Skip to contentSkip to conversation

libxl: Add suppress-vmdesc to QEMU machine

PATCH In review Jason Andryuk 3 participants 4 messages Oct 29, 2020 – Nov 3, 2020

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 libxl: Add suppress-vmdesc to QEMU machine

The device model state saved by QMP xen-save-devices-state doesn't include the vmdesc json. When restoring an HVM, xen-load-devices-state always triggers "Expected vmdescription section, but got 0". This is not a problem when restore comes from a file. However, when QEMU runs in a linux stubdom and comes over a console, EOF is not received. This causes a delay restoring - though it does restore.

Setting suppress-vmdesc skips looking for the vmdesc during restore and avoids the wait.

QEMU 5.2 enables suppress-vmdesc by default for xenfv, but this change sets it manually for xenfv and xen_platform_pci=0 when -machine pc is use.

QEMU commit 9850c6047b8b "migration: Allow to suppress vmdesc submission" added suppress-vmdesc in QEMU 2.3.

---
QEMU 2.3 came out in 2015, so setting suppress-vmdesc unilaterally should be okay... Is this okay?

---
 tools/libs/light/libxl_dm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/libs/light/libxl_dm.c +2 -2
index d1ff35dda3..3da83259c0 100644
@@ -1778,9 +1778,9 @@ static int libxl__build_device_model_args_new(libxl__gc *gc,
             /* Switching here to the machine "pc" which does not add
              * the xen-platform device instead of the default "xenfv" machine.
              */
-            machinearg = libxl__strdup(gc, "pc,accel=xen");
+            machinearg = libxl__strdup(gc, "pc,accel=xen,suppress-vmdesc=on");
         } else {
-            machinearg = libxl__strdup(gc, "xenfv");
+            machinearg = libxl__strdup(gc, "xenfv,suppress-vmdesc=on");
         }
         if (b_info->u.hvm.mmio_hole_memkb) {
             uint64_t max_ram_below_4g = (1ULL << 32) -
Signature · 1 line
2.25.1
#1 lore raw
Quoted 14 lines · Jason Andryuk

The device model state saved by QMP xen-save-devices-state doesn't include the vmdesc json. When restoring an HVM, xen-load-devices-state always triggers "Expected vmdescription section, but got 0". This is not a problem when restore comes from a file. However, when QEMU runs in a linux stubdom and comes over a console, EOF is not received. This causes a delay restoring - though it does restore.

Setting suppress-vmdesc skips looking for the vmdesc during restore and avoids the wait.

QEMU 5.2 enables suppress-vmdesc by default for xenfv, but this change sets it manually for xenfv and xen_platform_pci=0 when -machine pc is use.

Quoted 8 lines

QEMU commit 9850c6047b8b "migration: Allow to suppress vmdesc submission" added suppress-vmdesc in QEMU 2.3.

---
QEMU 2.3 came out in 2015, so setting suppress-vmdesc unilaterally should be okay... Is this okay?

Anthony, what is your opinion on this?

Wei.

#2 lore raw
Quoted 11 lines · Wei Liu
Quoted 10 lines · Jason Andryuk

The device model state saved by QMP xen-save-devices-state doesn't include the vmdesc json. When restoring an HVM, xen-load-devices-state always triggers "Expected vmdescription section, but got 0". This is not a problem when restore comes from a file. However, when QEMU runs in a linux stubdom and comes over a console, EOF is not received. This causes a delay restoring - though it does restore.

Setting suppress-vmdesc skips looking for the vmdesc during restore and avoids the wait.

Quoted 8 lines
Quoted 8 lines

QEMU 5.2 enables suppress-vmdesc by default for xenfv, but this change sets it manually for xenfv and xen_platform_pci=0 when -machine pc is use.

QEMU commit 9850c6047b8b "migration: Allow to suppress vmdesc submission" added suppress-vmdesc in QEMU 2.3.

:-(, sorry, I never received that email.

Quoted 6 lines
Quoted 3 lines

---
QEMU 2.3 came out in 2015, so setting suppress-vmdesc unilaterally should be okay... Is this okay?

Anthony, what is your opinion on this?

That it's fine, and I actually asked for the libxl patch. For reference, QEMU 2.3 is in qemu-xen-4.7.

So,

Thanks,

Signature · 1 line
Anthony PERARD
#3 lore raw
Quoted 17 lines · Anthony PERARD
Quoted 15 lines · Wei Liu
Quoted 14 lines · Jason Andryuk

The device model state saved by QMP xen-save-devices-state doesn't include the vmdesc json. When restoring an HVM, xen-load-devices-state always triggers "Expected vmdescription section, but got 0". This is not a problem when restore comes from a file. However, when QEMU runs in a linux stubdom and comes over a console, EOF is not received. This causes a delay restoring - though it does restore.

Setting suppress-vmdesc skips looking for the vmdesc during restore and avoids the wait.

QEMU 5.2 enables suppress-vmdesc by default for xenfv, but this change sets it manually for xenfv and xen_platform_pci=0 when -machine pc is use.

Quoted 6 lines
Quoted 4 lines
Quoted 4 lines

QEMU commit 9850c6047b8b "migration: Allow to suppress vmdesc submission" added suppress-vmdesc in QEMU 2.3.

:-(, sorry, I never received that email.

Sorry about that. I included your address , but Gmail said:
"""
Address not found

Your message wasn't delivered to [email protected] because the address couldn't be found, or is unable to receive mail.

The response from the remote server was:

550 Too many invalid recipients
"""

Quoted 5 lines
Quoted 3 lines
Quoted 3 lines

---
QEMU 2.3 came out in 2015, so setting suppress-vmdesc unilaterally should be okay... Is this okay?

Quoted 7 lines
Quoted 1 line

Anthony, what is your opinion on this?

That it's fine, and I actually asked for the libxl patch. For reference, QEMU 2.3 is in qemu-xen-4.7.

So,

Thanks,

Jason