Re: [PATCH] ext4: Check superblock mapped prior to committing

2018-07-02 Thread Theodore Y. Ts'o
On Fri, Jun 29, 2018 at 01:36:35PM -0600, Jon Derrick wrote: > This patch attempts to close a hole leading to a BUG seen with hot > removals during writes [1]. > > A block device (NVME namespace in this test case) is formatted to EXT4 > without partitions. It's mounted and write I/O is run to a

Re: [PATCH] ext4: Check superblock mapped prior to committing

2018-07-02 Thread Theodore Y. Ts'o
On Fri, Jun 29, 2018 at 01:36:35PM -0600, Jon Derrick wrote: > This patch attempts to close a hole leading to a BUG seen with hot > removals during writes [1]. > > A block device (NVME namespace in this test case) is formatted to EXT4 > without partitions. It's mounted and write I/O is run to a

Re: [PATCH] ext4: Check superblock mapped prior to committing

2018-06-29 Thread Andreas Dilger
On Jun 29, 2018, at 1:36 PM, Jon Derrick wrote: > > This patch attempts to close a hole leading to a BUG seen with hot > removals during writes [1]. > > A block device (NVME namespace in this test case) is formatted to EXT4 > without partitions. It's mounted and write I/O is run to a file, then

Re: [PATCH] ext4: Check superblock mapped prior to committing

2018-06-29 Thread Andreas Dilger
On Jun 29, 2018, at 1:36 PM, Jon Derrick wrote: > > This patch attempts to close a hole leading to a BUG seen with hot > removals during writes [1]. > > A block device (NVME namespace in this test case) is formatted to EXT4 > without partitions. It's mounted and write I/O is run to a file, then

[PATCH] ext4: Check superblock mapped prior to committing

2018-06-29 Thread Jon Derrick
This patch attempts to close a hole leading to a BUG seen with hot removals during writes [1]. A block device (NVME namespace in this test case) is formatted to EXT4 without partitions. It's mounted and write I/O is run to a file, then the device is hot removed from the slot. The superblock

[PATCH] ext4: Check superblock mapped prior to committing

2018-06-29 Thread Jon Derrick
This patch attempts to close a hole leading to a BUG seen with hot removals during writes [1]. A block device (NVME namespace in this test case) is formatted to EXT4 without partitions. It's mounted and write I/O is run to a file, then the device is hot removed from the slot. The superblock