[igt-dev] [PATCH i-g-t 1/2] lib/xe: Add support to reset all GT's

[Kamil Konieczny]

Hi Janga,

On 2023-07-25 at 12:46:21 +0530, janga.rahul.kumar at intel.com wrote:
> From: Janga Rahul Kumar <janga.rahul.kumar at intel.com>
> 
> Add library support to check support for GT reset and
> force reset all GT's.
> 
> Cc: Sai Gowtham Ch <sai.gowtham.ch at intel.com>
> Cc: Kunal Joshi <kunal1.joshi at intel.com>
> Signed-off-by: Janga Rahul Kumar <janga.rahul.kumar at intel.com>
> ---
>  lib/igt_gt.c      |  7 +++++++
>  lib/xe/xe_ioctl.c | 52 +++++++++++++++++++++++++++++++++++++++++++++++
>  lib/xe/xe_ioctl.h |  2 ++
>  3 files changed, 61 insertions(+)
> 
> diff --git a/lib/igt_gt.c b/lib/igt_gt.c
> index d4a825e66..2ce464ba6 100644
> --- a/lib/igt_gt.c
> +++ b/lib/igt_gt.c
> @@ -45,6 +45,7 @@
>  #include "intel_chipset.h"
>  #include "igt_dummyload.h"
>  
> +#include "xe/xe_ioctl.h"

Add newline here.

>  /**
>   * SECTION:igt_gt
>   * @short_description: GT support library
> @@ -60,6 +61,9 @@ static int reset_query_once = -1;
>  
>  static bool has_gpu_reset(int fd)
>  {
> +	if (is_xe_device(fd))
> +		has_xe_gt_reset(fd);
> +
>  	if (reset_query_once < 0) {
>  		reset_query_once = gem_gpu_reset_type(fd);
>  
> @@ -395,6 +399,9 @@ void igt_post_hang_ring(int fd, igt_hang_t arg)
>   */
>  void igt_force_gpu_reset(int drm_fd)
>  {
> +	if (is_xe_device(drm_fd))
> +		xe_force_gt_reset_all(drm_fd);
> +

Move above code after igt_debug below.

>  	int dir, wedged;
>  
>  	igt_debug("Triggering GPU reset\n");
> diff --git a/lib/xe/xe_ioctl.c b/lib/xe/xe_ioctl.c
> index 1f9240cd9..488aa218e 100644
> --- a/lib/xe/xe_ioctl.c
> +++ b/lib/xe/xe_ioctl.c
> @@ -41,6 +41,7 @@
>  #include "config.h"
>  #include "drmtest.h"
>  #include "igt_syncobj.h"
> +#include "igt_sysfs.h"
>  #include "ioctl_wrappers.h"
>  #include "xe_ioctl.h"
>  #include "xe_query.h"
> @@ -455,6 +456,57 @@ int64_t xe_wait_ufence_abstime(int fd, uint64_t *addr, uint64_t value,
>  	return ts.tv_sec * 1e9 + ts.tv_nsec;
>  }
>  
> +
> +/**
> + * has_xe_gt_reset:
> + * @fd: open xe drm file descriptor
> + *
> + * Check gt force reset syfs entry is available or not
> + *
> + * Returns: reset sysfs entry available
> + */
> +bool has_xe_gt_reset(int fd)
> +{
> +	char reset_sysfs_path[100];
> +	struct stat st;
> +	int gt;
> +	int reset_sysfs_fd = -1;
> +	int sysfs_fd = -1;

Add newline here.

> +	igt_assert_eq(fstat(fd, &st), 0);
> +
> +	sysfs_fd = igt_sysfs_open(fd);
> +	igt_assert(sysfs_fd != -1);
> +
> +	xe_for_each_gt(fd, gt) {
> +		sprintf(reset_sysfs_path, "/sys/kernel/debug/dri/%d/gt%d/force_reset", minor(st.st_rdev), gt);
> +		reset_sysfs_fd = openat(sysfs_fd, reset_sysfs_path, O_RDONLY);
> +
> +		if(reset_sysfs_fd == -1) {
> +			close(sysfs_fd);
> +			return 0;
> +		}
> +
> +		close(reset_sysfs_fd);
> +       }
> +
> +	close(sysfs_fd);
> +	return 1;
> +}
> +
> +/**
> + * xe_force_gt_reset_all:
> + *
> + * Forces reset of all the GT's.
> + *
> + */
> +void xe_force_gt_reset_all(int xe_fd)
> +{
> +	int gt;

Add newline.

Regards,
Kamil

> +	xe_for_each_gt(xe_fd, gt)
> +		xe_force_gt_reset(xe_fd, gt);
> +}
> +
> +
>  void xe_force_gt_reset(int fd, int gt)
>  {
>  	char reset_string[128];
> diff --git a/lib/xe/xe_ioctl.h b/lib/xe/xe_ioctl.h
> index 320e0f9f6..5a528b345 100644
> --- a/lib/xe/xe_ioctl.h
> +++ b/lib/xe/xe_ioctl.h
> @@ -87,6 +87,8 @@ int64_t xe_wait_ufence_abstime(int fd, uint64_t *addr, uint64_t value,
>  			       struct drm_xe_engine_class_instance *eci,
>  			       int64_t timeout);
>  void xe_force_gt_reset(int fd, int gt);
> +void xe_force_gt_reset_all(int fd);
> +bool has_xe_gt_reset(int fd);
>  void xe_vm_madvise(int fd, uint32_t vm, uint64_t addr, uint64_t size,
>  		   uint32_t property, uint32_t value);
>  
> -- 
> 2.25.1
>