From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chao Yu Subject: RE: [f2fs-dev] [PATCH] f2fs: check all ones or zeros bitmap with integer data type for better mount performance Date: Thu, 24 Oct 2013 11:28:19 +0800 Message-ID: <000101ced069$3d89fd40$b89df7c0$@samsung.com> References: <000701cecf09$2b2a2280$817e6780$@samsung.com> <1382444612.992.67.camel@kjgkr> <000001cecf9f$7113b630$533b2290$@samsung.com> <1382520734.992.78.camel@kjgkr> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-reply-to: <1382520734.992.78.camel@kjgkr> Content-language: zh-cn Sender: linux-fsdevel-owner@vger.kernel.org To: jaegeuk.kim@samsung.com Cc: linux-f2fs-devel@lists.sourceforge.net, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, =?utf-8?B?J+iwreWnnSc=?= List-Id: linux-f2fs-devel.lists.sourceforge.net Hi Kim, > -----Original Message----- > From: Jaegeuk Kim [mailto:jaegeuk.kim@samsung.com] > Sent: Wednesday, October 23, 2013 5:32 PM > To: Chao Yu > Cc: linux-f2fs-devel@lists.sourceforge.net; linux-fsdevel@vger.kernel= =2Eorg; > linux-kernel@vger.kernel.org; '=E8=B0=AD=E5=A7=9D' > Subject: RE: [f2fs-dev] [PATCH] f2fs: check all ones or zeros bitmap = with integer > data type for better mount performance >=20 > Hi, >=20 > 2013-10-23 (=EC=88=98), 11:23 +0800, Chao Yu: > > Hi, Kim: > > > > > -----Original Message----- > > > From: Jaegeuk Kim [mailto:jaegeuk.kim@samsung.com] > > > Sent: Tuesday, October 22, 2013 8:24 PM > > > To: Chao Yu > > > Cc: linux-f2fs-devel@lists.sourceforge.net; > > > linux-fsdevel@vger.kernel.org; linux-kernel@vger.kernel.org; =E8=B0= =AD=E5=A7=9D > > > Subject: Re: [f2fs-dev] [PATCH] f2fs: check all ones or zeros bit= map > > > with integer data type for better mount performance > > > > > > Hi, > > > > > > 2013-10-22 (=ED=99=94), 17:28 +0800, Chao Yu: > > > > Previously, check_block_count check valid_map with bit data typ= e > > > > in common scenario that sit has all ones or zeros bitmap, it ma= kes > > > > low mount performance. > > > > So let's check the special bitmap with integer data type instea= d > > > > of the bit one. > > > > > > > > Signed-off-by: Tan Shu > > > > Signed-off-by: Yu Chao > > > > --- > > > > fs/f2fs/segment.h | 13 +++++++++++++ > > > > 1 file changed, 13 insertions(+) > > > > > > > > diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index > > > > 7f94d78..d43ab9f 100644 > > > > --- a/fs/f2fs/segment.h > > > > +++ b/fs/f2fs/segment.h > > > > @@ -543,6 +543,7 @@ static inline void check_block_count(struct > > > > f2fs_sb_info *sbi, { > > > > struct f2fs_sm_info *sm_info =3D SM_I(sbi); > > > > unsigned int end_segno =3D sm_info->segment_count - 1; > > > > + int *valid_map =3D (int *)raw_sit->valid_map; > > > > int valid_blocks =3D 0; > > > > int i; > > > > > > > > @@ -552,6 +553,19 @@ static inline void check_block_count(struc= t > > > > f2fs_sb_info *sbi, > > > > /* check boundary of a given segment number */ > > > > BUG_ON(segno > end_segno); > > > > > > > > + /* check all ones or zeros valid_map */ > > > > + if (GET_SIT_VBLOCKS(raw_sit) =3D=3D 0) { > > > > + for (i =3D 0; i < SIT_VBLOCK_MAP_SIZE / sizeof(int); i++) > > > > > > We cannot guarantee all the time that SIT_VBLOCK_MAP_SIZE is > > > multiple of sizeof(int). > > Well, It's really large changes for f2fs if SIT_VBLOCK_MAP_SIZE val= ue is being > modified. >=20 > But, it can be changed. > Please do not add any unnecessary assumption. Got it, sorry for the unmeaning assumption. >=20 > > > > > How about using memcmp() with __u8? > > Do you mean that we can alloc all zeros or ones memory in > > SIT_VBLOCK_MAP_SIZE size, then memcmp() it with sit bitmap by __u8? >=20 > Yap. > Ah, but there is another one. > It would be better to use find_next_bit_le() and find_next_zero_bit_l= e(). > Any idea? Good point. I try to use memcmp(bitmap, bitmap+1, size-1) and bitmap[0],=20 But yours got better performance and readable. Thanks. >=20 > > > > > > > > > + if (unlikely(valid_map[i] !=3D 0)) > > > > + BUG(); > > > > + return; > > > > + } else if (GET_SIT_VBLOCKS(raw_sit) =3D=3D sbi->blocks_per_se= g) { > > > > + for (i =3D 0; i < SIT_VBLOCK_MAP_SIZE / sizeof(int); i++) > > > > + if (unlikely(valid_map[i] !=3D -1)) > > > > + BUG(); > > > > + return; > > > > + } > > > > + > > > > /* check bitmap with valid block count */ > > > > for (i =3D 0; i < sbi->blocks_per_seg; i++) > > > > if (f2fs_test_bit(i, raw_sit->valid_map)) > > > > --- > > > > > > > > -- > > > > To unsubscribe from this list: send the line "unsubscribe > > > > linux-fsdevel" in the body of a message to > > > > majordomo@vger.kernel.org More majordomo info at > > > > http://vger.kernel.org/majordomo-info.html > > > > > > -- > > > Jaegeuk Kim > > > Samsung > > >=20 > -- > Jaegeuk Kim > Samsung -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel= " in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html