From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754674Ab1ASPOI (ORCPT ); Wed, 19 Jan 2011 10:14:08 -0500 Received: from cpoproxy2-pub.bluehost.com ([67.222.39.38]:36364 "HELO cpoproxy2-pub.bluehost.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1754152Ab1ASPOG (ORCPT ); Wed, 19 Jan 2011 10:14:06 -0500 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=default; d=tao.ma; h=Received:Message-ID:Date:From:User-Agent:MIME-Version:To:CC:Subject:References:In-Reply-To:Content-Type:Content-Transfer-Encoding:X-Identified-User; b=Ixv9ba6zmDRxOb90SSThUluBuHAsXmOIDDXK8BCADNso3TWd/sSO9GT6F7q7Rj9rzaUIsQuVnRv3jiQPc8xpM/mZbys7mcYC3slHAdWvcaLTT2w+uK45wAsUqzW2XgpK; Message-ID: <4D36FFB2.8090303@tao.ma> Date: Wed, 19 Jan 2011 23:13:54 +0800 From: Tao Ma User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.13) Gecko/20101207 Thunderbird/3.1.7 MIME-Version: 1.0 To: Jeff Moyer CC: Jens Axboe , linux-kernel@vger.kernel.org Subject: Re: [PATCH v2] blktrace: Don't output messages if NOTIFY isn't set. References: <1295406590-7320-1-git-send-email-tm@tao.ma> <4D367873.90805@kernel.dk> <4D367FFF.2020505@tao.ma> <4D36EA84.1000605@kernel.dk> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Identified-User: {1390:box585.bluehost.com:colyli:tao.ma} {sentby:smtp auth 221.217.50.247 authed with tm@tao.ma} Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 01/19/2011 11:04 PM, Jeff Moyer wrote: > Jens Axboe writes: > >> On 2011-01-18 23:09, Tao Ma wrote: >>> On 01/19/2011 01:36 PM, Jens Axboe wrote: >>>> On 2011-01-18 20:09, Tao Ma wrote: >>>>> Hi Jeff, >>>>> On 01/18/2011 11:50 PM, Jeff Moyer wrote: >>>>>> Tao Ma writes: >>>>>> >>>>>>> + /* >>>>>>> + * If the user does specify some action masks, don't send any note >>>>>>> + * message to the trace so that it won't pollute what the user really >>>>>>> + * want to see. >>>>>>> + */ >>>>>>> + if (bt->act_mask != (u16) -1) >>>>>>> + return; >>>>>>> + >>>>>> I think you want to check for the BLK_TC_NOTIFY bit. >>>>> I was thinking of this, but actually there is no way for the user to set this flag. >>>>> At least from blktrace(8), -a doesn't have any flags like NOTIFY. So if the user >>>>> set any flag, NOTIFY will be cleared. So it does work. Thanks. >>>>> >>>>> Here is the updated patch. >>>>> >>>>> From 8693c19f165b09f60d5c26aed369d79a79e144ba Mon Sep 17 00:00:00 2001 >>>>> From: Tao Ma >>>>> Date: Wed, 19 Jan 2011 10:51:44 +0800 >>>>> Subject: [PATCH v2] blktrace: Don't output messages if NOTIFY isn't set. >>>>> >>>>> Now if we enable blktrace, cfq has too many messages output to the >>>>> trace buffer. It is fine if we don't specify any action mask. >>>>> But if I do like this: >>>>> blktrace /dev/sdb -a issue -a complete -o - | blkparse -i - >>>>> I only want to see 'D' and 'C', while with the following command >>>>> dd if=/mnt/ocfs2/test of=/dev/null bs=4k count=1 iflag=direct >>>>> >>>>> I will get(with a 2.6.37 vanilla kernel): >>>>> 8,16 0 0 0.000000000 0 m N cfq3805 alloced >>>>> 8,16 0 0 0.000004126 0 m N cfq3805 insert_request >>>>> 8,16 0 0 0.000004884 0 m N cfq3805 add_to_rr >>>>> 8,16 0 0 0.000008417 0 m