Cancel this month’s modularization topic, thanks.WW25ACRN Modularization IntroductionChen Jason6/15/2021 Special Notes: If you have Zoom connection issue by usingweb browser, please install &

Title: dm: xhci: Add max ep index checking in doorbell write emulation Check max ep index in doorbell write function to avoid overwriting the pci_xhci_dev_emu->eps[XHCI_MAX_ENDPOINTS]. UPRINTF(LWRN,

Title: dm: vsock: fix the memory leakage bug Fix the memory leakage issue caused by incomplete deinit flow. Acked-by: Wang, Yu1 <yu1.wang@...>

Add the endpint index value check in doorbell write function, because the max index value is 32. Signed-off-by: Liu Long <long.liu@...> --- devicemodel/hw/pci/xhci.c | 4 ++++ 1 file

In deinit function close the vhost_vsock fd and free the vhost_vsock struct no matter the above logic. Signed-off-by: Liu Long <long.liu@...> --- devicemodel/hw/pci/virtio/vhost_vsock.c

v1-->v2 1. Refine the vhost_vsock_devinit function return void 2. Add the endpoint index value check log. Liu Long (2): ACRN: DM: Refine the vhost_vsock_deinit function ACRN: DM: Add endpoint

Hi Yu,

In the vhost_vsock_deinit, it already close and free, why do it again?

Is there any capability register to indicate the max supported endpoint number? I haven't find it. And why FE driver ring the nonexisting ep? Another thing, this is not controller halted case,

return 0 for success or fail?

Because the endpoint index max value is 32, add the index value check in xhci register write function. Signed-off-by: Liu Long <long.liu@...> --- devicemodel/hw/pci/xhci.c | 2 +- 1 file

When deinit the vhost vsock release the vhost vsock struct to prevent memory leak. Signed-off-by: Liu Long <long.liu@...> --- devicemodel/hw/pci/virtio/vhost_vsock.c | 14 ++++++++------

1.Release the Vhost vsock when deinit the vsock. 2.Add endpoint index value check in xhci doorbell register write Liu Long (2): ACRN: DM: Release vsock struct in case memory leak ACRN: DM: Add

Sorry ，ignore this patch, will send out the formal version later.

Where's your commit message? Is this a debug patch? Why there are "XXXX" print?

Signed-off-by: Liu Long <long.liu@...> --- devicemodel/hw/pci/xhci.c | 1 + devicemodel/hw/platform/usb_mouse.c | 1 + devicemodel/hw/platform/usb_pmapper.c | 27

Cannot Yes. That's better. Please PR with the fix. Acked-by: Wang, Yu1 <yu1.wang@...>

update the hugepages algorithm as the following steps. 1. calculate the total hugepages of service vm using the formula. "total memory*0.98 - the memory consumed by pre-launched VMs -