lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <f48683d9-7854-ba5f-da3a-7ef987a539b8@windriver.com>
Date:   Wed, 11 Mar 2020 12:03:43 +0800
From:   He Zhe <zhe.he@...driver.com>
To:     Christoph Hellwig <hch@....de>
Cc:     jack@...e.cz, Jens Axboe <axboe@...nel.dk>,
        viro@...iv.linux.org.uk, bvanassche@....org, keith.busch@...el.com,
        tglx@...utronix.de, mwilck@...e.com, yuyufen@...wei.com,
        linux-block@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: disk revalidation updates and OOM



On 3/11/20 12:26 AM, Christoph Hellwig wrote:
> On Tue, Mar 10, 2020 at 11:30:27PM +0800, He Zhe wrote:
>>> So this is the exact requirement of commits to be reverted from a bisect
>>> or just a first guess?
>> Many commits failed to build or boot during bisection.
>>
>> At least the following four have to be reverted to make it work.
>>
>> 979c690d block: move clearing bd_invalidated into check_disk_size_change
>> f0b870d block: remove (__)blkdev_reread_part as an exported API
>> 142fe8f block: fix bdev_disk_changed for non-partitioned devices
>> a1548b6 block: move rescan_partitions to fs/block_dev.c
> Just to make sure we are on the same page:  if you revert all four it
> works, if you rever all but
>
> a1548b6 block: move rescan_partitions to fs/block_dev.c
>
> it doesn't?

After reverting 142fe8f, rescan_partitions would be called in block/ioctl.c
and cause a build failure. So I need to also revert a1548b6 to provide
rescan_partitions.

OR if I manually add the following diff instead of reverting a1548b6, then yes,
it works too.

diff --git a/block/ioctl.c b/block/ioctl.c
index 8d724d11c8f5..bac562604cd0 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -192,6 +192,7 @@ static int compat_blkpg_ioctl(struct block_device *bdev,
  * acquire bd_mutex. This API should be used in case that
  * caller has held bd_mutex already.
  */
+extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev, bool invalidate);
 int __blkdev_reread_part(struct block_device *bdev)
 {
        struct gendisk *disk = bdev->bd_disk;
diff --git a/fs/block_dev.c b/fs/block_dev.c
index ec10dacd18d0..30da0bc85c31 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1508,7 +1508,7 @@ EXPORT_SYMBOL(bd_set_size);

 static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);

-static int rescan_partitions(struct gendisk *disk, struct block_device *bdev,
+int rescan_partitions(struct gendisk *disk, struct block_device *bdev,
                bool invalidate)
 {
        int ret;


Zhe


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