manupa-arm commented on a change in pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#discussion_r619359010
##########
File path: src/runtime/crt/memory/stack_allocator.c
##########
@@ -16,17 +16,22 @@
* specific language governing permissions and limitations
* under the License.
*/
-
// LINT_C_FILE
-
#include <tvm/runtime/crt/stack_allocator.h>
+#ifdef TVM_CRT_DEBUG
+#include <tvm/runtime/crt/logging.h>
+#endif
void* StackMemoryManager_Allocate(tvm_workspace_t* tvm_runtime_workspace,
int32_t nbytes) {
uint32_t offset_bytes = (~nbytes + 1) & (TVM_RUNTIME_ALLOC_ALIGNMENT_BYTES -
1);
uint8_t* current_alloc = tvm_runtime_workspace->next_alloc;
uint8_t* next_alloc = tvm_runtime_workspace->next_alloc + nbytes +
offset_bytes;
uint8_t* workspace_end = tvm_runtime_workspace->workspace +
tvm_runtime_workspace->workspace_size;
-
+#ifdef TVM_CRT_DEBUG
Review comment:
I disagree -- user does not write the code that uses this -- its
generated by tvm -- if tvm generates code that does not follow the pattern it
should be fixed in the codegenerator -- the tree-scoped tir.allocates always
produce the code in this manner. Therefore, I think the 4-byte overhead for
each allocate should only be enabled to debug the codegenerator -- which I
don't think a user would be doing -- therefore default should be off, IMO.
This is aimed to be used in a deployment scenario.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
