From: Boaz Harrosh <bharrosh@panasas.com>
To: Benny Halevy <bhalevy@panasas.com>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>,
NFS list <linux-nfs@vger.kernel.org>
Subject: Re: [PATCHSET v6 0/26] pnfs for 2.6.40
Date: Mon, 23 May 2011 21:50:12 +0300 [thread overview]
Message-ID: <4DDAAC64.6050807@panasas.com> (raw)
In-Reply-To: <4DDA8C3D.5080706@panasas.com>
On 05/23/2011 07:33 PM, Benny Halevy wrote:
Benny Hi
I have a problem that the default wsize is very small 64K and
I get small IOs. I found that the governing member right now
is NFS_SERVER()->wsize
I did the below hack on My current code, but you took that away
from me.
diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c
index ec40408..f7b09e1 100644
--- a/fs/nfs/objlayout/objlayout.c
+++ b/fs/nfs/objlayout/objlayout.c
+ server->wsize = ((PAGE_SIZE - sizeof(struct bio)) / sizeof(struct bio_vec))
+ * PAGE_SIZE * 2;
- dprintk("%s: Return data=%p\n", __func__, data);
+ dprintk("%s: Return data=%p wsize=0x%x\n", __func__, data, server->wsize);
return 0;
}
What do you want that we do to replace this. The default 64K is to small.
I don't mind that for pnfs it will be ~0 and the pg_test() will test
for maxc_size as well. But then we'll also need the current size or the
start_index
Boaz
next prev parent reply other threads:[~2011-05-23 18:50 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-23 16:33 [PATCHSET v6 0/26] pnfs for 2.6.40 Benny Halevy
2011-05-23 16:34 ` [PATCH v6 01/26] NFSv4.1: use struct nfs_client to qualify deviceid Benny Halevy
2011-05-23 17:33 ` Benny Halevy
2011-05-23 16:34 ` [PATCH v6 02/26] pnfs: resolve header dependency in pnfs.h Benny Halevy
2011-05-23 16:34 ` [PATCH v6 03/26] NFSv4.1: make deviceid cache global Benny Halevy
2011-05-23 16:34 ` [PATCH v6 04/26] NFSv4.1: purge deviceid cache on nfs_free_client Benny Halevy
2011-05-23 17:21 ` Benny Halevy
2011-05-23 16:35 ` [PATCH v6 05/26] pnfs: CB_NOTIFY_DEVICEID Benny Halevy
2011-05-23 16:35 ` [PATCH v6 06/26] SUNRPC: introduce xdr_init_decode_pages Benny Halevy
2011-05-23 16:35 ` [PATCH v6 07/26] pnfs: Use byte-range for layoutget Benny Halevy
2011-05-23 16:35 ` [PATCH v6 08/26] pnfs: align layoutget requests on page boundaries Benny Halevy
2011-05-23 16:35 ` [PATCH v6 09/26] pnfs: Use byte-range for cb_layoutrecall Benny Halevy
2011-05-23 16:36 ` [PATCH v6 10/26] pnfs: client stats Benny Halevy
2011-05-23 16:36 ` [PATCH v6 11/26] pnfs-obj: objlayoutdriver module skeleton Benny Halevy
2011-05-23 16:36 ` [PATCH v6 12/26] pnfs-obj: pnfs_osd XDR definitions Benny Halevy
2011-05-23 16:36 ` [PATCH v6 13/26] pnfs-obj: pnfs_osd XDR client implementation Benny Halevy
2011-05-23 19:46 ` [PATCH] SQUASHME: pnf-obj xdr_cli: Wrong type in comments Boaz Harrosh
