From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
"Johannes Berg" <johannes.berg@intel.com>,
"David Hildenbrand" <david@redhat.com>,
"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
"Raphael Norwitz" <raphael.norwitz@nutanix.com>,
"Stefan Hajnoczi" <stefanha@redhat.com>,
"Marc-André Lureau" <marcandre.lureau@redhat.com>
Subject: [PULL 44/56] Support individual region unmap in libvhost-user
Date: Wed, 10 Jun 2020 00:28:11 -0400 [thread overview]
Message-ID: <20200610042613.1459309-45-mst@redhat.com> (raw)
In-Reply-To: <20200610042613.1459309-1-mst@redhat.com>
From: Raphael Norwitz <raphael.norwitz@nutanix.com>
When the VHOST_USER_PROTOCOL_F_CONFIGURE_MEM_SLOTS protocol feature is
enabled, on memory hot-unplug qemu will transmit memory regions to
remove individually using the new message VHOST_USER_REM_MEM_REG
message. With this change, vhost-user backends build with libvhost-user
can now unmap individual memory regions when receiving the
VHOST_USER_REM_MEM_REG message.
Qemu only sends VHOST_USER_REM_MEM_REG messages when the
VHOST_USER_PROTOCOL_F_CONFIGURE_MEM_SLOTS feature is negotiated, and
support for that feature has not yet been added in libvhost-user, this
new functionality is not yet used.
Signed-off-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Message-Id: <1588533678-23450-10-git-send-email-raphael.norwitz@nutanix.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
contrib/libvhost-user/libvhost-user.h | 1 +
contrib/libvhost-user/libvhost-user.c | 63 +++++++++++++++++++++++++++
2 files changed, 64 insertions(+)
diff --git a/contrib/libvhost-user/libvhost-user.h b/contrib/libvhost-user/libvhost-user.h
index 60ef7fd13e..f8439713a8 100644
--- a/contrib/libvhost-user/libvhost-user.h
+++ b/contrib/libvhost-user/libvhost-user.h
@@ -99,6 +99,7 @@ typedef enum VhostUserRequest {
VHOST_USER_VRING_KICK = 35,
VHOST_USER_GET_MAX_MEM_SLOTS = 36,
VHOST_USER_ADD_MEM_REG = 37,
+ VHOST_USER_REM_MEM_REG = 38,
VHOST_USER_MAX
} VhostUserRequest;
diff --git a/contrib/libvhost-user/libvhost-user.c b/contrib/libvhost-user/libvhost-user.c
index d8ee7a23a3..386449b697 100644
--- a/contrib/libvhost-user/libvhost-user.c
+++ b/contrib/libvhost-user/libvhost-user.c
@@ -139,6 +139,7 @@ vu_request_to_string(unsigned int req)
REQ(VHOST_USER_VRING_KICK),
REQ(VHOST_USER_GET_MAX_MEM_SLOTS),
REQ(VHOST_USER_ADD_MEM_REG),
+ REQ(VHOST_USER_REM_MEM_REG),
REQ(VHOST_USER_MAX),
};
#undef REQ
@@ -763,6 +764,66 @@ vu_add_mem_reg(VuDev *dev, VhostUserMsg *vmsg) {
}
}
+static inline bool reg_equal(VuDevRegion *vudev_reg,
+ VhostUserMemoryRegion *msg_reg)
+{
+ if (vudev_reg->gpa == msg_reg->guest_phys_addr &&
+ vudev_reg->qva == msg_reg->userspace_addr &&
+ vudev_reg->size == msg_reg->memory_size) {
+ return true;
+ }
+
+ return false;
+}
+
+static bool
+vu_rem_mem_reg(VuDev *dev, VhostUserMsg *vmsg) {
+ int i, j;
+ bool found = false;
+ VuDevRegion shadow_regions[VHOST_MEMORY_MAX_NREGIONS] = {};
+ VhostUserMemoryRegion m = vmsg->payload.memreg.region, *msg_region = &m;
+
+ DPRINT("Removing region:\n");
+ DPRINT(" guest_phys_addr: 0x%016"PRIx64"\n",
+ msg_region->guest_phys_addr);
+ DPRINT(" memory_size: 0x%016"PRIx64"\n",
+ msg_region->memory_size);
+ DPRINT(" userspace_addr 0x%016"PRIx64"\n",
+ msg_region->userspace_addr);
+ DPRINT(" mmap_offset 0x%016"PRIx64"\n",
+ msg_region->mmap_offset);
+
+ for (i = 0, j = 0; i < dev->nregions; i++) {
+ if (!reg_equal(&dev->regions[i], msg_region)) {
+ shadow_regions[j].gpa = dev->regions[i].gpa;
+ shadow_regions[j].size = dev->regions[i].size;
+ shadow_regions[j].qva = dev->regions[i].qva;
+ shadow_regions[j].mmap_offset = dev->regions[i].mmap_offset;
+ j++;
+ } else {
+ found = true;
+ VuDevRegion *r = &dev->regions[i];
+ void *m = (void *) (uintptr_t) r->mmap_addr;
+
+ if (m) {
+ munmap(m, r->size + r->mmap_offset);
+ }
+ }
+ }
+
+ if (found) {
+ memcpy(dev->regions, shadow_regions,
+ sizeof(VuDevRegion) * VHOST_MEMORY_MAX_NREGIONS);
+ DPRINT("Successfully removed a region\n");
+ dev->nregions--;
+ vmsg_set_reply_u64(vmsg, 0);
+ } else {
+ vu_panic(dev, "Specified region not found\n");
+ }
+
+ return true;
+}
+
static bool
vu_set_mem_table_exec_postcopy(VuDev *dev, VhostUserMsg *vmsg)
{
@@ -1771,6 +1832,8 @@ vu_process_message(VuDev *dev, VhostUserMsg *vmsg)
return vu_handle_get_max_memslots(dev, vmsg);
case VHOST_USER_ADD_MEM_REG:
return vu_add_mem_reg(dev, vmsg);
+ case VHOST_USER_REM_MEM_REG:
+ return vu_rem_mem_reg(dev, vmsg);
default:
vmsg_close_fds(vmsg);
vu_panic(dev, "Unhandled request: %d", vmsg->request);
--
MST
next prev parent reply other threads:[~2020-06-10 4:57 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-10 4:26 [PULL 00/56] virtio,acpi,pci: features, fixes, cleanups, tests Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 01/56] msix: allow qword MSI-X table accesses Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 02/56] diffs-allowed: add the SRAT AML to diffs-allowed Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 03/56] hw/acpi/nvdimm: add a helper to augment SRAT generation Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 04/56] tests/acpi: update expected SRAT files Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 05/56] qtest: allow DSDT acpi table changes Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 06/56] acpi: move aml builder code for rtc device Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 07/56] acpi: rtc: use a single crs range Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 08/56] acpi: serial: don't use _STA method Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 09/56] acpi: move aml builder code for serial device Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 10/56] acpi: parallel: don't use _STA method Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 11/56] acpi: move aml builder code for parallel device Michael S. Tsirkin
