From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45042 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229464AbhFXSk7 (ORCPT ); Thu, 24 Jun 2021 14:40:59 -0400 Received: from mail-vs1-xe2b.google.com (mail-vs1-xe2b.google.com [IPv6:2607:f8b0:4864:20::e2b]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 99774C061574 for ; Thu, 24 Jun 2021 11:38:39 -0700 (PDT) Received: by mail-vs1-xe2b.google.com with SMTP id x12so4038195vsp.4 for ; Thu, 24 Jun 2021 11:38:39 -0700 (PDT) MIME-Version: 1.0 From: Sunil Vettukalppurathu Date: Thu, 24 Jun 2021 11:38:27 -0700 Message-ID: Subject: random buffer_compress_percentage instead of static Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable List-Id: fio@vger.kernel.org To: fio@vger.kernel.org Hi, Currently we can=E2=80=99t vary % of compressible data from block to block = (as far as I know). The value provided for buffer_compress_percentage is used for all blocks. """ buffer_compress_percentage=3Dint If this is set, then fio will attempt to provide I/O buffer content (on WRITEs) that compresses to the specified level. Fio does this by providing a mix of random data followed by fixed pattern data. """ As we have a requirement to test with random compression %, we modified the code to accept a range of values for buffer_compress_percentage instead of one number so that the percentage changes from block to block =E2=80=93 random value between min a= nd max. Not sure if it=E2=80=99s useful for others; if so, will send the patch for= review. Thanks, --Sunil