[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20190123222315.1122-7-jglisse@redhat.com>
Date: Wed, 23 Jan 2019 17:23:12 -0500
From: jglisse@...hat.com
To: linux-mm@...ck.org
Cc: Andrew Morton <akpm@...ux-foundation.org>,
linux-kernel@...r.kernel.org,
Jérôme Glisse <jglisse@...hat.com>,
Christian König <christian.koenig@....com>,
Jan Kara <jack@...e.cz>,
Felix Kuehling <Felix.Kuehling@....com>,
Jason Gunthorpe <jgg@...lanox.com>,
Matthew Wilcox <mawilcox@...rosoft.com>,
Ross Zwisler <zwisler@...nel.org>,
Dan Williams <dan.j.williams@...el.com>,
Paolo Bonzini <pbonzini@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>,
Michal Hocko <mhocko@...nel.org>,
Ralph Campbell <rcampbell@...dia.com>,
John Hubbard <jhubbard@...dia.com>, kvm@...r.kernel.org,
dri-devel@...ts.freedesktop.org, linux-rdma@...r.kernel.org,
linux-fsdevel@...r.kernel.org, Arnd Bergmann <arnd@...db.de>
Subject: [PATCH v4 6/9] gpu/drm/radeon: optimize out the case when a range is updated to read only
From: Jérôme Glisse <jglisse@...hat.com>
When range of virtual address is updated read only and corresponding
user ptr object are already read only it is pointless to do anything.
Optimize this case out.
Signed-off-by: Jérôme Glisse <jglisse@...hat.com>
Cc: Christian König <christian.koenig@....com>
Cc: Jan Kara <jack@...e.cz>
Cc: Felix Kuehling <Felix.Kuehling@....com>
Cc: Jason Gunthorpe <jgg@...lanox.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Matthew Wilcox <mawilcox@...rosoft.com>
Cc: Ross Zwisler <zwisler@...nel.org>
Cc: Dan Williams <dan.j.williams@...el.com>
Cc: Paolo Bonzini <pbonzini@...hat.com>
Cc: Radim Krčmář <rkrcmar@...hat.com>
Cc: Michal Hocko <mhocko@...nel.org>
Cc: Ralph Campbell <rcampbell@...dia.com>
Cc: John Hubbard <jhubbard@...dia.com>
Cc: kvm@...r.kernel.org
Cc: dri-devel@...ts.freedesktop.org
Cc: linux-rdma@...r.kernel.org
Cc: linux-fsdevel@...r.kernel.org
Cc: Arnd Bergmann <arnd@...db.de>
---
drivers/gpu/drm/radeon/radeon_mn.c | 13 +++++++++++++
1 file changed, 13 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c
index b3019505065a..f77294f58e63 100644
--- a/drivers/gpu/drm/radeon/radeon_mn.c
+++ b/drivers/gpu/drm/radeon/radeon_mn.c
@@ -124,6 +124,7 @@ static int radeon_mn_invalidate_range_start(struct mmu_notifier *mn,
struct radeon_mn *rmn = container_of(mn, struct radeon_mn, mn);
struct ttm_operation_ctx ctx = { false, false };
struct interval_tree_node *it;
+ bool update_to_read_only;
unsigned long end;
int ret = 0;
@@ -138,6 +139,8 @@ static int radeon_mn_invalidate_range_start(struct mmu_notifier *mn,
else if (!mutex_trylock(&rmn->lock))
return -EAGAIN;
+ update_to_read_only = mmu_notifier_range_update_to_read_only(range);
+
it = interval_tree_iter_first(&rmn->objects, range->start, end);
while (it) {
struct radeon_mn_node *node;
@@ -153,10 +156,20 @@ static int radeon_mn_invalidate_range_start(struct mmu_notifier *mn,
it = interval_tree_iter_next(it, range->start, end);
list_for_each_entry(bo, &node->bos, mn_list) {
+ bool read_only;
if (!bo->tbo.ttm || bo->tbo.ttm->state != tt_bound)
continue;
+ /*
+ * If it is already read only and we are updating to
+ * read only then we do not need to change anything.
+ * So save time and skip this one.
+ */
+ read_only = radeon_ttm_tt_is_readonly(bo->tbo.ttm);
+ if (update_to_read_only && read_only)
+ continue;
+
r = radeon_bo_reserve(bo, true);
if (r) {
DRM_ERROR("(%ld) failed to reserve user bo\n", r);
--
2.17.2
Powered by blists - more mailing lists