btrfs: add fs state details to error messages.
PATCH 1/1 btrfs: add fs state details to error messages.
When a filesystem goes read-only due to an error, multiple errors tend to be reported, some of which are knock-on failures. Logging some fs_states, if any, in btrfs_handle_fs_error() and btrfs_printk() helps distinguish the first error from subsequent messages which may only exist due to an error state.
Under the new format, most initial errors will look like:BTRFS: error (device loop0) in ...
while subsequent errors will begin with:error (device loop0: state E) in ...
An initial transaction abort error will look likeerror (device loop0: state X) in ...
and subsequent messages will contain(device loop0: state EX) in ...
--- fs/btrfs/super.c | 49 +++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 40 insertions(+), 9 deletions(-)
fs/btrfs/super.c +40 -9
@@ -66,6 +66,31 @@ static struct file_system_type btrfs_root_fs_type;static int btrfs_remount (struct super_block *sb,int *flags,char *data); +#define STATE_STRING_PREFACE ": state " +#define MAX_STATE_CHARS 2 + +static void btrfs_state_to_string (const struct btrfs_fs_info *info,char *buf) +{ +unsigned long state = info->fs_state; +char *curr = buf; + +memcpy (curr,STATE_STRING_PREFACE ,sizeof (STATE_STRING_PREFACE )); + curr +=sizeof (STATE_STRING_PREFACE ) -1 ; + +/* If more states are reported, update MAX_STATE_CHARS also */ +if (test_and_clear_bit (BTRFS_FS_STATE_ERROR , &state)) + *curr++ ='E' ; + +if (test_and_clear_bit (BTRFS_FS_STATE_TRANS_ABORTED , &state)) + *curr++ ='X' ; + +/* If no states were printed, reset the buffer */ +if (state == info->fs_state) + curr = buf; + + *curr++ ='\0' ; +} +/* * Generally the error codes correspond to their respective errors, but there * are a few special cases.
@@ -128,6 +153,7 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function {struct super_block *sb = fs_info->sb;#ifdef CONFIG_PRINTK +char statestr[sizeof (STATE_STRING_PREFACE ) +MAX_STATE_CHARS ];const char *errstr;#endif
@@ -136,10 +162,11 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function* under SB_RDONLY, then it is safe here. */ if (errno == -EROFS &&sb_rdonly (sb)) -return ; +return ;#ifdef CONFIG_PRINTK errstr =btrfs_decode_error (errno); +btrfs_state_to_string (fs_info, statestr);if (fmt) {struct va_format vaf;va_list args;
@@ -148,12 +175,12 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function vaf.fmt = fmt; vaf.va = &args; -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s (%pV)\n" , - sb->s_id, function, line, errno, errstr, &vaf); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s (%pV)\n" , + sb->s_id, statestr, function, line, errno, errstr, &vaf); va_end(args); }else { -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s\n" , - sb->s_id, function, line, errno, errstr); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s\n" , + sb->s_id, statestr, function, line, errno, errstr); }#endif
@@ -240,11 +267,15 @@ void __cold btrfs_printk(const struct btrfs_fs_info *fs_info, const char *fmt, . vaf.va = &args;if (__ratelimit (ratelimit)) { -if (fs_info) -printk ("%sBTRFS %s (device %s): %pV\n" , lvl, type, - fs_info->sb->s_id, &vaf); -else +if (fs_info) { +char statestr[sizeof (STATE_STRING_PREFACE ) +MAX_STATE_CHARS ]; + +btrfs_state_to_string (fs_info, statestr); +printk ("%sBTRFS %s (device %s%s): %pV\n" , lvl, type, + fs_info->sb->s_id, statestr, &vaf); + }else {printk ("%sBTRFS %s: %pV\n" , lvl, type, &vaf); + } }va_end (args);
Signature · 1 line
2.30.2
Quoted 32 lines · Sweet Tea Dorminy
When a filesystem goes read-only due to an error, multiple errors tend to be reported, some of which are knock-on failures. Logging some
fs_states, if any, inbtrfs_handle_fs_error()andbtrfs_printk()helps distinguish the first error from subsequent messages which may only exist due to an error state.Under the new format, most initial errors will look like:
BTRFS: error (device loop0) in ...
while subsequent errors will begin with:error (device loop0: state E) in ...An initial transaction abort error will look like
error (device loop0: state X) in ...
and subsequent messages will contain(device loop0: state EX) in ...Signed-off-by Sweet Tea Dorminy <[email protected]>--- fs/btrfs/super.c | 49 +++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 40 insertions(+), 9 deletions(-)fs/btrfs/super.c +3
-0@@ -66,6 +66,31 @@ static struct file_system_type btrfs_root_fs_type;static int btrfs_remount (struct super_block *sb,int *flags,char *data); +#define STATE_STRING_PREFACE ": state " +#define MAX_STATE_CHARS 2 +
fs/btrfs/super.c static struct file_system_type btrfs_root_fs_type; · line 81
Quoted 10 lines
fs/btrfs/super.c +10
-0+static void btrfs_state_to_string (const struct btrfs_fs_info *info,char *buf) +{ +unsigned long state = info->fs_state; +char *curr = buf; + +memcpy (curr,STATE_STRING_PREFACE ,sizeof (STATE_STRING_PREFACE )); + curr +=sizeof (STATE_STRING_PREFACE ) -1 ; + +/* If more states are reported, update MAX_STATE_CHARS also */ +if (test_and_clear_bit (BTRFS_FS_STATE_ERROR , &state))
The state is supposed to be sticky, so can't be cleared. Also as I read the suggested change, the "state: " should be visible for all messages that are printed after the filesystem state changes.
Quoted 24 lines
fs/btrfs/super.c +13
-0+ *curr++ ='E' ; + +if (test_and_clear_bit (BTRFS_FS_STATE_TRANS_ABORTED , &state)) + *curr++ ='X' ; + +/* If no states were printed, reset the buffer */ +if (state == info->fs_state) + curr = buf; + + *curr++ ='\0' ; +} +/* * Generally the error codes correspond to their respective errors, but there * are a few special cases. @@ -128,6 +153,7 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function {struct super_block *sb = fs_info->sb;#ifdef CONFIG_PRINTK +char statestr[sizeof (STATE_STRING_PREFACE ) +MAX_STATE_CHARS ];const char *errstr;#endif
fs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 165
Quoted 5 lines
fs/btrfs/super.c +1
-1@@ -136,10 +162,11 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function* under SB_RDONLY, then it is safe here. */ if (errno == -EROFS &&sb_rdonly (sb)) -return ; +return ;
Unnecessary change.
Quoted 8 lines
fs/btrfs/super.c +1
-0#ifdef CONFIG_PRINTK errstr =btrfs_decode_error (errno); +btrfs_state_to_string (fs_info, statestr);if (fmt) {struct va_format vaf;va_list args;
fs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 179
Quoted 7 lines
fs/btrfs/super.c +2
-2@@ -148,12 +175,12 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function vaf.fmt = fmt; vaf.va = &args; -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s (%pV)\n" , - sb->s_id, function, line, errno, errstr, &vaf); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s (%pV)\n" , + sb->s_id, statestr, function, line, errno, errstr, &vaf);
Alternatively the state message can be built into the message itself so it does not require the extra array.
pr_crit("btrfs: error(device %s%s%s%s) ...",
sb->s_id,
statebits ? ", state" : "",
test_bit(FS_ERRROR) ? "E" : "",
test_bit(TRANS_ABORT) ? "T" : "", ...);This is the idea, final code can use some wrappers around the bit constatnts.
fs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 183
Quoted 6 lines
fs/btrfs/super.c +2
-2va_end (args); }else { -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s\n" , - sb->s_id, function, line, errno, errstr); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s\n" , + sb->s_id, statestr, function, line, errno, errstr);
Filling the temporary array makes sense as it's used twice, however I'm not sure if the 'else' branch is ever executed.
Quoted 16 lines
fs/btrfs/super.c static struct file_system_type btrfs_root_fs_type; · line 81
Quoted 12 lines
+static void
btrfs_state_to_string(const struct btrfs_fs_info *info, char *buf)fs/btrfs/super.c +9
-0+{ +unsigned long state = info->fs_state; +char *curr = buf; + +memcpy (curr,STATE_STRING_PREFACE ,sizeof (STATE_STRING_PREFACE )); + curr +=sizeof (STATE_STRING_PREFACE ) -1 ; + +/* If more states are reported, update MAX_STATE_CHARS also */ +if (test_and_clear_bit (BTRFS_FS_STATE_ERROR , &state))The state is supposed to be sticky, so can't be cleared. Also as I read the suggested change, the "state: " should be visible for all messages that are printed after the filesystem state changes.
'state' is a local copy of info->fs_state, so clearing bits on the local copy should be okay, and the "state: " will be present for everything after the fs state changes. Could instead use test_bit(&info->fs_state) and keep a count of how many states were printed (to know if we need to reset the buffer) if that is clearer.
Quoted 26 lines
Quoted 26 lines
fs/btrfs/super.c +12
-0+ *curr++ ='E' ; + +if (test_and_clear_bit (BTRFS_FS_STATE_TRANS_ABORTED , &state)) + *curr++ ='X' ; + +/* If no states were printed, reset the buffer */ +if (state == info->fs_state) + curr = buf; + + *curr++ ='\0' ; +} +/* * Generally the error codes correspond to their respective errors, but there
* are a few special cases.fs/btrfs/super.c +1
-0@@ -128,6 +153,7 @@ void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function {struct super_block *sb = fs_info->sb;#ifdef CONFIG_PRINTK +char statestr[sizeof (STATE_STRING_PREFACE ) +MAX_STATE_CHARS ];const char *errstr;#endif
Quoted 9 lines
fs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 165
Quoted 7 lines
(quoted diff) +0
-0@@ -136,10 +162,11 @@ void __btrfs_handle_fs_error(struct
btrfs_fs_info*fs_info, const char *functionfs/btrfs/super.c +1
-1* under SB_RDONLY, then it is safe here. */ if (errno == -EROFS &&sb_rdonly (sb)) -return ; +return ;Unnecessary change.
Yeah, there's a stray space at the start of that line, but I can take out fixing it.
Quoted 32 lines
Quoted 10 lines
#ifdef
CONFIG_PRINTKfs/btrfs/super.c +1
-0errstr =btrfs_decode_error (errno); +btrfs_state_to_string (fs_info, statestr);if (fmt) {struct va_format vaf;va_list args;@@ -148,12 +175,12 @@ void __btrfs_handle_fs_error(struct
btrfs_fs_info*fs_info, const char *functionfs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 179
Quoted 7 lines
fs/btrfs/super.c +2
-2vaf.fmt = fmt; vaf.va = &args; -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s (%pV)\n" , - sb->s_id, function, line, errno, errstr, &vaf); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s (%pV)\n" , + sb->s_id, statestr, function, line, errno, errstr, &vaf);Alternatively the state message can be built into the message itself so it does not require the extra array.
pr_crit("btrfs: error(device %s%s%s%s) ...", sb->s_id, statebits ? ", state" : "", test_bit(FS_ERRROR) ? "E" : "", test_bit(TRANS_ABORT) ? "T" : "", ...);This is the idea, final code can use some wrappers around the bit constatnts.
Quoted 9 lines
fs/btrfs/super.c void __btrfs_handle_fs_error(struct btrfs_fs_info *fs_info, const char *function · line 183
Quoted 6 lines
fs/btrfs/super.c +2
-2va_end (args); }else { -pr_crit ("BTRFS: error (device %s) in %s:%d: errno=%d %s\n" , - sb->s_id, function, line, errno, errstr); +pr_crit ("BTRFS: error (device %s%s) in %s:%d: errno=%d %s\n" , + sb->s_id, statestr, function, line, errno, errstr);Filling the temporary array makes sense as it's used twice, however I'm not sure if the 'else' branch is ever executed.
There are a bunch of calls with NULL format -> else branch, unfortunately.
Thanks!