On 11/30/2016 03:05 PM, Jaehoon Chung wrote:
> Before calling clk_round_rate(), put the spin_unlock_irq() in
> sdhci_s3c_consider_clock() function.
this patch is for fixing the below issue.
273.465261] BUG: scheduling while atomic: mmcqd/0/96/0x00000002
[ 273.469782] Modules linked in:
[ 273.472819] Preemption disabled at:[ 273.476121] [< (null)>] (null)
[ 273.479424] CPU: 3 PID: 96 Comm: mmcqd/0 Not tainted 4.9.0-rc5-00106-g4d698fc-dirty #6
[ 273.487316] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[ 273.493419] [<c010f0f4>] (unwind_backtrace) from [<c010ae24>] (show_stack+0x10/0x14)
[ 273.501127] [<c010ae24>] (show_stack) from [<c033cb68>] (dump_stack+0x78/0x8c)
[ 273.508333] [<c033cb68>] (dump_stack) from [<c013ddd4>] (__schedule_bug+0x84/0xd4)
[ 273.515884] [<c013ddd4>] (__schedule_bug) from [<c072e788>] (__schedule+0x398/0x480)
[ 273.523605] [<c072e788>] (__schedule) from [<c072e8bc>] (schedule+0x4c/0xac)
[ 273.530634] [<c072e8bc>] (schedule) from [<c072ed24>] (schedule_preempt_disabled+0x14/0x20)
[ 273.538969] [<c072ed24>] (schedule_preempt_disabled) from [<c07312a4>] (__mutex_lock_slowpath+0x19c/0x3fc)
[ 273.548603] [<c07312a4>] (__mutex_lock_slowpath) from [<c0731510>] (mutex_lock+0xc/0x24)
[ 273.556678] [<c0731510>] (mutex_lock) from [<c03a7614>] (clk_prepare_lock+0x50/0xf8)
[ 273.564403] [<c03a7614>] (clk_prepare_lock) from [<c03a8c34>] (clk_round_rate+0x18/0x60)
[ 273.572475] [<c03a8c34>] (clk_round_rate) from [<c057e274>] (sdhci_s3c_set_clock+0x1e0/0x1e8)
[ 273.580980] [<c057e274>] (sdhci_s3c_set_clock) from [<c057e2a0>] (sdhci_cmu_set_clock+0x24/0x17c)
[ 273.589831] [<c057e2a0>] (sdhci_cmu_set_clock) from [<c057d4d0>] (sdhci_set_ios+0x7c/0x408)
[ 273.598164] [<c057d4d0>] (sdhci_set_ios) from [<c057d378>] (sdhci_runtime_resume_host+0x70/0x14c)
[ 273.607025] [<c057d378>] (sdhci_runtime_resume_host) from [<c044a690>] (__rpm_callback+0x2c/0x60)
[ 273.615875] [<c044a690>] (__rpm_callback) from [<c044a71c>] (rpm_callback+0x58/0x80)
[ 273.623599] [<c044a71c>] (rpm_callback) from [<c044b5cc>] (rpm_resume+0x35c/0x560)
[ 273.631150] [<c044b5cc>] (rpm_resume) from [<c044b81c>] (__pm_runtime_resume+0x4c/0x64)
[ 273.639137] [<c044b81c>] (__pm_runtime_resume) from [<c056718c>] (__mmc_claim_host+0x17c/0x1b8)
[ 273.647821] [<c056718c>] (__mmc_claim_host) from [<c0578094>] (mmc_blk_issue_rq+0x2cc/0x570)
[ 273.656236] [<c0578094>] (mmc_blk_issue_rq) from [<c0578410>] (mmc_queue_thread+0xd8/0x188)
[ 273.664571] [<c0578410>] (mmc_queue_thread) from [<c0138bf8>] (kthread+0xdc/0xf4)
[ 273.672035] [<c0138bf8>] (kthread) from [<c0107878>] (ret_from_fork+0x14/0x3c)
[ 273.680046] ------------[ cut here ]------------
