Hi. I would like to report a suspected bug at let_transaction_grow()
in fs/reiserfs/journal.c of Linux 2.6.30.4.
I found this suspected bug while I survey the usage of atomic instruction
and memory barriers. I am not certain whether this report is real or not
since I do not have ReiserFS background. But I hope that this report is helpful.
Please examine the report and let me know your opinion.
In let_transaction_grow(), I found the following code for busy-waiting:
while ((atomic_read(&journal->j_wcount) > 0 ||
atomic_read(&journal->j_jlock)) &&
journal->j_trans_id == trans_id) {
queue_log_writer(sb);
}
However, it may result starvation(or infinite loop) since there is no
memory barrier for atomic_read()
(atomic_read() may read out-of-dates value).
Thank you.
Sincerely
Shin Hong
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
