--
 kernel/power/swap.c      |  2 +-
 3 files changed, 8 insertions(+), 14 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 07279506366255..78696aa04f5ca3 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -64,7 +64,6 @@ enum {
 static int hibernation_mode = HIBERNATION_SHUTDOWN;
 
 bool freezer_test_done;
-bool snapshot_test;
 
 static const struct platform_hibernation_ops *hibernation_ops;
 
@@ -684,7 +683,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(void)
+static int load_image_and_restore(bool snapshot_test)
 {
 	int error;
 	unsigned int flags;
@@ -721,6 +720,7 @@ static int load_image_and_restore(void)
  */
 int hibernate(void)
 {
+	bool snapshot_test = false;
 	unsigned int sleep_flags;
 	int error;
 
@@ -748,9 +748,6 @@ int hibernate(void)
 	if (error)
 		goto Exit;
 
-	/* protected by system_transition_mutex */
-	snapshot_test = false;
-
 	lock_device_hotplug();
 	/* Allocate memory management structures */
 	error = create_basic_memory_bitmaps();
@@ -792,9 +789,9 @@ int hibernate(void)
 	unlock_device_hotplug();
 	if (snapshot_test) {
 		pm_pr_dbg("Checking hibernation image\n");
-		error = swsusp_check();
+		error = swsusp_check(snapshot_test);
 		if (!error)
-			error = load_image_and_restore();
+			error = load_image_and_restore(snapshot_test);
 	}
 	thaw_processes();
 
@@ -982,8 +979,6 @@ static int software_resume(void)
 	 */
 	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
 
-	snapshot_test = false;
-
 	if (!swsusp_resume_device) {
 		error = find_resume_device();
 		if (error)
@@ -994,7 +989,7 @@ static int software_resume(void)
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
-	error = swsusp_check();
+	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
 
@@ -1022,7 +1017,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore();
+	error = load_image_and_restore(false);
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index b83c8d5e188dec..978189fcafd124 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -59,7 +59,6 @@ asmlinkage int swsusp_save(void);
 
 /* kernel/power/hibernate.c */
 extern bool freezer_test_done;
-extern bool snapshot_test;
 
 extern int hibernation_snapshot(int platform_mode);
 extern int hibernation_restore(int platform_mode);
@@ -174,7 +173,7 @@ extern int swsusp_swap_in_use(void);
 #define SF_HW_SIG		8
 
 /* kernel/power/hibernate.c */
-extern int swsusp_check(void);
+int swsusp_check(bool snapshot_test);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 92e41ed292ada8..efed11568bfc72 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -1514,7 +1514,7 @@ int swsusp_read(unsigned int *flags_p)
  *      swsusp_check - Check for swsusp signature in the resume device
  */
 
-int swsusp_check(void)
+int swsusp_check(bool snapshot_test)
 {
 	int error;
 	void *holder;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 03/24] PM: hibernate: remove the global snapshot_test variable

[Christoph Hellwig]

Passing call dependent variable in global variables is a huge
antipattern.  Fix it up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 17 ++++++-----------
 kernel/power/power.h     |  3 +--
 kernel/power/swap.c      |  2 +-
 3 files changed, 8 insertions(+), 14 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 07279506366255..78696aa04f5ca3 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -64,7 +64,6 @@ enum {
 static int hibernation_mode = HIBERNATION_SHUTDOWN;
 
 bool freezer_test_done;
-bool snapshot_test;
 
 static const struct platform_hibernation_ops *hibernation_ops;
 
@@ -684,7 +683,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(void)
+static int load_image_and_restore(bool snapshot_test)
 {
 	int error;
 	unsigned int flags;
@@ -721,6 +720,7 @@ static int load_image_and_restore(void)
  */
 int hibernate(void)
 {
+	bool snapshot_test = false;
 	unsigned int sleep_flags;
 	int error;
 
@@ -748,9 +748,6 @@ int hibernate(void)
 	if (error)
 		goto Exit;
 
-	/* protected by system_transition_mutex */
-	snapshot_test = false;
-
 	lock_device_hotplug();
 	/* Allocate memory management structures */
 	error = create_basic_memory_bitmaps();
@@ -792,9 +789,9 @@ int hibernate(void)
 	unlock_device_hotplug();
 	if (snapshot_test) {
 		pm_pr_dbg("Checking hibernation image\n");
-		error = swsusp_check();
+		error = swsusp_check(snapshot_test);
 		if (!error)
-			error = load_image_and_restore();
+			error = load_image_and_restore(snapshot_test);
 	}
 	thaw_processes();
 
@@ -982,8 +979,6 @@ static int software_resume(void)
 	 */
 	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
 
-	snapshot_test = false;
-
 	if (!swsusp_resume_device) {
 		error = find_resume_device();
 		if (error)
@@ -994,7 +989,7 @@ static int software_resume(void)
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
-	error = swsusp_check();
+	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
 
@@ -1022,7 +1017,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore();
+	error = load_image_and_restore(false);
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index b83c8d5e188dec..978189fcafd124 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -59,7 +59,6 @@ asmlinkage int swsusp_save(void);
 
 /* kernel/power/hibernate.c */
 extern bool freezer_test_done;
-extern bool snapshot_test;
 
 extern int hibernation_snapshot(int platform_mode);
 extern int hibernation_restore(int platform_mode);
@@ -174,7 +173,7 @@ extern int swsusp_swap_in_use(void);
 #define SF_HW_SIG		8
 
 /* kernel/power/hibernate.c */
-extern int swsusp_check(void);
+int swsusp_check(bool snapshot_test);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 92e41ed292ada8..efed11568bfc72 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -1514,7 +1514,7 @@ int swsusp_read(unsigned int *flags_p)
  *      swsusp_check - Check for swsusp signature in the resume device
  */
 
-int swsusp_check(void)
+int swsusp_check(bool snapshot_test)
 {
 	int error;
 	void *holder;
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 03/24] PM: hibernate: remove the global snapshot_test variable

[Christoph Hellwig]

Passing call dependent variable in global variables is a huge
antipattern.  Fix it up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 17 ++++++-----------
 kernel/power/power.h     |  3 +--
 kernel/power/swap.c      |  2 +-
 3 files changed, 8 insertions(+), 14 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 07279506366255..78696aa04f5ca3 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -64,7 +64,6 @@ enum {
 static int hibernation_mode = HIBERNATION_SHUTDOWN;
 
 bool freezer_test_done;
-bool snapshot_test;
 
 static const struct platform_hibernation_ops *hibernation_ops;
 
@@ -684,7 +683,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(void)
+static int load_image_and_restore(bool snapshot_test)
 {
 	int error;
 	unsigned int flags;
@@ -721,6 +720,7 @@ static int load_image_and_restore(void)
  */
 int hibernate(void)
 {
+	bool snapshot_test = false;
 	unsigned int sleep_flags;
 	int error;
 
@@ -748,9 +748,6 @@ int hibernate(void)
 	if (error)
 		goto Exit;
 
-	/* protected by system_transition_mutex */
-	snapshot_test = false;
-
 	lock_device_hotplug();
 	/* Allocate memory management structures */
 	error = create_basic_memory_bitmaps();
@@ -792,9 +789,9 @@ int hibernate(void)
 	unlock_device_hotplug();
 	if (snapshot_test) {
 		pm_pr_dbg("Checking hibernation image\n");
-		error = swsusp_check();
+		error = swsusp_check(snapshot_test);
 		if (!error)
-			error = load_image_and_restore();
+			error = load_image_and_restore(snapshot_test);
 	}
 	thaw_processes();
 
@@ -982,8 +979,6 @@ static int software_resume(void)
 	 */
 	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
 
-	snapshot_test = false;
-
 	if (!swsusp_resume_device) {
 		error = find_resume_device();
 		if (error)
@@ -994,7 +989,7 @@ static int software_resume(void)
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
-	error = swsusp_check();
+	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
 
@@ -1022,7 +1017,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore();
+	error = load_image_and_restore(false);
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index b83c8d5e188dec..978189fcafd124 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -59,7 +59,6 @@ asmlinkage int swsusp_save(void);
 
 /* kernel/power/hibernate.c */
 extern bool freezer_test_done;
-extern bool snapshot_test;
 
 extern int hibernation_snapshot(int platform_mode);
 extern int hibernation_restore(int platform_mode);
@@ -174,7 +173,7 @@ extern int swsusp_swap_in_use(void);
 #define SF_HW_SIG		8
 
 /* kernel/power/hibernate.c */
-extern int swsusp_check(void);
+int swsusp_check(bool snapshot_test);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 92e41ed292ada8..efed11568bfc72 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -1514,7 +1514,7 @@ int swsusp_read(unsigned int *flags_p)
  *      swsusp_check - Check for swsusp signature in the resume device
  */
 
-int swsusp_check(void)
+int swsusp_check(bool snapshot_test)
 {
 	int error;
 	void *holder;
-- 
2.39.2

[dm-devel] [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

software_resume can be called either from an init call in the boot code,
or from sysfs once the system has finished booting, and the two
invocation methods this can't race with each other.

For the latter case we did just parse the suspend device manually, while
the former might not have one.  Split software_resume so that the search
only happens for the boot case, which also means the special lockdep
nesting annotation can go away as the system transition mutex can be
taken a little later and doesn't have the sysfs locking nest inside it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 80 ++++++++++++++++++++--------------------
 1 file changed, 39 insertions(+), 41 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 78696aa04f5ca3..45e24b02cd50b6 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -907,7 +907,7 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
-static int find_resume_device(void)
+static int __init find_resume_device(void)
 {
 	if (!strlen(resume_file))
 		return -ENOENT;
@@ -942,53 +942,16 @@ static int find_resume_device(void)
 	return 0;
 }
 
-/**
- * software_resume - Resume from a saved hibernation image.
- *
- * This routine is called as a late initcall, when all devices have been
- * discovered and initialized already.
- *
- * The image reading code is called to see if there is a hibernation image
- * available for reading.  If that is the case, devices are quiesced and the
- * contents of memory is restored from the saved image.
- *
- * If this is successful, control reappears in the restored target kernel in
- * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
- * attempts to recover gracefully and make the kernel return to the normal mode
- * of operation.
- */
 static int software_resume(void)
 {
 	int error;
 
-	/*
-	 * If the user said "noresume".. bail out early.
-	 */
-	if (noresume || !hibernation_available())
-		return 0;
-
-	/*
-	 * name_to_dev_t() below takes a sysfs buffer mutex when sysfs
-	 * is configured into the kernel. Since the regular hibernate
-	 * trigger path is via sysfs which takes a buffer mutex before
-	 * calling hibernate functions (which take system_transition_mutex)
-	 * this can cause lockdep to complain about a possible ABBA deadlock
-	 * which cannot happen since we're in the boot code here and
-	 * sysfs can't be invoked yet. Therefore, we use a subclass
-	 * here to avoid lockdep complaining.
-	 */
-	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
-
-	if (!swsusp_resume_device) {
-		error = find_resume_device();
-		if (error)
-			goto Unlock;
-	}
-
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
+
+	mutex_lock(&system_transition_mutex);
 	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
@@ -1035,7 +998,39 @@ static int software_resume(void)
 	goto Finish;
 }
 
-late_initcall_sync(software_resume);
+/**
+ * software_resume_initcall - Resume from a saved hibernation image.
+ *
+ * This routine is called as a late initcall, when all devices have been
+ * discovered and initialized already.
+ *
+ * The image reading code is called to see if there is a hibernation image
+ * available for reading.  If that is the case, devices are quiesced and the
+ * contents of memory is restored from the saved image.
+ *
+ * If this is successful, control reappears in the restored target kernel in
+ * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
+ * attempts to recover gracefully and make the kernel return to the normal mode
+ * of operation.
+ */
+static int __init software_resume_initcall(void)
+{
+	/*
+	 * If the user said "noresume".. bail out early.
+	 */
+	if (noresume || !hibernation_available())
+		return 0;
+
+	if (!swsusp_resume_device) {
+		int error = find_resume_device();
+
+		if (error)
+			return error;
+	}
+
+	return software_resume();
+}
+late_initcall_sync(software_resume_initcall);
 
 
 static const char * const hibernation_modes[] = {
@@ -1176,6 +1171,9 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	char *name;
 	dev_t res;
 
+	if (!hibernation_available())
+		return 0;
+
 	if (len && buf[len-1] == '\n')
 		len--;
 	name = kstrndup(buf, len, GFP_KERNEL);
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

software_resume can be called either from an init call in the boot code,
or from sysfs once the system has finished booting, and the two
invocation methods this can't race with each other.

For the latter case we did just parse the suspend device manually, while
the former might not have one.  Split software_resume so that the search
only happens for the boot case, which also means the special lockdep
nesting annotation can go away as the system transition mutex can be
taken a little later and doesn't have the sysfs locking nest inside it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 80 ++++++++++++++++++++--------------------
 1 file changed, 39 insertions(+), 41 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 78696aa04f5ca3..45e24b02cd50b6 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -907,7 +907,7 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
-static int find_resume_device(void)
+static int __init find_resume_device(void)
 {
 	if (!strlen(resume_file))
 		return -ENOENT;
@@ -942,53 +942,16 @@ static int find_resume_device(void)
 	return 0;
 }
 
-/**
- * software_resume - Resume from a saved hibernation image.
- *
- * This routine is called as a late initcall, when all devices have been
- * discovered and initialized already.
- *
- * The image reading code is called to see if there is a hibernation image
- * available for reading.  If that is the case, devices are quiesced and the
- * contents of memory is restored from the saved image.
- *
- * If this is successful, control reappears in the restored target kernel in
- * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
- * attempts to recover gracefully and make the kernel return to the normal mode
- * of operation.
- */
 static int software_resume(void)
 {
 	int error;
 
-	/*
-	 * If the user said "noresume".. bail out early.
-	 */
-	if (noresume || !hibernation_available())
-		return 0;
-
-	/*
-	 * name_to_dev_t() below takes a sysfs buffer mutex when sysfs
-	 * is configured into the kernel. Since the regular hibernate
-	 * trigger path is via sysfs which takes a buffer mutex before
-	 * calling hibernate functions (which take system_transition_mutex)
-	 * this can cause lockdep to complain about a possible ABBA deadlock
-	 * which cannot happen since we're in the boot code here and
-	 * sysfs can't be invoked yet. Therefore, we use a subclass
-	 * here to avoid lockdep complaining.
-	 */
-	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
-
-	if (!swsusp_resume_device) {
-		error = find_resume_device();
-		if (error)
-			goto Unlock;
-	}
-
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
+
+	mutex_lock(&system_transition_mutex);
 	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
@@ -1035,7 +998,39 @@ static int software_resume(void)
 	goto Finish;
 }
 
-late_initcall_sync(software_resume);
+/**
+ * software_resume_initcall - Resume from a saved hibernation image.
+ *
+ * This routine is called as a late initcall, when all devices have been
+ * discovered and initialized already.
+ *
+ * The image reading code is called to see if there is a hibernation image
+ * available for reading.  If that is the case, devices are quiesced and the
+ * contents of memory is restored from the saved image.
+ *
+ * If this is successful, control reappears in the restored target kernel in
+ * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
+ * attempts to recover gracefully and make the kernel return to the normal mode
+ * of operation.
+ */
+static int __init software_resume_initcall(void)
+{
+	/*
+	 * If the user said "noresume".. bail out early.
+	 */
+	if (noresume || !hibernation_available())
+		return 0;
+
+	if (!swsusp_resume_device) {
+		int error = find_resume_device();
+
+		if (error)
+			return error;
+	}
+
+	return software_resume();
+}
+late_initcall_sync(software_resume_initcall);
 
 
 static const char * const hibernation_modes[] = {
@@ -1176,6 +1171,9 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	char *name;
 	dev_t res;
 
+	if (!hibernation_available())
+		return 0;
+
 	if (len && buf[len-1] == '\n')
 		len--;
 	name = kstrndup(buf, len, GFP_KERNEL);
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

software_resume can be called either from an init call in the boot code,
or from sysfs once the system has finished booting, and the two
invocation methods this can't race with each other.

For the latter case we did just parse the suspend device manually, while
the former might not have one.  Split software_resume so that the search
only happens for the boot case, which also means the special lockdep
nesting annotation can go away as the system transition mutex can be
taken a little later and doesn't have the sysfs locking nest inside it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 80 ++++++++++++++++++++--------------------
 1 file changed, 39 insertions(+), 41 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 78696aa04f5ca3..45e24b02cd50b6 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -907,7 +907,7 @@ int hibernate_quiet_exec(int (*func)(void *data), void *data)
 }
 EXPORT_SYMBOL_GPL(hibernate_quiet_exec);
 
-static int find_resume_device(void)
+static int __init find_resume_device(void)
 {
 	if (!strlen(resume_file))
 		return -ENOENT;
@@ -942,53 +942,16 @@ static int find_resume_device(void)
 	return 0;
 }
 
-/**
- * software_resume - Resume from a saved hibernation image.
- *
- * This routine is called as a late initcall, when all devices have been
- * discovered and initialized already.
- *
- * The image reading code is called to see if there is a hibernation image
- * available for reading.  If that is the case, devices are quiesced and the
- * contents of memory is restored from the saved image.
- *
- * If this is successful, control reappears in the restored target kernel in
- * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
- * attempts to recover gracefully and make the kernel return to the normal mode
- * of operation.
- */
 static int software_resume(void)
 {
 	int error;
 
-	/*
-	 * If the user said "noresume".. bail out early.
-	 */
-	if (noresume || !hibernation_available())
-		return 0;
-
-	/*
-	 * name_to_dev_t() below takes a sysfs buffer mutex when sysfs
-	 * is configured into the kernel. Since the regular hibernate
-	 * trigger path is via sysfs which takes a buffer mutex before
-	 * calling hibernate functions (which take system_transition_mutex)
-	 * this can cause lockdep to complain about a possible ABBA deadlock
-	 * which cannot happen since we're in the boot code here and
-	 * sysfs can't be invoked yet. Therefore, we use a subclass
-	 * here to avoid lockdep complaining.
-	 */
-	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
-
-	if (!swsusp_resume_device) {
-		error = find_resume_device();
-		if (error)
-			goto Unlock;
-	}
-
 	pm_pr_dbg("Hibernation image partition %d:%d present\n",
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
+
+	mutex_lock(&system_transition_mutex);
 	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
@@ -1035,7 +998,39 @@ static int software_resume(void)
 	goto Finish;
 }
 
-late_initcall_sync(software_resume);
+/**
+ * software_resume_initcall - Resume from a saved hibernation image.
+ *
+ * This routine is called as a late initcall, when all devices have been
+ * discovered and initialized already.
+ *
+ * The image reading code is called to see if there is a hibernation image
+ * available for reading.  If that is the case, devices are quiesced and the
+ * contents of memory is restored from the saved image.
+ *
+ * If this is successful, control reappears in the restored target kernel in
+ * hibernation_snapshot() which returns to hibernate().  Otherwise, the routine
+ * attempts to recover gracefully and make the kernel return to the normal mode
+ * of operation.
+ */
+static int __init software_resume_initcall(void)
+{
+	/*
+	 * If the user said "noresume".. bail out early.
+	 */
+	if (noresume || !hibernation_available())
+		return 0;
+
+	if (!swsusp_resume_device) {
+		int error = find_resume_device();
+
+		if (error)
+			return error;
+	}
+
+	return software_resume();
+}
+late_initcall_sync(software_resume_initcall);
 
 
 static const char * const hibernation_modes[] = {
@@ -1176,6 +1171,9 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	char *name;
 	dev_t res;
 
+	if (!hibernation_available())
+		return 0;
+
 	if (len && buf[len-1] == '\n')
 		len--;
 	name = kstrndup(buf, len, GFP_KERNEL);
-- 
2.39.2

Re: [dm-devel] [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Vlastimil Babka]

On 5/31/23 14:55, Christoph Hellwig wrote:
> software_resume can be called either from an init call in the boot code,
> or from sysfs once the system has finished booting, and the two
> invocation methods this can't race with each other.
> 
> For the latter case we did just parse the suspend device manually, while
> the former might not have one.  Split software_resume so that the search
> only happens for the boot case, which also means the special lockdep
> nesting annotation can go away as the system transition mutex can be
> taken a little later and doesn't have the sysfs locking nest inside it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>

This caused a regression for me in 6.5-rc1+, fix below.

----8<----
From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Wed, 2 Aug 2023 15:46:18 +0200
Subject: [PATCH] PM: hibernate: fix resume_store() return value when
 hibernation not available

On a laptop with hibernation set up but not actively used, and with
secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
the following repeated messages:

  A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
  lockdown_is_locked_down: 25311154 callbacks suppressed
  Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
  ...

Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
move finding the resume device out of software_resume") which
inadvertently changed the return value from resume_store() to 0 when
!hibernation_available(). This apparently translates to userspace
write() returning 0 as in number of bytes written, and userspace looping
indefinitely in the attempt to write the intended value.

Fix this by returning the full number of bytes that were to be written,
as that's what was done before the commit.

Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
 kernel/power/hibernate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index e1b4bfa938dd..2b4a946a6ff5 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1166,7 +1166,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	int error;
 
 	if (!hibernation_available())
-		return 0;
+		return n;
 
 	if (len && buf[len-1] == '\n')
 		len--;
-- 
2.41.0


--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Vlastimil Babka]

On 5/31/23 14:55, Christoph Hellwig wrote:
> software_resume can be called either from an init call in the boot code,
> or from sysfs once the system has finished booting, and the two
> invocation methods this can't race with each other.
> 
> For the latter case we did just parse the suspend device manually, while
> the former might not have one.  Split software_resume so that the search
> only happens for the boot case, which also means the special lockdep
> nesting annotation can go away as the system transition mutex can be
> taken a little later and doesn't have the sysfs locking nest inside it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>

This caused a regression for me in 6.5-rc1+, fix below.

----8<----
From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Wed, 2 Aug 2023 15:46:18 +0200
Subject: [PATCH] PM: hibernate: fix resume_store() return value when
 hibernation not available

On a laptop with hibernation set up but not actively used, and with
secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
the following repeated messages:

  A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
  lockdown_is_locked_down: 25311154 callbacks suppressed
  Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
  ...

Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
move finding the resume device out of software_resume") which
inadvertently changed the return value from resume_store() to 0 when
!hibernation_available(). This apparently translates to userspace
write() returning 0 as in number of bytes written, and userspace looping
indefinitely in the attempt to write the intended value.

Fix this by returning the full number of bytes that were to be written,
as that's what was done before the commit.

Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
 kernel/power/hibernate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index e1b4bfa938dd..2b4a946a6ff5 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1166,7 +1166,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	int error;
 
 	if (!hibernation_available())
-		return 0;
+		return n;
 
 	if (len && buf[len-1] == '\n')
 		len--;
-- 
2.41.0



______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Vlastimil Babka]

On 5/31/23 14:55, Christoph Hellwig wrote:
> software_resume can be called either from an init call in the boot code,
> or from sysfs once the system has finished booting, and the two
> invocation methods this can't race with each other.
> 
> For the latter case we did just parse the suspend device manually, while
> the former might not have one.  Split software_resume so that the search
> only happens for the boot case, which also means the special lockdep
> nesting annotation can go away as the system transition mutex can be
> taken a little later and doesn't have the sysfs locking nest inside it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> Acked-by: Rafael J. Wysocki <rafael@kernel.org>

This caused a regression for me in 6.5-rc1+, fix below.

----8<----
From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
From: Vlastimil Babka <vbabka@suse.cz>
Date: Wed, 2 Aug 2023 15:46:18 +0200
Subject: [PATCH] PM: hibernate: fix resume_store() return value when
 hibernation not available

On a laptop with hibernation set up but not actively used, and with
secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
the following repeated messages:

  A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
  lockdown_is_locked_down: 25311154 callbacks suppressed
  Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
  ...

Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
move finding the resume device out of software_resume") which
inadvertently changed the return value from resume_store() to 0 when
!hibernation_available(). This apparently translates to userspace
write() returning 0 as in number of bytes written, and userspace looping
indefinitely in the attempt to write the intended value.

Fix this by returning the full number of bytes that were to be written,
as that's what was done before the commit.

Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
 kernel/power/hibernate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index e1b4bfa938dd..2b4a946a6ff5 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1166,7 +1166,7 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	int error;
 
 	if (!hibernation_available())
-		return 0;
+		return n;
 
 	if (len && buf[len-1] == '\n')
 		len--;
-- 
2.41.0

Re: [dm-devel] [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

Looks good, thanks!

Reviewed-by: Christoph Hellwig <hch@lst.de>

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

Looks good, thanks!

Reviewed-by: Christoph Hellwig <hch@lst.de>


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Christoph Hellwig]

Looks good, thanks!

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [dm-devel] [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Greg Kroah-Hartman]

On Fri, Aug 04, 2023 at 12:31:01PM +0200, Christoph Hellwig wrote:
> Looks good, thanks!
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> 

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>


--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Greg Kroah-Hartman]

On Fri, Aug 04, 2023 at 12:31:01PM +0200, Christoph Hellwig wrote:
> Looks good, thanks!
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> 

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>



______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Greg Kroah-Hartman]

On Fri, Aug 04, 2023 at 12:31:01PM +0200, Christoph Hellwig wrote:
> Looks good, thanks!
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> 

Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Re: [dm-devel] [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Linux regression tracking #adding (Thorsten Leemhuis)]

[TLDR: I'm adding this report to the list of tracked Linux kernel
regressions; the text you find below is based on a few templates
paragraphs you might have encountered already in similar form.
See link in footer if these mails annoy you.]

On 03.08.23 10:27, Vlastimil Babka wrote:
> On 5/31/23 14:55, Christoph Hellwig wrote:
>> software_resume can be called either from an init call in the boot code,
>> or from sysfs once the system has finished booting, and the two
>> invocation methods this can't race with each other.
>>
>> For the latter case we did just parse the suspend device manually, while
>> the former might not have one.  Split software_resume so that the search
>> only happens for the boot case, which also means the special lockdep
>> nesting annotation can go away as the system transition mutex can be
>> taken a little later and doesn't have the sysfs locking nest inside it.
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> 
> This caused a regression for me in 6.5-rc1+, fix below.
> 
> ----8<----
>>From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
> From: Vlastimil Babka <vbabka@suse.cz>
> Date: Wed, 2 Aug 2023 15:46:18 +0200
> Subject: [PATCH] PM: hibernate: fix resume_store() return value when
>  hibernation not available
> 
> On a laptop with hibernation set up but not actively used, and with
> secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
> the following repeated messages:
> 
>   A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
>   lockdown_is_locked_down: 25311154 callbacks suppressed
>   Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
>   ...
> 
> Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
> move finding the resume device out of software_resume") which
> inadvertently changed the return value from resume_store() to 0 when
> !hibernation_available(). This apparently translates to userspace
> write() returning 0 as in number of bytes written, and userspace looping
> indefinitely in the attempt to write the intended value.
> 
> Fix this by returning the full number of bytes that were to be written,
> as that's what was done before the commit.
> 
> Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
> [...]

Thanks for the report. To be sure the issue doesn't fall through the
cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
tracking bot:

#regzbot ^introduced cc89c63e2fe3
#regzbot title pm: boot problems when hibernate is configured and kernel
locked down
#regzbot fix: PM: hibernate: fix resume_store() return value when
hibernation not available
#regzbot ignore-activity

This isn't a regression? This issue or a fix for it are already
discussed somewhere else? It was fixed already? You want to clarify when
the regression started to happen? Or point out I got the title or
something else totally wrong? Then just reply and tell me -- ideally
while also telling regzbot about it, as explained by the page listed in
the footer of this mail.

Developers: When fixing the issue, remember to add 'Link:' tags pointing
to the report (the parent of this mail). See page linked in footer for
details.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Linux regression tracking #adding (Thorsten Leemhuis)]

[TLDR: I'm adding this report to the list of tracked Linux kernel
regressions; the text you find below is based on a few templates
paragraphs you might have encountered already in similar form.
See link in footer if these mails annoy you.]

On 03.08.23 10:27, Vlastimil Babka wrote:
> On 5/31/23 14:55, Christoph Hellwig wrote:
>> software_resume can be called either from an init call in the boot code,
>> or from sysfs once the system has finished booting, and the two
>> invocation methods this can't race with each other.
>>
>> For the latter case we did just parse the suspend device manually, while
>> the former might not have one.  Split software_resume so that the search
>> only happens for the boot case, which also means the special lockdep
>> nesting annotation can go away as the system transition mutex can be
>> taken a little later and doesn't have the sysfs locking nest inside it.
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> 
> This caused a regression for me in 6.5-rc1+, fix below.
> 
> ----8<----
>>From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
> From: Vlastimil Babka <vbabka@suse.cz>
> Date: Wed, 2 Aug 2023 15:46:18 +0200
> Subject: [PATCH] PM: hibernate: fix resume_store() return value when
>  hibernation not available
> 
> On a laptop with hibernation set up but not actively used, and with
> secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
> the following repeated messages:
> 
>   A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
>   lockdown_is_locked_down: 25311154 callbacks suppressed
>   Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
>   ...
> 
> Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
> move finding the resume device out of software_resume") which
> inadvertently changed the return value from resume_store() to 0 when
> !hibernation_available(). This apparently translates to userspace
> write() returning 0 as in number of bytes written, and userspace looping
> indefinitely in the attempt to write the intended value.
> 
> Fix this by returning the full number of bytes that were to be written,
> as that's what was done before the commit.
> 
> Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
> [...]

Thanks for the report. To be sure the issue doesn't fall through the
cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
tracking bot:

#regzbot ^introduced cc89c63e2fe3
#regzbot title pm: boot problems when hibernate is configured and kernel
locked down
#regzbot fix: PM: hibernate: fix resume_store() return value when
hibernation not available
#regzbot ignore-activity

This isn't a regression? This issue or a fix for it are already
discussed somewhere else? It was fixed already? You want to clarify when
the regression started to happen? Or point out I got the title or
something else totally wrong? Then just reply and tell me -- ideally
while also telling regzbot about it, as explained by the page listed in
the footer of this mail.

Developers: When fixing the issue, remember to add 'Link:' tags pointing
to the report (the parent of this mail). See page linked in footer for
details.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 04/24] PM: hibernate: move finding the resume device out of software_resume

[Linux regression tracking #adding (Thorsten Leemhuis)]

[TLDR: I'm adding this report to the list of tracked Linux kernel
regressions; the text you find below is based on a few templates
paragraphs you might have encountered already in similar form.
See link in footer if these mails annoy you.]

On 03.08.23 10:27, Vlastimil Babka wrote:
> On 5/31/23 14:55, Christoph Hellwig wrote:
>> software_resume can be called either from an init call in the boot code,
>> or from sysfs once the system has finished booting, and the two
>> invocation methods this can't race with each other.
>>
>> For the latter case we did just parse the suspend device manually, while
>> the former might not have one.  Split software_resume so that the search
>> only happens for the boot case, which also means the special lockdep
>> nesting annotation can go away as the system transition mutex can be
>> taken a little later and doesn't have the sysfs locking nest inside it.
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> Acked-by: Rafael J. Wysocki <rafael@kernel.org>
> 
> This caused a regression for me in 6.5-rc1+, fix below.
> 
> ----8<----
>>From 95a310ae6cfae9b3cab61e54a1bce488c3ab93a1 Mon Sep 17 00:00:00 2001
> From: Vlastimil Babka <vbabka@suse.cz>
> Date: Wed, 2 Aug 2023 15:46:18 +0200
> Subject: [PATCH] PM: hibernate: fix resume_store() return value when
>  hibernation not available
> 
> On a laptop with hibernation set up but not actively used, and with
> secure boot and lockdown enabled kernel, 6.5-rc1 gets stuck on boot with
> the following repeated messages:
> 
>   A start job is running for Resume from hibernation using device /dev/system/swap (24s / no limit)
>   lockdown_is_locked_down: 25311154 callbacks suppressed
>   Lockdown: systemd-hiberna: hibernation is restricted; see man kernel_lockdown.7
>   ...
> 
> Checking the resume code leads to commit cc89c63e2fe3 ("PM: hibernate:
> move finding the resume device out of software_resume") which
> inadvertently changed the return value from resume_store() to 0 when
> !hibernation_available(). This apparently translates to userspace
> write() returning 0 as in number of bytes written, and userspace looping
> indefinitely in the attempt to write the intended value.
> 
> Fix this by returning the full number of bytes that were to be written,
> as that's what was done before the commit.
> 
> Fixes: cc89c63e2fe3 ("PM: hibernate: move finding the resume device out of software_resume")
> [...]

