gpu: nvgpu: create debugfs node early

Create debugfs node before platform->probe() is called.

Allow chip specific debugfs entries go to correct
directory.

bug 1525327
bug 1581799

Change-Id: I2d91bdc1e72dac6787938eff01218c9f871029cb
Signed-off-by: Kirill Artamonov <kartamonov@nvidia.com>
Reviewed-on: http://git-master/r/796092
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: http://git-master/r/778729
Reviewed-by: Automatic_Commit_Validation_User
GVS: Gerrit_Virtual_Submit

1 files changed, 2 insertions, 2 deletions

diff --git a/drivers/gpu/nvgpu/gk20a/gk20a.c b/drivers/gpu/nvgpu/gk20a/gk20a.c
index fb8b8b14..c606e027 100644
--- a/drivers/gpu/nvgpu/gk20a/gk20a.c
+++ b/drivers/gpu/nvgpu/gk20a/gk20a.c
@@ -1408,6 +1408,8 @@ static int gk20a_probe(struct platform_device *dev)
 
         spin_lock_init(&gk20a->mc_enable_lock);
 
+        gk20a_debug_init(dev);
+
         /* Initialize the platform interface. */
         err = platform->probe(dev);
         if (err) {
@@ -1440,8 +1442,6 @@ static int gk20a_probe(struct platform_device *dev)
                 dev_err(&dev->dev,
                         "failed to allocate secure buffer %d\n", err);
 
-        gk20a_debug_init(dev);
-
         /* Set DMA parameters to allow larger sgt lists */
         dev->dev.dma_parms = &gk20a->dma_parms;
         dma_set_max_seg_size(&dev->dev, UINT_MAX);