On 6/14/19 8:48 AM, Thierry Reding wrote:
On Fri, Jun 14, 2019 at 05:31:39AM -0700, Bitan Biswas wrote:
Early resume Tegra bpmp to fix Xavier clock rate error as follows:
[ 159.017101] of_serial 3110000.serial: calling platform_pm_resume+0x0/0x58 @
317, parent: cbb
[ 159.025501] of_serial 3110000.serial: platform_pm_resume+0x0/0x58 returned 0
after 14 usecs
[ 159.033694] tegra-i2c 31c0000.i2c: calling platform_pm_resume+0x0/0x58 @
317, parent: cbb
[ 159.042254] tegra-i2c 31c0000.i2c: failed changing clock rate: -22
[ 159.048216] PM: dpm_run_callback(): platform_pm_resume+0x0/0x58 returns -22
[ 159.055553] tegra-i2c 31c0000.i2c: platform_pm_resume+0x0/0x58 returned -22
after 13072 usecs
[ 159.063875] PM: Device 31c0000.i2c failed to resume: error -22
Signed-off-by: Bitan Biswas <bbis...@nvidia.com>
---
drivers/firmware/tegra/bpmp.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/firmware/tegra/bpmp.c b/drivers/firmware/tegra/bpmp.c
index dd775e8..de09036 100644
--- a/drivers/firmware/tegra/bpmp.c
+++ b/drivers/firmware/tegra/bpmp.c
@@ -811,7 +811,9 @@ static int __maybe_unused tegra_bpmp_resume(struct device
*dev)
return 0;
}
-static SIMPLE_DEV_PM_OPS(tegra_bpmp_pm_ops, NULL, tegra_bpmp_resume);
+const struct dev_pm_ops tegra_bpmp_pm_ops = {
+ .resume_early = tegra_bpmp_resume
+};
I've changed this to be static while applying to for-5.3/firmware.
Is this patch already part of some other patch? In that case I can
abandon the patch.
Else, I shall add static keyword.
-regards,
Bitan