Thanks for the report. To be sure the issue doesn't fall through the
cracks unnoticed, I'm adding it to regzbot, the Linux kernel regression
tracking bot:

#regzbot ^introduced cc89c63e2fe3
#regzbot title pm: boot problems when hibernate is configured and kernel
locked down
#regzbot fix: PM: hibernate: fix resume_store() return value when
hibernation not available
#regzbot ignore-activity

This isn't a regression? This issue or a fix for it are already
discussed somewhere else? It was fixed already? You want to clarify when
the regression started to happen? Or point out I got the title or
something else totally wrong? Then just reply and tell me -- ideally
while also telling regzbot about it, as explained by the page listed in
the footer of this mail.

Developers: When fixing the issue, remember to add 'Link:' tags pointing
to the report (the parent of this mail). See page linked in footer for
details.

Ciao, Thorsten (wearing his 'the Linux kernel's regression tracker' hat)
--
Everything you wanna know about Linux kernel regression tracking:
https://linux-regtracking.leemhuis.info/about/#tldr
That page also explains what to do if mails like this annoy you.

[dm-devel] [PATCH 05/24] init: remove pointless Root_* values

[Christoph Hellwig]

Remove all unused defines, and just use the expanded versions for
the SCSI disk majors.

I've decided to keep Root_RAM0 even if it could be expanded as there
is a lot of special casing for it in the init code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/alpha/kernel/setup.c               | 2 +-
 arch/ia64/kernel/setup.c                | 2 +-
 arch/powerpc/platforms/powermac/setup.c | 3 ++-
 include/linux/root_dev.h                | 8 --------
 4 files changed, 4 insertions(+), 11 deletions(-)

diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 33bf3a62700270..b650ff1cb022ee 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -658,7 +658,7 @@ setup_arch(char **cmdline_p)
 #endif
 
 	/* Default root filesystem to sda2.  */
-	ROOT_DEV = Root_SDA2;
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 #ifdef CONFIG_EISA
 	/* FIXME:  only set this when we actually have EISA in this box? */
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index c0572804427275..becdb4f33c2195 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -627,7 +627,7 @@ setup_arch (char **cmdline_p)
 	 * is physical disk 1 partition 1 and the Linux root disk is
 	 * physical disk 1 partition 2.
 	 */
-	ROOT_DEV = Root_SDA2;		/* default to second partition on first drive */
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 	if (is_uv_system())
 		uv_setup(cmdline_p);
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 193cc9c394221b..0c41f4b005bcf3 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -76,7 +76,8 @@ int pmac_newworld;
 
 static int current_root_goodness = -1;
 
-#define DEFAULT_ROOT_DEVICE Root_SDA1	/* sda1 - slightly silly choice */
+/* sda1 - slightly silly choice */
+#define DEFAULT_ROOT_DEVICE	MKDEV(SCSI_DISK0_MAJOR, 1)
 
 sys_ctrler_t sys_ctrler = SYS_CTRLER_UNKNOWN;
 EXPORT_SYMBOL(sys_ctrler);
diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index 4e78651371ba92..ed3ea8da642972 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -10,14 +10,6 @@ enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
-	Root_RAM1 = MKDEV(RAMDISK_MAJOR, 1),
-	Root_FD0 = MKDEV(FLOPPY_MAJOR, 0),
-	Root_HDA1 = MKDEV(IDE0_MAJOR, 1),
-	Root_HDA2 = MKDEV(IDE0_MAJOR, 2),
-	Root_SDA1 = MKDEV(SCSI_DISK0_MAJOR, 1),
-	Root_SDA2 = MKDEV(SCSI_DISK0_MAJOR, 2),
-	Root_HDC1 = MKDEV(IDE1_MAJOR, 1),
-	Root_SR0 = MKDEV(SCSI_CDROM_MAJOR, 0),
 };
 
 extern dev_t ROOT_DEV;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 05/24] init: remove pointless Root_* values

[Christoph Hellwig]

Remove all unused defines, and just use the expanded versions for
the SCSI disk majors.

I've decided to keep Root_RAM0 even if it could be expanded as there
is a lot of special casing for it in the init code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/alpha/kernel/setup.c               | 2 +-
 arch/ia64/kernel/setup.c                | 2 +-
 arch/powerpc/platforms/powermac/setup.c | 3 ++-
 include/linux/root_dev.h                | 8 --------
 4 files changed, 4 insertions(+), 11 deletions(-)

diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 33bf3a62700270..b650ff1cb022ee 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -658,7 +658,7 @@ setup_arch(char **cmdline_p)
 #endif
 
 	/* Default root filesystem to sda2.  */
-	ROOT_DEV = Root_SDA2;
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 #ifdef CONFIG_EISA
 	/* FIXME:  only set this when we actually have EISA in this box? */
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index c0572804427275..becdb4f33c2195 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -627,7 +627,7 @@ setup_arch (char **cmdline_p)
 	 * is physical disk 1 partition 1 and the Linux root disk is
 	 * physical disk 1 partition 2.
 	 */
-	ROOT_DEV = Root_SDA2;		/* default to second partition on first drive */
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 	if (is_uv_system())
 		uv_setup(cmdline_p);
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 193cc9c394221b..0c41f4b005bcf3 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -76,7 +76,8 @@ int pmac_newworld;
 
 static int current_root_goodness = -1;
 
-#define DEFAULT_ROOT_DEVICE Root_SDA1	/* sda1 - slightly silly choice */
+/* sda1 - slightly silly choice */
+#define DEFAULT_ROOT_DEVICE	MKDEV(SCSI_DISK0_MAJOR, 1)
 
 sys_ctrler_t sys_ctrler = SYS_CTRLER_UNKNOWN;
 EXPORT_SYMBOL(sys_ctrler);
diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index 4e78651371ba92..ed3ea8da642972 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -10,14 +10,6 @@ enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
-	Root_RAM1 = MKDEV(RAMDISK_MAJOR, 1),
-	Root_FD0 = MKDEV(FLOPPY_MAJOR, 0),
-	Root_HDA1 = MKDEV(IDE0_MAJOR, 1),
-	Root_HDA2 = MKDEV(IDE0_MAJOR, 2),
-	Root_SDA1 = MKDEV(SCSI_DISK0_MAJOR, 1),
-	Root_SDA2 = MKDEV(SCSI_DISK0_MAJOR, 2),
-	Root_HDC1 = MKDEV(IDE1_MAJOR, 1),
-	Root_SR0 = MKDEV(SCSI_CDROM_MAJOR, 0),
 };
 
 extern dev_t ROOT_DEV;
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 05/24] init: remove pointless Root_* values

[Christoph Hellwig]

Remove all unused defines, and just use the expanded versions for
the SCSI disk majors.

I've decided to keep Root_RAM0 even if it could be expanded as there
is a lot of special casing for it in the init code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 arch/alpha/kernel/setup.c               | 2 +-
 arch/ia64/kernel/setup.c                | 2 +-
 arch/powerpc/platforms/powermac/setup.c | 3 ++-
 include/linux/root_dev.h                | 8 --------
 4 files changed, 4 insertions(+), 11 deletions(-)

diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c
index 33bf3a62700270..b650ff1cb022ee 100644
--- a/arch/alpha/kernel/setup.c
+++ b/arch/alpha/kernel/setup.c
@@ -658,7 +658,7 @@ setup_arch(char **cmdline_p)
 #endif
 
 	/* Default root filesystem to sda2.  */
-	ROOT_DEV = Root_SDA2;
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 #ifdef CONFIG_EISA
 	/* FIXME:  only set this when we actually have EISA in this box? */
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index c0572804427275..becdb4f33c2195 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -627,7 +627,7 @@ setup_arch (char **cmdline_p)
 	 * is physical disk 1 partition 1 and the Linux root disk is
 	 * physical disk 1 partition 2.
 	 */
-	ROOT_DEV = Root_SDA2;		/* default to second partition on first drive */
+	ROOT_DEV = MKDEV(SCSI_DISK0_MAJOR, 2);
 
 	if (is_uv_system())
 		uv_setup(cmdline_p);
diff --git a/arch/powerpc/platforms/powermac/setup.c b/arch/powerpc/platforms/powermac/setup.c
index 193cc9c394221b..0c41f4b005bcf3 100644
--- a/arch/powerpc/platforms/powermac/setup.c
+++ b/arch/powerpc/platforms/powermac/setup.c
@@ -76,7 +76,8 @@ int pmac_newworld;
 
 static int current_root_goodness = -1;
 
-#define DEFAULT_ROOT_DEVICE Root_SDA1	/* sda1 - slightly silly choice */
+/* sda1 - slightly silly choice */
+#define DEFAULT_ROOT_DEVICE	MKDEV(SCSI_DISK0_MAJOR, 1)
 
 sys_ctrler_t sys_ctrler = SYS_CTRLER_UNKNOWN;
 EXPORT_SYMBOL(sys_ctrler);
diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index 4e78651371ba92..ed3ea8da642972 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -10,14 +10,6 @@ enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
-	Root_RAM1 = MKDEV(RAMDISK_MAJOR, 1),
-	Root_FD0 = MKDEV(FLOPPY_MAJOR, 0),
-	Root_HDA1 = MKDEV(IDE0_MAJOR, 1),
-	Root_HDA2 = MKDEV(IDE0_MAJOR, 2),
-	Root_SDA1 = MKDEV(SCSI_DISK0_MAJOR, 1),
-	Root_SDA2 = MKDEV(SCSI_DISK0_MAJOR, 2),
-	Root_HDC1 = MKDEV(IDE1_MAJOR, 1),
-	Root_SR0 = MKDEV(SCSI_CDROM_MAJOR, 0),
 };
 
 extern dev_t ROOT_DEV;
-- 
2.39.2

[dm-devel] [PATCH 06/24] init: rename mount_block_root to mount_root_generic

[Christoph Hellwig]

mount_block_root is also used to mount non-block file systems, so give
it a better name.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 6 +++---
 init/do_mounts.h        | 2 +-
 init/do_mounts_initrd.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 2fe7901b5bcfaf..a2c0baace0992c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -391,7 +391,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_block_root(char *name, int flags)
+void __init mount_root_generic(char *name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -589,7 +589,7 @@ void __init mount_root(void)
 
 		if (err < 0)
 			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_block_root("/dev/root", root_mountflags);
+		mount_root_generic("/dev/root", root_mountflags);
 	}
 #endif
 }
@@ -620,7 +620,7 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_block_root(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 7a29ac3e427bab..33623025f6951a 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,7 +10,7 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_block_root(char *name, int flags);
+void  mount_root_generic(char *name, int flags);
 void  mount_root(void);
 extern int root_mountflags;
 
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 34731241377d30..686d1ff3af4bb1 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -95,7 +95,7 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 06/24] init: rename mount_block_root to mount_root_generic

[Christoph Hellwig]

mount_block_root is also used to mount non-block file systems, so give
it a better name.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 6 +++---
 init/do_mounts.h        | 2 +-
 init/do_mounts_initrd.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 2fe7901b5bcfaf..a2c0baace0992c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -391,7 +391,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_block_root(char *name, int flags)
+void __init mount_root_generic(char *name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -589,7 +589,7 @@ void __init mount_root(void)
 
 		if (err < 0)
 			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_block_root("/dev/root", root_mountflags);
+		mount_root_generic("/dev/root", root_mountflags);
 	}
 #endif
 }
@@ -620,7 +620,7 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_block_root(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 7a29ac3e427bab..33623025f6951a 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,7 +10,7 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_block_root(char *name, int flags);
+void  mount_root_generic(char *name, int flags);
 void  mount_root(void);
 extern int root_mountflags;
 
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 34731241377d30..686d1ff3af4bb1 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -95,7 +95,7 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 06/24] init: rename mount_block_root to mount_root_generic

[Christoph Hellwig]

mount_block_root is also used to mount non-block file systems, so give
it a better name.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 6 +++---
 init/do_mounts.h        | 2 +-
 init/do_mounts_initrd.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 2fe7901b5bcfaf..a2c0baace0992c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -391,7 +391,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_block_root(char *name, int flags)
+void __init mount_root_generic(char *name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -589,7 +589,7 @@ void __init mount_root(void)
 
 		if (err < 0)
 			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_block_root("/dev/root", root_mountflags);
+		mount_root_generic("/dev/root", root_mountflags);
 	}
 #endif
 }
@@ -620,7 +620,7 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_block_root(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 7a29ac3e427bab..33623025f6951a 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,7 +10,7 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_block_root(char *name, int flags);
+void  mount_root_generic(char *name, int flags);
 void  mount_root(void);
 extern int root_mountflags;
 
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 34731241377d30..686d1ff3af4bb1 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -95,7 +95,7 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_block_root("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
-- 
2.39.2

[dm-devel] [PATCH 07/24] init: refactor mount_root

[Christoph Hellwig]

Provide stubs for all the lower level mount helpers, and just switch
on ROOT_DEV in the main function.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 104 +++++++++++++++++++++++++----------------------
 1 file changed, 56 insertions(+), 48 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index a2c0baace0992c..e708b02d9d6566 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -453,15 +453,14 @@ void __init mount_root_generic(char *name, int flags)
 #define NFSROOT_TIMEOUT_MAX	30
 #define NFSROOT_RETRY_MAX	5
 
-static int __init mount_nfs_root(void)
+static void __init mount_nfs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = nfs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (nfs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	/*
 	 * The server or network may not be ready, so try several
@@ -470,10 +469,8 @@ static int __init mount_nfs_root(void)
 	 */
 	timeout = NFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "nfs",
-					root_mountflags, root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "nfs", root_mountflags, root_data))
+			return;
 		if (try > NFSROOT_RETRY_MAX)
 			break;
 
@@ -483,9 +480,14 @@ static int __init mount_nfs_root(void)
 		if (timeout > NFSROOT_TIMEOUT_MAX)
 			timeout = NFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via NFS.\n");
 }
-#endif
+#else
+static inline void mount_nfs_root(void)
+{
+}
+#endif /* CONFIG_ROOT_NFS */
 
 #ifdef CONFIG_CIFS_ROOT
 
@@ -495,22 +497,20 @@ extern int cifs_root_data(char **dev, char **opts);
 #define CIFSROOT_TIMEOUT_MAX	30
 #define CIFSROOT_RETRY_MAX	5
 
-static int __init mount_cifs_root(void)
+static void __init mount_cifs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = cifs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (cifs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	timeout = CIFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "cifs", root_mountflags,
-				    root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "cifs", root_mountflags,
+				   root_data))
+			return;
 		if (try > CIFSROOT_RETRY_MAX)
 			break;
 
@@ -519,9 +519,14 @@ static int __init mount_cifs_root(void)
 		if (timeout > CIFSROOT_TIMEOUT_MAX)
 			timeout = CIFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via SMB.\n");
 }
-#endif
+#else
+static inline void mount_cifs_root(void)
+{
+}
+#endif /* CONFIG_CIFS_ROOT */
 
 static bool __init fs_is_nodev(char *fstype)
 {
@@ -563,35 +568,38 @@ static int __init mount_nodev_root(void)
 	return err;
 }
 
-void __init mount_root(void)
-{
-#ifdef CONFIG_ROOT_NFS
-	if (ROOT_DEV == Root_NFS) {
-		if (!mount_nfs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via NFS.\n");
-		return;
-	}
-#endif
-#ifdef CONFIG_CIFS_ROOT
-	if (ROOT_DEV == Root_CIFS) {
-		if (!mount_cifs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via SMB.\n");
-		return;
-	}
-#endif
-	if (ROOT_DEV == 0 && root_device_name && root_fs_names) {
-		if (mount_nodev_root() == 0)
-			return;
-	}
 #ifdef CONFIG_BLOCK
-	{
-		int err = create_dev("/dev/root", ROOT_DEV);
+static void __init mount_block_root(void)
+{
+	int err = create_dev("/dev/root", ROOT_DEV);
+
+	if (err < 0)
+		pr_emerg("Failed to create /dev/root: %d\n", err);
+	mount_root_generic("/dev/root", root_mountflags);
+}
+#else
+static inline void mount_block_root(void)
+{
+}
+#endif /* CONFIG_BLOCK */
 
-		if (err < 0)
-			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_root_generic("/dev/root", root_mountflags);
+void __init mount_root(void)
+{
+	switch (ROOT_DEV) {
+	case Root_NFS:
+		mount_nfs_root();
+		break;
+	case Root_CIFS:
+		mount_cifs_root();
+		break;
+	case 0:
+		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+			break;
+		fallthrough;
+	default:
+		mount_block_root();
+		break;
 	}
-#endif
 }
 
 /*
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 07/24] init: refactor mount_root

[Christoph Hellwig]

Provide stubs for all the lower level mount helpers, and just switch
on ROOT_DEV in the main function.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 104 +++++++++++++++++++++++++----------------------
 1 file changed, 56 insertions(+), 48 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index a2c0baace0992c..e708b02d9d6566 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -453,15 +453,14 @@ void __init mount_root_generic(char *name, int flags)
 #define NFSROOT_TIMEOUT_MAX	30
 #define NFSROOT_RETRY_MAX	5
 
-static int __init mount_nfs_root(void)
+static void __init mount_nfs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = nfs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (nfs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	/*
 	 * The server or network may not be ready, so try several
@@ -470,10 +469,8 @@ static int __init mount_nfs_root(void)
 	 */
 	timeout = NFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "nfs",
-					root_mountflags, root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "nfs", root_mountflags, root_data))
+			return;
 		if (try > NFSROOT_RETRY_MAX)
 			break;
 
@@ -483,9 +480,14 @@ static int __init mount_nfs_root(void)
 		if (timeout > NFSROOT_TIMEOUT_MAX)
 			timeout = NFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via NFS.\n");
 }
-#endif
+#else
+static inline void mount_nfs_root(void)
+{
+}
+#endif /* CONFIG_ROOT_NFS */
 
 #ifdef CONFIG_CIFS_ROOT
 
@@ -495,22 +497,20 @@ extern int cifs_root_data(char **dev, char **opts);
 #define CIFSROOT_TIMEOUT_MAX	30
 #define CIFSROOT_RETRY_MAX	5
 
-static int __init mount_cifs_root(void)
+static void __init mount_cifs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = cifs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (cifs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	timeout = CIFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "cifs", root_mountflags,
-				    root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "cifs", root_mountflags,
+				   root_data))
+			return;
 		if (try > CIFSROOT_RETRY_MAX)
 			break;
 
@@ -519,9 +519,14 @@ static int __init mount_cifs_root(void)
 		if (timeout > CIFSROOT_TIMEOUT_MAX)
 			timeout = CIFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via SMB.\n");
 }
-#endif
+#else
+static inline void mount_cifs_root(void)
+{
+}
+#endif /* CONFIG_CIFS_ROOT */
 
 static bool __init fs_is_nodev(char *fstype)
 {
@@ -563,35 +568,38 @@ static int __init mount_nodev_root(void)
 	return err;
 }
 
-void __init mount_root(void)
-{
-#ifdef CONFIG_ROOT_NFS
-	if (ROOT_DEV == Root_NFS) {
-		if (!mount_nfs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via NFS.\n");
-		return;
-	}
-#endif
-#ifdef CONFIG_CIFS_ROOT
-	if (ROOT_DEV == Root_CIFS) {
-		if (!mount_cifs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via SMB.\n");
-		return;
-	}
-#endif
-	if (ROOT_DEV == 0 && root_device_name && root_fs_names) {
-		if (mount_nodev_root() == 0)
-			return;
-	}
 #ifdef CONFIG_BLOCK
-	{
-		int err = create_dev("/dev/root", ROOT_DEV);
+static void __init mount_block_root(void)
+{
+	int err = create_dev("/dev/root", ROOT_DEV);
+
+	if (err < 0)
+		pr_emerg("Failed to create /dev/root: %d\n", err);
+	mount_root_generic("/dev/root", root_mountflags);
+}
+#else
+static inline void mount_block_root(void)
+{
+}
+#endif /* CONFIG_BLOCK */
 
-		if (err < 0)
-			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_root_generic("/dev/root", root_mountflags);
+void __init mount_root(void)
+{
+	switch (ROOT_DEV) {
+	case Root_NFS:
+		mount_nfs_root();
+		break;
+	case Root_CIFS:
+		mount_cifs_root();
+		break;
+	case 0:
+		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+			break;
+		fallthrough;
+	default:
+		mount_block_root();
+		break;
 	}
-#endif
 }
 
 /*
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 07/24] init: refactor mount_root

[Christoph Hellwig]

Provide stubs for all the lower level mount helpers, and just switch
on ROOT_DEV in the main function.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 104 +++++++++++++++++++++++++----------------------
 1 file changed, 56 insertions(+), 48 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index a2c0baace0992c..e708b02d9d6566 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -453,15 +453,14 @@ void __init mount_root_generic(char *name, int flags)
 #define NFSROOT_TIMEOUT_MAX	30
 #define NFSROOT_RETRY_MAX	5
 
-static int __init mount_nfs_root(void)
+static void __init mount_nfs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = nfs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (nfs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	/*
 	 * The server or network may not be ready, so try several
@@ -470,10 +469,8 @@ static int __init mount_nfs_root(void)
 	 */
 	timeout = NFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "nfs",
-					root_mountflags, root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "nfs", root_mountflags, root_data))
+			return;
 		if (try > NFSROOT_RETRY_MAX)
 			break;
 
@@ -483,9 +480,14 @@ static int __init mount_nfs_root(void)
 		if (timeout > NFSROOT_TIMEOUT_MAX)
 			timeout = NFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via NFS.\n");
 }
-#endif
+#else
+static inline void mount_nfs_root(void)
+{
+}
+#endif /* CONFIG_ROOT_NFS */
 
 #ifdef CONFIG_CIFS_ROOT
 
@@ -495,22 +497,20 @@ extern int cifs_root_data(char **dev, char **opts);
 #define CIFSROOT_TIMEOUT_MAX	30
 #define CIFSROOT_RETRY_MAX	5
 
-static int __init mount_cifs_root(void)
+static void __init mount_cifs_root(void)
 {
 	char *root_dev, *root_data;
 	unsigned int timeout;
-	int try, err;
+	int try;
 
-	err = cifs_root_data(&root_dev, &root_data);
-	if (err != 0)
-		return 0;
+	if (cifs_root_data(&root_dev, &root_data))
+		goto fail;
 
 	timeout = CIFSROOT_TIMEOUT_MIN;
 	for (try = 1; ; try++) {
-		err = do_mount_root(root_dev, "cifs", root_mountflags,
-				    root_data);
-		if (err == 0)
-			return 1;
+		if (!do_mount_root(root_dev, "cifs", root_mountflags,
+				   root_data))
+			return;
 		if (try > CIFSROOT_RETRY_MAX)
 			break;
 
@@ -519,9 +519,14 @@ static int __init mount_cifs_root(void)
 		if (timeout > CIFSROOT_TIMEOUT_MAX)
 			timeout = CIFSROOT_TIMEOUT_MAX;
 	}
-	return 0;
+fail:
+	pr_err("VFS: Unable to mount root fs via SMB.\n");
 }
-#endif
+#else
+static inline void mount_cifs_root(void)
+{
+}
+#endif /* CONFIG_CIFS_ROOT */
 
 static bool __init fs_is_nodev(char *fstype)
 {
@@ -563,35 +568,38 @@ static int __init mount_nodev_root(void)
 	return err;
 }
 
-void __init mount_root(void)
-{
-#ifdef CONFIG_ROOT_NFS
-	if (ROOT_DEV == Root_NFS) {
-		if (!mount_nfs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via NFS.\n");
-		return;
-	}
-#endif
-#ifdef CONFIG_CIFS_ROOT
-	if (ROOT_DEV == Root_CIFS) {
-		if (!mount_cifs_root())
-			printk(KERN_ERR "VFS: Unable to mount root fs via SMB.\n");
-		return;
-	}
-#endif
-	if (ROOT_DEV == 0 && root_device_name && root_fs_names) {
-		if (mount_nodev_root() == 0)
-			return;
-	}
 #ifdef CONFIG_BLOCK
-	{
-		int err = create_dev("/dev/root", ROOT_DEV);
+static void __init mount_block_root(void)
+{
+	int err = create_dev("/dev/root", ROOT_DEV);
+
+	if (err < 0)
+		pr_emerg("Failed to create /dev/root: %d\n", err);
+	mount_root_generic("/dev/root", root_mountflags);
+}
+#else
+static inline void mount_block_root(void)
+{
+}
+#endif /* CONFIG_BLOCK */
 
-		if (err < 0)
-			pr_emerg("Failed to create /dev/root: %d\n", err);
-		mount_root_generic("/dev/root", root_mountflags);
+void __init mount_root(void)
+{
+	switch (ROOT_DEV) {
+	case Root_NFS:
+		mount_nfs_root();
+		break;
+	case Root_CIFS:
+		mount_cifs_root();
+		break;
+	case 0:
+		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+			break;
+		fallthrough;
+	default:
+		mount_block_root();
+		break;
 	}
-#endif
 }
 
 /*
-- 
2.39.2

[dm-devel] [PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

Instead of declaring root_device_name as a global variable pass it as an
argument to the functions using it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 29 ++++++++++++++++-------------
 init/do_mounts.h        | 14 +++++++-------
 init/do_mounts_initrd.c | 11 ++++++-----
 3 files changed, 29 insertions(+), 25 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index e708b02d9d6566..1405ee7218bf00 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -28,7 +28,6 @@
 #include "do_mounts.h"
 
 int root_mountflags = MS_RDONLY | MS_SILENT;
-static char * __initdata root_device_name;
 static char __initdata saved_root_name[64];
 static int root_wait;
 
@@ -391,7 +390,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_root_generic(char *name, int flags)
+void __init mount_root_generic(char *name, char *pretty_name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -425,7 +424,7 @@ void __init mount_root_generic(char *name, int flags)
 		 * and give them a list of the available devices
 		 */
 		printk("VFS: Cannot open root device \"%s\" or %s: error %d\n",
-				root_device_name, b, err);
+				pretty_name, b, err);
 		printk("Please append a correct \"root=\" boot option; here are the available partitions:\n");
 
 		printk_all_partitions();
@@ -541,7 +540,7 @@ static bool __init fs_is_nodev(char *fstype)
 	return ret;
 }
 
-static int __init mount_nodev_root(void)
+static int __init mount_nodev_root(char *root_device_name)
 {
 	char *fs_names, *fstype;
 	int err = -EINVAL;
@@ -569,21 +568,21 @@ static int __init mount_nodev_root(void)
 }
 
 #ifdef CONFIG_BLOCK
-static void __init mount_block_root(void)
+static void __init mount_block_root(char *root_device_name)
 {
 	int err = create_dev("/dev/root", ROOT_DEV);
 
 	if (err < 0)
 		pr_emerg("Failed to create /dev/root: %d\n", err);
-	mount_root_generic("/dev/root", root_mountflags);
+	mount_root_generic("/dev/root", root_device_name, root_mountflags);
 }
 #else
-static inline void mount_block_root(void)
+static inline void mount_block_root(char *root_device_name)
 {
 }
 #endif /* CONFIG_BLOCK */
 
-void __init mount_root(void)
+void __init mount_root(char *root_device_name)
 {
 	switch (ROOT_DEV) {
 	case Root_NFS:
@@ -593,11 +592,12 @@ void __init mount_root(void)
 		mount_cifs_root();
 		break;
 	case 0:
-		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+		if (root_device_name && root_fs_names &&
+		    mount_nodev_root(root_device_name) == 0)
 			break;
 		fallthrough;
 	default:
-		mount_block_root();
+		mount_block_root(root_device_name);
 		break;
 	}
 }
@@ -607,6 +607,8 @@ void __init mount_root(void)
  */
 void __init prepare_namespace(void)
 {
+	char *root_device_name;
+
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -628,7 +630,8 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_device_name,
+					   root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
@@ -636,7 +639,7 @@ void __init prepare_namespace(void)
 			root_device_name += 5;
 	}
 
-	if (initrd_load())
+	if (initrd_load(root_device_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -649,7 +652,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root();
+	mount_root(root_device_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 33623025f6951a..15e372b00ce704 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,8 +10,8 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_root_generic(char *name, int flags);
-void  mount_root(void);
+void  mount_root_generic(char *name, char *pretty_name, int flags);
+void  mount_root(char *root_device_name);
 extern int root_mountflags;
 
 static inline __init int create_dev(char *name, dev_t dev)
@@ -33,11 +33,11 @@ static inline int rd_load_image(char *from) { return 0; }
 #endif
 
 #ifdef CONFIG_BLK_DEV_INITRD
-
-bool __init initrd_load(void);
-
+bool __init initrd_load(char *root_device_name);
 #else
-
-static inline bool initrd_load(void) { return false; }
+static inline bool initrd_load(char *root_device_name)
+{
+	return false;
+	}
 
 #endif
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 686d1ff3af4bb1..425f4bcf4b77e0 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -83,7 +83,7 @@ static int __init init_linuxrc(struct subprocess_info *info, struct cred *new)
 	return 0;
 }
 
-static void __init handle_initrd(void)
+static void __init handle_initrd(char *root_device_name)
 {
 	struct subprocess_info *info;
 	static char *argv[] = { "linuxrc", NULL, };
@@ -95,7 +95,8 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_device_name,
+			   root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
@@ -117,7 +118,7 @@ static void __init handle_initrd(void)
 
 	init_chdir("/");
 	ROOT_DEV = new_decode_dev(real_root_dev);
-	mount_root();
+	mount_root(root_device_name);
 
 	printk(KERN_NOTICE "Trying to move old root to /initrd ... ");
 	error = init_mount("/old", "/root/initrd", NULL, MS_MOVE, NULL);
@@ -133,7 +134,7 @@ static void __init handle_initrd(void)
 	}
 }
 
-bool __init initrd_load(void)
+bool __init initrd_load(char *root_device_name)
 {
 	if (mount_initrd) {
 		create_dev("/dev/ram", Root_RAM0);
@@ -145,7 +146,7 @@ bool __init initrd_load(void)
 		 */
 		if (rd_load_image("/initrd.image") && ROOT_DEV != Root_RAM0) {
 			init_unlink("/initrd.image");
-			handle_initrd();
+			handle_initrd(root_device_name);
 			return true;
 		}
 	}
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

Instead of declaring root_device_name as a global variable pass it as an
argument to the functions using it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 29 ++++++++++++++++-------------
 init/do_mounts.h        | 14 +++++++-------
 init/do_mounts_initrd.c | 11 ++++++-----
 3 files changed, 29 insertions(+), 25 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index e708b02d9d6566..1405ee7218bf00 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -28,7 +28,6 @@
 #include "do_mounts.h"
 
 int root_mountflags = MS_RDONLY | MS_SILENT;
-static char * __initdata root_device_name;
 static char __initdata saved_root_name[64];
 static int root_wait;
 
@@ -391,7 +390,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_root_generic(char *name, int flags)
+void __init mount_root_generic(char *name, char *pretty_name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -425,7 +424,7 @@ void __init mount_root_generic(char *name, int flags)
 		 * and give them a list of the available devices
 		 */
 		printk("VFS: Cannot open root device \"%s\" or %s: error %d\n",
-				root_device_name, b, err);
+				pretty_name, b, err);
 		printk("Please append a correct \"root=\" boot option; here are the available partitions:\n");
 
 		printk_all_partitions();
@@ -541,7 +540,7 @@ static bool __init fs_is_nodev(char *fstype)
 	return ret;
 }
 
-static int __init mount_nodev_root(void)
+static int __init mount_nodev_root(char *root_device_name)
 {
 	char *fs_names, *fstype;
 	int err = -EINVAL;
@@ -569,21 +568,21 @@ static int __init mount_nodev_root(void)
 }
 
 #ifdef CONFIG_BLOCK
-static void __init mount_block_root(void)
+static void __init mount_block_root(char *root_device_name)
 {
 	int err = create_dev("/dev/root", ROOT_DEV);
 
 	if (err < 0)
 		pr_emerg("Failed to create /dev/root: %d\n", err);
-	mount_root_generic("/dev/root", root_mountflags);
+	mount_root_generic("/dev/root", root_device_name, root_mountflags);
 }
 #else
-static inline void mount_block_root(void)
+static inline void mount_block_root(char *root_device_name)
 {
 }
 #endif /* CONFIG_BLOCK */
 
