[PATCH 00/56] Update Barebox UBI

[PATCH 00/56] Update Barebox UBI

[Teresa Remmet]

Hello, 

A new update of the Barebox UBI stack to the Kernel v4.19-rc1.
All patches are cherry-picks from the kernel tree.

Stumbled over some conflicts mostly due to non exsisting threads in barebox.
Run some tests and did not face any regressions so far.

Regards,
Teresa

Andrew F. Davis (1):
  UBI: Fix typos

Ben Dooks (1):
  ubi: pr_err() strings should end with newlines

Boris Brezillon (18):
  UBI: fastmap: use ubi_find_volume() instead of open coding it
  UBI: fix add_fastmap() to use the vid_hdr passed in argument
  UBI: fastmap: avoid multiple be32_to_cpu() when unneccesary
  UBI: fastmap: scrub PEB when bitflips are detected in a free PEB EC
    header
  UBI: factorize code used to manipulate volumes at attach time
  UBI: factorize destroy_av() and ubi_remove_av() code
  UBI: fastmap: use ubi_rb_for_each_entry() in unmap_peb()
  UBI: fastmap: use ubi_io_{read, write}_data() instead of ubi_io_{read,
    write}()
  UBI: provide helpers to allocate and free aeb elements
  UBI: move the global ech and vidh variables into struct
    ubi_attach_info
  UBI: simplify recover_peb() code
  UBI: simplify LEB write and atomic LEB change code
  UBI: add an helper to check lnum validity
  UBI: provide an helper to check whether a LEB is mapped or not
  UBI: provide an helper to query LEB information
  UBI: hide EBA internals
  UBI: introduce the VID buffer concept
  ubi: fastmap: Fix add_vol() return value test in ubi_attach_fastmap()

Colin Ian King (3):
  ubi: fix swapped arguments to call to ubi_alloc_aeb
  ubi: fastmap: fix spelling mistake: "invalidiate" -> "invalidate"
  ubi: fastmap: Clean up the initialization of pointer p

Dan Carpenter (1):
  UBI: Clean up return in ubi_remove_volume()

Geert Uytterhoeven (1):
  UBI: Fix crash in try_recover_peb()

Heiko Schocher (1):
  UBI: Set free_count to zero before walking through erase list

Iosif Harutyunov (1):
  ubi: Fix race condition between ubi device creation and udev

Lionel Debieve (1):
  mtd: ubi: Update ubi-media.h to dual license

Rabin Vincent (1):
  ubi: fastmap: Fix slab corruption

Richard Weinberger (20):
  UBI: Fastmap: Fix memory leak while attaching
  UBI: Remove alloc_ai() slab name from parameter list
  UBI: Don't read back all data in ubi_eba_copy_leb()
  UBI: Fix static volume checks when Fastmap is used
  ubi: Make recover_peb power cut aware
  ubi: Fix early logging
  ubi: Make volume resize power cut aware
  ubi: Fix scan_fast() comment
  ubi: Introduce vol_ignored()
  ubi: Fix whitespace issue in count_fastmap_pebs()
  ubi: Rework Fastmap attach base code
  ubi: Check whether the Fastmap anchor matches the super block
  ubi: Be more paranoid while seaching for the most recent Fastmap
  ubi: Deal with interrupted erasures in WL
  ubi: Fix races around ubi_refill_pools()
  ubi: Fix Fastmap's update_vol()
  ubi: fastmap: Correctly handle interrupted erasures in EBA
  ubi: fastmap: Check each mapping only once
  ubi: fastmap: Detect EBA mismatches on-the-fly
  ubi: Initialize Fastmap checkmapping correctly

Sascha Hauer (3):
  ubi: fastmap: Erase outdated anchor PEBs during attach
  ubi: Fastmap: Fix typo
  ubi: Fix copy/paste error in function documentation

Sebastian Siewior (1):
  ubi/upd: Always flush after prepared for an update

Uwe Kleine-König (1):
  UBI: Fix two typos in comments

Wei Yongjun (1):
  mtd: ubi: wl: Fix error return code in ubi_wl_init()

z00189512 (1):
  UBI: Modify wrong comment in ubi_leb_map function.

 drivers/mtd/ubi/Kconfig      |   2 +-
 drivers/mtd/ubi/attach.c     | 473 +++++++++++++++++---------
 drivers/mtd/ubi/build.c      |  17 +-
 drivers/mtd/ubi/eba.c        | 785 ++++++++++++++++++++++++++++---------------
 drivers/mtd/ubi/fastmap-wl.c |   6 +-
 drivers/mtd/ubi/fastmap.c    | 300 ++++++++++-------
 drivers/mtd/ubi/io.c         |  41 ++-
 drivers/mtd/ubi/kapi.c       |  18 +-
 drivers/mtd/ubi/ubi-media.h  |  22 +-
 drivers/mtd/ubi/ubi.h        | 194 +++++++++--
 drivers/mtd/ubi/upd.c        |   8 +-
 drivers/mtd/ubi/vmt.c        |  65 ++--
 drivers/mtd/ubi/vtbl.c       |  32 +-
 drivers/mtd/ubi/wl.c         | 161 +++++++--
 drivers/mtd/ubi/wl.h         |   2 +-
 15 files changed, 1419 insertions(+), 707 deletions(-)

-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 01/56] UBI: Fastmap: Fix memory leak while attaching

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Currently we leak a few ubi_ainf_pebs while attaching.

Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c  | 47 ++++++++++++++++++++++++++---------------------
 drivers/mtd/ubi/fastmap.c |  5 -----
 2 files changed, 26 insertions(+), 26 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index ff4b4e7acb76..daf0010d06da 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -1285,6 +1285,23 @@ out_ech:
 	return err;
 }
 
+static struct ubi_attach_info *alloc_ai(const char *slab_name)
+{
+	struct ubi_attach_info *ai;
+
+	ai = kzalloc(sizeof(struct ubi_attach_info), GFP_KERNEL);
+	if (!ai)
+		return ai;
+
+	INIT_LIST_HEAD(&ai->corr);
+	INIT_LIST_HEAD(&ai->free);
+	INIT_LIST_HEAD(&ai->erase);
+	INIT_LIST_HEAD(&ai->alien);
+	ai->volumes = RB_ROOT;
+
+	return ai;
+}
+
 #ifdef CONFIG_MTD_UBI_FASTMAP
 
 /**
@@ -1297,7 +1314,7 @@ out_ech:
  * UBI_NO_FASTMAP denotes that no fastmap was found.
  * UBI_BAD_FASTMAP denotes that the found fastmap was invalid.
  */
-static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info *ai)
+static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 {
 	int err, pnum, fm_anchor = -1;
 	unsigned long long max_sqnum = 0;
@@ -1317,7 +1334,7 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info *ai)
 		unsigned long long sqnum = -1;
 
 		dbg_gen("process PEB %d", pnum);
-		err = scan_peb(ubi, ai, pnum, &vol_id, &sqnum);
+		err = scan_peb(ubi, *ai, pnum, &vol_id, &sqnum);
 		if (err < 0)
 			goto out_vidh;
 
@@ -1333,7 +1350,12 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info *ai)
 	if (fm_anchor < 0)
 		return UBI_NO_FASTMAP;
 
-	return ubi_scan_fastmap(ubi, ai, fm_anchor);
+	destroy_ai(*ai);
+	*ai = alloc_ai("ubi_aeb_slab_cache");
+	if (!*ai)
+		return -ENOMEM;
+
+	return ubi_scan_fastmap(ubi, *ai, fm_anchor);
 
 out_vidh:
 	ubi_free_vid_hdr(ubi, vidh);
@@ -1345,23 +1367,6 @@ out:
 
 #endif
 
-static struct ubi_attach_info *alloc_ai(const char *slab_name)
-{
-	struct ubi_attach_info *ai;
-
-	ai = kzalloc(sizeof(struct ubi_attach_info), GFP_KERNEL);
-	if (!ai)
-		return ai;
-
-	INIT_LIST_HEAD(&ai->corr);
-	INIT_LIST_HEAD(&ai->free);
-	INIT_LIST_HEAD(&ai->erase);
-	INIT_LIST_HEAD(&ai->alien);
-	ai->volumes = RB_ROOT;
-
-	return ai;
-}
-
 /**
  * ubi_attach - attach an MTD device.
  * @ubi: UBI device descriptor
@@ -1389,7 +1394,7 @@ int ubi_attach(struct ubi_device *ubi, int force_scan)
 	if (force_scan)
 		err = scan_all(ubi, ai, 0);
 	else {
-		err = scan_fast(ubi, ai);
+		err = scan_fast(ubi, &ai);
 		if (err > 0 || mtd_is_eccerr(err)) {
 			if (err != UBI_NO_FASTMAP) {
 				destroy_ai(ai);
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 9975722698a0..2e0fdb7583e4 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -548,11 +548,6 @@ static int ubi_attach_fastmap(struct ubi_device *ubi,
 
 	INIT_LIST_HEAD(&used);
 	INIT_LIST_HEAD(&lfree);
-	INIT_LIST_HEAD(&ai->corr);
-	INIT_LIST_HEAD(&ai->free);
-	INIT_LIST_HEAD(&ai->erase);
-	INIT_LIST_HEAD(&ai->alien);
-	ai->volumes = RB_ROOT;
 	ai->min_ec = UBI_MAX_ERASECOUNTER;
 
 	fmsb = (struct ubi_fm_sb *)(fm_raw);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 02/56] UBI: Remove alloc_ai() slab name from parameter list

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

There is always exactly one ubi_attach_info object allocated,
therefore we don't have to care about the name.

Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index daf0010d06da..e9c34c0cc81e 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -1285,7 +1285,7 @@ out_ech:
 	return err;
 }
 
-static struct ubi_attach_info *alloc_ai(const char *slab_name)
+static struct ubi_attach_info *alloc_ai(void)
 {
 	struct ubi_attach_info *ai;
 
@@ -1351,7 +1351,7 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 		return UBI_NO_FASTMAP;
 
 	destroy_ai(*ai);
-	*ai = alloc_ai("ubi_aeb_slab_cache");
+	*ai = alloc_ai();
 	if (!*ai)
 		return -ENOMEM;
 
@@ -1380,7 +1380,7 @@ int ubi_attach(struct ubi_device *ubi, int force_scan)
 	int err;
 	struct ubi_attach_info *ai;
 
-	ai = alloc_ai("ubi_aeb_slab_cache");
+	ai = alloc_ai();
 	if (!ai)
 		return -ENOMEM;
 
@@ -1398,7 +1398,7 @@ int ubi_attach(struct ubi_device *ubi, int force_scan)
 		if (err > 0 || mtd_is_eccerr(err)) {
 			if (err != UBI_NO_FASTMAP) {
 				destroy_ai(ai);
-				ai = alloc_ai("ubi_aeb_slab_cache2");
+				ai = alloc_ai();
 				if (!ai)
 					return -ENOMEM;
 
@@ -1437,7 +1437,7 @@ int ubi_attach(struct ubi_device *ubi, int force_scan)
 	if (ubi->fm && ubi_dbg_chk_gen(ubi)) {
 		struct ubi_attach_info *scan_ai;
 
-		scan_ai = alloc_ai("ubi_ckh_aeb_slab_cache");
+		scan_ai = alloc_ai();
 		if (!scan_ai) {
 			err = -ENOMEM;
 			goto out_wl;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 03/56] UBI: Don't read back all data in ubi_eba_copy_leb()

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Drop this paranoia check from the old days.
If our MTD driver or the flash is so bad that we even cannot
trust it to write data we have bigger problems.

If one really does not trust the flash and wants write-verify
she can enable UBI io checks using debugfs.

Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed up conflicts due to non exsisting logging in barebox]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 31dbcd29c374..c242c460e849 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -1077,30 +1077,6 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 				err = MOVE_TARGET_WR_ERR;
 			goto out_unlock_buf;
 		}
-
-		/*
-		 * We've written the data and are going to read it back to make
-		 * sure it was written correctly.
-		 */
-		memset(ubi->peb_buf, 0xFF, aldata_size);
-		err = ubi_io_read_data(ubi, ubi->peb_buf, to, 0, aldata_size);
-		if (err) {
-			if (err != UBI_IO_BITFLIPS) {
-				ubi_warn(ubi, "error %d while reading data back from PEB %d",
-					 err, to);
-				if (is_error_sane(err))
-					err = MOVE_TARGET_RD_ERR;
-			} else
-				err = MOVE_TARGET_BITFLIPS;
-			goto out_unlock_buf;
-		}
-
-		if (crc != crc32(UBI_CRC32_INIT, ubi->peb_buf, data_size)) {
-			ubi_warn(ubi, "read data back from PEB %d and it is different",
-				 to);
-			err = -EINVAL;
-			goto out_unlock_buf;
-		}
 	}
 
 	ubi_assert(vol->eba_tbl[lnum] == from);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 04/56] UBI: Modify wrong comment in ubi_leb_map function.

[Teresa Remmet]

From: z00189512 <abc.zhangliang@huawei.com>

Signed-off-by: z00189512 <abc.zhangliang@huawei.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/kapi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index 60128a5c26fe..9ec9962a5966 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -586,7 +586,7 @@ int ubi_leb_map(struct ubi_volume_desc *desc, int lnum)
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
 
-	dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum);
+	dbg_gen("map LEB %d:%d", vol->vol_id, lnum);
 
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 05/56] UBI: Clean up return in ubi_remove_volume()

[Teresa Remmet]

From: Dan Carpenter <dan.carpenter@oracle.com>

My static checker says that "err" can be uninitialized if
"vol->reserved_pebs" is <= 0.  I don't think that can happen but
returning a literal is cleaner anyway.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/vmt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index ed043646b8d8..aed51c2fe3b2 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -264,7 +264,7 @@ int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl)
 	if (!no_vtbl)
 		self_check_volumes(ubi);
 
-	return err;
+	return 0;
 
 out_err:
 	ubi_err(ubi, "cannot remove volume %d, error %d", vol_id, err);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 06/56] UBI: Set free_count to zero before walking through erase list

[Teresa Remmet]

From: Heiko Schocher <hs@denx.de>

Set free_count to zero before walking through ai->erase list
in wl_init().

Found in U-Boot as U-Boot has no workqueue/threads, it immediately
calls erase_worker(), which increase for each erased block
free_count. Without this patch, free_count gets after
this initialized to zero in wl_init(), so the free_count
variable always has the maybe wrong value 0 in U-Boot.

Signed-off-by: Heiko Schocher <hs@denx.de>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/wl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index a368bf910806..eb1f24c74807 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -1399,6 +1399,7 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 		INIT_LIST_HEAD(&ubi->pq[i]);
 	ubi->pq_head = 0;
 
+	ubi->free_count = 0;
 	list_for_each_entry_safe(aeb, tmp, &ai->erase, u.list) {
 		e = kmalloc(sizeof(*e), GFP_KERNEL);
 		if (!e)
@@ -1415,7 +1416,6 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 		found_pebs++;
 	}
 
-	ubi->free_count = 0;
 	list_for_each_entry(aeb, &ai->free, u.list) {
 		e = kmalloc(sizeof(*e), GFP_KERNEL);
 		if (!e)
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 07/56] UBI: Fix static volume checks when Fastmap is used

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Ezequiel reported that he's facing UBI going into read-only
mode after power cut. It turned out that this behavior happens
only when updating a static volume is interrupted and Fastmap is
used.

A possible trace can look like:
ubi0 warning: ubi_io_read_vid_hdr [ubi]: no VID header found at PEB 2323, only 0xFF bytes
ubi0 warning: ubi_eba_read_leb [ubi]: switch to read-only mode
CPU: 0 PID: 833 Comm: ubiupdatevol Not tainted 4.6.0-rc2-ARCH #4
Hardware name: SAMSUNG ELECTRONICS CO., LTD. 300E4C/300E5C/300E7C/NP300E5C-AD8AR, BIOS P04RAP 10/15/2012
0000000000000286 00000000eba949bd ffff8800c45a7b38 ffffffff8140d841
ffff8801964be000 ffff88018eaa4800 ffff8800c45a7bb8 ffffffffa003abf6
ffffffff850e2ac0 8000000000000163 ffff8801850e2ac0 ffff8801850e2ac0
Call Trace:
[<ffffffff8140d841>] dump_stack+0x63/0x82
[<ffffffffa003abf6>] ubi_eba_read_leb+0x486/0x4a0 [ubi]
[<ffffffffa00453b3>] ubi_check_volume+0x83/0xf0 [ubi]
[<ffffffffa0039d97>] ubi_open_volume+0x177/0x350 [ubi]
[<ffffffffa00375d8>] vol_cdev_open+0x58/0xb0 [ubi]
[<ffffffff8124b08e>] chrdev_open+0xae/0x1d0
[<ffffffff81243bcf>] do_dentry_open+0x1ff/0x300
[<ffffffff8124afe0>] ? cdev_put+0x30/0x30
[<ffffffff81244d36>] vfs_open+0x56/0x60
[<ffffffff812545f4>] path_openat+0x4f4/0x1190
[<ffffffff81256621>] do_filp_open+0x91/0x100
[<ffffffff81263547>] ? __alloc_fd+0xc7/0x190
[<ffffffff812450df>] do_sys_open+0x13f/0x210
[<ffffffff812451ce>] SyS_open+0x1e/0x20
[<ffffffff81a99e32>] entry_SYSCALL_64_fastpath+0x1a/0xa4

UBI checks static volumes for data consistency and reads the
whole volume upon first open. If the volume is found erroneous
users of UBI cannot read from it, but another volume update is
possible to fix it. The check is performed by running
ubi_eba_read_leb() on every allocated LEB of the volume.
For static volumes ubi_eba_read_leb() computes the checksum of all
data stored in a LEB. To verify the computed checksum it has to read
the LEB's volume header which stores the original checksum.
If the volume header is not found UBI treats this as fatal internal
error and switches to RO mode. If the UBI device was attached via a
full scan the assumption is correct, the volume header has to be
present as it had to be there while scanning to get known as mapped.
If the attach operation happened via Fastmap the assumption is no
longer correct. When attaching via Fastmap UBI learns the mapping
table from Fastmap's snapshot of the system state and not via a full
scan. It can happen that a LEB got unmapped after a Fastmap was
written to the flash. Then UBI can learn the LEB still as mapped and
accessing it returns only 0xFF bytes. As UBI is not a FTL it is
allowed to have mappings to empty PEBs, it assumes that the layer
above takes care of LEB accounting and referencing.
UBIFS does so using the LEB property tree (LPT).
For static volumes UBI blindly assumes that all LEBs are present and
therefore special actions have to be taken.

The described situation can happen when updating a static volume is
interrupted, either by a user or a power cut.
The volume update code first unmaps all LEBs of a volume and then
writes LEB by LEB. If the sequence of operations is interrupted UBI
detects this either by the absence of LEBs, no volume header present
at scan time, or corrupted payload, detected via checksum.
In the Fastmap case the former method won't trigger as no scan
happened and UBI automatically thinks all LEBs are present.
Only by reading data from a LEB it detects that the volume header is
missing and incorrectly treats this as fatal error.
To deal with the situation ubi_eba_read_leb() from now on checks
whether we attached via Fastmap and handles the absence of a
volume header like a data corruption error.
This way interrupted static volume updates will correctly get detected
also when Fastmap is used.

Cc: <stable@vger.kernel.org>
Reported-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Tested-by: Ezequiel Garcia <ezequiel@vanguardiasur.com.ar>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c     | 21 +++++++++++++++++++--
 drivers/mtd/ubi/fastmap.c |  1 +
 drivers/mtd/ubi/ubi.h     |  2 ++
 3 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index c242c460e849..6e09cfb0a119 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -392,8 +392,25 @@ retry:
 						 pnum, vol_id, lnum);
 					err = -EBADMSG;
 				} else {
-					err = -EINVAL;
-					ubi_ro_mode(ubi);
+					/*
+					 * Ending up here in the non-Fastmap case
+					 * is a clear bug as the VID header had to
+					 * be present at scan time to have it referenced.
+					 * With fastmap the story is more complicated.
+					 * Fastmap has the mapping info without the need
+					 * of a full scan. So the LEB could have been
+					 * unmapped, Fastmap cannot know this and keeps
+					 * the LEB referenced.
+					 * This is valid and works as the layer above UBI
+					 * has to do bookkeeping about used/referenced
+					 * LEBs in any case.
+					 */
+					if (ubi->fast_attach) {
+						err = -EBADMSG;
+					} else {
+						err = -EINVAL;
+						ubi_ro_mode(ubi);
+					}
 				}
 			}
 			goto out_free;
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 2e0fdb7583e4..5bbbd7c6ae2a 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -991,6 +991,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	ubi_msg(ubi, "fastmap WL pool size: %d",
 		ubi->fm_wl_pool.max_size);
 	ubi->fm_disabled = 0;
+	ubi->fast_attach = 1;
 
 	ubi_free_vid_hdr(ubi, vh);
 	kfree(ech);
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 1f0ad386b45e..368b829fa17f 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -416,6 +416,7 @@ struct ubi_debug_info {
  * @fm_size: fastmap size in bytes
  * @fm_sem: allows ubi_update_fastmap() to block EBA table changes
  * @fm_work: fastmap work queue
+ * @fast_attach: non-zero if UBI was attached by fastmap
  *
  * @used: RB-tree of used physical eraseblocks
  * @erroneous: RB-tree of erroneous used physical eraseblocks
@@ -515,6 +516,7 @@ struct ubi_device {
 	struct ubi_fm_pool fm_wl_pool;
 	void *fm_buf;
 	size_t fm_size;
+	int fast_attach;
 
 	/* Wear-leveling sub-system's stuff */
 	struct rb_root used;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 08/56] ubi: Make recover_peb power cut aware

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

recover_peb() was never power cut aware,
if a power cut happened right after writing the VID header
upon next attach UBI would blindly use the new partial written
PEB and all data from the old PEB is lost.

In order to make recover_peb() power cut aware, write the new
VID with a proper crc and copy_flag set such that the UBI attach
process will detect whether the new PEB is completely written
or not.
We cannot directly use ubi_eba_atomic_leb_change() since we'd
have to unlock the LEB which is facing a write error.

Cc: stable@vger.kernel.org
Reported-by: Jörg Pfähler <pfaehler@isse.de>
Reviewed-by: Jörg Pfähler <pfaehler@isse.de>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed minor conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 6e09cfb0a119..9decc0e1ecce 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -486,6 +486,7 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
 	int err, idx = vol_id2idx(ubi, vol_id), new_pnum, data_size, tries = 0;
 	struct ubi_volume *vol = ubi->volumes[idx];
 	struct ubi_vid_hdr *vid_hdr;
+	uint32_t crc;
 
 	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
 	if (!vid_hdr)
@@ -508,12 +509,8 @@ retry:
 		goto out_put;
 	}
 
-	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
-	err = ubi_io_write_vid_hdr(ubi, new_pnum, vid_hdr);
-	if (err)
-		goto write_error;
+	ubi_assert(vid_hdr->vol_type == UBI_VID_DYNAMIC);
 
-	data_size = offset + len;
 	memset(ubi->peb_buf + offset, 0xFF, len);
 
 	/* Read everything before the area where the write failure happened */
@@ -525,6 +522,16 @@ retry:
 
 	memcpy(ubi->peb_buf + offset, buf, len);
 
+	data_size = offset + len;
+	crc = crc32(UBI_CRC32_INIT, ubi->peb_buf, data_size);
+	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
+	vid_hdr->copy_flag = 1;
+	vid_hdr->data_size = cpu_to_be32(data_size);
+	vid_hdr->data_crc = cpu_to_be32(crc);
+	err = ubi_io_write_vid_hdr(ubi, new_pnum, vid_hdr);
+	if (err)
+		goto write_error;
+
 	err = ubi_io_write_data(ubi, ubi->peb_buf, new_pnum, 0, data_size);
 	if (err)
 		goto write_error;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 09/56] ubi: Fix race condition between ubi device creation and udev

[Teresa Remmet]

From: Iosif Harutyunov <iharutyunov@SonicWALL.com>

Install the UBI device object before we arm sysfs.
Otherwise udev tries to read sysfs attributes before UBI is ready and
udev rules will not match.

Cc: <stable@vger.kernel.org>
Signed-off-by: Iosif Harutyunov <iharutyunov@sonicwall.com>
[rw: massaged commit message]
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed minor conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/build.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 40fa890c9e59..d5c3719482cf 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -633,6 +633,9 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 			goto out_detach;
 	}
 
+	/* Make device "available" before it becomes accessible via sysfs */
+	ubi_devices[ubi_num] = ubi;
+
 	err = uif_init(ubi, &ref);
 	if (err)
 		goto out_detach;
@@ -668,11 +671,10 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 	dev_add_param_uint32_ro(&ubi->dev, "available_pebs", &ubi->avail_pebs, "%u");
 	dev_add_param_uint32_ro(&ubi->dev, "reserved_pebs", &ubi->rsvd_pebs, "%u");
 
-	ubi_devices[ubi_num] = ubi;
-
 	return ubi_num;
 
 out_detach:
+	ubi_devices[ubi_num] = NULL;
 	ubi_wl_close(ubi);
 	ubi_free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 10/56] ubi: Fix early logging

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

We cannot use ubi_* logging functions before the UBI
object is initialized.

Cc: <stable@vger.kernel.org>
Fixes: 3260870331 ("UBI: Extend UBI layer debug/messaging capabilities")
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed minor conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/build.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index d5c3719482cf..ae6181e9d162 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -522,7 +522,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 	for (i = 0; i < UBI_MAX_DEVICES; i++) {
 		ubi = ubi_devices[i];
 		if (ubi && mtd == ubi->mtd) {
-			ubi_debug("mtd%d is already attached to ubi%d",
+			pr_err("ubi: mtd%d is already attached to ubi%d",
 				mtd->index, i);
 			return -EEXIST;
 		}
@@ -537,7 +537,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 	 * no sense to attach emulated MTD devices, so we prohibit this.
 	 */
 	if (mtd->type == MTD_UBIVOLUME) {
-		ubi_err(ubi, "refuse attaching mtd%d - it is already emulated on top of UBI",
+		pr_err("ubi: refuse attaching mtd%d - it is already emulated on top of UBI",
 			mtd->index);
 		return -EINVAL;
 	}
@@ -548,7 +548,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 			if (!ubi_devices[ubi_num])
 				break;
 		if (ubi_num == UBI_MAX_DEVICES) {
-			ubi_err(ubi, "only %d UBI devices may be created",
+			pr_err("ubi: only %d UBI devices may be created",
 				UBI_MAX_DEVICES);
 			return -ENFILE;
 		}
@@ -558,7 +558,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 
 		/* Make sure ubi_num is not busy */
 		if (ubi_devices[ubi_num]) {
-			ubi_err(ubi, "already exists");
+			pr_err("ubi: ubi%i already exists", ubi_num);
 			return -EEXIST;
 		}
 	}
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 11/56] ubi: Make volume resize power cut aware

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

When the volume resize operation shrinks a volume,
LEBs will be unmapped. Since unmapping will not erase these
LEBs immediately we have to wait for that operation to finish.
Otherwise in case of a power cut right after writing the new
volume table the UBI attach process can find more LEBs than the
volume table knows. This will render the UBI image unattachable.

Fix this issue by waiting for erase to complete and write the new
volume table afterward.

Cc: <stable@vger.kernel.org>
Reported-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/vmt.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index aed51c2fe3b2..05a61fccefc6 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -339,13 +339,6 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 		vol->eba_tbl = new_mapping;
 	}
 
-	/* Change volume table record */
-	vtbl_rec = ubi->vtbl[vol_id];
-	vtbl_rec.reserved_pebs = cpu_to_be32(reserved_pebs);
-	err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec);
-	if (err)
-		goto out_acc;
-
 	if (pebs < 0) {
 		for (i = 0; i < -pebs; i++) {
 			err = ubi_eba_unmap_leb(ubi, vol, reserved_pebs + i);
@@ -361,6 +354,24 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 		vol->eba_tbl = new_mapping;
 	}
 
+	/*
+	 * When we shrink a volume we have to flush all pending (erase) work.
+	 * Otherwise it can happen that upon next attach UBI finds a LEB with
+	 * lnum > highest_lnum and refuses to attach.
+	 */
+	if (pebs < 0) {
+		err = ubi_wl_flush(ubi, vol_id, UBI_ALL);
+		if (err)
+			goto out_acc;
+	}
+
+	/* Change volume table record */
+	vtbl_rec = ubi->vtbl[vol_id];
+	vtbl_rec.reserved_pebs = cpu_to_be32(reserved_pebs);
+	err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec);
+	if (err)
+		goto out_acc;
+
 	vol->reserved_pebs = reserved_pebs;
 	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
 		vol->used_ebs = reserved_pebs;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 12/56] ubi: Fix scan_fast() comment

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index e9c34c0cc81e..b6f03ed59f26 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -1305,7 +1305,7 @@ static struct ubi_attach_info *alloc_ai(void)
 #ifdef CONFIG_MTD_UBI_FASTMAP
 
 /**
- * scan_fastmap - try to find a fastmap and attach from it.
+ * scan_fast - try to find a fastmap and attach from it.
  * @ubi: UBI device description object
  * @ai: attach info object
  *
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 13/56] ubi: Introduce vol_ignored()

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

This makes the logic more easy to follow.

Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 24 ++++++++++++++++++------
 drivers/mtd/ubi/ubi.h    | 15 +++++++++++++++
 2 files changed, 33 insertions(+), 6 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index b6f03ed59f26..9cf9b2ae98e3 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -792,6 +792,20 @@ out_unlock:
 	return err;
 }
 
+static bool vol_ignored(int vol_id)
+{
+	switch (vol_id) {
+		case UBI_LAYOUT_VOLUME_ID:
+		return true;
+	}
+
+#ifdef CONFIG_MTD_UBI_FASTMAP
+	return ubi_is_fm_vol(vol_id);
+#else
+	return false;
+#endif
+}
+
 /**
  * scan_peb - scan and process UBI headers of a PEB.
  * @ubi: UBI device description object
@@ -984,17 +998,15 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		*vid = vol_id;
 	if (sqnum)
 		*sqnum = be64_to_cpu(vidh->sqnum);
-	if (vol_id > UBI_MAX_VOLUMES && vol_id != UBI_LAYOUT_VOLUME_ID) {
+	if (vol_id > UBI_MAX_VOLUMES && !vol_ignored(vol_id)) {
 		int lnum = be32_to_cpu(vidh->lnum);
 
 		/* Unsupported internal volume */
 		switch (vidh->compat) {
 		case UBI_COMPAT_DELETE:
-			if (vol_id != UBI_FM_SB_VOLUME_ID
-			    && vol_id != UBI_FM_DATA_VOLUME_ID) {
-				ubi_msg(ubi, "\"delete\" compatible internal volume %d:%d found, will remove it",
-					vol_id, lnum);
-			}
+			ubi_msg(ubi, "\"delete\" compatible internal volume %d:%d found, will remove it",
+				vol_id, lnum);
+
 			err = add_to_list(ai, pnum, vol_id, lnum,
 					  ec, 1, &ai->erase);
 			if (err)
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 368b829fa17f..5899ad7c92b8 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -1018,4 +1018,19 @@ static inline int idx2vol_id(const struct ubi_device *ubi, int idx)
 		return idx;
 }
 
+/**
+ * ubi_is_fm_vol - check whether a volume ID is a Fastmap volume.
+ * @vol_id: volume ID
+ */
+static inline bool ubi_is_fm_vol(int vol_id)
+{
+	switch (vol_id) {
+		case UBI_FM_SB_VOLUME_ID:
+		case UBI_FM_DATA_VOLUME_ID:
+		return true;
+	}
+
+	return false;
+}
+
 #endif /* !__UBI_UBI_H__ */
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 14/56] ubi: Fix whitespace issue in count_fastmap_pebs()

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 5bbbd7c6ae2a..205174495b0d 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -512,7 +512,7 @@ static int count_fastmap_pebs(struct ubi_attach_info *ai)
 	list_for_each_entry(aeb, &ai->free, u.list)
 		n++;
 
-	 ubi_rb_for_each_entry(rb1, av, &ai->volumes, rb)
+	ubi_rb_for_each_entry(rb1, av, &ai->volumes, rb)
 		ubi_rb_for_each_entry(rb2, aeb, &av->root, u.rb)
 			n++;
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 15/56] ubi: Rework Fastmap attach base code

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Introduce a new list to the UBI attach information
object to be able to deal better with old and corrupted
Fastmap eraseblocks.
Also move more Fastmap specific code into fastmap.c.

Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflicts in attach and fastmap]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c  | 100 ++++++++++++++++++++++++++++++++--------------
 drivers/mtd/ubi/fastmap.c |  36 +++++++++++++++--
 drivers/mtd/ubi/ubi.h     |  28 ++++++++++++-
 drivers/mtd/ubi/wl.c      |  39 ++++++++++++++----
 4 files changed, 160 insertions(+), 43 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index 9cf9b2ae98e3..4869c6778b3d 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -169,6 +169,40 @@ static int add_corrupted(struct ubi_attach_info *ai, int pnum, int ec)
 }
 
 /**
+ * add_fastmap - add a Fastmap related physical eraseblock.
+ * @ai: attaching information
+ * @pnum: physical eraseblock number the VID header came from
+ * @vid_hdr: the volume identifier header
+ * @ec: erase counter of the physical eraseblock
+ *
+ * This function allocates a 'struct ubi_ainf_peb' object for a Fastamp
+ * physical eraseblock @pnum and adds it to the 'fastmap' list.
+ * Such blocks can be Fastmap super and data blocks from both the most
+ * recent Fastmap we're attaching from or from old Fastmaps which will
+ * be erased.
+ */
+static int add_fastmap(struct ubi_attach_info *ai, int pnum,
+		       struct ubi_vid_hdr *vid_hdr, int ec)
+{
+	struct ubi_ainf_peb *aeb;
+
+	aeb = kmalloc(sizeof(*aeb), GFP_KERNEL);
+	if (!aeb)
+		return -ENOMEM;
+
+	aeb->pnum = pnum;
+	aeb->vol_id = be32_to_cpu(vidh->vol_id);
+	aeb->sqnum = be64_to_cpu(vidh->sqnum);
+	aeb->ec = ec;
+	list_add(&aeb->u.list, &ai->fastmap);
+
+	dbg_bld("add to fastmap list: PEB %d, vol_id %d, sqnum: %llu", pnum,
+		aeb->vol_id, aeb->sqnum);
+
+	return 0;
+}
+
+/**
  * validate_vid_hdr - check volume identifier header.
  * @ubi: UBI device description object
  * @vid_hdr: the volume identifier header to check
@@ -811,18 +845,15 @@ static bool vol_ignored(int vol_id)
  * @ubi: UBI device description object
  * @ai: attaching information
  * @pnum: the physical eraseblock number
- * @vid: The volume ID of the found volume will be stored in this pointer
- * @sqnum: The sqnum of the found volume will be stored in this pointer
  *
  * This function reads UBI headers of PEB @pnum, checks them, and adds
  * information about this PEB to the corresponding list or RB-tree in the
  * "attaching info" structure. Returns zero if the physical eraseblock was
  * successfully handled and a negative error code in case of failure.
  */
-static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
-		    int pnum, int *vid, unsigned long long *sqnum)
+static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum)
 {
-	long long uninitialized_var(ec);
+	long long ec;
 	int err, bitflips = 0, vol_id = -1, ec_err = 0;
 
 	dbg_bld("scan PEB %d", pnum);
@@ -994,10 +1025,6 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	}
 
 	vol_id = be32_to_cpu(vidh->vol_id);
-	if (vid)
-		*vid = vol_id;
-	if (sqnum)
-		*sqnum = be64_to_cpu(vidh->sqnum);
 	if (vol_id > UBI_MAX_VOLUMES && !vol_ignored(vol_id)) {
 		int lnum = be32_to_cpu(vidh->lnum);
 
@@ -1038,7 +1065,12 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (ec_err)
 		ubi_warn(ubi, "valid VID header but corrupted EC header at PEB %d",
 			 pnum);
-	err = ubi_add_to_av(ubi, ai, pnum, ec, vidh, bitflips);
+
+	if (ubi_is_fm_vol(vol_id))
+		err = add_fastmap(ai, pnum, vidh, ec);
+	else
+		err = ubi_add_to_av(ubi, ai, pnum, ec, vidh, bitflips);
+
 	if (err)
 		return err;
 
@@ -1187,6 +1219,10 @@ static void destroy_ai(struct ubi_attach_info *ai)
 		list_del(&aeb->u.list);
 		kfree(aeb);
 	}
+	list_for_each_entry_safe(aeb, aeb_tmp, &ai->fastmap, u.list) {
+		list_del(&aeb->u.list);
+		kfree(aeb);
+	}
 
 	/* Destroy the volume RB-tree */
 	rb = ai->volumes.rb_node;
@@ -1243,7 +1279,7 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 	for (pnum = start; pnum < ubi->peb_count; pnum++) {
 		dbg_gen("process PEB %d", pnum);
-		err = scan_peb(ubi, ai, pnum, NULL, NULL);
+		err = scan_peb(ubi, ai, pnum);
 		if (err < 0)
 			goto out_vidh;
 	}
@@ -1309,6 +1345,7 @@ static struct ubi_attach_info *alloc_ai(void)
 	INIT_LIST_HEAD(&ai->free);
 	INIT_LIST_HEAD(&ai->erase);
 	INIT_LIST_HEAD(&ai->alien);
+	INIT_LIST_HEAD(&ai->fastmap);
 	ai->volumes = RB_ROOT;
 
 	return ai;
@@ -1328,51 +1365,52 @@ static struct ubi_attach_info *alloc_ai(void)
  */
 static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 {
-	int err, pnum, fm_anchor = -1;
-	unsigned long long max_sqnum = 0;
+	int err, pnum;
+	struct ubi_attach_info *scan_ai;
 
 	err = -ENOMEM;
 
+	scan_ai = alloc_ai();
+	if (!scan_ai)
+		goto out;
+
 	ech = kzalloc(ubi->ec_hdr_alsize, GFP_KERNEL);
 	if (!ech)
-		goto out;
+		goto out_ai;
 
 	vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
 	if (!vidh)
 		goto out_ech;
 
 	for (pnum = 0; pnum < UBI_FM_MAX_START; pnum++) {
-		int vol_id = -1;
-		unsigned long long sqnum = -1;
-
 		dbg_gen("process PEB %d", pnum);
-		err = scan_peb(ubi, *ai, pnum, &vol_id, &sqnum);
+		err = scan_peb(ubi, scan_ai, pnum);
 		if (err < 0)
 			goto out_vidh;
-
-		if (vol_id == UBI_FM_SB_VOLUME_ID && sqnum > max_sqnum) {
-			max_sqnum = sqnum;
-			fm_anchor = pnum;
-		}
 	}
 
 	ubi_free_vid_hdr(ubi, vidh);
 	kfree(ech);
 
-	if (fm_anchor < 0)
-		return UBI_NO_FASTMAP;
-
-	destroy_ai(*ai);
-	*ai = alloc_ai();
-	if (!*ai)
-		return -ENOMEM;
+	err = ubi_scan_fastmap(ubi, *ai, scan_ai);
+	if (err) {
+		/*
+		 * Didn't attach via fastmap, do a full scan but reuse what
+		 * we've aready scanned.
+		 */
+		destroy_ai(*ai);
+		*ai = scan_ai;
+	} else
+		destroy_ai(scan_ai);
 
-	return ubi_scan_fastmap(ubi, *ai, fm_anchor);
+	return err;
 
 out_vidh:
 	ubi_free_vid_hdr(ubi, vidh);
 out_ech:
 	kfree(ech);
+out_ai:
+	destroy_ai(scan_ai);
 out:
 	return err;
 }
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 205174495b0d..4a8de1db407a 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -784,27 +784,57 @@ fail:
 }
 
 /**
+ * find_fm_anchor - find the most recent Fastmap superblock (anchor)
+ * @ai: UBI attach info to be filled
+ */
+static int find_fm_anchor(struct ubi_attach_info *ai)
+{
+	int ret = -1;
+	struct ubi_ainf_peb *aeb;
+	unsigned long long max_sqnum = 0;
+
+	list_for_each_entry(aeb, &ai->fastmap, u.list) {
+		if (aeb->vol_id == UBI_FM_SB_VOLUME_ID && aeb->sqnum > max_sqnum) {
+			max_sqnum = aeb->sqnum;
+			ret = aeb->pnum;
+		}
+	}
+
+	return ret;
+}
+
+/**
  * ubi_scan_fastmap - scan the fastmap.
  * @ubi: UBI device object
  * @ai: UBI attach info to be filled
- * @fm_anchor: The fastmap starts at this PEB
+ * @scan_ai: UBI attach info from the first 64 PEBs,
+ *           used to find the most recent Fastmap data structure
  *
  * Returns 0 on success, UBI_NO_FASTMAP if no fastmap was found,
  * UBI_BAD_FASTMAP if one was found but is not usable.
  * < 0 indicates an internal error.
  */
 int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
-		     int fm_anchor)
+		     struct ubi_attach_info *scan_ai)
 {
 	struct ubi_fm_sb *fmsb, *fmsb2;
 	struct ubi_vid_hdr *vh;
 	struct ubi_ec_hdr *ech;
 	struct ubi_fastmap_layout *fm;
-	int i, used_blocks, pnum, ret = 0;
+	struct ubi_ainf_peb *tmp_aeb, *aeb;
+	int i, used_blocks, pnum, fm_anchor, ret = 0;
 	size_t fm_size;
 	__be32 crc, tmp_crc;
 	unsigned long long sqnum = 0;
 
+	fm_anchor = find_fm_anchor(scan_ai);
+	if (fm_anchor < 0)
+		return UBI_NO_FASTMAP;
+
+	/* Move all (possible) fastmap blocks into our new attach structure. */
+	list_for_each_entry_safe(aeb, tmp_aeb, &scan_ai->fastmap, u.list)
+		list_move_tail(&aeb->u.list, &ai->fastmap);
+
 	memset(ubi->fm_buf, 0, ubi->fm_size);
 
 	fmsb = kmalloc(sizeof(*fmsb), GFP_KERNEL);
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 5899ad7c92b8..2bd2d9e1ed88 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -639,6 +639,8 @@ struct ubi_ainf_volume {
  * @erase: list of physical eraseblocks which have to be erased
  * @alien: list of physical eraseblocks which should not be used by UBI (e.g.,
  *         those belonging to "preserve"-compatible internal volumes)
+ * @fastmap: list of physical eraseblocks which relate to fastmap (e.g.,
+ *           eraseblocks of the current and not yet erased old fastmap blocks)
  * @corr_peb_count: count of PEBs in the @corr list
  * @empty_peb_count: count of PEBs which are presumably empty (contain only
  *                   0xFF bytes)
@@ -667,6 +669,7 @@ struct ubi_attach_info {
 	struct list_head free;
 	struct list_head erase;
 	struct list_head alien;
+	struct list_head fastmap;
 	int corr_peb_count;
 	int empty_peb_count;
 	int alien_peb_count;
@@ -843,7 +846,7 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 size_t ubi_calc_fm_size(struct ubi_device *ubi);
 int ubi_update_fastmap(struct ubi_device *ubi);
 int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
-		     int fm_anchor);
+		     struct ubi_attach_info *scan_ai);
 #else
 static inline int ubi_update_fastmap(struct ubi_device *ubi) { return 0; }
 #endif
@@ -1033,4 +1036,27 @@ static inline bool ubi_is_fm_vol(int vol_id)
 	return false;
 }
 
+/**
+ * ubi_find_fm_block - check whether a PEB is part of the current Fastmap.
+ * @ubi: UBI device description object
+ * @pnum: physical eraseblock to look for
+ *
+ * This function returns a wear leveling object if @pnum relates to the current
+ * fastmap, @NULL otherwise.
+ */
+static inline struct ubi_wl_entry *ubi_find_fm_block(const struct ubi_device *ubi,
+						     int pnum)
+{
+	int i;
+
+	if (ubi->fm) {
+		for (i = 0; i < ubi->fm->used_blocks; i++) {
+			if (ubi->fm->e[i]->pnum == pnum)
+				return ubi->fm->e[i];
+		}
+	}
+
+	return NULL;
+}
+
 #endif /* !__UBI_UBI_H__ */
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index eb1f24c74807..eef9c17c2d0f 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -1457,19 +1457,42 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 		}
 	}
 
-	dbg_wl("found %i PEBs", found_pebs);
+	list_for_each_entry(aeb, &ai->fastmap, u.list) {
+		e = ubi_find_fm_block(ubi, aeb->pnum);
+
+		if (e) {
+			ubi_assert(!ubi->lookuptbl[e->pnum]);
+			ubi->lookuptbl[e->pnum] = e;
+		} else {
+			/*
+			 * Usually old Fastmap PEBs are scheduled for erasure
+			 * and we don't have to care about them but if we face
+			 * an power cut before scheduling them we need to
+			 * take care of them here.
+			 */
+			if (ubi->lookuptbl[aeb->pnum])
+				continue;
 
-	if (ubi->fm) {
-		ubi_assert(ubi->good_peb_count ==
-			   found_pebs + ubi->fm->used_blocks);
+			e = kmalloc(sizeof(*e), GFP_KERNEL);
+			if (!e)
+				goto out_free;
 
-		for (i = 0; i < ubi->fm->used_blocks; i++) {
-			e = ubi->fm->e[i];
+			e->pnum = aeb->pnum;
+			e->ec = aeb->ec;
+			ubi_assert(!ubi->lookuptbl[e->pnum]);
 			ubi->lookuptbl[e->pnum] = e;
+			if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0)) {
+				wl_entry_destroy(ubi, e);
+				goto out_free;
+			}
 		}
+
+		found_pebs++;
 	}
-	else
-		ubi_assert(ubi->good_peb_count == found_pebs);
+
+	dbg_wl("found %i PEBs", found_pebs);
+
+	ubi_assert(ubi->good_peb_count == found_pebs);
 
 	reserved_pebs = WL_RESERVED_PEBS;
 	ubi_fastmap_init(ubi, &reserved_pebs);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 16/56] ubi: Check whether the Fastmap anchor matches the super block

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

This helps to detect cases where an user copies an UBI image to
another target with different bad blocks.

Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 4a8de1db407a..d60f89fdb369 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -908,6 +908,13 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			goto free_hdr;
 		}
 
+		if (i == 0 && pnum != fm_anchor) {
+			ubi_err(ubi, "Fastmap anchor PEB mismatch: PEB: %i vs. %i",
+				pnum, fm_anchor);
+			ret = UBI_BAD_FASTMAP;
+			goto free_hdr;
+		}
+
 		ret = ubi_io_read_ec_hdr(ubi, pnum, ech, 0);
 		if (ret && ret != UBI_IO_BITFLIPS) {
 			ubi_err(ubi, "unable to read fastmap block# %i EC (PEB: %i)",
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 17/56] ubi: Be more paranoid while seaching for the most recent Fastmap

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Since PEB erasure is asynchornous it can happen that there is
more than one Fastmap on the MTD. This is fine because the attach logic
will pick the Fastmap data structure with the highest sequence number.

On a not so well configured MTD stack spurious ECC errors are common.
Causes can be different, bad hardware, wrong operating modes, etc...
If the most current Fastmap renders bad due to ECC errors UBI might
pick an older Fastmap to attach from.
While this can only happen on an anyway broken setup it will show
completely different sympthoms and makes finding the root cause much
more difficult.
So, be debug friendly and fall back to scanning mode of we're facing
an ECC error while scanning for Fastmap.

Cc: <stable@vger.kernel.org>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 28 ++++++++++++++++++++++++----
 drivers/mtd/ubi/ubi.h    |  3 +++
 2 files changed, 27 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index 4869c6778b3d..a5d48932ed33 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -845,13 +845,15 @@ static bool vol_ignored(int vol_id)
  * @ubi: UBI device description object
  * @ai: attaching information
  * @pnum: the physical eraseblock number
+ * @fast: true if we're scanning for a Fastmap
  *
  * This function reads UBI headers of PEB @pnum, checks them, and adds
  * information about this PEB to the corresponding list or RB-tree in the
  * "attaching info" structure. Returns zero if the physical eraseblock was
  * successfully handled and a negative error code in case of failure.
  */
-static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum)
+static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
+		    int pnum, bool fast)
 {
 	long long ec;
 	int err, bitflips = 0, vol_id = -1, ec_err = 0;
@@ -969,6 +971,20 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum
 			 */
 			ai->maybe_bad_peb_count += 1;
 	case UBI_IO_BAD_HDR:
+			/*
+			 * If we're facing a bad VID header we have to drop *all*
+			 * Fastmap data structures we find. The most recent Fastmap
+			 * could be bad and therefore there is a chance that we attach
+			 * from an old one. On a fine MTD stack a PEB must not render
+			 * bad all of a sudden, but the reality is different.
+			 * So, let's be paranoid and help finding the root cause by
+			 * falling back to scanning mode instead of attaching with a
+			 * bad EBA table and cause data corruption which is hard to
+			 * analyze.
+			 */
+			if (fast)
+				ai->force_full_scan = 1;
+
 		if (ec_err)
 			/*
 			 * Both headers are corrupted. There is a possibility
@@ -1279,7 +1295,7 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 	for (pnum = start; pnum < ubi->peb_count; pnum++) {
 		dbg_gen("process PEB %d", pnum);
-		err = scan_peb(ubi, ai, pnum);
+		err = scan_peb(ubi, ai, pnum, false);
 		if (err < 0)
 			goto out_vidh;
 	}
@@ -1384,7 +1400,7 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 
 	for (pnum = 0; pnum < UBI_FM_MAX_START; pnum++) {
 		dbg_gen("process PEB %d", pnum);
-		err = scan_peb(ubi, scan_ai, pnum);
+		err = scan_peb(ubi, scan_ai, pnum, true);
 		if (err < 0)
 			goto out_vidh;
 	}
@@ -1392,7 +1408,11 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 	ubi_free_vid_hdr(ubi, vidh);
 	kfree(ech);
 
-	err = ubi_scan_fastmap(ubi, *ai, scan_ai);
+	if (scan_ai->force_full_scan)
+		err = UBI_NO_FASTMAP;
+	else
+		err = ubi_scan_fastmap(ubi, *ai, scan_ai);
+
 	if (err) {
 		/*
 		 * Didn't attach via fastmap, do a full scan but reuse what
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 2bd2d9e1ed88..c2f54a4b0b3f 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -651,6 +651,8 @@ struct ubi_ainf_volume {
  * @vols_found: number of volumes found
  * @highest_vol_id: highest volume ID
  * @is_empty: flag indicating whether the MTD device is empty or not
+ * @force_full_scan: flag indicating whether we need to do a full scan and drop
+		     all existing Fastmap data structures
  * @min_ec: lowest erase counter value
  * @max_ec: highest erase counter value
  * @max_sqnum: highest sequence number value
@@ -678,6 +680,7 @@ struct ubi_attach_info {
 	int vols_found;
 	int highest_vol_id;
 	int is_empty;
+	int force_full_scan;
 	int min_ec;
 	int max_ec;
 	unsigned long long max_sqnum;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 18/56] UBI: fastmap: use ubi_find_volume() instead of open coding it

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

process_pool_aeb() re-implements the logic found in ubi_find_volume().
Call ubi_find_volume() to avoid this duplication.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 23 +++--------------------
 1 file changed, 3 insertions(+), 20 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index d60f89fdb369..37bb93bc2a69 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -303,9 +303,7 @@ static int process_pool_aeb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			    struct ubi_vid_hdr *new_vh,
 			    struct ubi_ainf_peb *new_aeb)
 {
-	struct ubi_ainf_volume *av, *tmp_av = NULL;
-	struct rb_node **p = &ai->volumes.rb_node, *parent = NULL;
-	int found = 0;
+	struct ubi_ainf_volume *av;
 
 	if (be32_to_cpu(new_vh->vol_id) == UBI_FM_SB_VOLUME_ID ||
 		be32_to_cpu(new_vh->vol_id) == UBI_FM_DATA_VOLUME_ID) {
@@ -315,23 +313,8 @@ static int process_pool_aeb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	}
 
 	/* Find the volume this SEB belongs to */
-	while (*p) {
-		parent = *p;
-		tmp_av = rb_entry(parent, struct ubi_ainf_volume, rb);
-
-		if (be32_to_cpu(new_vh->vol_id) > tmp_av->vol_id)
-			p = &(*p)->rb_left;
-		else if (be32_to_cpu(new_vh->vol_id) < tmp_av->vol_id)
-			p = &(*p)->rb_right;
-		else {
-			found = 1;
-			break;
-		}
-	}
-
-	if (found)
-		av = tmp_av;
-	else {
+	av = ubi_find_av(ai, be32_to_cpu(new_vh->vol_id));
+	if (!av) {
 		ubi_err(ubi, "orphaned volume in fastmap pool!");
 		kfree(new_aeb);
 		return UBI_BAD_FASTMAP;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 19/56] UBI: fix add_fastmap() to use the vid_hdr passed in argument

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

add_fastmap() is passed a ubi_vid_hdr pointer in argument, but is
referencing the global vidh pointer.
Even if this is correct from a functional point of view (vidh and vid_hdr
point to the same object), it is confusing.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index a5d48932ed33..4d3a936ae861 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -191,8 +191,8 @@ static int add_fastmap(struct ubi_attach_info *ai, int pnum,
 		return -ENOMEM;
 
 	aeb->pnum = pnum;
-	aeb->vol_id = be32_to_cpu(vidh->vol_id);
-	aeb->sqnum = be64_to_cpu(vidh->sqnum);
+	aeb->vol_id = be32_to_cpu(vid_hdr->vol_id);
+	aeb->sqnum = be64_to_cpu(vid_hdr->sqnum);
 	aeb->ec = ec;
 	list_add(&aeb->u.list, &ai->fastmap);
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 20/56] UBI: fastmap: avoid multiple be32_to_cpu() when unneccesary

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

process_pool_aeb() does several times the be32_to_cpu(new_vh->vol_id)
operation. Create a temporary variable and do it once.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 37bb93bc2a69..6ef94189b612 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -303,24 +303,24 @@ static int process_pool_aeb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			    struct ubi_vid_hdr *new_vh,
 			    struct ubi_ainf_peb *new_aeb)
 {
+	int vol_id = be32_to_cpu(new_vh->vol_id);
 	struct ubi_ainf_volume *av;
 
-	if (be32_to_cpu(new_vh->vol_id) == UBI_FM_SB_VOLUME_ID ||
-		be32_to_cpu(new_vh->vol_id) == UBI_FM_DATA_VOLUME_ID) {
+	if (vol_id == UBI_FM_SB_VOLUME_ID || vol_id == UBI_FM_DATA_VOLUME_ID) {
 		kfree(new_aeb);
 
 		return 0;
 	}
 
 	/* Find the volume this SEB belongs to */
-	av = ubi_find_av(ai, be32_to_cpu(new_vh->vol_id));
+	av = ubi_find_av(ai, vol_id);
 	if (!av) {
 		ubi_err(ubi, "orphaned volume in fastmap pool!");
 		kfree(new_aeb);
 		return UBI_BAD_FASTMAP;
 	}
 
-	ubi_assert(be32_to_cpu(new_vh->vol_id) == av->vol_id);
+	ubi_assert(vol_id == av->vol_id);
 
 	return update_vol(ubi, ai, av, new_vh, new_aeb);
 }
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 21/56] UBI: fastmap: scrub PEB when bitflips are detected in a free PEB EC header

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

scan_pool() does not mark the PEB for scrubing when bitflips are
detected in the EC header of a free PEB (VID header region left to
0xff).
Make sure we scrub the PEB in this case.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Fixes: dbb7d2a88d2a ("UBI: Add fastmap core")
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 6ef94189b612..c10ef0650760 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -431,10 +431,11 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			unsigned long long ec = be64_to_cpu(ech->ec);
 			unmap_peb(ai, pnum);
 			dbg_bld("Adding PEB to free: %i", pnum);
+
 			if (err == UBI_IO_FF_BITFLIPS)
-				add_aeb(ai, lfree, pnum, ec, 1);
-			else
-				add_aeb(ai, lfree, pnum, ec, 0);
+				scrub = 1;
+
+			add_aeb(ai, lfree, pnum, ec, scrub);
 			continue;
 		} else if (err == 0 || err == UBI_IO_BITFLIPS) {
 			dbg_bld("Found non empty PEB:%i in pool", pnum);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 22/56] UBI: factorize code used to manipulate volumes at attach time

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Volume creation/search code is duplicated in a few places (fastmap and
non fastmap code). Create some helpers to factorize the code.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c  | 151 +++++++++++++++++++++++++++++++++-------------
 drivers/mtd/ubi/fastmap.c |  27 +--------
 drivers/mtd/ubi/ubi.h     |   1 +
 3 files changed, 112 insertions(+), 67 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index 4d3a936ae861..d450f6c0d65b 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -89,6 +89,92 @@ static int self_check_ai(struct ubi_device *ubi, struct ubi_attach_info *ai);
 static struct ubi_ec_hdr *ech;
 static struct ubi_vid_hdr *vidh;
 
