Received: by 2002:a25:e7d8:0:0:0:0:0 with SMTP id e207csp11505ybh; Tue, 10 Mar 2020 18:10:00 -0700 (PDT) X-Google-Smtp-Source: ADFU+vvwnHaCoc2K+yyGIEOfSGM7kIBskrYrhHOgZwwFuoYbmjgNQPNINfmdkxZTRgKSadB6oMhV X-Received: by 2002:aca:c415:: with SMTP id u21mr304633oif.49.1583889000470; Tue, 10 Mar 2020 18:10:00 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1583889000; cv=none; d=google.com; s=arc-20160816; b=iMnzLAPLFj4JfD31ed9R6BDD7Y4Z/Xs+T7qghxVSDIznYzsBAVLdBp4usrNiRiFV+R ooyLqQ6bpyMt5HAECuIj5O+cDbFXOa7iH683YOrsJczAUG9796aq8WKqeasKsux2aF4A Y9kKh/aiJI9EOLSAY8eyJYxOTDTNDj9J1Vkr0OQsUV8aYr5c47nM8Mo6p0WRCa3M6Mok m4AcHAs5+zJJs7MZjY0NkPiOUk0cx7JbkT+MgjTtJjYH9cWKA/kSH9upt/fVG8rgt7Bk RtvxlF6SCwvGgvfwMQsVTGu15gMeEqnComEPe5XghFUToapR3mbr4ha2sXR6B7Yi8QXD tebQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:content-transfer-encoding:mime-version :user-agent:references:in-reply-to:date:cc:to:from:subject :message-id; bh=5FUGOsUzw8JzHT6zQX+6v6wyY3Am4XWsMAZy2tS5kWc=; b=GUs0Y4ItIC7g//FTwbOXxEAti/qrBVstQTsp/0s/NkaHN+8LiGHjBkcgKUbSKD9s8R cNuoo68pjKaJTKh3ieM/YokAwDXoAXaJihJf3Hj/n2wu6TaOm92AkDkBF7SAYzVzLGDP CXPwvQ3Qbjkwtb+Q3siL1kGqOxkwVhem9o5PsDP5PtZ+EJWuxmaYM5KtpA3H/DXfcGU9 I3Wgtdvg61rbAcSVCcJBuZIaDf+FWeOf9HeXXfAJjY0Gs9hJTwTO8wx986uuY7Uf1bMi BLnk1ztINRE2u9Vv4aLkiLLgCl7EjMOjs86YQAikfJOC6ZwHPMDId8OcGrGM6If3AB1S 7rjQ== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id k3si308154oib.82.2020.03.10.18.09.42; Tue, 10 Mar 2020 18:10:00 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727702AbgCKBJY (ORCPT + 99 others); Tue, 10 Mar 2020 21:09:24 -0400 Received: from mga03.intel.com ([134.134.136.65]:12990 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726463AbgCKBJY (ORCPT ); Tue, 10 Mar 2020 21:09:24 -0400 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga003.jf.intel.com ([10.7.209.27]) by orsmga103.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 10 Mar 2020 18:09:23 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,539,1574150400"; d="scan'208";a="242515528" Received: from sai-dev-mach.sc.intel.com ([143.183.140.153]) by orsmga003.jf.intel.com with ESMTP; 10 Mar 2020 18:09:22 -0700 Message-ID: <7a1f93d4516a7de99c5dbc4afd6279d6fe7aa126.camel@intel.com> Subject: Re: [PATCH V1 09/13] selftests/resctrl: Modularize fill_buf for new CAT test case From: Sai Praneeth Prakhya To: Reinette Chatre , shuah@kernel.org, linux-kselftest@vger.kernel.org Cc: tglx@linutronix.de, mingo@redhat.com, bp@alien8.de, tony.luck@intel.com, babu.moger@amd.com, james.morse@arm.com, ravi.v.shankar@intel.com, fenghua.yu@intel.com, x86@kernel.org, linux-kernel@vger.kernel.org Date: Tue, 10 Mar 2020 18:04:34 -0700 In-Reply-To: <4c84be1d-8839-2c85-b294-7e3c454240bb@intel.com> References: <43b368952bb006ee973311d9c9ae0eb53d8e7f60.1583657204.git.sai.praneeth.prakhya@intel.com> <4c84be1d-8839-2c85-b294-7e3c454240bb@intel.com> Content-Type: text/plain; charset="UTF-8" User-Agent: Evolution 3.30.5-0ubuntu0.18.10.1 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Reinette, On Tue, 2020-03-10 at 14:59 -0700, Reinette Chatre wrote: > Hi Sai, > > On 3/6/2020 7:40 PM, Sai Praneeth Prakhya wrote: > > Currently