-void __init mount_root(void)
+void __init mount_root(char *root_device_name)
 {
 	switch (ROOT_DEV) {
 	case Root_NFS:
@@ -593,11 +592,12 @@ void __init mount_root(void)
 		mount_cifs_root();
 		break;
 	case 0:
-		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+		if (root_device_name && root_fs_names &&
+		    mount_nodev_root(root_device_name) == 0)
 			break;
 		fallthrough;
 	default:
-		mount_block_root();
+		mount_block_root(root_device_name);
 		break;
 	}
 }
@@ -607,6 +607,8 @@ void __init mount_root(void)
  */
 void __init prepare_namespace(void)
 {
+	char *root_device_name;
+
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -628,7 +630,8 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_device_name,
+					   root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
@@ -636,7 +639,7 @@ void __init prepare_namespace(void)
 			root_device_name += 5;
 	}
 
-	if (initrd_load())
+	if (initrd_load(root_device_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -649,7 +652,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root();
+	mount_root(root_device_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 33623025f6951a..15e372b00ce704 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,8 +10,8 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_root_generic(char *name, int flags);
-void  mount_root(void);
+void  mount_root_generic(char *name, char *pretty_name, int flags);
+void  mount_root(char *root_device_name);
 extern int root_mountflags;
 
 static inline __init int create_dev(char *name, dev_t dev)
@@ -33,11 +33,11 @@ static inline int rd_load_image(char *from) { return 0; }
 #endif
 
 #ifdef CONFIG_BLK_DEV_INITRD
-
-bool __init initrd_load(void);
-
+bool __init initrd_load(char *root_device_name);
 #else
-
-static inline bool initrd_load(void) { return false; }
+static inline bool initrd_load(char *root_device_name)
+{
+	return false;
+	}
 
 #endif
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 686d1ff3af4bb1..425f4bcf4b77e0 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -83,7 +83,7 @@ static int __init init_linuxrc(struct subprocess_info *info, struct cred *new)
 	return 0;
 }
 
-static void __init handle_initrd(void)
+static void __init handle_initrd(char *root_device_name)
 {
 	struct subprocess_info *info;
 	static char *argv[] = { "linuxrc", NULL, };
@@ -95,7 +95,8 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_device_name,
+			   root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
@@ -117,7 +118,7 @@ static void __init handle_initrd(void)
 
 	init_chdir("/");
 	ROOT_DEV = new_decode_dev(real_root_dev);
-	mount_root();
+	mount_root(root_device_name);
 
 	printk(KERN_NOTICE "Trying to move old root to /initrd ... ");
 	error = init_mount("/old", "/root/initrd", NULL, MS_MOVE, NULL);
@@ -133,7 +134,7 @@ static void __init handle_initrd(void)
 	}
 }
 
-bool __init initrd_load(void)
+bool __init initrd_load(char *root_device_name)
 {
 	if (mount_initrd) {
 		create_dev("/dev/ram", Root_RAM0);
@@ -145,7 +146,7 @@ bool __init initrd_load(void)
 		 */
 		if (rd_load_image("/initrd.image") && ROOT_DEV != Root_RAM0) {
 			init_unlink("/initrd.image");
-			handle_initrd();
+			handle_initrd(root_device_name);
 			return true;
 		}
 	}
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

Instead of declaring root_device_name as a global variable pass it as an
argument to the functions using it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c        | 29 ++++++++++++++++-------------
 init/do_mounts.h        | 14 +++++++-------
 init/do_mounts_initrd.c | 11 ++++++-----
 3 files changed, 29 insertions(+), 25 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index e708b02d9d6566..1405ee7218bf00 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -28,7 +28,6 @@
 #include "do_mounts.h"
 
 int root_mountflags = MS_RDONLY | MS_SILENT;
-static char * __initdata root_device_name;
 static char __initdata saved_root_name[64];
 static int root_wait;
 
@@ -391,7 +390,7 @@ static int __init do_mount_root(const char *name, const char *fs,
 	return ret;
 }
 
-void __init mount_root_generic(char *name, int flags)
+void __init mount_root_generic(char *name, char *pretty_name, int flags)
 {
 	struct page *page = alloc_page(GFP_KERNEL);
 	char *fs_names = page_address(page);
@@ -425,7 +424,7 @@ void __init mount_root_generic(char *name, int flags)
 		 * and give them a list of the available devices
 		 */
 		printk("VFS: Cannot open root device \"%s\" or %s: error %d\n",
-				root_device_name, b, err);
+				pretty_name, b, err);
 		printk("Please append a correct \"root=\" boot option; here are the available partitions:\n");
 
 		printk_all_partitions();
@@ -541,7 +540,7 @@ static bool __init fs_is_nodev(char *fstype)
 	return ret;
 }
 
-static int __init mount_nodev_root(void)
+static int __init mount_nodev_root(char *root_device_name)
 {
 	char *fs_names, *fstype;
 	int err = -EINVAL;
@@ -569,21 +568,21 @@ static int __init mount_nodev_root(void)
 }
 
 #ifdef CONFIG_BLOCK
-static void __init mount_block_root(void)
+static void __init mount_block_root(char *root_device_name)
 {
 	int err = create_dev("/dev/root", ROOT_DEV);
 
 	if (err < 0)
 		pr_emerg("Failed to create /dev/root: %d\n", err);
-	mount_root_generic("/dev/root", root_mountflags);
+	mount_root_generic("/dev/root", root_device_name, root_mountflags);
 }
 #else
-static inline void mount_block_root(void)
+static inline void mount_block_root(char *root_device_name)
 {
 }
 #endif /* CONFIG_BLOCK */
 
-void __init mount_root(void)
+void __init mount_root(char *root_device_name)
 {
 	switch (ROOT_DEV) {
 	case Root_NFS:
@@ -593,11 +592,12 @@ void __init mount_root(void)
 		mount_cifs_root();
 		break;
 	case 0:
-		if (root_device_name && root_fs_names && mount_nodev_root() == 0)
+		if (root_device_name && root_fs_names &&
+		    mount_nodev_root(root_device_name) == 0)
 			break;
 		fallthrough;
 	default:
-		mount_block_root();
+		mount_block_root(root_device_name);
 		break;
 	}
 }
@@ -607,6 +607,8 @@ void __init mount_root(void)
  */
 void __init prepare_namespace(void)
 {
+	char *root_device_name;
+
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -628,7 +630,8 @@ void __init prepare_namespace(void)
 		root_device_name = saved_root_name;
 		if (!strncmp(root_device_name, "mtd", 3) ||
 		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_mountflags);
+			mount_root_generic(root_device_name, root_device_name,
+					   root_mountflags);
 			goto out;
 		}
 		ROOT_DEV = name_to_dev_t(root_device_name);
@@ -636,7 +639,7 @@ void __init prepare_namespace(void)
 			root_device_name += 5;
 	}
 
-	if (initrd_load())
+	if (initrd_load(root_device_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -649,7 +652,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root();
+	mount_root(root_device_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 33623025f6951a..15e372b00ce704 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -10,8 +10,8 @@
 #include <linux/root_dev.h>
 #include <linux/init_syscalls.h>
 
-void  mount_root_generic(char *name, int flags);
-void  mount_root(void);
+void  mount_root_generic(char *name, char *pretty_name, int flags);
+void  mount_root(char *root_device_name);
 extern int root_mountflags;
 
 static inline __init int create_dev(char *name, dev_t dev)
@@ -33,11 +33,11 @@ static inline int rd_load_image(char *from) { return 0; }
 #endif
 
 #ifdef CONFIG_BLK_DEV_INITRD
-
-bool __init initrd_load(void);
-
+bool __init initrd_load(char *root_device_name);
 #else
-
-static inline bool initrd_load(void) { return false; }
+static inline bool initrd_load(char *root_device_name)
+{
+	return false;
+	}
 
 #endif
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index 686d1ff3af4bb1..425f4bcf4b77e0 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -83,7 +83,7 @@ static int __init init_linuxrc(struct subprocess_info *info, struct cred *new)
 	return 0;
 }
 
-static void __init handle_initrd(void)
+static void __init handle_initrd(char *root_device_name)
 {
 	struct subprocess_info *info;
 	static char *argv[] = { "linuxrc", NULL, };
@@ -95,7 +95,8 @@ static void __init handle_initrd(void)
 	real_root_dev = new_encode_dev(ROOT_DEV);
 	create_dev("/dev/root.old", Root_RAM0);
 	/* mount initrd on rootfs' /root */
-	mount_root_generic("/dev/root.old", root_mountflags & ~MS_RDONLY);
+	mount_root_generic("/dev/root.old", root_device_name,
+			   root_mountflags & ~MS_RDONLY);
 	init_mkdir("/old", 0700);
 	init_chdir("/old");
 
@@ -117,7 +118,7 @@ static void __init handle_initrd(void)
 
 	init_chdir("/");
 	ROOT_DEV = new_decode_dev(real_root_dev);
-	mount_root();
+	mount_root(root_device_name);
 
 	printk(KERN_NOTICE "Trying to move old root to /initrd ... ");
 	error = init_mount("/old", "/root/initrd", NULL, MS_MOVE, NULL);
@@ -133,7 +134,7 @@ static void __init handle_initrd(void)
 	}
 }
 
-bool __init initrd_load(void)
+bool __init initrd_load(char *root_device_name)
 {
 	if (mount_initrd) {
 		create_dev("/dev/ram", Root_RAM0);
@@ -145,7 +146,7 @@ bool __init initrd_load(void)
 		 */
 		if (rd_load_image("/initrd.image") && ROOT_DEV != Root_RAM0) {
 			init_unlink("/initrd.image");
-			handle_initrd();
+			handle_initrd(root_device_name);
 			return true;
 		}
 	}
-- 
2.39.2

Re: [dm-devel] [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

Hi,

On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> Instead of declaring root_device_name as a global variable pass it as an
> argument to the functions using it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>

This patch results in the following build error when trying to build
xtensa:tinyconfig.

WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
ERROR: modpost: Section mismatches detected.

Unfortunately, reverting it is not possible due to conflicts,
so I can not confirm the bisect results.

Bisect log attached.

Guenter

---
# bad: [8d2be868b42c08290509c60515865f4de24ea704] Add linux-next specific files for 20230623
# good: [45a3e24f65e90a047bef86f927ebdc4c710edaa1] Linux 6.4-rc7
git bisect start 'HEAD' 'v6.4-rc7'
# good: [a5838c78db6a3a02e8d221e588c948f792e7f256] Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git
git bisect good a5838c78db6a3a02e8d221e588c948f792e7f256
# bad: [cca41cc0b5485a0ec20707316c1a00082c01a2af] Merge branch 'for-next' of git://git.kernel.dk/linux-block.git
git bisect bad cca41cc0b5485a0ec20707316c1a00082c01a2af
# good: [901bdf5ea1a836400ee69aa32b04e9c209271ec7] Merge tag 'amd-drm-next-6.5-2023-06-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
git bisect good 901bdf5ea1a836400ee69aa32b04e9c209271ec7
# good: [b4666c320b8113d94b3f4624054562e7add57e4a] Merge branch 'for-next' of https://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394.git
git bisect good b4666c320b8113d94b3f4624054562e7add57e4a
# good: [b2c28785b125acb28a681462510297410cbbabd7] ASoC: dt-bindings: microchip,sama7g5-pdmc: Simplify "microchip,mic-pos" constraints
git bisect good b2c28785b125acb28a681462510297410cbbabd7
# bad: [9d217fb0e778d69b2e3988efbc441976c0fb29b5] nvme: reorder fields in 'struct nvme_ctrl'
git bisect bad 9d217fb0e778d69b2e3988efbc441976c0fb29b5
# good: [20d099756b98fa6b5b838448b1ffbce46f4f3283] block: Replace all non-returning strlcpy with strscpy
git bisect good 20d099756b98fa6b5b838448b1ffbce46f4f3283
# bad: [93c8f6f38be67e30adf8d8eb5e7e9ccb89326119] pktcdvd: Drop redundant castings for sector_t
git bisect bad 93c8f6f38be67e30adf8d8eb5e7e9ccb89326119
# bad: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly
git bisect bad c8643c72bc42781fc169c6498a3902bec447099e
# good: [87efb39075be6a288cd7f23858f15bd01c83028a] fs: add a method to shut down the file system
git bisect good 87efb39075be6a288cd7f23858f15bd01c83028a
# good: [aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d] driver core: return bool from driver_probe_done
git bisect good aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
# good: [cc89c63e2fe37d476357c82390dfb12edcd41cdd] PM: hibernate: move finding the resume device out of software_resume
git bisect good cc89c63e2fe37d476357c82390dfb12edcd41cdd
# good: [e3102722ffe77094ba9e7e46380792b3dd8a7abd] init: rename mount_block_root to mount_root_generic
git bisect good e3102722ffe77094ba9e7e46380792b3dd8a7abd
# good: [a6a41d39c2d91ff2543d31b6cc6070f3957e3aea] init: refactor mount_root
git bisect good a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
# first bad commit: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

Hi,

On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> Instead of declaring root_device_name as a global variable pass it as an
> argument to the functions using it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>

This patch results in the following build error when trying to build
xtensa:tinyconfig.

WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
ERROR: modpost: Section mismatches detected.

Unfortunately, reverting it is not possible due to conflicts,
so I can not confirm the bisect results.

Bisect log attached.

Guenter

---
# bad: [8d2be868b42c08290509c60515865f4de24ea704] Add linux-next specific files for 20230623
# good: [45a3e24f65e90a047bef86f927ebdc4c710edaa1] Linux 6.4-rc7
git bisect start 'HEAD' 'v6.4-rc7'
# good: [a5838c78db6a3a02e8d221e588c948f792e7f256] Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git
git bisect good a5838c78db6a3a02e8d221e588c948f792e7f256
# bad: [cca41cc0b5485a0ec20707316c1a00082c01a2af] Merge branch 'for-next' of git://git.kernel.dk/linux-block.git
git bisect bad cca41cc0b5485a0ec20707316c1a00082c01a2af
# good: [901bdf5ea1a836400ee69aa32b04e9c209271ec7] Merge tag 'amd-drm-next-6.5-2023-06-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
git bisect good 901bdf5ea1a836400ee69aa32b04e9c209271ec7
# good: [b4666c320b8113d94b3f4624054562e7add57e4a] Merge branch 'for-next' of https://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394.git
git bisect good b4666c320b8113d94b3f4624054562e7add57e4a
# good: [b2c28785b125acb28a681462510297410cbbabd7] ASoC: dt-bindings: microchip,sama7g5-pdmc: Simplify "microchip,mic-pos" constraints
git bisect good b2c28785b125acb28a681462510297410cbbabd7
# bad: [9d217fb0e778d69b2e3988efbc441976c0fb29b5] nvme: reorder fields in 'struct nvme_ctrl'
git bisect bad 9d217fb0e778d69b2e3988efbc441976c0fb29b5
# good: [20d099756b98fa6b5b838448b1ffbce46f4f3283] block: Replace all non-returning strlcpy with strscpy
git bisect good 20d099756b98fa6b5b838448b1ffbce46f4f3283
# bad: [93c8f6f38be67e30adf8d8eb5e7e9ccb89326119] pktcdvd: Drop redundant castings for sector_t
git bisect bad 93c8f6f38be67e30adf8d8eb5e7e9ccb89326119
# bad: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly
git bisect bad c8643c72bc42781fc169c6498a3902bec447099e
# good: [87efb39075be6a288cd7f23858f15bd01c83028a] fs: add a method to shut down the file system
git bisect good 87efb39075be6a288cd7f23858f15bd01c83028a
# good: [aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d] driver core: return bool from driver_probe_done
git bisect good aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
# good: [cc89c63e2fe37d476357c82390dfb12edcd41cdd] PM: hibernate: move finding the resume device out of software_resume
git bisect good cc89c63e2fe37d476357c82390dfb12edcd41cdd
# good: [e3102722ffe77094ba9e7e46380792b3dd8a7abd] init: rename mount_block_root to mount_root_generic
git bisect good e3102722ffe77094ba9e7e46380792b3dd8a7abd
# good: [a6a41d39c2d91ff2543d31b6cc6070f3957e3aea] init: refactor mount_root
git bisect good a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
# first bad commit: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

Hi,

On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> Instead of declaring root_device_name as a global variable pass it as an
> argument to the functions using it.
> 
> Signed-off-by: Christoph Hellwig <hch@lst.de>

This patch results in the following build error when trying to build
xtensa:tinyconfig.

WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
ERROR: modpost: Section mismatches detected.

Unfortunately, reverting it is not possible due to conflicts,
so I can not confirm the bisect results.

Bisect log attached.

Guenter

---
# bad: [8d2be868b42c08290509c60515865f4de24ea704] Add linux-next specific files for 20230623
# good: [45a3e24f65e90a047bef86f927ebdc4c710edaa1] Linux 6.4-rc7
git bisect start 'HEAD' 'v6.4-rc7'
# good: [a5838c78db6a3a02e8d221e588c948f792e7f256] Merge branch 'for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next.git
git bisect good a5838c78db6a3a02e8d221e588c948f792e7f256
# bad: [cca41cc0b5485a0ec20707316c1a00082c01a2af] Merge branch 'for-next' of git://git.kernel.dk/linux-block.git
git bisect bad cca41cc0b5485a0ec20707316c1a00082c01a2af
# good: [901bdf5ea1a836400ee69aa32b04e9c209271ec7] Merge tag 'amd-drm-next-6.5-2023-06-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next
git bisect good 901bdf5ea1a836400ee69aa32b04e9c209271ec7
# good: [b4666c320b8113d94b3f4624054562e7add57e4a] Merge branch 'for-next' of https://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394.git
git bisect good b4666c320b8113d94b3f4624054562e7add57e4a
# good: [b2c28785b125acb28a681462510297410cbbabd7] ASoC: dt-bindings: microchip,sama7g5-pdmc: Simplify "microchip,mic-pos" constraints
git bisect good b2c28785b125acb28a681462510297410cbbabd7
# bad: [9d217fb0e778d69b2e3988efbc441976c0fb29b5] nvme: reorder fields in 'struct nvme_ctrl'
git bisect bad 9d217fb0e778d69b2e3988efbc441976c0fb29b5
# good: [20d099756b98fa6b5b838448b1ffbce46f4f3283] block: Replace all non-returning strlcpy with strscpy
git bisect good 20d099756b98fa6b5b838448b1ffbce46f4f3283
# bad: [93c8f6f38be67e30adf8d8eb5e7e9ccb89326119] pktcdvd: Drop redundant castings for sector_t
git bisect bad 93c8f6f38be67e30adf8d8eb5e7e9ccb89326119
# bad: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly
git bisect bad c8643c72bc42781fc169c6498a3902bec447099e
# good: [87efb39075be6a288cd7f23858f15bd01c83028a] fs: add a method to shut down the file system
git bisect good 87efb39075be6a288cd7f23858f15bd01c83028a
# good: [aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d] driver core: return bool from driver_probe_done
git bisect good aa5f6ed8c21ec1aa5fd688118d8d5cd87c5ffc1d
# good: [cc89c63e2fe37d476357c82390dfb12edcd41cdd] PM: hibernate: move finding the resume device out of software_resume
git bisect good cc89c63e2fe37d476357c82390dfb12edcd41cdd
# good: [e3102722ffe77094ba9e7e46380792b3dd8a7abd] init: rename mount_block_root to mount_root_generic
git bisect good e3102722ffe77094ba9e7e46380792b3dd8a7abd
# good: [a6a41d39c2d91ff2543d31b6cc6070f3957e3aea] init: refactor mount_root
git bisect good a6a41d39c2d91ff2543d31b6cc6070f3957e3aea
# first bad commit: [c8643c72bc42781fc169c6498a3902bec447099e] init: pass root_device_name explicitly

Re: [dm-devel] [PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> Hi,
> 
> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> > Instead of declaring root_device_name as a global variable pass it as an
> > argument to the functions using it.
> > 
> > Signed-off-by: Christoph Hellwig <hch@lst.de>
> 
> This patch results in the following build error when trying to build
> xtensa:tinyconfig.
> 
> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> ERROR: modpost: Section mismatches detected.

I can reproduce these with gcc 13.1 on xtensa, but the report makes
no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
a similar warning for put_page intead of strcpy which seems just as
arcance. 

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> Hi,
> 
> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> > Instead of declaring root_device_name as a global variable pass it as an
> > argument to the functions using it.
> > 
> > Signed-off-by: Christoph Hellwig <hch@lst.de>
> 
> This patch results in the following build error when trying to build
> xtensa:tinyconfig.
> 
> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> ERROR: modpost: Section mismatches detected.

I can reproduce these with gcc 13.1 on xtensa, but the report makes
no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
a similar warning for put_page intead of strcpy which seems just as
arcance. 


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Christoph Hellwig]

On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> Hi,
> 
> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> > Instead of declaring root_device_name as a global variable pass it as an
> > argument to the functions using it.
> > 
> > Signed-off-by: Christoph Hellwig <hch@lst.de>
> 
> This patch results in the following build error when trying to build
> xtensa:tinyconfig.
> 
> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> ERROR: modpost: Section mismatches detected.

I can reproduce these with gcc 13.1 on xtensa, but the report makes
no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
a similar warning for put_page intead of strcpy which seems just as
arcance. 

Re: [dm-devel] [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

On 6/26/23 00:53, Christoph Hellwig wrote:
> On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
>> Hi,
>>
>> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
>>> Instead of declaring root_device_name as a global variable pass it as an
>>> argument to the functions using it.
>>>
>>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>>
>> This patch results in the following build error when trying to build
>> xtensa:tinyconfig.
>>
>> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
>> ERROR: modpost: Section mismatches detected.
> 
> I can reproduce these with gcc 13.1 on xtensa, but the report makes
> no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> a similar warning for put_page intead of strcpy which seems just as
> arcance.
> 

I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
for xtensa builds instead.

Guenter

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

On 6/26/23 00:53, Christoph Hellwig wrote:
> On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
>> Hi,
>>
>> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
>>> Instead of declaring root_device_name as a global variable pass it as an
>>> argument to the functions using it.
>>>
>>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>>
>> This patch results in the following build error when trying to build
>> xtensa:tinyconfig.
>>
>> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
>> ERROR: modpost: Section mismatches detected.
> 
> I can reproduce these with gcc 13.1 on xtensa, but the report makes
> no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> a similar warning for put_page intead of strcpy which seems just as
> arcance.
> 

I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
for xtensa builds instead.

Guenter


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Guenter Roeck]

On 6/26/23 00:53, Christoph Hellwig wrote:
> On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
>> Hi,
>>
>> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
>>> Instead of declaring root_device_name as a global variable pass it as an
>>> argument to the functions using it.
>>>
>>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>>
>> This patch results in the following build error when trying to build
>> xtensa:tinyconfig.
>>
>> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
>> ERROR: modpost: Section mismatches detected.
> 
> I can reproduce these with gcc 13.1 on xtensa, but the report makes
> no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> a similar warning for put_page intead of strcpy which seems just as
> arcance.
> 

I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
for xtensa builds instead.

Guenter

Re: [dm-devel] [PATCH 08/24] init: pass root_device_name explicitly

[Max Filippov]

On Mon, Jun 26, 2023 at 8:10 AM Guenter Roeck <linux@roeck-us.net> wrote:
>
> On 6/26/23 00:53, Christoph Hellwig wrote:
> > On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> >> Hi,
> >>
> >> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> >>> Instead of declaring root_device_name as a global variable pass it as an
> >>> argument to the functions using it.
> >>>
> >>> Signed-off-by: Christoph Hellwig <hch@lst.de>
> >>
> >> This patch results in the following build error when trying to build
> >> xtensa:tinyconfig.
> >>
> >> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> >> ERROR: modpost: Section mismatches detected.
> >
> > I can reproduce these with gcc 13.1 on xtensa, but the report makes
> > no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> > a similar warning for put_page intead of strcpy which seems just as
> > arcance.
> >
>
> I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
> if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
> for xtensa builds instead.

I believe it's yet another manifestation of the following issue:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92938

Hunting is still on my todo list, but it's quite low, so I guess
forcing CONFIG_SECTION_MISMATCH_WARN_ONLY=y for xtensa
is the right thing to do for now.

-- 
Thanks.
-- Max

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Max Filippov]

On Mon, Jun 26, 2023 at 8:10 AM Guenter Roeck <linux@roeck-us.net> wrote:
>
> On 6/26/23 00:53, Christoph Hellwig wrote:
> > On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> >> Hi,
> >>
> >> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> >>> Instead of declaring root_device_name as a global variable pass it as an
> >>> argument to the functions using it.
> >>>
> >>> Signed-off-by: Christoph Hellwig <hch@lst.de>
> >>
> >> This patch results in the following build error when trying to build
> >> xtensa:tinyconfig.
> >>
> >> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> >> ERROR: modpost: Section mismatches detected.
> >
> > I can reproduce these with gcc 13.1 on xtensa, but the report makes
> > no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> > a similar warning for put_page intead of strcpy which seems just as
> > arcance.
> >
>
> I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
> if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
> for xtensa builds instead.

I believe it's yet another manifestation of the following issue:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92938

Hunting is still on my todo list, but it's quite low, so I guess
forcing CONFIG_SECTION_MISMATCH_WARN_ONLY=y for xtensa
is the right thing to do for now.

-- 
Thanks.
-- Max

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

Re: [PATCH 08/24] init: pass root_device_name explicitly

[Max Filippov]

On Mon, Jun 26, 2023 at 8:10 AM Guenter Roeck <linux@roeck-us.net> wrote:
>
> On 6/26/23 00:53, Christoph Hellwig wrote:
> > On Fri, Jun 23, 2023 at 05:08:59PM -0700, Guenter Roeck wrote:
> >> Hi,
> >>
> >> On Wed, May 31, 2023 at 02:55:19PM +0200, Christoph Hellwig wrote:
> >>> Instead of declaring root_device_name as a global variable pass it as an
> >>> argument to the functions using it.
> >>>
> >>> Signed-off-by: Christoph Hellwig <hch@lst.de>
> >>
> >> This patch results in the following build error when trying to build
> >> xtensa:tinyconfig.
> >>
> >> WARNING: modpost: vmlinux: section mismatch in reference: strcpy.isra.0+0x14 (section: .text.unlikely) -> initcall_level_names (section: .init.data)
> >> ERROR: modpost: Section mismatches detected.
> >
> > I can reproduce these with gcc 13.1 on xtensa, but the report makes
> > no sense to me.  If I disable CONFIG_CC_OPTIMIZE_FOR_SIZE it now reports
> > a similar warning for put_page intead of strcpy which seems just as
> > arcance.
> >
>
> I don't see that (I tried 11.3, 11.4, 12.3, and 13.1), but then I am not sure
> if this is worth tracking down. I just force CONFIG_SECTION_MISMATCH_WARN_ONLY=y
> for xtensa builds instead.

I believe it's yet another manifestation of the following issue:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92938

Hunting is still on my todo list, but it's quite low, so I guess
forcing CONFIG_SECTION_MISMATCH_WARN_ONLY=y for xtensa
is the right thing to do for now.

-- 
Thanks.
-- Max

[dm-devel] [PATCH 09/24] init: don't remove the /dev/ prefix from error messages

[Christoph Hellwig]

Remove the code that drops the /dev/ prefix from root_device_name, which
is only used for error messages when mounting the root device fails.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 1405ee7218bf00..74cc96bffbdd71 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -607,8 +607,6 @@ void __init mount_root(char *root_device_name)
  */
 void __init prepare_namespace(void)
 {
-	char *root_device_name;
-
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -627,19 +625,16 @@ void __init prepare_namespace(void)
 	md_run_setup();
 
 	if (saved_root_name[0]) {
-		root_device_name = saved_root_name;
-		if (!strncmp(root_device_name, "mtd", 3) ||
-		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_device_name,
+		if (!strncmp(saved_root_name, "mtd", 3) ||
+		    !strncmp(saved_root_name, "ubi", 3)) {
+			mount_root_generic(saved_root_name, saved_root_name,
 					   root_mountflags);
 			goto out;
 		}
-		ROOT_DEV = name_to_dev_t(root_device_name);
-		if (strncmp(root_device_name, "/dev/", 5) == 0)
-			root_device_name += 5;
+		ROOT_DEV = name_to_dev_t(saved_root_name);
 	}
 
-	if (initrd_load(root_device_name))
+	if (initrd_load(saved_root_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -652,7 +647,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root(root_device_name);
+	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 09/24] init: don't remove the /dev/ prefix from error messages

[Christoph Hellwig]

Remove the code that drops the /dev/ prefix from root_device_name, which
is only used for error messages when mounting the root device fails.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 1405ee7218bf00..74cc96bffbdd71 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -607,8 +607,6 @@ void __init mount_root(char *root_device_name)
  */
 void __init prepare_namespace(void)
 {
-	char *root_device_name;
-
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -627,19 +625,16 @@ void __init prepare_namespace(void)
 	md_run_setup();
 
 	if (saved_root_name[0]) {
-		root_device_name = saved_root_name;
-		if (!strncmp(root_device_name, "mtd", 3) ||
-		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_device_name,
+		if (!strncmp(saved_root_name, "mtd", 3) ||
+		    !strncmp(saved_root_name, "ubi", 3)) {
+			mount_root_generic(saved_root_name, saved_root_name,
 					   root_mountflags);
 			goto out;
 		}
-		ROOT_DEV = name_to_dev_t(root_device_name);
-		if (strncmp(root_device_name, "/dev/", 5) == 0)
-			root_device_name += 5;
+		ROOT_DEV = name_to_dev_t(saved_root_name);
 	}
 
-	if (initrd_load(root_device_name))
+	if (initrd_load(saved_root_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -652,7 +647,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root(root_device_name);
+	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 09/24] init: don't remove the /dev/ prefix from error messages

[Christoph Hellwig]

Remove the code that drops the /dev/ prefix from root_device_name, which
is only used for error messages when mounting the root device fails.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index 1405ee7218bf00..74cc96bffbdd71 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -607,8 +607,6 @@ void __init mount_root(char *root_device_name)
  */
 void __init prepare_namespace(void)
 {
-	char *root_device_name;
-
 	if (root_delay) {
 		printk(KERN_INFO "Waiting %d sec before mounting root device...\n",
 		       root_delay);
@@ -627,19 +625,16 @@ void __init prepare_namespace(void)
 	md_run_setup();
 
 	if (saved_root_name[0]) {
-		root_device_name = saved_root_name;
-		if (!strncmp(root_device_name, "mtd", 3) ||
-		    !strncmp(root_device_name, "ubi", 3)) {
-			mount_root_generic(root_device_name, root_device_name,
+		if (!strncmp(saved_root_name, "mtd", 3) ||
+		    !strncmp(saved_root_name, "ubi", 3)) {
+			mount_root_generic(saved_root_name, saved_root_name,
 					   root_mountflags);
 			goto out;
 		}
-		ROOT_DEV = name_to_dev_t(root_device_name);
-		if (strncmp(root_device_name, "/dev/", 5) == 0)
-			root_device_name += 5;
+		ROOT_DEV = name_to_dev_t(saved_root_name);
 	}
 
-	if (initrd_load(root_device_name))
+	if (initrd_load(saved_root_name))
 		goto out;
 
 	/* wait for any asynchronous scanning to complete */
@@ -652,7 +647,7 @@ void __init prepare_namespace(void)
 		async_synchronize_full();
 	}
 
-	mount_root(root_device_name);
+	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
 	init_mount(".", "/", NULL, MS_MOVE, NULL);
-- 
2.39.2

[dm-devel] [PATCH 10/24] init: handle ubi/mtd root mounting like all other root types

[Christoph Hellwig]

Assign a Root_Generic magic value for UBI/MTD root and handle the root
mounting in mount_root like all other root types.  Besides making the
code more clear this also means that UBI/MTD root can be used together
with an initrd (not that anyone should care).