+#define AV_FIND		BIT(0)
+#define AV_ADD		BIT(1)
+#define AV_FIND_OR_ADD	(AV_FIND | AV_ADD)
+
+/**
+ * find_or_add_av - internal function to find a volume, add a volume or do
+ *		    both (find and add if missing).
+ * @ai: attaching information
+ * @vol_id: the requested volume ID
+ * @flags: a combination of the %AV_FIND and %AV_ADD flags describing the
+ *	   expected operation. If only %AV_ADD is set, -EEXIST is returned
+ *	   if the volume already exists. If only %AV_FIND is set, NULL is
+ *	   returned if the volume does not exist. And if both flags are
+ *	   set, the helper first tries to find an existing volume, and if
+ *	   it does not exist it creates a new one.
+ * @created: in value used to inform the caller whether it"s a newly created
+ *	     volume or not.
+ *
+ * This function returns a pointer to a volume description or an ERR_PTR if
+ * the operation failed. It can also return NULL if only %AV_FIND is set and
+ * the volume does not exist.
+ */
+static struct ubi_ainf_volume *find_or_add_av(struct ubi_attach_info *ai,
+					      int vol_id, unsigned int flags,
+					      bool *created)
+{
+	struct ubi_ainf_volume *av;
+	struct rb_node **p = &ai->volumes.rb_node, *parent = NULL;
+
+	/* Walk the volume RB-tree to look if this volume is already present */
+	while (*p) {
+		parent = *p;
+		av = rb_entry(parent, struct ubi_ainf_volume, rb);
+
+		if (vol_id == av->vol_id) {
+			*created = false;
+
+			if (!(flags & AV_FIND))
+				return ERR_PTR(-EEXIST);
+
+			return av;
+		}
+
+		if (vol_id > av->vol_id)
+			p = &(*p)->rb_left;
+		else
+			p = &(*p)->rb_right;
+	}
+
+	if (!(flags & AV_ADD))
+		return NULL;
+
+	/* The volume is absent - add it */
+	av = kzalloc(sizeof(*av), GFP_KERNEL);
+	if (!av)
+		return ERR_PTR(-ENOMEM);
+
+	av->vol_id = vol_id;
+
+	if (vol_id > ai->highest_vol_id)
+		ai->highest_vol_id = vol_id;
+
+	rb_link_node(&av->rb, parent, p);
+	rb_insert_color(&av->rb, &ai->volumes);
+	ai->vols_found += 1;
+	*created = true;
+	dbg_bld("added volume %d", vol_id);
+	return av;
+}
+
+/**
+ * ubi_find_or_add_av - search for a volume in the attaching information and
+ *			add one if it does not exist.
+ * @ai: attaching information
+ * @vol_id: the requested volume ID
+ * @created: whether the volume has been created or not
+ *
+ * This function returns a pointer to the new volume description or an
+ * ERR_PTR if the operation failed.
+ */
+static struct ubi_ainf_volume *ubi_find_or_add_av(struct ubi_attach_info *ai,
+						  int vol_id, bool *created)
+{
+	return find_or_add_av(ai, vol_id, AV_FIND_OR_ADD, created);
+}
+
 /**
  * add_to_list - add physical eraseblock to a list.
  * @ai: attaching information
@@ -288,44 +374,20 @@ static struct ubi_ainf_volume *add_volume(struct ubi_attach_info *ai,
 					  const struct ubi_vid_hdr *vid_hdr)
 {
 	struct ubi_ainf_volume *av;
-	struct rb_node **p = &ai->volumes.rb_node, *parent = NULL;
+	bool created;
 
 	ubi_assert(vol_id == be32_to_cpu(vid_hdr->vol_id));
 
-	/* Walk the volume RB-tree to look if this volume is already present */
-	while (*p) {
-		parent = *p;
-		av = rb_entry(parent, struct ubi_ainf_volume, rb);
-
-		if (vol_id == av->vol_id)
-			return av;
+	av = ubi_find_or_add_av(ai, vol_id, &created);
+	if (IS_ERR(av) || !created)
+		return av;
 
-		if (vol_id > av->vol_id)
-			p = &(*p)->rb_left;
-		else
-			p = &(*p)->rb_right;
-	}
-
-	/* The volume is absent - add it */
-	av = kmalloc(sizeof(struct ubi_ainf_volume), GFP_KERNEL);
-	if (!av)
-		return ERR_PTR(-ENOMEM);
-
-	av->highest_lnum = av->leb_count = 0;
-	av->vol_id = vol_id;
-	av->root = RB_ROOT;
 	av->used_ebs = be32_to_cpu(vid_hdr->used_ebs);
 	av->data_pad = be32_to_cpu(vid_hdr->data_pad);
 	av->compat = vid_hdr->compat;
 	av->vol_type = vid_hdr->vol_type == UBI_VID_DYNAMIC ? UBI_DYNAMIC_VOLUME
 							    : UBI_STATIC_VOLUME;
-	if (vol_id > ai->highest_vol_id)
-		ai->highest_vol_id = vol_id;
 
-	rb_link_node(&av->rb, parent, p);
-	rb_insert_color(&av->rb, &ai->volumes);
-	ai->vols_found += 1;
-	dbg_bld("added volume %d", vol_id);
 	return av;
 }
 
@@ -620,6 +682,21 @@ int ubi_add_to_av(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum,
 }
 
 /**
+ * ubi_add_av - add volume to the attaching information.
+ * @ai: attaching information
+ * @vol_id: the requested volume ID
+ *
+ * This function returns a pointer to the new volume description or an
+ * ERR_PTR if the operation failed.
+ */
+struct ubi_ainf_volume *ubi_add_av(struct ubi_attach_info *ai, int vol_id)
+{
+	bool created;
+
+	return find_or_add_av(ai, vol_id, AV_ADD, &created);
+}
+
+/**
  * ubi_find_av - find volume in the attaching information.
  * @ai: attaching information
  * @vol_id: the requested volume ID
@@ -630,22 +707,10 @@ int ubi_add_to_av(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum,
 struct ubi_ainf_volume *ubi_find_av(const struct ubi_attach_info *ai,
 				    int vol_id)
 {
-	struct ubi_ainf_volume *av;
-	struct rb_node *p = ai->volumes.rb_node;
-
-	while (p) {
-		av = rb_entry(p, struct ubi_ainf_volume, rb);
-
-		if (vol_id == av->vol_id)
-			return av;
-
-		if (vol_id > av->vol_id)
-			p = p->rb_left;
-		else
-			p = p->rb_right;
-	}
+	bool created;
 
-	return NULL;
+	return find_or_add_av((struct ubi_attach_info *)ai, vol_id, AV_FIND,
+			      &created);
 }
 
 /**
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index c10ef0650760..b0979af62f18 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -120,40 +120,19 @@ static struct ubi_ainf_volume *add_vol(struct ubi_attach_info *ai, int vol_id,
 				       int last_eb_bytes)
 {
 	struct ubi_ainf_volume *av;
-	struct rb_node **p = &ai->volumes.rb_node, *parent = NULL;
 
-	while (*p) {
-		parent = *p;
-		av = rb_entry(parent, struct ubi_ainf_volume, rb);
-
-		if (vol_id > av->vol_id)
-			p = &(*p)->rb_left;
-		else if (vol_id < av->vol_id)
-			p = &(*p)->rb_right;
-		else
-			return ERR_PTR(-EINVAL);
-	}
+	av = ubi_add_av(ai, vol_id);
+	if (IS_ERR(av))
+		return av;
 
-	av = kmalloc(sizeof(struct ubi_ainf_volume), GFP_KERNEL);
-	if (!av)
-		goto out;
-
-	av->highest_lnum = av->leb_count = av->used_ebs = 0;
-	av->vol_id = vol_id;
 	av->data_pad = data_pad;
 	av->last_data_size = last_eb_bytes;
 	av->compat = 0;
 	av->vol_type = vol_type;
-	av->root = RB_ROOT;
 	if (av->vol_type == UBI_STATIC_VOLUME)
 		av->used_ebs = used_ebs;
 
 	dbg_bld("found volume (ID %i)", vol_id);
-
-	rb_link_node(&av->rb, parent, p);
-	rb_insert_color(&av->rb, &ai->volumes);
-
-out:
 	return av;
 }
 
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index c2f54a4b0b3f..eb81298beff3 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -729,6 +729,7 @@ extern struct blocking_notifier_head ubi_notifiers;
 /* attach.c */
 int ubi_add_to_av(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum,
 		  int ec, const struct ubi_vid_hdr *vid_hdr, int bitflips);
+struct ubi_ainf_volume *ubi_add_av(struct ubi_attach_info *ai, int vol_id);
 struct ubi_ainf_volume *ubi_find_av(const struct ubi_attach_info *ai,
 				    int vol_id);
 void ubi_remove_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 23/56] UBI: factorize destroy_av() and ubi_remove_av() code

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Those functions are pretty much doing the same thing, except
ubi_remove_av() is putting the aeb elements attached to the volume into
the ai->erase list and the destroy_av() is freeing them.

Rework destroy_av() to handle both cases.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 25 ++++++++++++-------------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index d450f6c0d65b..f42c33a10fef 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -713,6 +713,9 @@ struct ubi_ainf_volume *ubi_find_av(const struct ubi_attach_info *ai,
 			      &created);
 }
 
+static void destroy_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av,
+		       struct list_head *list);
+
 /**
  * ubi_remove_av - delete attaching information about a volume.
  * @ai: attaching information
@@ -720,19 +723,10 @@ struct ubi_ainf_volume *ubi_find_av(const struct ubi_attach_info *ai,
  */
 void ubi_remove_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av)
 {
-	struct rb_node *rb;
-	struct ubi_ainf_peb *aeb;
-
 	dbg_bld("remove attaching information about volume %d", av->vol_id);
 
-	while ((rb = rb_first(&av->root))) {
-		aeb = rb_entry(rb, struct ubi_ainf_peb, u.rb);
-		rb_erase(&aeb->u.rb, &av->root);
-		list_add_tail(&aeb->u.list, &ai->erase);
-	}
-
 	rb_erase(&av->rb, &ai->volumes);
-	kfree(av);
+	destroy_av(ai, av, &ai->erase);
 	ai->vols_found -= 1;
 }
 
@@ -1245,10 +1239,12 @@ static int late_analysis(struct ubi_device *ubi, struct ubi_attach_info *ai)
  * destroy_av - free volume attaching information.
  * @av: volume attaching information
  * @ai: attaching information
+ * @list: put the aeb elements in there if !NULL, otherwise free them
  *
  * This function destroys the volume attaching information.
  */
-static void destroy_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av)
+static void destroy_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av,
+		       struct list_head *list)
 {
 	struct ubi_ainf_peb *aeb;
 	struct rb_node *this = av->root.rb_node;
@@ -1268,7 +1264,10 @@ static void destroy_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av)
 					this->rb_right = NULL;
 			}
 
-			kfree(aeb);
+			if (list)
+				list_add_tail(&aeb->u.list, list);
+			else
+				kfree(aeb);
 		}
 	}
 	kfree(av);
@@ -1323,7 +1322,7 @@ static void destroy_ai(struct ubi_attach_info *ai)
 					rb->rb_right = NULL;
 			}
 
-			destroy_av(ai, av);
+			destroy_av(ai, av, NULL);
 		}
 	}
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 24/56] UBI: fastmap: use ubi_rb_for_each_entry() in unmap_peb()

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Use the ubi_rb_for_each_entry() macro instead of open-coding it.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index b0979af62f18..683ac0d9c4ab 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -318,12 +318,8 @@ static void unmap_peb(struct ubi_attach_info *ai, int pnum)
 	struct rb_node *node, *node2;
 	struct ubi_ainf_peb *aeb;
 
-	for (node = rb_first(&ai->volumes); node; node = rb_next(node)) {
-		av = rb_entry(node, struct ubi_ainf_volume, rb);
-
-		for (node2 = rb_first(&av->root); node2;
-		     node2 = rb_next(node2)) {
-			aeb = rb_entry(node2, struct ubi_ainf_peb, u.rb);
+	ubi_rb_for_each_entry(node, av, &ai->volumes, rb) {
+		ubi_rb_for_each_entry(node2, aeb, &av->root, u.rb) {
 			if (aeb->pnum == pnum) {
 				rb_erase(&aeb->u.rb, &av->root);
 				av->leb_count--;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 25/56] UBI: fastmap: use ubi_io_{read, write}_data() instead of ubi_io_{read, write}()

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

ubi_io_{read,write}_data() are wrappers around ubi_io_{read/write}() that
are used to read/write eraseblock payload data, which is exactly what
fastmap does when calling ubi_io_{read,write}().

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 683ac0d9c4ab..2bf626f8075e 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -809,7 +809,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		goto out;
 	}
 
-	ret = ubi_io_read(ubi, fmsb, fm_anchor, ubi->leb_start, sizeof(*fmsb));
+	ret = ubi_io_read_data(ubi, fmsb, fm_anchor, 0, sizeof(*fmsb));
 	if (ret && ret != UBI_IO_BITFLIPS)
 		goto free_fm_sb;
 	else if (ret == UBI_IO_BITFLIPS)
@@ -927,8 +927,8 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		if (sqnum < be64_to_cpu(vh->sqnum))
 			sqnum = be64_to_cpu(vh->sqnum);
 
-		ret = ubi_io_read(ubi, ubi->fm_buf + (ubi->leb_size * i), pnum,
-				  ubi->leb_start, ubi->leb_size);
+		ret = ubi_io_read_data(ubi, ubi->fm_buf + (ubi->leb_size * i),
+				       pnum, 0, ubi->leb_size);
 		if (ret && ret != UBI_IO_BITFLIPS) {
 			ubi_err(ubi, "unable to read fastmap block# %i (PEB: %i, "
 				"err: %i)", i, pnum, ret);
@@ -1218,8 +1218,8 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 	}
 
 	for (i = 0; i < new_fm->used_blocks; i++) {
-		ret = ubi_io_write(ubi, fm_raw + (i * ubi->leb_size),
-			new_fm->e[i]->pnum, ubi->leb_start, ubi->leb_size);
+		ret = ubi_io_write_data(ubi, fm_raw + (i * ubi->leb_size),
+					new_fm->e[i]->pnum, 0, ubi->leb_size);
 		if (ret) {
 			ubi_err(ubi, "unable to write fastmap to PEB %i!",
 				new_fm->e[i]->pnum);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 26/56] UBI: provide helpers to allocate and free aeb elements

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

This not only hides the aeb allocation internals (which is always good in
case we ever want to change the allocation system), but also helps us
factorize the initialization of some common fields (ec and pnum).

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed kzalloc/free conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c  | 69 ++++++++++++++++++++++++++++++++++-------------
 drivers/mtd/ubi/fastmap.c | 29 +++++++++-----------
 drivers/mtd/ubi/ubi.h     |  3 +++
 drivers/mtd/ubi/vtbl.c    |  3 ++-
 4 files changed, 69 insertions(+), 35 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index f42c33a10fef..2b1271218804 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -176,6 +176,47 @@ static struct ubi_ainf_volume *ubi_find_or_add_av(struct ubi_attach_info *ai,
 }
 
 /**
+ * ubi_alloc_aeb - allocate an aeb element
+ * @ai: attaching information
+ * @pnum: physical eraseblock number
+ * @ec: erase counter of the physical eraseblock
+ *
+ * Allocate an aeb object and initialize the pnum and ec information.
+ * vol_id and lnum are set to UBI_UNKNOWN, and the other fields are
+ * initialized to zero.
+ * Note that the element is not added in any list or RB tree.
+ */
+struct ubi_ainf_peb *ubi_alloc_aeb(struct ubi_attach_info *ai, int pnum,
+				   int ec)
+{
+	struct ubi_ainf_peb *aeb;
+
+	aeb = kzalloc(sizeof(*aeb), GFP_KERNEL);
+	if (!aeb)
+		return NULL;
+
+	aeb->pnum = pnum;
+	aeb->ec = ec;
+	aeb->vol_id = UBI_UNKNOWN;
+	aeb->lnum = UBI_UNKNOWN;
+
+	return aeb;
+}
+
+/**
+ * ubi_free_aeb - free an aeb element
+ * @ai: attaching information
+ * @aeb: the element to free
+ *
+ * Free an aeb object. The caller must have removed the element from any list
+ * or RB tree.
+ */
+void ubi_free_aeb(struct ubi_attach_info *ai, struct ubi_ainf_peb *aeb)
+{
+	kfree(aeb);
+}
+
+/**
  * add_to_list - add physical eraseblock to a list.
  * @ai: attaching information
  * @pnum: physical eraseblock number to add
@@ -211,14 +252,12 @@ static int add_to_list(struct ubi_attach_info *ai, int pnum, int vol_id,
 	} else
 		BUG();
 
-	aeb = kzalloc(sizeof(*aeb), GFP_KERNEL);
+	aeb = ubi_alloc_aeb(ai, pnum, ec);
 	if (!aeb)
 		return -ENOMEM;
 
-	aeb->pnum = pnum;
 	aeb->vol_id = vol_id;
 	aeb->lnum = lnum;
-	aeb->ec = ec;
 	if (to_head)
 		list_add(&aeb->u.list, list);
 	else
@@ -243,13 +282,11 @@ static int add_corrupted(struct ubi_attach_info *ai, int pnum, int ec)
 
 	dbg_bld("add to corrupted: PEB %d, EC %d", pnum, ec);
 
-	aeb = kzalloc(sizeof(aeb), GFP_KERNEL);
+	aeb = ubi_alloc_aeb(ai, pnum, ec);
 	if (!aeb)
 		return -ENOMEM;
 
 	ai->corr_peb_count += 1;
-	aeb->pnum = pnum;
-	aeb->ec = ec;
 	list_add(&aeb->u.list, &ai->corr);
 	return 0;
 }
@@ -272,14 +309,12 @@ static int add_fastmap(struct ubi_attach_info *ai, int pnum,
 {
 	struct ubi_ainf_peb *aeb;
 
-	aeb = kmalloc(sizeof(*aeb), GFP_KERNEL);
+	aeb = ubi_alloc_aeb(ai, pnum, ec);
 	if (!aeb)
 		return -ENOMEM;
 
-	aeb->pnum = pnum;
 	aeb->vol_id = be32_to_cpu(vid_hdr->vol_id);
 	aeb->sqnum = be64_to_cpu(vid_hdr->sqnum);
-	aeb->ec = ec;
 	list_add(&aeb->u.list, &ai->fastmap);
 
 	dbg_bld("add to fastmap list: PEB %d, vol_id %d, sqnum: %llu", pnum,
@@ -658,12 +693,10 @@ int ubi_add_to_av(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum,
 	if (err)
 		return err;
 
-	aeb = kzalloc(sizeof(*aeb), GFP_KERNEL);
+	aeb = ubi_alloc_aeb(ai, pnum, ec);
 	if (!aeb)
 		return -ENOMEM;
 
-	aeb->ec = ec;
-	aeb->pnum = pnum;
 	aeb->vol_id = vol_id;
 	aeb->lnum = lnum;
 	aeb->scrub = bitflips;
@@ -1267,7 +1300,7 @@ static void destroy_av(struct ubi_attach_info *ai, struct ubi_ainf_volume *av,
 			if (list)
 				list_add_tail(&aeb->u.list, list);
 			else
-				kfree(aeb);
+				ubi_free_aeb(ai, aeb);
 		}
 	}
 	kfree(av);
@@ -1285,23 +1318,23 @@ static void destroy_ai(struct ubi_attach_info *ai)
 
 	list_for_each_entry_safe(aeb, aeb_tmp, &ai->alien, u.list) {
 		list_del(&aeb->u.list);
-		kfree(aeb);
+		ubi_free_aeb(ai, aeb);
 	}
 	list_for_each_entry_safe(aeb, aeb_tmp, &ai->erase, u.list) {
 		list_del(&aeb->u.list);
-		kfree(aeb);
+		ubi_free_aeb(ai, aeb);
 	}
 	list_for_each_entry_safe(aeb, aeb_tmp, &ai->corr, u.list) {
 		list_del(&aeb->u.list);
-		kfree(aeb);
+		ubi_free_aeb(ai, aeb);
 	}
 	list_for_each_entry_safe(aeb, aeb_tmp, &ai->free, u.list) {
 		list_del(&aeb->u.list);
-		kfree(aeb);
+		ubi_free_aeb(ai, aeb);
 	}
 	list_for_each_entry_safe(aeb, aeb_tmp, &ai->fastmap, u.list) {
 		list_del(&aeb->u.list);
-		kfree(aeb);
+		ubi_free_aeb(ai, aeb);
 	}
 
 	/* Destroy the volume RB-tree */
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 2bf626f8075e..d77592bb19ce 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -79,12 +79,10 @@ static int add_aeb(struct ubi_attach_info *ai, struct list_head *list,
 {
 	struct ubi_ainf_peb *aeb;
 
-	aeb = kzalloc(sizeof(*aeb), GFP_KERNEL);
+	aeb = ubi_alloc_aeb(ai, pnum, ec);
 	if (!aeb)
 		return -ENOMEM;
 
-	aeb->pnum = pnum;
-	aeb->ec = ec;
 	aeb->lnum = -1;
 	aeb->scrub = scrub;
 	aeb->copy_flag = aeb->sqnum = 0;
@@ -210,7 +208,8 @@ static int update_vol(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		 */
 		if (aeb->pnum == new_aeb->pnum) {
 			ubi_assert(aeb->lnum == new_aeb->lnum);
-			kfree(new_aeb);
+
+			ubi_free_aeb(ai, new_aeb);
 
 			return 0;
 		}
@@ -221,12 +220,10 @@ static int update_vol(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 		/* new_aeb is newer */
 		if (cmp_res & 1) {
-			victim = kzalloc(sizeof(*victim), GFP_KERNEL);
+			victim = ubi_alloc_aeb(ai, aeb->ec, aeb->pnum);
 			if (!victim)
 				return -ENOMEM;
 
-			victim->ec = aeb->ec;
-			victim->pnum = aeb->pnum;
 			list_add_tail(&victim->u.list, &ai->erase);
 
 			if (av->highest_lnum == be32_to_cpu(new_vh->lnum))
@@ -240,7 +237,8 @@ static int update_vol(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			aeb->pnum = new_aeb->pnum;
 			aeb->copy_flag = new_vh->copy_flag;
 			aeb->scrub = new_aeb->scrub;
-			kfree(new_aeb);
+			
+			ubi_free_aeb(ai, new_aeb);
 
 		/* new_aeb is older */
 		} else {
@@ -286,7 +284,7 @@ static int process_pool_aeb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	struct ubi_ainf_volume *av;
 
 	if (vol_id == UBI_FM_SB_VOLUME_ID || vol_id == UBI_FM_DATA_VOLUME_ID) {
-		kfree(new_aeb);
+		ubi_free_aeb(ai, new_aeb);
 
 		return 0;
 	}
@@ -295,7 +293,7 @@ static int process_pool_aeb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	av = ubi_find_av(ai, vol_id);
 	if (!av) {
 		ubi_err(ubi, "orphaned volume in fastmap pool!");
-		kfree(new_aeb);
+		ubi_free_aeb(ai, new_aeb);
 		return UBI_BAD_FASTMAP;
 	}
 
@@ -323,7 +321,8 @@ static void unmap_peb(struct ubi_attach_info *ai, int pnum)
 			if (aeb->pnum == pnum) {
 				rb_erase(&aeb->u.rb, &av->root);
 				av->leb_count--;
-				kfree(aeb);
+
+				ubi_free_aeb(ai, aeb);
 				return;
 			}
 		}
@@ -418,14 +417,12 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			if (err == UBI_IO_BITFLIPS)
 				scrub = 1;
 
-			new_aeb = kzalloc(sizeof(*new_aeb), GFP_KERNEL);
+			new_aeb = ubi_alloc_aeb(ai, pnum, be64_to_cpu(ech->ec));
 			if (!new_aeb) {
 				ret = -ENOMEM;
 				goto out;
 			}
 
-			new_aeb->ec = be64_to_cpu(ech->ec);
-			new_aeb->pnum = pnum;
 			new_aeb->lnum = be32_to_cpu(vh->lnum);
 			new_aeb->sqnum = be64_to_cpu(vh->sqnum);
 			new_aeb->copy_flag = vh->copy_flag;
@@ -732,11 +729,11 @@ fail_bad:
 fail:
 	list_for_each_entry_safe(tmp_aeb, _tmp_aeb, &used, u.list) {
 		list_del(&tmp_aeb->u.list);
-		free(tmp_aeb);
+		ubi_free_aeb(ai, tmp_aeb);
 	}
 	list_for_each_entry_safe(tmp_aeb, _tmp_aeb, &lfree, u.list) {
 		list_del(&tmp_aeb->u.list);
-		free(tmp_aeb);
+		ubi_free_aeb(ai, tmp_aeb);
 	}
 
 	return ret;
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index eb81298beff3..9539d5676306 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -727,6 +727,9 @@ extern struct class *ubi_class;
 extern struct blocking_notifier_head ubi_notifiers;
 
 /* attach.c */
+struct ubi_ainf_peb *ubi_alloc_aeb(struct ubi_attach_info *ai, int pnum,
+				   int ec);
+void ubi_free_aeb(struct ubi_attach_info *ai, struct ubi_ainf_peb *aeb);
 int ubi_add_to_av(struct ubi_device *ubi, struct ubi_attach_info *ai, int pnum,
 		  int ec, const struct ubi_vid_hdr *vid_hdr, int bitflips);
 struct ubi_ainf_volume *ubi_add_av(struct ubi_attach_info *ai, int vol_id);
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index b5eb4545efa1..ea30edef87bf 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -329,7 +329,7 @@ retry:
 	 * of this LEB as it will be deleted and freed in 'ubi_add_to_av()'.
 	 */
 	err = ubi_add_to_av(ubi, ai, new_aeb->pnum, new_aeb->ec, vid_hdr, 0);
-	kfree(new_aeb);
+	ubi_free_aeb(ai, new_aeb);
 	ubi_free_vid_hdr(ubi, vid_hdr);
 	return err;
 
@@ -343,6 +343,7 @@ write_error:
 		goto retry;
 	}
 	kfree(new_aeb);
+	ubi_free_aeb(ai, new_aeb);
 out_free:
 	ubi_free_vid_hdr(ubi, vid_hdr);
 	return err;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 27/56] UBI: move the global ech and vidh variables into struct ubi_attach_info

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Even if it works fine with those global variables, attaching the
temporary ech and vidh objects used during UBI scan to the
ubi_attach_info object sounds like a more future-proof option.

