qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: Paolo Bonzini <pbonzini@redhat.com>, peterx@redhat.com
Subject: [Qemu-devel] [PATCH v3 1/4] memory: Split zones when do coalesced_io_del()
Date: Tue, 20 Aug 2019 22:13:25 +0800	[thread overview]
Message-ID: <20190820141328.10009-2-peterx@redhat.com> (raw)
In-Reply-To: <20190820141328.10009-1-peterx@redhat.com>

It is a workaround of current KVM's KVM_UNREGISTER_COALESCED_MMIO
interface.  The kernel interface only allows to unregister an mmio
device with exactly the zone size when registered, or any smaller zone
that is included in the device mmio zone.  It does not support the
userspace to specify a very large zone to remove all the small mmio
devices within the zone covered.

Logically speaking it would be nicer to fix this from KVM side, though
in all cases we still need to coop with old kernels so let's do this.

Fixes: 3ac7d43a6fbb5d4a3
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 memory.c | 49 +++++++++++++++++++++++++++++++++++--------------
 1 file changed, 35 insertions(+), 14 deletions(-)

diff --git a/memory.c b/memory.c
index 8141486832..8173f6be62 100644
--- a/memory.c
+++ b/memory.c
@@ -855,8 +855,39 @@ static void address_space_update_ioeventfds(AddressSpace *as)
     flatview_unref(view);
 }
 
+/*
+ * Notify the memory listeners about the coalesced IO change events of
+ * range `cmr'.  Only the part that has intersection of the specified
+ * FlatRange will be sent.
+ */
+static void flat_range_coalesced_io_notify(FlatRange *fr, AddressSpace *as,
+                                           CoalescedMemoryRange *cmr, bool add)
+{
+    AddrRange tmp;
+
+    tmp = addrrange_shift(cmr->addr,
+                          int128_sub(fr->addr.start,
+                                     int128_make64(fr->offset_in_region)));
+    if (!addrrange_intersects(tmp, fr->addr)) {
+        return;
+    }
+    tmp = addrrange_intersection(tmp, fr->addr);
+
+    if (add) {
+        MEMORY_LISTENER_UPDATE_REGION(fr, as, Forward, coalesced_io_add,
+                                      int128_get64(tmp.start),
+                                      int128_get64(tmp.size));
+    } else {
+        MEMORY_LISTENER_UPDATE_REGION(fr, as, Reverse, coalesced_io_del,
+                                      int128_get64(tmp.start),
+                                      int128_get64(tmp.size));
+    }
+}
+
 static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
 {
+    CoalescedMemoryRange *cmr;
+
     if (!fr->has_coalesced_range) {
         return;
     }
@@ -865,16 +896,15 @@ static void flat_range_coalesced_io_del(FlatRange *fr, AddressSpace *as)
         return;
     }
 
-    MEMORY_LISTENER_UPDATE_REGION(fr, as, Reverse, coalesced_io_del,
-                                  int128_get64(fr->addr.start),
-                                  int128_get64(fr->addr.size));
+    QTAILQ_FOREACH(cmr, &fr->mr->coalesced, link) {
+        flat_range_coalesced_io_notify(fr, as, cmr, false);
+    }
 }
 
 static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
 {
     MemoryRegion *mr = fr->mr;
     CoalescedMemoryRange *cmr;
-    AddrRange tmp;
 
     if (QTAILQ_EMPTY(&mr->coalesced)) {
         return;
@@ -885,16 +915,7 @@ static void flat_range_coalesced_io_add(FlatRange *fr, AddressSpace *as)
     }
 
     QTAILQ_FOREACH(cmr, &mr->coalesced, link) {
-        tmp = addrrange_shift(cmr->addr,
-                              int128_sub(fr->addr.start,
-                                         int128_make64(fr->offset_in_region)));
-        if (!addrrange_intersects(tmp, fr->addr)) {
-            continue;
-        }
-        tmp = addrrange_intersection(tmp, fr->addr);
-        MEMORY_LISTENER_UPDATE_REGION(fr, as, Forward, coalesced_io_add,
-                                      int128_get64(tmp.start),
-                                      int128_get64(tmp.size));
+        flat_range_coalesced_io_notify(fr, as, cmr, true);
     }
 }
 
-- 
2.21.0



  reply	other threads:[~2019-08-20 14:17 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-20 14:13 [Qemu-devel] [PATCH v3 0/4] memory: Fix up coalesced_io_del not working for KVM Peter Xu
2019-08-20 14:13 ` Peter Xu [this message]
2019-08-20 14:13 ` [Qemu-devel] [PATCH v3 2/4] memory: Remove has_coalesced_range counter Peter Xu
2019-08-20 14:13 ` [Qemu-devel] [PATCH v3 3/4] memory: Refactor memory_region_clear_coalescing Peter Xu
2019-08-20 14:13 ` [Qemu-devel] [PATCH v3 4/4] memory: Fix up memory_region_{add|del}_coalescing Peter Xu
2019-08-21  8:14 ` [Qemu-devel] [PATCH v3 0/4] memory: Fix up coalesced_io_del not working for KVM Paolo Bonzini

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=20190820141328.10009-2-peterx@redhat.com \
    --to=peterx@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.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).