Skip to content
Snippets Groups Projects
Commit b1db125e authored by girish's avatar girish
Browse files

JBD: Fix NULL pointer bh->b_data on NUMA box with journal checksumming. 

Current journal checksumming patch failed fsstress test on NUMA. The
bh->b_data passed to the crc32_be () function could be NULL pointer,
which caused kernel oops immediately when running fsstress with -o
journal_checksum. It is because the page is part of highmem on NUMA box.
We need to kmap the page before access the bh->b_data to calculate
the checksums.
parent df6b10a5
No related branches found
No related tags found
No related merge requests found
Hide whitespace changes
Inline Side-by-side
lustre/kernel_patches/patches/jbd-journal-chksum-2.6-sles10.patch
+ 45
− 35
View file @ b1db125e
Index: linux-2.6.16.46-0.14/fs/jbd/commit.c Index: linux-2.6.16.53-0.16/fs/jbd/commit.c
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/fs/jbd/commit.c --- linux-2.6.16.53-0.16.orig/fs/jbd/commit.c
+++ linux-2.6.16.46-0.14/fs/jbd/commit.c +++ linux-2.6.16.53-0.16/fs/jbd/commit.c
@@ -22,6 +22,7 @@ @@ -22,6 +22,7 @@
#include <linux/pagemap.h> #include <linux/pagemap.h>
#include <linux/smp_lock.h> #include <linux/smp_lock.h>
...@@ -82,7 +82,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -82,7 +82,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
/* is it possible for another commit to fail at roughly /* is it possible for another commit to fail at roughly
* the same time as this one? If so, we don't want to * the same time as this one? If so, we don't want to
* trust the barrier flag in the super, but instead want * trust the barrier flag in the super, but instead want
@@ -153,15 +171,74 @@ static int journal_write_commit_record(j @@ -153,12 +171,84 @@ static int journal_write_commit_record(j
clear_buffer_ordered(bh); clear_buffer_ordered(bh);
set_buffer_uptodate(bh); set_buffer_uptodate(bh);
set_buffer_dirty(bh); set_buffer_dirty(bh);
...@@ -113,9 +113,9 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -113,9 +113,9 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
+ journal_put_journal_head(bh2jh(bh)); + journal_put_journal_head(bh2jh(bh));
+ +
+ return ret; + return ret;
} +}
+
/* +/*
+ * Wait for all submitted IO to complete. + * Wait for all submitted IO to complete.
+ */ + */
+static int journal_wait_on_locked_list(journal_t *journal, +static int journal_wait_on_locked_list(journal_t *journal,
...@@ -156,12 +156,22 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -156,12 +156,22 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
+ return ret; + return ret;
+} +}
+ +
+ +static inline __u32 jbd_checksum_data(__u32 crc32_sum, struct buffer_head *bh)
+/* +{
* journal_commit_transaction + struct page *page = bh->b_page;
* + char *addr;
* The primary function for committing a transaction to the log. This + __u32 checksum;
@@ -184,6 +261,8 @@ void journal_commit_transaction(journal_ +
+ addr = kmap(page);
+ checksum = crc32_be(crc32_sum,
+ (void *)(addr + offset_in_page(bh->b_data)),
+ bh->b_size);
+ kunmap(page);
+ return checksum;
}
/*
@@ -184,6 +274,8 @@ void journal_commit_transaction(journal_
int first_tag = 0; int first_tag = 0;
int tag_flag; int tag_flag;
int i; int i;
...@@ -170,7 +180,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -170,7 +180,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
/* /*
* First job: lock down the current transaction and wait for * First job: lock down the current transaction and wait for
@@ -395,37 +474,14 @@ write_out_data: @@ -395,37 +487,14 @@ write_out_data:
} }
/* /*
...@@ -214,7 +224,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -214,7 +224,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
spin_unlock(&journal->j_list_lock); spin_unlock(&journal->j_list_lock);
if (err) if (err)
@@ -598,6 +654,16 @@ write_out_data: @@ -598,6 +667,16 @@ write_out_data:
start_journal_io: start_journal_io:
for (i = 0; i < bufs; i++) { for (i = 0; i < bufs; i++) {
struct buffer_head *bh = wbuf[i]; struct buffer_head *bh = wbuf[i];
...@@ -223,15 +233,15 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -223,15 +233,15 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