For example, attaching several UBI devices in parallel is prevented by
this use of global variable. And also because global variables should
be avoided in general.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c | 39 +++++++++++++++++++--------------------
 drivers/mtd/ubi/ubi.h    |  4 ++++
 2 files changed, 23 insertions(+), 20 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index 2b1271218804..36483b804fb1 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -85,10 +85,6 @@
 
 static int self_check_ai(struct ubi_device *ubi, struct ubi_attach_info *ai);
 
-/* Temporary variables used during scanning */
-static struct ubi_ec_hdr *ech;
-static struct ubi_vid_hdr *vidh;
-
 #define AV_FIND		BIT(0)
 #define AV_ADD		BIT(1)
 #define AV_FIND_OR_ADD	(AV_FIND | AV_ADD)
@@ -947,6 +943,8 @@ static bool vol_ignored(int vol_id)
 static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		    int pnum, bool fast)
 {
+	struct ubi_ec_hdr *ech = ai->ech;
+	struct ubi_vid_hdr *vidh = ai->vidh;
 	long long ec;
 	int err, bitflips = 0, vol_id = -1, ec_err = 0;
 
@@ -1382,12 +1380,12 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 	err = -ENOMEM;
 
-	ech = kzalloc(ubi->ec_hdr_alsize, GFP_KERNEL);
-	if (!ech)
+	ai->ech = kzalloc(ubi->ec_hdr_alsize, GFP_KERNEL);
+	if (!ai->ech)
 		return err;
 
-	vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!vidh)
+	ai->vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
+	if (!ai->vidh)
 		goto out_ech;
 
 	for (pnum = start; pnum < ubi->peb_count; pnum++) {
@@ -1434,15 +1432,15 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (err)
 		goto out_vidh;
 
-	ubi_free_vid_hdr(ubi, vidh);
-	kfree(ech);
+	ubi_free_vid_hdr(ubi, ai->vidh);
+	kfree(ai->ech);
 
 	return 0;
 
 out_vidh:
-	ubi_free_vid_hdr(ubi, vidh);
+	ubi_free_vid_hdr(ubi, ai->vidh);
 out_ech:
-	kfree(ech);
+	kfree(ai->ech);
 	return err;
 }
 
@@ -1487,12 +1485,12 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 	if (!scan_ai)
 		goto out;
 
-	ech = kzalloc(ubi->ec_hdr_alsize, GFP_KERNEL);
-	if (!ech)
+	scan_ai->ech = kzalloc(ubi->ec_hdr_alsize, GFP_KERNEL);
+	if (!scan_ai->ech)
 		goto out_ai;
 
-	vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!vidh)
+	scan_ai->vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
+	if (!scan_ai->vidh)
 		goto out_ech;
 
 	for (pnum = 0; pnum < UBI_FM_MAX_START; pnum++) {
@@ -1502,8 +1500,8 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 			goto out_vidh;
 	}
 
-	ubi_free_vid_hdr(ubi, vidh);
-	kfree(ech);
+	ubi_free_vid_hdr(ubi, scan_ai->vidh);
+	kfree(scan_ai->ech);
 
 	if (scan_ai->force_full_scan)
 		err = UBI_NO_FASTMAP;
@@ -1523,9 +1521,9 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 	return err;
 
 out_vidh:
-	ubi_free_vid_hdr(ubi, vidh);
+	ubi_free_vid_hdr(ubi, scan_ai->vidh);
 out_ech:
-	kfree(ech);
+	kfree(scan_ai->ech);
 out_ai:
 	destroy_ai(scan_ai);
 out:
@@ -1647,6 +1645,7 @@ out_ai:
  */
 static int self_check_ai(struct ubi_device *ubi, struct ubi_attach_info *ai)
 {
+	struct ubi_vid_hdr *vidh = ai->vidh;
 	int pnum, err, vols_found = 0;
 	struct rb_node *rb1, *rb2;
 	struct ubi_ainf_volume *av;
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 9539d5676306..d8e895ae47d7 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -660,6 +660,8 @@ struct ubi_ainf_volume {
  * @ec_sum: a temporary variable used when calculating @mean_ec
  * @ec_count: a temporary variable used when calculating @mean_ec
  * @aeb_slab_cache: slab cache for &struct ubi_ainf_peb objects
+ * @ech: temporary EC header. Only available during scan
+ * @vidh: temporary VID header. Only available during scan
  *
  * This data structure contains the result of attaching an MTD device and may
  * be used by other UBI sub-systems to build final UBI data structures, further
@@ -688,6 +690,8 @@ struct ubi_attach_info {
 	uint64_t ec_sum;
 	int ec_count;
 	struct kmem_cache *aeb_slab_cache;
+	struct ubi_ec_hdr *ech;
+	struct ubi_vid_hdr *vidh;
 };
 
 /**
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 28/56] UBI: simplify recover_peb() code

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

recover_peb() is using a convoluted retry/exit path. Add try_recover_peb()
to simplify the retry logic and make sure we have a single exit path
instead of manually releasing the resource in each error path.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 114 ++++++++++++++++++++++++++++++++------------------
 1 file changed, 74 insertions(+), 40 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 9decc0e1ecce..bc2dc2ef4371 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -465,38 +465,37 @@ out_unlock:
 }
 
 /**
- * recover_peb - recover from write failure.
- * @ubi: UBI device description object
+ * try_recover_peb - try to recover from write failure.
+ * @vol: volume description object
  * @pnum: the physical eraseblock to recover
- * @vol_id: volume ID
  * @lnum: logical eraseblock number
  * @buf: data which was not written because of the write failure
  * @offset: offset of the failed write
  * @len: how many bytes should have been written
+ * @vid: VID header
+ * @retry: whether the caller should retry in case of failure
  *
  * This function is called in case of a write failure and moves all good data
  * from the potentially bad physical eraseblock to a good physical eraseblock.
  * This function also writes the data which was not written due to the failure.
- * Returns new physical eraseblock number in case of success, and a negative
- * error code in case of failure.
+ * Returns 0 in case of success, and a negative error code in case of failure.
+ * In case of failure, the %retry parameter is set to false if this is a fatal
+ * error (retrying won't help), and true otherwise.
  */
-static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
-		       const void *buf, int offset, int len)
+static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
+			   const void *buf, int offset, int len,
+			   struct ubi_vid_hdr *vid_hdr, bool *retry)
 {
-	int err, idx = vol_id2idx(ubi, vol_id), new_pnum, data_size, tries = 0;
-	struct ubi_volume *vol = ubi->volumes[idx];
-	struct ubi_vid_hdr *vid_hdr;
+	struct ubi_device *ubi = vol->ubi;
+	int new_pnum, err, vol_id = vol->vol_id, data_size;
 	uint32_t crc;
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
-		return -ENOMEM;
+	*retry = false;
 
-retry:
 	new_pnum = ubi_wl_get_peb(ubi);
 	if (new_pnum < 0) {
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return new_pnum;
+		err = new_pnum;
+		goto out_put;
 	}
 
 	ubi_msg(ubi, "recover PEB %d, move data to PEB %d",
@@ -520,6 +519,8 @@ retry:
 			goto out_unlock;
 	}
 
+	*retry = true;
+
 	memcpy(ubi->peb_buf + offset, buf, len);
 
 	data_size = offset + len;
@@ -530,39 +531,72 @@ retry:
 	vid_hdr->data_crc = cpu_to_be32(crc);
 	err = ubi_io_write_vid_hdr(ubi, new_pnum, vid_hdr);
 	if (err)
-		goto write_error;
+		goto out_unlock;
 
 	err = ubi_io_write_data(ubi, ubi->peb_buf, new_pnum, 0, data_size);
-	if (err)
-		goto write_error;
 
-	ubi_free_vid_hdr(ubi, vid_hdr);
+out_unlock:
+	if (!err)
+		vol->eba_tbl[lnum] = new_pnum;
 
-	vol->eba_tbl[lnum] = new_pnum;
-	ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
+out_put:
 
-	ubi_msg(ubi, "data was successfully recovered");
-	return 0;
+	if (!err) {
+		ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
+		ubi_msg(ubi, "data was successfully recovered");
+	} else if (new_pnum >= 0) {
+		/*
+		 * Bad luck? This physical eraseblock is bad too? Crud. Let's
+		 * try to get another one.
+		 */
+		ubi_wl_put_peb(ubi, vol_id, lnum, new_pnum, 1);
+		ubi_warn(ubi, "failed to write to PEB %d", new_pnum);
+	}
 
-out_unlock:
-out_put:
-	ubi_wl_put_peb(ubi, vol_id, lnum, new_pnum, 1);
-	ubi_free_vid_hdr(ubi, vid_hdr);
 	return err;
+}
 
-write_error:
-	/*
-	 * Bad luck? This physical eraseblock is bad too? Crud. Let's try to
-	 * get another one.
-	 */
-	ubi_warn(ubi, "failed to write to PEB %d", new_pnum);
-	ubi_wl_put_peb(ubi, vol_id, lnum, new_pnum, 1);
-	if (++tries > UBI_IO_RETRIES) {
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
+/**
+ * recover_peb - recover from write failure.
+ * @ubi: UBI device description object
+ * @pnum: the physical eraseblock to recover
+ * @vol_id: volume ID
+ * @lnum: logical eraseblock number
+ * @buf: data which was not written because of the write failure
+ * @offset: offset of the failed write
+ * @len: how many bytes should have been written
+ *
+ * This function is called in case of a write failure and moves all good data
+ * from the potentially bad physical eraseblock to a good physical eraseblock.
+ * This function also writes the data which was not written due to the failure.
+ * Returns 0 in case of success, and a negative error code in case of failure.
+ * This function tries %UBI_IO_RETRIES before giving up.
+ */
+static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
+		       const void *buf, int offset, int len)
+{
+	int err, idx = vol_id2idx(ubi, vol_id), tries;
+	struct ubi_volume *vol = ubi->volumes[idx];
+	struct ubi_vid_hdr *vid_hdr;
+
+	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
+	if (!vid_hdr)
+		return -ENOMEM;
+
+	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
+		bool retry;
+
+		err = try_recover_peb(vol, pnum, lnum, buf, offset, len,
+				      vid_hdr, &retry);
+		if (!err || !retry)
+			break;
+
+		ubi_msg(ubi, "try again");
 	}
-	ubi_msg(ubi, "try again");
-	goto retry;
+
+	ubi_free_vid_hdr(ubi, vid_hdr);
+
+	return err;
 }
 
 /**
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 29/56] UBI: simplify LEB write and atomic LEB change code

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

ubi_eba_write_leb(), ubi_eba_write_leb_st() and
ubi_eba_atomic_leb_change() are using a convoluted retry/exit path.
Add the try_write_vid_and_data() function to simplify the retry logic
and make sure we have a single exit path instead of manually releasing
the resources in each error path.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 271 +++++++++++++++++++++-----------------------------
 1 file changed, 113 insertions(+), 158 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index bc2dc2ef4371..fd8368286825 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -600,6 +600,67 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
 }
 
 /**
+ * try_write_vid_and_data - try to write VID header and data to a new PEB.
+ * @vol: volume description object
+ * @lnum: logical eraseblock number
+ * @vid_hdr: VID header to write
+ * @buf: buffer containing the data
+ * @offset: where to start writing data
+ * @len: how many bytes should be written
+ *
+ * This function tries to write VID header and data belonging to logical
+ * eraseblock @lnum of volume @vol to a new physical eraseblock. Returns zero
+ * in case of success and a negative error code in case of failure.
+ * In case of error, it is possible that something was still written to the
+ * flash media, but may be some garbage.
+ */
+static int try_write_vid_and_data(struct ubi_volume *vol, int lnum,
+				  struct ubi_vid_hdr *vid_hdr, const void *buf,
+				  int offset, int len)
+{
+	struct ubi_device *ubi = vol->ubi;
+	int pnum, opnum, err, vol_id = vol->vol_id;
+
+	pnum = ubi_wl_get_peb(ubi);
+	if (pnum < 0) {
+		err = pnum;
+		goto out_put;
+	}
+
+	opnum = vol->eba_tbl[lnum];
+
+	dbg_eba("write VID hdr and %d bytes at offset %d of LEB %d:%d, PEB %d",
+		len, offset, vol_id, lnum, pnum);
+
+	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
+	if (err) {
+		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
+			 vol_id, lnum, pnum);
+		goto out_put;
+	}
+
+	if (len) {
+		err = ubi_io_write_data(ubi, buf, pnum, offset, len);
+		if (err) {
+			ubi_warn(ubi,
+				 "failed to write %d bytes at offset %d of LEB %d:%d, PEB %d",
+				 len, offset, vol_id, lnum, pnum);
+			goto out_put;
+		}
+	}
+
+	vol->eba_tbl[lnum] = pnum;
+
+out_put:
+	if (err && pnum >= 0)
+		err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
+	else if (!err && opnum >= 0)
+		err = ubi_wl_put_peb(ubi, vol_id, lnum, opnum, 0);
+
+	return err;
+}
+
+/**
  * ubi_eba_write_leb - write data to dynamic volume.
  * @ubi: UBI device description object
  * @vol: volume description object
@@ -612,11 +673,12 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
  * @vol. Returns zero in case of success and a negative error code in case
  * of failure. In case of error, it is possible that something was still
  * written to the flash media, but may be some garbage.
+ * This function retries %UBI_IO_RETRIES times before giving up.
  */
 int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		      const void *buf, int offset, int len)
 {
-	int err, pnum, tries = 0, vol_id = vol->vol_id;
+	int err, pnum, tries, vol_id = vol->vol_id;
 	struct ubi_vid_hdr *vid_hdr;
 
 	if (ubi->ro_mode)
@@ -637,11 +699,9 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 			if (err == -EIO && ubi->bad_allowed)
 				err = recover_peb(ubi, pnum, vol_id, lnum, buf,
 						  offset, len);
-			if (err)
-				ubi_ro_mode(ubi);
 		}
-		leb_write_unlock(ubi, vol_id, lnum);
-		return err;
+
+		goto out;
 	}
 
 	/*
@@ -661,63 +721,31 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	vid_hdr->compat = ubi_get_compat(ubi, vol_id);
 	vid_hdr->data_pad = cpu_to_be32(vol->data_pad);
 
-retry:
-	pnum = ubi_wl_get_peb(ubi);
-	if (pnum < 0) {
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		leb_write_unlock(ubi, vol_id, lnum);
-		return pnum;
-	}
-
-	dbg_eba("write VID hdr and %d bytes at offset %d of LEB %d:%d, PEB %d",
-		len, offset, vol_id, lnum, pnum);
-
-	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
-	if (err) {
-		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
-			 vol_id, lnum, pnum);
-		goto write_error;
-	}
+	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
+		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, offset,
+					     len);
+		if (err != -EIO || !ubi->bad_allowed)
+			break;
 
-	if (len) {
-		err = ubi_io_write_data(ubi, buf, pnum, offset, len);
-		if (err) {
-			ubi_warn(ubi, "failed to write %d bytes at offset %d of LEB %d:%d, PEB %d",
-				 len, offset, vol_id, lnum, pnum);
-			goto write_error;
-		}
+		/*
+		 * Fortunately, this is the first write operation to this
+		 * physical eraseblock, so just put it and request a new one.
+		 * We assume that if this physical eraseblock went bad, the
+		 * erase code will handle that.
+		 */
+		vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
+		ubi_msg(ubi, "try another PEB");
 	}
 
-	vol->eba_tbl[lnum] = pnum;
-
-	leb_write_unlock(ubi, vol_id, lnum);
 	ubi_free_vid_hdr(ubi, vid_hdr);
-	return 0;
 
-write_error:
-	if (err != -EIO || !ubi->bad_allowed) {
+out:
+	if (err)
 		ubi_ro_mode(ubi);
-		leb_write_unlock(ubi, vol_id, lnum);
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
-	}
 
-	/*
-	 * Fortunately, this is the first write operation to this physical
-	 * eraseblock, so just put it and request a new one. We assume that if
-	 * this physical eraseblock went bad, the erase code will handle that.
-	 */
-	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
-	if (err || ++tries > UBI_IO_RETRIES) {
-		ubi_ro_mode(ubi);
-		leb_write_unlock(ubi, vol_id, lnum);
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
-	}
+	leb_write_unlock(ubi, vol_id, lnum);
 
-	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
-	ubi_msg(ubi, "try another PEB");
-	goto retry;
+	return err;
 }
 
 /**
@@ -745,7 +773,7 @@ write_error:
 int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 			 int lnum, const void *buf, int len, int used_ebs)
 {
-	int err, pnum, tries = 0, data_size = len, vol_id = vol->vol_id;
+	int err, tries, data_size = len, vol_id = vol->vol_id;
 	struct ubi_vid_hdr *vid_hdr;
 	uint32_t crc;
 
@@ -763,10 +791,8 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 		return -ENOMEM;
 
 	err = leb_write_lock(ubi, vol_id, lnum);
-	if (err) {
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
-	}
+	if (err)
+		goto out;
 
 	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
 	vid_hdr->vol_id = cpu_to_be32(vol_id);
@@ -780,62 +806,26 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 	vid_hdr->used_ebs = cpu_to_be32(used_ebs);
 	vid_hdr->data_crc = cpu_to_be32(crc);
 
-retry:
-	pnum = ubi_wl_get_peb(ubi);
-	if (pnum < 0) {
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		leb_write_unlock(ubi, vol_id, lnum);
-		return pnum;
-	}
-
-	dbg_eba("write VID hdr and %d bytes at LEB %d:%d, PEB %d, used_ebs %d",
-		len, vol_id, lnum, pnum, used_ebs);
+	ubi_assert(vol->eba_tbl[lnum] < 0);
 
-	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
-	if (err) {
-		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
-			 vol_id, lnum, pnum);
-		goto write_error;
-	}
+	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
+		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, 0, len);
+		if (err != -EIO || !ubi->bad_allowed)
+			break;
 
-	err = ubi_io_write_data(ubi, buf, pnum, 0, len);
-	if (err) {
-		ubi_warn(ubi, "failed to write %d bytes of data to PEB %d",
-			 len, pnum);
-		goto write_error;
+		vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
+		ubi_msg(ubi, "try another PEB");
 	}
 
-	ubi_assert(vol->eba_tbl[lnum] < 0);
-	vol->eba_tbl[lnum] = pnum;
+	if (err)
+		ubi_ro_mode(ubi);
 
 	leb_write_unlock(ubi, vol_id, lnum);
-	ubi_free_vid_hdr(ubi, vid_hdr);
-	return 0;
-
-write_error:
-	if (err != -EIO || !ubi->bad_allowed) {
-		/*
-		 * This flash device does not admit of bad eraseblocks or
-		 * something nasty and unexpected happened. Switch to read-only
-		 * mode just in case.
-		 */
-		ubi_ro_mode(ubi);
-		leb_write_unlock(ubi, vol_id, lnum);
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
-	}
 
-	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
-	if (err || ++tries > UBI_IO_RETRIES) {
-		ubi_ro_mode(ubi);
-		leb_write_unlock(ubi, vol_id, lnum);
-		ubi_free_vid_hdr(ubi, vid_hdr);
-		return err;
-	}
+out:
+	ubi_free_vid_hdr(ubi, vid_hdr);
 
-	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
-	ubi_msg(ubi, "try another PEB");
-	goto retry;
+	return err;
 }
 
 /*
@@ -858,7 +848,7 @@ write_error:
 int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 			      int lnum, const void *buf, int len)
 {
-	int err, pnum, old_pnum, tries = 0, vol_id = vol->vol_id;
+	int err, tries, vol_id = vol->vol_id;
 	struct ubi_vid_hdr *vid_hdr;
 	uint32_t crc;
 
@@ -896,65 +886,30 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 	vid_hdr->copy_flag = 1;
 	vid_hdr->data_crc = cpu_to_be32(crc);
 
-retry:
-	pnum = ubi_wl_get_peb(ubi);
-	if (pnum < 0) {
-		err = pnum;
-		goto out_leb_unlock;
-	}
-
-	dbg_eba("change LEB %d:%d, PEB %d, write VID hdr to PEB %d",
-		vol_id, lnum, vol->eba_tbl[lnum], pnum);
+	dbg_eba("change LEB %d:%d", vol_id, lnum);
 
-	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
-	if (err) {
-		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
-			 vol_id, lnum, pnum);
-		goto write_error;
-	}
+	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
+		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, 0, len);
+		if (err != -EIO || !ubi->bad_allowed)
+			break;
 
-	err = ubi_io_write_data(ubi, buf, pnum, 0, len);
-	if (err) {
-		ubi_warn(ubi, "failed to write %d bytes of data to PEB %d",
-			 len, pnum);
-		goto write_error;
+		vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
+		ubi_msg(ubi, "try another PEB");
 	}
 
-	old_pnum = vol->eba_tbl[lnum];
-	vol->eba_tbl[lnum] = pnum;
-
-	if (old_pnum >= 0) {
-		err = ubi_wl_put_peb(ubi, vol_id, lnum, old_pnum, 0);
-		if (err)
-			goto out_leb_unlock;
-	}
+	/*
+	 * This flash device does not admit of bad eraseblocks or
+	 * something nasty and unexpected happened. Switch to read-only
+	 * mode just in case.
+	 */
+	if (err)
+		ubi_ro_mode(ubi);
 
-out_leb_unlock:
 	leb_write_unlock(ubi, vol_id, lnum);
+
 out_mutex:
 	ubi_free_vid_hdr(ubi, vid_hdr);
 	return err;
-
-write_error:
-	if (err != -EIO || !ubi->bad_allowed) {
-		/*
-		 * This flash device does not admit of bad eraseblocks or
-		 * something nasty and unexpected happened. Switch to read-only
-		 * mode just in case.
-		 */
-		ubi_ro_mode(ubi);
-		goto out_leb_unlock;
-	}
-
-	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
-	if (err || ++tries > UBI_IO_RETRIES) {
-		ubi_ro_mode(ubi);
-		goto out_leb_unlock;
-	}
-
-	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
-	ubi_msg(ubi, "try another PEB");
-	goto retry;
 }
 
 /**
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 30/56] UBI: add an helper to check lnum validity

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

ubi_leb_valid() is here to replace the
lnum < 0 || lnum >= vol->reserved_pebs checks.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflicts due missing cdev]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/kapi.c | 12 ++++++------
 drivers/mtd/ubi/ubi.h  |  5 +++++
 2 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index 9ec9962a5966..a7be941908fc 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -413,7 +413,7 @@ int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs || offset < 0 || len < 0 ||
+	if (!ubi_leb_valid(vol, lnum) || offset < 0 || len < 0 ||
 	    offset + len > vol->usable_leb_size ||
 	    offset & (ubi->min_io_size - 1) || len & (ubi->min_io_size - 1))
 		return -EINVAL;
@@ -458,7 +458,7 @@ int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs || len < 0 ||
+	if (!ubi_leb_valid(vol, lnum) || len < 0 ||
 	    len > vol->usable_leb_size || len & (ubi->min_io_size - 1))
 		return -EINVAL;
 
@@ -495,7 +495,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum)
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs)
+	if (!ubi_leb_valid(vol, lnum))
 		return -EINVAL;
 
 	if (vol->upd_marker)
@@ -555,7 +555,7 @@ int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum)
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs)
+	if (!ubi_leb_valid(vol, lnum))
 		return -EINVAL;
 
 	if (vol->upd_marker)
@@ -591,7 +591,7 @@ int ubi_leb_map(struct ubi_volume_desc *desc, int lnum)
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs)
+	if (!ubi_leb_valid(vol, lnum))
 		return -EINVAL;
 
 	if (vol->upd_marker)
@@ -626,7 +626,7 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
 
 	dbg_gen("test LEB %d:%d", vol->vol_id, lnum);
 
-	if (lnum < 0 || lnum >= vol->reserved_pebs)
+	if (!ubi_leb_valid(vol, lnum))
 		return -EINVAL;
 
 	if (vol->upd_marker)
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index d8e895ae47d7..4ef692fc7275 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -778,6 +778,11 @@ int ubi_check_volume(struct ubi_device *ubi, int vol_id);
 void ubi_update_reserved(struct ubi_device *ubi);
 void ubi_calculate_reserved(struct ubi_device *ubi);
 
+static inline bool ubi_leb_valid(struct ubi_volume *vol, int lnum)
+{
+	return lnum >= 0 && lnum < vol->reserved_pebs;
+}
+
 /* eba.c */
 int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
 		      int lnum);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 31/56] UBI: provide an helper to check whether a LEB is mapped or not

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

This is part of the process of hiding UBI EBA's internal to other part of
the UBI implementation, so that we can add new information to the EBA
table without having to patch different places in the UBI code.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c  | 12 ++++++++++++
 drivers/mtd/ubi/kapi.c |  4 ++--
 drivers/mtd/ubi/ubi.h  |  1 +
 3 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index fd8368286825..704f67dfb6a1 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -280,6 +280,18 @@ static void leb_write_unlock(struct ubi_device *ubi, int vol_id, int lnum)
 }
 
 /**
+ * ubi_eba_is_mapped - check if a LEB is mapped.
+ * @vol: volume description object
+ * @lnum: logical eraseblock number
+ *
+ * This function returns true if the LEB is mapped, false otherwise.
+ */
+bool ubi_eba_is_mapped(struct ubi_volume *vol, int lnum)
+{
+	return vol->eba_tbl[lnum] >= 0;
+}
+
+/**
  * ubi_eba_unmap_leb - un-map logical eraseblock.
  * @ubi: UBI device description object
  * @vol: volume description object
diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index a7be941908fc..e1cab763ebd2 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -597,7 +597,7 @@ int ubi_leb_map(struct ubi_volume_desc *desc, int lnum)
 	if (vol->upd_marker)
 		return -EBADF;
 
-	if (vol->eba_tbl[lnum] >= 0)
+	if (ubi_eba_is_mapped(vol, lnum))
 		return -EBADMSG;
 
 	return ubi_eba_write_leb(ubi, vol, lnum, NULL, 0, 0);
@@ -632,7 +632,7 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
 	if (vol->upd_marker)
 		return -EBADF;
 
-	return vol->eba_tbl[lnum] >= 0;
+	return ubi_eba_is_mapped(vol, lnum);
 }
 EXPORT_SYMBOL_GPL(ubi_is_mapped);
 
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 4ef692fc7275..5d39b0d55b01 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -784,6 +784,7 @@ static inline bool ubi_leb_valid(struct ubi_volume *vol, int lnum)
 }
 
 /* eba.c */
+bool ubi_eba_is_mapped(struct ubi_volume *vol, int lnum);
 int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
 		      int lnum);
 int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 32/56] UBI: provide an helper to query LEB information

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

This is part of our attempt to hide EBA internals from other part of the
implementation in order to easily adapt it to the MLC needs.

Here we are creating an ubi_eba_leb_desc struct to hide the way we keep
track of the LEB to PEB mapping.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c     | 17 +++++++++++++++++
 drivers/mtd/ubi/fastmap.c |  8 ++++++--
 drivers/mtd/ubi/ubi.h     | 17 +++++++++++++++++
 3 files changed, 40 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 704f67dfb6a1..11b85f4a6b03 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -76,6 +76,23 @@ static int ubi_get_compat(const struct ubi_device *ubi, int vol_id)
 }
 
 /**
+ * ubi_eba_get_ldesc - get information about a LEB
+ * @vol: volume description object
+ * @lnum: logical eraseblock number
+ * @ldesc: the LEB descriptor to fill
+ *
+ * Used to query information about a specific LEB.
+ * It is currently only returning the physical position of the LEB, but will be
+ * extended to provide more information.
+ */
+void ubi_eba_get_ldesc(struct ubi_volume *vol, int lnum,
+		       struct ubi_eba_leb_desc *ldesc)
+{
+	ldesc->lnum = lnum;
+	ldesc->pnum = vol->eba_tbl[lnum];
+}
+
+/**
  * ltree_lookup - look up the lock tree.
  * @ubi: UBI device description object
  * @vol_id: volume ID
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index d77592bb19ce..b319ae672cf2 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -1173,8 +1173,12 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 		fm_pos += sizeof(*feba) + (sizeof(__be32) * vol->reserved_pebs);
 		ubi_assert(fm_pos <= ubi->fm_size);
 
-		for (j = 0; j < vol->reserved_pebs; j++)
-			feba->pnum[j] = cpu_to_be32(vol->eba_tbl[j]);
+		for (j = 0; j < vol->reserved_pebs; j++) {
+			struct ubi_eba_leb_desc ldesc;
+
+			ubi_eba_get_ldesc(vol, j, &ldesc);
+			feba->pnum[j] = cpu_to_be32(ldesc.pnum);
+		}
 
 		feba->reserved_pebs = cpu_to_be32(j);
 		feba->magic = cpu_to_be32(UBI_FM_EBA_MAGIC);
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 5d39b0d55b01..0d2c5e8f2244 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -239,6 +239,21 @@ struct ubi_fm_pool {
 };
 
 /**
+ * struct ubi_eba_leb_desc - EBA logical eraseblock descriptor
+ * @lnum: the logical eraseblock number
+ * @pnum: the physical eraseblock where the LEB can be found
+ *
+ * This structure is here to hide EBA's internal from other part of the
+ * UBI implementation.
+ *
+ * One can query the position of a LEB by calling ubi_eba_get_ldesc().
+ */
+struct ubi_eba_leb_desc {
+	int lnum;
+	int pnum;
+};
+
+/**
  * struct ubi_volume - UBI volume description data structure.
  * @dev: device object to make use of the the Linux device model
  * @cdev: character device object to create character device
@@ -784,6 +799,8 @@ static inline bool ubi_leb_valid(struct ubi_volume *vol, int lnum)
 }
 
 /* eba.c */
+void ubi_eba_get_ldesc(struct ubi_volume *vol, int lnum,
+		       struct ubi_eba_leb_desc *ldesc);
 bool ubi_eba_is_mapped(struct ubi_volume *vol, int lnum);
 int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
 		      int lnum);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 33/56] UBI: hide EBA internals

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Create a private ubi_eba_table struct to hide EBA internals and provide
helpers to allocate, destroy, copy and assing an EBA table to a volume.

