mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-28 11:18:45 +07:00
d480b28a41
When add 'x-no-mmap=on' for vfio-pci option, aperture access in guest
is emulated. But the vgpu_aperture_rw() function take wrong offset when
do memcpy, since vgpu->gm.aperture_va is not the base of entire aperture.
This mistake cause GPU command in guest get lost and so the seqno is not
updated in engine HWSP.
This patch fix this, and it also move the emulation code to kvmgt.
Because only vfio need to emulate it. Put aperture rw to MMIO emulation
path breaks assumptions in xengt.
v2: Remove PAGE_ALIGN for size (zhenyu)
Fixes: f090a00df9
("drm/i915/gvt: Add emulation for BAR2 (aperture) with normal file RW approach")
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
413 lines
11 KiB
C
413 lines
11 KiB
C
/*
|
|
* Copyright(c) 2011-2016 Intel Corporation. All rights reserved.
|
|
*
|
|
* Permission is hereby granted, free of charge, to any person obtaining a
|
|
* copy of this software and associated documentation files (the "Software"),
|
|
* to deal in the Software without restriction, including without limitation
|
|
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
|
* and/or sell copies of the Software, and to permit persons to whom the
|
|
* Software is furnished to do so, subject to the following conditions:
|
|
*
|
|
* The above copyright notice and this permission notice (including the next
|
|
* paragraph) shall be included in all copies or substantial portions of the
|
|
* Software.
|
|
*
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
* SOFTWARE.
|
|
*
|
|
* Authors:
|
|
* Eddie Dong <eddie.dong@intel.com>
|
|
* Jike Song <jike.song@intel.com>
|
|
*
|
|
* Contributors:
|
|
* Zhi Wang <zhi.a.wang@intel.com>
|
|
* Min He <min.he@intel.com>
|
|
* Bing Niu <bing.niu@intel.com>
|
|
*
|
|
*/
|
|
|
|
#include "i915_drv.h"
|
|
#include "gvt.h"
|
|
|
|
enum {
|
|
INTEL_GVT_PCI_BAR_GTTMMIO = 0,
|
|
INTEL_GVT_PCI_BAR_APERTURE,
|
|
INTEL_GVT_PCI_BAR_PIO,
|
|
INTEL_GVT_PCI_BAR_MAX,
|
|
};
|
|
|
|
/* bitmap for writable bits (RW or RW1C bits, but cannot co-exist in one
|
|
* byte) byte by byte in standard pci configuration space. (not the full
|
|
* 256 bytes.)
|
|
*/
|
|
static const u8 pci_cfg_space_rw_bmp[PCI_INTERRUPT_LINE + 4] = {
|
|
[PCI_COMMAND] = 0xff, 0x07,
|
|
[PCI_STATUS] = 0x00, 0xf9, /* the only one RW1C byte */
|
|
[PCI_CACHE_LINE_SIZE] = 0xff,
|
|
[PCI_BASE_ADDRESS_0 ... PCI_CARDBUS_CIS - 1] = 0xff,
|
|
[PCI_ROM_ADDRESS] = 0x01, 0xf8, 0xff, 0xff,
|
|
[PCI_INTERRUPT_LINE] = 0xff,
|
|
};
|
|
|
|
/**
|
|
* vgpu_pci_cfg_mem_write - write virtual cfg space memory
|
|
*
|
|
* Use this function to write virtual cfg space memory.
|
|
* For standard cfg space, only RW bits can be changed,
|
|
* and we emulates the RW1C behavior of PCI_STATUS register.
|
|
*/
|
|
static void vgpu_pci_cfg_mem_write(struct intel_vgpu *vgpu, unsigned int off,
|
|
u8 *src, unsigned int bytes)
|
|
{
|
|
u8 *cfg_base = vgpu_cfg_space(vgpu);
|
|
u8 mask, new, old;
|
|
int i = 0;
|
|
|
|
for (; i < bytes && (off + i < sizeof(pci_cfg_space_rw_bmp)); i++) {
|
|
mask = pci_cfg_space_rw_bmp[off + i];
|
|
old = cfg_base[off + i];
|
|
new = src[i] & mask;
|
|
|
|
/**
|
|
* The PCI_STATUS high byte has RW1C bits, here
|
|
* emulates clear by writing 1 for these bits.
|
|
* Writing a 0b to RW1C bits has no effect.
|
|
*/
|
|
if (off + i == PCI_STATUS + 1)
|
|
new = (~new & old) & mask;
|
|
|
|
cfg_base[off + i] = (old & ~mask) | new;
|
|
}
|
|
|
|
/* For other configuration space directly copy as it is. */
|
|
if (i < bytes)
|
|
memcpy(cfg_base + off + i, src + i, bytes - i);
|
|
}
|
|
|
|
/**
|
|
* intel_vgpu_emulate_cfg_read - emulate vGPU configuration space read
|
|
*
|
|
* Returns:
|
|
* Zero on success, negative error code if failed.
