linux-kernel - [PATCH 1/3] block: blk_interposer

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: <1607518911-30692-2-git-send-email-sergei.shtepa@veeam.com>
Date:   Wed, 9 Dec 2020 16:01:49 +0300
From:   Sergei Shtepa <sergei.shtepa@...am.com>
To:     <axboe@...nel.dk>, <johannes.thumshirn@....com>,
        <koct9i@...il.com>, <ming.lei@...hat.com>, <snitzer@...hat.com>,
        <hare@...e.de>, <josef@...icpanda.com>, <steve@....org>,
        <linux-block@...r.kernel.org>, <linux-kernel@...r.kernel.org>
CC:     <pavel.tide@...am.com>, <sergei.shtepa@...am.com>
Subject: [PATCH 1/3] block: blk_interposer - Block Layer Interposer

The Block Layer Interposer (blk_interposer) allows to intercept bio.
This allows to connect device mapper and other kernel modules to
the block device stack on the fly.

This simple hook allows to work modules that previously relied on the
interception of the make_request_fn() function from the
request_queue structure. This interception capability was lost in
the v5.8.

Changes:
  * A new __submit_bio_interposed() function call is added
    to the submit_bio_noacct(). The call is made only if the
    interposer was connected.
  * A new bio flag BIO_INTERPOSED can indicate that bio has
    already been interposed or cannot be interposed.
  * The bi_flags member of the BIO structure is expanded
    to a 32-bit variable to accommodate the new flag.
  * New structure blk_interposer contains the interposer
    callback function.
Signed-off-by: Sergei Shtepa <sergei.shtepa@...am.com>
---
 block/blk-core.c          | 32 ++++++++++++++++++++++++++++++++
 include/linux/blk_types.h |  6 ++++--
 include/linux/genhd.h     | 12 +++++++++++-
 3 files changed, 47 insertions(+), 3 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index 2db8bda43b6e..682b9b5a93ff 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1030,6 +1030,35 @@ static blk_qc_t __submit_bio_noacct_mq(struct bio *bio)
 	return ret;
 }
 
+static blk_qc_t __submit_bio_interposed(struct bio *bio)
+{
+	struct bio_list bio_list[2] = { };
+	blk_qc_t ret = BLK_QC_T_NONE;
+
+	current->bio_list = bio_list;
+	if (likely(bio_queue_enter(bio) == 0)) {
+		struct gendisk *disk = bio->bi_disk;
+
+		bio_set_flag(bio, BIO_INTERPOSED);
+		if (likely(blk_has_interposer(disk))) {
+			struct blk_interposer *ip = disk->interposer;
+
+			ip->ip_submit_bio(ip, bio);
+		} else
+			/* interposer was removed */
+			bio_list_add(&current->bio_list[0], bio);
+
+		blk_queue_exit(disk->queue);
+	}
+	current->bio_list = NULL;
+
+	/* Resubmit the remaining bios */
+	while ((bio = bio_list_pop(&bio_list[0])))
+		ret = submit_bio_noacct(bio);
+
+	return ret;
+}
+
 /**
  * submit_bio_noacct - re-submit a bio to the block device layer for I/O
  * @bio:  The bio describing the location in memory and on the device.
@@ -1055,6 +1084,9 @@ blk_qc_t submit_bio_noacct(struct bio *bio)
 		return BLK_QC_T_NONE;
 	}
 
+	if (blk_has_interposer(bio->bi_disk) &&
+	    !bio_flagged(bio, BIO_INTERPOSED))
+		return __submit_bio_interposed(bio);
 	if (!bio->bi_disk->fops->submit_bio)
 		return __submit_bio_noacct_mq(bio);
 	return __submit_bio_noacct(bio);
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index d9b69bbde5cc..996b803e5aa1 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -207,7 +207,7 @@ struct bio {
 						 * top bits REQ_OP. Use
 						 * accessors.
 						 */
-	unsigned short		bi_flags;	/* status, etc and bvec pool number */
+	unsigned int		bi_flags;	/* status, etc and bvec pool number */
 	unsigned short		bi_ioprio;
 	unsigned short		bi_write_hint;
 	blk_status_t		bi_status;
@@ -284,6 +284,8 @@ enum {
 				 * of this bio. */
 	BIO_CGROUP_ACCT,	/* has been accounted to a cgroup */
 	BIO_TRACKED,		/* set if bio goes through the rq_qos path */
+	BIO_INTERPOSED,		/* bio has been interposed and can be moved to
+				 * a different disk */
 	BIO_FLAG_LAST
 };
 
@@ -302,7 +304,7 @@ enum {
  * freed.
  */
 #define BVEC_POOL_BITS		(3)
-#define BVEC_POOL_OFFSET	(16 - BVEC_POOL_BITS)
+#define BVEC_POOL_OFFSET	(32 - BVEC_POOL_BITS)
 #define BVEC_POOL_IDX(bio)	((bio)->bi_flags >> BVEC_POOL_OFFSET)
 #if (1<< BVEC_POOL_BITS) < (BVEC_POOL_NR+1)
 # error "BVEC_POOL_BITS is too small"
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 03da3f603d30..ea5fc57c3451 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -4,7 +4,7 @@
 
 /*
  * 	genhd.h Copyright (C) 1992 Drew Eckhardt
- *	Generic hard disk header file by  
+ *	Generic hard disk header file by
  * 		Drew Eckhardt
  *
  *		<drew@...orado.edu>
@@ -164,6 +164,13 @@ struct blk_integrity {
 	unsigned char				tag_size;
 };
 
+struct blk_interposer;
+typedef void (*ip_submit_bio_t) (struct blk_interposer *ip, struct bio *bio);
+
+struct blk_interposer {
+	ip_submit_bio_t ip_submit_bio;
+};
+
 struct gendisk {
 	/* major, first_minor and minors are input parameters only,
 	 * don't use directly.  Use disk_devt() and disk_max_parts().
@@ -188,6 +195,7 @@ struct gendisk {
 
 	const struct block_device_operations *fops;
 	struct request_queue *queue;
+	struct blk_interposer *interposer;
 	void *private_data;
 
 	int flags;
@@ -409,4 +417,6 @@ static inline dev_t blk_lookup_devt(const char *name, int partno)
 }
 #endif /* CONFIG_BLOCK */
 
+#define blk_has_interposer(d) ((d)->interposer != NULL)
+
 #endif /* _LINUX_GENHD_H */
-- 
2.20.1