Now that external EBA users are using helpers to query/modify the EBA
state we can safely change the internal representation, which will be
needed to support the LEB consolidation concept.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/build.c |   2 +-
 drivers/mtd/ubi/eba.c   | 166 ++++++++++++++++++++++++++++++++++++++++--------
 drivers/mtd/ubi/ubi.h   |   8 ++-
 drivers/mtd/ubi/vmt.c   |  38 +++++------
 4 files changed, 164 insertions(+), 50 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index ae6181e9d162..13f39b39c8fb 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -218,7 +218,7 @@ void ubi_free_internal_volumes(struct ubi_device *ubi)
 
 	for (i = ubi->vtbl_slots;
 	     i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
-		kfree(ubi->volumes[i]->eba_tbl);
+		ubi_eba_replace_table(ubi->volumes[i], NULL);
 		kfree(ubi->volumes[i]);
 	}
 }
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 11b85f4a6b03..652474b4ba45 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -44,6 +44,30 @@
 #define EBA_RESERVED_PEBS 1
 
 /**
+ * struct ubi_eba_entry - structure encoding a single LEB -> PEB association
+ * @pnum: the physical eraseblock number attached to the LEB
+ *
+ * This structure is encoding a LEB -> PEB association. Note that the LEB
+ * number is not stored here, because it is the index used to access the
+ * entries table.
+ */
+struct ubi_eba_entry {
+	int pnum;
+};
+
+/**
+ * struct ubi_eba_table - LEB -> PEB association information
+ * @entries: the LEB to PEB mapping (one entry per LEB).
+ *
+ * This structure is private to the EBA logic and should be kept here.
+ * It is encoding the LEB to PEB association table, and is subject to
+ * changes.
+ */
+struct ubi_eba_table {
+	struct ubi_eba_entry *entries;
+};
+
+/**
  * next_sqnum - get next sequence number.
  * @ubi: UBI device description object
  *
@@ -89,7 +113,94 @@ void ubi_eba_get_ldesc(struct ubi_volume *vol, int lnum,
 		       struct ubi_eba_leb_desc *ldesc)
 {
 	ldesc->lnum = lnum;
-	ldesc->pnum = vol->eba_tbl[lnum];
+	ldesc->pnum = vol->eba_tbl->entries[lnum].pnum;
+}
+
+/**
+ * ubi_eba_create_table - allocate a new EBA table and initialize it with all
+ *			  LEBs unmapped
+ * @vol: volume containing the EBA table to copy
+ * @nentries: number of entries in the table
+ *
+ * Allocate a new EBA table and initialize it with all LEBs unmapped.
+ * Returns a valid pointer if it succeed, an ERR_PTR() otherwise.
+ */
+struct ubi_eba_table *ubi_eba_create_table(struct ubi_volume *vol,
+					   int nentries)
+{
+	struct ubi_eba_table *tbl;
+	int err = -ENOMEM;
+	int i;
+
+	tbl = kzalloc(sizeof(*tbl), GFP_KERNEL);
+	if (!tbl)
+		return ERR_PTR(-ENOMEM);
+
+	tbl->entries = kzalloc(nentries * sizeof(*tbl->entries),
+				     GFP_KERNEL);
+	if (!tbl->entries)
+		goto err;
+
+	for (i = 0; i < nentries; i++)
+		tbl->entries[i].pnum = UBI_LEB_UNMAPPED;
+
+	return tbl;
+
+err:
+	kfree(tbl->entries);
+	kfree(tbl);
+
+	return ERR_PTR(err);
+}
+
+/**
+ * ubi_eba_destroy_table - destroy an EBA table
+ * @tbl: the table to destroy
+ *
+ * Destroy an EBA table.
+ */
+void ubi_eba_destroy_table(struct ubi_eba_table *tbl)
+{
+	if (!tbl)
+		return;
+
+	kfree(tbl->entries);
+	kfree(tbl);
+}
+
+/**
+ * ubi_eba_copy_table - copy the EBA table attached to vol into another table
+ * @vol: volume containing the EBA table to copy
+ * @dst: destination
+ * @nentries: number of entries to copy
+ *
+ * Copy the EBA table stored in vol into the one pointed by dst.
+ */
+void ubi_eba_copy_table(struct ubi_volume *vol, struct ubi_eba_table *dst,
+			int nentries)
+{
+	struct ubi_eba_table *src;
+	int i;
+
+	ubi_assert(dst && vol && vol->eba_tbl);
+
+	src = vol->eba_tbl;
+
+	for (i = 0; i < nentries; i++)
+		dst->entries[i].pnum = src->entries[i].pnum;
+}
+
+/**
+ * ubi_eba_replace_table - assign a new EBA table to a volume
+ * @vol: volume containing the EBA table to copy
+ * @tbl: new EBA table
+ *
+ * Assign a new EBA table to the volume and release the old one.
+ */
+void ubi_eba_replace_table(struct ubi_volume *vol, struct ubi_eba_table *tbl)
+{
+	ubi_eba_destroy_table(vol->eba_tbl);
+	vol->eba_tbl = tbl;
 }
 
 /**
@@ -305,7 +416,7 @@ static void leb_write_unlock(struct ubi_device *ubi, int vol_id, int lnum)
  */
 bool ubi_eba_is_mapped(struct ubi_volume *vol, int lnum)
 {
-	return vol->eba_tbl[lnum] >= 0;
+	return vol->eba_tbl->entries[lnum].pnum >= 0;
 }
 
 /**
@@ -330,14 +441,14 @@ int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
 	if (err)
 		return err;
 
-	pnum = vol->eba_tbl[lnum];
+	pnum = vol->eba_tbl->entries[lnum].pnum;
 	if (pnum < 0)
 		/* This logical eraseblock is already unmapped */
 		goto out_unlock;
 
 	dbg_eba("erase LEB %d:%d, PEB %d", vol_id, lnum, pnum);
 
-	vol->eba_tbl[lnum] = UBI_LEB_UNMAPPED;
+	vol->eba_tbl->entries[lnum].pnum = UBI_LEB_UNMAPPED;
 	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 0);
 
 out_unlock:
@@ -375,7 +486,7 @@ int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	if (err)
 		return err;
 
-	pnum = vol->eba_tbl[lnum];
+	pnum = vol->eba_tbl->entries[lnum].pnum;
 	if (pnum < 0) {
 		/*
 		 * The logical eraseblock is not mapped, fill the whole buffer
@@ -566,7 +677,7 @@ static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
 
 out_unlock:
 	if (!err)
-		vol->eba_tbl[lnum] = new_pnum;
+		vol->eba_tbl->entries[lnum].pnum = new_pnum;
 
 out_put:
 
@@ -656,7 +767,7 @@ static int try_write_vid_and_data(struct ubi_volume *vol, int lnum,
 		goto out_put;
 	}
 
-	opnum = vol->eba_tbl[lnum];
+	opnum = vol->eba_tbl->entries[lnum].pnum;
 
 	dbg_eba("write VID hdr and %d bytes at offset %d of LEB %d:%d, PEB %d",
 		len, offset, vol_id, lnum, pnum);
@@ -678,7 +789,7 @@ static int try_write_vid_and_data(struct ubi_volume *vol, int lnum,
 		}
 	}
 
-	vol->eba_tbl[lnum] = pnum;
+	vol->eba_tbl->entries[lnum].pnum = pnum;
 
 out_put:
 	if (err && pnum >= 0)
@@ -717,7 +828,7 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	if (err)
 		return err;
 
-	pnum = vol->eba_tbl[lnum];
+	pnum = vol->eba_tbl->entries[lnum].pnum;
 	if (pnum >= 0) {
 		dbg_eba("write %d bytes at offset %d of LEB %d:%d, PEB %d",
 			len, offset, vol_id, lnum, pnum);
@@ -835,7 +946,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 	vid_hdr->used_ebs = cpu_to_be32(used_ebs);
 	vid_hdr->data_crc = cpu_to_be32(crc);
 
-	ubi_assert(vol->eba_tbl[lnum] < 0);
+	ubi_assert(vol->eba_tbl->entries[lnum].pnum < 0);
 
 	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
 		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, 0, len);
@@ -1041,9 +1152,9 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 	 * probably waiting on @ubi->move_mutex. No need to continue the work,
 	 * cancel it.
 	 */
-	if (vol->eba_tbl[lnum] != from) {
+	if (vol->eba_tbl->entries[lnum].pnum != from) {
 		dbg_wl("LEB %d:%d is no longer mapped to PEB %d, mapped to PEB %d, cancel",
-		       vol_id, lnum, from, vol->eba_tbl[lnum]);
+		       vol_id, lnum, from, vol->eba_tbl->entries[lnum].pnum);
 		err = MOVE_CANCEL_RACE;
 		goto out_unlock_leb;
 	}
@@ -1121,8 +1232,8 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 		}
 	}
 
-	ubi_assert(vol->eba_tbl[lnum] == from);
-	vol->eba_tbl[lnum] = to;
+	ubi_assert(vol->eba_tbl->entries[lnum].pnum == from);
+	vol->eba_tbl->entries[lnum].pnum = to;
 
 out_unlock_buf:
 out_unlock_leb:
@@ -1277,7 +1388,7 @@ out_free:
  */
 int ubi_eba_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 {
-	int i, j, err, num_volumes;
+	int i, err, num_volumes;
 	struct ubi_ainf_volume *av;
 	struct ubi_volume *vol;
 	struct ubi_ainf_peb *aeb;
@@ -1291,33 +1402,37 @@ int ubi_eba_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 	num_volumes = ubi->vtbl_slots + UBI_INT_VOL_COUNT;
 
 	for (i = 0; i < num_volumes; i++) {
+		struct ubi_eba_table *tbl;
+
 		vol = ubi->volumes[i];
 		if (!vol)
 			continue;
 
-		vol->eba_tbl = kmalloc(vol->reserved_pebs * sizeof(int),
-				       GFP_KERNEL);
-		if (!vol->eba_tbl) {
-			err = -ENOMEM;
+		tbl = ubi_eba_create_table(vol, vol->reserved_pebs);
+		if (IS_ERR(tbl)) {
+			err = PTR_ERR(tbl);
 			goto out_free;
 		}
 
-		for (j = 0; j < vol->reserved_pebs; j++)
-			vol->eba_tbl[j] = UBI_LEB_UNMAPPED;
+		ubi_eba_replace_table(vol, tbl);
 
 		av = ubi_find_av(ai, idx2vol_id(ubi, i));
 		if (!av)
 			continue;
 
 		ubi_rb_for_each_entry(rb, aeb, &av->root, u.rb) {
-			if (aeb->lnum >= vol->reserved_pebs)
+			if (aeb->lnum >= vol->reserved_pebs) {
 				/*
 				 * This may happen in case of an unclean reboot
 				 * during re-size.
 				 */
 				ubi_move_aeb_to_list(av, aeb, &ai->erase);
-			else
-				vol->eba_tbl[aeb->lnum] = aeb->pnum;
+			} else {
+				struct ubi_eba_entry *entry;
+
+				entry = &vol->eba_tbl->entries[aeb->lnum];
+				entry->pnum = aeb->pnum;
+			}
 		}
 	}
 
@@ -1354,8 +1469,7 @@ out_free:
 	for (i = 0; i < num_volumes; i++) {
 		if (!ubi->volumes[i])
 			continue;
-		kfree(ubi->volumes[i]->eba_tbl);
-		ubi->volumes[i]->eba_tbl = NULL;
+		ubi_eba_replace_table(ubi->volumes[i], NULL);
 	}
 	return err;
 }
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 0d2c5e8f2244..c61b2d9d0a28 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -329,7 +329,7 @@ struct ubi_volume {
 	long long upd_received;
 	void *upd_buf;
 
-	int *eba_tbl;
+	struct ubi_eba_table *eba_tbl;
 	unsigned int checked:1;
 	unsigned int corrupted:1;
 	unsigned int upd_marker:1;
@@ -799,6 +799,12 @@ static inline bool ubi_leb_valid(struct ubi_volume *vol, int lnum)
 }
 
 /* eba.c */
+struct ubi_eba_table *ubi_eba_create_table(struct ubi_volume *vol,
+					   int nentries);
+void ubi_eba_destroy_table(struct ubi_eba_table *tbl);
+void ubi_eba_copy_table(struct ubi_volume *vol, struct ubi_eba_table *dst,
+			int nentries);
+void ubi_eba_replace_table(struct ubi_volume *vol, struct ubi_eba_table *tbl);
 void ubi_eba_get_ldesc(struct ubi_volume *vol, int lnum,
 		       struct ubi_eba_leb_desc *ldesc);
 bool ubi_eba_is_mapped(struct ubi_volume *vol, int lnum);
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 05a61fccefc6..61770482149e 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -41,6 +41,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	int i, err, vol_id = req->vol_id, do_free = 1;
 	struct ubi_volume *vol;
 	struct ubi_vtbl_record vtbl_rec;
+	struct ubi_eba_table *eba_tbl = NULL;
 
 	if (ubi->ro_mode)
 		return -EROFS;
@@ -121,14 +122,13 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	if (err)
 		goto out_acc;
 
-	vol->eba_tbl = kmalloc(vol->reserved_pebs * sizeof(int), GFP_KERNEL);
-	if (!vol->eba_tbl) {
-		err = -ENOMEM;
+	eba_tbl = ubi_eba_create_table(vol, vol->reserved_pebs);
+	if (IS_ERR(eba_tbl)) {
+		err = PTR_ERR(eba_tbl);
 		goto out_acc;
 	}
 
-	for (i = 0; i < vol->reserved_pebs; i++)
-		vol->eba_tbl[i] = UBI_LEB_UNMAPPED;
+	ubi_eba_replace_table(vol, eba_tbl);
 
 	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
 		vol->used_ebs = vol->reserved_pebs;
@@ -194,7 +194,7 @@ out_sysfs:
 	do_free = 0;
 out_mapping:
 	if (do_free)
-		kfree(vol->eba_tbl);
+		ubi_eba_destroy_table(eba_tbl);
 	unregister_device(&vol->dev);
 out_acc:
 	ubi->rsvd_pebs -= vol->reserved_pebs;
@@ -284,10 +284,11 @@ out_unlock:
  */
 int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 {
-	int i, err, pebs, *new_mapping;
+	int i, err, pebs;
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
 	struct ubi_vtbl_record vtbl_rec;
+	struct ubi_eba_table *new_eba_tbl = NULL;
 	int vol_id = vol->vol_id;
 
 	if (ubi->ro_mode)
@@ -307,12 +308,9 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 	if (reserved_pebs == vol->reserved_pebs)
 		return 0;
 
-	new_mapping = kmalloc(reserved_pebs * sizeof(int), GFP_KERNEL);
-	if (!new_mapping)
-		return -ENOMEM;
-
-	for (i = 0; i < reserved_pebs; i++)
-		new_mapping[i] = UBI_LEB_UNMAPPED;
+	new_eba_tbl = ubi_eba_create_table(vol, reserved_pebs);
+	if (IS_ERR(new_eba_tbl))
+		return PTR_ERR(new_eba_tbl);
 
 	if (vol->ref_count > 1) {
 		err = -EBUSY;
@@ -333,10 +331,8 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 		}
 		ubi->avail_pebs -= pebs;
 		ubi->rsvd_pebs += pebs;
-		for (i = 0; i < vol->reserved_pebs; i++)
-			new_mapping[i] = vol->eba_tbl[i];
-		kfree(vol->eba_tbl);
-		vol->eba_tbl = new_mapping;
+		ubi_eba_copy_table(vol, new_eba_tbl, vol->reserved_pebs);
+		ubi_eba_replace_table(vol, new_eba_tbl);
 	}
 
 	if (pebs < 0) {
@@ -348,10 +344,8 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 		ubi->rsvd_pebs += pebs;
 		ubi->avail_pebs -= pebs;
 		ubi_update_reserved(ubi);
-		for (i = 0; i < reserved_pebs; i++)
-			new_mapping[i] = vol->eba_tbl[i];
-		kfree(vol->eba_tbl);
-		vol->eba_tbl = new_mapping;
+		ubi_eba_copy_table(vol, new_eba_tbl, reserved_pebs);
+		ubi_eba_replace_table(vol, new_eba_tbl);
 	}
 
 	/*
@@ -390,7 +384,7 @@ out_acc:
 		ubi->avail_pebs += pebs;
 	}
 out_free:
-	kfree(new_mapping);
+	kfree(new_eba_tbl);
 	return err;
 }
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 34/56] UBI: introduce the VID buffer concept

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Currently, all VID headers are allocated and freed using the
ubi_zalloc_vid_hdr() and ubi_free_vid_hdr() function. These functions
make sure to align allocation on ubi->vid_hdr_alsize and adjust the
vid_hdr pointer to match the ubi->vid_hdr_shift requirements.
This works fine, but is a bit convoluted.
Moreover, the future introduction of LEB consolidation (needed to support
MLC/TLC NANDs) will allows a VID buffer to contain more than one VID
header.

Hence the creation of a ubi_vid_io_buf struct to attach extra information
to the VID header.

We currently only store the actual pointer of the underlying buffer, but
will soon add the number of VID headers contained in the buffer.

Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/attach.c  | 40 +++++++++----------
 drivers/mtd/ubi/eba.c     | 81 ++++++++++++++++++++++-----------------
 drivers/mtd/ubi/fastmap.c | 71 ++++++++++++++++++++--------------
 drivers/mtd/ubi/io.c      | 39 ++++++++++---------
 drivers/mtd/ubi/ubi.h     | 97 ++++++++++++++++++++++++++++++++---------------
 drivers/mtd/ubi/vtbl.c    | 13 ++++---
 drivers/mtd/ubi/wl.c      | 19 ++++++----
 7 files changed, 218 insertions(+), 142 deletions(-)

diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c
index 36483b804fb1..eee0e55ba9ce 100644
--- a/drivers/mtd/ubi/attach.c
+++ b/drivers/mtd/ubi/attach.c
@@ -447,7 +447,7 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 {
 	int len, err, second_is_newer, bitflips = 0, corrupted = 0;
 	uint32_t data_crc, crc;
-	struct ubi_vid_hdr *vh = NULL;
+	struct ubi_vid_io_buf *vidb = NULL;
 	unsigned long long sqnum2 = be64_to_cpu(vid_hdr->sqnum);
 
 	if (sqnum2 == aeb->sqnum) {
@@ -490,12 +490,12 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 			return bitflips << 1;
 		}
 
-		vh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-		if (!vh)
+		vidb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+		if (!vidb)
 			return -ENOMEM;
 
 		pnum = aeb->pnum;
-		err = ubi_io_read_vid_hdr(ubi, pnum, vh, 0);
+		err = ubi_io_read_vid_hdr(ubi, pnum, vidb, 0);
 		if (err) {
 			if (err == UBI_IO_BITFLIPS)
 				bitflips = 1;
@@ -509,7 +509,7 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 			}
 		}
 
-		vid_hdr = vh;
+		vid_hdr = ubi_get_vid_hdr(vidb);
 	}
 
 	/* Read the data of the copy and check the CRC */
@@ -533,7 +533,7 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 		bitflips |= !!err;
 	}
 
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vidb);
 
 	if (second_is_newer)
 		dbg_bld("second PEB %d is newer, copy_flag is set", pnum);
@@ -544,7 +544,7 @@ int ubi_compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb,
 
 out_unlock:
 out_free_vidh:
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vidb);
 	return err;
 }
 
@@ -944,7 +944,8 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		    int pnum, bool fast)
 {
 	struct ubi_ec_hdr *ech = ai->ech;
-	struct ubi_vid_hdr *vidh = ai->vidh;
+	struct ubi_vid_io_buf *vidb = ai->vidb;
+	struct ubi_vid_hdr *vidh = ubi_get_vid_hdr(vidb);
 	long long ec;
 	int err, bitflips = 0, vol_id = -1, ec_err = 0;
 
@@ -1042,7 +1043,7 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 	/* OK, we've done with the EC header, let's look at the VID header */
 
-	err = ubi_io_read_vid_hdr(ubi, pnum, vidh, 0);
+	err = ubi_io_read_vid_hdr(ubi, pnum, vidb, 0);
 	if (err < 0)
 		return err;
 	switch (err) {
@@ -1384,8 +1385,8 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (!ai->ech)
 		return err;
 
-	ai->vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!ai->vidh)
+	ai->vidb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+	if (!ai->vidb)
 		goto out_ech;
 
 	for (pnum = start; pnum < ubi->peb_count; pnum++) {
@@ -1432,13 +1433,13 @@ static int scan_all(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (err)
 		goto out_vidh;
 
-	ubi_free_vid_hdr(ubi, ai->vidh);
+	ubi_free_vid_buf(ai->vidb);
 	kfree(ai->ech);
 
 	return 0;
 
 out_vidh:
-	ubi_free_vid_hdr(ubi, ai->vidh);
+	ubi_free_vid_buf(ai->vidb);
 out_ech:
 	kfree(ai->ech);
 	return err;
@@ -1489,8 +1490,8 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 	if (!scan_ai->ech)
 		goto out_ai;
 
-	scan_ai->vidh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!scan_ai->vidh)
+	scan_ai->vidb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+	if (!scan_ai->vidb)
 		goto out_ech;
 
 	for (pnum = 0; pnum < UBI_FM_MAX_START; pnum++) {
@@ -1500,7 +1501,7 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 			goto out_vidh;
 	}
 
-	ubi_free_vid_hdr(ubi, scan_ai->vidh);
+	ubi_free_vid_buf(scan_ai->vidb);
 	kfree(scan_ai->ech);
 
 	if (scan_ai->force_full_scan)
@@ -1521,7 +1522,7 @@ static int scan_fast(struct ubi_device *ubi, struct ubi_attach_info **ai)
 	return err;
 
 out_vidh:
-	ubi_free_vid_hdr(ubi, scan_ai->vidh);
+	ubi_free_vid_buf(scan_ai->vidb);
 out_ech:
 	kfree(scan_ai->ech);
 out_ai:
@@ -1645,7 +1646,8 @@ out_ai:
  */
 static int self_check_ai(struct ubi_device *ubi, struct ubi_attach_info *ai)
 {
-	struct ubi_vid_hdr *vidh = ai->vidh;
+	struct ubi_vid_io_buf *vidb = ai->vidb;
+	struct ubi_vid_hdr *vidh = ubi_get_vid_hdr(vidb);
 	int pnum, err, vols_found = 0;
 	struct rb_node *rb1, *rb2;
 	struct ubi_ainf_volume *av;
@@ -1775,7 +1777,7 @@ static int self_check_ai(struct ubi_device *ubi, struct ubi_attach_info *ai)
 
 			last_aeb = aeb;
 
-			err = ubi_io_read_vid_hdr(ubi, aeb->pnum, vidh, 1);
+			err = ubi_io_read_vid_hdr(ubi, aeb->pnum, vidb, 1);
 			if (err && err != UBI_IO_BITFLIPS) {
 				ubi_err(ubi, "VID header is not OK (%d)",
 					err);
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 652474b4ba45..7c0fdefa2576 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -479,6 +479,7 @@ int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		     void *buf, int offset, int len, int check)
 {
 	int err, pnum, scrub = 0, vol_id = vol->vol_id;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	uint32_t uninitialized_var(crc);
 
@@ -509,13 +510,15 @@ int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 
 retry:
 	if (check) {
-		vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-		if (!vid_hdr) {
+		vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+		if (!vidb) {
 			err = -ENOMEM;
 			goto out_unlock;
 		}
 
-		err = ubi_io_read_vid_hdr(ubi, pnum, vid_hdr, 1);
+		vid_hdr = ubi_get_vid_hdr(vidb);
+
+		err = ubi_io_read_vid_hdr(ubi, pnum, vidb, 1);
 		if (err && err != UBI_IO_BITFLIPS) {
 			if (err > 0) {
 				/*
@@ -561,7 +564,7 @@ retry:
 		ubi_assert(len == be32_to_cpu(vid_hdr->data_size));
 
 		crc = be32_to_cpu(vid_hdr->data_crc);
-		ubi_free_vid_hdr(ubi, vid_hdr);
+		ubi_free_vid_buf(vidb);
 	}
 
 	err = ubi_io_read_data(ubi, buf, pnum, offset, len);
@@ -598,7 +601,7 @@ retry:
 	return err;
 
 out_free:
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 out_unlock:
 	leb_read_unlock(ubi, vol_id, lnum);
 	return err;
@@ -612,7 +615,7 @@ out_unlock:
  * @buf: data which was not written because of the write failure
  * @offset: offset of the failed write
  * @len: how many bytes should have been written
- * @vid: VID header
+ * @vidb: VID buffer
  * @retry: whether the caller should retry in case of failure
  *
  * This function is called in case of a write failure and moves all good data
@@ -624,9 +627,10 @@ out_unlock:
  */
 static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
 			   const void *buf, int offset, int len,
-			   struct ubi_vid_hdr *vid_hdr, bool *retry)
+			   struct ubi_vid_io_buf *vidb, bool *retry)
 {
 	struct ubi_device *ubi = vol->ubi;
+	struct ubi_vid_hdr *vid_hdr;
 	int new_pnum, err, vol_id = vol->vol_id, data_size;
 	uint32_t crc;
 
@@ -641,7 +645,7 @@ static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
 	ubi_msg(ubi, "recover PEB %d, move data to PEB %d",
 		pnum, new_pnum);
 
-	err = ubi_io_read_vid_hdr(ubi, pnum, vid_hdr, 1);
+	err = ubi_io_read_vid_hdr(ubi, pnum, vidb, 1);
 	if (err && err != UBI_IO_BITFLIPS) {
 		if (err > 0)
 			err = -EIO;
@@ -669,7 +673,7 @@ static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
 	vid_hdr->copy_flag = 1;
 	vid_hdr->data_size = cpu_to_be32(data_size);
 	vid_hdr->data_crc = cpu_to_be32(crc);
-	err = ubi_io_write_vid_hdr(ubi, new_pnum, vid_hdr);
+	err = ubi_io_write_vid_hdr(ubi, new_pnum, vidb);
 	if (err)
 		goto out_unlock;
 
@@ -717,24 +721,24 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
 {
 	int err, idx = vol_id2idx(ubi, vol_id), tries;
 	struct ubi_volume *vol = ubi->volumes[idx];
-	struct ubi_vid_hdr *vid_hdr;
+	struct ubi_vid_io_buf *vidb;
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
 		return -ENOMEM;
 
 	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
 		bool retry;
 
-		err = try_recover_peb(vol, pnum, lnum, buf, offset, len,
-				      vid_hdr, &retry);
+		err = try_recover_peb(vol, pnum, lnum, buf, offset, len, vidb,
+				      &retry);
 		if (!err || !retry)
 			break;
 
 		ubi_msg(ubi, "try again");
 	}
 
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 
 	return err;
 }
@@ -743,7 +747,7 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
  * try_write_vid_and_data - try to write VID header and data to a new PEB.
  * @vol: volume description object
  * @lnum: logical eraseblock number
- * @vid_hdr: VID header to write
+ * @vidb: the VID buffer to write
  * @buf: buffer containing the data
  * @offset: where to start writing data
  * @len: how many bytes should be written
@@ -755,7 +759,7 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
  * flash media, but may be some garbage.
  */
 static int try_write_vid_and_data(struct ubi_volume *vol, int lnum,
-				  struct ubi_vid_hdr *vid_hdr, const void *buf,
+				  struct ubi_vid_io_buf *vidb, const void *buf,
 				  int offset, int len)
 {
 	struct ubi_device *ubi = vol->ubi;
@@ -772,7 +776,7 @@ static int try_write_vid_and_data(struct ubi_volume *vol, int lnum,
 	dbg_eba("write VID hdr and %d bytes at offset %d of LEB %d:%d, PEB %d",
 		len, offset, vol_id, lnum, pnum);
 
-	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
+	err = ubi_io_write_vid_hdr(ubi, pnum, vidb);
 	if (err) {
 		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
 			 vol_id, lnum, pnum);
@@ -819,6 +823,7 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		      const void *buf, int offset, int len)
 {
 	int err, pnum, tries, vol_id = vol->vol_id;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 
 	if (ubi->ro_mode)
@@ -848,12 +853,14 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	 * The logical eraseblock is not mapped. We have to get a free physical
 	 * eraseblock and write the volume identifier header there first.
 	 */
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr) {
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb) {
 		leb_write_unlock(ubi, vol_id, lnum);
 		return -ENOMEM;
 	}
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
+
 	vid_hdr->vol_type = UBI_VID_DYNAMIC;
 	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
 	vid_hdr->vol_id = cpu_to_be32(vol_id);
@@ -862,8 +869,7 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	vid_hdr->data_pad = cpu_to_be32(vol->data_pad);
 
 	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
-		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, offset,
-					     len);
+		err = try_write_vid_and_data(vol, lnum, vidb, buf, offset, len);
 		if (err != -EIO || !ubi->bad_allowed)
 			break;
 
@@ -877,7 +883,7 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		ubi_msg(ubi, "try another PEB");
 	}
 
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 
 out:
 	if (err)
@@ -914,6 +920,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 			 int lnum, const void *buf, int len, int used_ebs)
 {
 	int err, tries, data_size = len, vol_id = vol->vol_id;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	uint32_t crc;
 
@@ -926,10 +933,12 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 	else
 		ubi_assert(!(len & (ubi->min_io_size - 1)));
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
 		return -ENOMEM;
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
+
 	err = leb_write_lock(ubi, vol_id, lnum);
 	if (err)
 		goto out;
@@ -949,7 +958,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 	ubi_assert(vol->eba_tbl->entries[lnum].pnum < 0);
 
 	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
-		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, 0, len);
+		err = try_write_vid_and_data(vol, lnum, vidb, buf, 0, len);
 		if (err != -EIO || !ubi->bad_allowed)
 			break;
 
@@ -963,7 +972,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 	leb_write_unlock(ubi, vol_id, lnum);
 
 out:
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 
 	return err;
 }
@@ -989,6 +998,7 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 			      int lnum, const void *buf, int len)
 {
 	int err, tries, vol_id = vol->vol_id;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	uint32_t crc;
 
