From: Chenyi Qiang <chenyi.qiang@intel.com>
To: "David Hildenbrand" <david@redhat.com>,
"Alexey Kardashevskiy" <aik@amd.com>,
"Peter Xu" <peterx@redhat.com>,
"Gupta Pankaj" <pankaj.gupta@amd.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Philippe Mathieu-Daudé" <philmd@linaro.org>,
"Michael Roth" <michael.roth@amd.com>
Cc: Chenyi Qiang <chenyi.qiang@intel.com>,
qemu-devel@nongnu.org, kvm@vger.kernel.org,
Williams Dan J <dan.j.williams@intel.com>,
Peng Chao P <chao.p.peng@intel.com>,
Gao Chao <chao.gao@intel.com>, Xu Yilun <yilun.xu@intel.com>,
Li Xiaoyao <xiaoyao.li@intel.com>
Subject: [PATCH v4 09/13] memory: Attach RamBlockAttribute to guest_memfd-backed RAMBlocks
Date: Mon, 7 Apr 2025 15:49:29 +0800 [thread overview]
Message-ID: <20250407074939.18657-10-chenyi.qiang@intel.com> (raw)
In-Reply-To: <20250407074939.18657-1-chenyi.qiang@intel.com>
A new field, ram_block_attribute, is introduced in RAMBlock to link to a
RamBlockAttribute object. This change centralizes all guest_memfd state
information (such as fd and shared_bitmap) within a RAMBlock,
simplifying management.
The realize()/unrealized() helpers are used to initialize/uninitialize
the RamBlockAttribute object. The object is registered/unregistered in
the target RAMBlock's MemoryRegion when creating guest_memfd.
Additionally, use the private_shared_manager_state_change() helper to
notify the registered PrivateSharedListener of these changes.
Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
---
Changes in v4:
- Remove the replay operations for attribute changes which will be
handled in a listener in following patches.
- Add some comment in the error path of realize() to remind the
future development of the unified error path.
Changes in v3:
- Use ram_discard_manager_reply_populated/discarded() to set the
memory attribute and add the undo support if state_change()
failed.
- Didn't add Reviewed-by from Alexey due to the new changes in this
commit.
Changes in v2:
- Introduce a new field memory_attribute_manager in RAMBlock.
- Move the state_change() handling during page conversion in this patch.
- Undo what we did if it fails to set.
- Change the order of close(guest_memfd) and memory_attribute_manager cleanup.
---
accel/kvm/kvm-all.c | 9 +++++++++
include/exec/ramblock.h | 1 +
system/physmem.c | 16 ++++++++++++++++
3 files changed, 26 insertions(+)
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index c1fea69d58..546b58b737 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -3088,6 +3088,15 @@ int kvm_convert_memory(hwaddr start, hwaddr size, bool to_private)
addr = memory_region_get_ram_ptr(mr) + section.offset_within_region;
rb = qemu_ram_block_from_host(addr, false, &offset);
+ ret = private_shared_manager_state_change(PRIVATE_SHARED_MANAGER(mr->gsm),
+ offset, size, to_private);
+ if (ret) {
+ error_report("Failed to notify the listener the state change of "
+ "(0x%"HWADDR_PRIx" + 0x%"HWADDR_PRIx") to %s",
+ start, size, to_private ? "private" : "shared");
+ goto out_unref;
+ }
+
if (to_private) {
if (rb->page_size != qemu_real_host_page_size()) {
/*
diff --git a/include/exec/ramblock.h b/include/exec/ramblock.h
index b8b5469db9..78eb031819 100644
--- a/include/exec/ramblock.h
+++ b/include/exec/ramblock.h
@@ -46,6 +46,7 @@ struct RAMBlock {
int fd;
uint64_t fd_offset;
int guest_memfd;
+ RamBlockAttribute *ram_block_attribute;
size_t page_size;
/* dirty bitmap used during migration */
unsigned long *bmap;
diff --git a/system/physmem.c b/system/physmem.c
index c76503aea8..fb74321e10 100644
--- a/system/physmem.c
+++ b/system/physmem.c
@@ -1885,6 +1885,20 @@ static void ram_block_add(RAMBlock *new_block, Error **errp)
qemu_mutex_unlock_ramlist();
goto out_free;
}
+
+ new_block->ram_block_attribute = RAM_BLOCK_ATTRIBUTE(object_new(TYPE_RAM_BLOCK_ATTRIBUTE));
+ if (ram_block_attribute_realize(new_block->ram_block_attribute, new_block->mr)) {
+ error_setg(errp, "Failed to realize ram block attribute");
+ /*
+ * The error path could be unified if the rest of ram_block_add() ever
+ * develops a need to check for errors.
