gpu: nvgpu: use correct units for bios wait

nvgpu_flcn_wait_for_halt() takes a timeout value in milliseconds, so
pass it the millisecond value instead of number of iterations to sleep
if one iteration took a specific amount of time, a bug left over from
refactoring.

Change-Id: Ifddea16522c14c52d765f1c1bfaa69d1f7dfd011
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: https://git-master/r/1514396
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 1 insertions, 1 deletions

diff --git a/drivers/gpu/nvgpu/gm206/bios_gm206.c b/drivers/gpu/nvgpu/gm206/bios_gm206.c
index bcc46c83..9407bf3a 100644
--- a/drivers/gpu/nvgpu/gm206/bios_gm206.c
+++ b/drivers/gpu/nvgpu/gm206/bios_gm206.c
@@ -153,7 +153,7 @@ static int gm206_bios_preos(struct gk20a *g)
         nvgpu_flcn_bootstrap(g->pmu.flcn, g->bios.preos.code_entry_point);
 
         if (nvgpu_flcn_wait_for_halt(g->pmu.flcn,
-                PMU_BOOT_TIMEOUT_MAX / PMU_BOOT_TIMEOUT_DEFAULT)) {
+                PMU_BOOT_TIMEOUT_MAX / 1000)) {
                 err = -ETIMEDOUT;
                 goto out;
         }