Skip to content

ggml : add view_src and view_offs to ggml_tensor for views #2874

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Aug 29, 2023
Merged

ggml : add view_src and view_offs to ggml_tensor for views #2874

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
1f5a0f0
ggml : add view_src and view_offs
slaren Aug 29, 2023
28c6e32
update ggml-alloc to use view_src
slaren Aug 29, 2023
5460aec
update ggml_diag_mask to work correctly with automatic inplace
slaren Aug 29, 2023
b5cd70d
exclude other ops that set an inplace flag from automatic inplace
slaren Aug 29, 2023
9fca82b
formatting
slaren Aug 29, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 6 additions & 47 deletions ggml-alloc.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -321,8 +321,7 @@ bool ggml_allocr_is_measure(struct ggml_allocr * alloc) {
//////////// compute graph allocator

static bool ggml_is_view(struct ggml_tensor * t) {
return t->op == GGML_OP_RESHAPE || t->op == GGML_OP_VIEW || t->op == GGML_OP_TRANSPOSE ||
t->op == GGML_OP_PERMUTE || t->op == GGML_OP_CPY;
return t->view_src != NULL;
}

static bool ggml_are_same_layout(const struct ggml_tensor * a, const struct ggml_tensor * b) {
Expand All @@ -340,36 +339,13 @@ static bool ggml_are_same_layout(const struct ggml_tensor * a, const struct ggml
return true;
}

static struct ggml_tensor * get_view_parent(struct ggml_tensor * t) {
switch (t->op) {
case GGML_OP_PERMUTE:
case GGML_OP_RESHAPE:
case GGML_OP_TRANSPOSE:
case GGML_OP_VIEW:
return t->src[0];
case GGML_OP_CPY:
return t->src[1];
default:
return NULL;
}
}

static struct ggml_tensor * get_view_source(struct ggml_tensor * t) {
struct ggml_tensor * parent = t;
do {
parent = get_view_parent(parent);
} while (ggml_is_view(parent));
return parent;
}

static bool ggml_op_can_inplace(enum ggml_op op) {
switch (op) {
case GGML_OP_SCALE:
case GGML_OP_DIAG_MASK_ZERO:
case GGML_OP_DIAG_MASK_INF:
case GGML_OP_ADD:
case GGML_OP_ADD1:
case GGML_OP_ACC:
case GGML_OP_SUB:
case GGML_OP_MUL:
case GGML_OP_DIV:
Expand All @@ -379,7 +355,6 @@ static bool ggml_op_can_inplace(enum ggml_op op) {
case GGML_OP_UNARY:
case GGML_OP_ROPE:
case GGML_OP_RMS_NORM:
case GGML_OP_SET:
case GGML_OP_SOFT_MAX:
case GGML_OP_CONT:
return true;
Expand All @@ -393,24 +368,8 @@ static void allocate_node(struct ggml_allocr * alloc, struct ggml_tensor * node)
struct hash_node * ht = alloc->hash_table;
if (node->data == NULL) {
if (ggml_is_view(node)) {
size_t offset;
switch(node->op) {
case GGML_OP_VIEW:
memcpy(&offset, node->op_params, sizeof(size_t));
node->data = (char *) node->src[0]->data + offset;
break;
case GGML_OP_PERMUTE:
case GGML_OP_RESHAPE:
case GGML_OP_TRANSPOSE:
node->data = node->src[0]->data;
break;
case GGML_OP_CPY:
node->data = node->src[1]->data;
break;
default:
GGML_ASSERT(!"unknown view op");
break;
}
assert(node->view_src->data != NULL);
node->data = (char *)node->view_src->data + node->view_offs;
} else {
// see if we can reuse a parent's buffer (inplace)
if (ggml_op_can_inplace(node->op)) {
Expand All @@ -430,7 +389,7 @@ static void allocate_node(struct ggml_allocr * alloc, struct ggml_tensor * node)
struct hash_node * p_hn = hash_get(ht, parent);
if (parent->data != NULL && p_hn->n_children == 1 && p_hn->n_views == 0 && ggml_are_same_layout(node, parent)) {
if (ggml_is_view(parent)) {
struct ggml_tensor * view_src = get_view_source(parent);
struct ggml_tensor * view_src = parent->view_src;
struct hash_node * view_src_hn = hash_get(ht, view_src);
if (view_src_hn->n_views == 1 && view_src_hn->n_children == 0 && view_src->data == parent->data) {
// TODO: the offset of the view parent must be kept to ensure that the op doesn't overwrite
Expand Down Expand Up @@ -472,7 +431,7 @@ static size_t ggml_allocator_alloc_graph_tensors_n(
struct ggml_tensor * node = gf->nodes[i];

if (ggml_is_view(node)) {
struct ggml_tensor * view_src = get_view_source(node);
struct ggml_tensor * view_src = node->view_src;
hash_get(ht, view_src)->n_views += 1;
}

Expand Down Expand Up @@ -557,7 +516,7 @@ static size_t ggml_allocator_alloc_graph_tensors_n(

if (p_hn->n_children == 0 && p_hn->n_views == 0) {
if (ggml_is_view(parent)) {
struct ggml_tensor * view_src = get_view_source(parent);
struct ggml_tensor * view_src = parent->view_src;
struct hash_node * view_src_hn = hash_get(ht, view_src);
view_src_hn->n_views -= 1;
AT_PRINTF("view_src %s: %d children, %d views\n", view_src->name, view_src_hn->n_children, view_src_hn->n_views);
Expand Down
Loading