+ */
+ object_unref(OBJECT(new_block->ram_block_attribute));
+ close(new_block->guest_memfd);
+ ram_block_discard_require(false);
+ qemu_mutex_unlock_ramlist();
+ goto out_free;
+ }
}
ram_size = (new_block->offset + new_block->max_length) >> TARGET_PAGE_BITS;
@@ -2138,6 +2152,8 @@ static void reclaim_ramblock(RAMBlock *block)
}
if (block->guest_memfd >= 0) {
+ ram_block_attribute_unrealize(block->ram_block_attribute);
+ object_unref(OBJECT(block->ram_block_attribute));
close(block->guest_memfd);
ram_block_discard_require(false);
}
--
2.43.5
next prev parent reply other threads:[~2025-04-07 7:51 UTC|newest]
Thread overview: 67+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-07 7:49 [PATCH v4 00/13] Enable shared device assignment Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 01/13] memory: Export a helper to get intersection of a MemoryRegionSection with a given range Chenyi Qiang
2025-04-09 2:47 ` Alexey Kardashevskiy
2025-04-09 6:26 ` Chenyi Qiang
2025-04-09 6:45 ` Alexey Kardashevskiy
2025-04-09 7:38 ` Chenyi Qiang
2025-05-12 3:24 ` Zhao Liu
2025-04-07 7:49 ` [PATCH v4 02/13] memory: Change memory_region_set_ram_discard_manager() to return the result Chenyi Qiang
2025-04-07 9:53 ` Xiaoyao Li
2025-04-08 0:50 ` Chenyi Qiang
2025-04-09 5:35 ` Alexey Kardashevskiy
2025-04-09 5:52 ` Chenyi Qiang
2025-04-25 12:35 ` David Hildenbrand
2025-04-07 7:49 ` [PATCH v4 03/13] memory: Unify the definiton of ReplayRamPopulate() and ReplayRamDiscard() Chenyi Qiang
2025-04-09 5:43 ` Alexey Kardashevskiy
2025-04-09 6:56 ` Chenyi Qiang
2025-04-25 12:44 ` David Hildenbrand
2025-04-25 12:42 ` David Hildenbrand
2025-04-27 2:13 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 04/13] memory: Introduce generic state change parent class for RamDiscardManager Chenyi Qiang
2025-04-09 9:56 ` Alexey Kardashevskiy
2025-04-09 12:57 ` Chenyi Qiang
2025-04-10 0:11 ` Alexey Kardashevskiy
2025-04-10 1:44 ` Chenyi Qiang
2025-04-16 3:32 ` Chenyi Qiang
2025-04-17 23:10 ` Alexey Kardashevskiy
2025-04-18 3:49 ` Chenyi Qiang
2025-04-25 12:54 ` David Hildenbrand
2025-04-25 12:49 ` David Hildenbrand
2025-04-27 1:33 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 05/13] memory: Introduce PrivateSharedManager Interface as child of GenericStateManager Chenyi Qiang
2025-04-09 9:56 ` Alexey Kardashevskiy
2025-04-10 3:47 ` Chenyi Qiang
2025-04-25 12:57 ` David Hildenbrand
2025-04-27 1:40 ` Chenyi Qiang
2025-04-29 10:01 ` David Hildenbrand
2025-04-07 7:49 ` [PATCH v4 06/13] vfio: Add the support for PrivateSharedManager Interface Chenyi Qiang
2025-04-09 9:58 ` Alexey Kardashevskiy
2025-04-10 5:53 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 07/13] ram-block-attribute: Introduce RamBlockAttribute to manage RAMBLock with guest_memfd Chenyi Qiang
2025-04-09 9:57 ` Alexey Kardashevskiy
2025-04-10 7:37 ` Chenyi Qiang
2025-05-09 6:41 ` Baolu Lu
2025-05-09 7:55 ` Chenyi Qiang
2025-05-09 8:18 ` David Hildenbrand
2025-05-09 10:37 ` Chenyi Qiang
2025-05-12 8:07 ` Zhao Liu
2025-05-12 9:43 ` Chenyi Qiang
2025-05-13 8:31 ` Zhao Liu
2025-05-14 1:39 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 08/13] ram-block-attribute: Introduce a callback to notify shared/private state changes Chenyi Qiang
2025-04-07 7:49 ` Chenyi Qiang [this message]
2025-04-07 7:49 ` [PATCH v4 10/13] memory: Change NotifyStateClear() definition to return the result Chenyi Qiang
2025-04-27 2:26 ` Chenyi Qiang
2025-05-09 2:38 ` Chao Gao
2025-05-09 8:20 ` David Hildenbrand
2025-05-09 9:19 ` Chenyi Qiang
2025-05-09 8:22 ` Baolu Lu
2025-05-09 10:04 ` Chenyi Qiang
2025-05-12 7:54 ` David Hildenbrand
2025-04-07 7:49 ` [PATCH v4 11/13] KVM: Introduce CVMPrivateSharedListener for attribute changes during page conversions Chenyi Qiang
2025-05-09 9:03 ` Baolu Lu
2025-05-12 3:18 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 12/13] ram-block-attribute: Add priority listener support for PrivateSharedListener Chenyi Qiang
2025-05-09 9:23 ` Baolu Lu
2025-05-09 9:39 ` Chenyi Qiang
2025-04-07 7:49 ` [PATCH v4 13/13] RAMBlock: Make guest_memfd require coordinate discard Chenyi Qiang
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=20250407074939.18657-10-chenyi.qiang@intel.com \
--to=chenyi.qiang@intel.com \
--cc=aik@amd.com \
--cc=chao.gao@intel.com \
--cc=chao.p.peng@intel.com \
--cc=dan.j.williams@intel.com \
--cc=david@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=michael.roth@amd.com \
--cc=pankaj.gupta@amd.com \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=xiaoyao.li@intel.com \
--cc=yilun.xu@intel.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 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).