Commit Graph

9 Commits

Author SHA1 Message Date
Shuo A Liu
98c80d75b8 hv: dm: Use new virtual device management ioctls
IC_ADD_HV_VDEV		->	ACRN_IOCTL_CREATE_VDEV
IC_REMOVE_HV_VDEV	->	ACRN_IOCTL_DESTROY_VDEV
struct acrn_emul_dev	->	struct acrn_vdev

Also, move struct acrn_vdev to acrn_common.h as this structure is used
by both DM and HV.

Tracked-On: #6282
Signed-off-by: Shuo A Liu <shuo.a.liu@intel.com>
2021-07-15 11:53:54 +08:00
Shuo A Liu
9c910bae44 hv: dm: Use new I/O request data structures
struct vhm_request		->	struct acrn_io_request
union vhm_request_buffer	->	struct acrn_io_request_buffer
struct pio_request		->	struct acrn_pio_request
struct mmio_request		->	struct acrn_mmio_request
struct ioreq_notify		->	struct acrn_ioreq_notify

VHM_REQ_PIO_INVAL		->	IOREQ_PIO_INVAL
VHM_REQ_MMIO_INVAL		->	IOREQ_MMIO_INVAL
REQ_PORTIO			->	ACRN_IOREQ_TYPE_PORTIO
REQ_MMIO			->	ACRN_IOREQ_TYPE_MMIO
REQ_PCICFG			->	ACRN_IOREQ_TYPE_PCICFG
REQ_WP				->	ACRN_IOREQ_TYPE_WP

REQUEST_READ			->	ACRN_IOREQ_DIR_READ
REQUEST_WRITE			->	ACRN_IOREQ_DIR_WRITE
REQ_STATE_PROCESSING		->	ACRN_IOREQ_STATE_PROCESSING
REQ_STATE_PENDING		->	ACRN_IOREQ_STATE_PENDING
REQ_STATE_COMPLETE		->	ACRN_IOREQ_STATE_COMPLETE
REQ_STATE_FREE			->	ACRN_IOREQ_STATE_FREE

IC_CREATE_IOREQ_CLIENT		->	ACRN_IOCTL_CREATE_IOREQ_CLIENT
IC_DESTROY_IOREQ_CLIENT		->	ACRN_IOCTL_DESTROY_IOREQ_CLIENT
IC_ATTACH_IOREQ_CLIENT		->	ACRN_IOCTL_ATTACH_IOREQ_CLIENT
IC_NOTIFY_REQUEST_FINISH	->	ACRN_IOCTL_NOTIFY_REQUEST_FINISH
IC_CLEAR_VM_IOREQ		->	ACRN_IOCTL_CLEAR_VM_IOREQ
HYPERVISOR_CALLBACK_VHM_VECTOR	->	HYPERVISOR_CALLBACK_HSM_VECTOR

arch_fire_vhm_interrupt()	->	arch_fire_hsm_interrupt()
get_vhm_notification_vector()	->	get_hsm_notification_vector()
set_vhm_notification_vector()	->	set_hsm_notification_vector()
acrn_vhm_notification_vector	->	acrn_hsm_notification_vector
get_vhm_req_state()		->	get_io_req_state()
set_vhm_req_state()		->	set_io_req_state()

Below structures have slight difference with former ones.

  struct acrn_ioreq_notify
  strcut acrn_io_request

Tracked-On: #6282
Signed-off-by: Shuo A Liu <shuo.a.liu@intel.com>
2021-07-15 11:53:54 +08:00
Tao Yuhong
cb75de2163 HV: vpci: refine vbar sizing
For a pci BAR, its size aligned bits have fixed to 0(except the memory
type bits, they have another fixed value), they are read-only.
When write ~0U to BAR for sizing, (type_bits | size_mask) is written
into BAR.
So do not need to distinguish between sizing vBAR and programming vBAR.
When write a value to vBAR, always store (value & size_mask | type_bit)
to vfcg.
pci_vdev_read_vbar() is unnecessary, because it is only need to read
vcfg.

