diff mbox series

mmc: meson: move mmc_request_done() call to irq thread

Message ID 20230606065918.460866-1-martin@geanix.com
State New
Headers show
Series mmc: meson: move mmc_request_done() call to irq thread | expand

Commit Message

Martin Hundebøll June 6, 2023, 6:59 a.m. UTC 
From: Martin Hundebøll <mhu@rtx.dk>

The call to mmc_request_done() can schedule, so it cannot be called from
irq context. Wake the irq thread if it needs to be called, and call it
from there instead.

Fixes the following kernel bug, which appears when running an RT patched
kernel on the AmLogic Meson AXG A113X SoC:
[   11.111407] BUG: scheduling while atomic: kworker/0:1H/75/0x00010001
[   11.111438] Modules linked in:
[   11.111451] CPU: 0 PID: 75 Comm: kworker/0:1H Not tainted 6.4.0-rc3-rt2-rtx-00081-gfd07f41ed6b4-dirty #1
[   11.111461] Hardware name: RTX AXG A113X Linux Platform Board (DT)
[   11.111469] Workqueue: kblockd blk_mq_run_work_fn
[   11.111492] Call trace:
[   11.111497]  dump_backtrace+0xac/0xe8
[   11.111510]  show_stack+0x18/0x28
[   11.111518]  dump_stack_lvl+0x48/0x60
[   11.111530]  dump_stack+0x18/0x24
[   11.111537]  __schedule_bug+0x4c/0x68
[   11.111548]  __schedule+0x80/0x574
[   11.111558]  schedule_loop+0x2c/0x50
[   11.111567]  schedule_rtlock+0x14/0x20
[   11.111576]  rtlock_slowlock_locked+0x468/0x730
[   11.111587]  rt_spin_lock+0x40/0x64
[   11.111596]  __wake_up_common_lock+0x5c/0xc4
[   11.111610]  __wake_up+0x18/0x24
[   11.111620]  mmc_blk_mq_req_done+0x68/0x138
[   11.111633]  mmc_request_done+0x104/0x118
[   11.111644]  meson_mmc_request_done+0x38/0x48
[   11.111654]  meson_mmc_irq+0x128/0x1f0
[   11.111663]  __handle_irq_event_percpu+0x70/0x114
[   11.111674]  handle_irq_event_percpu+0x18/0x4c
[   11.111683]  handle_irq_event+0x80/0xb8
[   11.111691]  handle_fasteoi_irq+0xa4/0x120
[   11.111704]  handle_irq_desc+0x20/0x38
[   11.111712]  generic_handle_domain_irq+0x1c/0x28
[   11.111721]  gic_handle_irq+0x8c/0xa8
[   11.111735]  call_on_irq_stack+0x24/0x4c
[   11.111746]  do_interrupt_handler+0x88/0x94
[   11.111757]  el1_interrupt+0x34/0x64
[   11.111769]  el1h_64_irq_handler+0x18/0x24
[   11.111779]  el1h_64_irq+0x64/0x68
[   11.111786]  __add_wait_queue+0x0/0x4c
[   11.111795]  mmc_blk_rw_wait+0x84/0x118
[   11.111804]  mmc_blk_mq_issue_rq+0x5c4/0x654
[   11.111814]  mmc_mq_queue_rq+0x194/0x214
[   11.111822]  blk_mq_dispatch_rq_list+0x3ac/0x528
[   11.111834]  __blk_mq_sched_dispatch_requests+0x340/0x4d0
[   11.111847]  blk_mq_sched_dispatch_requests+0x38/0x70
[   11.111858]  blk_mq_run_work_fn+0x3c/0x70
[   11.111865]  process_one_work+0x17c/0x1f0
[   11.111876]  worker_thread+0x1d4/0x26c
[   11.111885]  kthread+0xe4/0xf4
[   11.111894]  ret_from_fork+0x10/0x20

Fixes: 51c5d8447bd71b ("MMC: meson: initial support for GX platforms")
Cc: stable@vger.kernel.org
Signed-off-by: Martin Hundebøll <martin@geanix.com>
---
 drivers/mmc/host/meson-gx-mmc.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

Comments

Martin Blumenstingl June 6, 2023, 7:22 p.m. UTC | #1 
Hi Martin,

On Tue, Jun 6, 2023 at 8:59 AM Martin Hundebøll <martin@geanix.com> wrote:
[...]
> @@ -1040,6 +1034,13 @@ static irqreturn_t meson_mmc_irq_thread(int irq, void *dev_id)
>         if (WARN_ON(!cmd))
>                 return IRQ_NONE;
>
> +       if (!meson_mmc_bounce_buf_read(cmd->data) &&
> +           !meson_mmc_get_next_command(cmd)) {
> +               meson_mmc_request_done(host->mmc, cmd->mrq);
> +
> +               return IRQ_HANDLED;
> +       }
> +
I'm wondering if this is changing the behavior when cmd->error is non-zero.

Let's take the "status & IRQ_CRC_ERR" case from meson_mmc_irq() - previously:
- cmd->error is set
- regmap_update_bits(host->regmap, SD_EMMC_START, START_DESC_BUSY, 0);
- IRQ_WAKE_THREAD
- then inside meson_mmc_irq_thread():
-- meson_mmc_wait_desc_stop(host);
-- meson_mmc_request_done(host->mmc, cmd->mrq);

With this patch:
- cmd->error is set
- regmap_update_bits(host->regmap, SD_EMMC_START, START_DESC_BUSY, 0);
- IRQ_WAKE_THREAD
- then inside meson_mmc_irq_thread():
-- !meson_mmc_bounce_buf_read(cmd->data) &&
!meson_mmc_get_next_command(cmd) can be true, which means we only
call:
--- meson_mmc_request_done(host->mmc, cmd->mrq);
--- (note the missing meson_mmc_wait_desc_stop() call before
meson_mmc_request_done())

What do you think?
Maybe we just need to remove the code from meson_mmc_irq() and just
omit any modifications to meson_mmc_irq_thread() (as it seems all
checks that you're adding are already present in that function below
the cmd->error case).


Best regards,
Martin
diff mbox series

Patch

diff --git a/drivers/mmc/host/meson-gx-mmc.c b/drivers/mmc/host/meson-gx-mmc.c
index b8514d9d5e736..77b2c23084566 100644
--- a/drivers/mmc/host/meson-gx-mmc.c
+++ b/drivers/mmc/host/meson-gx-mmc.c
@@ -991,11 +991,8 @@  static irqreturn_t meson_mmc_irq(int irq, void *dev_id)
 
 		if (data && !cmd->error)
 			data->bytes_xfered = data->blksz * data->blocks;
-		if (meson_mmc_bounce_buf_read(data) ||
-		    meson_mmc_get_next_command(cmd))
-			ret = IRQ_WAKE_THREAD;
-		else
-			ret = IRQ_HANDLED;
+
+		ret = IRQ_WAKE_THREAD;
 	}
 
 out:
@@ -1007,9 +1004,6 @@  static irqreturn_t meson_mmc_irq(int irq, void *dev_id)
 		writel(start, host->regs + SD_EMMC_START);
 	}
 
-	if (ret == IRQ_HANDLED)
-		meson_mmc_request_done(host->mmc, cmd->mrq);
-
 	return ret;
 }
 
@@ -1040,6 +1034,13 @@  static irqreturn_t meson_mmc_irq_thread(int irq, void *dev_id)
 	if (WARN_ON(!cmd))
 		return IRQ_NONE;
 
+	if (!meson_mmc_bounce_buf_read(cmd->data) &&
+	    !meson_mmc_get_next_command(cmd)) {
+		meson_mmc_request_done(host->mmc, cmd->mrq);
+
+		return IRQ_HANDLED;
+	}
+
 	if (cmd->error) {
 		meson_mmc_wait_desc_stop(host);
 		meson_mmc_request_done(host->mmc, cmd->mrq);