linux-kernel - The meaning of new_batch in Linux AS I/O scheduler

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Thu, 19 Mar 2009 11:04:13 +0800
From:	谢纲 <xiegang112@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: The meaning of new_batch in Linux AS I/O scheduler

Hi,

I'm little confused about the semantic of new_batch in Linux AS I/O scheduler.
As the comments mentioned, if it's 1, then it stands for that AS will
wait for the first read to complete. There is comments in defination:
/* 1: waiting on first read complete */

But there is another comments in as_completed_request:
/*
        * Start counting the batch from when a request of that direction is
        * actually serviced. This should help devices with big TCQ windows
        * and writeback caches
        */
       if (ad->new_batch && ad->batch_data_dir == rq_is_sync(rq)) {
               update_write_batch(ad);
               ad->current_batch_expires = jiffies +
                               ad->batch_expire[REQ_SYNC];
               ad->new_batch = 0;
       }

>From this comments and code, it seems that, new_batch does not only
work for read batch but also work for write batch, which conflicts
with the comments in defination. If it's only used when read batching,
why do we need to check the batch dir: ad->batch_data_dir ==
rq_is_sync(rq) (now, we are definitely in read batching ).

Thanks



-- 
Xie Gang
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/