From mboxrd@z Thu Jan 1 00:00:00 1970 From: Martin Schwidefsky Subject: Re: [PATCH] bitmap: bitmap_equal memcmp optimization Date: Thu, 9 Jun 2016 08:31:39 +0200 Message-ID: <20160609083139.665d996b@mschwide> References: <1465288661-3273-1-git-send-email-schwidefsky@de.ibm.com> <201606070837.u578YJKt047781@mx0a-001b2d01.pphosted.com> <20160608134212.9f868a8dd494b88e402d5d31@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20160608134212.9f868a8dd494b88e402d5d31@linux-foundation.org> Sender: linux-kernel-owner@vger.kernel.org To: Andrew Morton Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org, Benjamin Herrenschmidt , "David S. Miller" List-Id: linux-arch.vger.kernel.org On Wed, 8 Jun 2016 13:42:12 -0700 Andrew Morton wrote: > On Tue, 7 Jun 2016 10:37:41 +0200 Martin Schwidefsky wrote: > > > The bitmap_equal function has optimized code for small bitmaps with less > > than BITS_PER_LONG bits. For larger bitmaps the out-of-line function > > __bitmap_equal is called. > > > > For a constant number of bits divisible by BITS_PER_LONG the memcmp > > function can be used. For s390 gcc knows how to optimize this function, > > memcmp calls with up to 256 bytes / 2048 bits are translated into a > > single instruction. > > Patch looks simple enough, although it would benefit from a comment > explaining what's special about s390. The explanation from the git commit could be transfered into a comment. > > --- a/include/linux/bitmap.h > > +++ b/include/linux/bitmap.h > > @@ -267,6 +267,10 @@ static inline int bitmap_equal(const unsigned long *src1, > > { > > if (small_const_nbits(nbits)) > > return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > > +#ifdef CONFIG_S390 > > + else if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > > + return !memcmp(src1, src2, nbits / 8); > > +#endif > > else > > return __bitmap_equal(src1, src2, nbits); > > } > > Those elses are a bit daffy. This: > > --- a/include/linux/bitmap.h~bitmap-bitmap_equal-memcmp-optimization-fix > +++ a/include/linux/bitmap.h > @@ -266,13 +266,12 @@ static inline int bitmap_equal(const uns > const unsigned long *src2, unsigned int nbits) > { > if (small_const_nbits(nbits)) > - return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > + return !((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > #ifdef CONFIG_S390 > - else if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > + if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > return !memcmp(src1, src2, nbits / 8); > #endif > - else > - return __bitmap_equal(src1, src2, nbits); > + return __bitmap_equal(src1, src2, nbits); > } > > static inline int bitmap_intersects(const unsigned long *src1, > _ > Yeah that looks better. Thanks for picking this up! -- blue skies, Martin. "Reality continues to ruin my life." - Calvin. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:58824 "EHLO mx0a-001b2d01.pphosted.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1161696AbcFIGbp (ORCPT ); Thu, 9 Jun 2016 02:31:45 -0400 Received: from pps.filterd (m0049461.ppops.net [127.0.0.1]) by mx0b-001b2d01.pphosted.com (8.16.0.11/8.16.0.11) with SMTP id u596TkDG021828 for ; Thu, 9 Jun 2016 02:31:44 -0400 Received: from e06smtp16.uk.ibm.com (e06smtp16.uk.ibm.com [195.75.94.112]) by mx0b-001b2d01.pphosted.com with ESMTP id 23e9mx9nec-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Thu, 09 Jun 2016 02:31:44 -0400 Received: from localhost by e06smtp16.uk.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Thu, 9 Jun 2016 07:31:43 +0100 Date: Thu, 9 Jun 2016 08:31:39 +0200 From: Martin Schwidefsky Subject: Re: [PATCH] bitmap: bitmap_equal memcmp optimization In-Reply-To: <20160608134212.9f868a8dd494b88e402d5d31@linux-foundation.org> References: <1465288661-3273-1-git-send-email-schwidefsky@de.ibm.com> <201606070837.u578YJKt047781@mx0a-001b2d01.pphosted.com> <20160608134212.9f868a8dd494b88e402d5d31@linux-foundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Message-ID: <20160609083139.665d996b@mschwide> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Andrew Morton Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org, Benjamin Herrenschmidt , "David S. Miller" Message-ID: <20160609063139.Brw-2bjtErDxw6CnU7ABqspfz11jEjUS-kugiy7mLR0@z> On Wed, 8 Jun 2016 13:42:12 -0700 Andrew Morton wrote: > On Tue, 7 Jun 2016 10:37:41 +0200 Martin Schwidefsky wrote: > > > The bitmap_equal function has optimized code for small bitmaps with less > > than BITS_PER_LONG bits. For larger bitmaps the out-of-line function > > __bitmap_equal is called. > > > > For a constant number of bits divisible by BITS_PER_LONG the memcmp > > function can be used. For s390 gcc knows how to optimize this function, > > memcmp calls with up to 256 bytes / 2048 bits are translated into a > > single instruction. > > Patch looks simple enough, although it would benefit from a comment > explaining what's special about s390. The explanation from the git commit could be transfered into a comment. > > --- a/include/linux/bitmap.h > > +++ b/include/linux/bitmap.h > > @@ -267,6 +267,10 @@ static inline int bitmap_equal(const unsigned long *src1, > > { > > if (small_const_nbits(nbits)) > > return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > > +#ifdef CONFIG_S390 > > + else if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > > + return !memcmp(src1, src2, nbits / 8); > > +#endif > > else > > return __bitmap_equal(src1, src2, nbits); > > } > > Those elses are a bit daffy. This: > > --- a/include/linux/bitmap.h~bitmap-bitmap_equal-memcmp-optimization-fix > +++ a/include/linux/bitmap.h > @@ -266,13 +266,12 @@ static inline int bitmap_equal(const uns > const unsigned long *src2, unsigned int nbits) > { > if (small_const_nbits(nbits)) > - return ! ((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > + return !((*src1 ^ *src2) & BITMAP_LAST_WORD_MASK(nbits)); > #ifdef CONFIG_S390 > - else if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > + if (__builtin_constant_p(nbits) && (nbits % BITS_PER_LONG) == 0) > return !memcmp(src1, src2, nbits / 8); > #endif > - else > - return __bitmap_equal(src1, src2, nbits); > + return __bitmap_equal(src1, src2, nbits); > } > > static inline int bitmap_intersects(const unsigned long *src1, > _ > Yeah that looks better. Thanks for picking this up! -- blue skies, Martin. "Reality continues to ruin my life." - Calvin.