2011-05-23 16:37 ` [PATCH v6 14/26] pnfs-obj: decode layout, alloc/free lseg Benny Halevy
2011-05-23 19:45 ` [PATCH] SQUASHME: objio read/write patch: Bugs fixes Boaz Harrosh
2011-05-24 13:10 ` Benny Halevy
2011-05-24 14:37 ` Boaz Harrosh
2011-05-24 15:57 ` Benny Halevy
2011-05-24 16:04 ` Boaz Harrosh
2011-05-23 16:37 ` [PATCH v6 15/26] pnfs-obj: objio_osd device information retrieval and caching Benny Halevy
2011-05-23 16:37 ` [PATCH v6 16/26] NFSv4.1: use layout driver in global device cache Benny Halevy
2011-05-23 16:37 ` [PATCH v6 17/26] pnfs: alloc and free layout_hdr layoutdriver methods Benny Halevy
2011-05-23 16:37 ` [PATCH v6 18/26] pnfs-obj: define per-inode private structure Benny Halevy
2011-05-23 16:38 ` [PATCH v6 19/26] pnfs: support for non-rpc layout drivers Benny Halevy
2011-05-23 18:10 ` Boaz Harrosh
2011-05-23 19:22 ` Benny Halevy
2011-05-23 19:43 ` [PATCH] SQUASHME: into pnfs: pnfs: support for non-rpc layout drivers: de-ref not needed Boaz Harrosh
2011-05-23 16:38 ` [PATCH v6 20/26] pnfs-obj: osd raid engine read/write implementation Benny Halevy
2011-05-25 13:39 ` Boaz Harrosh
2011-05-25 13:41 ` [PATCH] SQUASHME: pnfs-obj: pg_test check for max_io_size Boaz Harrosh
2011-05-25 16:57 ` Benny Halevy
2011-05-23 16:38 ` [PATCH v6 21/26] pnfs: layoutreturn Benny Halevy
2011-05-25 16:07 ` [PATCH] SQUASHME: pnfs: Fix NULL dereference in the -ENOMEM path Boaz Harrosh
2011-05-25 16:12 ` Boaz Harrosh
2011-05-25 16:19 ` [PATCH V2] SQUASHME: pnfs: Fix NULL dereference and leak " Boaz Harrosh
2011-05-25 16:37 ` Boaz Harrosh
2011-05-25 16:47 ` Benny Halevy
2011-05-25 16:40 ` [PATCH V3] " Boaz Harrosh
2011-05-25 16:55 ` Benny Halevy
2011-05-23 16:38 ` [PATCH v6 22/26] pnfs: layoutret_on_setattr Benny Halevy
2011-05-23 16:38 ` [PATCH v6 23/26] pnfs: encode_layoutreturn Benny Halevy
2011-05-23 16:39 ` [PATCH v6 24/26] pnfs-obj: report errors and .encode_layoutreturn Implementation Benny Halevy
2011-05-23 16:39 ` [PATCH v6 25/26] pnfs: encode_layoutcommit Benny Halevy
2011-05-23 16:39 ` [PATCH v6 26/26] pnfs-obj: objlayout_encode_layoutcommit implementation Benny Halevy
2011-05-23 18:20 ` [PATCHSET v6 0/26] pnfs for 2.6.40 Boaz Harrosh
2011-05-23 18:50 ` Boaz Harrosh [this message]
2011-05-24 15:16 ` Benny Halevy
[not found] ` <2E1EB2CF9ED1CB4AA966F0EB76EAB443080D6E54@SACMVEXC2-PRD.hq.netapp.com>
2011-05-24 15:49 ` Benny Halevy
2011-05-24 17:07 ` Fred Isaman
2011-05-24 15:56 ` Boaz Harrosh
2011-05-24 16:21 ` Trond Myklebust
2011-05-24 16:58 ` Boaz Harrosh
2011-05-24 17:05 ` Trond Myklebust
2011-05-24 17:07 ` Boaz Harrosh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4DDAAC64.6050807@panasas.com \
--to=bharrosh@panasas.com \
--cc=Trond.Myklebust@netapp.com \
--cc=bhalevy@panasas.com \
--cc=linux-nfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).