From: Michael Dagitses <mikeyd@fb.com>
Date: Thu, 26 Aug 2021 11:42:36 +0000 (-0700)
Subject: use `const auto&` as type for grad alias (#63949)
X-Git-Tag: accepted/tizen/8.0/unified/20231005.095509~699
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=61d88cdd1c5fe7cf91b6ee0a71a250e3a6f61878;p=platform%2Fupstream%2Fpytorch.git

use `const auto&` as type for grad alias (#63949)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/63949

This is an extension of the discussion in
https://github.com/pytorch/pytorch/pull/63040#discussion_r687793027.

Test Plan: Imported from OSS

Reviewed By: albanD

Differential Revision: D30546789

Pulled By: dagitses

fbshipit-source-id: 3046aff4f129d5492d73dfb67717a824e16ffee8
---

diff --git a/tools/autograd/gen_autograd_functions.py b/tools/autograd/gen_autograd_functions.py
index 7d852ad..08136ab 100644
--- a/tools/autograd/gen_autograd_functions.py
+++ b/tools/autograd/gen_autograd_functions.py
@@ -479,7 +479,7 @@ def process_function(info: DifferentiabilityInfo, template: CodeTemplate) -> str
     body: List[str] = []
 
     if uses_single_grad(info):
-        body.append('auto& grad = grads[0];')
+        body.append('const auto& grad = grads[0];')
 
     def emit_derivative(
         derivative: Derivative,