fill_buf (in-built benchmark) runs as a separate process and it > > runs indefinitely looping around given buffer either reading it or writing > > to it. But, some future test cases might want to start and stop looping > > around the buffer as they see fit. So, modularize fill_buf to support this > > use case. > > > > Signed-off-by: Sai Praneeth Prakhya > > --- > > tools/testing/selftests/resctrl/fill_buf.c | 66 ++++++++++++++++++++----- > > ----- > > 1 file changed, 44 insertions(+), 22 deletions(-) > > > > diff --git a/tools/testing/selftests/resctrl/fill_buf.c > > b/tools/testing/selftests/resctrl/fill_buf.c > > index 9ede7b63f059..204ae8870a32 100644 > > --- a/tools/testing/selftests/resctrl/fill_buf.c > > +++ b/tools/testing/selftests/resctrl/fill_buf.c > > @@ -23,7 +23,7 @@ > > #define PAGE_SIZE (4 * 1024) > > #define MB (1024 * 1024) > > > > -static unsigned char *startptr; > > +static unsigned char *startptr, *endptr; [Snipped.. assuming code over here might not be needed for discussion] > > +static int use_buffer_forever(int op, char *resctrl_val) > > +{ > > + int ret; > > + > > if (op == 0) > > - ret = fill_cache_read(start_ptr, end_ptr, resctrl_val); > > + ret = fill_cache_read(resctrl_val); > > else > > - ret = fill_cache_write(start_ptr, end_ptr, resctrl_val); > > + ret = fill_cache_write(resctrl_val); > > > > if (ret) { > > printf("\n Errror in fill cache read/write...\n"); > > return -1; > > } > > > > + return 0; > > +} > > + > > +static int > > +fill_cache(unsigned long long buf_size, int malloc_and_init, int > > memflush, > > + int op, char *resctrl_val) > > +{ > > + int ret; > > + > > + ret = init_buffer(buf_size, malloc_and_init, memflush); > > + if (ret) > > + return ret; > > + > > + ret = use_buffer_forever(op, resctrl_val); > > + if (ret) > > + return ret; > > Should buffer be freed on this error path? Yes, that's right.. my bad. Will fix it. But the right fix is, use_buffer_forever() should not return at all. It's meant to loop around the buffer _forever_. > I think the asymmetrical nature of the memory allocation and release > creates traps like this. > > It may be less error prone to have the pointer returned by init_buffer > and the acted on and released within fill_cache(), passed to > "use_buffer_forever()" as a parameter. The buffer size is known here, > there is no need to keep an "end pointer" around. The main reason for having "startptr" as a global variable is to free memory when fill_buf is killed. fill_buf runs as a separate process (for test cases like MBM, MBA and CQM) and when user issues Ctrl_c or when the test kills benchmark_pid (i.e. fill_buf), the buffer is freed (please see ctrl_handler()). So, I thought, as "startptr" is anyways global, why pass it around as an argument? While making this change I thought it's natural to make "endptr" global as well because the function didn't really look good to just take endptr as an argument without startptr. I do agree that asymmetrical nature of the memory allocation and release might create traps, I will try to overcome this for CAT test case (other test cases will not need it). Regards, Sai