[ 273.684681] WARNING: CPU: 3 PID: 96 at kernel/sched/core.c:3182 _raw_spin_unlock_irq+0x14/0x40
[ 273.692452] DEBUG_LOCKS_WARN_ON(val > preempt_count())
Modules linked in:
[ 273.699285] CPU: 3 PID: 96 Comm: mmcqd/0 Tainted: G W 4.9.0-rc5-00106-g4d698fc-dirty #6
[ 273.708394] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[ 273.714488] [<c010f0f4>] (unwind_backtrace) from [<c010ae24>] (show_stack+0x10/0x14)
[ 273.722202] [<c010ae24>] (show_stack) from [<c033cb68>] (dump_stack+0x78/0x8c)
[ 273.729406] [<c033cb68>] (dump_stack) from [<c011d3ec>] (__warn+0xe8/0x100)
[ 273.736349] [<c011d3ec>] (__warn) from [<c011d43c>] (warn_slowpath_fmt+0x38/0x48)
[ 273.743814] [<c011d43c>] (warn_slowpath_fmt) from [<c0733404>] (_raw_spin_unlock_irq+0x14/0x40)
[ 273.752496] [<c0733404>] (_raw_spin_unlock_irq) from [<c057e2e0>] (sdhci_cmu_set_clock+0x64/0x17c)
[ 273.761433] [<c057e2e0>] (sdhci_cmu_set_clock) from [<c057d4d0>] (sdhci_set_ios+0x7c/0x408)
[ 273.769765] [<c057d4d0>] (sdhci_set_ios) from [<c057d378>] (sdhci_runtime_resume_host+0x70/0x14c)
[ 273.778623] [<c057d378>] (sdhci_runtime_resume_host) from [<c044a690>] (__rpm_callback+0x2c/0x60)
[ 273.787475] [<c044a690>] (__rpm_callback) from [<c044a71c>] (rpm_callback+0x58/0x80)
[ 273.795199] [<c044a71c>] (rpm_callback) from [<c044b5cc>] (rpm_resume+0x35c/0x560)
[ 273.802750] [<c044b5cc>] (rpm_resume) from [<c044b81c>] (__pm_runtime_resume+0x4c/0x64)
[ 273.810737] [<c044b81c>] (__pm_runtime_resume) from [<c056718c>] (__mmc_claim_host+0x17c/0x1b8)
[ 273.819420] [<c056718c>] (__mmc_claim_host) from [<c0578094>] (mmc_blk_issue_rq+0x2cc/0x570)
[ 273.827837] [<c0578094>] (mmc_blk_issue_rq) from [<c0578410>] (mmc_queue_thread+0xd8/0x188)
[ 273.836171] [<c0578410>] (mmc_queue_thread) from [<c0138bf8>] (kthread+0xdc/0xf4)
[ 273.843635] [<c0138bf8>] (kthread) from [<c0107878>] (ret_from_fork+0x14/0x3c)
If my approach is wrong, let me know, plz.
Best Regards,
Jaehoon Chung
>
> Signed-off-by: Jaehoon Chung <jh80.chung@xxxxxxxxxxx>
> ---
> drivers/mmc/host/sdhci-s3c.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
> index 784c5a8..de219ca 100644
> --- a/drivers/mmc/host/sdhci-s3c.c
> +++ b/drivers/mmc/host/sdhci-s3c.c
> @@ -121,7 +121,9 @@ static unsigned int sdhci_s3c_consider_clock(struct sdhci_s3c *ourhost,
> * speed possible with selected clock source and skip the division.
> */
> if (ourhost->no_divider) {
> + spin_unlock_irq(&ourhost->host->lock);
> rate = clk_round_rate(clksrc, wanted);
> + spin_lock_irq(&ourhost->host->lock);
> return wanted - rate;
> }
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-mmc" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
