Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Sender: quic_akhilpo=quicinc.com@mg.codeaurora.org
DMARC-Filter: OpenDMARC Filter v1.4.1 smtp.codeaurora.org 81583C4361A
From:   Akhil P Oommen <quic_akhilpo@quicinc.com>
To:     freedreno <freedreno@lists.freedesktop.org>,
        dri-devel@lists.freedesktop.org, linux-arm-msm@vger.kernel.org,
        Rob Clark <robdclark@gmail.com>,
        Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
        Bjorn Andersson <bjorn.andersson@linaro.org>
Cc:     Abhinav Kumar <quic_abhinavk@quicinc.com>,
        Daniel Vetter <daniel@ffwll.ch>,
        David Airlie <airlied@linux.ie>, Sean Paul <sean@poorly.run>,
        linux-kernel@vger.kernel.org
Subject: [PATCH v1 05/10] drm/msm: Do recovery on hw_init failure
Date:   Wed,  2 Mar 2022 22:57:31 +0530
Message-Id: <20220302225551.v1.5.Ib02d5c2b453e8d4ef3f20e48fef7d9e0be09178e@changeid>
In-Reply-To: <1646242056-2456-1-git-send-email-quic_akhilpo@quicinc.com>
References: <1646242056-2456-1-git-send-email-quic_akhilpo@quicinc.com>
Precedence: bulk

Schedule the recover worker when there is hw init failure in
msm_gpu_submit(). The recover worker will take care of capturing
coredump, gpu recovery and resubmission of pending IBs.

Signed-off-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
---

 drivers/gpu/drm/msm/msm_gpu.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index e8a442a..4d24fa1 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -757,12 +757,15 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_ringbuffer *ring = submit->ring;
 	unsigned long flags;
+	int ret;
 
 	WARN_ON(!mutex_is_locked(&gpu->lock));
 
 	pm_runtime_get_sync(&gpu->pdev->dev);
 
-	msm_gpu_hw_init(gpu);
+	ret = msm_gpu_hw_init(gpu);
+	if (ret)
+		kthread_queue_work(gpu->worker, &gpu->recover_work);
 
 	submit->seqno = ++ring->seqno;
 
-- 
2.7.4