All of lore.kernel.org
 help / color / mirror / Atom feed
From: Olaf Hering <olaf@aepfle.de>
To: xen-devel@lists.xensource.com
Subject: [PATCH 10 of 22] xenpaging: correct dropping of pages to avoid full ring buffer
Date: Fri, 10 Jun 2011 11:12:59 +0200	[thread overview]
Message-ID: <1de8de108d152fe915fc.1307697179@probook.site> (raw)
In-Reply-To: <patchbomb.1307697169@probook.site>

# HG changeset patch
# User Olaf Hering <olaf@aepfle.de>
# Date 1307695634 -7200
# Node ID 1de8de108d152fe915fc7f78044c406fed872bca
# Parent  9c42376aac05dc21a617d1d9fd62037cb8a9700d
xenpaging: correct dropping of pages to avoid full ring buffer

Doing a one-way channel from Xen to xenpaging is not possible with the
current ring buffer implementation. xenpaging uses the mem_event ring
buffer, which expects request/response pairs to make progress. The
previous patch, which tried to establish a one-way communication from
Xen to xenpaging, stalled the guest once the buffer was filled up with
requests. Correct page-dropping by taking the slow path and let
p2m_mem_paging_resume() consume the response from xenpaging. This makes
room for yet another request/response pair and avoids hanging guests.

Signed-off-by: Olaf Hering <olaf@aepfle.de>

diff -r 9c42376aac05 -r 1de8de108d15 tools/xenpaging/xenpaging.c
--- a/tools/xenpaging/xenpaging.c	Fri Jun 10 10:47:12 2011 +0200
+++ b/tools/xenpaging/xenpaging.c	Fri Jun 10 10:47:14 2011 +0200
@@ -690,19 +690,19 @@ int main(int argc, char *argv[])
                         ERROR("Error populating page");
                         goto out;
                     }
+                }
 
-                    /* Prepare the response */
-                    rsp.gfn = req.gfn;
-                    rsp.p2mt = req.p2mt;
-                    rsp.vcpu_id = req.vcpu_id;
-                    rsp.flags = req.flags;
+                /* Prepare the response */
+                rsp.gfn = req.gfn;
+                rsp.p2mt = req.p2mt;
+                rsp.vcpu_id = req.vcpu_id;
+                rsp.flags = req.flags;
 
-                    rc = xenpaging_resume_page(paging, &rsp, 1);
-                    if ( rc != 0 )
-                    {
-                        ERROR("Error resuming page");
-                        goto out;
-                    }
+                rc = xenpaging_resume_page(paging, &rsp, 1);
+                if ( rc != 0 )
+                {
+                    ERROR("Error resuming page");
+                    goto out;
                 }
 
                 /* Evict a new page to replace the one we just paged in */

  parent reply	other threads:[~2011-06-10  9:12 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-10  9:12 [PATCH 00 of 22] xenpaging and libxc fixes for xen-unstable Olaf Hering
2011-06-10  9:12 ` [PATCH 01 of 22] xenpaging: remove unused spinlock in pager Olaf Hering
2011-06-21 16:24   ` Ian Jackson
2011-06-10  9:12 ` [PATCH 02 of 22] tools: merge several bitop functions into xc_bitops.h Olaf Hering
2011-06-21 16:16   ` Ian Jackson
2011-06-21 16:20     ` Olaf Hering
2011-06-21 16:25       ` Ian Jackson
2011-06-21 17:17     ` Olaf Hering
2011-06-21 17:18       ` Ian Jackson
2011-06-22 14:03         ` Ian Jackson
2011-06-10  9:12 ` [PATCH 03 of 22] xenpaging: add xs_handle to struct xenpaging Olaf Hering
2011-06-14 11:05   ` Ian Campbell
2011-06-20  9:58     ` [PATCH 03 of 22 v2] " Olaf Hering
2011-06-21 13:17       ` Ian Campbell
2011-06-21 15:33         ` [PATCH 03 of 22 v3] " Olaf Hering
2011-06-10  9:12 ` [PATCH 04 of 22] xenpaging: drop xc.c, remove ASSERT Olaf Hering
2011-06-10  9:12 ` [PATCH 05 of 22] xenpaging: drop xc.c, remove xc_platform_info_t Olaf Hering
2011-06-10  9:12 ` [PATCH 06 of 22] xenpaging: drop xc.c, remove xc_wait_for_event Olaf Hering
2011-06-10  9:12 ` [PATCH 07 of 22] xenpaging: drop xc.c, move xc_mem_paging_flush_ioemu_cache Olaf Hering
2011-06-10  9:12 ` [PATCH 08 of 22] xenpaging: drop xc.c, move xc_wait_for_event_or_timeout Olaf Hering
2011-06-10  9:12 ` [PATCH 09 of 22] xenpaging: drop xc.c, remove xc files Olaf Hering
2011-06-10  9:12 ` Olaf Hering [this message]
2011-06-10  9:13 ` [PATCH 11 of 22] xenpaging: do not bounce p2mt to xenpaging Olaf Hering
2011-06-13 10:48   ` Tim Deegan
2011-06-20  9:57     ` [PATCH 11 of 22 v2] xenpaging: do not bounce p2mt back to the hypervisor Olaf Hering
2011-06-20 10:36       ` Tim Deegan
2011-06-10  9:13 ` [PATCH 12 of 22] xenpaging: remove srand call Olaf Hering
2011-06-10  9:13 ` [PATCH 13 of 22] xenpaging: remove return values from functions that can not fail Olaf Hering
2011-06-10  9:13 ` [PATCH 14 of 22] xenpaging: catch xc_mem_paging_resume errors Olaf Hering
2011-06-10  9:13 ` [PATCH 15 of 22] xenpaging: remove local domain_id variable Olaf Hering
2011-06-10  9:13 ` [PATCH 16 of 22] xenpaging: move num_pages into xenpaging struct Olaf Hering
2011-06-10  9:13 ` [PATCH 17 of 22] xenpaging: start paging in the middle of gfn range Olaf Hering
2011-06-10  9:13 ` [PATCH 18 of 22] xenpaging: pass integer to xenpaging_populate_page Olaf Hering
2011-06-10  9:13 ` [PATCH 19 of 22] xenpaging: add helper function for unlinking pagefile Olaf Hering
2011-06-10  9:13 ` [PATCH 20 of 22] xenpaging: add watch thread to catch guest shutdown Olaf Hering
2011-06-10  9:13 ` [PATCH 21 of 22] xenpaging: implement stopping of pager by sending SIGTERM/SIGINT Olaf Hering
2011-06-10  9:13 ` [PATCH 22 of 22] xenpaging: remove private mem_event.h Olaf Hering

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=1de8de108d152fe915fc.1307697179@probook.site \
    --to=olaf@aepfle.de \
    --cc=xen-devel@lists.xensource.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.