Also factor parsing of the root name into a helper now that it can
be easily done and will get more complicated with subsequent patches.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 include/linux/root_dev.h |  1 +
 init/do_mounts.c         | 23 ++++++++++++++---------
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index ed3ea8da642972..847c9a06101b76 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -9,6 +9,7 @@
 enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
+	Root_Generic = MKDEV(UNNAMED_MAJOR, 253),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
 };
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 74cc96bffbdd71..be6d14733ba02f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -591,6 +591,10 @@ void __init mount_root(char *root_device_name)
 	case Root_CIFS:
 		mount_cifs_root();
 		break;
+	case Root_Generic:
+		mount_root_generic(root_device_name, root_device_name,
+				   root_mountflags);
+		break;
 	case 0:
 		if (root_device_name && root_fs_names &&
 		    mount_nodev_root(root_device_name) == 0)
@@ -602,6 +606,14 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+static dev_t __init parse_root_device(char *root_device_name)
+{
+	if (!strncmp(root_device_name, "mtd", 3) ||
+	    !strncmp(root_device_name, "ubi", 3))
+		return Root_Generic;
+	return name_to_dev_t(root_device_name);
+}
+
 /*
  * Prepare the namespace - decide what/where to mount, load ramdisks, etc.
  */
@@ -624,15 +636,8 @@ void __init prepare_namespace(void)
 
 	md_run_setup();
 
-	if (saved_root_name[0]) {
-		if (!strncmp(saved_root_name, "mtd", 3) ||
-		    !strncmp(saved_root_name, "ubi", 3)) {
-			mount_root_generic(saved_root_name, saved_root_name,
-					   root_mountflags);
-			goto out;
-		}
-		ROOT_DEV = name_to_dev_t(saved_root_name);
-	}
+	if (saved_root_name[0])
+		ROOT_DEV = parse_root_device(saved_root_name);
 
 	if (initrd_load(saved_root_name))
 		goto out;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 10/24] init: handle ubi/mtd root mounting like all other root types

[Christoph Hellwig]

Assign a Root_Generic magic value for UBI/MTD root and handle the root
mounting in mount_root like all other root types.  Besides making the
code more clear this also means that UBI/MTD root can be used together
with an initrd (not that anyone should care).

Also factor parsing of the root name into a helper now that it can
be easily done and will get more complicated with subsequent patches.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 include/linux/root_dev.h |  1 +
 init/do_mounts.c         | 23 ++++++++++++++---------
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index ed3ea8da642972..847c9a06101b76 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -9,6 +9,7 @@
 enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
+	Root_Generic = MKDEV(UNNAMED_MAJOR, 253),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
 };
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 74cc96bffbdd71..be6d14733ba02f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -591,6 +591,10 @@ void __init mount_root(char *root_device_name)
 	case Root_CIFS:
 		mount_cifs_root();
 		break;
+	case Root_Generic:
+		mount_root_generic(root_device_name, root_device_name,
+				   root_mountflags);
+		break;
 	case 0:
 		if (root_device_name && root_fs_names &&
 		    mount_nodev_root(root_device_name) == 0)
@@ -602,6 +606,14 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+static dev_t __init parse_root_device(char *root_device_name)
+{
+	if (!strncmp(root_device_name, "mtd", 3) ||
+	    !strncmp(root_device_name, "ubi", 3))
+		return Root_Generic;
+	return name_to_dev_t(root_device_name);
+}
+
 /*
  * Prepare the namespace - decide what/where to mount, load ramdisks, etc.
  */
@@ -624,15 +636,8 @@ void __init prepare_namespace(void)
 
 	md_run_setup();
 
-	if (saved_root_name[0]) {
-		if (!strncmp(saved_root_name, "mtd", 3) ||
-		    !strncmp(saved_root_name, "ubi", 3)) {
-			mount_root_generic(saved_root_name, saved_root_name,
-					   root_mountflags);
-			goto out;
-		}
-		ROOT_DEV = name_to_dev_t(saved_root_name);
-	}
+	if (saved_root_name[0])
+		ROOT_DEV = parse_root_device(saved_root_name);
 
 	if (initrd_load(saved_root_name))
 		goto out;
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 10/24] init: handle ubi/mtd root mounting like all other root types

[Christoph Hellwig]

Assign a Root_Generic magic value for UBI/MTD root and handle the root
mounting in mount_root like all other root types.  Besides making the
code more clear this also means that UBI/MTD root can be used together
with an initrd (not that anyone should care).

Also factor parsing of the root name into a helper now that it can
be easily done and will get more complicated with subsequent patches.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 include/linux/root_dev.h |  1 +
 init/do_mounts.c         | 23 ++++++++++++++---------
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/include/linux/root_dev.h b/include/linux/root_dev.h
index ed3ea8da642972..847c9a06101b76 100644
--- a/include/linux/root_dev.h
+++ b/include/linux/root_dev.h
@@ -9,6 +9,7 @@
 enum {
 	Root_NFS = MKDEV(UNNAMED_MAJOR, 255),
 	Root_CIFS = MKDEV(UNNAMED_MAJOR, 254),
+	Root_Generic = MKDEV(UNNAMED_MAJOR, 253),
 	Root_RAM0 = MKDEV(RAMDISK_MAJOR, 0),
 };
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 74cc96bffbdd71..be6d14733ba02f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -591,6 +591,10 @@ void __init mount_root(char *root_device_name)
 	case Root_CIFS:
 		mount_cifs_root();
 		break;
+	case Root_Generic:
+		mount_root_generic(root_device_name, root_device_name,
+				   root_mountflags);
+		break;
 	case 0:
 		if (root_device_name && root_fs_names &&
 		    mount_nodev_root(root_device_name) == 0)
@@ -602,6 +606,14 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+static dev_t __init parse_root_device(char *root_device_name)
+{
+	if (!strncmp(root_device_name, "mtd", 3) ||
+	    !strncmp(root_device_name, "ubi", 3))
+		return Root_Generic;
+	return name_to_dev_t(root_device_name);
+}
+
 /*
  * Prepare the namespace - decide what/where to mount, load ramdisks, etc.
  */
@@ -624,15 +636,8 @@ void __init prepare_namespace(void)
 
 	md_run_setup();
 
-	if (saved_root_name[0]) {
-		if (!strncmp(saved_root_name, "mtd", 3) ||
-		    !strncmp(saved_root_name, "ubi", 3)) {
-			mount_root_generic(saved_root_name, saved_root_name,
-					   root_mountflags);
-			goto out;
-		}
-		ROOT_DEV = name_to_dev_t(saved_root_name);
-	}
+	if (saved_root_name[0])
+		ROOT_DEV = parse_root_device(saved_root_name);
 
 	if (initrd_load(saved_root_name))
 		goto out;
-- 
2.39.2

[dm-devel] [PATCH 11/24] init: factor the root_wait logic in prepare_namespace into a helper

[Christoph Hellwig]

The root_wait logic is a bit obsfucated right now.  Expand it and move it
into a helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index be6d14733ba02f..d5c06c1546e82c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -606,6 +606,26 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+/* wait for any asynchronous scanning to complete */
+static void __init wait_for_root(char *root_device_name)
+{
+	if (ROOT_DEV != 0)
+		return;
+
+	pr_info("Waiting for root device %s...\n", root_device_name);
+
+	for (;;) {
+		if (driver_probe_done()) {
+			ROOT_DEV = name_to_dev_t(root_device_name);
+			if (ROOT_DEV)
+				break;
+		}
+		msleep(5);
+	}
+	async_synchronize_full();
+
+}
+
 static dev_t __init parse_root_device(char *root_device_name)
 {
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -642,16 +662,8 @@ void __init prepare_namespace(void)
 	if (initrd_load(saved_root_name))
 		goto out;
 
-	/* wait for any asynchronous scanning to complete */
-	if ((ROOT_DEV == 0) && root_wait) {
-		printk(KERN_INFO "Waiting for root device %s...\n",
-			saved_root_name);
-		while (!driver_probe_done() ||
-			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
-			msleep(5);
-		async_synchronize_full();
-	}
-
+	if (root_wait)
+		wait_for_root(saved_root_name);
 	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 11/24] init: factor the root_wait logic in prepare_namespace into a helper

[Christoph Hellwig]

The root_wait logic is a bit obsfucated right now.  Expand it and move it
into a helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index be6d14733ba02f..d5c06c1546e82c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -606,6 +606,26 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+/* wait for any asynchronous scanning to complete */
+static void __init wait_for_root(char *root_device_name)
+{
+	if (ROOT_DEV != 0)
+		return;
+
+	pr_info("Waiting for root device %s...\n", root_device_name);
+
+	for (;;) {
+		if (driver_probe_done()) {
+			ROOT_DEV = name_to_dev_t(root_device_name);
+			if (ROOT_DEV)
+				break;
+		}
+		msleep(5);
+	}
+	async_synchronize_full();
+
+}
+
 static dev_t __init parse_root_device(char *root_device_name)
 {
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -642,16 +662,8 @@ void __init prepare_namespace(void)
 	if (initrd_load(saved_root_name))
 		goto out;
 
-	/* wait for any asynchronous scanning to complete */
-	if ((ROOT_DEV == 0) && root_wait) {
-		printk(KERN_INFO "Waiting for root device %s...\n",
-			saved_root_name);
-		while (!driver_probe_done() ||
-			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
-			msleep(5);
-		async_synchronize_full();
-	}
-
+	if (root_wait)
+		wait_for_root(saved_root_name);
 	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 11/24] init: factor the root_wait logic in prepare_namespace into a helper

[Christoph Hellwig]

The root_wait logic is a bit obsfucated right now.  Expand it and move it
into a helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 32 ++++++++++++++++++++++----------
 1 file changed, 22 insertions(+), 10 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index be6d14733ba02f..d5c06c1546e82c 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -606,6 +606,26 @@ void __init mount_root(char *root_device_name)
 	}
 }
 
+/* wait for any asynchronous scanning to complete */
+static void __init wait_for_root(char *root_device_name)
+{
+	if (ROOT_DEV != 0)
+		return;
+
+	pr_info("Waiting for root device %s...\n", root_device_name);
+
+	for (;;) {
+		if (driver_probe_done()) {
+			ROOT_DEV = name_to_dev_t(root_device_name);
+			if (ROOT_DEV)
+				break;
+		}
+		msleep(5);
+	}
+	async_synchronize_full();
+
+}
+
 static dev_t __init parse_root_device(char *root_device_name)
 {
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -642,16 +662,8 @@ void __init prepare_namespace(void)
 	if (initrd_load(saved_root_name))
 		goto out;
 
-	/* wait for any asynchronous scanning to complete */
-	if ((ROOT_DEV == 0) && root_wait) {
-		printk(KERN_INFO "Waiting for root device %s...\n",
-			saved_root_name);
-		while (!driver_probe_done() ||
-			(ROOT_DEV = name_to_dev_t(saved_root_name)) == 0)
-			msleep(5);
-		async_synchronize_full();
-	}
-
+	if (root_wait)
+		wait_for_root(saved_root_name);
 	mount_root(saved_root_name);
 out:
 	devtmpfs_mount();
-- 
2.39.2

[dm-devel] [PATCH 12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t

[Christoph Hellwig]

The nfs/cifs/ram special case only needs to be parsed once, and only in
the boot code.  Move them out of name_to_dev_t and into
prepare_namespace.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 Documentation/admin-guide/kernel-parameters.txt |  7 ++++++-
 init/do_mounts.c                                | 14 ++++++--------
 2 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 9e5bab29685ff0..457c342d15977e 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5452,7 +5452,12 @@
 			port and the regular usb controller gets disabled.
 
 	root=		[KNL] Root filesystem
-			See name_to_dev_t comment in init/do_mounts.c.
+			Usually this a a block device specifier of some kind,
+			see the name_to_dev_t comment in init/do_mounts.c for
+			details.
+			Alternatively this can be "ram" for the legacy initial
+			ramdisk, "nfs" and "cifs" for root on a network file
+			system, or "mtd" and "ubi" for mounting from raw flash.
 
 	rootdelay=	[KNL] Delay (in seconds) to pause before attempting to
 			mount the root filesystem
diff --git a/init/do_mounts.c b/init/do_mounts.c
index d5c06c1546e82c..86599faf2bf8a1 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -248,7 +248,6 @@ static dev_t devt_from_devnum(const char *name)
  *
  *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
  *         no leading 0x, for example b302.
- *	2) /dev/nfs represents Root_NFS (0xff)
  *	3) /dev/<disk_name> represents the device number of disk
  *	4) /dev/<disk_name><decimal> represents the device number
  *         of partition - device number of disk plus the partition number
@@ -266,7 +265,6 @@ static dev_t devt_from_devnum(const char *name)
  *	   a colon.
  *	9) PARTLABEL=<name> with name being the GPT partition label.
  *	   MSDOS partitions do not support labels!
- *	10) /dev/cifs represents Root_CIFS (0xfe)
  *
  *	If name doesn't have fall into the categories above, we return (0,0).
  *	block_class is used to check if something is a disk name. If the disk
@@ -275,12 +273,6 @@ static dev_t devt_from_devnum(const char *name)
  */
 dev_t name_to_dev_t(const char *name)
 {
-	if (strcmp(name, "/dev/nfs") == 0)
-		return Root_NFS;
-	if (strcmp(name, "/dev/cifs") == 0)
-		return Root_CIFS;
-	if (strcmp(name, "/dev/ram") == 0)
-		return Root_RAM0;
 #ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9);
@@ -631,6 +623,12 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
+	if (strcmp(root_device_name, "/dev/nfs") == 0)
+		return Root_NFS;
+	if (strcmp(root_device_name, "/dev/cifs") == 0)
+		return Root_CIFS;
+	if (strcmp(root_device_name, "/dev/ram") == 0)
+		return Root_RAM0;
 	return name_to_dev_t(root_device_name);
 }
 
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t

[Christoph Hellwig]

The nfs/cifs/ram special case only needs to be parsed once, and only in
the boot code.  Move them out of name_to_dev_t and into
prepare_namespace.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 Documentation/admin-guide/kernel-parameters.txt |  7 ++++++-
 init/do_mounts.c                                | 14 ++++++--------
 2 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 9e5bab29685ff0..457c342d15977e 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5452,7 +5452,12 @@
 			port and the regular usb controller gets disabled.
 
 	root=		[KNL] Root filesystem
-			See name_to_dev_t comment in init/do_mounts.c.
+			Usually this a a block device specifier of some kind,
+			see the name_to_dev_t comment in init/do_mounts.c for
+			details.
+			Alternatively this can be "ram" for the legacy initial
+			ramdisk, "nfs" and "cifs" for root on a network file
+			system, or "mtd" and "ubi" for mounting from raw flash.
 
 	rootdelay=	[KNL] Delay (in seconds) to pause before attempting to
 			mount the root filesystem
diff --git a/init/do_mounts.c b/init/do_mounts.c
index d5c06c1546e82c..86599faf2bf8a1 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -248,7 +248,6 @@ static dev_t devt_from_devnum(const char *name)
  *
  *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
  *         no leading 0x, for example b302.
- *	2) /dev/nfs represents Root_NFS (0xff)
  *	3) /dev/<disk_name> represents the device number of disk
  *	4) /dev/<disk_name><decimal> represents the device number
  *         of partition - device number of disk plus the partition number
@@ -266,7 +265,6 @@ static dev_t devt_from_devnum(const char *name)
  *	   a colon.
  *	9) PARTLABEL=<name> with name being the GPT partition label.
  *	   MSDOS partitions do not support labels!
- *	10) /dev/cifs represents Root_CIFS (0xfe)
  *
  *	If name doesn't have fall into the categories above, we return (0,0).
  *	block_class is used to check if something is a disk name. If the disk
@@ -275,12 +273,6 @@ static dev_t devt_from_devnum(const char *name)
  */
 dev_t name_to_dev_t(const char *name)
 {
-	if (strcmp(name, "/dev/nfs") == 0)
-		return Root_NFS;
-	if (strcmp(name, "/dev/cifs") == 0)
-		return Root_CIFS;
-	if (strcmp(name, "/dev/ram") == 0)
-		return Root_RAM0;
 #ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9);
@@ -631,6 +623,12 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
+	if (strcmp(root_device_name, "/dev/nfs") == 0)
+		return Root_NFS;
+	if (strcmp(root_device_name, "/dev/cifs") == 0)
+		return Root_CIFS;
+	if (strcmp(root_device_name, "/dev/ram") == 0)
+		return Root_RAM0;
 	return name_to_dev_t(root_device_name);
 }
 
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 12/24] init: move the nfs/cifs/ram special cases out of name_to_dev_t

[Christoph Hellwig]

The nfs/cifs/ram special case only needs to be parsed once, and only in
the boot code.  Move them out of name_to_dev_t and into
prepare_namespace.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 Documentation/admin-guide/kernel-parameters.txt |  7 ++++++-
 init/do_mounts.c                                | 14 ++++++--------
 2 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 9e5bab29685ff0..457c342d15977e 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5452,7 +5452,12 @@
 			port and the regular usb controller gets disabled.
 
 	root=		[KNL] Root filesystem
-			See name_to_dev_t comment in init/do_mounts.c.
+			Usually this a a block device specifier of some kind,
+			see the name_to_dev_t comment in init/do_mounts.c for
+			details.
+			Alternatively this can be "ram" for the legacy initial
+			ramdisk, "nfs" and "cifs" for root on a network file
+			system, or "mtd" and "ubi" for mounting from raw flash.
 
 	rootdelay=	[KNL] Delay (in seconds) to pause before attempting to
 			mount the root filesystem
diff --git a/init/do_mounts.c b/init/do_mounts.c
index d5c06c1546e82c..86599faf2bf8a1 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -248,7 +248,6 @@ static dev_t devt_from_devnum(const char *name)
  *
  *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
  *         no leading 0x, for example b302.
- *	2) /dev/nfs represents Root_NFS (0xff)
  *	3) /dev/<disk_name> represents the device number of disk
  *	4) /dev/<disk_name><decimal> represents the device number
  *         of partition - device number of disk plus the partition number
@@ -266,7 +265,6 @@ static dev_t devt_from_devnum(const char *name)
  *	   a colon.
  *	9) PARTLABEL=<name> with name being the GPT partition label.
  *	   MSDOS partitions do not support labels!
- *	10) /dev/cifs represents Root_CIFS (0xfe)
  *
  *	If name doesn't have fall into the categories above, we return (0,0).
  *	block_class is used to check if something is a disk name. If the disk
@@ -275,12 +273,6 @@ static dev_t devt_from_devnum(const char *name)
  */
 dev_t name_to_dev_t(const char *name)
 {
-	if (strcmp(name, "/dev/nfs") == 0)
-		return Root_NFS;
-	if (strcmp(name, "/dev/cifs") == 0)
-		return Root_CIFS;
-	if (strcmp(name, "/dev/ram") == 0)
-		return Root_RAM0;
 #ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9);
@@ -631,6 +623,12 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
+	if (strcmp(root_device_name, "/dev/nfs") == 0)
+		return Root_NFS;
+	if (strcmp(root_device_name, "/dev/cifs") == 0)
+		return Root_CIFS;
+	if (strcmp(root_device_name, "/dev/ram") == 0)
+		return Root_RAM0;
 	return name_to_dev_t(root_device_name);
 }
 
-- 
2.39.2

[dm-devel] [PATCH 13/24] init: improve the name_to_dev_t interface

[Christoph Hellwig]

name_to_dev_t has a very misleading name, that doesn't make clear
it should only be used by the early init code, and also has a bad
calling convention that doesn't allow returning different kinds of
errors.  Rename it to early_lookup_bdev to make the use case clear,
and return an errno, where -EINVAL means the string could not be
parsed, and -ENODEV means it the string was valid, but there was
no device found for it.

Also stub out the whole call for !CONFIG_BLOCK as all the non-block
root cases are always covered in the caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 drivers/md/dm-table.c                         |   5 +-
 drivers/md/md-autodetect.c                    |   3 +-
 drivers/mtd/devices/block2mtd.c               |   3 +-
 fs/pstore/blk.c                               |   4 +-
 include/linux/blkdev.h                        |   5 +
 include/linux/mount.h                         |   1 -
 init/do_mounts.c                              | 102 +++++++++---------
 kernel/power/hibernate.c                      |  22 ++--
 9 files changed, 74 insertions(+), 75 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 457c342d15977e..a6bc31349cbb76 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the name_to_dev_t comment in init/do_mounts.c for
-			details.
+			see the early_lookup_bdev comment in init/do_mounts.c
+			for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1398f1d6e83e7f..05aa16da43b0d5 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -330,8 +330,9 @@ dev_t dm_get_dev_t(const char *path)
 {
 	dev_t dev;
 
-	if (lookup_bdev(path, &dev))
-		dev = name_to_dev_t(path);
+	if (lookup_bdev(path, &dev) &&
+	    early_lookup_bdev(path, &dev))
+		return 0;
 	return dev;
 }
 EXPORT_SYMBOL_GPL(dm_get_dev_t);
diff --git a/drivers/md/md-autodetect.c b/drivers/md/md-autodetect.c
index 91836e6de3260f..6eaa0eab40f962 100644
--- a/drivers/md/md-autodetect.c
+++ b/drivers/md/md-autodetect.c
@@ -147,7 +147,8 @@ static void __init md_setup_drive(struct md_setup_args *args)
 		if (p)
 			*p++ = 0;
 
-		dev = name_to_dev_t(devname);
+		if (early_lookup_bdev(devname, &dev))
+			dev = 0;
 		if (strncmp(devname, "/dev/", 5) == 0)
 			devname += 5;
 		snprintf(comp_name, 63, "/dev/%s", devname);
diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4cd37ec45762b6..4c21e9f13bead5 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -254,8 +254,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		devt = name_to_dev_t(devname);
-		if (!devt)
+		if (early_lookup_bdev(devname, &devt))
 			continue;
 		bdev = blkdev_get_by_dev(devt, mode, dev);
 	}
diff --git a/fs/pstore/blk.c b/fs/pstore/blk.c
index 4ae0cfcd15f20b..de8cf5d75f34d5 100644
--- a/fs/pstore/blk.c
+++ b/fs/pstore/blk.c
@@ -263,9 +263,9 @@ static __init const char *early_boot_devpath(const char *initial_devname)
 	 * same scheme to find the device that we use for mounting
 	 * the root file system.
 	 */
-	dev_t dev = name_to_dev_t(initial_devname);
+	dev_t dev;
 
-	if (!dev) {
+	if (early_lookup_bdev(initial_devname, &dev)) {
 		pr_err("failed to resolve '%s'!\n", initial_devname);
 		return initial_devname;
 	}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d89c2da1469872..0bda6cb98d7eb8 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1495,6 +1495,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
+int early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1516,6 +1517,10 @@ static inline void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
 static inline void printk_all_partitions(void)
 {
 }
+static inline int early_lookup_bdev(const char *pathname, dev_t *dev)
+{
+	return -EINVAL;
+}
 #endif /* CONFIG_BLOCK */
 
 int fsync_bdev(struct block_device *bdev);
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 1ea326c368f726..4b81ea90440e45 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -107,7 +107,6 @@ extern struct vfsmount *vfs_submount(const struct dentry *mountpoint,
 extern void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list);
 extern void mark_mounts_for_expiry(struct list_head *mounts);
 
-extern dev_t name_to_dev_t(const char *name);
 extern bool path_is_mountpoint(const struct path *path);
 
 extern bool our_mnt(struct vfsmount *mnt);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 86599faf2bf8a1..f1953aeb321978 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -96,11 +96,10 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static dev_t devt_from_partuuid(const char *uuid_str)
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
-	dev_t devt = 0;
 	int offset = 0;
 	char *slash;
 
@@ -124,21 +123,21 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
-		return 0;
+		return -ENODEV;
 
 	if (offset) {
 		/*
 		 * Attempt to find the requested partition by adding an offset
 		 * to the partition number found by UUID.
 		 */
-		devt = part_devt(dev_to_disk(dev),
-				 dev_to_bdev(dev)->bd_partno + offset);
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
 	} else {
-		devt = dev->devt;
+		*devt = dev->devt;
 	}
 
 	put_device(dev);
-	return devt;
+	return 0;
 
 clear_root_wait:
 	pr_err("VFS: PARTUUID= is invalid.\n"
@@ -146,7 +145,7 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 	if (root_wait)
 		pr_err("Disabling rootwait; root= is invalid.\n");
 	root_wait = 0;
-	return 0;
+	return -EINVAL;
 }
 
 /**
@@ -166,38 +165,35 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static dev_t devt_from_partlabel(const char *label)
+static int devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
-	dev_t devt = 0;
 
 	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (dev) {
-		devt = dev->devt;
-		put_device(dev);
-	}
-
-	return devt;
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
 }
 
-static dev_t devt_from_devname(const char *name)
+static int devt_from_devname(const char *name, dev_t *devt)
 {
-	dev_t devt = 0;
 	int part;
 	char s[32];
 	char *p;
 
 	if (strlen(name) > 31)
-		return 0;
+		return -EINVAL;
 	strcpy(s, name);
 	for (p = s; *p; p++) {
 		if (*p == '/')
 			*p = '!';
 	}
 
-	devt = blk_lookup_devt(s, 0);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
 
 	/*
 	 * Try non-existent, but valid partition, which may only exist after
@@ -206,41 +202,42 @@ static dev_t devt_from_devname(const char *name)
 	while (p > s && isdigit(p[-1]))
 		p--;
 	if (p == s || !*p || *p == '0')
-		return 0;
+		return -EINVAL;
 
 	/* try disk name without <part number> */
 	part = simple_strtoul(p, NULL, 10);
 	*p = '\0';
-	devt = blk_lookup_devt(s, part);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
 
 	/* try disk name without p<part number> */
 	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return 0;
+		return -EINVAL;
 	p[-1] = '\0';
-	return blk_lookup_devt(s, part);
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
 }
-#endif /* CONFIG_BLOCK */
 
-static dev_t devt_from_devnum(const char *name)
+static int devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
-	dev_t devt = 0;
 	char *p, dummy;
 
 	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
 	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		devt = MKDEV(maj, min);
-		if (maj != MAJOR(devt) || min != MINOR(devt))
-			return 0;
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
 	} else {
-		devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
 		if (*p)
-			return 0;
+			return -EINVAL;
 	}
 
-	return devt;
+	return 0;
 }
 
 /*
@@ -271,19 +268,18 @@ static dev_t devt_from_devnum(const char *name)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-dev_t name_to_dev_t(const char *name)
+int early_lookup_bdev(const char *name, dev_t *devt)
 {
-#ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9);
+		return devt_from_partuuid(name + 9, devt);
 	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10);
+		return devt_from_partlabel(name + 10, devt);
 	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5);
-#endif
-	return devt_from_devnum(name);
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(name_to_dev_t);
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
+#endif
 
 static int __init root_dev_setup(char *line)
 {
@@ -606,20 +602,17 @@ static void __init wait_for_root(char *root_device_name)
 
 	pr_info("Waiting for root device %s...\n", root_device_name);
 
-	for (;;) {
-		if (driver_probe_done()) {
-			ROOT_DEV = name_to_dev_t(root_device_name);
-			if (ROOT_DEV)
-				break;
-		}
+	while (!driver_probe_done() ||
+	       early_lookup_bdev(root_device_name, &ROOT_DEV) < 0)
 		msleep(5);
-	}
 	async_synchronize_full();
 
 }
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	dev_t dev;
+
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
@@ -629,7 +622,10 @@ static dev_t __init parse_root_device(char *root_device_name)
 		return Root_CIFS;
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
-	return name_to_dev_t(root_device_name);
+
+	if (early_lookup_bdev(root_device_name, &dev))
+		return 0;
+	return dev;
 }
 
 /*
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 45e24b02cd50b6..c52dedb9f7c8e8 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -11,6 +11,7 @@
 
 #define pr_fmt(fmt) "PM: hibernation: " fmt
 
+#include <linux/blkdev.h>
 #include <linux/export.h>
 #include <linux/suspend.h>
 #include <linux/reboot.h>
@@ -921,8 +922,7 @@ static int __init find_resume_device(void)
 	}
 
 	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (swsusp_resume_device)
+	if (!early_lookup_bdev(resume_file, &swsusp_resume_device))
 		return 0;
 
 	/*
@@ -931,15 +931,12 @@ static int __init find_resume_device(void)
 	 */
 	wait_for_device_probe();
 	if (resume_wait) {
-		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+		while (early_lookup_bdev(resume_file, &swsusp_resume_device))
 			msleep(10);
 		async_synchronize_full();
 	}
 
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (!swsusp_resume_device)
-		return -ENODEV;
-	return 0;
+	return early_lookup_bdev(resume_file, &swsusp_resume_device);
 }
 
 static int software_resume(void)
@@ -1169,7 +1166,8 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	unsigned int sleep_flags;
 	int len = n;
 	char *name;
-	dev_t res;
+	dev_t dev;
+	int error;
 
 	if (!hibernation_available())
 		return 0;
@@ -1180,13 +1178,13 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	res = name_to_dev_t(name);
+	error = early_lookup_bdev(name, &dev);
 	kfree(name);
-	if (!res)
-		return -EINVAL;
+	if (error)
+		return error;
 
 	sleep_flags = lock_system_sleep();
-	swsusp_resume_device = res;
+	swsusp_resume_device = dev;
 	unlock_system_sleep(sleep_flags);
 
 	pm_pr_dbg("Configured hibernation resume from disk to %u\n",
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 13/24] init: improve the name_to_dev_t interface

[Christoph Hellwig]

name_to_dev_t has a very misleading name, that doesn't make clear
it should only be used by the early init code, and also has a bad
calling convention that doesn't allow returning different kinds of
errors.  Rename it to early_lookup_bdev to make the use case clear,
and return an errno, where -EINVAL means the string could not be
parsed, and -ENODEV means it the string was valid, but there was
no device found for it.

