From: Ashwin Bharambe <ashwin@fb.com>
Date: Thu, 6 Dec 2018 21:44:33 +0000 (-0800)
Subject: Kill GPU memory logs in normal runs (#14838)
X-Git-Tag: accepted/tizen/6.5/unified/20211028.231830~2414
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=5f0bff96391621fcdad638a71e1ab6e42f553ef5;p=platform%2Fupstream%2Fpytorch.git

Kill GPU memory logs in normal runs (#14838)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/14838

The GPU memory tracking logs are incredibly annoying and merely serve
to pollute output. I `VLOG(1)`ed them. Hopefully, this is non-controversial.

Reviewed By: kuttas

Differential Revision: D13343290

fbshipit-source-id: b3cae99346c97b66e97ea660061e15dc5c99b9fc
---

diff --git a/caffe2/core/context_gpu.cu b/caffe2/core/context_gpu.cu
index 9893828..287e17e 100644
--- a/caffe2/core/context_gpu.cu
+++ b/caffe2/core/context_gpu.cu
@@ -384,14 +384,14 @@ void TrackMemoryAlloc(size_t nbytes) {
       long max_t = g_max_by_gpu_map[gpu];
       if (max_t > 0) {
         if (max_t != t) {
-          LOG(INFO) << "GPU " << gpu << ": " << t / 1024 / 1024 << " MB"
-                    << " (max: " << max_t / 1024 / 1024 << " MB)";
+          VLOG(1) << "GPU " << gpu << ": " << t / 1024 / 1024 << " MB"
+                  << " (max: " << max_t / 1024 / 1024 << " MB)";
         } else {
-          LOG(INFO) << "GPU " << gpu << ": " << t / 1024 / 1024 << " MB";
+          VLOG(1) << "GPU " << gpu << ": " << t / 1024 / 1024 << " MB";
         }
       }
     }
-    LOG(INFO) << "Total: " << g_total_mem / 1024 / 1024 << " MB";
+    VLOG(1) << "Total: " << g_total_mem / 1024 / 1024 << " MB";
     g_last_rep = g_total_mem;
   }
 }