@@ -1006,10 +1016,12 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 		return ubi_eba_write_leb(ubi, vol, lnum, NULL, 0, 0);
 	}
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
 		return -ENOMEM;
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
+
 	err = leb_write_lock(ubi, vol_id, lnum);
 	if (err)
 		goto out_mutex;
@@ -1029,7 +1041,7 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 	dbg_eba("change LEB %d:%d", vol_id, lnum);
 
 	for (tries = 0; tries <= UBI_IO_RETRIES; tries++) {
-		err = try_write_vid_and_data(vol, lnum, vid_hdr, buf, 0, len);
+		err = try_write_vid_and_data(vol, lnum, vidb, buf, 0, len);
 		if (err != -EIO || !ubi->bad_allowed)
 			break;
 
@@ -1048,7 +1060,7 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 	leb_write_unlock(ubi, vol_id, lnum);
 
 out_mutex:
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	return err;
 }
 
@@ -1094,9 +1106,10 @@ static int is_error_sane(int err)
  *   o a negative error code in case of failure.
  */
 int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
-		     struct ubi_vid_hdr *vid_hdr)
+		     struct ubi_vid_io_buf *vidb)
 {
 	int err, vol_id, lnum, data_size, aldata_size, idx;
+	struct ubi_vid_hdr *vid_hdr = ubi_get_vid_hdr(vidb);
 	struct ubi_volume *vol;
 	uint32_t crc;
 
@@ -1203,7 +1216,7 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 	}
 	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
 
-	err = ubi_io_write_vid_hdr(ubi, to, vid_hdr);
+	err = ubi_io_write_vid_hdr(ubi, to, vidb);
 	if (err) {
 		if (err == -EIO)
 			err = MOVE_TARGET_WR_ERR;
@@ -1211,7 +1224,7 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 	}
 
 	/* Read the VID header back and check if it was written correctly */
-	err = ubi_io_read_vid_hdr(ubi, to, vid_hdr, 1);
+	err = ubi_io_read_vid_hdr(ubi, to, vidb, 1);
 	if (err) {
 		if (err != UBI_IO_BITFLIPS) {
 			ubi_warn(ubi, "error %d while reading VID header back from PEB %d",
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index b319ae672cf2..a0b5fe784a6f 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -44,21 +44,23 @@ size_t ubi_calc_fm_size(struct ubi_device *ubi)
  * Returns a new struct ubi_vid_hdr on success.
  * NULL indicates out of memory.
  */
-static struct ubi_vid_hdr *new_fm_vhdr(struct ubi_device *ubi, int vol_id)
+static struct ubi_vid_io_buf *new_fm_vbuf(struct ubi_device *ubi, int vol_id)
 {
-	struct ubi_vid_hdr *new;
+	struct ubi_vid_io_buf *new;
+	struct ubi_vid_hdr *vh;
 
-	new = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
+	new = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
 	if (!new)
 		goto out;
 
-	new->vol_type = UBI_VID_DYNAMIC;
-	new->vol_id = cpu_to_be32(vol_id);
+	vh = ubi_get_vid_hdr(new);
+	vh->vol_type = UBI_VID_DYNAMIC;
+	vh->vol_id = cpu_to_be32(vol_id);
 
 	/* UBI implementations without fastmap support have to delete the
 	 * fastmap.
 	 */
-	new->compat = UBI_COMPAT_DELETE;
+	vh->compat = UBI_COMPAT_DELETE;
 
 out:
 	return new;
@@ -345,6 +347,7 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		     __be32 *pebs, int pool_size, unsigned long long *max_sqnum,
 		     struct list_head *lfree)
 {
+	struct ubi_vid_io_buf *vb;
 	struct ubi_vid_hdr *vh;
 	struct ubi_ec_hdr *ech;
 	struct ubi_ainf_peb *new_aeb;
@@ -354,12 +357,14 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (!ech)
 		return -ENOMEM;
 
-	vh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!vh) {
+	vb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+	if (!vb) {
 		kfree(ech);
 		return -ENOMEM;
 	}
 
+	vh = ubi_get_vid_hdr(vb);
+
 	dbg_bld("scanning fastmap pool: size = %i", pool_size);
 
 	/*
@@ -400,7 +405,7 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			goto out;
 		}
 
-		err = ubi_io_read_vid_hdr(ubi, pnum, vh, 0);
+		err = ubi_io_read_vid_hdr(ubi, pnum, vb, 0);
 		if (err == UBI_IO_FF || err == UBI_IO_FF_BITFLIPS) {
 			unsigned long long ec = be64_to_cpu(ech->ec);
 			unmap_peb(ai, pnum);
@@ -446,7 +451,7 @@ static int scan_pool(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	}
 
 out:
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vb);
 	kfree(ech);
 	return ret;
 }
@@ -774,6 +779,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		     struct ubi_attach_info *scan_ai)
 {
 	struct ubi_fm_sb *fmsb, *fmsb2;
+	struct ubi_vid_io_buf *vb;
 	struct ubi_vid_hdr *vh;
 	struct ubi_ec_hdr *ech;
 	struct ubi_fastmap_layout *fm;
@@ -848,12 +854,14 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		goto free_fm_sb;
 	}
 
-	vh = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!vh) {
+	vb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+	if (!vb) {
 		ret = -ENOMEM;
 		goto free_hdr;
 	}
 
+	vh = ubi_get_vid_hdr(vb);
+
 	for (i = 0; i < used_blocks; i++) {
 		int image_seq;
 
@@ -896,7 +904,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			goto free_hdr;
 		}
 
-		ret = ubi_io_read_vid_hdr(ubi, pnum, vh, 0);
+		ret = ubi_io_read_vid_hdr(ubi, pnum, vb, 0);
 		if (ret && ret != UBI_IO_BITFLIPS) {
 			ubi_err(ubi, "unable to read fastmap block# %i (PEB: %i)",
 				i, pnum);
@@ -986,7 +994,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	ubi->fm_disabled = 0;
 	ubi->fast_attach = 1;
 
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vb);
 	kfree(ech);
 out:
 	if (ret == UBI_BAD_FASTMAP)
@@ -994,7 +1002,7 @@ out:
 	return ret;
 
 free_hdr:
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vb);
 	kfree(ech);
 free_fm_sb:
 	kfree(fmsb);
@@ -1022,6 +1030,7 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 	struct ubi_fm_eba *feba;
 	struct ubi_wl_entry *wl_e;
 	struct ubi_volume *vol;
+	struct ubi_vid_io_buf *avbuf, *dvbuf;
 	struct ubi_vid_hdr *avhdr, *dvhdr;
 	struct ubi_work *ubi_wrk;
 	struct rb_node *tmp_rb;
@@ -1031,18 +1040,21 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 	fm_raw = ubi->fm_buf;
 	memset(ubi->fm_buf, 0, ubi->fm_size);
 
-	avhdr = new_fm_vhdr(ubi, UBI_FM_SB_VOLUME_ID);
-	if (!avhdr) {
+	avbuf = new_fm_vbuf(ubi, UBI_FM_SB_VOLUME_ID);
+	if (!avbuf) {
 		ret = -ENOMEM;
 		goto out;
 	}
 
-	dvhdr = new_fm_vhdr(ubi, UBI_FM_DATA_VOLUME_ID);
-	if (!dvhdr) {
+	dvbuf = new_fm_vbuf(ubi, UBI_FM_DATA_VOLUME_ID);
+	if (!dvbuf) {
 		ret = -ENOMEM;
 		goto out_kfree;
 	}
 
+	avhdr = ubi_get_vid_hdr(avbuf);
+	dvhdr = ubi_get_vid_hdr(dvbuf);
+
 	fmsb = (struct ubi_fm_sb *)fm_raw;
 	fm_pos += sizeof(*fmsb);
 	ubi_assert(fm_pos <= ubi->fm_size);
@@ -1190,7 +1202,7 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 	avhdr->lnum = 0;
 
 	dbg_bld("writing fastmap SB to PEB %i", new_fm->e[0]->pnum);
-	ret = ubi_io_write_vid_hdr(ubi, new_fm->e[0]->pnum, avhdr);
+	ret = ubi_io_write_vid_hdr(ubi, new_fm->e[0]->pnum, avbuf);
 	if (ret) {
 		ubi_err(ubi, "unable to write vid_hdr to fastmap SB!");
 		goto out_kfree;
@@ -1210,7 +1222,7 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 		dvhdr->lnum = cpu_to_be32(i);
 		dbg_bld("writing fastmap data to PEB %i sqnum %llu",
 			new_fm->e[i]->pnum, be64_to_cpu(dvhdr->sqnum));
-		ret = ubi_io_write_vid_hdr(ubi, new_fm->e[i]->pnum, dvhdr);
+		ret = ubi_io_write_vid_hdr(ubi, new_fm->e[i]->pnum, dvbuf);
 		if (ret) {
 			ubi_err(ubi, "unable to write vid_hdr to PEB %i!",
 				new_fm->e[i]->pnum);
@@ -1234,8 +1246,8 @@ static int ubi_write_fastmap(struct ubi_device *ubi,
 	dbg_bld("fastmap written!");
 
 out_kfree:
-	ubi_free_vid_hdr(ubi, avhdr);
-	ubi_free_vid_hdr(ubi, dvhdr);
+	ubi_free_vid_buf(avbuf);
+	ubi_free_vid_buf(dvbuf);
 out:
 	return ret;
 }
@@ -1304,7 +1316,8 @@ static int invalidate_fastmap(struct ubi_device *ubi)
 	int ret;
 	struct ubi_fastmap_layout *fm;
 	struct ubi_wl_entry *e;
-	struct ubi_vid_hdr *vh = NULL;
+	struct ubi_vid_io_buf *vb = NULL;
+	struct ubi_vid_hdr *vh;
 
 	if (!ubi->fm)
 		return 0;
@@ -1316,10 +1329,12 @@ static int invalidate_fastmap(struct ubi_device *ubi)
 	if (!fm)
 		goto out;
 
-	vh = new_fm_vhdr(ubi, UBI_FM_SB_VOLUME_ID);
-	if (!vh)
+	vb = new_fm_vbuf(ubi, UBI_FM_SB_VOLUME_ID);
+	if (!vb)
 		goto out_free_fm;
 
+	vh = ubi_get_vid_hdr(vb);
+
 	ret = -ENOSPC;
 	e = ubi_wl_get_fm_peb(ubi, 1);
 	if (!e)
@@ -1330,7 +1345,7 @@ static int invalidate_fastmap(struct ubi_device *ubi)
 	 * to scanning mode.
 	 */
 	vh->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
-	ret = ubi_io_write_vid_hdr(ubi, e->pnum, vh);
+	ret = ubi_io_write_vid_hdr(ubi, e->pnum, vb);
 	if (ret < 0) {
 		ubi_wl_put_fm_peb(ubi, e, 0, 0);
 		goto out_free_fm;
@@ -1342,7 +1357,7 @@ static int invalidate_fastmap(struct ubi_device *ubi)
 	ubi->fm = fm;
 
 out:
-	ubi_free_vid_hdr(ubi, vh);
+	ubi_free_vid_buf(vb);
 	return ret;
 
 out_free_fm:
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 6d08f92ea647..4c2731c4172a 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -226,6 +226,7 @@ static int nor_erase_prepare(struct ubi_device *ubi, int pnum)
 	loff_t addr;
 	uint32_t data = 0;
 	struct ubi_ec_hdr ec_hdr;
+	struct ubi_vid_io_buf vidb;
 
 	/*
 	 * Note, we cannot generally define VID header buffers on stack,
@@ -252,7 +253,10 @@ static int nor_erase_prepare(struct ubi_device *ubi, int pnum)
 			goto error;
 	}
 
-	err = ubi_io_read_vid_hdr(ubi, pnum, &vid_hdr, 0);
+	ubi_init_vid_buf(ubi, &vidb, &vid_hdr);
+	ubi_assert(&vid_hdr == ubi_get_vid_hdr(&vidb));
+
+	err = ubi_io_read_vid_hdr(ubi, pnum, &vidb, 0);
 	if (err != UBI_IO_BAD_HDR_EBADMSG && err != UBI_IO_BAD_HDR &&
 	    err != UBI_IO_FF){
 		addr += ubi->vid_hdr_aloffset;
@@ -687,12 +691,11 @@ bad:
  * ubi_io_read_vid_hdr - read and check a volume identifier header.
  * @ubi: UBI device description object
  * @pnum: physical eraseblock number to read from
- * @vid_hdr: &struct ubi_vid_hdr object where to store the read volume
- * identifier header
+ * @vidb: the volume identifier buffer to store data in
  * @verbose: be verbose if the header is corrupted or wasn't found
  *
  * This function reads the volume identifier header from physical eraseblock
- * @pnum and stores it in @vid_hdr. It also checks CRC checksum of the read
+ * @pnum and stores it in @vidb. It also checks CRC checksum of the read
  * volume identifier header. The error codes are the same as in
  * 'ubi_io_read_ec_hdr()'.
  *
@@ -700,16 +703,16 @@ bad:
  * 'ubi_io_read_ec_hdr()', so refer commentaries in 'ubi_io_read_ec_hdr()'.
  */
 int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum,
-			struct ubi_vid_hdr *vid_hdr, int verbose)
+			struct ubi_vid_io_buf *vidb, int verbose)
 {
 	int err, read_err;
 	uint32_t crc, magic, hdr_crc;
-	void *p;
+	struct ubi_vid_hdr *vid_hdr = ubi_get_vid_hdr(vidb);
+	void *p = vidb->buffer;
 
 	dbg_io("read VID header from PEB %d", pnum);
 	ubi_assert(pnum >= 0 &&  pnum < ubi->peb_count);
 
-	p = (char *)vid_hdr - ubi->vid_hdr_shift;
 	read_err = ubi_io_read(ubi, p, pnum, ubi->vid_hdr_aloffset,
 			  ubi->vid_hdr_shift + UBI_VID_HDR_SIZE);
 	if (read_err && read_err != UBI_IO_BITFLIPS && !mtd_is_eccerr(read_err))
@@ -772,23 +775,24 @@ int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum,
  * ubi_io_write_vid_hdr - write a volume identifier header.
  * @ubi: UBI device description object
  * @pnum: the physical eraseblock number to write to
- * @vid_hdr: the volume identifier header to write
+ * @vidb: the volume identifier buffer to write
  *
  * This function writes the volume identifier header described by @vid_hdr to
  * physical eraseblock @pnum. This function automatically fills the
- * @vid_hdr->magic and the @vid_hdr->version fields, as well as calculates
- * header CRC checksum and stores it at vid_hdr->hdr_crc.
+ * @vidb->hdr->magic and the @vidb->hdr->version fields, as well as calculates
+ * header CRC checksum and stores it at vidb->hdr->hdr_crc.
  *
  * This function returns zero in case of success and a negative error code in
  * case of failure. If %-EIO is returned, the physical eraseblock probably went
  * bad.
  */
 int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum,
-			 struct ubi_vid_hdr *vid_hdr)
+			 struct ubi_vid_io_buf *vidb)
 {
+	struct ubi_vid_hdr *vid_hdr = ubi_get_vid_hdr(vidb);
 	int err;
 	uint32_t crc;
-	void *p;
+	void *p = vidb->buffer;
 
 	dbg_io("write VID header to PEB %d", pnum);
 	ubi_assert(pnum >= 0 &&  pnum < ubi->peb_count);
@@ -806,7 +810,6 @@ int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum,
 	if (err)
 		return err;
 
-	p = (char *)vid_hdr - ubi->vid_hdr_shift;
 	err = ubi_io_write(ubi, p, pnum, ubi->vid_hdr_aloffset,
 			   ubi->vid_hdr_alsize);
 	return err;
@@ -972,17 +975,19 @@ static int self_check_peb_vid_hdr(const struct ubi_device *ubi, int pnum)
 {
 	int err;
 	uint32_t crc, hdr_crc;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	void *p;
 
 	if (!ubi_dbg_chk_io(ubi))
 		return 0;
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
 		return -ENOMEM;
 
-	p = (char *)vid_hdr - ubi->vid_hdr_shift;
+	vid_hdr = ubi_get_vid_hdr(vidb);
+	p = vidb->buffer;
 	err = ubi_io_read(ubi, p, pnum, ubi->vid_hdr_aloffset,
 			  ubi->vid_hdr_alsize);
 	if (err && err != UBI_IO_BITFLIPS && !mtd_is_eccerr(err))
@@ -1003,7 +1008,7 @@ static int self_check_peb_vid_hdr(const struct ubi_device *ubi, int pnum)
 	err = self_check_vid_hdr(ubi, pnum, vid_hdr);
 
 exit:
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	return err;
 }
 
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index c61b2d9d0a28..4f1ba745ccc9 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -140,6 +140,17 @@ enum {
 };
 
 /**
+ * struct ubi_vid_io_buf - VID buffer used to read/write VID info to/from the
+ *			   flash.
+ * @hdr: a pointer to the VID header stored in buffer
+ * @buffer: underlying buffer
+ */
+struct ubi_vid_io_buf {
+	struct ubi_vid_hdr *hdr;
+	void *buffer;
+};
+
+/**
  * struct ubi_wl_entry - wear-leveling entry.
  * @u.rb: link in the corresponding (free/used) RB-tree
  * @u.list: link in the protection queue
@@ -676,7 +687,7 @@ struct ubi_ainf_volume {
  * @ec_count: a temporary variable used when calculating @mean_ec
  * @aeb_slab_cache: slab cache for &struct ubi_ainf_peb objects
  * @ech: temporary EC header. Only available during scan
- * @vidh: temporary VID header. Only available during scan
+ * @vidh: temporary VID buffer. Only available during scan
  *
  * This data structure contains the result of attaching an MTD device and may
  * be used by other UBI sub-systems to build final UBI data structures, further
@@ -706,7 +717,7 @@ struct ubi_attach_info {
 	int ec_count;
 	struct kmem_cache *aeb_slab_cache;
 	struct ubi_ec_hdr *ech;
-	struct ubi_vid_hdr *vidh;
+	struct ubi_vid_io_buf *vidb;
 };
 
 /**
@@ -819,7 +830,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 			      int lnum, const void *buf, int len);
 int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
-		     struct ubi_vid_hdr *vid_hdr);
+		     struct ubi_vid_io_buf *vidb);
 int ubi_eba_init(struct ubi_device *ubi, struct ubi_attach_info *ai);
 unsigned long long ubi_next_sqnum(struct ubi_device *ubi);
 int self_check_eba(struct ubi_device *ubi, struct ubi_attach_info *ai_fastmap,
@@ -853,9 +864,9 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum,
 int ubi_io_write_ec_hdr(struct ubi_device *ubi, int pnum,
 			struct ubi_ec_hdr *ec_hdr);
 int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum,
-			struct ubi_vid_hdr *vid_hdr, int verbose);
+			struct ubi_vid_io_buf *vidb, int verbose);
 int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum,
-			 struct ubi_vid_hdr *vid_hdr);
+			 struct ubi_vid_io_buf *vidb);
 
 /* build.c */
 int ubi_detach_mtd_dev(int ubi_num, int anyway);
@@ -958,44 +969,68 @@ static inline void ubi_move_aeb_to_list(struct ubi_ainf_volume *av,
 }
 
 /**
- * ubi_zalloc_vid_hdr - allocate a volume identifier header object.
- * @ubi: UBI device description object
- * @gfp_flags: GFP flags to allocate with
- *
- * This function returns a pointer to the newly allocated and zero-filled
- * volume identifier header object in case of success and %NULL in case of
- * failure.
+ * ubi_init_vid_buf - Initialize a VID buffer
+ * @ubi: the UBI device
+ * @vidb: the VID buffer to initialize
+ * @buf: the underlying buffer
+ */
+static inline void ubi_init_vid_buf(const struct ubi_device *ubi,
+				    struct ubi_vid_io_buf *vidb,
+				    void *buf)
+{
+	if (buf)
+		memset(buf, 0, ubi->vid_hdr_alsize);
+
+	vidb->buffer = buf;
+	vidb->hdr = buf + ubi->vid_hdr_shift;
+}
+
+/**
+ * ubi_init_vid_buf - Allocate a VID buffer
+ * @ubi: the UBI device
+ * @gfp_flags: GFP flags to use for the allocation
  */
-static inline struct ubi_vid_hdr *
-ubi_zalloc_vid_hdr(const struct ubi_device *ubi, gfp_t gfp_flags)
+static inline struct ubi_vid_io_buf *
+ubi_alloc_vid_buf(const struct ubi_device *ubi, gfp_t gfp_flags)
 {
-	void *vid_hdr;
+	struct ubi_vid_io_buf *vidb;
+	void *buf;
 
-	vid_hdr = kzalloc(ubi->vid_hdr_alsize, gfp_flags);
-	if (!vid_hdr)
+	vidb = kzalloc(sizeof(*vidb), gfp_flags);
+	if (!vidb)
 		return NULL;
 
-	/*
-	 * VID headers may be stored at un-aligned flash offsets, so we shift
-	 * the pointer.
-	 */
-	return vid_hdr + ubi->vid_hdr_shift;
+	buf = kmalloc(ubi->vid_hdr_alsize, gfp_flags);
+	if (!buf) {
+		kfree(vidb);
+		return NULL;
+	}
+
+	ubi_init_vid_buf(ubi, vidb, buf);
+
+	return vidb;
 }
 
 /**
- * ubi_free_vid_hdr - free a volume identifier header object.
- * @ubi: UBI device description object
- * @vid_hdr: the object to free
+ * ubi_free_vid_buf - Free a VID buffer
+ * @vidb: the VID buffer to free
  */
-static inline void ubi_free_vid_hdr(const struct ubi_device *ubi,
-				    struct ubi_vid_hdr *vid_hdr)
+static inline void ubi_free_vid_buf(struct ubi_vid_io_buf *vidb)
 {
-	void *p = vid_hdr;
-
-	if (!p)
+	if (!vidb)
 		return;
 
-	kfree(p - ubi->vid_hdr_shift);
+	kfree(vidb->buffer);
+	kfree(vidb);
+}
+
+/**
+ * ubi_get_vid_hdr - Get the VID header attached to a VID buffer
+ * @vidb: VID buffer
+ */
+static inline struct ubi_vid_hdr *ubi_get_vid_hdr(struct ubi_vid_io_buf *vidb)
+{
+	return vidb->hdr;
 }
 
 /*
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index ea30edef87bf..41af1f4d9204 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -290,15 +290,18 @@ static int create_vtbl(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		       int copy, void *vtbl)
 {
 	int err, tries = 0;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	struct ubi_ainf_peb *new_aeb;
 
 	dbg_gen("create volume table (copy #%d)", copy + 1);
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_KERNEL);
+	if (!vidb)
 		return -ENOMEM;
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
+
 retry:
 	new_aeb = ubi_early_get_peb(ubi, ai);
 	if (IS_ERR(new_aeb)) {
@@ -315,7 +318,7 @@ retry:
 	vid_hdr->sqnum = cpu_to_be64(++ai->max_sqnum);
 
 	/* The EC header is already there, write the VID header */
-	err = ubi_io_write_vid_hdr(ubi, new_aeb->pnum, vid_hdr);
+	err = ubi_io_write_vid_hdr(ubi, new_aeb->pnum, vidb);
 	if (err)
 		goto write_error;
 
@@ -330,7 +333,7 @@ retry:
 	 */
 	err = ubi_add_to_av(ubi, ai, new_aeb->pnum, new_aeb->ec, vid_hdr, 0);
 	ubi_free_aeb(ai, new_aeb);
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	return err;
 
 write_error:
@@ -345,7 +348,7 @@ write_error:
 	kfree(new_aeb);
 	ubi_free_aeb(ai, new_aeb);
 out_free:
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	return err;
 
 }
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index eef9c17c2d0f..32cea29e8827 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -623,6 +623,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	int anchor = wrk->anchor;
 #endif
 	struct ubi_wl_entry *e1, *e2;
+	struct ubi_vid_io_buf *vidb;
 	struct ubi_vid_hdr *vid_hdr;
 	int dst_leb_clean = 0;
 
@@ -630,10 +631,12 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	if (shutdown)
 		return 0;
 
-	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr)
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
 		return -ENOMEM;
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
+
 	ubi_assert(!ubi->move_from && !ubi->move_to);
 	ubi_assert(!ubi->move_to_put);
 
@@ -724,7 +727,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	 * which is being moved was unmapped.
 	 */
 
-	err = ubi_io_read_vid_hdr(ubi, e1->pnum, vid_hdr, 0);
+	err = ubi_io_read_vid_hdr(ubi, e1->pnum, vidb, 0);
 	if (err && err != UBI_IO_BITFLIPS) {
 		dst_leb_clean = 1;
 		if (err == UBI_IO_FF) {
@@ -761,7 +764,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	vol_id = be32_to_cpu(vid_hdr->vol_id);
 	lnum = be32_to_cpu(vid_hdr->lnum);
 
-	err = ubi_eba_copy_leb(ubi, e1->pnum, e2->pnum, vid_hdr);
+	err = ubi_eba_copy_leb(ubi, e1->pnum, e2->pnum, vidb);
 	if (err) {
 		if (err == MOVE_CANCEL_RACE) {
 			/*
@@ -818,7 +821,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	if (scrubbing)
 		ubi_msg(ubi, "scrubbed PEB %d (LEB %d:%d), data moved to PEB %d",
 			e1->pnum, vol_id, lnum, e2->pnum);
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 
 	if (!ubi->move_to_put) {
 		wl_tree_add(e2, &ubi->used);
@@ -879,7 +882,7 @@ out_not_moved:
 	ubi->move_from = ubi->move_to = NULL;
 	ubi->wl_scheduled = 0;
 
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	if (dst_leb_clean) {
 		ensure_wear_leveling(ubi, 1);
 	} else {
@@ -900,7 +903,7 @@ out_error:
 	ubi->move_from = ubi->move_to = NULL;
 	ubi->move_to_put = ubi->wl_scheduled = 0;
 
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	wl_entry_destroy(ubi, e1);
 	wl_entry_destroy(ubi, e2);
 
@@ -911,7 +914,7 @@ out_ro:
 
 out_cancel:
 	ubi->wl_scheduled = 0;
-	ubi_free_vid_hdr(ubi, vid_hdr);
+	ubi_free_vid_buf(vidb);
 	return 0;
 }
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 35/56] ubi: Deal with interrupted erasures in WL

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

When Fastmap is used we can face here an -EBADMSG
since Fastmap cannot know about unmaps.
If the erasure was interrupted the PEB may show ECC
errors and UBI would go to ro-mode as it assumes
that the PEB was check during attach time, which is
not the case with Fastmap.

Cc: <stable@vger.kernel.org>
Fixes: dbb7d2a88d ("UBI: Add fastmap core")
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/wl.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 32cea29e8827..764e8dddc2b5 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -618,7 +618,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 				int shutdown)
 {
 	int err, scrubbing = 0, torture = 0, protect = 0, erroneous = 0;
-	int vol_id = -1, lnum = -1;
+	int erase = 0, keep = 0, vol_id = -1, lnum = -1;
 #ifdef CONFIG_MTD_UBI_FASTMAP
 	int anchor = wrk->anchor;
 #endif
@@ -754,6 +754,16 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 			       e1->pnum);
 			scrubbing = 1;
 			goto out_not_moved;
+		} else if (ubi->fast_attach && err == UBI_IO_BAD_HDR_EBADMSG) {
+			/*
+			 * While a full scan would detect interrupted erasures
+			 * at attach time we can face them here when attached from
+			 * Fastmap.
+			 */
+			dbg_wl("PEB %d has ECC errors, maybe from an interrupted erasure",
+			       e1->pnum);
+			erase = 1;
+			goto out_not_moved;
 		}
 
 		ubi_err(ubi, "error %d while reading VID header from PEB %d",
@@ -789,6 +799,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 			 * Target PEB had bit-flips or write error - torture it.
 			 */
 			torture = 1;
+			keep = 1;
 			goto out_not_moved;
 		}
 
@@ -871,7 +882,7 @@ out_not_moved:
 		ubi->erroneous_peb_count += 1;
 	} else if (scrubbing)
 		wl_tree_add(e1, &ubi->scrub);
-	else
+	else if (keep)
 		wl_tree_add(e1, &ubi->used);
 	if (dst_leb_clean) {
 		wl_tree_add(e2, &ubi->free);
@@ -891,6 +902,12 @@ out_not_moved:
 			goto out_ro;
 	}
 
+	if (erase) {
+		err = do_sync_erase(ubi, e1, vol_id, lnum, 1);
+		if (err)
+			goto out_ro;
+	}
+
 	return 0;
 
 out_error:
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 36/56] ubi: Fix races around ubi_refill_pools()

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

When writing a new Fastmap the first thing that happens
is refilling the pools in memory.
At this stage it is possible that new PEBs from the new pools
get already claimed and written with data.
If this happens before the new Fastmap data structure hits the
flash and we face power cut the freshly written PEB will not
scanned and unnoticed.

Solve the issue by locking the pools until Fastmap is written.