Also stub out the whole call for !CONFIG_BLOCK as all the non-block
root cases are always covered in the caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 drivers/md/dm-table.c                         |   5 +-
 drivers/md/md-autodetect.c                    |   3 +-
 drivers/mtd/devices/block2mtd.c               |   3 +-
 fs/pstore/blk.c                               |   4 +-
 include/linux/blkdev.h                        |   5 +
 include/linux/mount.h                         |   1 -
 init/do_mounts.c                              | 102 +++++++++---------
 kernel/power/hibernate.c                      |  22 ++--
 9 files changed, 74 insertions(+), 75 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 457c342d15977e..a6bc31349cbb76 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the name_to_dev_t comment in init/do_mounts.c for
-			details.
+			see the early_lookup_bdev comment in init/do_mounts.c
+			for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1398f1d6e83e7f..05aa16da43b0d5 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -330,8 +330,9 @@ dev_t dm_get_dev_t(const char *path)
 {
 	dev_t dev;
 
-	if (lookup_bdev(path, &dev))
-		dev = name_to_dev_t(path);
+	if (lookup_bdev(path, &dev) &&
+	    early_lookup_bdev(path, &dev))
+		return 0;
 	return dev;
 }
 EXPORT_SYMBOL_GPL(dm_get_dev_t);
diff --git a/drivers/md/md-autodetect.c b/drivers/md/md-autodetect.c
index 91836e6de3260f..6eaa0eab40f962 100644
--- a/drivers/md/md-autodetect.c
+++ b/drivers/md/md-autodetect.c
@@ -147,7 +147,8 @@ static void __init md_setup_drive(struct md_setup_args *args)
 		if (p)
 			*p++ = 0;
 
-		dev = name_to_dev_t(devname);
+		if (early_lookup_bdev(devname, &dev))
+			dev = 0;
 		if (strncmp(devname, "/dev/", 5) == 0)
 			devname += 5;
 		snprintf(comp_name, 63, "/dev/%s", devname);
diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4cd37ec45762b6..4c21e9f13bead5 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -254,8 +254,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		devt = name_to_dev_t(devname);
-		if (!devt)
+		if (early_lookup_bdev(devname, &devt))
 			continue;
 		bdev = blkdev_get_by_dev(devt, mode, dev);
 	}
diff --git a/fs/pstore/blk.c b/fs/pstore/blk.c
index 4ae0cfcd15f20b..de8cf5d75f34d5 100644
--- a/fs/pstore/blk.c
+++ b/fs/pstore/blk.c
@@ -263,9 +263,9 @@ static __init const char *early_boot_devpath(const char *initial_devname)
 	 * same scheme to find the device that we use for mounting
 	 * the root file system.
 	 */
-	dev_t dev = name_to_dev_t(initial_devname);
+	dev_t dev;
 
-	if (!dev) {
+	if (early_lookup_bdev(initial_devname, &dev)) {
 		pr_err("failed to resolve '%s'!\n", initial_devname);
 		return initial_devname;
 	}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d89c2da1469872..0bda6cb98d7eb8 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1495,6 +1495,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
+int early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1516,6 +1517,10 @@ static inline void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
 static inline void printk_all_partitions(void)
 {
 }
+static inline int early_lookup_bdev(const char *pathname, dev_t *dev)
+{
+	return -EINVAL;
+}
 #endif /* CONFIG_BLOCK */
 
 int fsync_bdev(struct block_device *bdev);
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 1ea326c368f726..4b81ea90440e45 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -107,7 +107,6 @@ extern struct vfsmount *vfs_submount(const struct dentry *mountpoint,
 extern void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list);
 extern void mark_mounts_for_expiry(struct list_head *mounts);
 
-extern dev_t name_to_dev_t(const char *name);
 extern bool path_is_mountpoint(const struct path *path);
 
 extern bool our_mnt(struct vfsmount *mnt);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 86599faf2bf8a1..f1953aeb321978 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -96,11 +96,10 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static dev_t devt_from_partuuid(const char *uuid_str)
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
-	dev_t devt = 0;
 	int offset = 0;
 	char *slash;
 
@@ -124,21 +123,21 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
-		return 0;
+		return -ENODEV;
 
 	if (offset) {
 		/*
 		 * Attempt to find the requested partition by adding an offset
 		 * to the partition number found by UUID.
 		 */
-		devt = part_devt(dev_to_disk(dev),
-				 dev_to_bdev(dev)->bd_partno + offset);
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
 	} else {
-		devt = dev->devt;
+		*devt = dev->devt;
 	}
 
 	put_device(dev);
-	return devt;
+	return 0;
 
 clear_root_wait:
 	pr_err("VFS: PARTUUID= is invalid.\n"
@@ -146,7 +145,7 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 	if (root_wait)
 		pr_err("Disabling rootwait; root= is invalid.\n");
 	root_wait = 0;
-	return 0;
+	return -EINVAL;
 }
 
 /**
@@ -166,38 +165,35 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static dev_t devt_from_partlabel(const char *label)
+static int devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
-	dev_t devt = 0;
 
 	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (dev) {
-		devt = dev->devt;
-		put_device(dev);
-	}
-
-	return devt;
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
 }
 
-static dev_t devt_from_devname(const char *name)
+static int devt_from_devname(const char *name, dev_t *devt)
 {
-	dev_t devt = 0;
 	int part;
 	char s[32];
 	char *p;
 
 	if (strlen(name) > 31)
-		return 0;
+		return -EINVAL;
 	strcpy(s, name);
 	for (p = s; *p; p++) {
 		if (*p == '/')
 			*p = '!';
 	}
 
-	devt = blk_lookup_devt(s, 0);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
 
 	/*
 	 * Try non-existent, but valid partition, which may only exist after
@@ -206,41 +202,42 @@ static dev_t devt_from_devname(const char *name)
 	while (p > s && isdigit(p[-1]))
 		p--;
 	if (p == s || !*p || *p == '0')
-		return 0;
+		return -EINVAL;
 
 	/* try disk name without <part number> */
 	part = simple_strtoul(p, NULL, 10);
 	*p = '\0';
-	devt = blk_lookup_devt(s, part);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
 
 	/* try disk name without p<part number> */
 	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return 0;
+		return -EINVAL;
 	p[-1] = '\0';
-	return blk_lookup_devt(s, part);
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
 }
-#endif /* CONFIG_BLOCK */
 
-static dev_t devt_from_devnum(const char *name)
+static int devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
-	dev_t devt = 0;
 	char *p, dummy;
 
 	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
 	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		devt = MKDEV(maj, min);
-		if (maj != MAJOR(devt) || min != MINOR(devt))
-			return 0;
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
 	} else {
-		devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
 		if (*p)
-			return 0;
+			return -EINVAL;
 	}
 
-	return devt;
+	return 0;
 }
 
 /*
@@ -271,19 +268,18 @@ static dev_t devt_from_devnum(const char *name)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-dev_t name_to_dev_t(const char *name)
+int early_lookup_bdev(const char *name, dev_t *devt)
 {
-#ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9);
+		return devt_from_partuuid(name + 9, devt);
 	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10);
+		return devt_from_partlabel(name + 10, devt);
 	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5);
-#endif
-	return devt_from_devnum(name);
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(name_to_dev_t);
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
+#endif
 
 static int __init root_dev_setup(char *line)
 {
@@ -606,20 +602,17 @@ static void __init wait_for_root(char *root_device_name)
 
 	pr_info("Waiting for root device %s...\n", root_device_name);
 
-	for (;;) {
-		if (driver_probe_done()) {
-			ROOT_DEV = name_to_dev_t(root_device_name);
-			if (ROOT_DEV)
-				break;
-		}
+	while (!driver_probe_done() ||
+	       early_lookup_bdev(root_device_name, &ROOT_DEV) < 0)
 		msleep(5);
-	}
 	async_synchronize_full();
 
 }
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	dev_t dev;
+
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
@@ -629,7 +622,10 @@ static dev_t __init parse_root_device(char *root_device_name)
 		return Root_CIFS;
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
-	return name_to_dev_t(root_device_name);
+
+	if (early_lookup_bdev(root_device_name, &dev))
+		return 0;
+	return dev;
 }
 
 /*
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 45e24b02cd50b6..c52dedb9f7c8e8 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -11,6 +11,7 @@
 
 #define pr_fmt(fmt) "PM: hibernation: " fmt
 
+#include <linux/blkdev.h>
 #include <linux/export.h>
 #include <linux/suspend.h>
 #include <linux/reboot.h>
@@ -921,8 +922,7 @@ static int __init find_resume_device(void)
 	}
 
 	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (swsusp_resume_device)
+	if (!early_lookup_bdev(resume_file, &swsusp_resume_device))
 		return 0;
 
 	/*
@@ -931,15 +931,12 @@ static int __init find_resume_device(void)
 	 */
 	wait_for_device_probe();
 	if (resume_wait) {
-		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+		while (early_lookup_bdev(resume_file, &swsusp_resume_device))
 			msleep(10);
 		async_synchronize_full();
 	}
 
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (!swsusp_resume_device)
-		return -ENODEV;
-	return 0;
+	return early_lookup_bdev(resume_file, &swsusp_resume_device);
 }
 
 static int software_resume(void)
@@ -1169,7 +1166,8 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	unsigned int sleep_flags;
 	int len = n;
 	char *name;
-	dev_t res;
+	dev_t dev;
+	int error;
 
 	if (!hibernation_available())
 		return 0;
@@ -1180,13 +1178,13 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	res = name_to_dev_t(name);
+	error = early_lookup_bdev(name, &dev);
 	kfree(name);
-	if (!res)
-		return -EINVAL;
+	if (error)
+		return error;
 
 	sleep_flags = lock_system_sleep();
-	swsusp_resume_device = res;
+	swsusp_resume_device = dev;
 	unlock_system_sleep(sleep_flags);
 
 	pm_pr_dbg("Configured hibernation resume from disk to %u\n",
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 13/24] init: improve the name_to_dev_t interface

[Christoph Hellwig]

name_to_dev_t has a very misleading name, that doesn't make clear
it should only be used by the early init code, and also has a bad
calling convention that doesn't allow returning different kinds of
errors.  Rename it to early_lookup_bdev to make the use case clear,
and return an errno, where -EINVAL means the string could not be
parsed, and -ENODEV means it the string was valid, but there was
no device found for it.

Also stub out the whole call for !CONFIG_BLOCK as all the non-block
root cases are always covered in the caller.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 drivers/md/dm-table.c                         |   5 +-
 drivers/md/md-autodetect.c                    |   3 +-
 drivers/mtd/devices/block2mtd.c               |   3 +-
 fs/pstore/blk.c                               |   4 +-
 include/linux/blkdev.h                        |   5 +
 include/linux/mount.h                         |   1 -
 init/do_mounts.c                              | 102 +++++++++---------
 kernel/power/hibernate.c                      |  22 ++--
 9 files changed, 74 insertions(+), 75 deletions(-)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 457c342d15977e..a6bc31349cbb76 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the name_to_dev_t comment in init/do_mounts.c for
-			details.
+			see the early_lookup_bdev comment in init/do_mounts.c
+			for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1398f1d6e83e7f..05aa16da43b0d5 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -330,8 +330,9 @@ dev_t dm_get_dev_t(const char *path)
 {
 	dev_t dev;
 
-	if (lookup_bdev(path, &dev))
-		dev = name_to_dev_t(path);
+	if (lookup_bdev(path, &dev) &&
+	    early_lookup_bdev(path, &dev))
+		return 0;
 	return dev;
 }
 EXPORT_SYMBOL_GPL(dm_get_dev_t);
diff --git a/drivers/md/md-autodetect.c b/drivers/md/md-autodetect.c
index 91836e6de3260f..6eaa0eab40f962 100644
--- a/drivers/md/md-autodetect.c
+++ b/drivers/md/md-autodetect.c
@@ -147,7 +147,8 @@ static void __init md_setup_drive(struct md_setup_args *args)
 		if (p)
 			*p++ = 0;
 
-		dev = name_to_dev_t(devname);
+		if (early_lookup_bdev(devname, &dev))
+			dev = 0;
 		if (strncmp(devname, "/dev/", 5) == 0)
 			devname += 5;
 		snprintf(comp_name, 63, "/dev/%s", devname);
diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4cd37ec45762b6..4c21e9f13bead5 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -254,8 +254,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		devt = name_to_dev_t(devname);
-		if (!devt)
+		if (early_lookup_bdev(devname, &devt))
 			continue;
 		bdev = blkdev_get_by_dev(devt, mode, dev);
 	}
diff --git a/fs/pstore/blk.c b/fs/pstore/blk.c
index 4ae0cfcd15f20b..de8cf5d75f34d5 100644
--- a/fs/pstore/blk.c
+++ b/fs/pstore/blk.c
@@ -263,9 +263,9 @@ static __init const char *early_boot_devpath(const char *initial_devname)
 	 * same scheme to find the device that we use for mounting
 	 * the root file system.
 	 */
-	dev_t dev = name_to_dev_t(initial_devname);
+	dev_t dev;
 
-	if (!dev) {
+	if (early_lookup_bdev(initial_devname, &dev)) {
 		pr_err("failed to resolve '%s'!\n", initial_devname);
 		return initial_devname;
 	}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d89c2da1469872..0bda6cb98d7eb8 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1495,6 +1495,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
+int early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1516,6 +1517,10 @@ static inline void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
 static inline void printk_all_partitions(void)
 {
 }
+static inline int early_lookup_bdev(const char *pathname, dev_t *dev)
+{
+	return -EINVAL;
+}
 #endif /* CONFIG_BLOCK */
 
 int fsync_bdev(struct block_device *bdev);
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 1ea326c368f726..4b81ea90440e45 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -107,7 +107,6 @@ extern struct vfsmount *vfs_submount(const struct dentry *mountpoint,
 extern void mnt_set_expiry(struct vfsmount *mnt, struct list_head *expiry_list);
 extern void mark_mounts_for_expiry(struct list_head *mounts);
 
-extern dev_t name_to_dev_t(const char *name);
 extern bool path_is_mountpoint(const struct path *path);
 
 extern bool our_mnt(struct vfsmount *mnt);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 86599faf2bf8a1..f1953aeb321978 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -96,11 +96,10 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static dev_t devt_from_partuuid(const char *uuid_str)
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
-	dev_t devt = 0;
 	int offset = 0;
 	char *slash;
 
@@ -124,21 +123,21 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
-		return 0;
+		return -ENODEV;
 
 	if (offset) {
 		/*
 		 * Attempt to find the requested partition by adding an offset
 		 * to the partition number found by UUID.
 		 */
-		devt = part_devt(dev_to_disk(dev),
-				 dev_to_bdev(dev)->bd_partno + offset);
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
 	} else {
-		devt = dev->devt;
+		*devt = dev->devt;
 	}
 
 	put_device(dev);
-	return devt;
+	return 0;
 
 clear_root_wait:
 	pr_err("VFS: PARTUUID= is invalid.\n"
@@ -146,7 +145,7 @@ static dev_t devt_from_partuuid(const char *uuid_str)
 	if (root_wait)
 		pr_err("Disabling rootwait; root= is invalid.\n");
 	root_wait = 0;
-	return 0;
+	return -EINVAL;
 }
 
 /**
@@ -166,38 +165,35 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static dev_t devt_from_partlabel(const char *label)
+static int devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
-	dev_t devt = 0;
 
 	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (dev) {
-		devt = dev->devt;
-		put_device(dev);
-	}
-
-	return devt;
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
 }
 
-static dev_t devt_from_devname(const char *name)
+static int devt_from_devname(const char *name, dev_t *devt)
 {
-	dev_t devt = 0;
 	int part;
 	char s[32];
 	char *p;
 
 	if (strlen(name) > 31)
-		return 0;
+		return -EINVAL;
 	strcpy(s, name);
 	for (p = s; *p; p++) {
 		if (*p == '/')
 			*p = '!';
 	}
 
-	devt = blk_lookup_devt(s, 0);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
 
 	/*
 	 * Try non-existent, but valid partition, which may only exist after
@@ -206,41 +202,42 @@ static dev_t devt_from_devname(const char *name)
 	while (p > s && isdigit(p[-1]))
 		p--;
 	if (p == s || !*p || *p == '0')
-		return 0;
+		return -EINVAL;
 
 	/* try disk name without <part number> */
 	part = simple_strtoul(p, NULL, 10);
 	*p = '\0';
-	devt = blk_lookup_devt(s, part);
-	if (devt)
-		return devt;
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
 
 	/* try disk name without p<part number> */
 	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return 0;
+		return -EINVAL;
 	p[-1] = '\0';
-	return blk_lookup_devt(s, part);
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
 }
-#endif /* CONFIG_BLOCK */
 
-static dev_t devt_from_devnum(const char *name)
+static int devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
-	dev_t devt = 0;
 	char *p, dummy;
 
 	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
 	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		devt = MKDEV(maj, min);
-		if (maj != MAJOR(devt) || min != MINOR(devt))
-			return 0;
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
 	} else {
-		devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
 		if (*p)
-			return 0;
+			return -EINVAL;
 	}
 
-	return devt;
+	return 0;
 }
 
 /*
@@ -271,19 +268,18 @@ static dev_t devt_from_devnum(const char *name)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-dev_t name_to_dev_t(const char *name)
+int early_lookup_bdev(const char *name, dev_t *devt)
 {
-#ifdef CONFIG_BLOCK
 	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9);
+		return devt_from_partuuid(name + 9, devt);
 	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10);
+		return devt_from_partlabel(name + 10, devt);
 	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5);
-#endif
-	return devt_from_devnum(name);
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(name_to_dev_t);
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
+#endif
 
 static int __init root_dev_setup(char *line)
 {
@@ -606,20 +602,17 @@ static void __init wait_for_root(char *root_device_name)
 
 	pr_info("Waiting for root device %s...\n", root_device_name);
 
-	for (;;) {
-		if (driver_probe_done()) {
-			ROOT_DEV = name_to_dev_t(root_device_name);
-			if (ROOT_DEV)
-				break;
-		}
+	while (!driver_probe_done() ||
+	       early_lookup_bdev(root_device_name, &ROOT_DEV) < 0)
 		msleep(5);
-	}
 	async_synchronize_full();
 
 }
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	dev_t dev;
+
 	if (!strncmp(root_device_name, "mtd", 3) ||
 	    !strncmp(root_device_name, "ubi", 3))
 		return Root_Generic;
@@ -629,7 +622,10 @@ static dev_t __init parse_root_device(char *root_device_name)
 		return Root_CIFS;
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
-	return name_to_dev_t(root_device_name);
+
+	if (early_lookup_bdev(root_device_name, &dev))
+		return 0;
+	return dev;
 }
 
 /*
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 45e24b02cd50b6..c52dedb9f7c8e8 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -11,6 +11,7 @@
 
 #define pr_fmt(fmt) "PM: hibernation: " fmt
 
+#include <linux/blkdev.h>
 #include <linux/export.h>
 #include <linux/suspend.h>
 #include <linux/reboot.h>
@@ -921,8 +922,7 @@ static int __init find_resume_device(void)
 	}
 
 	/* Check if the device is there */
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (swsusp_resume_device)
+	if (!early_lookup_bdev(resume_file, &swsusp_resume_device))
 		return 0;
 
 	/*
@@ -931,15 +931,12 @@ static int __init find_resume_device(void)
 	 */
 	wait_for_device_probe();
 	if (resume_wait) {
-		while (!(swsusp_resume_device = name_to_dev_t(resume_file)))
+		while (early_lookup_bdev(resume_file, &swsusp_resume_device))
 			msleep(10);
 		async_synchronize_full();
 	}
 
-	swsusp_resume_device = name_to_dev_t(resume_file);
-	if (!swsusp_resume_device)
-		return -ENODEV;
-	return 0;
+	return early_lookup_bdev(resume_file, &swsusp_resume_device);
 }
 
 static int software_resume(void)
@@ -1169,7 +1166,8 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	unsigned int sleep_flags;
 	int len = n;
 	char *name;
-	dev_t res;
+	dev_t dev;
+	int error;
 
 	if (!hibernation_available())
 		return 0;
@@ -1180,13 +1178,13 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	res = name_to_dev_t(name);
+	error = early_lookup_bdev(name, &dev);
 	kfree(name);
-	if (!res)
-		return -EINVAL;
+	if (error)
+		return error;
 
 	sleep_flags = lock_system_sleep();
-	swsusp_resume_device = res;
+	swsusp_resume_device = dev;
 	unlock_system_sleep(sleep_flags);
 
 	pm_pr_dbg("Configured hibernation resume from disk to %u\n",
-- 
2.39.2

[dm-devel] [PATCH 14/24] init: clear root_wait on all invalid root= strings

[Christoph Hellwig]

Instead of only clearing root_wait in devt_from_partuuid when the UUID
format was invalid, do that in parse_root_device for all strings that
failed to parse.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index f1953aeb321978..0b36a5f39ee8e2 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -112,14 +112,14 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 
 		/* Explicitly fail on poor PARTUUID syntax. */
 		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto clear_root_wait;
+			goto out_invalid;
 		cmp.len = slash - uuid_str;
 	} else {
 		cmp.len = strlen(uuid_str);
 	}
 
 	if (!cmp.len)
-		goto clear_root_wait;
+		goto out_invalid;
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
@@ -139,12 +139,9 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 	put_device(dev);
 	return 0;
 
-clear_root_wait:
+out_invalid:
 	pr_err("VFS: PARTUUID= is invalid.\n"
 	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	if (root_wait)
-		pr_err("Disabling rootwait; root= is invalid.\n");
-	root_wait = 0;
 	return -EINVAL;
 }
 
@@ -611,6 +608,7 @@ static void __init wait_for_root(char *root_device_name)
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	int error;
 	dev_t dev;
 
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -623,8 +621,14 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
 
-	if (early_lookup_bdev(root_device_name, &dev))
+	error = early_lookup_bdev(root_device_name, &dev);
+	if (error) {
+		if (error == -EINVAL && root_wait) {
+			pr_err("Disabling rootwait; root= is invalid.\n");
+			root_wait = 0;
+		}
 		return 0;
+	}
 	return dev;
 }
 
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 14/24] init: clear root_wait on all invalid root= strings

[Christoph Hellwig]

Instead of only clearing root_wait in devt_from_partuuid when the UUID
format was invalid, do that in parse_root_device for all strings that
failed to parse.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index f1953aeb321978..0b36a5f39ee8e2 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -112,14 +112,14 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 
 		/* Explicitly fail on poor PARTUUID syntax. */
 		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto clear_root_wait;
+			goto out_invalid;
 		cmp.len = slash - uuid_str;
 	} else {
 		cmp.len = strlen(uuid_str);
 	}
 
 	if (!cmp.len)
-		goto clear_root_wait;
+		goto out_invalid;
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
@@ -139,12 +139,9 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 	put_device(dev);
 	return 0;
 
-clear_root_wait:
+out_invalid:
 	pr_err("VFS: PARTUUID= is invalid.\n"
 	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	if (root_wait)
-		pr_err("Disabling rootwait; root= is invalid.\n");
-	root_wait = 0;
 	return -EINVAL;
 }
 
@@ -611,6 +608,7 @@ static void __init wait_for_root(char *root_device_name)
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	int error;
 	dev_t dev;
 
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -623,8 +621,14 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
 
-	if (early_lookup_bdev(root_device_name, &dev))
+	error = early_lookup_bdev(root_device_name, &dev);
+	if (error) {
+		if (error == -EINVAL && root_wait) {
+			pr_err("Disabling rootwait; root= is invalid.\n");
+			root_wait = 0;
+		}
 		return 0;
+	}
 	return dev;
 }
 
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 14/24] init: clear root_wait on all invalid root= strings

[Christoph Hellwig]

Instead of only clearing root_wait in devt_from_partuuid when the UUID
format was invalid, do that in parse_root_device for all strings that
failed to parse.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 init/do_mounts.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index f1953aeb321978..0b36a5f39ee8e2 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -112,14 +112,14 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 
 		/* Explicitly fail on poor PARTUUID syntax. */
 		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto clear_root_wait;
+			goto out_invalid;
 		cmp.len = slash - uuid_str;
 	} else {
 		cmp.len = strlen(uuid_str);
 	}
 
 	if (!cmp.len)
-		goto clear_root_wait;
+		goto out_invalid;
 
 	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
 	if (!dev)
@@ -139,12 +139,9 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
 	put_device(dev);
 	return 0;
 
-clear_root_wait:
+out_invalid:
 	pr_err("VFS: PARTUUID= is invalid.\n"
 	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	if (root_wait)
-		pr_err("Disabling rootwait; root= is invalid.\n");
-	root_wait = 0;
 	return -EINVAL;
 }
 
@@ -611,6 +608,7 @@ static void __init wait_for_root(char *root_device_name)
 
 static dev_t __init parse_root_device(char *root_device_name)
 {
+	int error;
 	dev_t dev;
 
 	if (!strncmp(root_device_name, "mtd", 3) ||
@@ -623,8 +621,14 @@ static dev_t __init parse_root_device(char *root_device_name)
 	if (strcmp(root_device_name, "/dev/ram") == 0)
 		return Root_RAM0;
 
-	if (early_lookup_bdev(root_device_name, &dev))
+	error = early_lookup_bdev(root_device_name, &dev);
+	if (error) {
+		if (error == -EINVAL && root_wait) {
+			pr_err("Disabling rootwait; root= is invalid.\n");
+			root_wait = 0;
+		}
 		return 0;
+	}
 	return dev;
 }
 
-- 
2.39.2

[dm-devel] [PATCH 15/24] block: move the code to do early boot lookup of block devices to block/

[Christoph Hellwig]

Create a new block/early-lookup.c to keep the early block device lookup
code instead of having this code sit with the early mount code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 block/Makefile                                |   2 +-
 block/early-lookup.c                          | 224 ++++++++++++++++++
 init/do_mounts.c                              | 219 -----------------
 4 files changed, 227 insertions(+), 222 deletions(-)
 create mode 100644 block/early-lookup.c

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index a6bc31349cbb76..911c54829c7c59 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the early_lookup_bdev comment in init/do_mounts.c
-			for details.
+			see the early_lookup_bdev comment in
+			block/early-lookup.c for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/block/Makefile b/block/Makefile
index b31b05390749a1..46ada9dc8bbfe2 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -9,7 +9,7 @@ obj-y		:= bdev.o fops.o bio.o elevator.o blk-core.o blk-sysfs.o \
 			blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \
 			blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \
 			genhd.o ioprio.o badblocks.o partitions/ blk-rq-qos.o \
-			disk-events.o blk-ia-ranges.o
+			disk-events.o blk-ia-ranges.o early-lookup.o
 
 obj-$(CONFIG_BOUNCE)		+= bounce.o
 obj-$(CONFIG_BLK_DEV_BSG_COMMON) += bsg.o
diff --git a/block/early-lookup.c b/block/early-lookup.c
new file mode 100644
index 00000000000000..9fc30d039508af
--- /dev/null
+++ b/block/early-lookup.c
@@ -0,0 +1,224 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Code for looking up block devices in the early boot code before mounting the
+ * root file system.  Unfortunately currently also abused in a few other places.
+ */
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+
+struct uuidcmp {
+	const char *uuid;
+	int len;
+};
+
+/**
+ * match_dev_by_uuid - callback for finding a partition using its uuid
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the desired struct uuidcmp to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_uuid(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const struct uuidcmp *cmp = data;
+
+	if (!bdev->bd_meta_info ||
+	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
+		return 0;
+	return 1;
+}
+
+/**
+ * devt_from_partuuid - looks up the dev_t of a partition by its UUID
+ * @uuid_str:	char array containing ascii UUID
+ *
+ * The function will return the first partition which contains a matching
+ * UUID value in its partition_meta_info struct.  This does not search
+ * by filesystem UUIDs.
+ *
+ * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
+ * extracted and used as an offset from the partition identified by the UUID.
+ *
+ * Returns the matching dev_t on success or 0 on failure.
+ */
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+{
+	struct uuidcmp cmp;
+	struct device *dev = NULL;
+	int offset = 0;
+	char *slash;
+
+	cmp.uuid = uuid_str;
+
+	slash = strchr(uuid_str, '/');
+	/* Check for optional partition number offset attributes. */
+	if (slash) {
+		char c = 0;
+
+		/* Explicitly fail on poor PARTUUID syntax. */
+		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
+			goto out_invalid;
+		cmp.len = slash - uuid_str;
+	} else {
+		cmp.len = strlen(uuid_str);
+	}
+
+	if (!cmp.len)
+		goto out_invalid;
+
+	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
+	if (!dev)
+		return -ENODEV;
+
+	if (offset) {
+		/*
+		 * Attempt to find the requested partition by adding an offset
+		 * to the partition number found by UUID.
+		 */
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
+	} else {
+		*devt = dev->devt;
+	}
+
+	put_device(dev);
+	return 0;
+
+out_invalid:
+	pr_err("VFS: PARTUUID= is invalid.\n"
+	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
+	return -EINVAL;
+}
+
+/**
+ * match_dev_by_label - callback for finding a partition using its label
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the label to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_label(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const char *label = data;
+
+	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
+		return 0;
+	return 1;
+}
+
+static int devt_from_partlabel(const char *label, dev_t *devt)
+{
+	struct device *dev;
+
+	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
+}
+
+static int devt_from_devname(const char *name, dev_t *devt)
+{
+	int part;
+	char s[32];
+	char *p;
+
+	if (strlen(name) > 31)
+		return -EINVAL;
+	strcpy(s, name);
+	for (p = s; *p; p++) {
+		if (*p == '/')
+			*p = '!';
+	}
+
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
+
+	/*
+	 * Try non-existent, but valid partition, which may only exist after
+	 * opening the device, like partitioned md devices.
+	 */
+	while (p > s && isdigit(p[-1]))
+		p--;
+	if (p == s || !*p || *p == '0')
+		return -EINVAL;
+
+	/* try disk name without <part number> */
+	part = simple_strtoul(p, NULL, 10);
+	*p = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+
+	/* try disk name without p<part number> */
+	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
+		return -EINVAL;
+	p[-1] = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
+}
+
+static int devt_from_devnum(const char *name, dev_t *devt)
+{
+	unsigned maj, min, offset;
+	char *p, dummy;
+
+	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
+	} else {
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		if (*p)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+/*
+ *	Convert a name into device number.  We accept the following variants:
+ *
+ *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
+ *         no leading 0x, for example b302.
+ *	3) /dev/<disk_name> represents the device number of disk
+ *	4) /dev/<disk_name><decimal> represents the device number
+ *         of partition - device number of disk plus the partition number
+ *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
+ *	   used when disk name of partitioned disk ends on a digit.
+ *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
+ *	   unique id of a partition if the partition table provides it.
+ *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
+ *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
+ *	   filled hex representation of the 32-bit "NT disk signature", and PP
+ *	   is a zero-filled hex representation of the 1-based partition number.
+ *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
+ *	   a partition with a known unique id.
+ *	8) <major>:<minor> major and minor number of the device separated by
+ *	   a colon.
+ *	9) PARTLABEL=<name> with name being the GPT partition label.
+ *	   MSDOS partitions do not support labels!
+ *
+ *	If name doesn't have fall into the categories above, we return (0,0).
+ *	block_class is used to check if something is a disk name. If the disk
+ *	name contains slashes, the device name has them replaced with
+ *	bangs.
+ */
+int early_lookup_bdev(const char *name, dev_t *devt)
+{
+	if (strncmp(name, "PARTUUID=", 9) == 0)
+		return devt_from_partuuid(name + 9, devt);
+	if (strncmp(name, "PARTLABEL=", 10) == 0)
+		return devt_from_partlabel(name + 10, devt);
+	if (strncmp(name, "/dev/", 5) == 0)
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
+}
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 0b36a5f39ee8e2..780546a6cbfb6f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -59,225 +59,6 @@ static int __init readwrite(char *str)
 __setup("ro", readonly);
 __setup("rw", readwrite);
 
-#ifdef CONFIG_BLOCK
-struct uuidcmp {
-	const char *uuid;
-	int len;
-};
-
-/**
- * match_dev_by_uuid - callback for finding a partition using its uuid
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the desired struct uuidcmp to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_uuid(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const struct uuidcmp *cmp = data;
-
-	if (!bdev->bd_meta_info ||
-	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
-		return 0;
-	return 1;
-}
-
-/**
- * devt_from_partuuid - looks up the dev_t of a partition by its UUID
- * @uuid_str:	char array containing ascii UUID
- *
- * The function will return the first partition which contains a matching
- * UUID value in its partition_meta_info struct.  This does not search
- * by filesystem UUIDs.
- *
- * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
- * extracted and used as an offset from the partition identified by the UUID.
- *
- * Returns the matching dev_t on success or 0 on failure.
- */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
-{
-	struct uuidcmp cmp;
-	struct device *dev = NULL;
-	int offset = 0;
-	char *slash;
-
-	cmp.uuid = uuid_str;
-
-	slash = strchr(uuid_str, '/');
-	/* Check for optional partition number offset attributes. */
-	if (slash) {
-		char c = 0;
-
-		/* Explicitly fail on poor PARTUUID syntax. */
-		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto out_invalid;
-		cmp.len = slash - uuid_str;
-	} else {
-		cmp.len = strlen(uuid_str);
-	}
-
-	if (!cmp.len)
-		goto out_invalid;
-
-	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
-	if (!dev)
-		return -ENODEV;
-
-	if (offset) {
-		/*
-		 * Attempt to find the requested partition by adding an offset
-		 * to the partition number found by UUID.
-		 */
-		*devt = part_devt(dev_to_disk(dev),
-				  dev_to_bdev(dev)->bd_partno + offset);
-	} else {
-		*devt = dev->devt;
-	}
-
-	put_device(dev);
-	return 0;
-
-out_invalid:
-	pr_err("VFS: PARTUUID= is invalid.\n"
-	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	return -EINVAL;
-}
-
-/**
- * match_dev_by_label - callback for finding a partition using its label
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the label to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_label(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const char *label = data;
-
-	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
-		return 0;
-	return 1;
-}
-
-static int devt_from_partlabel(const char *label, dev_t *devt)
-{
-	struct device *dev;
-
-	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (!dev)
-		return -ENODEV;
-	*devt = dev->devt;
-	put_device(dev);
-	return 0;
-}
-
-static int devt_from_devname(const char *name, dev_t *devt)
-{
-	int part;
-	char s[32];
-	char *p;
-
-	if (strlen(name) > 31)
-		return -EINVAL;
-	strcpy(s, name);
-	for (p = s; *p; p++) {
-		if (*p == '/')
-			*p = '!';
-	}
-
-	*devt = blk_lookup_devt(s, 0);
-	if (*devt)
-		return 0;
-
-	/*
-	 * Try non-existent, but valid partition, which may only exist after
-	 * opening the device, like partitioned md devices.
-	 */
-	while (p > s && isdigit(p[-1]))
-		p--;
-	if (p == s || !*p || *p == '0')
-		return -EINVAL;
-
-	/* try disk name without <part number> */
-	part = simple_strtoul(p, NULL, 10);
-	*p = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-
-	/* try disk name without p<part number> */
-	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return -EINVAL;
-	p[-1] = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-	return -EINVAL;
-}
-
-static int devt_from_devnum(const char *name, dev_t *devt)
-{
-	unsigned maj, min, offset;
-	char *p, dummy;
-
-	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
-	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		*devt = MKDEV(maj, min);
-		if (maj != MAJOR(*devt) || min != MINOR(*devt))
-			return -EINVAL;
-	} else {
-		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
-		if (*p)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-/*
- *	Convert a name into device number.  We accept the following variants:
- *
- *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
- *         no leading 0x, for example b302.
- *	3) /dev/<disk_name> represents the device number of disk
- *	4) /dev/<disk_name><decimal> represents the device number
- *         of partition - device number of disk plus the partition number
- *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
- *	   used when disk name of partitioned disk ends on a digit.
- *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
- *	   unique id of a partition if the partition table provides it.
- *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
- *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
- *	   filled hex representation of the 32-bit "NT disk signature", and PP
- *	   is a zero-filled hex representation of the 1-based partition number.
- *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
- *	   a partition with a known unique id.
- *	8) <major>:<minor> major and minor number of the device separated by
- *	   a colon.
- *	9) PARTLABEL=<name> with name being the GPT partition label.
- *	   MSDOS partitions do not support labels!
- *
- *	If name doesn't have fall into the categories above, we return (0,0).
- *	block_class is used to check if something is a disk name. If the disk
- *	name contains slashes, the device name has them replaced with
- *	bangs.
- */
-int early_lookup_bdev(const char *name, dev_t *devt)
-{
-	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9, devt);
-	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10, devt);
-	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5, devt);
-	return devt_from_devnum(name, devt);
-}
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
-#endif
-
 static int __init root_dev_setup(char *line)
 {
 	strscpy(saved_root_name, line, sizeof(saved_root_name));
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 15/24] block: move the code to do early boot lookup of block devices to block/

