[Intel-gfx] [PATCH] tests/gem_exec_big: Re-add gem_sync

Daniel Vetter daniel.vetter at ffwll.ch
Thu Jun 19 10:17:11 CEST 2014


We need this to avoid hitting the slowpath and ending up with a
presumed_offset == -1. Regression reported by PRTS, bisected to

commit eb36fc993d7ae1988c80ba5b767989059c91d0ec
Author:     Chris Wilson <chris at chris-wilson.co.uk>
AuthorDate: Mon Jun 16 10:49:16 2014 +0100
Commit:     Chris Wilson <chris at chris-wilson.co.uk>
CommitDate: Mon Jun 16 10:51:02 2014 +0100

    igt/gem_exec_big: Update to new igt_assert_eq

    Use igt_assert_eq for better test output on failures.

    Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>

v2: igt_warn_on unexpected reloc offsets.

Cc: shuang.he at intel.com
Acked-by: Chris Wilson <chris at chris-wilson.co.uk> (on irc)
Signed-off-by: Daniel Vetter <daniel.vetter at ffwll.ch>
---
 tests/gem_exec_big.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/tests/gem_exec_big.c b/tests/gem_exec_big.c
index 615295e9f117..b82774fb8c7c 100644
--- a/tests/gem_exec_big.c
+++ b/tests/gem_exec_big.c
@@ -84,8 +84,15 @@ static void exec(int fd, uint32_t handle, uint32_t reloc_ofs)
 	i915_execbuffer2_set_context_id(execbuf, 0);
 	execbuf.rsvd2 = 0;
 
+	/* Avoid hitting slowpaths in the reloc processing which might yield a
+	 * presumed_offset of -1. Happens when the batch is still busy from the
+	 * last round. */
+	gem_sync(fd, handle);
+
 	gem_execbuf(fd, &execbuf);
 
+	igt_warn_on(gem_reloc[0].presumed_offset == -1);
+
 	gem_read(fd, handle, reloc_ofs, &tmp, 4);
 	igt_assert_eq(tmp, gem_reloc[0].presumed_offset);
 }
@@ -106,8 +113,11 @@ igt_simple_main
 		handle = gem_create(fd, batch_size);
 		gem_write(fd, handle, 0, batch, sizeof(batch));
 
-		for (reloc_ofs = 4096; reloc_ofs < batch_size; reloc_ofs += 4096)
+		for (reloc_ofs = 4096; reloc_ofs < batch_size; reloc_ofs += 4096) {
+			igt_debug("batch_size %u, reloc_ofs %u\n",
+				  batch_size, reloc_ofs);
 			exec(fd, handle, reloc_ofs);
+		}
 	}
 
 	gem_close(fd, handle);
-- 
1.9.3




More information about the Intel-gfx mailing list