+ */ + */
+ if (JFS_HAS_COMPAT_FEATURE(journal, + if (JFS_HAS_COMPAT_FEATURE(journal,
+ JFS_FEATURE_COMPAT_CHECKSUM)) { + JFS_FEATURE_COMPAT_CHECKSUM)) {
+ crc32_sum = crc32_be(crc32_sum, + crc32_sum =
+ (void *)bh->b_data, + jbd_checksum_data(crc32_sum,
+ bh->b_size); + bh);
+ } + }
+ +
lock_buffer(bh); lock_buffer(bh);
clear_buffer_dirty(bh); clear_buffer_dirty(bh);
set_buffer_uptodate(bh); set_buffer_uptodate(bh);
@@ -614,6 +680,23 @@ start_journal_io: @@ -614,6 +693,23 @@ start_journal_io:
} }
} }
...@@ -255,7 +265,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -255,7 +265,7 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
/* Lo and behold: we have just managed to send a transaction to /* Lo and behold: we have just managed to send a transaction to
the log. Before we can commit it, wait for the IO so far to the log. Before we can commit it, wait for the IO so far to
complete. Control buffers being written are on the complete. Control buffers being written are on the
@@ -712,9 +795,15 @@ wait_for_iobuf: @@ -712,9 +808,15 @@ wait_for_iobuf:
} }
jbd_debug(3, "JBD: commit phase 6\n"); jbd_debug(3, "JBD: commit phase 6\n");
...@@ -274,10 +284,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c ...@@ -274,10 +284,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/commit.c
if (err) if (err)
__journal_abort_hard(journal); __journal_abort_hard(journal);
Index: linux-2.6.16.46-0.14/include/linux/jbd.h Index: linux-2.6.16.53-0.16/include/linux/jbd.h
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/include/linux/jbd.h --- linux-2.6.16.53-0.16.orig/include/linux/jbd.h
+++ linux-2.6.16.46-0.14/include/linux/jbd.h +++ linux-2.6.16.53-0.16/include/linux/jbd.h
@@ -142,6 +142,29 @@ typedef struct journal_header_s @@ -142,6 +142,29 @@ typedef struct journal_header_s
__be32 h_sequence; __be32 h_sequence;
} journal_header_t; } journal_header_t;
...@@ -337,10 +347,10 @@ Index: linux-2.6.16.46-0.14/include/linux/jbd.h ...@@ -337,10 +347,10 @@ Index: linux-2.6.16.46-0.14/include/linux/jbd.h
extern int journal_create (journal_t *); extern int journal_create (journal_t *);
extern int journal_load (journal_t *journal); extern int journal_load (journal_t *journal);
extern void journal_destroy (journal_t *); extern void journal_destroy (journal_t *);
Index: linux-2.6.16.46-0.14/fs/jbd/recovery.c Index: linux-2.6.16.53-0.16/fs/jbd/recovery.c
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/fs/jbd/recovery.c --- linux-2.6.16.53-0.16.orig/fs/jbd/recovery.c
+++ linux-2.6.16.46-0.14/fs/jbd/recovery.c +++ linux-2.6.16.53-0.16/fs/jbd/recovery.c
@@ -21,6 +21,7 @@ @@ -21,6 +21,7 @@
#include <linux/jbd.h> #include <linux/jbd.h>
#include <linux/errno.h> #include <linux/errno.h>
...@@ -536,10 +546,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/recovery.c ...@@ -536,10 +546,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/recovery.c
/* It's really bad news if different passes end up at /* It's really bad news if different passes end up at
* different places (but possible due to IO errors). */ * different places (but possible due to IO errors). */
if (info->end_transaction != next_commit_ID) { if (info->end_transaction != next_commit_ID) {
Index: linux-2.6.16.46-0.14/fs/jbd/journal.c Index: linux-2.6.16.53-0.16/fs/jbd/journal.c
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/fs/jbd/journal.c --- linux-2.6.16.53-0.16.orig/fs/jbd/journal.c
+++ linux-2.6.16.46-0.14/fs/jbd/journal.c +++ linux-2.6.16.53-0.16/fs/jbd/journal.c
@@ -64,6 +64,7 @@ EXPORT_SYMBOL(journal_update_format); @@ -64,6 +64,7 @@ EXPORT_SYMBOL(journal_update_format);
EXPORT_SYMBOL(journal_check_used_features); EXPORT_SYMBOL(journal_check_used_features);
EXPORT_SYMBOL(journal_check_available_features); EXPORT_SYMBOL(journal_check_available_features);
...@@ -582,10 +592,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/journal.c ...@@ -582,10 +592,10 @@ Index: linux-2.6.16.46-0.14/fs/jbd/journal.c
/** /**
* int journal_update_format () - Update on-disk journal structure. * int journal_update_format () - Update on-disk journal structure.