[Christoph Hellwig]

Create a new block/early-lookup.c to keep the early block device lookup
code instead of having this code sit with the early mount code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 block/Makefile                                |   2 +-
 block/early-lookup.c                          | 224 ++++++++++++++++++
 init/do_mounts.c                              | 219 -----------------
 4 files changed, 227 insertions(+), 222 deletions(-)
 create mode 100644 block/early-lookup.c

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index a6bc31349cbb76..911c54829c7c59 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the early_lookup_bdev comment in init/do_mounts.c
-			for details.
+			see the early_lookup_bdev comment in
+			block/early-lookup.c for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/block/Makefile b/block/Makefile
index b31b05390749a1..46ada9dc8bbfe2 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -9,7 +9,7 @@ obj-y		:= bdev.o fops.o bio.o elevator.o blk-core.o blk-sysfs.o \
 			blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \
 			blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \
 			genhd.o ioprio.o badblocks.o partitions/ blk-rq-qos.o \
-			disk-events.o blk-ia-ranges.o
+			disk-events.o blk-ia-ranges.o early-lookup.o
 
 obj-$(CONFIG_BOUNCE)		+= bounce.o
 obj-$(CONFIG_BLK_DEV_BSG_COMMON) += bsg.o
diff --git a/block/early-lookup.c b/block/early-lookup.c
new file mode 100644
index 00000000000000..9fc30d039508af
--- /dev/null
+++ b/block/early-lookup.c
@@ -0,0 +1,224 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Code for looking up block devices in the early boot code before mounting the
+ * root file system.  Unfortunately currently also abused in a few other places.
+ */
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+
+struct uuidcmp {
+	const char *uuid;
+	int len;
+};
+
+/**
+ * match_dev_by_uuid - callback for finding a partition using its uuid
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the desired struct uuidcmp to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_uuid(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const struct uuidcmp *cmp = data;
+
+	if (!bdev->bd_meta_info ||
+	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
+		return 0;
+	return 1;
+}
+
+/**
+ * devt_from_partuuid - looks up the dev_t of a partition by its UUID
+ * @uuid_str:	char array containing ascii UUID
+ *
+ * The function will return the first partition which contains a matching
+ * UUID value in its partition_meta_info struct.  This does not search
+ * by filesystem UUIDs.
+ *
+ * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
+ * extracted and used as an offset from the partition identified by the UUID.
+ *
+ * Returns the matching dev_t on success or 0 on failure.
+ */
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+{
+	struct uuidcmp cmp;
+	struct device *dev = NULL;
+	int offset = 0;
+	char *slash;
+
+	cmp.uuid = uuid_str;
+
+	slash = strchr(uuid_str, '/');
+	/* Check for optional partition number offset attributes. */
+	if (slash) {
+		char c = 0;
+
+		/* Explicitly fail on poor PARTUUID syntax. */
+		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
+			goto out_invalid;
+		cmp.len = slash - uuid_str;
+	} else {
+		cmp.len = strlen(uuid_str);
+	}
+
+	if (!cmp.len)
+		goto out_invalid;
+
+	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
+	if (!dev)
+		return -ENODEV;
+
+	if (offset) {
+		/*
+		 * Attempt to find the requested partition by adding an offset
+		 * to the partition number found by UUID.
+		 */
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
+	} else {
+		*devt = dev->devt;
+	}
+
+	put_device(dev);
+	return 0;
+
+out_invalid:
+	pr_err("VFS: PARTUUID= is invalid.\n"
+	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
+	return -EINVAL;
+}
+
+/**
+ * match_dev_by_label - callback for finding a partition using its label
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the label to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_label(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const char *label = data;
+
+	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
+		return 0;
+	return 1;
+}
+
+static int devt_from_partlabel(const char *label, dev_t *devt)
+{
+	struct device *dev;
+
+	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
+}
+
+static int devt_from_devname(const char *name, dev_t *devt)
+{
+	int part;
+	char s[32];
+	char *p;
+
+	if (strlen(name) > 31)
+		return -EINVAL;
+	strcpy(s, name);
+	for (p = s; *p; p++) {
+		if (*p == '/')
+			*p = '!';
+	}
+
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
+
+	/*
+	 * Try non-existent, but valid partition, which may only exist after
+	 * opening the device, like partitioned md devices.
+	 */
+	while (p > s && isdigit(p[-1]))
+		p--;
+	if (p == s || !*p || *p == '0')
+		return -EINVAL;
+
+	/* try disk name without <part number> */
+	part = simple_strtoul(p, NULL, 10);
+	*p = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+
+	/* try disk name without p<part number> */
+	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
+		return -EINVAL;
+	p[-1] = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
+}
+
+static int devt_from_devnum(const char *name, dev_t *devt)
+{
+	unsigned maj, min, offset;
+	char *p, dummy;
+
+	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
+	} else {
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		if (*p)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+/*
+ *	Convert a name into device number.  We accept the following variants:
+ *
+ *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
+ *         no leading 0x, for example b302.
+ *	3) /dev/<disk_name> represents the device number of disk
+ *	4) /dev/<disk_name><decimal> represents the device number
+ *         of partition - device number of disk plus the partition number
+ *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
+ *	   used when disk name of partitioned disk ends on a digit.
+ *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
+ *	   unique id of a partition if the partition table provides it.
+ *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
+ *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
+ *	   filled hex representation of the 32-bit "NT disk signature", and PP
+ *	   is a zero-filled hex representation of the 1-based partition number.
+ *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
+ *	   a partition with a known unique id.
+ *	8) <major>:<minor> major and minor number of the device separated by
+ *	   a colon.
+ *	9) PARTLABEL=<name> with name being the GPT partition label.
+ *	   MSDOS partitions do not support labels!
+ *
+ *	If name doesn't have fall into the categories above, we return (0,0).
+ *	block_class is used to check if something is a disk name. If the disk
+ *	name contains slashes, the device name has them replaced with
+ *	bangs.
+ */
+int early_lookup_bdev(const char *name, dev_t *devt)
+{
+	if (strncmp(name, "PARTUUID=", 9) == 0)
+		return devt_from_partuuid(name + 9, devt);
+	if (strncmp(name, "PARTLABEL=", 10) == 0)
+		return devt_from_partlabel(name + 10, devt);
+	if (strncmp(name, "/dev/", 5) == 0)
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
+}
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 0b36a5f39ee8e2..780546a6cbfb6f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -59,225 +59,6 @@ static int __init readwrite(char *str)
 __setup("ro", readonly);
 __setup("rw", readwrite);
 
-#ifdef CONFIG_BLOCK
-struct uuidcmp {
-	const char *uuid;
-	int len;
-};
-
-/**
- * match_dev_by_uuid - callback for finding a partition using its uuid
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the desired struct uuidcmp to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_uuid(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const struct uuidcmp *cmp = data;
-
-	if (!bdev->bd_meta_info ||
-	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
-		return 0;
-	return 1;
-}
-
-/**
- * devt_from_partuuid - looks up the dev_t of a partition by its UUID
- * @uuid_str:	char array containing ascii UUID
- *
- * The function will return the first partition which contains a matching
- * UUID value in its partition_meta_info struct.  This does not search
- * by filesystem UUIDs.
- *
- * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
- * extracted and used as an offset from the partition identified by the UUID.
- *
- * Returns the matching dev_t on success or 0 on failure.
- */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
-{
-	struct uuidcmp cmp;
-	struct device *dev = NULL;
-	int offset = 0;
-	char *slash;
-
-	cmp.uuid = uuid_str;
-
-	slash = strchr(uuid_str, '/');
-	/* Check for optional partition number offset attributes. */
-	if (slash) {
-		char c = 0;
-
-		/* Explicitly fail on poor PARTUUID syntax. */
-		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto out_invalid;
-		cmp.len = slash - uuid_str;
-	} else {
-		cmp.len = strlen(uuid_str);
-	}
-
-	if (!cmp.len)
-		goto out_invalid;
-
-	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
-	if (!dev)
-		return -ENODEV;
-
-	if (offset) {
-		/*
-		 * Attempt to find the requested partition by adding an offset
-		 * to the partition number found by UUID.
-		 */
-		*devt = part_devt(dev_to_disk(dev),
-				  dev_to_bdev(dev)->bd_partno + offset);
-	} else {
-		*devt = dev->devt;
-	}
-
-	put_device(dev);
-	return 0;
-
-out_invalid:
-	pr_err("VFS: PARTUUID= is invalid.\n"
-	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	return -EINVAL;
-}
-
-/**
- * match_dev_by_label - callback for finding a partition using its label
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the label to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_label(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const char *label = data;
-
-	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
-		return 0;
-	return 1;
-}
-
-static int devt_from_partlabel(const char *label, dev_t *devt)
-{
-	struct device *dev;
-
-	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (!dev)
-		return -ENODEV;
-	*devt = dev->devt;
-	put_device(dev);
-	return 0;
-}
-
-static int devt_from_devname(const char *name, dev_t *devt)
-{
-	int part;
-	char s[32];
-	char *p;
-
-	if (strlen(name) > 31)
-		return -EINVAL;
-	strcpy(s, name);
-	for (p = s; *p; p++) {
-		if (*p == '/')
-			*p = '!';
-	}
-
-	*devt = blk_lookup_devt(s, 0);
-	if (*devt)
-		return 0;
-
-	/*
-	 * Try non-existent, but valid partition, which may only exist after
-	 * opening the device, like partitioned md devices.
-	 */
-	while (p > s && isdigit(p[-1]))
-		p--;
-	if (p == s || !*p || *p == '0')
-		return -EINVAL;
-
-	/* try disk name without <part number> */
-	part = simple_strtoul(p, NULL, 10);
-	*p = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-
-	/* try disk name without p<part number> */
-	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return -EINVAL;
-	p[-1] = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-	return -EINVAL;
-}
-
-static int devt_from_devnum(const char *name, dev_t *devt)
-{
-	unsigned maj, min, offset;
-	char *p, dummy;
-
-	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
-	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		*devt = MKDEV(maj, min);
-		if (maj != MAJOR(*devt) || min != MINOR(*devt))
-			return -EINVAL;
-	} else {
-		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
-		if (*p)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-/*
- *	Convert a name into device number.  We accept the following variants:
- *
- *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
- *         no leading 0x, for example b302.
- *	3) /dev/<disk_name> represents the device number of disk
- *	4) /dev/<disk_name><decimal> represents the device number
- *         of partition - device number of disk plus the partition number
- *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
- *	   used when disk name of partitioned disk ends on a digit.
- *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
- *	   unique id of a partition if the partition table provides it.
- *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
- *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
- *	   filled hex representation of the 32-bit "NT disk signature", and PP
- *	   is a zero-filled hex representation of the 1-based partition number.
- *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
- *	   a partition with a known unique id.
- *	8) <major>:<minor> major and minor number of the device separated by
- *	   a colon.
- *	9) PARTLABEL=<name> with name being the GPT partition label.
- *	   MSDOS partitions do not support labels!
- *
- *	If name doesn't have fall into the categories above, we return (0,0).
- *	block_class is used to check if something is a disk name. If the disk
- *	name contains slashes, the device name has them replaced with
- *	bangs.
- */
-int early_lookup_bdev(const char *name, dev_t *devt)
-{
-	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9, devt);
-	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10, devt);
-	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5, devt);
-	return devt_from_devnum(name, devt);
-}
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
-#endif
-
 static int __init root_dev_setup(char *line)
 {
 	strscpy(saved_root_name, line, sizeof(saved_root_name));
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 15/24] block: move the code to do early boot lookup of block devices to block/

[Christoph Hellwig]

Create a new block/early-lookup.c to keep the early block device lookup
code instead of having this code sit with the early mount code.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 .../admin-guide/kernel-parameters.txt         |   4 +-
 block/Makefile                                |   2 +-
 block/early-lookup.c                          | 224 ++++++++++++++++++
 init/do_mounts.c                              | 219 -----------------
 4 files changed, 227 insertions(+), 222 deletions(-)
 create mode 100644 block/early-lookup.c

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index a6bc31349cbb76..911c54829c7c59 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5453,8 +5453,8 @@
 
 	root=		[KNL] Root filesystem
 			Usually this a a block device specifier of some kind,
-			see the early_lookup_bdev comment in init/do_mounts.c
-			for details.
+			see the early_lookup_bdev comment in
+			block/early-lookup.c for details.
 			Alternatively this can be "ram" for the legacy initial
 			ramdisk, "nfs" and "cifs" for root on a network file
 			system, or "mtd" and "ubi" for mounting from raw flash.
diff --git a/block/Makefile b/block/Makefile
index b31b05390749a1..46ada9dc8bbfe2 100644
--- a/block/Makefile
+++ b/block/Makefile
@@ -9,7 +9,7 @@ obj-y		:= bdev.o fops.o bio.o elevator.o blk-core.o blk-sysfs.o \
 			blk-lib.o blk-mq.o blk-mq-tag.o blk-stat.o \
 			blk-mq-sysfs.o blk-mq-cpumap.o blk-mq-sched.o ioctl.o \
 			genhd.o ioprio.o badblocks.o partitions/ blk-rq-qos.o \
-			disk-events.o blk-ia-ranges.o
+			disk-events.o blk-ia-ranges.o early-lookup.o
 
 obj-$(CONFIG_BOUNCE)		+= bounce.o
 obj-$(CONFIG_BLK_DEV_BSG_COMMON) += bsg.o
diff --git a/block/early-lookup.c b/block/early-lookup.c
new file mode 100644
index 00000000000000..9fc30d039508af
--- /dev/null
+++ b/block/early-lookup.c
@@ -0,0 +1,224 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Code for looking up block devices in the early boot code before mounting the
+ * root file system.  Unfortunately currently also abused in a few other places.
+ */
+#include <linux/blkdev.h>
+#include <linux/ctype.h>
+
+struct uuidcmp {
+	const char *uuid;
+	int len;
+};
+
+/**
+ * match_dev_by_uuid - callback for finding a partition using its uuid
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the desired struct uuidcmp to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_uuid(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const struct uuidcmp *cmp = data;
+
+	if (!bdev->bd_meta_info ||
+	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
+		return 0;
+	return 1;
+}
+
+/**
+ * devt_from_partuuid - looks up the dev_t of a partition by its UUID
+ * @uuid_str:	char array containing ascii UUID
+ *
+ * The function will return the first partition which contains a matching
+ * UUID value in its partition_meta_info struct.  This does not search
+ * by filesystem UUIDs.
+ *
+ * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
+ * extracted and used as an offset from the partition identified by the UUID.
+ *
+ * Returns the matching dev_t on success or 0 on failure.
+ */
+static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+{
+	struct uuidcmp cmp;
+	struct device *dev = NULL;
+	int offset = 0;
+	char *slash;
+
+	cmp.uuid = uuid_str;
+
+	slash = strchr(uuid_str, '/');
+	/* Check for optional partition number offset attributes. */
+	if (slash) {
+		char c = 0;
+
+		/* Explicitly fail on poor PARTUUID syntax. */
+		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
+			goto out_invalid;
+		cmp.len = slash - uuid_str;
+	} else {
+		cmp.len = strlen(uuid_str);
+	}
+
+	if (!cmp.len)
+		goto out_invalid;
+
+	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
+	if (!dev)
+		return -ENODEV;
+
+	if (offset) {
+		/*
+		 * Attempt to find the requested partition by adding an offset
+		 * to the partition number found by UUID.
+		 */
+		*devt = part_devt(dev_to_disk(dev),
+				  dev_to_bdev(dev)->bd_partno + offset);
+	} else {
+		*devt = dev->devt;
+	}
+
+	put_device(dev);
+	return 0;
+
+out_invalid:
+	pr_err("VFS: PARTUUID= is invalid.\n"
+	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
+	return -EINVAL;
+}
+
+/**
+ * match_dev_by_label - callback for finding a partition using its label
+ * @dev:	device passed in by the caller
+ * @data:	opaque pointer to the label to match
+ *
+ * Returns 1 if the device matches, and 0 otherwise.
+ */
+static int match_dev_by_label(struct device *dev, const void *data)
+{
+	struct block_device *bdev = dev_to_bdev(dev);
+	const char *label = data;
+
+	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
+		return 0;
+	return 1;
+}
+
+static int devt_from_partlabel(const char *label, dev_t *devt)
+{
+	struct device *dev;
+
+	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
+	if (!dev)
+		return -ENODEV;
+	*devt = dev->devt;
+	put_device(dev);
+	return 0;
+}
+
+static int devt_from_devname(const char *name, dev_t *devt)
+{
+	int part;
+	char s[32];
+	char *p;
+
+	if (strlen(name) > 31)
+		return -EINVAL;
+	strcpy(s, name);
+	for (p = s; *p; p++) {
+		if (*p == '/')
+			*p = '!';
+	}
+
+	*devt = blk_lookup_devt(s, 0);
+	if (*devt)
+		return 0;
+
+	/*
+	 * Try non-existent, but valid partition, which may only exist after
+	 * opening the device, like partitioned md devices.
+	 */
+	while (p > s && isdigit(p[-1]))
+		p--;
+	if (p == s || !*p || *p == '0')
+		return -EINVAL;
+
+	/* try disk name without <part number> */
+	part = simple_strtoul(p, NULL, 10);
+	*p = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+
+	/* try disk name without p<part number> */
+	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
+		return -EINVAL;
+	p[-1] = '\0';
+	*devt = blk_lookup_devt(s, part);
+	if (*devt)
+		return 0;
+	return -EINVAL;
+}
+
+static int devt_from_devnum(const char *name, dev_t *devt)
+{
+	unsigned maj, min, offset;
+	char *p, dummy;
+
+	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
+		*devt = MKDEV(maj, min);
+		if (maj != MAJOR(*devt) || min != MINOR(*devt))
+			return -EINVAL;
+	} else {
+		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
+		if (*p)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+/*
+ *	Convert a name into device number.  We accept the following variants:
+ *
+ *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
+ *         no leading 0x, for example b302.
+ *	3) /dev/<disk_name> represents the device number of disk
+ *	4) /dev/<disk_name><decimal> represents the device number
+ *         of partition - device number of disk plus the partition number
+ *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
+ *	   used when disk name of partitioned disk ends on a digit.
+ *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
+ *	   unique id of a partition if the partition table provides it.
+ *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
+ *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
+ *	   filled hex representation of the 32-bit "NT disk signature", and PP
+ *	   is a zero-filled hex representation of the 1-based partition number.
+ *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
+ *	   a partition with a known unique id.
+ *	8) <major>:<minor> major and minor number of the device separated by
+ *	   a colon.
+ *	9) PARTLABEL=<name> with name being the GPT partition label.
+ *	   MSDOS partitions do not support labels!
+ *
+ *	If name doesn't have fall into the categories above, we return (0,0).
+ *	block_class is used to check if something is a disk name. If the disk
+ *	name contains slashes, the device name has them replaced with
+ *	bangs.
+ */
+int early_lookup_bdev(const char *name, dev_t *devt)
+{
+	if (strncmp(name, "PARTUUID=", 9) == 0)
+		return devt_from_partuuid(name + 9, devt);
+	if (strncmp(name, "PARTLABEL=", 10) == 0)
+		return devt_from_partlabel(name + 10, devt);
+	if (strncmp(name, "/dev/", 5) == 0)
+		return devt_from_devname(name + 5, devt);
+	return devt_from_devnum(name, devt);
+}
+EXPORT_SYMBOL_GPL(early_lookup_bdev);
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 0b36a5f39ee8e2..780546a6cbfb6f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -59,225 +59,6 @@ static int __init readwrite(char *str)
 __setup("ro", readonly);
 __setup("rw", readwrite);
 
-#ifdef CONFIG_BLOCK
-struct uuidcmp {
-	const char *uuid;
-	int len;
-};
-
-/**
- * match_dev_by_uuid - callback for finding a partition using its uuid
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the desired struct uuidcmp to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_uuid(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const struct uuidcmp *cmp = data;
-
-	if (!bdev->bd_meta_info ||
-	    strncasecmp(cmp->uuid, bdev->bd_meta_info->uuid, cmp->len))
-		return 0;
-	return 1;
-}
-
-/**
- * devt_from_partuuid - looks up the dev_t of a partition by its UUID
- * @uuid_str:	char array containing ascii UUID
- *
- * The function will return the first partition which contains a matching
- * UUID value in its partition_meta_info struct.  This does not search
- * by filesystem UUIDs.
- *
- * If @uuid_str is followed by a "/PARTNROFF=%d", then the number will be
- * extracted and used as an offset from the partition identified by the UUID.
- *
- * Returns the matching dev_t on success or 0 on failure.
- */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
-{
-	struct uuidcmp cmp;
-	struct device *dev = NULL;
-	int offset = 0;
-	char *slash;
-
-	cmp.uuid = uuid_str;
-
-	slash = strchr(uuid_str, '/');
-	/* Check for optional partition number offset attributes. */
-	if (slash) {
-		char c = 0;
-
-		/* Explicitly fail on poor PARTUUID syntax. */
-		if (sscanf(slash + 1, "PARTNROFF=%d%c", &offset, &c) != 1)
-			goto out_invalid;
-		cmp.len = slash - uuid_str;
-	} else {
-		cmp.len = strlen(uuid_str);
-	}
-
-	if (!cmp.len)
-		goto out_invalid;
-
-	dev = class_find_device(&block_class, NULL, &cmp, &match_dev_by_uuid);
-	if (!dev)
-		return -ENODEV;
-
-	if (offset) {
-		/*
-		 * Attempt to find the requested partition by adding an offset
-		 * to the partition number found by UUID.
-		 */
-		*devt = part_devt(dev_to_disk(dev),
-				  dev_to_bdev(dev)->bd_partno + offset);
-	} else {
-		*devt = dev->devt;
-	}
-
-	put_device(dev);
-	return 0;
-
-out_invalid:
-	pr_err("VFS: PARTUUID= is invalid.\n"
-	       "Expected PARTUUID=<valid-uuid-id>[/PARTNROFF=%%d]\n");
-	return -EINVAL;
-}
-
-/**
- * match_dev_by_label - callback for finding a partition using its label
- * @dev:	device passed in by the caller
- * @data:	opaque pointer to the label to match
- *
- * Returns 1 if the device matches, and 0 otherwise.
- */
-static int match_dev_by_label(struct device *dev, const void *data)
-{
-	struct block_device *bdev = dev_to_bdev(dev);
-	const char *label = data;
-
-	if (!bdev->bd_meta_info || strcmp(label, bdev->bd_meta_info->volname))
-		return 0;
-	return 1;
-}
-
-static int devt_from_partlabel(const char *label, dev_t *devt)
-{
-	struct device *dev;
-
-	dev = class_find_device(&block_class, NULL, label, &match_dev_by_label);
-	if (!dev)
-		return -ENODEV;
-	*devt = dev->devt;
-	put_device(dev);
-	return 0;
-}
-
-static int devt_from_devname(const char *name, dev_t *devt)
-{
-	int part;
-	char s[32];
-	char *p;
-
-	if (strlen(name) > 31)
-		return -EINVAL;
-	strcpy(s, name);
-	for (p = s; *p; p++) {
-		if (*p == '/')
-			*p = '!';
-	}
-
-	*devt = blk_lookup_devt(s, 0);
-	if (*devt)
-		return 0;
-
-	/*
-	 * Try non-existent, but valid partition, which may only exist after
-	 * opening the device, like partitioned md devices.
-	 */
-	while (p > s && isdigit(p[-1]))
-		p--;
-	if (p == s || !*p || *p == '0')
-		return -EINVAL;
-
-	/* try disk name without <part number> */
-	part = simple_strtoul(p, NULL, 10);
-	*p = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-
-	/* try disk name without p<part number> */
-	if (p < s + 2 || !isdigit(p[-2]) || p[-1] != 'p')
-		return -EINVAL;
-	p[-1] = '\0';
-	*devt = blk_lookup_devt(s, part);
-	if (*devt)
-		return 0;
-	return -EINVAL;
-}
-
-static int devt_from_devnum(const char *name, dev_t *devt)
-{
-	unsigned maj, min, offset;
-	char *p, dummy;
-
-	if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
-	    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset, &dummy) == 3) {
-		*devt = MKDEV(maj, min);
-		if (maj != MAJOR(*devt) || min != MINOR(*devt))
-			return -EINVAL;
-	} else {
-		*devt = new_decode_dev(simple_strtoul(name, &p, 16));
-		if (*p)
-			return -EINVAL;
-	}
-
-	return 0;
-}
-
-/*
- *	Convert a name into device number.  We accept the following variants:
- *
- *	1) <hex_major><hex_minor> device number in hexadecimal represents itself
- *         no leading 0x, for example b302.
- *	3) /dev/<disk_name> represents the device number of disk
- *	4) /dev/<disk_name><decimal> represents the device number
- *         of partition - device number of disk plus the partition number
- *	5) /dev/<disk_name>p<decimal> - same as the above, that form is
- *	   used when disk name of partitioned disk ends on a digit.
- *	6) PARTUUID=00112233-4455-6677-8899-AABBCCDDEEFF representing the
- *	   unique id of a partition if the partition table provides it.
- *	   The UUID may be either an EFI/GPT UUID, or refer to an MSDOS
- *	   partition using the format SSSSSSSS-PP, where SSSSSSSS is a zero-
- *	   filled hex representation of the 32-bit "NT disk signature", and PP
- *	   is a zero-filled hex representation of the 1-based partition number.
- *	7) PARTUUID=<UUID>/PARTNROFF=<int> to select a partition in relation to
- *	   a partition with a known unique id.
- *	8) <major>:<minor> major and minor number of the device separated by
- *	   a colon.
- *	9) PARTLABEL=<name> with name being the GPT partition label.
- *	   MSDOS partitions do not support labels!
- *
- *	If name doesn't have fall into the categories above, we return (0,0).
- *	block_class is used to check if something is a disk name. If the disk
- *	name contains slashes, the device name has them replaced with
- *	bangs.
- */
-int early_lookup_bdev(const char *name, dev_t *devt)
-{
-	if (strncmp(name, "PARTUUID=", 9) == 0)
-		return devt_from_partuuid(name + 9, devt);
-	if (strncmp(name, "PARTLABEL=", 10) == 0)
-		return devt_from_partlabel(name + 10, devt);
-	if (strncmp(name, "/dev/", 5) == 0)
-		return devt_from_devname(name + 5, devt);
-	return devt_from_devnum(name, devt);
-}
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
-#endif
-
 static int __init root_dev_setup(char *line)
 {
 	strscpy(saved_root_name, line, sizeof(saved_root_name));
-- 
2.39.2

[dm-devel] [PATCH 16/24] block: move more code to early-lookup.c

[Christoph Hellwig]

blk_lookup_devt is only used by code in early-lookup.c, so move it
there.

printk_all_partitions and it's helper bdevt_str are only used by the
early init code in init/do_mounts.c, so they should go there as well.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 92 ++++++++++++++++++++++++++++++++++++++++++
 block/genhd.c          | 92 ------------------------------------------
 include/linux/blkdev.h |  1 -
 3 files changed, 92 insertions(+), 93 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 9fc30d039508af..6016e781b6a0e2 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -120,6 +120,35 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
+static dev_t blk_lookup_devt(const char *name, int partno)
+{
+	dev_t devt = MKDEV(0, 0);
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+
+		if (strcmp(dev_name(dev), name))
+			continue;
+
+		if (partno < disk->minors) {
+			/* We need to return the right devno, even
+			 * if the partition doesn't exist yet.
+			 */
+			devt = MKDEV(MAJOR(dev->devt),
+				     MINOR(dev->devt) + partno);
+		} else {
+			devt = part_devt(disk, partno);
+			if (devt)
+				break;
+		}
+	}
+	class_dev_iter_exit(&iter);
+	return devt;
+}
+
 static int devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
@@ -222,3 +251,66 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 	return devt_from_devnum(name, devt);
 }
 EXPORT_SYMBOL_GPL(early_lookup_bdev);
+
+static char __init *bdevt_str(dev_t devt, char *buf)
+{
+	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
+		char tbuf[BDEVT_SIZE];
+		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
+		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
+	} else
+		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
+
+	return buf;
+}
+
+/*
+ * print a full list of all partitions - intended for places where the root
+ * filesystem can't be mounted and thus to give the victim some idea of what
+ * went wrong
+ */
+void __init printk_all_partitions(void)
+{
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+		struct block_device *part;
+		char devt_buf[BDEVT_SIZE];
+		unsigned long idx;
+
+		/*
+		 * Don't show empty devices or things that have been
+		 * suppressed
+		 */
+		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
+			continue;
+
+		/*
+		 * Note, unlike /proc/partitions, I am showing the numbers in
+		 * hex - the same format as the root= option takes.
+		 */
+		rcu_read_lock();
+		xa_for_each(&disk->part_tbl, idx, part) {
+			if (!bdev_nr_sectors(part))
+				continue;
+			printk("%s%s %10llu %pg %s",
+			       bdev_is_partition(part) ? "  " : "",
+			       bdevt_str(part->bd_dev, devt_buf),
+			       bdev_nr_sectors(part) >> 1, part,
+			       part->bd_meta_info ?
+					part->bd_meta_info->uuid : "");
+			if (bdev_is_partition(part))
+				printk("\n");
+			else if (dev->parent && dev->parent->driver)
+				printk(" driver: %s\n",
+					dev->parent->driver->name);
+			else
+				printk(" (driver?)\n");
+		}
+		rcu_read_unlock();
+	}
+	class_dev_iter_exit(&iter);
+}
diff --git a/block/genhd.c b/block/genhd.c
index 1cb489b927d50a..aa28f296fe391b 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -318,18 +318,6 @@ void blk_free_ext_minor(unsigned int minor)
 	ida_free(&ext_devt_ida, minor);
 }
 
-static char *bdevt_str(dev_t devt, char *buf)
-{
-	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
-		char tbuf[BDEVT_SIZE];
-		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
-		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
-	} else
-		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
-
-	return buf;
-}
-
 void disk_uevent(struct gendisk *disk, enum kobject_action action)
 {
 	struct block_device *part;
@@ -755,57 +743,6 @@ void blk_request_module(dev_t devt)
 }
 #endif /* CONFIG_BLOCK_LEGACY_AUTOLOAD */
 