N cfq workload slice:300 >>>>> 8,16 0 0 0.000009557 0 m N cfq3805 set_active wl_prio:0 wl_type:2 >>>>> 8,16 0 0 0.000010640 0 m N cfq3805 fifo= (null) >>>>> 8,16 0 0 0.000011193 0 m N cfq3805 dispatch_insert >>>>> 8,16 0 0 0.000012221 0 m N cfq3805 dispatched a request >>>>> 8,16 0 0 0.000012802 0 m N cfq3805 activate rq, drv=1 >>>>> 8,16 0 1 0.000013181 3805 D R 114759 + 8 [dd] >>>>> 8,16 0 2 0.000164244 0 C R 114759 + 8 [0] >>>>> 8,16 0 0 0.000167997 0 m N cfq3805 complete rqnoidle 0 >>>>> 8,16 0 0 0.000168782 0 m N cfq3805 set_slice=100 >>>>> 8,16 0 0 0.000169874 0 m N cfq3805 arm_idle: 8 group_idle: 0 >>>>> 8,16 0 0 0.000170189 0 m N cfq schedule dispatch >>>>> 8,16 0 0 0.000397938 0 m N cfq3805 slice expired t=0 >>>>> 8,16 0 0 0.000399763 0 m N cfq3805 sl_used=1 disp=1 charge=1 iops=0 sect=8 >>>>> 8,16 0 0 0.000400227 0 m N cfq3805 del_from_rr >>>>> 8,16 0 0 0.000400882 0 m N cfq3805 put_queue >>>>> >>>>> See, there are 19 lines while I only need 2. I don't think it is >>>>> appropriate for a user. >>>>> >>>>> So this patch will disable any messages if the BLK_TC_NOTIFY isn't set. >>>>> Now the output for the same command will look like: >>>>> 8,16 0 1 0.000000000 4908 D R 114759 + 8 [dd] >>>>> 8,16 0 2 0.000146827 0 C R 114759 + 8 [0] >>>>> >>>>> Yes, it is what I want to see. >>>>> >>>>> Cc: Jens Axboe >>>>> Cc: Steven Rostedt >>>>> Cc: Jeff Moyer >>>>> Signed-off-by: Tao Ma >>>>> --- >>>>> kernel/trace/blktrace.c | 7 +++++++ >>>>> 1 files changed, 7 insertions(+), 0 deletions(-) >>>>> >>>>> diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c >>>>> index 153562d..d95721f 100644 >>>>> --- a/kernel/trace/blktrace.c >>>>> +++ b/kernel/trace/blktrace.c >>>>> @@ -138,6 +138,13 @@ void __trace_note_message(struct blk_trace *bt, const char *fmt, ...) >>>>> !blk_tracer_enabled)) >>>>> return; >>>>> >>>>> + /* >>>>> + * If the BLK_TC_NOTIFY action mask isn't set, don't send any note >>>>> + * message to the trace. >>>>> + */ >>>>> + if (!(bt->act_mask& BLK_TC_NOTIFY)) >>>>> + return; >>>>> + >>>>> local_irq_save(flags); >>>>> buf = per_cpu_ptr(bt->msg_data, smp_processor_id()); >>>>> va_start(args, fmt); >>>> Hmm, it should only mask messages, not task/timestamp notifies. This >>>> will mask all of them. >>> uh, yes, that's the reason why I put the check in __trace_note_message, >>> not in trace_note. trace_note_time and trace_note_tsk will handle >>> time/task issue and then call trace_note with the appropriate action >>> mask. Do I miss something here? >> You did, I missed that. Patch looks fine then, I'll queue it up. Thanks! > I was guessing at the use of BLK_TC_NOTIFY for this. Should there be > another flag for these messages, so they aren't conflated with > task/timestamp notifications? I was thinking of this and gave up. Now we already have 15 different TC_FLAGS(yes, barrier may be removed, but for the compatibility problem, it have to stay there for some time I guess). and it seems we can have 16 flags totally(in blk_io_trace we have a u32 for action and the high 16 bit are used for TC_FLAGS). So I guess a new flag isn't suitable here. Regards, Tao