gpu: nvgpu: Hard code regops max batch size

We set the regops limit in common code to a hard coded value and access
it in Linux code. Change the responsibility so that regops limit is
set in Linux code in the GPU characteristics query to a hard coded value
and just use the same hard coded value in the IOCTL limit check.

JIRA NVGPU-259

Change-Id: I2f78a7ea8f1cb68a08633a2dc74b71b3b001e5c9
Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1593682
Reviewed-by: Automatic_Commit_Validation_User
Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
Reviewed-by: Peter Daifuku <pdaifuku@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>

1 files changed, 2 insertions, 0 deletions

diff --git a/drivers/gpu/nvgpu/common/linux/ioctl_ctrl.c b/drivers/gpu/nvgpu/common/linux/ioctl_ctrl.c
index 02dedc9f..83282b87 100644
--- a/drivers/gpu/nvgpu/common/linux/ioctl_ctrl.c
+++ b/drivers/gpu/nvgpu/common/linux/ioctl_ctrl.c
@@ -33,6 +33,7 @@
 #include <nvgpu/linux/vidmem.h>
 
 #include "ioctl_ctrl.h"
+#include "ioctl_dbg.h"
 #ifdef CONFIG_TEGRA_19x_GPU
 #include "common/linux/ioctl_ctrl_t19x.h"
 #endif
@@ -207,6 +208,7 @@ gk20a_ctrl_ioctl_gpu_characteristics(
         pgpu->arch = g->params.gpu_arch;
         pgpu->impl = g->params.gpu_impl;
         pgpu->rev = g->params.gpu_rev;
+        pgpu->reg_ops_limit = NVGPU_IOCTL_DBG_REG_OPS_LIMIT;
 
         pgpu->vbios_version = g->bios.vbios_version;
         pgpu->vbios_oem_version = g->bios.vbios_oem_version;