Tracked-On: #6011
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Eddie Dong <eddie.dong@intel.com>
Reviewed-by: Li Fei <fei1.li@intel.com>
2021-06-08 08:39:01 +08:00
Liang Yi
688a41c290 hv: mod: do not use explicit arch name when including headers
Instead of "#include <x86/foo.h>", use "#include <asm/foo.h>".

In other words, we are adopting the same practice in Linux kernel.

Tracked-On: #5920
Signed-off-by: Liang Yi <yi.liang@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
2021-05-08 11:15:46 +08:00
Liang Yi
33ef656462 hv/mod-irq: use arch specific header files
Requires explicit arch path name in the include directive.

The config scripts was also updated to reflect this change.

Tracked-On: #5825
Signed-off-by: Peter Fang <peter.fang@intel.com>
Reviewed-by: Jason Chen CJ <jason.cj.chen@intel.com>
2021-03-24 11:38:14 +08:00
Tao Yuhong
84752ab229 Fix: HV: VM OS failed to assign new address to pci-vuart BARs
When wrong BAR address is set for pci-vuart, OS may assign a
new BAR address to it. Pci-vuart BAR can't be reprogrammed,
for its wrong fixed value. That can may because pci_vbar.fixed and
pci_vbar.type has overlap in abstraction, pci_vbar.fixed
has a confusing name, pci_vbar.type has PCIBAR_MEM64HI which is not
really a type of pci BARs.
So replace pci_vbar.type with pci_vbar.is_mem64hi, and change
pci_vbar.fixed to an union type with new name pci_vbar.bar_type.

Tracked-On: #5491
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
2021-01-08 17:20:56 +08:00
Tao Yuhong
996e8f680c HV: pci-vuart support create vdev hcall
Add cteate method for vmcs9900 vdev in hypercalls.

The destroy method of ivshmem is also suitable for other emulated vdev,
move it into hcall_destroy_vdev() for all emulated vdevs

Tracked-On: #5394
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Wang, Yu1 <yu1.wang@intel.com>
2020-10-30 20:41:34 +08:00
Tao Yuhong
691abe90ff HV: vuart: send msi for pci vuart type
if vuart type is pci-vuart, then use MSI interrupt

split vuart_toggle_intr() control flow into vuart_trigger_level_intr() &
trigger_vmcs9900_msix(), because MSI is edge triggered, no deassertion
operation. Only trigger MSI for pci-vuart when assert interrupt.

Tracked-On: #5394
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Wang, Yu1 <yu1.wang@intel.com>
Acked-by: Eddie Dong <eddie.dong@Intel.com>
2020-10-30 20:41:34 +08:00
Tao Yuhong
55b7fae67a HV: pci-vuart: pci based vuart emulation
Add emulation for pci based vuart device mcs9900 at hv land.
add struct pci_vdev_ops vuart_pci_ops, the vdev callbalks for vuart.

How to use
In misc/vm_configs/scenarios/<SCENARIO>/<BOARD>/pci_dev.c, add pci
vuart config to vm_pci_devs[] array. For example:

struct acrn_vm_pci_dev_config vm0_pci_devs[] = {
       /* console vuart setting*/
       {
               .emu_type = PCI_DEV_TYPE_HVEMUL,
               .vbdf.bits = {.b = 0x00U, .d = 0x04U, .f = 0x00U},
               .vdev_ops = &vmcs_ops,
               .vbar_base[0] = 0x80001000,	/* mmio bar */
               .vbar_base[1] = 0x80002000,	/* msix bar */
               .vuart_idx = 0,
       },
       /* communication vuart setting */
       {
               .emu_type = PCI_DEV_TYPE_HVEMUL,
               .vbdf.bits = {.b = 0x00U, .d = 0x05U, .f = 0x00U},
               .vdev_ops = &vmcs_ops,
               .vbar_base[0] = 0x80003000,
               .vbar_base[1] = 0x80004000,
               .vuart_idx = 1,
               .t_vuart.vm_id = 1U,
               .t_vuart.vuart_id = 1U,
       },
}

Tracked-On: #5394
Signed-off-by: Tao Yuhong <yuhong.tao@intel.com>
Reviewed-by: Wang, Yu1 <yu1.wang@intel.com>
Acked-by: Eddie Dong <eddie.dong@Intel.com>
2020-10-30 20:41:34 +08:00