Commit 9dedc73a authored by Vishal Verma's avatar Vishal Verma Committed by Dan Williams
Browse files

libnvdimm/btt: Fix LBA masking during 'free list' population

The Linux BTT implementation assumes that log entries will never have
the 'zero' flag set, and indeed it never sets that flag for log entries

However, the UEFI spec is ambiguous on the exact format of the LBA field
of a log entry, specifically as to whether it should include the
additional flag bits or not. While a zero bit doesn't make sense in the
context of a log entry, other BTT implementations might still have it set.

If an implementation does happen to have it set, we would happily read
it in as the next block to write to for writes. Since a high bit is set,
it pushes the block number out of the range of an 'arena', and we fail
such a write with an EIO.

Follow the robustness principle, and tolerate such implementations by
stripping out the zero flag when populating the free list during
initialization. Additionally, use the same stripped out entries for
detection of incomplete writes and map restoration that happens at this

Add a sysfs file 'log_zero_flags' that indicates the ability to accept
such a layout to userspace applications. This enables 'ndctl
check-namespace' to recognize whether the kernel is able to handle zero
flags, or whether it should attempt a fix-up under the --repair option.

Cc: Dan Williams <>
Reported-by: default avatarDexuan Cui <>
Reported-by: default avatarPedro d'Aquino Filocre F S Barbuda <>
Tested-by: default avatarDexuan Cui <>
Signed-off-by: default avatarVishal Verma <>
Signed-off-by: default avatarDan Williams <>
parent 2f8c9011
......@@ -542,8 +542,8 @@ static int arena_clear_freelist_error(struct arena_info *arena, u32 lane)
static int btt_freelist_init(struct arena_info *arena)
int new, ret;
u32 i, map_entry;
struct log_entry log_new;
u32 i, map_entry, log_oldmap, log_newmap;
arena->freelist = kcalloc(arena->nfree, sizeof(struct free_entry),
......@@ -555,16 +555,22 @@ static int btt_freelist_init(struct arena_info *arena)
if (new < 0)
return new;
/* old and new map entries with any flags stripped out */
log_oldmap = ent_lba(le32_to_cpu(log_new.old_map));
log_newmap = ent_lba(le32_to_cpu(log_new.new_map));
/* sub points to the next one to be overwritten */
arena->freelist[i].sub = 1 - new;
arena->freelist[i].seq = nd_inc_seq(le32_to_cpu(log_new.seq));
arena->freelist[i].block = le32_to_cpu(log_new.old_map);
arena->freelist[i].block = log_oldmap;
* FIXME: if error clearing fails during init, we want to make
* the BTT read-only
if (ent_e_flag(log_new.old_map)) {
if (ent_e_flag(log_new.old_map) &&
!ent_normal(log_new.old_map)) {
arena->freelist[i].has_err = 1;
ret = arena_clear_freelist_error(arena, i);
if (ret)
......@@ -572,7 +578,7 @@ static int btt_freelist_init(struct arena_info *arena)
/* This implies a newly created or untouched flog entry */
if (log_new.old_map == log_new.new_map)
if (log_oldmap == log_newmap)
/* Check if map recovery is needed */
......@@ -580,8 +586,15 @@ static int btt_freelist_init(struct arena_info *arena)
if (ret)
return ret;
if ((le32_to_cpu(log_new.new_map) != map_entry) &&
(le32_to_cpu(log_new.old_map) == map_entry)) {
* The map_entry from btt_read_map is stripped of any flag bits,
* so use the stripped out versions from the log as well for
* testing whether recovery is needed. For restoration, use the
* 'raw' version of the log entries as that captured what we
* were going to write originally.
if ((log_newmap != map_entry) && (log_oldmap == map_entry)) {
* Last transaction wrote the flog, but wasn't able
* to complete the map write. So fix up the map.
......@@ -44,6 +44,8 @@
#define ent_e_flag(ent) (!!(ent & MAP_ERR_MASK))
#define ent_z_flag(ent) (!!(ent & MAP_TRIM_MASK))
#define set_e_flag(ent) (ent |= MAP_ERR_MASK)
/* 'normal' is both e and z flags set */
#define ent_normal(ent) (ent_e_flag(ent) && ent_z_flag(ent))
enum btt_init_state {
......@@ -159,11 +159,19 @@ static ssize_t size_show(struct device *dev,
static DEVICE_ATTR_RO(size);
static ssize_t log_zero_flags_show(struct device *dev,
struct device_attribute *attr, char *buf)
return sprintf(buf, "Y\n");
static DEVICE_ATTR_RO(log_zero_flags);
static struct attribute *nd_btt_attributes[] = {
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment