manupa-arm commented on a change in pull request #7785:
URL: https://github.com/apache/tvm/pull/7785#discussion_r619372859
##########
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 agree if illegal code is "generated", it will result in garbage
calculation or exceeding the provided buffer for allocator if we are using this
to handle workspace data.
Even this gives an error (of illegal free) that should mean a bug in the
codegen (a very unlikely one if everything is lowered from TIR). In which case,
the codegen developer needs to fix. Therefore, I do think the developer should
investigate with DEBUG mode to see if this is causing the error and fix the
codegen.
--
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]