-/*
- * print a full list of all partitions - intended for places where the root
- * filesystem can't be mounted and thus to give the victim some idea of what
- * went wrong
- */
-void __init printk_all_partitions(void)
-{
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-		struct block_device *part;
-		char devt_buf[BDEVT_SIZE];
-		unsigned long idx;
-
-		/*
-		 * Don't show empty devices or things that have been
-		 * suppressed
-		 */
-		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
-			continue;
-
-		/*
-		 * Note, unlike /proc/partitions, I am showing the numbers in
-		 * hex - the same format as the root= option takes.
-		 */
-		rcu_read_lock();
-		xa_for_each(&disk->part_tbl, idx, part) {
-			if (!bdev_nr_sectors(part))
-				continue;
-			printk("%s%s %10llu %pg %s",
-			       bdev_is_partition(part) ? "  " : "",
-			       bdevt_str(part->bd_dev, devt_buf),
-			       bdev_nr_sectors(part) >> 1, part,
-			       part->bd_meta_info ?
-					part->bd_meta_info->uuid : "");
-			if (bdev_is_partition(part))
-				printk("\n");
-			else if (dev->parent && dev->parent->driver)
-				printk(" driver: %s\n",
-					dev->parent->driver->name);
-			else
-				printk(" (driver?)\n");
-		}
-		rcu_read_unlock();
-	}
-	class_dev_iter_exit(&iter);
-}
-
 #ifdef CONFIG_PROC_FS
 /* iterator */
 static void *disk_seqf_start(struct seq_file *seqf, loff_t *pos)
@@ -1339,35 +1276,6 @@ dev_t part_devt(struct gendisk *disk, u8 partno)
 	return devt;
 }
 
-dev_t blk_lookup_devt(const char *name, int partno)
-{
-	dev_t devt = MKDEV(0, 0);
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-
-		if (strcmp(dev_name(dev), name))
-			continue;
-
-		if (partno < disk->minors) {
-			/* We need to return the right devno, even
-			 * if the partition doesn't exist yet.
-			 */
-			devt = MKDEV(MAJOR(dev->devt),
-				     MINOR(dev->devt) + partno);
-		} else {
-			devt = part_devt(disk, partno);
-			if (devt)
-				break;
-		}
-	}
-	class_dev_iter_exit(&iter);
-	return devt;
-}
-
 struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
 		struct lock_class_key *lkclass)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0bda6cb98d7eb8..0ce09ecfa3a7af 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -838,7 +838,6 @@ static inline void bd_unlink_disk_holder(struct block_device *bdev,
 
 dev_t part_devt(struct gendisk *disk, u8 partno);
 void inc_diskseq(struct gendisk *disk);
-dev_t blk_lookup_devt(const char *name, int partno);
 void blk_request_module(dev_t devt);
 
 extern int blk_register_queue(struct gendisk *disk);
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 16/24] block: move more code to early-lookup.c

[Christoph Hellwig]

blk_lookup_devt is only used by code in early-lookup.c, so move it
there.

printk_all_partitions and it's helper bdevt_str are only used by the
early init code in init/do_mounts.c, so they should go there as well.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 92 ++++++++++++++++++++++++++++++++++++++++++
 block/genhd.c          | 92 ------------------------------------------
 include/linux/blkdev.h |  1 -
 3 files changed, 92 insertions(+), 93 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 9fc30d039508af..6016e781b6a0e2 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -120,6 +120,35 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
+static dev_t blk_lookup_devt(const char *name, int partno)
+{
+	dev_t devt = MKDEV(0, 0);
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+
+		if (strcmp(dev_name(dev), name))
+			continue;
+
+		if (partno < disk->minors) {
+			/* We need to return the right devno, even
+			 * if the partition doesn't exist yet.
+			 */
+			devt = MKDEV(MAJOR(dev->devt),
+				     MINOR(dev->devt) + partno);
+		} else {
+			devt = part_devt(disk, partno);
+			if (devt)
+				break;
+		}
+	}
+	class_dev_iter_exit(&iter);
+	return devt;
+}
+
 static int devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
@@ -222,3 +251,66 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 	return devt_from_devnum(name, devt);
 }
 EXPORT_SYMBOL_GPL(early_lookup_bdev);
+
+static char __init *bdevt_str(dev_t devt, char *buf)
+{
+	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
+		char tbuf[BDEVT_SIZE];
+		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
+		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
+	} else
+		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
+
+	return buf;
+}
+
+/*
+ * print a full list of all partitions - intended for places where the root
+ * filesystem can't be mounted and thus to give the victim some idea of what
+ * went wrong
+ */
+void __init printk_all_partitions(void)
+{
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+		struct block_device *part;
+		char devt_buf[BDEVT_SIZE];
+		unsigned long idx;
+
+		/*
+		 * Don't show empty devices or things that have been
+		 * suppressed
+		 */
+		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
+			continue;
+
+		/*
+		 * Note, unlike /proc/partitions, I am showing the numbers in
+		 * hex - the same format as the root= option takes.
+		 */
+		rcu_read_lock();
+		xa_for_each(&disk->part_tbl, idx, part) {
+			if (!bdev_nr_sectors(part))
+				continue;
+			printk("%s%s %10llu %pg %s",
+			       bdev_is_partition(part) ? "  " : "",
+			       bdevt_str(part->bd_dev, devt_buf),
+			       bdev_nr_sectors(part) >> 1, part,
+			       part->bd_meta_info ?
+					part->bd_meta_info->uuid : "");
+			if (bdev_is_partition(part))
+				printk("\n");
+			else if (dev->parent && dev->parent->driver)
+				printk(" driver: %s\n",
+					dev->parent->driver->name);
+			else
+				printk(" (driver?)\n");
+		}
+		rcu_read_unlock();
+	}
+	class_dev_iter_exit(&iter);
+}
diff --git a/block/genhd.c b/block/genhd.c
index 1cb489b927d50a..aa28f296fe391b 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -318,18 +318,6 @@ void blk_free_ext_minor(unsigned int minor)
 	ida_free(&ext_devt_ida, minor);
 }
 
-static char *bdevt_str(dev_t devt, char *buf)
-{
-	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
-		char tbuf[BDEVT_SIZE];
-		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
-		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
-	} else
-		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
-
-	return buf;
-}
-
 void disk_uevent(struct gendisk *disk, enum kobject_action action)
 {
 	struct block_device *part;
@@ -755,57 +743,6 @@ void blk_request_module(dev_t devt)
 }
 #endif /* CONFIG_BLOCK_LEGACY_AUTOLOAD */
 
-/*
- * print a full list of all partitions - intended for places where the root
- * filesystem can't be mounted and thus to give the victim some idea of what
- * went wrong
- */
-void __init printk_all_partitions(void)
-{
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-		struct block_device *part;
-		char devt_buf[BDEVT_SIZE];
-		unsigned long idx;
-
-		/*
-		 * Don't show empty devices or things that have been
-		 * suppressed
-		 */
-		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
-			continue;
-
-		/*
-		 * Note, unlike /proc/partitions, I am showing the numbers in
-		 * hex - the same format as the root= option takes.
-		 */
-		rcu_read_lock();
-		xa_for_each(&disk->part_tbl, idx, part) {
-			if (!bdev_nr_sectors(part))
-				continue;
-			printk("%s%s %10llu %pg %s",
-			       bdev_is_partition(part) ? "  " : "",
-			       bdevt_str(part->bd_dev, devt_buf),
-			       bdev_nr_sectors(part) >> 1, part,
-			       part->bd_meta_info ?
-					part->bd_meta_info->uuid : "");
-			if (bdev_is_partition(part))
-				printk("\n");
-			else if (dev->parent && dev->parent->driver)
-				printk(" driver: %s\n",
-					dev->parent->driver->name);
-			else
-				printk(" (driver?)\n");
-		}
-		rcu_read_unlock();
-	}
-	class_dev_iter_exit(&iter);
-}
-
 #ifdef CONFIG_PROC_FS
 /* iterator */
 static void *disk_seqf_start(struct seq_file *seqf, loff_t *pos)
@@ -1339,35 +1276,6 @@ dev_t part_devt(struct gendisk *disk, u8 partno)
 	return devt;
 }
 
-dev_t blk_lookup_devt(const char *name, int partno)
-{
-	dev_t devt = MKDEV(0, 0);
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-
-		if (strcmp(dev_name(dev), name))
-			continue;
-
-		if (partno < disk->minors) {
-			/* We need to return the right devno, even
-			 * if the partition doesn't exist yet.
-			 */
-			devt = MKDEV(MAJOR(dev->devt),
-				     MINOR(dev->devt) + partno);
-		} else {
-			devt = part_devt(disk, partno);
-			if (devt)
-				break;
-		}
-	}
-	class_dev_iter_exit(&iter);
-	return devt;
-}
-
 struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
 		struct lock_class_key *lkclass)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0bda6cb98d7eb8..0ce09ecfa3a7af 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -838,7 +838,6 @@ static inline void bd_unlink_disk_holder(struct block_device *bdev,
 
 dev_t part_devt(struct gendisk *disk, u8 partno);
 void inc_diskseq(struct gendisk *disk);
-dev_t blk_lookup_devt(const char *name, int partno);
 void blk_request_module(dev_t devt);
 
 extern int blk_register_queue(struct gendisk *disk);
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 16/24] block: move more code to early-lookup.c

[Christoph Hellwig]

blk_lookup_devt is only used by code in early-lookup.c, so move it
there.

printk_all_partitions and it's helper bdevt_str are only used by the
early init code in init/do_mounts.c, so they should go there as well.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 92 ++++++++++++++++++++++++++++++++++++++++++
 block/genhd.c          | 92 ------------------------------------------
 include/linux/blkdev.h |  1 -
 3 files changed, 92 insertions(+), 93 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 9fc30d039508af..6016e781b6a0e2 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -120,6 +120,35 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
+static dev_t blk_lookup_devt(const char *name, int partno)
+{
+	dev_t devt = MKDEV(0, 0);
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+
+		if (strcmp(dev_name(dev), name))
+			continue;
+
+		if (partno < disk->minors) {
+			/* We need to return the right devno, even
+			 * if the partition doesn't exist yet.
+			 */
+			devt = MKDEV(MAJOR(dev->devt),
+				     MINOR(dev->devt) + partno);
+		} else {
+			devt = part_devt(disk, partno);
+			if (devt)
+				break;
+		}
+	}
+	class_dev_iter_exit(&iter);
+	return devt;
+}
+
 static int devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
@@ -222,3 +251,66 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 	return devt_from_devnum(name, devt);
 }
 EXPORT_SYMBOL_GPL(early_lookup_bdev);
+
+static char __init *bdevt_str(dev_t devt, char *buf)
+{
+	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
+		char tbuf[BDEVT_SIZE];
+		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
+		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
+	} else
+		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
+
+	return buf;
+}
+
+/*
+ * print a full list of all partitions - intended for places where the root
+ * filesystem can't be mounted and thus to give the victim some idea of what
+ * went wrong
+ */
+void __init printk_all_partitions(void)
+{
+	struct class_dev_iter iter;
+	struct device *dev;
+
+	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
+	while ((dev = class_dev_iter_next(&iter))) {
+		struct gendisk *disk = dev_to_disk(dev);
+		struct block_device *part;
+		char devt_buf[BDEVT_SIZE];
+		unsigned long idx;
+
+		/*
+		 * Don't show empty devices or things that have been
+		 * suppressed
+		 */
+		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
+			continue;
+
+		/*
+		 * Note, unlike /proc/partitions, I am showing the numbers in
+		 * hex - the same format as the root= option takes.
+		 */
+		rcu_read_lock();
+		xa_for_each(&disk->part_tbl, idx, part) {
+			if (!bdev_nr_sectors(part))
+				continue;
+			printk("%s%s %10llu %pg %s",
+			       bdev_is_partition(part) ? "  " : "",
+			       bdevt_str(part->bd_dev, devt_buf),
+			       bdev_nr_sectors(part) >> 1, part,
+			       part->bd_meta_info ?
+					part->bd_meta_info->uuid : "");
+			if (bdev_is_partition(part))
+				printk("\n");
+			else if (dev->parent && dev->parent->driver)
+				printk(" driver: %s\n",
+					dev->parent->driver->name);
+			else
+				printk(" (driver?)\n");
+		}
+		rcu_read_unlock();
+	}
+	class_dev_iter_exit(&iter);
+}
diff --git a/block/genhd.c b/block/genhd.c
index 1cb489b927d50a..aa28f296fe391b 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -318,18 +318,6 @@ void blk_free_ext_minor(unsigned int minor)
 	ida_free(&ext_devt_ida, minor);
 }
 
-static char *bdevt_str(dev_t devt, char *buf)
-{
-	if (MAJOR(devt) <= 0xff && MINOR(devt) <= 0xff) {
-		char tbuf[BDEVT_SIZE];
-		snprintf(tbuf, BDEVT_SIZE, "%02x%02x", MAJOR(devt), MINOR(devt));
-		snprintf(buf, BDEVT_SIZE, "%-9s", tbuf);
-	} else
-		snprintf(buf, BDEVT_SIZE, "%03x:%05x", MAJOR(devt), MINOR(devt));
-
-	return buf;
-}
-
 void disk_uevent(struct gendisk *disk, enum kobject_action action)
 {
 	struct block_device *part;
@@ -755,57 +743,6 @@ void blk_request_module(dev_t devt)
 }
 #endif /* CONFIG_BLOCK_LEGACY_AUTOLOAD */
 
-/*
- * print a full list of all partitions - intended for places where the root
- * filesystem can't be mounted and thus to give the victim some idea of what
- * went wrong
- */
-void __init printk_all_partitions(void)
-{
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-		struct block_device *part;
-		char devt_buf[BDEVT_SIZE];
-		unsigned long idx;
-
-		/*
-		 * Don't show empty devices or things that have been
-		 * suppressed
-		 */
-		if (get_capacity(disk) == 0 || (disk->flags & GENHD_FL_HIDDEN))
-			continue;
-
-		/*
-		 * Note, unlike /proc/partitions, I am showing the numbers in
-		 * hex - the same format as the root= option takes.
-		 */
-		rcu_read_lock();
-		xa_for_each(&disk->part_tbl, idx, part) {
-			if (!bdev_nr_sectors(part))
-				continue;
-			printk("%s%s %10llu %pg %s",
-			       bdev_is_partition(part) ? "  " : "",
-			       bdevt_str(part->bd_dev, devt_buf),
-			       bdev_nr_sectors(part) >> 1, part,
-			       part->bd_meta_info ?
-					part->bd_meta_info->uuid : "");
-			if (bdev_is_partition(part))
-				printk("\n");
-			else if (dev->parent && dev->parent->driver)
-				printk(" driver: %s\n",
-					dev->parent->driver->name);
-			else
-				printk(" (driver?)\n");
-		}
-		rcu_read_unlock();
-	}
-	class_dev_iter_exit(&iter);
-}
-
 #ifdef CONFIG_PROC_FS
 /* iterator */
 static void *disk_seqf_start(struct seq_file *seqf, loff_t *pos)
@@ -1339,35 +1276,6 @@ dev_t part_devt(struct gendisk *disk, u8 partno)
 	return devt;
 }
 
-dev_t blk_lookup_devt(const char *name, int partno)
-{
-	dev_t devt = MKDEV(0, 0);
-	struct class_dev_iter iter;
-	struct device *dev;
-
-	class_dev_iter_init(&iter, &block_class, NULL, &disk_type);
-	while ((dev = class_dev_iter_next(&iter))) {
-		struct gendisk *disk = dev_to_disk(dev);
-
-		if (strcmp(dev_name(dev), name))
-			continue;
-
-		if (partno < disk->minors) {
-			/* We need to return the right devno, even
-			 * if the partition doesn't exist yet.
-			 */
-			devt = MKDEV(MAJOR(dev->devt),
-				     MINOR(dev->devt) + partno);
-		} else {
-			devt = part_devt(disk, partno);
-			if (devt)
-				break;
-		}
-	}
-	class_dev_iter_exit(&iter);
-	return devt;
-}
-
 struct gendisk *__alloc_disk_node(struct request_queue *q, int node_id,
 		struct lock_class_key *lkclass)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0bda6cb98d7eb8..0ce09ecfa3a7af 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -838,7 +838,6 @@ static inline void bd_unlink_disk_holder(struct block_device *bdev,
 
 dev_t part_devt(struct gendisk *disk, u8 partno);
 void inc_diskseq(struct gendisk *disk);
-dev_t blk_lookup_devt(const char *name, int partno);
 void blk_request_module(dev_t devt);
 
 extern int blk_register_queue(struct gendisk *disk);
-- 
2.39.2

[dm-devel] [PATCH 17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr

[Christoph Hellwig]

Use the block_device acquired in dm_get_device for the check instead
of doing an extra lookup.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-snap.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 9c49f53760d066..7832974b73eb03 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -1241,7 +1241,6 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	int i;
 	int r = -EINVAL;
 	char *origin_path, *cow_path;
-	dev_t origin_dev, cow_dev;
 	unsigned int args_used, num_flush_bios = 1;
 	fmode_t origin_mode = FMODE_READ;
 
@@ -1279,24 +1278,21 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		ti->error = "Cannot get origin device";
 		goto bad_origin;
 	}
-	origin_dev = s->origin->bdev->bd_dev;
 
 	cow_path = argv[0];
 	argv++;
 	argc--;
 
-	cow_dev = dm_get_dev_t(cow_path);
-	if (cow_dev && cow_dev == origin_dev) {
-		ti->error = "COW device cannot be the same as origin device";
-		r = -EINVAL;
-		goto bad_cow;
-	}
-
 	r = dm_get_device(ti, cow_path, dm_table_get_mode(ti->table), &s->cow);
 	if (r) {
 		ti->error = "Cannot get COW device";
 		goto bad_cow;
 	}
+	if (s->cow->bdev && s->cow->bdev == s->origin->bdev) {
+		ti->error = "COW device cannot be the same as origin device";
+		r = -EINVAL;
+		goto bad_store;
+	}
 
 	r = dm_exception_store_create(ti, argc, argv, s, &args_used, &s->store);
 	if (r) {
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr

[Christoph Hellwig]

Use the block_device acquired in dm_get_device for the check instead
of doing an extra lookup.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-snap.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 9c49f53760d066..7832974b73eb03 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -1241,7 +1241,6 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	int i;
 	int r = -EINVAL;
 	char *origin_path, *cow_path;
-	dev_t origin_dev, cow_dev;
 	unsigned int args_used, num_flush_bios = 1;
 	fmode_t origin_mode = FMODE_READ;
 
@@ -1279,24 +1278,21 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		ti->error = "Cannot get origin device";
 		goto bad_origin;
 	}
-	origin_dev = s->origin->bdev->bd_dev;
 
 	cow_path = argv[0];
 	argv++;
 	argc--;
 
-	cow_dev = dm_get_dev_t(cow_path);
-	if (cow_dev && cow_dev == origin_dev) {
-		ti->error = "COW device cannot be the same as origin device";
-		r = -EINVAL;
-		goto bad_cow;
-	}
-
 	r = dm_get_device(ti, cow_path, dm_table_get_mode(ti->table), &s->cow);
 	if (r) {
 		ti->error = "Cannot get COW device";
 		goto bad_cow;
 	}
+	if (s->cow->bdev && s->cow->bdev == s->origin->bdev) {
+		ti->error = "COW device cannot be the same as origin device";
+		r = -EINVAL;
+		goto bad_store;
+	}
 
 	r = dm_exception_store_create(ti, argc, argv, s, &args_used, &s->store);
 	if (r) {
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 17/24] dm-snap: simplify the origin_dev == cow_dev check in snapshot_ctr

[Christoph Hellwig]

Use the block_device acquired in dm_get_device for the check instead
of doing an extra lookup.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-snap.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index 9c49f53760d066..7832974b73eb03 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -1241,7 +1241,6 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	int i;
 	int r = -EINVAL;
 	char *origin_path, *cow_path;
-	dev_t origin_dev, cow_dev;
 	unsigned int args_used, num_flush_bios = 1;
 	fmode_t origin_mode = FMODE_READ;
 
@@ -1279,24 +1278,21 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		ti->error = "Cannot get origin device";
 		goto bad_origin;
 	}
-	origin_dev = s->origin->bdev->bd_dev;
 
 	cow_path = argv[0];
 	argv++;
 	argc--;
 
-	cow_dev = dm_get_dev_t(cow_path);
-	if (cow_dev && cow_dev == origin_dev) {
-		ti->error = "COW device cannot be the same as origin device";
-		r = -EINVAL;
-		goto bad_cow;
-	}
-
 	r = dm_get_device(ti, cow_path, dm_table_get_mode(ti->table), &s->cow);
 	if (r) {
 		ti->error = "Cannot get COW device";
 		goto bad_cow;
 	}
+	if (s->cow->bdev && s->cow->bdev == s->origin->bdev) {
+		ti->error = "COW device cannot be the same as origin device";
+		r = -EINVAL;
+		goto bad_store;
+	}
 
 	r = dm_exception_store_create(ti, argc, argv, s, &args_used, &s->store);
 	if (r) {
-- 
2.39.2

[dm-devel] [PATCH 18/24] dm: open code dm_get_dev_t in dm_init_init

[Christoph Hellwig]

dm_init_init is called from early boot code, and thus lookup_bdev
will never succeed.  Just open code that call to early_lookup_bdev
instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-init.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-init.c b/drivers/md/dm-init.c
index d369457dbed0ed..2a71bcdba92d14 100644
--- a/drivers/md/dm-init.c
+++ b/drivers/md/dm-init.c
@@ -293,8 +293,10 @@ static int __init dm_init_init(void)
 
 	for (i = 0; i < ARRAY_SIZE(waitfor); i++) {
 		if (waitfor[i]) {
+			dev_t dev;
+
 			DMINFO("waiting for device %s ...", waitfor[i]);
-			while (!dm_get_dev_t(waitfor[i]))
+			while (early_lookup_bdev(waitfor[i], &dev))
 				fsleep(5000);
 		}
 	}
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 18/24] dm: open code dm_get_dev_t in dm_init_init

[Christoph Hellwig]

dm_init_init is called from early boot code, and thus lookup_bdev
will never succeed.  Just open code that call to early_lookup_bdev
instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-init.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-init.c b/drivers/md/dm-init.c
index d369457dbed0ed..2a71bcdba92d14 100644
--- a/drivers/md/dm-init.c
+++ b/drivers/md/dm-init.c
@@ -293,8 +293,10 @@ static int __init dm_init_init(void)
 
 	for (i = 0; i < ARRAY_SIZE(waitfor); i++) {
 		if (waitfor[i]) {
+			dev_t dev;
+
 			DMINFO("waiting for device %s ...", waitfor[i]);
-			while (!dm_get_dev_t(waitfor[i]))
+			while (early_lookup_bdev(waitfor[i], &dev))
 				fsleep(5000);
 		}
 	}
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 18/24] dm: open code dm_get_dev_t in dm_init_init

[Christoph Hellwig]

dm_init_init is called from early boot code, and thus lookup_bdev
will never succeed.  Just open code that call to early_lookup_bdev
instead.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-init.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm-init.c b/drivers/md/dm-init.c
index d369457dbed0ed..2a71bcdba92d14 100644
--- a/drivers/md/dm-init.c
+++ b/drivers/md/dm-init.c
@@ -293,8 +293,10 @@ static int __init dm_init_init(void)
 
 	for (i = 0; i < ARRAY_SIZE(waitfor); i++) {
 		if (waitfor[i]) {
+			dev_t dev;
+
 			DMINFO("waiting for device %s ...", waitfor[i]);
-			while (!dm_get_dev_t(waitfor[i]))
+			while (early_lookup_bdev(waitfor[i], &dev))
 				fsleep(5000);
 		}
 	}
-- 
2.39.2

[dm-devel] [PATCH 19/24] dm: remove dm_get_dev_t

[Christoph Hellwig]

Open code dm_get_dev_t in the only remaining caller, and propagate the
exact error code from lookup_bdev and early_lookup_bdev.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/md/dm-table.c         | 22 +++++-----------------
 include/linux/device-mapper.h |  2 --
 2 files changed, 5 insertions(+), 19 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 05aa16da43b0d5..1576b408768d4b 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -323,20 +323,6 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 	return 0;
 }
 
-/*
- * Convert the path to a device
- */
-dev_t dm_get_dev_t(const char *path)
-{
-	dev_t dev;
-
-	if (lookup_bdev(path, &dev) &&
-	    early_lookup_bdev(path, &dev))
-		return 0;
-	return dev;
-}
-EXPORT_SYMBOL_GPL(dm_get_dev_t);
-
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
@@ -359,9 +345,11 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		if (MAJOR(dev) != major || MINOR(dev) != minor)
 			return -EOVERFLOW;
 	} else {
-		dev = dm_get_dev_t(path);
-		if (!dev)
-			return -ENODEV;
+		r = lookup_bdev(path, &dev);
+		if (r)
+			r = early_lookup_bdev(path, &dev);
+		if (r)
+			return r;
 	}
 	if (dev == disk_devt(t->md->disk))
 		return -EINVAL;
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index a52d2b9a68460a..c27b84002d8382 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -170,8 +170,6 @@ struct dm_dev {
 	char name[16];
 };
 
-dev_t dm_get_dev_t(const char *path);
-
 /*
  * Constructors should call these functions to ensure destination devices
  * are opened/closed correctly.
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 19/24] dm: remove dm_get_dev_t

[Christoph Hellwig]

Open code dm_get_dev_t in the only remaining caller, and propagate the
exact error code from lookup_bdev and early_lookup_bdev.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/md/dm-table.c         | 22 +++++-----------------
 include/linux/device-mapper.h |  2 --
 2 files changed, 5 insertions(+), 19 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 05aa16da43b0d5..1576b408768d4b 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -323,20 +323,6 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 	return 0;
 }
 
-/*
- * Convert the path to a device
- */
-dev_t dm_get_dev_t(const char *path)
-{
-	dev_t dev;
-
-	if (lookup_bdev(path, &dev) &&
-	    early_lookup_bdev(path, &dev))
-		return 0;
-	return dev;
-}
-EXPORT_SYMBOL_GPL(dm_get_dev_t);
-
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
@@ -359,9 +345,11 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		if (MAJOR(dev) != major || MINOR(dev) != minor)
 			return -EOVERFLOW;
 	} else {
-		dev = dm_get_dev_t(path);
-		if (!dev)
-			return -ENODEV;
+		r = lookup_bdev(path, &dev);
+		if (r)
+			r = early_lookup_bdev(path, &dev);
+		if (r)
+			return r;
 	}
 	if (dev == disk_devt(t->md->disk))
 		return -EINVAL;
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index a52d2b9a68460a..c27b84002d8382 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -170,8 +170,6 @@ struct dm_dev {
 	char name[16];
 };
 
-dev_t dm_get_dev_t(const char *path);
-
 /*
  * Constructors should call these functions to ensure destination devices
  * are opened/closed correctly.
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 19/24] dm: remove dm_get_dev_t

[Christoph Hellwig]

Open code dm_get_dev_t in the only remaining caller, and propagate the
exact error code from lookup_bdev and early_lookup_bdev.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/md/dm-table.c         | 22 +++++-----------------
 include/linux/device-mapper.h |  2 --
 2 files changed, 5 insertions(+), 19 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 05aa16da43b0d5..1576b408768d4b 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -323,20 +323,6 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 	return 0;
 }
 
-/*
- * Convert the path to a device
- */
-dev_t dm_get_dev_t(const char *path)
-{
-	dev_t dev;
-
-	if (lookup_bdev(path, &dev) &&
-	    early_lookup_bdev(path, &dev))
-		return 0;
-	return dev;
-}
-EXPORT_SYMBOL_GPL(dm_get_dev_t);
-
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
@@ -359,9 +345,11 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		if (MAJOR(dev) != major || MINOR(dev) != minor)
 			return -EOVERFLOW;
 	} else {
-		dev = dm_get_dev_t(path);
-		if (!dev)
-			return -ENODEV;
+		r = lookup_bdev(path, &dev);
+		if (r)
+			r = early_lookup_bdev(path, &dev);
+		if (r)
+			return r;
 	}
 	if (dev == disk_devt(t->md->disk))
 		return -EINVAL;
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index a52d2b9a68460a..c27b84002d8382 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -170,8 +170,6 @@ struct dm_dev {
 	char name[16];
 };
 
-dev_t dm_get_dev_t(const char *path);
-
 /*
  * Constructors should call these functions to ensure destination devices
  * are opened/closed correctly.
-- 
2.39.2

[dm-devel] [PATCH 20/24] dm: only call early_lookup_bdev from early boot context

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but dm_get_device calls it as a general fallback when lookup_bdev
fails, which is problematic because early_lookup_bdev bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to only call early_lookup_bdev when dm is built-in and the system
state in not running yet.  This means it is still available when tables
are constructed by dm-init.c from the kernel command line, but not
otherwise.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-table.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1576b408768d4b..2fd5826bfce175 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -326,8 +326,11 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
+ *
+ * Note: the __ref annotation is because this function can call the __init
+ * marked early_lookup_bdev when called during early boot code from dm-init.c.
  */
-int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
+int __ref dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		  struct dm_dev **result)
 {
 	int r;
@@ -346,8 +349,10 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 			return -EOVERFLOW;
 	} else {
 		r = lookup_bdev(path, &dev);
-		if (r)
+#ifndef MODULE
+		if (r && system_state < SYSTEM_RUNNING)
 			r = early_lookup_bdev(path, &dev);
+#endif
 		if (r)
 			return r;
 	}
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 20/24] dm: only call early_lookup_bdev from early boot context

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but dm_get_device calls it as a general fallback when lookup_bdev
fails, which is problematic because early_lookup_bdev bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to only call early_lookup_bdev when dm is built-in and the system
state in not running yet.  This means it is still available when tables
are constructed by dm-init.c from the kernel command line, but not
otherwise.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-table.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1576b408768d4b..2fd5826bfce175 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -326,8 +326,11 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
+ *
+ * Note: the __ref annotation is because this function can call the __init
+ * marked early_lookup_bdev when called during early boot code from dm-init.c.
  */
-int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
+int __ref dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		  struct dm_dev **result)
 {
 	int r;
@@ -346,8 +349,10 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 			return -EOVERFLOW;
 	} else {
 		r = lookup_bdev(path, &dev);
-		if (r)
+#ifndef MODULE
+		if (r && system_state < SYSTEM_RUNNING)
 			r = early_lookup_bdev(path, &dev);
+#endif
 		if (r)
 			return r;
 	}
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 20/24] dm: only call early_lookup_bdev from early boot context

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but dm_get_device calls it as a general fallback when lookup_bdev
fails, which is problematic because early_lookup_bdev bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to only call early_lookup_bdev when dm is built-in and the system
state in not running yet.  This means it is still available when tables
are constructed by dm-init.c from the kernel command line, but not
otherwise.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Snitzer <snitzer@kernel.org>
---
 drivers/md/dm-table.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index 1576b408768d4b..2fd5826bfce175 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -326,8 +326,11 @@ static int upgrade_mode(struct dm_dev_internal *dd, fmode_t new_mode,
 /*
  * Add a device to the list, or just increment the usage count if
  * it's already present.
+ *
+ * Note: the __ref annotation is because this function can call the __init
+ * marked early_lookup_bdev when called during early boot code from dm-init.c.
  */
-int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
+int __ref dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 		  struct dm_dev **result)
 {
 	int r;
@@ -346,8 +349,10 @@ int dm_get_device(struct dm_target *ti, const char *path, fmode_t mode,
 			return -EOVERFLOW;
 	} else {
 		r = lookup_bdev(path, &dev);
-		if (r)
+#ifndef MODULE
+		if (r && system_state < SYSTEM_RUNNING)
 			r = early_lookup_bdev(path, &dev);
+#endif
 		if (r)
 			return r;
 	}
-- 
2.39.2

[dm-devel] [PATCH 21/24] PM: hibernate: don't use early_lookup_bdev in resume_store

[Christoph Hellwig]

resume_store is a sysfs attribute written during normal kernel runtime,
and it should not use the early_lookup_bdev API that bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to lookup_bdev, which does a normal path lookup instead, and fall
back to trying to parse a numeric dev_t just like early_lookup_bdev did.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Fixes: 421a5fa1a6cf ("PM / hibernate: use name_to_dev_t to parse resume")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index c52dedb9f7c8e8..7ae95ec72f9902 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1178,7 +1178,23 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	error = early_lookup_bdev(name, &dev);
+	error = lookup_bdev(name, &dev);
+	if (error) {
+		unsigned maj, min, offset;
+		char *p, dummy;
+
+		if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+		    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset,
+				&dummy) == 3) {
+			dev = MKDEV(maj, min);
+			if (maj != MAJOR(dev) || min != MINOR(dev))
+				error = -EINVAL;
+		} else {
+			dev = new_decode_dev(simple_strtoul(name, &p, 16));
+			if (*p)
+				error = -EINVAL;
+		}
+	}
 	kfree(name);
 	if (error)
 		return error;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 21/24] PM: hibernate: don't use early_lookup_bdev in resume_store

[Christoph Hellwig]

resume_store is a sysfs attribute written during normal kernel runtime,
and it should not use the early_lookup_bdev API that bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to lookup_bdev, which does a normal path lookup instead, and fall
back to trying to parse a numeric dev_t just like early_lookup_bdev did.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Fixes: 421a5fa1a6cf ("PM / hibernate: use name_to_dev_t to parse resume")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index c52dedb9f7c8e8..7ae95ec72f9902 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1178,7 +1178,23 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	error = early_lookup_bdev(name, &dev);
+	error = lookup_bdev(name, &dev);
+	if (error) {
+		unsigned maj, min, offset;
+		char *p, dummy;
+
+		if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+		    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset,
+				&dummy) == 3) {
+			dev = MKDEV(maj, min);
+			if (maj != MAJOR(dev) || min != MINOR(dev))
+				error = -EINVAL;
+		} else {
+			dev = new_decode_dev(simple_strtoul(name, &p, 16));
+			if (*p)
+				error = -EINVAL;
+		}
+	}
 	kfree(name);
 	if (error)
 		return error;
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 21/24] PM: hibernate: don't use early_lookup_bdev in resume_store

[Christoph Hellwig]

resume_store is a sysfs attribute written during normal kernel runtime,
and it should not use the early_lookup_bdev API that bypasses all normal
path based permission checking, and might cause problems with certain
container environments renaming devices.