Index: linux-2.6.16.46-0.14/fs/Kconfig Index: linux-2.6.16.53-0.16/fs/Kconfig
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/fs/Kconfig --- linux-2.6.16.53-0.16.orig/fs/Kconfig
+++ linux-2.6.16.46-0.14/fs/Kconfig +++ linux-2.6.16.53-0.16/fs/Kconfig
@@ -140,6 +140,7 @@ config EXT3_FS_SECURITY @@ -140,6 +140,7 @@ config EXT3_FS_SECURITY
config JBD config JBD
...@@ -594,10 +604,10 @@ Index: linux-2.6.16.46-0.14/fs/Kconfig ...@@ -594,10 +604,10 @@ Index: linux-2.6.16.46-0.14/fs/Kconfig
help help
This is a generic journaling layer for block devices. It is This is a generic journaling layer for block devices. It is
currently used by the ext3 and OCFS2 file systems, but it could currently used by the ext3 and OCFS2 file systems, but it could
Index: linux-2.6.16.46-0.14/Documentation/filesystems/ext3.txt Index: linux-2.6.16.53-0.16/Documentation/filesystems/ext3.txt
=================================================================== ===================================================================
--- linux-2.6.16.46-0.14.orig/Documentation/filesystems/ext3.txt --- linux-2.6.16.53-0.16.orig/Documentation/filesystems/ext3.txt
+++ linux-2.6.16.46-0.14/Documentation/filesystems/ext3.txt +++ linux-2.6.16.53-0.16/Documentation/filesystems/ext3.txt
@@ -14,6 +14,16 @@ Options @@ -14,6 +14,16 @@ Options
When mounting an ext3 filesystem, the following option are accepted: When mounting an ext3 filesystem, the following option are accepted:
(*) == default (*) == default
......
lustre/kernel_patches/patches/jbd-journal-chksum-2.6.18-vanilla.patch
+ 61
− 42
View file @ b1db125e
Index: linux-2.6.18-8.1.8/fs/jbd/commit.c Index: linux-2.6.18.8/fs/jbd/commit.c
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/fs/jbd/commit.c --- linux-2.6.18.8.orig/fs/jbd/commit.c
+++ linux-2.6.18-8.1.8/fs/jbd/commit.c +++ linux-2.6.18.8/fs/jbd/commit.c
@@ -21,6 +21,7 @@ @@ -22,6 +22,7 @@
#include <linux/mm.h> #include <linux/mm.h>
#include <linux/pagemap.h> #include <linux/pagemap.h>
#include <linux/smp_lock.h> #include <linux/smp_lock.h>
+#include <linux/crc32.h> +#include <linux/crc32.h>
/* /*
* Default IO end handler for temporary BJ_IO buffer_heads. @@ -95,19 +96,23 @@ static int inverted_lock(journal_t *jour
@@ -93,19 +94,23 @@ static int inverted_lock(journal_t *jour
return 1; return 1;
} }
...@@ -38,7 +38,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -38,7 +38,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
int barrier_done = 0; int barrier_done = 0;
if (is_journal_aborted(journal)) if (is_journal_aborted(journal))
@@ -117,21 +122,34 @@ static int journal_write_commit_record(j @@ -119,21 +124,34 @@ static int journal_write_commit_record(j
bh = jh2bh(descriptor); bh = jh2bh(descriptor);
...@@ -82,7 +82,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -82,7 +82,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
/* is it possible for another commit to fail at roughly /* is it possible for another commit to fail at roughly
* the same time as this one? If so, we don't want to * the same time as this one? If so, we don't want to
* trust the barrier flag in the super, but instead want * trust the barrier flag in the super, but instead want
@@ -152,14 +170,72 @@ static int journal_write_commit_record(j @@ -154,12 +172,70 @@ static int journal_write_commit_record(j
clear_buffer_ordered(bh); clear_buffer_ordered(bh);
set_buffer_uptodate(bh); set_buffer_uptodate(bh);
set_buffer_dirty(bh); set_buffer_dirty(bh);
...@@ -94,8 +94,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -94,8 +94,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
+ *cbh = bh; + *cbh = bh;
+ return ret; + return ret;
+} +}
+
- return (ret == -EIO);
+/* +/*
+ * This function along with journal_submit_commit_record + * This function along with journal_submit_commit_record
+ * allows to write the commit record asynchronously. + * allows to write the commit record asynchronously.
...@@ -113,8 +112,8 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -113,8 +112,8 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
+ journal_put_journal_head(bh2jh(bh)); + journal_put_journal_head(bh2jh(bh));
+ +
+ return ret; + return ret;
} +}
+
+/* +/*
+ * Wait for all submitted IO to complete. + * Wait for all submitted IO to complete.
+ */ + */
...@@ -123,7 +122,8 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -123,7 +122,8 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
+{ +{
+ int ret = 0; + int ret = 0;
+ struct journal_head *jh; + struct journal_head *jh;
+
- return (ret == -EIO);
+ while (commit_transaction->t_locked_list) { + while (commit_transaction->t_locked_list) {
+ struct buffer_head *bh; + struct buffer_head *bh;
+ +
...@@ -154,12 +154,31 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -154,12 +154,31 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
+ cond_resched_lock(&journal->j_list_lock); + cond_resched_lock(&journal->j_list_lock);
+ } + }
+ return ret; + return ret;
}
void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
@@ -273,6 +349,20 @@ write_out_data:
journal_do_submit_data(wbuf, bufs);
}
+static inline __u32 jbd_checksum_data(__u32 crc32_sum, struct buffer_head *bh)
+{
+ struct page *page = bh->b_page;
+ char *addr;
+ __u32 checksum;
+
+ addr = kmap(page);
+ checksum = crc32_be(crc32_sum,
+ (void *)(addr + offset_in_page(bh->b_data)),
+ bh->b_size);
+ kunmap(page);
+ return checksum;
+} +}
+ +
void journal_do_submit_data(struct buffer_head **wbuf, int bufs) /*
{ * journal_commit_transaction
int i; *
@@ -293,6 +369,8 @@ void journal_commit_transaction(journal_ @@ -296,6 +386,8 @@ void journal_commit_transaction(journal_
int first_tag = 0; int first_tag = 0;
int tag_flag; int tag_flag;
int i; int i;
...@@ -168,7 +187,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -168,7 +187,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
/* /*
* First job: lock down the current transaction and wait for * First job: lock down the current transaction and wait for
@@ -428,38 +506,14 @@ void journal_commit_transaction(journal_ @@ -439,38 +531,14 @@ void journal_commit_transaction(journal_
journal_submit_data_buffers(journal, commit_transaction); journal_submit_data_buffers(journal, commit_transaction);
/* /*
...@@ -213,7 +232,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -213,7 +232,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
spin_unlock(&journal->j_list_lock); spin_unlock(&journal->j_list_lock);
if (err) if (err)
@@ -627,6 +681,16 @@ void journal_commit_transaction(journal_ @@ -643,6 +711,16 @@ void journal_commit_transaction(journal_
start_journal_io: start_journal_io:
for (i = 0; i < bufs; i++) { for (i = 0; i < bufs; i++) {
struct buffer_head *bh = wbuf[i]; struct buffer_head *bh = wbuf[i];
...@@ -222,15 +241,15 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -222,15 +241,15 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
+ */ + */
+ if (JFS_HAS_COMPAT_FEATURE(journal, + if (JFS_HAS_COMPAT_FEATURE(journal,
+ JFS_FEATURE_COMPAT_CHECKSUM)) { + JFS_FEATURE_COMPAT_CHECKSUM)) {
+ crc32_sum = crc32_be(crc32_sum, + crc32_sum =
+ (void *)bh->b_data, + jbd_checksum_data(crc32_sum,
+ bh->b_size); + bh);
+ } + }
+ +
lock_buffer(bh); lock_buffer(bh);
clear_buffer_dirty(bh); clear_buffer_dirty(bh);
set_buffer_uptodate(bh); set_buffer_uptodate(bh);
@@ -642,6 +706,23 @@ start_journal_io: @@ -659,6 +737,23 @@ start_journal_io:
} }
} }
...@@ -254,7 +273,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -254,7 +273,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
/* Lo and behold: we have just managed to send a transaction to /* Lo and behold: we have just managed to send a transaction to
the log. Before we can commit it, wait for the IO so far to the log. Before we can commit it, wait for the IO so far to
complete. Control buffers being written are on the complete. Control buffers being written are on the
@@ -740,9 +821,15 @@ wait_for_iobuf: @@ -757,9 +852,15 @@ wait_for_iobuf:
} }
jbd_debug(3, "JBD: commit phase 6\n"); jbd_debug(3, "JBD: commit phase 6\n");
...@@ -273,10 +292,10 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c ...@@ -273,10 +292,10 @@ Index: linux-2.6.18-8.1.8/fs/jbd/commit.c
if (err) if (err)
__journal_abort_hard(journal); __journal_abort_hard(journal);
Index: linux-2.6.18-8.1.8/include/linux/jbd.h Index: linux-2.6.18.8/include/linux/jbd.h
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/include/linux/jbd.h --- linux-2.6.18.8.orig/include/linux/jbd.h
+++ linux-2.6.18-8.1.8/include/linux/jbd.h +++ linux-2.6.18.8/include/linux/jbd.h
@@ -148,6 +148,29 @@ typedef struct journal_header_s @@ -148,6 +148,29 @@ typedef struct journal_header_s
__be32 h_sequence; __be32 h_sequence;
} journal_header_t; } journal_header_t;
...@@ -327,7 +346,7 @@ Index: linux-2.6.18-8.1.8/include/linux/jbd.h ...@@ -327,7 +346,7 @@ Index: linux-2.6.18-8.1.8/include/linux/jbd.h
#ifdef __KERNEL__ #ifdef __KERNEL__
@@ -967,6 +994,8 @@ extern int journal_check_available_fe @@ -1053,6 +1080,8 @@ extern int journal_check_available_fe
(journal_t *, unsigned long, unsigned long, unsigned long); (journal_t *, unsigned long, unsigned long, unsigned long);
extern int journal_set_features extern int journal_set_features
(journal_t *, unsigned long, unsigned long, unsigned long); (journal_t *, unsigned long, unsigned long, unsigned long);
...@@ -336,10 +355,10 @@ Index: linux-2.6.18-8.1.8/include/linux/jbd.h ...@@ -336,10 +355,10 @@ Index: linux-2.6.18-8.1.8/include/linux/jbd.h
extern int journal_create (journal_t *); extern int journal_create (journal_t *);
extern int journal_load (journal_t *journal); extern int journal_load (journal_t *journal);
extern void journal_destroy (journal_t *); extern void journal_destroy (journal_t *);
Index: linux-2.6.18-8.1.8/fs/jbd/recovery.c Index: linux-2.6.18.8/fs/jbd/recovery.c
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/fs/jbd/recovery.c --- linux-2.6.18.8.orig/fs/jbd/recovery.c
+++ linux-2.6.18-8.1.8/fs/jbd/recovery.c +++ linux-2.6.18.8/fs/jbd/recovery.c
@@ -21,6 +21,7 @@ @@ -21,6 +21,7 @@
#include <linux/jbd.h> #include <linux/jbd.h>
#include <linux/errno.h> #include <linux/errno.h>
...@@ -535,11 +554,11 @@ Index: linux-2.6.18-8.1.8/fs/jbd/recovery.c ...@@ -535,11 +554,11 @@ Index: linux-2.6.18-8.1.8/fs/jbd/recovery.c
/* It's really bad news if different passes end up at /* It's really bad news if different passes end up at
* different places (but possible due to IO errors). */ * different places (but possible due to IO errors). */
if (info->end_transaction != next_commit_ID) { if (info->end_transaction != next_commit_ID) {
Index: linux-2.6.18-8.1.8/fs/jbd/journal.c Index: linux-2.6.18.8/fs/jbd/journal.c
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/fs/jbd/journal.c --- linux-2.6.18.8.orig/fs/jbd/journal.c
+++ linux-2.6.18-8.1.8/fs/jbd/journal.c +++ linux-2.6.18.8/fs/jbd/journal.c
@@ -66,6 +66,7 @@ EXPORT_SYMBOL(journal_update_format); @@ -67,6 +67,7 @@ EXPORT_SYMBOL(journal_update_format);
EXPORT_SYMBOL(journal_check_used_features); EXPORT_SYMBOL(journal_check_used_features);
EXPORT_SYMBOL(journal_check_available_features); EXPORT_SYMBOL(journal_check_available_features);
EXPORT_SYMBOL(journal_set_features); EXPORT_SYMBOL(journal_set_features);
...@@ -547,7 +566,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/journal.c ...@@ -547,7 +566,7 @@ Index: linux-2.6.18-8.1.8/fs/jbd/journal.c
EXPORT_SYMBOL(journal_create); EXPORT_SYMBOL(journal_create);
EXPORT_SYMBOL(journal_load); EXPORT_SYMBOL(journal_load);
EXPORT_SYMBOL(journal_destroy); EXPORT_SYMBOL(journal_destroy);
@@ -1271,6 +1272,33 @@ int journal_set_features (journal_t *jou @@ -1573,6 +1574,33 @@ int journal_set_features (journal_t *jou
return 1; return 1;
} }
...@@ -581,10 +600,10 @@ Index: linux-2.6.18-8.1.8/fs/jbd/journal.c ...@@ -581,10 +600,10 @@ Index: linux-2.6.18-8.1.8/fs/jbd/journal.c
/** /**
* int journal_update_format () - Update on-disk journal structure. * int journal_update_format () - Update on-disk journal structure.
Index: linux-2.6.18-8.1.8/fs/Kconfig Index: linux-2.6.18.8/fs/Kconfig
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/fs/Kconfig --- linux-2.6.18.8.orig/fs/Kconfig
+++ linux-2.6.18-8.1.8/fs/Kconfig +++ linux-2.6.18.8/fs/Kconfig
@@ -140,6 +140,7 @@ config EXT3_FS_SECURITY @@ -140,6 +140,7 @@ config EXT3_FS_SECURITY
config JBD config JBD
...@@ -593,10 +612,10 @@ Index: linux-2.6.18-8.1.8/fs/Kconfig ...@@ -593,10 +612,10 @@ Index: linux-2.6.18-8.1.8/fs/Kconfig
help help
This is a generic journaling layer for block devices. It is This is a generic journaling layer for block devices. It is
currently used by the ext3 and OCFS2 file systems, but it could currently used by the ext3 and OCFS2 file systems, but it could
Index: linux-2.6.18-8.1.8/Documentation/filesystems/ext3.txt Index: linux-2.6.18.8/Documentation/filesystems/ext3.txt
=================================================================== ===================================================================
--- linux-2.6.18-8.1.8.orig/Documentation/filesystems/ext3.txt --- linux-2.6.18.8.orig/Documentation/filesystems/ext3.txt
+++ linux-2.6.18-8.1.8/Documentation/filesystems/ext3.txt +++ linux-2.6.18.8/Documentation/filesystems/ext3.txt
@@ -14,6 +14,16 @@ Options @@ -14,6 +14,16 @@ Options
When mounting an ext3 filesystem, the following option are accepted: When mounting an ext3 filesystem, the following option are accepted:
(*) == default (*) == default
......
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

https://git.gsi.de is provided by CIT→Linux&Web | GSI Helmholtzzentrum fuer Schwerionenforschung GmbH | Imprint (in German) | Privacy policy