Re: Mount ext3 with barrier=1 doesn't send real barrier bio?

[OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>]

Eric Sandeen <sandeen@sandeen.net> writes:

>> --- a/fs/buffer.c
>> +++ b/fs/buffer.c
>> @@ -2926,16 +2926,16 @@ int submit_bh(int rw, struct buffer_head * bh)
>>  	BUG_ON(!buffer_mapped(bh));
>>  	BUG_ON(!bh->b_end_io);
>>  
>> -	if (buffer_ordered(bh) && (rw == WRITE))
>> -		rw = WRITE_BARRIER;
>> -
>>  	/*
>>  	 * Only clear out a write error when rewriting, should this
>>  	 * include WRITE_SYNC as well?
>>  	 */
>> -	if (test_set_buffer_req(bh) && (rw == WRITE || rw == WRITE_BARRIER))
>> +	if (test_set_buffer_req(bh) && rw == WRITE)
>>  		clear_buffer_write_io_error(bh);

This should be ((rw & RW_MASK) == WRITE) too?  Anyway, this seems change
behavior of submit_bh(WRITE_BARRIER) (maybe reiserfs only), it wouldn't
be your intent...

>> +	if (buffer_ordered(bh) && ((rw & RW_MASK) == WRITE))
>> +		rw |= (1 << BIO_RW_BARRIER);
>> +
>>  	/*
>>  	 * from here on down, it's all bio -- do the initial mapping,
>>  	 * submit_bio -> generic_make_request may further map this bio around
-- 
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>