Switch to lookup_bdev, which does a normal path lookup instead, and fall
back to trying to parse a numeric dev_t just like early_lookup_bdev did.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Fixes: 421a5fa1a6cf ("PM / hibernate: use name_to_dev_t to parse resume")
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
---
 kernel/power/hibernate.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index c52dedb9f7c8e8..7ae95ec72f9902 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -1178,7 +1178,23 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
 	if (!name)
 		return -ENOMEM;
 
-	error = early_lookup_bdev(name, &dev);
+	error = lookup_bdev(name, &dev);
+	if (error) {
+		unsigned maj, min, offset;
+		char *p, dummy;
+
+		if (sscanf(name, "%u:%u%c", &maj, &min, &dummy) == 2 ||
+		    sscanf(name, "%u:%u:%u:%c", &maj, &min, &offset,
+				&dummy) == 3) {
+			dev = MKDEV(maj, min);
+			if (maj != MAJOR(dev) || min != MINOR(dev))
+				error = -EINVAL;
+		} else {
+			dev = new_decode_dev(simple_strtoul(name, &p, 16));
+			if (*p)
+				error = -EINVAL;
+		}
+	}
 	kfree(name);
 	if (error)
 		return error;
-- 
2.39.2

[dm-devel] [PATCH 22/24] mtd: block2mtd: factor the early block device open logic into a helper

[Christoph Hellwig]

Simplify add_device a bit by splitting out the cumbersome early boot logic
into a separate helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 53 +++++++++++++++++++--------------
 1 file changed, 30 insertions(+), 23 deletions(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4c21e9f13bead5..182eed68c75634 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,34 +215,18 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-
-static struct block2mtd_dev *add_device(char *devname, int erase_size,
-		char *label, int timeout)
+static struct block_device *mdtblock_early_get_bdev(const char *devname,
+		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
+	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
-#endif
-	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
-	struct block_device *bdev;
-	struct block2mtd_dev *dev;
-	char *name;
-
-	if (!devname)
-		return NULL;
-
-	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
-	if (!dev)
-		return NULL;
 
-	/* Get a handle on the device */
-	bdev = blkdev_get_by_path(devname, mode, dev);
-
-#ifndef MODULE
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
 	 */
-	for (i = 0; IS_ERR(bdev) && i <= timeout; i++) {
+	for (i = 0; i <= timeout; i++) {
 		dev_t devt;
 
 		if (i)
@@ -254,12 +238,35 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		if (early_lookup_bdev(devname, &devt))
-			continue;
-		bdev = blkdev_get_by_dev(devt, mode, dev);
+		if (!early_lookup_bdev(devname, &devt)) {
+			bdev = blkdev_get_by_dev(devt, mode, dev);
+			if (!IS_ERR(bdev))
+				break;
+		}
 	}
 #endif
+	return bdev;
+}
+
+static struct block2mtd_dev *add_device(char *devname, int erase_size,
+		char *label, int timeout)
+{
+	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
+	struct block_device *bdev;
+	struct block2mtd_dev *dev;
+	char *name;
 
+	if (!devname)
+		return NULL;
+
+	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
+	if (!dev)
+		return NULL;
+
+	/* Get a handle on the device */
+	bdev = blkdev_get_by_path(devname, mode, dev);
+	if (IS_ERR(bdev))
+		bdev = mdtblock_early_get_bdev(devname, mode, timeout, dev);
 	if (IS_ERR(bdev)) {
 		pr_err("error: cannot open device %s\n", devname);
 		goto err_free_block2mtd;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 22/24] mtd: block2mtd: factor the early block device open logic into a helper

[Christoph Hellwig]

Simplify add_device a bit by splitting out the cumbersome early boot logic
into a separate helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 53 +++++++++++++++++++--------------
 1 file changed, 30 insertions(+), 23 deletions(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4c21e9f13bead5..182eed68c75634 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,34 +215,18 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-
-static struct block2mtd_dev *add_device(char *devname, int erase_size,
-		char *label, int timeout)
+static struct block_device *mdtblock_early_get_bdev(const char *devname,
+		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
+	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
-#endif
-	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
-	struct block_device *bdev;
-	struct block2mtd_dev *dev;
-	char *name;
-
-	if (!devname)
-		return NULL;
-
-	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
-	if (!dev)
-		return NULL;
 
-	/* Get a handle on the device */
-	bdev = blkdev_get_by_path(devname, mode, dev);
-
-#ifndef MODULE
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
 	 */
-	for (i = 0; IS_ERR(bdev) && i <= timeout; i++) {
+	for (i = 0; i <= timeout; i++) {
 		dev_t devt;
 
 		if (i)
@@ -254,12 +238,35 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		if (early_lookup_bdev(devname, &devt))
-			continue;
-		bdev = blkdev_get_by_dev(devt, mode, dev);
+		if (!early_lookup_bdev(devname, &devt)) {
+			bdev = blkdev_get_by_dev(devt, mode, dev);
+			if (!IS_ERR(bdev))
+				break;
+		}
 	}
 #endif
+	return bdev;
+}
+
+static struct block2mtd_dev *add_device(char *devname, int erase_size,
+		char *label, int timeout)
+{
+	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
+	struct block_device *bdev;
+	struct block2mtd_dev *dev;
+	char *name;
 
+	if (!devname)
+		return NULL;
+
+	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
+	if (!dev)
+		return NULL;
+
+	/* Get a handle on the device */
+	bdev = blkdev_get_by_path(devname, mode, dev);
+	if (IS_ERR(bdev))
+		bdev = mdtblock_early_get_bdev(devname, mode, timeout, dev);
 	if (IS_ERR(bdev)) {
 		pr_err("error: cannot open device %s\n", devname);
 		goto err_free_block2mtd;
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 22/24] mtd: block2mtd: factor the early block device open logic into a helper

[Christoph Hellwig]

Simplify add_device a bit by splitting out the cumbersome early boot logic
into a separate helper.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 53 +++++++++++++++++++--------------
 1 file changed, 30 insertions(+), 23 deletions(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 4c21e9f13bead5..182eed68c75634 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,34 +215,18 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-
-static struct block2mtd_dev *add_device(char *devname, int erase_size,
-		char *label, int timeout)
+static struct block_device *mdtblock_early_get_bdev(const char *devname,
+		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
+	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
-#endif
-	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
-	struct block_device *bdev;
-	struct block2mtd_dev *dev;
-	char *name;
-
-	if (!devname)
-		return NULL;
-
-	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
-	if (!dev)
-		return NULL;
 
-	/* Get a handle on the device */
-	bdev = blkdev_get_by_path(devname, mode, dev);
-
-#ifndef MODULE
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
 	 */
-	for (i = 0; IS_ERR(bdev) && i <= timeout; i++) {
+	for (i = 0; i <= timeout; i++) {
 		dev_t devt;
 
 		if (i)
@@ -254,12 +238,35 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 			msleep(1000);
 		wait_for_device_probe();
 
-		if (early_lookup_bdev(devname, &devt))
-			continue;
-		bdev = blkdev_get_by_dev(devt, mode, dev);
+		if (!early_lookup_bdev(devname, &devt)) {
+			bdev = blkdev_get_by_dev(devt, mode, dev);
+			if (!IS_ERR(bdev))
+				break;
+		}
 	}
 #endif
+	return bdev;
+}
+
+static struct block2mtd_dev *add_device(char *devname, int erase_size,
+		char *label, int timeout)
+{
+	const fmode_t mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL;
+	struct block_device *bdev;
+	struct block2mtd_dev *dev;
+	char *name;
 
+	if (!devname)
+		return NULL;
+
+	dev = kzalloc(sizeof(struct block2mtd_dev), GFP_KERNEL);
+	if (!dev)
+		return NULL;
+
+	/* Get a handle on the device */
+	bdev = blkdev_get_by_path(devname, mode, dev);
+	if (IS_ERR(bdev))
+		bdev = mdtblock_early_get_bdev(devname, mode, timeout, dev);
 	if (IS_ERR(bdev)) {
 		pr_err("error: cannot open device %s\n", devname);
 		goto err_free_block2mtd;
-- 
2.39.2

[dm-devel] [PATCH 23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but mdtblock_early_get_bdev is called as a general fallback when
lookup_bdev fails, which is problematic because early_lookup_bdev
bypasses all normal path based permission checking, and might cause
problems with certain container environments renaming devices.

Switch to only call early_lookup_bdev when block2mtd is built-in and the
system state in not running yet.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 182eed68c75634..59e4c71cfc6f53 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,13 +215,23 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-static struct block_device *mdtblock_early_get_bdev(const char *devname,
+/*
+ * This function is marked __ref because it calls the __init marked
+ * early_lookup_bdev when called from the early boot code.
+ */
+static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
 	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
 
+	/*
+	 * We can't use early_lookup_bdev from a running system.
+	 */
+	if (system_state >= SYSTEM_RUNNING)
+		return bdev;
+
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but mdtblock_early_get_bdev is called as a general fallback when
lookup_bdev fails, which is problematic because early_lookup_bdev
bypasses all normal path based permission checking, and might cause
problems with certain container environments renaming devices.

Switch to only call early_lookup_bdev when block2mtd is built-in and the
system state in not running yet.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 182eed68c75634..59e4c71cfc6f53 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,13 +215,23 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-static struct block_device *mdtblock_early_get_bdev(const char *devname,
+/*
+ * This function is marked __ref because it calls the __init marked
+ * early_lookup_bdev when called from the early boot code.
+ */
+static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
 	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
 
+	/*
+	 * We can't use early_lookup_bdev from a running system.
+	 */
+	if (system_state >= SYSTEM_RUNNING)
+		return bdev;
+
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 23/24] mtd: block2mtd: don't call early_lookup_bdev after the system is running

[Christoph Hellwig]

early_lookup_bdev is supposed to only be called from the early boot
code, but mdtblock_early_get_bdev is called as a general fallback when
lookup_bdev fails, which is problematic because early_lookup_bdev
bypasses all normal path based permission checking, and might cause
problems with certain container environments renaming devices.

Switch to only call early_lookup_bdev when block2mtd is built-in and the
system state in not running yet.

Note that this strictly speaking changes the kernel ABI as the PARTUUID=
and PARTLABEL= style syntax is now not available during a running
systems.  They never were intended for that, but this breaks things
we'll have to figure out a way to make them available again.  But if
avoidable in any way I'd rather avoid that.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
---
 drivers/mtd/devices/block2mtd.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 182eed68c75634..59e4c71cfc6f53 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -215,13 +215,23 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev);
 }
 
-static struct block_device *mdtblock_early_get_bdev(const char *devname,
+/*
+ * This function is marked __ref because it calls the __init marked
+ * early_lookup_bdev when called from the early boot code.
+ */
+static struct block_device __ref *mdtblock_early_get_bdev(const char *devname,
 		fmode_t mode, int timeout, struct block2mtd_dev *dev)
 {
 	struct block_device *bdev = ERR_PTR(-ENODEV);
 #ifndef MODULE
 	int i;
 
+	/*
+	 * We can't use early_lookup_bdev from a running system.
+	 */
+	if (system_state >= SYSTEM_RUNNING)
+		return bdev;
+
 	/*
 	 * We might not have the root device mounted at this point.
 	 * Try to resolve the device name by other means.
-- 
2.39.2

[dm-devel] [PATCH 24/24] block: mark early_lookup_bdev as __init

[Christoph Hellwig]

early_lookup_bdev is now only used during the early boot code as it
should, so mark it __init to not waste run time memory on it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 19 +++++++++----------
 include/linux/blkdev.h |  2 +-
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 6016e781b6a0e2..3ff0d2e4dcbfb8 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
  * Code for looking up block devices in the early boot code before mounting the
- * root file system.  Unfortunately currently also abused in a few other places.
+ * root file system.
  */
 #include <linux/blkdev.h>
 #include <linux/ctype.h>
@@ -18,7 +18,7 @@ struct uuidcmp {
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_uuid(struct device *dev, const void *data)
+static int __init match_dev_by_uuid(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const struct uuidcmp *cmp = data;
@@ -42,7 +42,7 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+static int __init devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
@@ -98,7 +98,7 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_label(struct device *dev, const void *data)
+static int __init match_dev_by_label(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const char *label = data;
@@ -108,7 +108,7 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static int devt_from_partlabel(const char *label, dev_t *devt)
+static int __init devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
 
@@ -120,7 +120,7 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
-static dev_t blk_lookup_devt(const char *name, int partno)
+static dev_t __init blk_lookup_devt(const char *name, int partno)
 {
 	dev_t devt = MKDEV(0, 0);
 	struct class_dev_iter iter;
@@ -149,7 +149,7 @@ static dev_t blk_lookup_devt(const char *name, int partno)
 	return devt;
 }
 
-static int devt_from_devname(const char *name, dev_t *devt)
+static int __init devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
 	char s[32];
@@ -193,7 +193,7 @@ static int devt_from_devname(const char *name, dev_t *devt)
 	return -EINVAL;
 }
 
-static int devt_from_devnum(const char *name, dev_t *devt)
+static int __init devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
 	char *p, dummy;
@@ -240,7 +240,7 @@ static int devt_from_devnum(const char *name, dev_t *devt)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-int early_lookup_bdev(const char *name, dev_t *devt)
+int __init early_lookup_bdev(const char *name, dev_t *devt)
 {
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9, devt);
@@ -250,7 +250,6 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 		return devt_from_devname(name + 5, devt);
 	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
 
 static char __init *bdevt_str(dev_t devt, char *buf)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0ce09ecfa3a7af..ca84a135ddc80f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1494,7 +1494,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
-int early_lookup_bdev(const char *pathname, dev_t *dev);
+int __init early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[PATCH 24/24] block: mark early_lookup_bdev as __init

[Christoph Hellwig]

early_lookup_bdev is now only used during the early boot code as it
should, so mark it __init to not waste run time memory on it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 19 +++++++++----------
 include/linux/blkdev.h |  2 +-
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 6016e781b6a0e2..3ff0d2e4dcbfb8 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
  * Code for looking up block devices in the early boot code before mounting the
- * root file system.  Unfortunately currently also abused in a few other places.
+ * root file system.
  */
 #include <linux/blkdev.h>
 #include <linux/ctype.h>
@@ -18,7 +18,7 @@ struct uuidcmp {
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_uuid(struct device *dev, const void *data)
+static int __init match_dev_by_uuid(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const struct uuidcmp *cmp = data;
@@ -42,7 +42,7 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+static int __init devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
@@ -98,7 +98,7 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_label(struct device *dev, const void *data)
+static int __init match_dev_by_label(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const char *label = data;
@@ -108,7 +108,7 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static int devt_from_partlabel(const char *label, dev_t *devt)
+static int __init devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
 
@@ -120,7 +120,7 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
-static dev_t blk_lookup_devt(const char *name, int partno)
+static dev_t __init blk_lookup_devt(const char *name, int partno)
 {
 	dev_t devt = MKDEV(0, 0);
 	struct class_dev_iter iter;
@@ -149,7 +149,7 @@ static dev_t blk_lookup_devt(const char *name, int partno)
 	return devt;
 }
 
-static int devt_from_devname(const char *name, dev_t *devt)
+static int __init devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
 	char s[32];
@@ -193,7 +193,7 @@ static int devt_from_devname(const char *name, dev_t *devt)
 	return -EINVAL;
 }
 
-static int devt_from_devnum(const char *name, dev_t *devt)
+static int __init devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
 	char *p, dummy;
@@ -240,7 +240,7 @@ static int devt_from_devnum(const char *name, dev_t *devt)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-int early_lookup_bdev(const char *name, dev_t *devt)
+int __init early_lookup_bdev(const char *name, dev_t *devt)
 {
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9, devt);
@@ -250,7 +250,6 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 		return devt_from_devname(name + 5, devt);
 	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
 
 static char __init *bdevt_str(dev_t devt, char *buf)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0ce09ecfa3a7af..ca84a135ddc80f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1494,7 +1494,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
-int early_lookup_bdev(const char *pathname, dev_t *dev);
+int __init early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
-- 
2.39.2


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

[PATCH 24/24] block: mark early_lookup_bdev as __init

[Christoph Hellwig]

early_lookup_bdev is now only used during the early boot code as it
should, so mark it __init to not waste run time memory on it.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 block/early-lookup.c   | 19 +++++++++----------
 include/linux/blkdev.h |  2 +-
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/block/early-lookup.c b/block/early-lookup.c
index 6016e781b6a0e2..3ff0d2e4dcbfb8 100644
--- a/block/early-lookup.c
+++ b/block/early-lookup.c
@@ -1,7 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0-only
 /*
  * Code for looking up block devices in the early boot code before mounting the
- * root file system.  Unfortunately currently also abused in a few other places.
+ * root file system.
  */
 #include <linux/blkdev.h>
 #include <linux/ctype.h>
@@ -18,7 +18,7 @@ struct uuidcmp {
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_uuid(struct device *dev, const void *data)
+static int __init match_dev_by_uuid(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const struct uuidcmp *cmp = data;
@@ -42,7 +42,7 @@ static int match_dev_by_uuid(struct device *dev, const void *data)
  *
  * Returns the matching dev_t on success or 0 on failure.
  */
-static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
+static int __init devt_from_partuuid(const char *uuid_str, dev_t *devt)
 {
 	struct uuidcmp cmp;
 	struct device *dev = NULL;
@@ -98,7 +98,7 @@ static int devt_from_partuuid(const char *uuid_str, dev_t *devt)
  *
  * Returns 1 if the device matches, and 0 otherwise.
  */
-static int match_dev_by_label(struct device *dev, const void *data)
+static int __init match_dev_by_label(struct device *dev, const void *data)
 {
 	struct block_device *bdev = dev_to_bdev(dev);
 	const char *label = data;
@@ -108,7 +108,7 @@ static int match_dev_by_label(struct device *dev, const void *data)
 	return 1;
 }
 
-static int devt_from_partlabel(const char *label, dev_t *devt)
+static int __init devt_from_partlabel(const char *label, dev_t *devt)
 {
 	struct device *dev;
 
@@ -120,7 +120,7 @@ static int devt_from_partlabel(const char *label, dev_t *devt)
 	return 0;
 }
 
-static dev_t blk_lookup_devt(const char *name, int partno)
+static dev_t __init blk_lookup_devt(const char *name, int partno)
 {
 	dev_t devt = MKDEV(0, 0);
 	struct class_dev_iter iter;
@@ -149,7 +149,7 @@ static dev_t blk_lookup_devt(const char *name, int partno)
 	return devt;
 }
 
-static int devt_from_devname(const char *name, dev_t *devt)
+static int __init devt_from_devname(const char *name, dev_t *devt)
 {
 	int part;
 	char s[32];
@@ -193,7 +193,7 @@ static int devt_from_devname(const char *name, dev_t *devt)
 	return -EINVAL;
 }
 
-static int devt_from_devnum(const char *name, dev_t *devt)
+static int __init devt_from_devnum(const char *name, dev_t *devt)
 {
 	unsigned maj, min, offset;
 	char *p, dummy;
@@ -240,7 +240,7 @@ static int devt_from_devnum(const char *name, dev_t *devt)
  *	name contains slashes, the device name has them replaced with
  *	bangs.
  */
-int early_lookup_bdev(const char *name, dev_t *devt)
+int __init early_lookup_bdev(const char *name, dev_t *devt)
 {
 	if (strncmp(name, "PARTUUID=", 9) == 0)
 		return devt_from_partuuid(name + 9, devt);
@@ -250,7 +250,6 @@ int early_lookup_bdev(const char *name, dev_t *devt)
 		return devt_from_devname(name + 5, devt);
 	return devt_from_devnum(name, devt);
 }
-EXPORT_SYMBOL_GPL(early_lookup_bdev);
 
 static char __init *bdevt_str(dev_t devt, char *buf)
 {
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 0ce09ecfa3a7af..ca84a135ddc80f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1494,7 +1494,7 @@ int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
 void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
-int early_lookup_bdev(const char *pathname, dev_t *dev);
+int __init early_lookup_bdev(const char *pathname, dev_t *dev);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
-- 
2.39.2

[dm-devel] fix the name_to_dev_t mess

[Christoph Hellwig]

Hi all,

this series tries to sort out accumulated mess around the name_to_dev_t
function.  This function is intended to allow looking up the dev_t of a
block device based on a name string before the root file systems is
mounted and thus the normal path based lookup is available.

Unfortunately a few years ago it managed to get exported and used in
non-init contexts, leading to the something looking like a path name
also beeing lookuped up by a different and potential dangerous
algorithm.

This series does a fair amount of refactoring and finally ends up with
the renamed and improved name_to_dev_t only beeing available for the
early init code again.

The series is against Jens' for-6.5/block tree but probably applies
against current mainline just fine as well.

A git tree is also available here:

    git://git.infradead.org/users/hch/block.git blk-init-cleanup

Gitweb:

    http://git.infradead.org/users/hch/block.git/shortlog/refs/heads/blk-init-cleanup

Diffstat:
 Documentation/admin-guide/kernel-parameters.txt |    2 
 arch/alpha/kernel/setup.c                       |    2 
 arch/ia64/kernel/setup.c                        |    2 
 arch/powerpc/platforms/powermac/setup.c         |    3 
 block/Makefile                                  |    2 
 block/early-lookup.c                            |  315 ++++++++++++++++++
 block/genhd.c                                   |   92 -----
 drivers/base/dd.c                               |    6 
 drivers/md/dm-init.c                            |    4 
 drivers/md/dm-snap.c                            |   14 
 drivers/md/dm-table.c                           |   26 -
 drivers/md/md-autodetect.c                      |    3 
 drivers/mtd/devices/block2mtd.c                 |   62 ++-
 fs/pstore/blk.c                                 |    4 
 include/linux/blkdev.h                          |    6 
 include/linux/device-mapper.h                   |    2 
 include/linux/device/driver.h                   |    2 
 include/linux/mount.h                           |    1 
 include/linux/root_dev.h                        |    9 
 init/do_mounts.c                                |  416 ++++++------------------
 init/do_mounts.h                                |   14 
 init/do_mounts_initrd.c                         |   11 
 kernel/power/hibernate.c                        |  167 +++++----
 kernel/power/power.h                            |    3 
 kernel/power/swap.c                             |    2 
 25 files changed, 603 insertions(+), 567 deletions(-)

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

[dm-devel] [PATCH 03/24] PM: hibernate: remove the global snapshot_test variable

[Christoph Hellwig]

Passing call dependent variable in global variables is a huge
antipattern.  Fix it up.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 kernel/power/hibernate.c | 17 ++++++-----------
 kernel/power/power.h     |  3 +--
 kernel/power/swap.c      |  2 +-
 3 files changed, 8 insertions(+), 14 deletions(-)

diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 07279506366255..78696aa04f5ca3 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -64,7 +64,6 @@ enum {
 static int hibernation_mode = HIBERNATION_SHUTDOWN;
 
 bool freezer_test_done;
-bool snapshot_test;
 
 static const struct platform_hibernation_ops *hibernation_ops;
 
@@ -684,7 +683,7 @@ static void power_down(void)
 		cpu_relax();
 }
 
-static int load_image_and_restore(void)
+static int load_image_and_restore(bool snapshot_test)
 {
 	int error;
 	unsigned int flags;
@@ -721,6 +720,7 @@ static int load_image_and_restore(void)
  */
 int hibernate(void)
 {
+	bool snapshot_test = false;
 	unsigned int sleep_flags;
 	int error;
 
@@ -748,9 +748,6 @@ int hibernate(void)
 	if (error)
 		goto Exit;
 
-	/* protected by system_transition_mutex */
-	snapshot_test = false;
-
 	lock_device_hotplug();
 	/* Allocate memory management structures */
 	error = create_basic_memory_bitmaps();
@@ -792,9 +789,9 @@ int hibernate(void)
 	unlock_device_hotplug();
 	if (snapshot_test) {
 		pm_pr_dbg("Checking hibernation image\n");
-		error = swsusp_check();
+		error = swsusp_check(snapshot_test);
 		if (!error)
-			error = load_image_and_restore();
+			error = load_image_and_restore(snapshot_test);
 	}
 	thaw_processes();
 
@@ -982,8 +979,6 @@ static int software_resume(void)
 	 */
 	mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
 
-	snapshot_test = false;
-
 	if (!swsusp_resume_device) {
 		error = find_resume_device();
 		if (error)
@@ -994,7 +989,7 @@ static int software_resume(void)
 		MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
 
 	pm_pr_dbg("Looking for hibernation image.\n");
-	error = swsusp_check();
+	error = swsusp_check(false);
 	if (error)
 		goto Unlock;
 
@@ -1022,7 +1017,7 @@ static int software_resume(void)
 		goto Close_Finish;
 	}
 
-	error = load_image_and_restore();
+	error = load_image_and_restore(false);
 	thaw_processes();
  Finish:
 	pm_notifier_call_chain(PM_POST_RESTORE);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index b83c8d5e188dec..978189fcafd124 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -59,7 +59,6 @@ asmlinkage int swsusp_save(void);
 
 /* kernel/power/hibernate.c */
 extern bool freezer_test_done;
-extern bool snapshot_test;
 
 extern int hibernation_snapshot(int platform_mode);
 extern int hibernation_restore(int platform_mode);
@@ -174,7 +173,7 @@ extern int swsusp_swap_in_use(void);
 #define SF_HW_SIG		8
 
 /* kernel/power/hibernate.c */
-extern int swsusp_check(void);
+int swsusp_check(bool snapshot_test);
 extern void swsusp_free(void);
 extern int swsusp_read(unsigned int *flags_p);
 extern int swsusp_write(unsigned int flags);
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 92e41ed292ada8..efed11568bfc72 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -1514,7 +1514,7 @@ int swsusp_read(unsigned int *flags_p)
  *      swsusp_check - Check for swsusp signature in the resume device
  */
 
-int swsusp_check(void)
+int swsusp_check(bool snapshot_test)
 {
 	int error;
 	void *holder;
-- 
2.39.2

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel

Re: [dm-devel] [PATCH 03/24] PM: hibernate: remove the global snapshot_test variable

[Rafael J. Wysocki]

On Tue, May 23, 2023 at 9:45 AM Christoph Hellwig <hch@lst.de> wrote:
>
> Passing call dependent variable in global variables is a huge
> antipattern.  Fix it up.
>
> Signed-off-by: Christoph Hellwig <hch@lst.de>

Acked-by: Rafael J. Wysocki <rafael@kernel.org>

> ---
>  kernel/power/hibernate.c | 17 ++++++-----------
>  kernel/power/power.h     |  3 +--
>  kernel/power/swap.c      |  2 +-
>  3 files changed, 8 insertions(+), 14 deletions(-)
>
> diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
> index 07279506366255..78696aa04f5ca3 100644
> --- a/kernel/power/hibernate.c
> +++ b/kernel/power/hibernate.c
> @@ -64,7 +64,6 @@ enum {
>  static int hibernation_mode = HIBERNATION_SHUTDOWN;
>
>  bool freezer_test_done;
> -bool snapshot_test;
>
>  static const struct platform_hibernation_ops *hibernation_ops;
>
> @@ -684,7 +683,7 @@ static void power_down(void)
>                 cpu_relax();
>  }
>
> -static int load_image_and_restore(void)
> +static int load_image_and_restore(bool snapshot_test)
>  {
>         int error;
>         unsigned int flags;
> @@ -721,6 +720,7 @@ static int load_image_and_restore(void)
>   */
>  int hibernate(void)
>  {
> +       bool snapshot_test = false;
>         unsigned int sleep_flags;
>         int error;
>
> @@ -748,9 +748,6 @@ int hibernate(void)
>         if (error)
>                 goto Exit;
>
> -       /* protected by system_transition_mutex */
> -       snapshot_test = false;
> -
>         lock_device_hotplug();
>         /* Allocate memory management structures */
>         error = create_basic_memory_bitmaps();
> @@ -792,9 +789,9 @@ int hibernate(void)
>         unlock_device_hotplug();
>         if (snapshot_test) {
>                 pm_pr_dbg("Checking hibernation image\n");
> -               error = swsusp_check();
> +               error = swsusp_check(snapshot_test);
>                 if (!error)
> -                       error = load_image_and_restore();
> +                       error = load_image_and_restore(snapshot_test);
>         }
>         thaw_processes();
>
> @@ -982,8 +979,6 @@ static int software_resume(void)
>          */
>         mutex_lock_nested(&system_transition_mutex, SINGLE_DEPTH_NESTING);
>
> -       snapshot_test = false;
> -
>         if (!swsusp_resume_device) {
>                 error = find_resume_device();
>                 if (error)
> @@ -994,7 +989,7 @@ static int software_resume(void)
>                 MAJOR(swsusp_resume_device), MINOR(swsusp_resume_device));
>
>         pm_pr_dbg("Looking for hibernation image.\n");
> -       error = swsusp_check();
> +       error = swsusp_check(false);
>         if (error)
>                 goto Unlock;
>
> @@ -1022,7 +1017,7 @@ static int software_resume(void)
>                 goto Close_Finish;
>         }
>
> -       error = load_image_and_restore();
> +       error = load_image_and_restore(false);
>         thaw_processes();
>   Finish:
>         pm_notifier_call_chain(PM_POST_RESTORE);
> diff --git a/kernel/power/power.h b/kernel/power/power.h
> index b83c8d5e188dec..978189fcafd124 100644
> --- a/kernel/power/power.h
> +++ b/kernel/power/power.h
> @@ -59,7 +59,6 @@ asmlinkage int swsusp_save(void);
>
>  /* kernel/power/hibernate.c */
>  extern bool freezer_test_done;
> -extern bool snapshot_test;
>
>  extern int hibernation_snapshot(int platform_mode);
>  extern int hibernation_restore(int platform_mode);
> @@ -174,7 +173,7 @@ extern int swsusp_swap_in_use(void);
>  #define SF_HW_SIG              8
>
>  /* kernel/power/hibernate.c */
> -extern int swsusp_check(void);
> +int swsusp_check(bool snapshot_test);
>  extern void swsusp_free(void);
>  extern int swsusp_read(unsigned int *flags_p);
>  extern int swsusp_write(unsigned int flags);
> diff --git a/kernel/power/swap.c b/kernel/power/swap.c
> index 92e41ed292ada8..efed11568bfc72 100644
> --- a/kernel/power/swap.c
> +++ b/kernel/power/swap.c
> @@ -1514,7 +1514,7 @@ int swsusp_read(unsigned int *flags_p)
>   *      swsusp_check - Check for swsusp signature in the resume device
>   */
>
> -int swsusp_check(void)
> +int swsusp_check(bool snapshot_test)
>  {
>         int error;
>         void *holder;
> --
> 2.39.2
>

--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel