summaryrefslogtreecommitdiff
path: root/tests/i915
diff options
context:
space:
mode:
authorChris Wilson <chris@chris-wilson.co.uk>2019-03-26 18:54:19 +0000
committerChris Wilson <chris@chris-wilson.co.uk>2019-03-27 07:50:21 +0000
commit7c8f2616fa0fd3ddb16e050c5b7ea9ce707abbe4 (patch)
treef4122da2a5e77c1dc6f73238bfc9b9f3100209d8 /tests/i915
parent55192ddf19444767c039386cae6864d934b35930 (diff)
i915/gem_exec_big: 128MiB not enough slack? Let out the rope!
Even with 128MiB reserved for other use, a single pass of gem_exec_big runs out of memory. Give in and halve our batch size, that has to be enough slack! As to why it keeps on failing, is left as an exercise to the reader -- we have to solve the mm/ mystery one day, as eventually it will be our only remaining source of bugs! Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Diffstat (limited to 'tests/i915')
-rw-r--r--tests/i915/gem_exec_big.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/tests/i915/gem_exec_big.c b/tests/i915/gem_exec_big.c
index 015f59e2..440136ee 100644
--- a/tests/i915/gem_exec_big.c
+++ b/tests/i915/gem_exec_big.c
@@ -260,7 +260,7 @@ static void single(int i915)
uint32_t handle;
void *ptr;
- batch_size = (intel_get_avail_ram_mb() - 128) << 20; /* CI slack */
+ batch_size = (intel_get_avail_ram_mb() / 2) << 20; /* XXX CI slack? */
limit = gem_aperture_size(i915) - (256 << 10); /* low pages reserved */
if (!gem_uses_full_ppgtt(i915))
limit = 3 * limit / 4;