|
|
*/
|
|
int intel_vgpu_emulate_cfg_read(struct intel_vgpu *vgpu, unsigned int offset,
|
|
void *p_data, unsigned int bytes)
|
|
{
|
|
if (WARN_ON(bytes > 4))
|
|
return -EINVAL;
|
|
|
|
if (WARN_ON(offset + bytes > vgpu->gvt->device_info.cfg_space_size))
|
|
return -EINVAL;
|
|
|
|
memcpy(p_data, vgpu_cfg_space(vgpu) + offset, bytes);
|
|
return 0;
|
|
}
|
|
|
|
static int map_aperture(struct intel_vgpu *vgpu, bool map)
|
|
{
|
|
phys_addr_t aperture_pa = vgpu_aperture_pa_base(vgpu);
|
|
unsigned long aperture_sz = vgpu_aperture_sz(vgpu);
|
|
u64 first_gfn;
|
|
u64 val;
|
|
int ret;
|
|
|
|
if (map == vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_APERTURE].tracked)
|
|
return 0;
|
|
|
|
val = vgpu_cfg_space(vgpu)[PCI_BASE_ADDRESS_2];
|
|
if (val & PCI_BASE_ADDRESS_MEM_TYPE_64)
|
|
val = *(u64 *)(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_2);
|
|
else
|
|
val = *(u32 *)(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_2);
|
|
|
|
first_gfn = (val + vgpu_aperture_offset(vgpu)) >> PAGE_SHIFT;
|
|
|
|
ret = intel_gvt_hypervisor_map_gfn_to_mfn(vgpu, first_gfn,
|
|
aperture_pa >> PAGE_SHIFT,
|
|
aperture_sz >> PAGE_SHIFT,
|
|
map);
|
|
if (ret)
|
|
return ret;
|
|
|
|
vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_APERTURE].tracked = map;
|
|
return 0;
|
|
}
|
|
|
|
static int trap_gttmmio(struct intel_vgpu *vgpu, bool trap)
|
|
{
|
|
u64 start, end;
|
|
u64 val;
|
|
int ret;
|
|
|
|
if (trap == vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].tracked)
|
|
return 0;
|
|
|
|
val = vgpu_cfg_space(vgpu)[PCI_BASE_ADDRESS_0];
|
|
if (val & PCI_BASE_ADDRESS_MEM_TYPE_64)
|
|
start = *(u64 *)(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_0);
|
|
else
|
|
start = *(u32 *)(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_0);
|
|
|
|
start &= ~GENMASK(3, 0);
|
|
end = start + vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].size - 1;
|
|
|
|
ret = intel_gvt_hypervisor_set_trap_area(vgpu, start, end, trap);
|
|
if (ret)
|
|
return ret;
|
|
|
|
vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].tracked = trap;
|
|
return 0;
|
|
}
|
|
|
|
static int emulate_pci_command_write(struct intel_vgpu *vgpu,
|
|
unsigned int offset, void *p_data, unsigned int bytes)
|
|
{
|
|
u8 old = vgpu_cfg_space(vgpu)[offset];
|
|
u8 new = *(u8 *)p_data;
|
|
u8 changed = old ^ new;
|
|
int ret;
|
|
|
|
vgpu_pci_cfg_mem_write(vgpu, offset, p_data, bytes);
|
|
if (!(changed & PCI_COMMAND_MEMORY))
|
|
return 0;
|
|
|
|
if (old & PCI_COMMAND_MEMORY) {
|
|
ret = trap_gttmmio(vgpu, false);
|
|
if (ret)
|
|
return ret;
|
|
ret = map_aperture(vgpu, false);
|
|
if (ret)
|
|
return ret;
|
|
} else {
|
|
ret = trap_gttmmio(vgpu, true);
|
|
if (ret)
|
|
return ret;
|
|
ret = map_aperture(vgpu, true);
|
|
if (ret)
|
|
return ret;
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int emulate_pci_rom_bar_write(struct intel_vgpu *vgpu,
|
|
unsigned int offset, void *p_data, unsigned int bytes)
|
|
{
|
|
u32 *pval = (u32 *)(vgpu_cfg_space(vgpu) + offset);
|
|
u32 new = *(u32 *)(p_data);
|
|
|
|
if ((new & PCI_ROM_ADDRESS_MASK) == PCI_ROM_ADDRESS_MASK)
|
|
/* We don't have rom, return size of 0. */
|
|
*pval = 0;
|
|
else
|
|
vgpu_pci_cfg_mem_write(vgpu, offset, p_data, bytes);
|
|
return 0;
|
|
}
|
|
|
|
static int emulate_pci_bar_write(struct intel_vgpu *vgpu, unsigned int offset,
|
|
void *p_data, unsigned int bytes)
|
|
{
|
|
u32 new = *(u32 *)(p_data);
|
|
bool lo = IS_ALIGNED(offset, 8);
|
|
u64 size;
|
|
int ret = 0;
|
|
bool mmio_enabled =
|
|
vgpu_cfg_space(vgpu)[PCI_COMMAND] & PCI_COMMAND_MEMORY;
|
|
struct intel_vgpu_pci_bar *bars = vgpu->cfg_space.bar;
|
|
|
|
/*
|
|
* Power-up software can determine how much address
|
|
* space the device requires by writing a value of
|
|
* all 1's to the register and then reading the value
|
|
* back. The device will return 0's in all don't-care
|
|
* address bits.
|
|
*/
|
|
if (new == 0xffffffff) {
|
|
switch (offset) {
|
|
case PCI_BASE_ADDRESS_0:
|
|
case PCI_BASE_ADDRESS_1:
|
|
size = ~(bars[INTEL_GVT_PCI_BAR_GTTMMIO].size -1);
|
|
intel_vgpu_write_pci_bar(vgpu, offset,
|
|
size >> (lo ? 0 : 32), lo);
|
|
/*
|
|
* Untrap the BAR, since guest hasn't configured a
|
|
* valid GPA
|
|
*/
|
|
ret = trap_gttmmio(vgpu, false);
|
|
break;
|
|
case PCI_BASE_ADDRESS_2:
|
|
case PCI_BASE_ADDRESS_3:
|
|
size = ~(bars[INTEL_GVT_PCI_BAR_APERTURE].size -1);
|
|
intel_vgpu_write_pci_bar(vgpu, offset,
|
|
size >> (lo ? 0 : 32), lo);
|
|
ret = map_aperture(vgpu, false);
|
|
break;
|
|
default:
|
|
/* Unimplemented BARs */
|
|
intel_vgpu_write_pci_bar(vgpu, offset, 0x0, false);
|
|
}
|
|
} else {
|
|
switch (offset) {
|
|
case PCI_BASE_ADDRESS_0:
|
|
case PCI_BASE_ADDRESS_1:
|
|
/*
|
|
* Untrap the old BAR first, since guest has
|
|
* re-configured the BAR
|
|
*/
|
|
trap_gttmmio(vgpu, false);
|
|
intel_vgpu_write_pci_bar(vgpu, offset, new, lo);
|
|
ret = trap_gttmmio(vgpu, mmio_enabled);
|
|
break;
|
|
case PCI_BASE_ADDRESS_2:
|
|
case PCI_BASE_ADDRESS_3:
|
|
map_aperture(vgpu, false);
|
|
intel_vgpu_write_pci_bar(vgpu, offset, new, lo);
|
|
ret = map_aperture(vgpu, mmio_enabled);
|
|
break;
|
|
default:
|
|
intel_vgpu_write_pci_bar(vgpu, offset, new, lo);
|
|
}
|
|
}
|
|
return ret;
|
|
}
|
|
|
|
/**
|
|
* intel_vgpu_emulate_cfg_read - emulate vGPU configuration space write
|
|
*
|
|
* Returns:
|
|
* Zero on success, negative error code if failed.
|
|
*/
|
|
int intel_vgpu_emulate_cfg_write(struct intel_vgpu *vgpu, unsigned int offset,
|
|
void *p_data, unsigned int bytes)
|
|
{
|
|
int ret;
|
|
|
|
if (WARN_ON(bytes > 4))
|
|
return -EINVAL;
|
|
|
|
if (WARN_ON(offset + bytes > vgpu->gvt->device_info.cfg_space_size))
|
|
return -EINVAL;
|
|
|
|
/* First check if it's PCI_COMMAND */
|
|
if (IS_ALIGNED(offset, 2) && offset == PCI_COMMAND) {
|
|
if (WARN_ON(bytes > 2))
|
|
return -EINVAL;
|
|
return emulate_pci_command_write(vgpu, offset, p_data, bytes);
|
|
}
|
|
|
|
switch (rounddown(offset, 4)) {
|
|
case PCI_ROM_ADDRESS:
|
|
if (WARN_ON(!IS_ALIGNED(offset, 4)))
|
|
return -EINVAL;
|
|
return emulate_pci_rom_bar_write(vgpu, offset, p_data, bytes);
|
|
|
|
case PCI_BASE_ADDRESS_0 ... PCI_BASE_ADDRESS_5:
|
|
if (WARN_ON(!IS_ALIGNED(offset, 4)))
|
|
return -EINVAL;
|
|
return emulate_pci_bar_write(vgpu, offset, p_data, bytes);
|
|
|
|
case INTEL_GVT_PCI_SWSCI:
|
|
if (WARN_ON(!IS_ALIGNED(offset, 4)))
|
|
return -EINVAL;
|
|
ret = intel_vgpu_emulate_opregion_request(vgpu, *(u32 *)p_data);
|
|
if (ret)
|
|
return ret;
|
|
break;
|
|
|
|
case INTEL_GVT_PCI_OPREGION:
|
|
if (WARN_ON(!IS_ALIGNED(offset, 4)))
|
|
return -EINVAL;
