Due this typo we don't save/restore the GFX_MAX_REQ_COUNT register across
suspend/resume, so fix this.
This was introduced in
commit 
ddeea5b0c36f3665446518c609be91f9336ef674
Author: Imre Deak <imre.deak@intel.com>
Date:   Mon May 5 15:19:56 2014 +0300
    drm/i915: vlv: add runtime PM support
I noticed this only by reading the code. To my knowledge it shouldn't
cause any real problems at the moment, since the power well backing this
register remains on across a runtime s/r. This may change once
system-wide s0ix functionality is enabled in the kernel.
v2:
- resend after a missing git add -u :/
Cc: stable@vger.kernel.org
Signed-off-by: Imre Deak <imre.deak@intel.com>
Tested-By: PRC QA PRTS (Patch Regression Test System Contact: shuang.he@intel.com)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
                s->lra_limits[i] = I915_READ(GEN7_LRA_LIMITS_BASE + i * 4);
 
        s->media_max_req_count  = I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
-       s->gfx_max_req_count    = I915_READ(GEN7_MEDIA_MAX_REQ_COUNT);
+       s->gfx_max_req_count    = I915_READ(GEN7_GFX_MAX_REQ_COUNT);
 
        s->render_hwsp          = I915_READ(RENDER_HWS_PGA_GEN7);
        s->ecochk               = I915_READ(GAM_ECOCHK);
                I915_WRITE(GEN7_LRA_LIMITS_BASE + i * 4, s->lra_limits[i]);
 
        I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->media_max_req_count);
-       I915_WRITE(GEN7_MEDIA_MAX_REQ_COUNT, s->gfx_max_req_count);
+       I915_WRITE(GEN7_GFX_MAX_REQ_COUNT, s->gfx_max_req_count);
 
        I915_WRITE(RENDER_HWS_PGA_GEN7, s->render_hwsp);
        I915_WRITE(GAM_ECOCHK,          s->ecochk);