drm/i915/gvt: keep oa config in shadow ctx

When populating shadow ctx from guest, we should handle oa related
registers in hw ctx, so that they will not be overlapped by guest oa
configs. This patch made it possible to capture oa data from host for
both host and guests.

Signed-off-by: Min He <min.he@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
This commit is contained in:
Min He 2018-03-02 10:00:25 +08:00 коммит произвёл Zhi Wang
Родитель b24881e0b0
Коммит fa3dd623e5
2 изменённых файлов: 54 добавлений и 0 удалений

Просмотреть файл

@ -52,6 +52,54 @@ static void set_context_pdp_root_pointer(
pdp_pair[i].val = pdp[7 - i];
}
/*
* when populating shadow ctx from guest, we should not overrride oa related
* registers, so that they will not be overlapped by guest oa configs. Thus
* made it possible to capture oa data from host for both host and guests.
*/
static void sr_oa_regs(struct intel_vgpu_workload *workload,
u32 *reg_state, bool save)
{
struct drm_i915_private *dev_priv = workload->vgpu->gvt->dev_priv;
u32 ctx_oactxctrl = dev_priv->perf.oa.ctx_oactxctrl_offset;
u32 ctx_flexeu0 = dev_priv->perf.oa.ctx_flexeu0_offset;
int i = 0;
u32 flex_mmio[] = {
i915_mmio_reg_offset(EU_PERF_CNTL0),
i915_mmio_reg_offset(EU_PERF_CNTL1),
i915_mmio_reg_offset(EU_PERF_CNTL2),
i915_mmio_reg_offset(EU_PERF_CNTL3),
i915_mmio_reg_offset(EU_PERF_CNTL4),
i915_mmio_reg_offset(EU_PERF_CNTL5),
i915_mmio_reg_offset(EU_PERF_CNTL6),
};
if (!workload || !reg_state || workload->ring_id != RCS)
return;
if (save) {
workload->oactxctrl = reg_state[ctx_oactxctrl + 1];
for (i = 0; i < ARRAY_SIZE(workload->flex_mmio); i++) {
u32 state_offset = ctx_flexeu0 + i * 2;
workload->flex_mmio[i] = reg_state[state_offset + 1];
}
} else {
reg_state[ctx_oactxctrl] =
i915_mmio_reg_offset(GEN8_OACTXCONTROL);
reg_state[ctx_oactxctrl + 1] = workload->oactxctrl;
for (i = 0; i < ARRAY_SIZE(workload->flex_mmio); i++) {
u32 state_offset = ctx_flexeu0 + i * 2;
u32 mmio = flex_mmio[i];
reg_state[state_offset] = mmio;
reg_state[state_offset + 1] = workload->flex_mmio[i];
}
}
}
static int populate_shadow_context(struct intel_vgpu_workload *workload)
{
struct intel_vgpu *vgpu = workload->vgpu;
@ -98,6 +146,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
page = i915_gem_object_get_page(ctx_obj, LRC_STATE_PN);
shadow_ring_context = kmap(page);
sr_oa_regs(workload, (u32 *)shadow_ring_context, true);
#define COPY_REG(name) \
intel_gvt_hypervisor_read_gpa(vgpu, workload->ring_context_gpa \
+ RING_CTX_OFF(name.val), &shadow_ring_context->name.val, 4)
@ -122,6 +171,7 @@ static int populate_shadow_context(struct intel_vgpu_workload *workload)
sizeof(*shadow_ring_context),
I915_GTT_PAGE_SIZE - sizeof(*shadow_ring_context));
sr_oa_regs(workload, (u32 *)shadow_ring_context, false);
kunmap(page);
return 0;
}

Просмотреть файл

@ -110,6 +110,10 @@ struct intel_vgpu_workload {
/* shadow batch buffer */
struct list_head shadow_bb;
struct intel_shadow_wa_ctx wa_ctx;
/* oa registers */
u32 oactxctrl;
u32 flex_mmio[7];
};
struct intel_vgpu_shadow_bb {