Return-Path: Received: from mail-qk0-f175.google.com ([209.85.220.175]:36707 "EHLO mail-qk0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751498AbcEESY6 (ORCPT ); Thu, 5 May 2016 14:24:58 -0400 Received: by mail-qk0-f175.google.com with SMTP id x7so47665084qkd.3 for ; Thu, 05 May 2016 11:24:58 -0700 (PDT) From: Jeff Layton To: linux-nfs@vger.kernel.org Cc: hch@lst.de, Anna.Schumaker@netapp.com Subject: [RFC PATCH 3/6] nfs: keep track of the return sequence number in pnfs_layout_hdr Date: Thu, 5 May 2016 14:24:45 -0400 Message-Id: <1462472688-5663-4-git-send-email-jeff.layton@primarydata.com> In-Reply-To: <1462472688-5663-1-git-send-email-jeff.layton@primarydata.com> References: <1462472688-5663-1-git-send-email-jeff.layton@primarydata.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: When we want to selectively do a LAYOUTRETURN, we need to specify a stateid that represents most recent layout acquisition that is to be returned. When we mark a layout stateid to be returned, we update the return sequence number in the layout header with that value, if it's newer than the existing one. Then, when we go to do a LAYOUTRETURN on layout header put, we overwrite the seqid in the stateid with the saved one, and then zero it out. Signed-off-by: Jeff Layton --- fs/nfs/pnfs.c | 11 ++++++++--- fs/nfs/pnfs.h | 1 + 2 files changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 7cc2c431f355..12016ed54e1d 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -899,6 +899,7 @@ pnfs_prepare_layoutreturn(struct pnfs_layout_hdr *lo) if (test_and_set_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) return false; lo->plh_return_iomode = 0; + lo->plh_return_seq = 0; pnfs_get_layout_hdr(lo); clear_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); return true; @@ -969,6 +970,7 @@ static void pnfs_layoutreturn_before_put_layout_hdr(struct pnfs_layout_hdr *lo) bool send; nfs4_stateid_copy(&stateid, &lo->plh_stateid); + stateid.seqid = cpu_to_be32(lo->plh_return_seq); iomode = lo->plh_return_iomode; send = pnfs_prepare_layoutreturn(lo); spin_unlock(&inode->i_lock); @@ -1747,7 +1749,8 @@ out_forget_reply: } static void -pnfs_set_plh_return_iomode(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode) +pnfs_set_plh_return_info(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode, + u32 seq) { if (lo->plh_return_iomode == iomode) return; @@ -1755,6 +1758,8 @@ pnfs_set_plh_return_iomode(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode) iomode = IOMODE_ANY; lo->plh_return_iomode = iomode; set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags); + if (!lo->plh_return_seq || pnfs_seqid_is_newer(seq, lo->plh_return_seq)) + lo->plh_return_seq = seq; } /** @@ -1793,7 +1798,7 @@ pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo, continue; remaining++; set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags); - pnfs_set_plh_return_iomode(lo, return_range->iomode); + pnfs_set_plh_return_info(lo, return_range->iomode, lseg->pls_seq); } return remaining; } @@ -1811,7 +1816,7 @@ void pnfs_error_mark_layout_for_return(struct inode *inode, bool return_now = false; spin_lock(&inode->i_lock); - pnfs_set_plh_return_iomode(lo, range.iomode); + pnfs_set_plh_return_info(lo, range.iomode, lseg->pls_seq); /* * mark all matching lsegs so that we are sure to have no live * segments at hand when sending layoutreturn. See pnfs_put_lseg() diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 62e61cf0ad98..47042084888c 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -195,6 +195,7 @@ struct pnfs_layout_hdr { unsigned long plh_flags; nfs4_stateid plh_stateid; u32 plh_barrier; /* ignore lower seqids */ + u32 plh_return_seq; enum pnfs_iomode plh_return_iomode; loff_t plh_lwb; /* last write byte for layoutcommit */ struct rpc_cred *plh_lc_cred; /* layoutcommit cred */ -- 2.5.5