|
|
ret = intel_vgpu_opregion_base_write_handler(vgpu,
|
|
*(u32 *)p_data);
|
|
if (ret)
|
|
return ret;
|
|
|
|
vgpu_pci_cfg_mem_write(vgpu, offset, p_data, bytes);
|
|
break;
|
|
default:
|
|
vgpu_pci_cfg_mem_write(vgpu, offset, p_data, bytes);
|
|
break;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
/**
|
|
* intel_vgpu_init_cfg_space - init vGPU configuration space when create vGPU
|
|
*
|
|
* @vgpu: a vGPU
|
|
* @primary: is the vGPU presented as primary
|
|
*
|
|
*/
|
|
void intel_vgpu_init_cfg_space(struct intel_vgpu *vgpu,
|
|
bool primary)
|
|
{
|
|
struct intel_gvt *gvt = vgpu->gvt;
|
|
const struct intel_gvt_device_info *info = &gvt->device_info;
|
|
u16 *gmch_ctl;
|
|
|
|
memcpy(vgpu_cfg_space(vgpu), gvt->firmware.cfg_space,
|
|
info->cfg_space_size);
|
|
|
|
if (!primary) {
|
|
vgpu_cfg_space(vgpu)[PCI_CLASS_DEVICE] =
|
|
INTEL_GVT_PCI_CLASS_VGA_OTHER;
|
|
vgpu_cfg_space(vgpu)[PCI_CLASS_PROG] =
|
|
INTEL_GVT_PCI_CLASS_VGA_OTHER;
|
|
}
|
|
|
|
/* Show guest that there isn't any stolen memory.*/
|
|
gmch_ctl = (u16 *)(vgpu_cfg_space(vgpu) + INTEL_GVT_PCI_GMCH_CONTROL);
|
|
*gmch_ctl &= ~(BDW_GMCH_GMS_MASK << BDW_GMCH_GMS_SHIFT);
|
|
|
|
intel_vgpu_write_pci_bar(vgpu, PCI_BASE_ADDRESS_2,
|
|
gvt_aperture_pa_base(gvt), true);
|
|
|
|
vgpu_cfg_space(vgpu)[PCI_COMMAND] &= ~(PCI_COMMAND_IO
|
|
| PCI_COMMAND_MEMORY
|
|
| PCI_COMMAND_MASTER);
|
|
/*
|
|
* Clear the bar upper 32bit and let guest to assign the new value
|
|
*/
|
|
memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_1, 0, 4);
|
|
memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_3, 0, 4);
|
|
memset(vgpu_cfg_space(vgpu) + PCI_BASE_ADDRESS_4, 0, 8);
|
|
memset(vgpu_cfg_space(vgpu) + INTEL_GVT_PCI_OPREGION, 0, 4);
|
|
|
|
vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_GTTMMIO].size =
|
|
pci_resource_len(gvt->dev_priv->drm.pdev, 0);
|
|
vgpu->cfg_space.bar[INTEL_GVT_PCI_BAR_APERTURE].size =
|
|
pci_resource_len(gvt->dev_priv->drm.pdev, 2);
|
|
|
|
memset(vgpu_cfg_space(vgpu) + PCI_ROM_ADDRESS, 0, 4);
|
|
}
|
|
|
|
/**
|
|
* intel_vgpu_reset_cfg_space - reset vGPU configuration space
|
|
*
|
|
* @vgpu: a vGPU
|
|
*
|
|
*/
|
|
void intel_vgpu_reset_cfg_space(struct intel_vgpu *vgpu)
|
|
{
|
|
u8 cmd = vgpu_cfg_space(vgpu)[PCI_COMMAND];
|
|
bool primary = vgpu_cfg_space(vgpu)[PCI_CLASS_DEVICE] !=
|
|
INTEL_GVT_PCI_CLASS_VGA_OTHER;
|
|
|
|
if (cmd & PCI_COMMAND_MEMORY) {
|
|
trap_gttmmio(vgpu, false);
|
|
map_aperture(vgpu, false);
|
|
}
|
|
|
|
/**
|
|
* Currently we only do such reset when vGPU is not
|
|
* owned by any VM, so we simply restore entire cfg
|
|
* space to default value.
|
|
*/
|
|
intel_vgpu_init_cfg_space(vgpu, primary);
|
|
}
|