From: Lizhi Hou <lizhi.hou@amd.com>
To: Max Zhen <max.zhen@amd.com>, <ogabbay@kernel.org>,
<quic_jhugo@quicinc.com>, <dri-devel@lists.freedesktop.org>,
<mario.limonciello@amd.com>, <karol.wachowski@linux.intel.com>
Cc: <linux-kernel@vger.kernel.org>, <sonal.santan@amd.com>
Subject: Re: [PATCH V1] accel/amdxdna: Fix deadlock on debug BO command timeout
Date: Tue, 7 Jul 2026 10:13:03 -0700 [thread overview]
Message-ID: <af44ba99-b220-cbba-a147-d9618620ae49@amd.com> (raw)
In-Reply-To: <5cbef92d-485d-49a9-ba51-001c959e0b28@amd.com>
Applied to drm-misc-fixes
On 7/7/26 08:12, Max Zhen wrote:
>
>
> On 7/6/2026 Mon 22:57, Lizhi Hou wrote:
>> Both amdxdna_hwctx_sync_debug_bo() and amdxdna_drm_config_hwctx_ioctl()
>> hold xdna->dev_lock while invoking backend operations. If the hardware
>> hangs, aie2_cmd_wait() blocks waiting for a firmware response. When the
>> DRM scheduler timeout expires, aie2_sched_job_timedout() is invoked to
>> reset the hardware. However, the timeout handler also attempts to
>> acquire
>> dev_lock, resulting in a deadlock.
>>
>> Avoid this by releasing dev_lock before waiting for the firmware
>> response and reacquiring it after the wait completes. This allows the
>> timeout handler to proceed with device recovery when a debug BO command
>> times out.
>>
>> Fixes: 7ea046838021 ("accel/amdxdna: Support firmware debug buffer")
>> Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
> Reviewed-by: Max Zhen <max.zhen@amd.com>
>> ---
>> drivers/accel/amdxdna/aie2_ctx.c | 5 ++++-
>> drivers/accel/amdxdna/amdxdna_ctx.c | 6 ++++++
>> 2 files changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/accel/amdxdna/aie2_ctx.c
>> b/drivers/accel/amdxdna/aie2_ctx.c
>> index 7bf635634e64..30ccb8d5e23d 100644
>> --- a/drivers/accel/amdxdna/aie2_ctx.c
>> +++ b/drivers/accel/amdxdna/aie2_ctx.c
>> @@ -900,13 +900,16 @@ static int aie2_hwctx_cu_config(struct
>> amdxdna_hwctx *hwctx, void *buf, u32 size
>> static void aie2_cmd_wait(struct amdxdna_hwctx *hwctx, u64 seq)
>> {
>> struct dma_fence *out_fence = aie2_cmd_get_out_fence(hwctx, seq);
>> + struct amdxdna_dev *xdna = hwctx->client->xdna;
>> if (!out_fence) {
>> - XDNA_ERR(hwctx->client->xdna, "Failed to get fence");
>> + XDNA_ERR(xdna, "Failed to get fence");
>> return;
>> }
>> + mutex_unlock(&xdna->dev_lock);
>> dma_fence_wait_timeout(out_fence, false, MAX_SCHEDULE_TIMEOUT);
>> + mutex_lock(&xdna->dev_lock);
>> dma_fence_put(out_fence);
>> }
>> diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c
>> b/drivers/accel/amdxdna/amdxdna_ctx.c
>> index 67a2abcf173e..9ae19393e488 100644
>> --- a/drivers/accel/amdxdna/amdxdna_ctx.c
>> +++ b/drivers/accel/amdxdna/amdxdna_ctx.c
>> @@ -310,6 +310,7 @@ int amdxdna_drm_destroy_hwctx_ioctl(struct
>> drm_device *dev, void *data, struct d
>> if (!drm_dev_enter(dev, &idx))
>> return -ENODEV;
>> + mutex_lock(&xdna->client_lock);
>> mutex_lock(&xdna->dev_lock);
>> hwctx = xa_erase(&client->hwctx_xa, args->handle);
>> if (!hwctx) {
>> @@ -328,6 +329,7 @@ int amdxdna_drm_destroy_hwctx_ioctl(struct
>> drm_device *dev, void *data, struct d
>> XDNA_DBG(xdna, "PID %d destroyed HW context %d", client->pid,
>> args->handle);
>> out:
>> mutex_unlock(&xdna->dev_lock);
>> + mutex_unlock(&xdna->client_lock);
>> drm_dev_exit(idx);
>> return ret;
>> }
>> @@ -388,6 +390,7 @@ int amdxdna_drm_config_hwctx_ioctl(struct
>> drm_device *dev, void *data, struct dr
>> goto free_buf;
>> }
>> + mutex_lock(&xdna->client_lock);
>> mutex_lock(&xdna->dev_lock);
>> hwctx = xa_load(&client->hwctx_xa, args->handle);
>> if (!hwctx) {
>> @@ -400,6 +403,7 @@ int amdxdna_drm_config_hwctx_ioctl(struct
>> drm_device *dev, void *data, struct dr
>> unlock:
>> mutex_unlock(&xdna->dev_lock);
>> + mutex_unlock(&xdna->client_lock);
>> amdxdna_pm_suspend_put(xdna);
>> free_buf:
>> kfree(buf);
>> @@ -428,6 +432,7 @@ int amdxdna_hwctx_sync_debug_bo(struct
>> amdxdna_client *client, u32 debug_bo_hdl)
>> }
>> abo = to_xdna_obj(gobj);
>> + mutex_lock(&xdna->client_lock);
>> mutex_lock(&xdna->dev_lock);
>> hwctx = xa_load(&client->hwctx_xa, abo->assigned_hwctx);
>> if (!hwctx) {
>> @@ -439,6 +444,7 @@ int amdxdna_hwctx_sync_debug_bo(struct
>> amdxdna_client *client, u32 debug_bo_hdl)
>> unlock:
>> mutex_unlock(&xdna->dev_lock);
>> + mutex_unlock(&xdna->client_lock);
>> amdxdna_pm_suspend_put(xdna);
>> put_obj:
>> drm_gem_object_put(gobj);
>
prev parent reply other threads:[~2026-07-07 17:13 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-07 5:57 Lizhi Hou
2026-07-07 15:12 ` Max Zhen
2026-07-07 17:13 ` Lizhi Hou [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=af44ba99-b220-cbba-a147-d9618620ae49@amd.com \
--to=lizhi.hou@amd.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=karol.wachowski@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mario.limonciello@amd.com \
--cc=max.zhen@amd.com \
--cc=ogabbay@kernel.org \
--cc=quic_jhugo@quicinc.com \
--cc=sonal.santan@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®