From: Roland Dreier <roland@topspin.com>
To: "Michael S. Tsirkin" <mst@mellanox.co.il>
Cc: akpm@osdl.org, linux-kernel@vger.kernel.org, openib-general@openib.org
Subject: Re: [openib-general] [PATCH][5/18] InfiniBand/mthca: add needed rmb() in event queue poll
Date: Thu, 13 Jan 2005 07:32:15 -0800 [thread overview]
Message-ID: <52acrds5ts.fsf@topspin.com> (raw)
In-Reply-To: <20050113094532.GA31298@mellanox.co.il> (Michael S. Tsirkin's message of "Thu, 13 Jan 2005 11:45:32 +0200")
Michael> Since we are using the eqe here, it seems that
Michael> read_barrier_depends shall be sufficient (as well as in
Michael> the cq case)?
Michael> However, I see that read_barrier_depends is a nop on ppc,
Michael> and the comment indicates that problems were seen on ppc
Michael> 970. What gives? do I misunderstand what a dependency
Michael> is?
There is no dependency between the EQE ownership field and the rest of
the EQE, so read_barrier_depends() is not sufficient. I think you are
misunderstanding what a dependency is. The comments in
asm-i386/system.h or http://lse.sourceforge.net/locking/wmbdd.html may
help clear things up.
- R.
prev parent reply other threads:[~2005-01-13 15:36 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-01-12 21:47 [PATCH][0/18] InfiniBand: updates for 2.6.11-rc1 Roland Dreier
2005-01-12 21:47 ` [PATCH][1/18] InfiniBand/IPoIB: use correct static rate in IpoIB Roland Dreier
2005-01-12 21:47 ` [PATCH][2/18] InfiniBand/mthca: trivial formatting fix Roland Dreier
2005-01-12 21:47 ` [PATCH][3/18] InfiniBand/mthca: support RDMA/atomic attributes in QP modify Roland Dreier
2005-01-12 21:47 ` [PATCH][4/18] InfiniBand/mthca: clean up allocation mapping of HCA context memory Roland Dreier
2005-01-12 21:47 ` [PATCH][5/18] InfiniBand/mthca: add needed rmb() in event queue poll Roland Dreier
2005-01-12 21:47 ` [PATCH][6/18] InfiniBand/core: remove debug printk Roland Dreier
2005-01-12 21:47 ` [PATCH][7/18] InfiniBand: make more code static Roland Dreier
2005-01-12 21:47 ` [PATCH][8/18] InfiniBand/core: set byte_cnt correctly in MAD completion Roland Dreier
2005-01-12 21:47 ` [PATCH][9/18] InfiniBand/core: add QP number to work completion struct Roland Dreier
2005-01-12 21:47 ` [PATCH][10/18] InfiniBand/core: add node_type and phys_state sysfs attrs Roland Dreier
2005-01-12 21:48 ` [PATCH][11/18] InfiniBand/mthca: clean up computation of HCA memory map Roland Dreier
2005-01-12 21:48 ` [PATCH][12/18] InfiniBand/core: fix handling of 0-hop directed route MADs Roland Dreier
2005-01-12 21:48 ` [PATCH][13/18] InfiniBand/core: add more parameters to process_mad Roland Dreier
2005-01-12 21:48 ` [PATCH][14/18] InfiniBand/core: add qp_type to struct ib_qp Roland Dreier
2005-01-12 21:48 ` [PATCH][15/18] InfiniBand/core: add ib_find_cached_gid function Roland Dreier
2005-01-12 21:48 ` [PATCH][16/18] InfiniBand: update copyrights for new year Roland Dreier
2005-01-12 21:48 ` [PATCH][17/18] InfiniBand/ipoib: move structs from stack to device private struct Roland Dreier
2005-01-12 21:48 ` [PATCH][18/18] InfiniBand/core: rename handle_outgoing_smp Roland Dreier
2005-01-13 9:45 ` [openib-general] [PATCH][5/18] InfiniBand/mthca: add needed rmb() in event queue poll Michael S. Tsirkin
2005-01-13 15:32 ` Roland Dreier [this message]
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=52acrds5ts.fsf@topspin.com \
--to=roland@topspin.com \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@mellanox.co.il \
--cc=openib-general@openib.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