[igt-dev] [PATCH i-g-t] Check caching ioctl before forking

[Vikas Srivastava]

    Several subtests for gem_exec_flush require the SET_CACHING
    ioctl to run.  However, the function that calls it is
    wrapped by an igt_require and thus causes a skip when the
    ioctl fails.  Unfortunately, MTL currently fails this check
    with -EOPNOTSUPP.

    While the ioctl failing and resulting in a skip wouldn't
    normally be a problem, the failing call is made within a
    forked child process, which causes an error.  By checking
    whether or not caching is supported before forking, this
    error can be downgraded to a graceful skip.

    v2: Fix formatting

    References: VLK-34086
    Signed-off-by: vikas srivastava <vikas.srivastava at intel.com>
    Signed-off-by: Jonathan Cavitt <jonathan.cavitt at intel.com>
    CC: Fei Yang <fei.yang at intel.com>

---
 tests/i915/gem_exec_flush.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/tests/i915/gem_exec_flush.c b/tests/i915/gem_exec_flush.c
index 40c58db2..50cabcbe 100644
--- a/tests/i915/gem_exec_flush.c
+++ b/tests/i915/gem_exec_flush.c
@@ -81,6 +81,8 @@ static void run(int fd, unsigned ring, int nchild, int timeout,
 		unsigned flags)
 {
 	const unsigned int gen = intel_gen(intel_get_drm_devid(fd));
+	if (!(flags & WC))
+		 gem_require_caching(fd);
 
 	/* The crux of this testing is whether writes by the GPU are coherent
 	 * from the CPU.
@@ -371,7 +373,8 @@ static void batch(int fd, unsigned ring, int nchild, int timeout,
 		  enum batch_mode mode, unsigned flags)
 {
 	const unsigned int gen = intel_gen(intel_get_drm_devid(fd));
-
+	gem_require_caching(fd);
+	
 	if (mode == BATCH_GTT)
 		gem_require_mappable_ggtt(fd);
 
-- 
2.25.1