2020-06-10 4:26 ` [PULL 12/56] tests/acpi: update DSDT expected files Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 13/56] acpi: tpm: Do not build TCPA table for TPM 2 Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 14/56] acpi: Convert build_tpm2() to build_append* API Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 15/56] acpi: Move build_tpm2() in the generic part Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 16/56] arm/acpi: TPM2 ACPI table support Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 17/56] test/tpm-emu: include sockets and channel headers in tpm-emu header Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 18/56] tests/acpi: Add void tables for Q35/TPM-TIS bios-tables-test Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 19/56] tests: tpm-emu: Remove assert on TPM2_ST_NO_SESSIONS Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 20/56] bios-tables-test: Add Q35/TPM-TIS test Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 21/56] bios-tables-test: Generate reference tables for Q35/TPM-TIS Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 22/56] virtio-balloon: fix free page hinting without an iothread Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 23/56] virtio-balloon: fix free page hinting check on unrealize Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 24/56] virtio-balloon: unref the iothread when unrealizing Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 25/56] virtio-balloon: Implement support for page poison reporting feature Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 26/56] virtio-balloon: Provide an interface for free page reporting Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 27/56] MAINTAINERS: Fix the classification of bios-tables-test-allowed-diff.h Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 28/56] hw/pci/pcie: Move hot plug capability check to pre_plug callback Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 29/56] pci: assert configuration access is within bounds Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 30/56] hw/pci-host/prep: Correct RAVEN bus bridge memory region size Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 31/56] hw/pci/pci_bridge: Correct pci_bridge_io " Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 32/56] hw/pci/pci_bridge: Use the IEC binary prefix definitions Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 33/56] hw/pci-host: " Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 34/56] char-socket: return -1 in case of disconnect during tcp_chr_write Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 35/56] vhost-user-blk: delay vhost_user_blk_disconnect Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 36/56] Add helper to populate vhost-user message regions Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 37/56] Add vhost-user helper to get MemoryRegion data Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 38/56] Add VHOST_USER_PROTOCOL_F_CONFIGURE_MEM_SLOTS Michael S. Tsirkin
2020-06-10 4:27 ` [PULL 39/56] Transmit vhost-user memory regions individually Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 40/56] Lift max memory slots limit imposed by vhost-user Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 41/56] Refactor out libvhost-user fault generation logic Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 42/56] Support ram slot configuration in libvhost-user Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 43/56] Support adding individual regions " Michael S. Tsirkin
2020-06-10 4:28 ` Michael S. Tsirkin [this message]
2020-06-10 4:28 ` [PULL 45/56] Lift max ram slots limit " Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 46/56] libvhost-user: advertise vring features Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 47/56] hw/pci: Fix crash when running QEMU with "-nic model=rocker" Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 48/56] vhost-vsock: add vhost-vsock-common abstraction Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 49/56] virtio: add vhost-user-vsock base device Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 50/56] virtio: add vhost-user-vsock-pci device Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 51/56] acpi: make build_madt() more generic Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 52/56] acpi: create acpi-common.c and move madt code Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 53/56] acpi: madt: skip pci override on pci-less systems Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 54/56] acpi: fadt: add hw-reduced sleep register support Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 55/56] acpi: ged: rename event memory region Michael S. Tsirkin
2020-06-10 4:28 ` [PULL 56/56] Fix parameter type in vhost migration log path Michael S. Tsirkin
2020-06-10 5:30 ` [PULL 00/56] virtio,acpi,pci: features, fixes, cleanups, tests no-reply
2020-06-11 18:13 ` Peter Maydell
2020-06-12 16:12 ` Michael S. Tsirkin
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=20200610042613.1459309-45-mst@redhat.com \
--to=mst@redhat.com \
--cc=david@redhat.com \
--cc=dgilbert@redhat.com \
--cc=johannes.berg@intel.com \
--cc=marcandre.lureau@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=raphael.norwitz@nutanix.com \
--cc=stefanha@redhat.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).