igt/prime_vgem: Split out the fine-grain coherency check

We don't expect every machine to be able to pass the WC/GTT coherency
check, see

kernel commit 3b5724d702ef24ee41ca008a1fab1cf94f3d31b5
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Thu Aug 18 17:16:49 2016 +0100

    drm/i915: Wait for writes through the GTT to land before reading back

    If we quickly switch from writing through the GTT to a read of the
    physical page directly with the CPU (e.g. performing relocations through
    the GTT and then running the command parser), we can observe that the
    writes are not visible to the CPU. It is not a coherency problem, as
    extensive investigations with clflush have demonstrated, but a mere
    timing issue - we have to wait for the GTT to complete it's write before
    we start our read from the CPU.

    The issue can be illustrated in userspace with:

            gtt = gem_mmap__gtt(fd, handle, 0, OBJECT_SIZE, PROT_READ | PROT_WRITE);
            cpu = gem_mmap__cpu(fd, handle, 0, OBJECT_SIZE, PROT_READ | PROT_WRITE);
            gem_set_domain(fd, handle, I915_GEM_DOMAIN_GTT, I915_GEM_DOMAIN_GTT);

            for (i = 0; i < OBJECT_SIZE / 64; i++) {
                    int x = 16*i + (i%16);
                    gtt[x] = i;
                    clflush(&cpu[x], sizeof(cpu[x]));
                    assert(cpu[x] == i);
            }

    Experimenting with that shows that this behaviour is indeed limited to
    recent Atom-class hardware.

so split out the interleave coherency check from the basic
interopability check.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102577
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>

1 files changed, 36 insertions, 5 deletions

diff --git a/tests/prime_vgem.c b/tests/prime_vgem.c
index 95557ef9..0ffaee90 100644
--- a/tests/prime_vgem.c
+++ b/tests/prime_vgem.c
@@ -203,7 +203,7 @@ static void test_gtt(int vgem, int i915)
 {
 	struct vgem_bo scratch;
 	uint32_t handle;
-	uint32_t *ptr, *gtt;
+	uint32_t *ptr;
 	int dmabuf, i;
 
 	scratch.width = 1024;
@@ -232,18 +232,46 @@ static void test_gtt(int vgem, int i915)
 		igt_assert_eq(ptr[1024*i], ~i);
 	munmap(ptr, scratch.size);
 
+	gem_close(i915, handle);
+	gem_close(vgem, scratch.handle);
+}
+
+static void test_gtt_interleaved(int vgem, int i915)
+{
+	struct vgem_bo scratch;
+	uint32_t handle;
+	uint32_t *ptr, *gtt;
+	int dmabuf, i;
+
+	scratch.width = 1024;
+	scratch.height = 1024;
+	scratch.bpp = 32;
+	vgem_create(vgem, &scratch);
+
+	dmabuf = prime_handle_to_fd(vgem, scratch.handle);
+	handle = prime_fd_to_handle(i915, dmabuf);
+	close(dmabuf);
+
+	/* This assumes that GTT is perfectedly coherent. On certain machines,
+	 * it is possible for a direct acces to bypass the GTT indirection.
+	 *
+	 * This test may fail. It tells us how far userspace can trust
+	 * concurrent dmabuf/i915 access. In the future, we may have a kernel
+	 * param to indicate whether or not this interleaving is possible.
+	 * However, the mmaps may be passed around to third parties that do
+	 * not know about the shortcommings...
+	 */
 	ptr = vgem_mmap(vgem, &scratch, PROT_WRITE);
 	gtt = gem_mmap__gtt(i915, handle, scratch.size, PROT_WRITE);
-#if defined(__x86_64__)
 	for (i = 0; i < 1024; i++) {
 		gtt[1024*i] = i;
-		__builtin_ia32_sfence();
+		/* The read from WC should act as a flush for the GTT wcb */
 		igt_assert_eq(ptr[1024*i], i);
+
 		ptr[1024*i] = ~i;
-		__builtin_ia32_sfence();
+		/* The read from GTT should act as a flush for the WC wcb */
 		igt_assert_eq(gtt[1024*i], ~i);
 	}
-#endif
 	munmap(gtt, scratch.size);
 	munmap(ptr, scratch.size);
 
@@ -753,6 +781,9 @@ igt_main
 	igt_subtest("basic-gtt")
 		test_gtt(vgem, i915);
 
+	igt_subtest("coherency-gtt")
+		test_gtt_interleaved(vgem, i915);
+
 	for (e = intel_execution_engines; e->name; e++) {
 		igt_subtest_f("%ssync-%s",
 			      e->exec_id == 0 ? "basic-" : "",