Messages in this thread | | | Date | Wed, 19 May 1999 23:44:19 +0200 (CEST) | From | Andrea Arcangeli <> | Subject | Re: andrea buffer code (2.2.9-C.gz) |
| |
On Wed, 19 May 1999, Manfred Spraul wrote:
>That would be something for 2.[34], and your patch is for the stable >kernel tree. >Therefore, I'd prefer not to add new entries to mem_map unless there >is no way around. In this case: one spinlock or a small array >are the better solution.
Since the wastage will be only in SMP I don't care too much, I would prefer to remove the buffer field that is useless instead of dropping a nice per-pagemap spinlock.
>> >BTW, I think I found a major SMP-race, I've posted it to >> >linux-kernel@vger. >> >Since your buffer code fixes parts of that problem, I've attached my >> >mail. >> >> I would be glad if you would explain _why_ my new SMP per-pagemap io-async >> locking is not yet safe enough according to you. > >You code is safe, but there are several functions which call
Ok.
>The most frequent one was the old buffer code, that one is fixed.
The problem is not how it is frequent. The problem if the code is buggy or not. If it's not buggy then you could get 2000 printk/sec without having to worry about them. The case of the buffer code was not safe at least using IDE software-raids in SMP (I bet it would be not safe also with scsi raids since the scsi layer local-cli most of the time sigh!).
Thanks for your helpful comments!
Andrea Arcangeli
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |