From: Haren Myneni Subject: Re: [PATCH V3 6/6] crypto/nx: Add P9 NX support for 842 compression engine Date: Tue, 29 Aug 2017 18:02:09 -0700 Message-ID: <59A60E91.8020803@linux.vnet.ibm.com> References: <1500699702.23205.8.camel@hbabu-laptop> <1504041817.2358.32.camel@kernel.crashing.org> <59A5E2AE.5060400@linux.vnet.ibm.com> <1504043837.2358.38.camel@kernel.crashing.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Cc: Dan Streetman , Michael Ellerman , Herbert Xu , Linux Crypto Mailing List , "linuxppc-dev@lists.ozlabs.org" , mikey@neuling.org, suka@us.ibm.com, Ram Pai , npiggin@gmail.com, Haren Myneni To: Benjamin Herrenschmidt Return-path: Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]:40023 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751378AbdH3BCY (ORCPT ); Tue, 29 Aug 2017 21:02:24 -0400 Received: from pps.filterd (m0098396.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.21/8.16.0.21) with SMTP id v7U10DBW108604 for ; Tue, 29 Aug 2017 21:02:24 -0400 Received: from e38.co.us.ibm.com (e38.co.us.ibm.com [32.97.110.159]) by mx0a-001b2d01.pphosted.com with ESMTP id 2cn9g047y8-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Tue, 29 Aug 2017 21:02:24 -0400 Received: from localhost by e38.co.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Tue, 29 Aug 2017 19:02:23 -0600 In-Reply-To: <1504043837.2358.38.camel@kernel.crashing.org> Sender: linux-crypto-owner@vger.kernel.org List-ID: On 08/29/2017 02:57 PM, Benjamin Herrenschmidt wrote: > On Tue, 2017-08-29 at 14:54 -0700, Haren Myneni wrote: >> Opening send window for each crypto transform (crypto_alloc, >> compression/decompression, ..., crypto_free) so that does not have to >> wait for the previous copy/paste complete. VAS will map send and >> receive windows, and can cache in send windows (up to 128). So I >> thought using the same send window (per chip) for more requests (say >> 1000) may be adding overhead. >> >> I will make changes if you prefer using 1 send window per chip. > > Did you check the cost of opening/closing a window ? No, Not yet. opening / closing happens only during alloc/free, but not for each compression/decompression. Hence used separate send windows. Thanks Haren > > Cheers, > Ben. >