Cc: <stable@vger.kernel.org>
Fixes: dbb7d2a88d ("UBI: Add fastmap core")
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap-wl.c |  4 ++--
 drivers/mtd/ubi/wl.c         | 15 +++++++++------
 2 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap-wl.c b/drivers/mtd/ubi/fastmap-wl.c
index 1885b155ddf9..07982aab169a 100644
--- a/drivers/mtd/ubi/fastmap-wl.c
+++ b/drivers/mtd/ubi/fastmap-wl.c
@@ -230,7 +230,7 @@ int ubi_ensure_anchor_pebs(struct ubi_device *ubi)
 
 	wrk->anchor = 1;
 	wrk->func = &wear_leveling_worker;
-	schedule_ubi_work(ubi, wrk);
+	__schedule_ubi_work(ubi, wrk);
 	return 0;
 }
 
@@ -268,7 +268,7 @@ int ubi_wl_put_fm_peb(struct ubi_device *ubi, struct ubi_wl_entry *fm_e,
 	}
 
 	vol_id = lnum ? UBI_FM_DATA_VOLUME_ID : UBI_FM_SB_VOLUME_ID;
-	return schedule_erase(ubi, e, vol_id, lnum, torture);
+	return schedule_erase(ubi, e, vol_id, lnum, torture, true);
 }
 
 /**
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 764e8dddc2b5..54c119d20b07 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -554,7 +554,7 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk,
  * failure.
  */
 static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e,
-			  int vol_id, int lnum, int torture)
+			  int vol_id, int lnum, int torture, bool nested)
 {
 	struct ubi_work *wl_wrk;
 
@@ -573,7 +573,10 @@ static int schedule_erase(struct ubi_device *ubi, struct ubi_wl_entry *e,
 	wl_wrk->lnum = lnum;
 	wl_wrk->torture = torture;
 
-	schedule_ubi_work(ubi, wl_wrk);
+	if (nested)
+		__schedule_ubi_work(ubi, wl_wrk);
+	else
+		schedule_ubi_work(ubi, wl_wrk);
 	return 0;
 }
 
@@ -1047,7 +1050,7 @@ static int __erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk)
 		int err1;
 
 		/* Re-schedule the LEB for erasure */
-		err1 = schedule_erase(ubi, e, vol_id, lnum, 0);
+		err1 = schedule_erase(ubi, e, vol_id, lnum, 0, false);
 		if (err1) {
 			wl_entry_destroy(ubi, e);
 			err = err1;
@@ -1204,7 +1207,7 @@ retry:
 		}
 	}
 
-	err = schedule_erase(ubi, e, vol_id, lnum, torture);
+	err = schedule_erase(ubi, e, vol_id, lnum, torture, false);
 	if (err)
 		wl_tree_add(e, &ubi->used);
 
@@ -1428,7 +1431,7 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 		e->pnum = aeb->pnum;
 		e->ec = aeb->ec;
 		ubi->lookuptbl[e->pnum] = e;
-		if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0)) {
+		if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0, false)) {
 			wl_entry_destroy(ubi, e);
 			goto out_free;
 		}
@@ -1501,7 +1504,7 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 			e->ec = aeb->ec;
 			ubi_assert(!ubi->lookuptbl[e->pnum]);
 			ubi->lookuptbl[e->pnum] = e;
-			if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0)) {
+			if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0, false)) {
 				wl_entry_destroy(ubi, e);
 				goto out_free;
 			}
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 37/56] ubi: Fix Fastmap's update_vol()

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Usually Fastmap is free to consider every PEB in one of the pools
as newer than the existing PEB. Since PEBs in a pool are by definition
newer than everything else.
But update_vol() missed the case that a pool can contain more than
one candidate.

Cc: <stable@vger.kernel.org>
Fixes: dbb7d2a88d ("UBI: Add fastmap core")
Signed-off-by: Richard Weinberger <richard@nod.at>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index a0b5fe784a6f..bd7abf55b768 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -239,7 +239,7 @@ static int update_vol(struct ubi_device *ubi, struct ubi_attach_info *ai,
 			aeb->pnum = new_aeb->pnum;
 			aeb->copy_flag = new_vh->copy_flag;
 			aeb->scrub = new_aeb->scrub;
-			
+			aeb->sqnum = new_aeb->sqnum;
 			ubi_free_aeb(ai, new_aeb);
 
 		/* new_aeb is older */
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 38/56] ubi: fix swapped arguments to call to ubi_alloc_aeb

[Teresa Remmet]

From: Colin Ian King <colin.king@canonical.com>

Static analysis by CoverityScan detected the ec and pnum
arguments are in the wrong order on a call to ubi_alloc_aeb.
Swap the order to fix this.

Fixes: 91f4285fe389a27 ("UBI: provide helpers to allocate and free aeb elements")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index bd7abf55b768..cf5abef8c2a7 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -222,7 +222,7 @@ static int update_vol(struct ubi_device *ubi, struct ubi_attach_info *ai,
 
 		/* new_aeb is newer */
 		if (cmp_res & 1) {
-			victim = ubi_alloc_aeb(ai, aeb->ec, aeb->pnum);
+			victim = ubi_alloc_aeb(ai, aeb->pnum, aeb->ec);
 			if (!victim)
 				return -ENOMEM;
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 39/56] UBI: Fix crash in try_recover_peb()

[Teresa Remmet]

From: Geert Uytterhoeven <geert@linux-m68k.org>

    drivers/mtd/ubi/eba.c: In function ‘try_recover_peb’:
    drivers/mtd/ubi/eba.c:744: warning: ‘vid_hdr’ is used uninitialized in this function

The pointer vid_hdr is indeed not initialized, leading to a crash when
it is dereferenced.

Fix this by obtaining the pointer from the VID buffer, like is done
everywhere else.

Fixes: 3291b52f9ff0acc8 ("UBI: introduce the VID buffer concept")
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 7c0fdefa2576..d5b47cc20519 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -652,6 +652,7 @@ static int try_recover_peb(struct ubi_volume *vol, int pnum, int lnum,
 		goto out_put;
 	}
 
+	vid_hdr = ubi_get_vid_hdr(vidb);
 	ubi_assert(vid_hdr->vol_type == UBI_VID_DYNAMIC);
 
 	memset(ubi->peb_buf + offset, 0xFF, len);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 40/56] ubi: fastmap: Fix add_vol() return value test in ubi_attach_fastmap()

[Teresa Remmet]

From: Boris Brezillon <boris.brezillon@free-electrons.com>

Commit e96a8a3bb671 ("UBI: Fastmap: Do not add vol if it already
exists") introduced a bug by changing the possible error codes returned
by add_vol():
- this function no longer returns NULL in case of allocation failure
  but return ERR_PTR(-ENOMEM)
- when a duplicate entry in the volume RB tree is found it returns
  ERR_PTR(-EEXIST) instead of ERR_PTR(-EINVAL)

Fix the tests done on add_vol() return val to match this new behavior.

Fixes: e96a8a3bb671 ("UBI: Fastmap: Do not add vol if it already exists")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Acked-by: Sheng Yong <shengyong1@huawei.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index cf5abef8c2a7..52f24a6be74f 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -643,11 +643,11 @@ static int ubi_attach_fastmap(struct ubi_device *ubi,
 			     fmvhdr->vol_type,
 			     be32_to_cpu(fmvhdr->last_eb_bytes));
 
-		if (!av)
-			goto fail_bad;
-		if (PTR_ERR(av) == -EINVAL) {
-			ubi_err(ubi, "volume (ID %i) already exists",
-				fmvhdr->vol_id);
+		if (IS_ERR(av)) {
+			if (PTR_ERR(av) == -EEXIST)
+				ubi_err(ubi, "volume (ID %i) already exists",
+					fmvhdr->vol_id);
+
 			goto fail_bad;
 		}
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 41/56] UBI: Fix typos

[Teresa Remmet]

From: "Andrew F. Davis" <afd@ti.com>

Signed-off-by: Andrew F. Davis <afd@ti.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/Kconfig | 2 +-
 drivers/mtd/ubi/io.c    | 2 +-
 drivers/mtd/ubi/ubi.h   | 3 +--
 3 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/Kconfig b/drivers/mtd/ubi/Kconfig
index 4c497931f192..9a344082b7bb 100644
--- a/drivers/mtd/ubi/Kconfig
+++ b/drivers/mtd/ubi/Kconfig
@@ -2,7 +2,7 @@ menuconfig MTD_UBI
 	tristate "Enable UBI - Unsorted block images"
 	select CRC32
 	help
-	  UBI is a software layer above MTD layer which admits of LVM-like
+	  UBI is a software layer above MTD layer which admits use of LVM-like
 	  logical volumes on top of MTD devices, hides some complexities of
 	  flash chips like wear and bad blocks and provides some other useful
 	  capabilities. Please, consult the MTD web site for more details
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 4c2731c4172a..d0ea23b124f2 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -41,7 +41,7 @@
  * About minimal I/O units. In general, UBI assumes flash device model where
  * there is only one minimal I/O unit size. E.g., in case of NOR flash it is 1,
  * in case of NAND flash it is a NAND page, etc. This is reported by MTD in the
- * @ubi->mtd->writesize field. But as an exception, UBI admits of using another
+ * @ubi->mtd->writesize field. But as an exception, UBI admits use of another
  * (smaller) minimal I/O unit size for EC and VID headers to make it possible
  * to do different optimizations.
  *
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 4f1ba745ccc9..a5096934e2de 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -493,8 +493,7 @@ struct ubi_debug_info {
  * @vid_hdr_aloffset: starting offset of the VID header aligned to
  *                    @hdrs_min_io_size
  * @vid_hdr_shift: contains @vid_hdr_offset - @vid_hdr_aloffset
- * @bad_allowed: whether the MTD device admits of bad physical eraseblocks or
- *               not
+ * @bad_allowed: whether the MTD device admits bad physical eraseblocks or not
  * @nor_flash: non-zero if working on top of NOR flash
  * @max_write_size: maximum amount of bytes the underlying flash can write at a
  *                  time (MTD write buffer size)
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 42/56] ubi/upd: Always flush after prepared for an update

[Teresa Remmet]

From: Sebastian Siewior <bigeasy@linutronix.de>

In commit 6afaf8a484cb ("UBI: flush wl before clearing update marker") I
managed to trigger and fix a similar bug. Now here is another version of
which I assumed it wouldn't matter back then but it turns out UBI has a
check for it and will error out like this:

|ubi0 warning: validate_vid_hdr: inconsistent used_ebs
|ubi0 error: validate_vid_hdr: inconsistent VID header at PEB 592

All you need to trigger this is? "ubiupdatevol /dev/ubi0_0 file" + a
powercut in the middle of the operation.
ubi_start_update() sets the update-marker and puts all EBs on the erase
list. After that userland can proceed to write new data while the old EB
aren't erased completely. A powercut at this point is usually not that
much of a tragedy. UBI won't give read access to the static volume
because it has the update marker. It will most likely set the corrupted
flag because it misses some EBs.
So we are all good. Unless the size of the image that has been written
differs from the old image in the magnitude of at least one EB. In that
case UBI will find two different values for `used_ebs' and refuse to
attach the image with the error message mentioned above.

So in order not to get in the situation, the patch will ensure that we
wait until everything is removed before it tries to write any data.
The alternative would be to detect such a case and remove all EBs at the
attached time after we processed the volume-table and see the
update-marker set. The patch looks bigger and I doubt it is worth it
since usually the write() will wait from time to time for a new EB since
usually there not that many spare EB that can be used.

Cc: stable@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/upd.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index e3deb3e931c0..609c789ae4c3 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -139,11 +139,11 @@ int ubi_start_update(struct ubi_device *ubi, struct ubi_volume *vol,
 			return err;
 	}
 
-	if (bytes == 0) {
-		err = ubi_wl_flush(ubi, UBI_ALL, UBI_ALL);
-		if (err)
-			return err;
+	err = ubi_wl_flush(ubi, UBI_ALL, UBI_ALL);
+	if (err)
+		return err;
 
+	if (bytes == 0) {
 		err = clear_update_marker(ubi, vol, 0);
 		if (err)
 			return err;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 43/56] ubi: fastmap: Fix slab corruption

[Teresa Remmet]

From: Rabin Vincent <rabinv@axis.com>

Booting with UBI fastmap and SLUB debugging enabled results in the
following splats.  The problem is that ubi_scan_fastmap() moves the
fastmap blocks from the scan_ai (allocated in scan_fast()) to the ai
allocated in ubi_attach().  This results in two problems:

 - When the scan_ai is freed, aebs which were allocated from its slab
   cache are still in use.

 - When the other ai is being destroyed in destroy_ai(), the
   arguments to kmem_cache_free() call are incorrect since aebs on its
   ->fastmap list were allocated with a slab cache from a differnt ai.

Fix this by making a copy of the aebs in ubi_scan_fastmap() instead of
moving them.

 =============================================================================
 BUG ubi_aeb_slab_cache (Not tainted): Objects remaining in ubi_aeb_slab_cache on __kmem_cache_shutdown()
 -----------------------------------------------------------------------------

 INFO: Slab 0xbfd2da3c objects=17 used=1 fp=0xb33d7748 flags=0x40000080
 CPU: 1 PID: 118 Comm: ubiattach Tainted: G    B           4.9.15 #3
 [<80111910>] (unwind_backtrace) from [<8010d498>] (show_stack+0x18/0x1c)
 [<8010d498>] (show_stack) from [<804a3274>] (dump_stack+0xb4/0xe0)
 [<804a3274>] (dump_stack) from [<8026c47c>] (slab_err+0x78/0x88)
 [<8026c47c>] (slab_err) from [<802735bc>] (__kmem_cache_shutdown+0x180/0x3e0)
 [<802735bc>] (__kmem_cache_shutdown) from [<8024e13c>] (shutdown_cache+0x1c/0x60)
 [<8024e13c>] (shutdown_cache) from [<8024ed64>] (kmem_cache_destroy+0x19c/0x20c)
 [<8024ed64>] (kmem_cache_destroy) from [<8057cc14>] (destroy_ai+0x1dc/0x1e8)
 [<8057cc14>] (destroy_ai) from [<8057f04c>] (ubi_attach+0x3f4/0x450)
 [<8057f04c>] (ubi_attach) from [<8056fe70>] (ubi_attach_mtd_dev+0x60c/0xff8)
 [<8056fe70>] (ubi_attach_mtd_dev) from [<80571d78>] (ctrl_cdev_ioctl+0x110/0x2b8)
 [<80571d78>] (ctrl_cdev_ioctl) from [<8029c77c>] (do_vfs_ioctl+0xac/0xa00)
 [<8029c77c>] (do_vfs_ioctl) from [<8029d10c>] (SyS_ioctl+0x3c/0x64)
 [<8029d10c>] (SyS_ioctl) from [<80108860>] (ret_fast_syscall+0x0/0x1c)
 INFO: Object 0xb33d7e88 @offset=3720
 INFO: Allocated in scan_peb+0x608/0x81c age=72 cpu=1 pid=118
 	kmem_cache_alloc+0x3b0/0x43c
 	scan_peb+0x608/0x81c
 	ubi_attach+0x124/0x450
 	ubi_attach_mtd_dev+0x60c/0xff8
 	ctrl_cdev_ioctl+0x110/0x2b8
 	do_vfs_ioctl+0xac/0xa00
 	SyS_ioctl+0x3c/0x64
 	ret_fast_syscall+0x0/0x1c
 kmem_cache_destroy ubi_aeb_slab_cache: Slab cache still has objects
 CPU: 1 PID: 118 Comm: ubiattach Tainted: G    B           4.9.15 #3
 [<80111910>] (unwind_backtrace) from [<8010d498>] (show_stack+0x18/0x1c)
 [<8010d498>] (show_stack) from [<804a3274>] (dump_stack+0xb4/0xe0)
 [<804a3274>] (dump_stack) from [<8024ed80>] (kmem_cache_destroy+0x1b8/0x20c)
 [<8024ed80>] (kmem_cache_destroy) from [<8057cc14>] (destroy_ai+0x1dc/0x1e8)
 [<8057cc14>] (destroy_ai) from [<8057f04c>] (ubi_attach+0x3f4/0x450)
 [<8057f04c>] (ubi_attach) from [<8056fe70>] (ubi_attach_mtd_dev+0x60c/0xff8)
 [<8056fe70>] (ubi_attach_mtd_dev) from [<80571d78>] (ctrl_cdev_ioctl+0x110/0x2b8)
 [<80571d78>] (ctrl_cdev_ioctl) from [<8029c77c>] (do_vfs_ioctl+0xac/0xa00)
 [<8029c77c>] (do_vfs_ioctl) from [<8029d10c>] (SyS_ioctl+0x3c/0x64)
 [<8029d10c>] (SyS_ioctl) from [<80108860>] (ret_fast_syscall+0x0/0x1c)
 cache_from_obj: Wrong slab cache. ubi_aeb_slab_cache but object is from ubi_aeb_slab_cache
 ------------[ cut here ]------------
 WARNING: CPU: 1 PID: 118 at mm/slab.h:354 kmem_cache_free+0x39c/0x450
 Modules linked in:
 CPU: 1 PID: 118 Comm: ubiattach Tainted: G    B           4.9.15 #3
 [<80111910>] (unwind_backtrace) from [<8010d498>] (show_stack+0x18/0x1c)
 [<8010d498>] (show_stack) from [<804a3274>] (dump_stack+0xb4/0xe0)
 [<804a3274>] (dump_stack) from [<80120e40>] (__warn+0xf4/0x10c)
 [<80120e40>] (__warn) from [<80120f20>] (warn_slowpath_null+0x28/0x30)
 [<80120f20>] (warn_slowpath_null) from [<80271fe0>] (kmem_cache_free+0x39c/0x450)
 [<80271fe0>] (kmem_cache_free) from [<8057cb88>] (destroy_ai+0x150/0x1e8)
 [<8057cb88>] (destroy_ai) from [<8057ef1c>] (ubi_attach+0x2c4/0x450)
 [<8057ef1c>] (ubi_attach) from [<8056fe70>] (ubi_attach_mtd_dev+0x60c/0xff8)
 [<8056fe70>] (ubi_attach_mtd_dev) from [<80571d78>] (ctrl_cdev_ioctl+0x110/0x2b8)
 [<80571d78>] (ctrl_cdev_ioctl) from [<8029c77c>] (do_vfs_ioctl+0xac/0xa00)
 [<8029c77c>] (do_vfs_ioctl) from [<8029d10c>] (SyS_ioctl+0x3c/0x64)
 [<8029d10c>] (SyS_ioctl) from [<80108860>] (ret_fast_syscall+0x0/0x1c)
 ---[ end trace 2bd8396277fd0a0b ]---
 =============================================================================
 BUG ubi_aeb_slab_cache (Tainted: G    B   W      ): page slab pointer corrupt.
 -----------------------------------------------------------------------------

 INFO: Allocated in scan_peb+0x608/0x81c age=104 cpu=1 pid=118
 	kmem_cache_alloc+0x3b0/0x43c
 	scan_peb+0x608/0x81c
 	ubi_attach+0x124/0x450
 	ubi_attach_mtd_dev+0x60c/0xff8
 	ctrl_cdev_ioctl+0x110/0x2b8
 	do_vfs_ioctl+0xac/0xa00
 	SyS_ioctl+0x3c/0x64
 	ret_fast_syscall+0x0/0x1c
 INFO: Slab 0xbfd2da3c objects=17 used=1 fp=0xb33d7748 flags=0x40000081
 INFO: Object 0xb33d7e88 @offset=3720 fp=0xb33d7da0

 Redzone b33d7e80: cc cc cc cc cc cc cc cc                          ........
 Object b33d7e88: 02 00 00 00 01 00 00 00 00 f0 ff 7f ff ff ff ff  ................
 Object b33d7e98: 00 00 00 00 00 00 00 00 bd 16 00 00 00 00 00 00  ................
 Object b33d7ea8: 00 01 00 00 00 02 00 00 00 00 00 00 00 00 00 00  ................
 Redzone b33d7eb8: cc cc cc cc                                      ....
 Padding b33d7f60: 5a 5a 5a 5a 5a 5a 5a 5a                          ZZZZZZZZ
 CPU: 1 PID: 118 Comm: ubiattach Tainted: G    B   W       4.9.15 #3
 [<80111910>] (unwind_backtrace) from [<8010d498>] (show_stack+0x18/0x1c)
 [<8010d498>] (show_stack) from [<804a3274>] (dump_stack+0xb4/0xe0)
 [<804a3274>] (dump_stack) from [<80271770>] (free_debug_processing+0x320/0x3c4)
 [<80271770>] (free_debug_processing) from [<80271ad0>] (__slab_free+0x2bc/0x430)
 [<80271ad0>] (__slab_free) from [<80272024>] (kmem_cache_free+0x3e0/0x450)
 [<80272024>] (kmem_cache_free) from [<8057cb88>] (destroy_ai+0x150/0x1e8)
 [<8057cb88>] (destroy_ai) from [<8057ef1c>] (ubi_attach+0x2c4/0x450)
 [<8057ef1c>] (ubi_attach) from [<8056fe70>] (ubi_attach_mtd_dev+0x60c/0xff8)
 [<8056fe70>] (ubi_attach_mtd_dev) from [<80571d78>] (ctrl_cdev_ioctl+0x110/0x2b8)
 [<80571d78>] (ctrl_cdev_ioctl) from [<8029c77c>] (do_vfs_ioctl+0xac/0xa00)
 [<8029c77c>] (do_vfs_ioctl) from [<8029d10c>] (SyS_ioctl+0x3c/0x64)
 [<8029d10c>] (SyS_ioctl) from [<80108860>] (ret_fast_syscall+0x0/0x1c)
 FIX ubi_aeb_slab_cache: Object at 0xb33d7e88 not freed

Signed-off-by: Rabin Vincent <rabinv@axis.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 33 +++++++++++++++++++++++++++++----
 1 file changed, 29 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 52f24a6be74f..2debf42a06b4 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -764,6 +764,24 @@ static int find_fm_anchor(struct ubi_attach_info *ai)
 	return ret;
 }
 
+static struct ubi_ainf_peb *clone_aeb(struct ubi_attach_info *ai,
+				      struct ubi_ainf_peb *old)
+{
+	struct ubi_ainf_peb *new;
+
+	new = ubi_alloc_aeb(ai, old->pnum, old->ec);
+	if (!new)
+		return NULL;
+
+	new->vol_id = old->vol_id;
+	new->sqnum = old->sqnum;
+	new->lnum = old->lnum;
+	new->scrub = old->scrub;
+	new->copy_flag = old->copy_flag;
+
+	return new;
+}
+
 /**
  * ubi_scan_fastmap - scan the fastmap.
  * @ubi: UBI device object
@@ -783,7 +801,7 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	struct ubi_vid_hdr *vh;
 	struct ubi_ec_hdr *ech;
 	struct ubi_fastmap_layout *fm;
-	struct ubi_ainf_peb *tmp_aeb, *aeb;
+	struct ubi_ainf_peb *aeb;
 	int i, used_blocks, pnum, fm_anchor, ret = 0;
 	size_t fm_size;
 	__be32 crc, tmp_crc;
@@ -793,9 +811,16 @@ int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 	if (fm_anchor < 0)
 		return UBI_NO_FASTMAP;
 
-	/* Move all (possible) fastmap blocks into our new attach structure. */
-	list_for_each_entry_safe(aeb, tmp_aeb, &scan_ai->fastmap, u.list)
-		list_move_tail(&aeb->u.list, &ai->fastmap);
+	/* Copy all (possible) fastmap blocks into our new attach structure. */
+	list_for_each_entry(aeb, &scan_ai->fastmap, u.list) {
+		struct ubi_ainf_peb *new;
+
+		new = clone_aeb(ai, aeb);
+		if (!new)
+			return -ENOMEM;
+
+		list_add(&new->u.list, &ai->fastmap);
+	}
 
 	memset(ubi->fm_buf, 0, ubi->fm_size);
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 44/56] ubi: pr_err() strings should end with newlines

[Teresa Remmet]

From: Ben Dooks <ben.dooks@codethink.co.uk>

In ubi_attach_mtd_dev() the pr_err() calls should have their
messgaes terminated with a new-line to avoid other messages
being concatenated onto the end.

Signed-off-by: Ben Dooks <ben.dooks@codethink.co.uk>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fix conflics]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/build.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 13f39b39c8fb..5917fe67918e 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -522,7 +522,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 	for (i = 0; i < UBI_MAX_DEVICES; i++) {
 		ubi = ubi_devices[i];
 		if (ubi && mtd == ubi->mtd) {
-			pr_err("ubi: mtd%d is already attached to ubi%d",
+			pr_err("ubi: mtd%d is already attached to ubi%d\n",
 				mtd->index, i);
 			return -EEXIST;
 		}
@@ -537,7 +537,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 	 * no sense to attach emulated MTD devices, so we prohibit this.
 	 */
 	if (mtd->type == MTD_UBIVOLUME) {
-		pr_err("ubi: refuse attaching mtd%d - it is already emulated on top of UBI",
+		pr_err("ubi: refuse attaching mtd%d - it is already emulated on top of UBI\n",
 			mtd->index);
 		return -EINVAL;
 	}
@@ -548,7 +548,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 			if (!ubi_devices[ubi_num])
 				break;
 		if (ubi_num == UBI_MAX_DEVICES) {
-			pr_err("ubi: only %d UBI devices may be created",
+			pr_err("ubi: only %d UBI devices may be created\n",
 				UBI_MAX_DEVICES);
 			return -ENFILE;
 		}
@@ -558,7 +558,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num,
 
 		/* Make sure ubi_num is not busy */
 		if (ubi_devices[ubi_num]) {
-			pr_err("ubi: ubi%i already exists", ubi_num);
+			pr_err("ubi: ubi%i already exists\n", ubi_num);
 			return -EEXIST;
 		}
 	}
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 45/56] ubi: fastmap: fix spelling mistake: "invalidiate" -> "invalidate"

[Teresa Remmet]

From: Colin Ian King <colin.king@canonical.com>

Trivial fix to spelling mistake in ubi_err error message

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 2debf42a06b4..c50684a19401 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -1552,7 +1552,7 @@ err:
 
 	ret = invalidate_fastmap(ubi);
 	if (ret < 0) {
-		ubi_err(ubi, "Unable to invalidiate current fastmap!");
+		ubi_err(ubi, "Unable to invalidate current fastmap!");
 		ubi_ro_mode(ubi);
 	} else {
 		return_fm_pebs(ubi, old_fm);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 46/56] UBI: Fix two typos in comments

[Teresa Remmet]

From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/ubi-media.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index 27c49beafe17..43584843542f 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -225,7 +225,7 @@ struct ubi_ec_hdr {
  * copy. UBI also calculates data CRC when the data is moved and stores it at
  * the @data_crc field of the copy (P1). So when UBI needs to pick one physical
  * eraseblock of two (P or P1), the @copy_flag of the newer one (P1) is
- * examined. If it is cleared, the situation* is simple and the newer one is
+ * examined. If it is cleared, the situation is simple and the newer one is
  * picked. If it is set, the data CRC of the copy (P1) is examined. If the CRC
  * checksum is correct, this physical eraseblock is selected (P1). Otherwise
  * the older one (P) is selected.
@@ -385,7 +385,7 @@ struct ubi_vtbl_record {
 #define UBI_FM_POOL_MAGIC	0x67AF4D08
 #define UBI_FM_EBA_MAGIC	0xf0c040a8
 
-/* A fastmap supber block can be located between PEB 0 and
+/* A fastmap super block can be located between PEB 0 and
  * UBI_FM_MAX_START */
 #define UBI_FM_MAX_START	64
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 47/56] ubi: fastmap: Clean up the initialization of pointer p

[Teresa Remmet]

From: Colin Ian King <colin.king@canonical.com>

The pointer p is being initialized with one value and a few lines
later being set to a newer replacement value. Clean up the code by
using the latter assignment to p as the initial value. Cleans up
clang warning:

drivers/mtd/ubi/fastmap.c:217:19: warning: Value stored to 'p'
during its initialization is never read

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index c50684a19401..5f31f2081437 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -148,9 +148,8 @@ static void assign_aeb_to_av(struct ubi_attach_info *ai,
 			     struct ubi_ainf_volume *av)
 {
 	struct ubi_ainf_peb *tmp_aeb;
-	struct rb_node **p = &ai->volumes.rb_node, *parent = NULL;
+	struct rb_node **p = &av->root.rb_node, *parent = NULL;
 
-	p = &av->root.rb_node;
 	while (*p) {
 		parent = *p;
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 48/56] ubi: fastmap: Erase outdated anchor PEBs during attach

[Teresa Remmet]

From: Sascha Hauer <s.hauer@pengutronix.de>

The fastmap update code might erase the current fastmap anchor PEB
in case it doesn't find any new free PEB. When a power cut happens
in this situation we must not have any outdated fastmap anchor PEB
on the device, because that would be used to attach during next
boot.
The easiest way to make that sure is to erase all outdated fastmap
anchor PEBs synchronously during attach.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Reviewed-by: Richard Weinberger <richard@nod.at>
Fixes: dbb7d2a88d2a ("UBI: Add fastmap core")
Cc: <stable@vger.kernel.org>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/wl.c | 77 ++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 57 insertions(+), 20 deletions(-)

diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 54c119d20b07..b8513615b64d 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -1392,6 +1392,46 @@ static void shutdown_work(struct ubi_device *ubi)
 }
 
 /**
+ * erase_aeb - erase a PEB given in UBI attach info PEB
+ * @ubi: UBI device description object
+ * @aeb: UBI attach info PEB
+ * @sync: If true, erase synchronously. Otherwise schedule for erasure
+ */
+static int erase_aeb(struct ubi_device *ubi, struct ubi_ainf_peb *aeb, bool sync)
+{
+	struct ubi_wl_entry *e;
+	int err;
+
+	e = kmalloc(sizeof(*e), GFP_KERNEL);
+	if (!e)
+		return -ENOMEM;
+
+	e->pnum = aeb->pnum;
+	e->ec = aeb->ec;
+	ubi->lookuptbl[e->pnum] = e;
+
+	if (sync) {
+		err = sync_erase(ubi, e, false);
+		if (err)
+			goto out_free;
+
+		wl_tree_add(e, &ubi->free);
+		ubi->free_count++;
+	} else {
+		err = schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0, false);
+		if (err)
+			goto out_free;
+	}
+
+	return 0;
+
+out_free:
+	wl_entry_destroy(ubi, e);
+
+	return err;
+}
+
+/**
  * ubi_wl_init - initialize the WL sub-system using attaching information.
  * @ubi: UBI device description object
  * @ai: attaching information
@@ -1424,18 +1464,10 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 
 	ubi->free_count = 0;
 	list_for_each_entry_safe(aeb, tmp, &ai->erase, u.list) {
-		e = kmalloc(sizeof(*e), GFP_KERNEL);
-		if (!e)
+		err = erase_aeb(ubi, aeb, false);
+		if (err)
 			goto out_free;
 
-		e->pnum = aeb->pnum;
-		e->ec = aeb->ec;
-		ubi->lookuptbl[e->pnum] = e;
-		if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0, false)) {
-			wl_entry_destroy(ubi, e);
-			goto out_free;
-		}
-
 		found_pebs++;
 	}
 
@@ -1487,6 +1519,8 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 			ubi_assert(!ubi->lookuptbl[e->pnum]);
 			ubi->lookuptbl[e->pnum] = e;
 		} else {
+			bool sync = false;
+
 			/*
 			 * Usually old Fastmap PEBs are scheduled for erasure
 			 * and we don't have to care about them but if we face
@@ -1496,18 +1530,21 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 			if (ubi->lookuptbl[aeb->pnum])
 				continue;
 
-			e = kmalloc(sizeof(*e), GFP_KERNEL);
-			if (!e)
-				goto out_free;
+			/*
+			 * The fastmap update code might not find a free PEB for
+			 * writing the fastmap anchor to and then reuses the
+			 * current fastmap anchor PEB. When this PEB gets erased
+			 * and a power cut happens before it is written again we
+			 * must make sure that the fastmap attach code doesn't
+			 * find any outdated fastmap anchors, hence we erase the
+			 * outdated fastmap anchor PEBs synchronously here.
+			 */
+			if (aeb->vol_id == UBI_FM_SB_VOLUME_ID)
+				sync = true;
 
-			e->pnum = aeb->pnum;
-			e->ec = aeb->ec;
-			ubi_assert(!ubi->lookuptbl[e->pnum]);
-			ubi->lookuptbl[e->pnum] = e;
-			if (schedule_erase(ubi, e, aeb->vol_id, aeb->lnum, 0, false)) {
-				wl_entry_destroy(ubi, e);
+			err = erase_aeb(ubi, aeb, sync);
+			if (err)
 				goto out_free;
-			}
 		}
 
 		found_pebs++;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 49/56] ubi: Fastmap: Fix typo

[Teresa Remmet]

From: Sascha Hauer <s.hauer@pengutronix.de>

Fix misspelling of 'available' in function name.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conficts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/fastmap-wl.c | 2 +-
 drivers/mtd/ubi/wl.c         | 2 +-
 drivers/mtd/ubi/wl.h         | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/ubi/fastmap-wl.c b/drivers/mtd/ubi/fastmap-wl.c
index 07982aab169a..08593ea23aee 100644
--- a/drivers/mtd/ubi/fastmap-wl.c
+++ b/drivers/mtd/ubi/fastmap-wl.c
@@ -52,7 +52,7 @@ static void return_unused_pool_pebs(struct ubi_device *ubi,
 	}
 }
 
-static int anchor_pebs_avalible(struct rb_root *root)
+static int anchor_pebs_available(struct rb_root *root)
 {
 	struct rb_node *p;
 	struct ubi_wl_entry *e;
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index b8513615b64d..8d45b318f5c1 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -663,7 +663,7 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 #ifdef CONFIG_MTD_UBI_FASTMAP
 	/* Check whether we need to produce an anchor PEB */
 	if (!anchor)
-		anchor = !anchor_pebs_avalible(&ubi->free);
+		anchor = !anchor_pebs_available(&ubi->free);
 
 	if (anchor) {
 		e1 = find_anchor_wl_entry(&ubi->used);
diff --git a/drivers/mtd/ubi/wl.h b/drivers/mtd/ubi/wl.h
index 2a1899e9877d..60168116dbf9 100644
--- a/drivers/mtd/ubi/wl.h
+++ b/drivers/mtd/ubi/wl.h
@@ -1,7 +1,7 @@
 #ifndef UBI_WL_H
 #define UBI_WL_H
 #ifdef CONFIG_MTD_UBI_FASTMAP
-static int anchor_pebs_avalible(struct rb_root *root);
+static int anchor_pebs_available(struct rb_root *root);
 static struct ubi_wl_entry *find_anchor_wl_entry(struct rb_root *root);
 static struct ubi_wl_entry *get_peb_for_wl(struct ubi_device *ubi);
 static void ubi_fastmap_close(struct ubi_device *ubi);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 50/56] ubi: Fix copy/paste error in function documentation

[Teresa Remmet]

From: Sascha Hauer <s.hauer@pengutronix.de>

The function documentation of leb_write_trylock is copied from
leb_write_lock. Replace the function name with the correct one.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index d5b47cc20519..69a6242e054a 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -368,7 +368,7 @@ static int leb_write_lock(struct ubi_device *ubi, int vol_id, int lnum)
 }
 
 /**
- * leb_write_lock - lock logical eraseblock for writing.
+ * leb_write_trylock - try to lock logical eraseblock for writing.
  * @ubi: UBI device description object
  * @vol_id: volume ID
  * @lnum: logical eraseblock number
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 51/56] mtd: ubi: wl: Fix error return code in ubi_wl_init()

[Teresa Remmet]

From: Wei Yongjun <weiyongjun1@huawei.com>

Fix to return error code -ENOMEM from the kmem_cache_alloc() error
handling case instead of 0, as done elsewhere in this function.

Fixes: f78e5623f45b ("ubi: fastmap: Erase outdated anchor PEBs during
attach")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Reviewed-by: Boris Brezillon <boris.brezillon@free-electrons.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflicts]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/wl.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 8d45b318f5c1..fa1b8137024a 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -1473,8 +1473,10 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 
 	list_for_each_entry(aeb, &ai->free, u.list) {
 		e = kmalloc(sizeof(*e), GFP_KERNEL);
-		if (!e)
+		if (!e) {
+			err = -ENOMEM;
 			goto out_free;
+		}
 
 		e->pnum = aeb->pnum;
 		e->ec = aeb->ec;
@@ -1491,8 +1493,10 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
 	ubi_rb_for_each_entry(rb1, av, &ai->volumes, rb) {
 		ubi_rb_for_each_entry(rb2, aeb, &av->root, u.rb) {
 			e = kmalloc(sizeof(*e), GFP_KERNEL);
-			if (!e)
+			if (!e) {
+				err = -ENOMEM;
 				goto out_free;
+			}
 
 			e->pnum = aeb->pnum;
 			e->ec = aeb->ec;
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 52/56] ubi: fastmap: Correctly handle interrupted erasures in EBA

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Fastmap cannot track the LEB unmap operation, therefore it can
happen that after an interrupted erasure the mapping still looks
good from Fastmap's point of view, while reading from the PEB will
cause an ECC error and confuses the upper layer.

Instead of teaching users of UBI how to deal with that, we read back
the VID header and check for errors. If the PEB is empty or shows ECC
errors we fixup the mapping and schedule the PEB for erasure.

Fixes: dbb7d2a88d2a ("UBI: Add fastmap core")
Cc: <stable@vger.kernel.org>
Reported-by: martin bayern <Martinbayern@outlook.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed compile issue]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 88 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 87 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 69a6242e054a..c0653122ffc5 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -456,6 +456,80 @@ out_unlock:
 	return err;
 }
 
+#ifdef CONFIG_MTD_UBI_FASTMAP
+/**
+ * check_mapping - check and fixup a mapping
+ * @ubi: UBI device description object
+ * @vol: volume description object
+ * @lnum: logical eraseblock number
+ * @pnum: physical eraseblock number
+ *
+ * Checks whether a given mapping is valid. Fastmap cannot track LEB unmap
+ * operations, if such an operation is interrupted the mapping still looks
+ * good, but upon first read an ECC is reported to the upper layer.
+ * Normaly during the full-scan at attach time this is fixed, for Fastmap
+ * we have to deal with it while reading.
+ * If the PEB behind a LEB shows this symthom we change the mapping to
+ * %UBI_LEB_UNMAPPED and schedule the PEB for erasure.
+ *
+ * Returns 0 on success, negative error code in case of failure.
+ */
+static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
+			 int *pnum)
+{
+	int err;
+	struct ubi_vid_io_buf *vidb;
+
+	if (!ubi->fast_attach)
+		return 0;
+
+	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
+	if (!vidb)
+		return -ENOMEM;
+
+	err = ubi_io_read_vid_hdr(ubi, *pnum, vidb, 0);
+	if (err > 0 && err != UBI_IO_BITFLIPS) {
+		int torture = 0;
+
+		switch (err) {
+			case UBI_IO_FF:
+			case UBI_IO_FF_BITFLIPS:
+			case UBI_IO_BAD_HDR:
+			case UBI_IO_BAD_HDR_EBADMSG:
+				break;
+			default:
+				ubi_assert(0);
+		}
+
+		if (err == UBI_IO_BAD_HDR_EBADMSG || err == UBI_IO_FF_BITFLIPS)
+			torture = 1;
+
+		vol->eba_tbl->entries[lnum].pnum = UBI_LEB_UNMAPPED;
+		ubi_wl_put_peb(ubi, vol->vol_id, lnum, *pnum, torture);
+
+		*pnum = UBI_LEB_UNMAPPED;
+	} else if (err < 0) {
+		ubi_err(ubi, "unable to read VID header back from PEB %i: %i",
+			*pnum, err);
+
+		goto out_free;
+	}
+
+	err = 0;
+
+out_free:
+	ubi_free_vid_buf(vidb);
+
+	return err;
+}
+#else
+static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
+		  int *pnum)
+{
+	return 0;
+}
+#endif
+
 /**
  * ubi_eba_read_leb - read data.
  * @ubi: UBI device description object
@@ -488,7 +562,13 @@ int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		return err;
 
 	pnum = vol->eba_tbl->entries[lnum].pnum;
-	if (pnum < 0) {
+	if (pnum >= 0) {
+		err = check_mapping(ubi, vol, lnum, &pnum);
+		if (err < 0)
+			goto out_unlock;
+	}
+
+	if (pnum == UBI_LEB_UNMAPPED) {
 		/*
 		 * The logical eraseblock is not mapped, fill the whole buffer
 		 * with 0xFF bytes. The exception is static volumes for which
@@ -836,6 +916,12 @@ int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 
 	pnum = vol->eba_tbl->entries[lnum].pnum;
 	if (pnum >= 0) {
+		err = check_mapping(ubi, vol, lnum, &pnum);
+		if (err < 0)
+			goto out;
+	}
+
+	if (pnum >= 0) {
 		dbg_eba("write %d bytes at offset %d of LEB %d:%d, PEB %d",
 			len, offset, vol_id, lnum, pnum);
 
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 53/56] ubi: fastmap: Check each mapping only once

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Maintain a bitmap to keep track of which LEB->PEB mapping
was checked already.
That way we have to read back VID headers only once.

Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed Conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/build.c   |  1 +
 drivers/mtd/ubi/eba.c     |  4 ++++
 drivers/mtd/ubi/fastmap.c | 20 ++++++++++++++++++++
 drivers/mtd/ubi/ubi.h     | 11 +++++++++++
 drivers/mtd/ubi/vtbl.c    | 16 +++++++++++++++-
 5 files changed, 51 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 5917fe67918e..536077ebf3f3 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -219,6 +219,7 @@ void ubi_free_internal_volumes(struct ubi_device *ubi)
 	for (i = ubi->vtbl_slots;
 	     i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
 		ubi_eba_replace_table(ubi->volumes[i], NULL);
+		ubi_fastmap_destroy_checkmap(ubi->volumes[i]);
 		kfree(ubi->volumes[i]);
 	}
 }
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index c0653122ffc5..d29df1d1550b 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -483,6 +483,9 @@ static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnu
 	if (!ubi->fast_attach)
 		return 0;
 
+	if (!vol->checkmap || test_bit(lnum, vol->checkmap))
+		return 0;
+
 	vidb = ubi_alloc_vid_buf(ubi, GFP_NOFS);
 	if (!vidb)
 		return -ENOMEM;
@@ -515,6 +518,7 @@ static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnu
 		goto out_free;
 	}
 
+	set_bit(lnum, vol->checkmap);
 	err = 0;
 
 out_free:
diff --git a/drivers/mtd/ubi/fastmap.c b/drivers/mtd/ubi/fastmap.c
index 5f31f2081437..84c2912bf560 100644
--- a/drivers/mtd/ubi/fastmap.c
+++ b/drivers/mtd/ubi/fastmap.c
@@ -1034,6 +1034,26 @@ free_fm_sb:
 	goto out;
 }
 
+int ubi_fastmap_init_checkmap(struct ubi_volume *vol, int leb_count)
+{
+	struct ubi_device *ubi = vol->ubi;
+
+	if (!ubi->fast_attach)
+		return 0;
+
+	vol->checkmap = kcalloc(BITS_TO_LONGS(leb_count), sizeof(unsigned long),
+				GFP_KERNEL);
+	if (!vol->checkmap)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void ubi_fastmap_destroy_checkmap(struct ubi_volume *vol)
+{
+	kfree(vol->checkmap);
+}
+
 /**
  * ubi_write_fastmap - writes a fastmap.
  * @ubi: UBI device object
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index a5096934e2de..b85d60205583 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -305,6 +305,9 @@ struct ubi_eba_leb_desc {
  * @changing_leb: %1 if the atomic LEB change ioctl command is in progress
  * @direct_writes: %1 if direct writes are enabled for this volume
  *
+ * @checkmap: bitmap to remember which PEB->LEB mappings got checked,
+ *            protected by UBI LEB lock tree.
+ *
  * The @corrupted field indicates that the volume's contents is corrupted.
  * Since UBI protects only static volumes, this field is not relevant to
  * dynamic volumes - it is user's responsibility to assure their data
@@ -348,6 +351,10 @@ struct ubi_volume {
 	unsigned int changing_leb:1;
 	unsigned int direct_writes:1;
 
+#ifdef CONFIG_MTD_UBI_FASTMAP
+	unsigned long *checkmap;
+#endif
+
 	struct list_head list;
 };
 
@@ -897,8 +904,12 @@ size_t ubi_calc_fm_size(struct ubi_device *ubi);
 int ubi_update_fastmap(struct ubi_device *ubi);
 int ubi_scan_fastmap(struct ubi_device *ubi, struct ubi_attach_info *ai,
 		     struct ubi_attach_info *scan_ai);
+int ubi_fastmap_init_checkmap(struct ubi_volume *vol, int leb_count);
+void ubi_fastmap_destroy_checkmap(struct ubi_volume *vol);
 #else
 static inline int ubi_update_fastmap(struct ubi_device *ubi) { return 0; }
+int static inline ubi_fastmap_init_checkmap(struct ubi_volume *vol, int leb_count) { return 0; }
+static inline void ubi_fastmap_destroy_checkmap(struct ubi_volume *vol) {}
 #endif
 
 /*
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 41af1f4d9204..4ca5bb897463 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -526,7 +526,7 @@ static int init_volumes(struct ubi_device *ubi,
 			const struct ubi_attach_info *ai,
 			const struct ubi_vtbl_record *vtbl)
 {
-	int i, reserved_pebs = 0;
+	int i, err, reserved_pebs = 0;
 	struct ubi_ainf_volume *av;
 	struct ubi_volume *vol;
 
@@ -610,6 +610,16 @@ static int init_volumes(struct ubi_device *ubi,
 			(long long)(vol->used_ebs - 1) * vol->usable_leb_size;
 		vol->used_bytes += av->last_data_size;
 		vol->last_eb_bytes = av->last_data_size;
+
+		/*
+		 * We use ubi->peb_count and not vol->reserved_pebs because
+		 * we want to keep the code simple. Otherwise we'd have to
+		 * resize/check the bitmap upon volume resize too.
+		 * Allocating a few bytes more does not hurt.
+		 */
+		err = ubi_fastmap_init_checkmap(vol, ubi->peb_count);
+		if (err)
+			return err;
 	}
 
 	/* And add the layout volume */
@@ -635,6 +645,9 @@ static int init_volumes(struct ubi_device *ubi,
 	reserved_pebs += vol->reserved_pebs;
 	ubi->vol_count += 1;
 	vol->ubi = ubi;
+	err = ubi_fastmap_init_checkmap(vol, UBI_LAYOUT_VOLUME_EBS);
+	if (err)
+		return err;
 
 	if (reserved_pebs > ubi->avail_pebs) {
 		ubi_err(ubi, "not enough PEBs, required %d, available %d",
@@ -837,6 +850,7 @@ int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_attach_info *ai)
 out_free:
 	vfree(ubi->vtbl);
 	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
+		ubi_fastmap_destroy_checkmap(ubi->volumes[i]);
 		kfree(ubi->volumes[i]);
 		ubi->volumes[i] = NULL;
 	}
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 54/56] ubi: fastmap: Detect EBA mismatches on-the-fly

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

Now we have the machinery to detect EBA mismatches on-the-fly
by comparing the in-memory volume ID and LEB number with the found
VID header.
This helps to detect malfunction of Fastmap.

Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/eba.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index d29df1d1550b..3aae3029e5b6 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -479,6 +479,7 @@ static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnu
 {
 	int err;
 	struct ubi_vid_io_buf *vidb;
+	struct ubi_vid_hdr *vid_hdr;
 
 	if (!ubi->fast_attach)
 		return 0;
@@ -516,6 +517,22 @@ static int check_mapping(struct ubi_device *ubi, struct ubi_volume *vol, int lnu
 			*pnum, err);
 
 		goto out_free;
+	} else {
+		int found_vol_id, found_lnum;
+
+		ubi_assert(err == 0 || err == UBI_IO_BITFLIPS);
+
+		vid_hdr = ubi_get_vid_hdr(vidb);
+		found_vol_id = be32_to_cpu(vid_hdr->vol_id);
+		found_lnum = be32_to_cpu(vid_hdr->lnum);
+
+		if (found_lnum != lnum || found_vol_id != vol->vol_id) {
+			ubi_err(ubi, "EBA mismatch! PEB %i is LEB %i:%i instead of LEB %i:%i",
+				*pnum, found_vol_id, found_lnum, vol->vol_id, lnum);
+			ubi_ro_mode(ubi);
+			err = -EINVAL;
+			goto out_free;
+		}
 	}
 
 	set_bit(lnum, vol->checkmap);
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 55/56] mtd: ubi: Update ubi-media.h to dual license

[Teresa Remmet]

From: Lionel Debieve <lionel.debieve@st.com>

Update license template using SPDX. Move the global layout
of UBI headers to dual license helping UBI to be the standard
solution for raw NAND management.

Signed-off-by: Lionel Debieve <lionel.debieve@st.com>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Artem Bityutskiy <dedekind1@gmail.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Richard Weinberger <richard@nod.at>
[Fixed conflict]
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/ubi-media.h | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index 43584843542f..195ff8ca8211 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -1,24 +1,12 @@
+/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */
 /*
- * Copyright (c) International Business Machines Corp., 2006
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
- * the GNU General Public License for more details.
- *
+ * Copyright (C) International Business Machines Corp., 2006
  * Authors: Artem Bityutskiy (Битюцкий Артём)
  *          Thomas Gleixner
  *          Frank Haverkamp
  *          Oliver Lohmann
  *          Andreas Arnez
- */
-
-/*
+ *
  * This file defines the layout of UBI headers and all the other UBI on-flash
  * data structures.
  */
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

[PATCH 56/56] ubi: Initialize Fastmap checkmapping correctly

[Teresa Remmet]

From: Richard Weinberger <richard@nod.at>

We cannot do it last, otherwithse it will be skipped for dynamic
volumes.

Reported-by: Lachmann, Juergen <juergen.lachmann@harman.com>
Fixes: 34653fd8c46e ("ubi: fastmap: Check each mapping only once")
Signed-off-by: Richard Weinberger <richard@nod.at>
Signed-off-by: Teresa Remmet <t.remmet@phytec.de>
---
 drivers/mtd/ubi/vtbl.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 4ca5bb897463..6959564a1354 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -569,6 +569,16 @@ static int init_volumes(struct ubi_device *ubi,
 		reserved_pebs += vol->reserved_pebs;
 
 		/*
+		 * We use ubi->peb_count and not vol->reserved_pebs because
+		 * we want to keep the code simple. Otherwise we'd have to
+		 * resize/check the bitmap upon volume resize too.
+		 * Allocating a few bytes more does not hurt.
+		 */
+		err = ubi_fastmap_init_checkmap(vol, ubi->peb_count);
+		if (err)
+			return err;
+
+		/*
 		 * In case of dynamic volume UBI knows nothing about how many
 		 * data is stored there. So assume the whole volume is used.
 		 */
@@ -610,16 +620,6 @@ static int init_volumes(struct ubi_device *ubi,
 			(long long)(vol->used_ebs - 1) * vol->usable_leb_size;
 		vol->used_bytes += av->last_data_size;
 		vol->last_eb_bytes = av->last_data_size;
-
-		/*
-		 * We use ubi->peb_count and not vol->reserved_pebs because
-		 * we want to keep the code simple. Otherwise we'd have to
-		 * resize/check the bitmap upon volume resize too.
-		 * Allocating a few bytes more does not hurt.
-		 */
-		err = ubi_fastmap_init_checkmap(vol, ubi->peb_count);
-		if (err)
-			return err;
 	}
 
 	/* And add the layout volume */
-- 
2.7.4


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

Re: [PATCH 00/56] Update Barebox UBI

[Sascha Hauer]

On Wed, Aug 29, 2018 at 02:19:16PM +0200, Teresa Remmet wrote:
> Hello, 
> 
> A new update of the Barebox UBI stack to the Kernel v4.19-rc1.
> All patches are cherry-picks from the kernel tree.
> 
> Stumbled over some conflicts mostly due to non exsisting threads in barebox.
> Run some tests and did not face any regressions so far.

Applied, thanks. I didn't look very close at these patches, just relying
on the fact that they work on the Kernel.

Sascha

> 
> Regards,
> Teresa
> 
> Andrew F. Davis (1):
>   UBI: Fix typos
> 
> Ben Dooks (1):
>   ubi: pr_err() strings should end with newlines
> 
> Boris Brezillon (18):
>   UBI: fastmap: use ubi_find_volume() instead of open coding it
>   UBI: fix add_fastmap() to use the vid_hdr passed in argument
>   UBI: fastmap: avoid multiple be32_to_cpu() when unneccesary
>   UBI: fastmap: scrub PEB when bitflips are detected in a free PEB EC
>     header
>   UBI: factorize code used to manipulate volumes at attach time
>   UBI: factorize destroy_av() and ubi_remove_av() code
>   UBI: fastmap: use ubi_rb_for_each_entry() in unmap_peb()
>   UBI: fastmap: use ubi_io_{read, write}_data() instead of ubi_io_{read,
>     write}()
>   UBI: provide helpers to allocate and free aeb elements
>   UBI: move the global ech and vidh variables into struct
>     ubi_attach_info
>   UBI: simplify recover_peb() code
>   UBI: simplify LEB write and atomic LEB change code
>   UBI: add an helper to check lnum validity
>   UBI: provide an helper to check whether a LEB is mapped or not
>   UBI: provide an helper to query LEB information
>   UBI: hide EBA internals
>   UBI: introduce the VID buffer concept
>   ubi: fastmap: Fix add_vol() return value test in ubi_attach_fastmap()
> 
> Colin Ian King (3):
>   ubi: fix swapped arguments to call to ubi_alloc_aeb
>   ubi: fastmap: fix spelling mistake: "invalidiate" -> "invalidate"
>   ubi: fastmap: Clean up the initialization of pointer p
> 
> Dan Carpenter (1):
>   UBI: Clean up return in ubi_remove_volume()
> 
> Geert Uytterhoeven (1):
>   UBI: Fix crash in try_recover_peb()
> 
> Heiko Schocher (1):
>   UBI: Set free_count to zero before walking through erase list
> 
> Iosif Harutyunov (1):
>   ubi: Fix race condition between ubi device creation and udev
> 
> Lionel Debieve (1):
>   mtd: ubi: Update ubi-media.h to dual license
> 
> Rabin Vincent (1):
>   ubi: fastmap: Fix slab corruption
> 
> Richard Weinberger (20):
>   UBI: Fastmap: Fix memory leak while attaching
>   UBI: Remove alloc_ai() slab name from parameter list
>   UBI: Don't read back all data in ubi_eba_copy_leb()
>   UBI: Fix static volume checks when Fastmap is used
>   ubi: Make recover_peb power cut aware
>   ubi: Fix early logging
>   ubi: Make volume resize power cut aware
>   ubi: Fix scan_fast() comment
>   ubi: Introduce vol_ignored()
>   ubi: Fix whitespace issue in count_fastmap_pebs()
>   ubi: Rework Fastmap attach base code
>   ubi: Check whether the Fastmap anchor matches the super block
>   ubi: Be more paranoid while seaching for the most recent Fastmap
>   ubi: Deal with interrupted erasures in WL
>   ubi: Fix races around ubi_refill_pools()
>   ubi: Fix Fastmap's update_vol()
>   ubi: fastmap: Correctly handle interrupted erasures in EBA
>   ubi: fastmap: Check each mapping only once
>   ubi: fastmap: Detect EBA mismatches on-the-fly
>   ubi: Initialize Fastmap checkmapping correctly
> 
> Sascha Hauer (3):
>   ubi: fastmap: Erase outdated anchor PEBs during attach
>   ubi: Fastmap: Fix typo
>   ubi: Fix copy/paste error in function documentation
> 
> Sebastian Siewior (1):
>   ubi/upd: Always flush after prepared for an update
> 
> Uwe Kleine-König (1):
>   UBI: Fix two typos in comments
> 
> Wei Yongjun (1):
>   mtd: ubi: wl: Fix error return code in ubi_wl_init()
> 
> z00189512 (1):
>   UBI: Modify wrong comment in ubi_leb_map function.
> 
>  drivers/mtd/ubi/Kconfig      |   2 +-
>  drivers/mtd/ubi/attach.c     | 473 +++++++++++++++++---------
>  drivers/mtd/ubi/build.c      |  17 +-
>  drivers/mtd/ubi/eba.c        | 785 ++++++++++++++++++++++++++++---------------
>  drivers/mtd/ubi/fastmap-wl.c |   6 +-
>  drivers/mtd/ubi/fastmap.c    | 300 ++++++++++-------
>  drivers/mtd/ubi/io.c         |  41 ++-
>  drivers/mtd/ubi/kapi.c       |  18 +-
>  drivers/mtd/ubi/ubi-media.h  |  22 +-
>  drivers/mtd/ubi/ubi.h        | 194 +++++++++--
>  drivers/mtd/ubi/upd.c        |   8 +-
>  drivers/mtd/ubi/vmt.c        |  65 ++--
>  drivers/mtd/ubi/vtbl.c       |  32 +-
>  drivers/mtd/ubi/wl.c         | 161 +++++++--
>  drivers/mtd/ubi/wl.h         |   2 +-
>  15 files changed, 1419 insertions(+), 707 deletions(-)
> 
> -- 
> 2.7.4
> 
> 
> _______________________________________________
> barebox mailing list
> barebox@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/barebox

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox