erofs-utils: lib/diskbuf: fix MT data race in erofs_diskbuf_reserve()
PATCH 1/1 erofs-utils: lib/diskbuf: fix MT data race in erofs_diskbuf_reserve()
Two threads calling erofs_diskbuf_reserve() concurrently can both observe strm->locked == false and both advance tailoffset, silently producing an incorrect offset in the output image. Add a pthread_mutex_t to serialize access.
The commit path also gains the missing strm->locked = false reset; without it, locked was set once and never cleared, making the flag a latch rather than a guard.
libpthread is already a build requirement (-lpthread in lib/Makefile.am) and pthread_mutex_t is used identically in lib/workqueue.c and lib/compress.c, so no build system changes are required.
--- lib/diskbuf.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-)
lib/diskbuf.c +15 -3
@@ -2,6 +2,7 @@#include "erofs/diskbuf.h" #include "erofs/internal.h" #include "erofs/print.h" +#include <pthread.h> #include <stdio.h> #include <errno.h> #include <sys/stat.h>
@@ -15,6 +16,7 @@ static struct erofs_diskbufstrm {int fd;unsigned int alignsize;bool locked; +pthread_mutex_t lock; } *dbufstrm;int erofs_diskbuf_getfd (struct erofs_diskbuf *db,u64 *fpos)
@@ -34,15 +36,20 @@ int erofs_diskbuf_reserve(struct erofs_diskbuf *db, int sid, u64 *off) {struct erofs_diskbufstrm *strm = dbufstrm + sid; -if (strm->tailoffset & (strm->alignsize -1 )) { - strm->tailoffset =round_up (strm->tailoffset, strm->alignsize); +pthread_mutex_lock (&strm->lock); +if (strm->locked) { +pthread_mutex_unlock (&strm->lock); +return -EBUSY ; } +if (strm->tailoffset & (strm->alignsize -1 )) + strm->tailoffset =round_up (strm->tailoffset, strm->alignsize); db->offset = strm->tailoffset;if (off) *off = db->offset + strm->devpos; db->sp = strm; (void )erofs_atomic_inc_return (&strm->count); - strm->locked = true;/* TODO: need a real lock for MT */ + strm->locked = true; +pthread_mutex_unlock (&strm->lock);return strm->fd; }
@@ -51,9 +58,12 @@ void erofs_diskbuf_commit(struct erofs_diskbuf *db, u64 len)struct erofs_diskbufstrm *strm = db->sp;DBG_BUGON (!strm); +pthread_mutex_lock (&strm->lock);DBG_BUGON (!strm->locked);DBG_BUGON (strm->tailoffset != db->offset); strm->tailoffset += len; + strm->locked = false; +pthread_mutex_unlock (&strm->lock); }void erofs_diskbuf_close (struct erofs_diskbuf *db)
@@ -115,6 +125,7 @@ int erofs_diskbuf_init(unsigned int nstrms) setupone: strm->tailoffset =0 ;erofs_atomic_set (&strm->count,1 ); +pthread_mutex_init (&strm->lock,NULL );if (fstat (strm->fd, &st))return -errno; strm->alignsize =max_t (u32, st.st_blksize,getpagesize ());
@@ -132,6 +143,7 @@ void erofs_diskbuf_exit(void)for (strm = dbufstrm; strm->fd >=0 ; ++strm) {DBG_BUGON (erofs_atomic_read (&strm->count) !=1 ); +pthread_mutex_destroy (&strm->lock);close (strm->fd); strm->fd =-1 ; }
Signature · 1 line
2.43.0
Quoted 2 lines · Utkal Singh
Two threads calling
erofs_diskbuf_reserve()concurrently can both observestrm->locked== false and both advance tailoffset, silently
How?
Quoted 31 lines
producing an incorrect offset in the output image. Add a
pthread_mutex_tto serialize access.The commit path also gains the missing
strm->locked= false reset; without it, locked was set once and never cleared, making the flag a latch rather than a guard.libpthread is already a build requirement (-lpthread in
lib/Makefile.am) andpthread_mutex_tis used identically inlib/workqueue.candlib/compress.c, so no build system changes are required.Fixes 13f7268 ("erofs-utils: lib: introduce multi-threaded I/O framework")Signed-off-by Utkal Singh <[email protected]>--- lib/diskbuf.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-)lib/diskbuf.c +1
-0@@ -2,6 +2,7 @@#include "erofs/diskbuf.h" #include "erofs/internal.h" #include "erofs/print.h" +#include <pthread.h> #include <stdio.h> #include <errno.h> #include <sys/stat.h>
lib/diskbuf.c static struct erofs_diskbufstrm { · line 19
Quoted 4 lines
lib/diskbuf.c +1
-0@@ -15,6 +16,7 @@ static struct erofs_diskbufstrm {int fd;unsigned int alignsize;bool locked; +pthread_mutex_t lock;
Do you ever notice erofs_mutex_t and include/erofs/lock.h?
By the way, I sincerely hope you pay less attention of this project since I've told you the reasons in emails before.
It won't help with your gsoc proposal. Thanks for your efforts and contributions to the EROFS project.
Thanks,
Gao Xiang