diff --git a/.github/workflows/s390x.yml b/.github/workflows/s390x.yml
index 6bdc2be..bcce4c6 100644
--- a/.github/workflows/s390x.yml
+++ b/.github/workflows/s390x.yml
@@ -3,10 +3,10 @@ name: Ubuntu s390x (GCC 11)
 on:
   push:
     branches:
-      - master
+      - main
   pull_request:
     branches:
-      - master
+      - main
 
 jobs:
   build:
diff --git a/CMakeLists.txt b/CMakeLists.txt
index f1f7a4a..a220f60 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,7 +1,7 @@
 cmake_minimum_required(VERSION 3.14)
 
 
-project(fast_float VERSION 8.2.0 LANGUAGES CXX)
+project(fast_float VERSION 8.2.1 LANGUAGES CXX)
 set(FASTFLOAT_CXX_STANDARD 11 CACHE STRING "the C++ standard to use for fastfloat")
 set(CMAKE_CXX_STANDARD ${FASTFLOAT_CXX_STANDARD})
 option(FASTFLOAT_TEST "Enable tests" OFF)
diff --git a/README.md b/README.md
index 905ef09..c4b3e0d 100644
--- a/README.md
+++ b/README.md
@@ -569,7 +569,7 @@ sufficiently recent version of CMake (3.11 or better at least):
 FetchContent_Declare(
   fast_float
   GIT_REPOSITORY https://github.com/fastfloat/fast_float.git
-  GIT_TAG tags/v8.2.0
+  GIT_TAG tags/v8.2.1
   GIT_SHALLOW TRUE)
 
 FetchContent_MakeAvailable(fast_float)
@@ -585,7 +585,7 @@ You may also use [CPM](https://github.com/cpm-cmake/CPM.cmake), like so:
 CPMAddPackage(
   NAME fast_float
   GITHUB_REPOSITORY "fastfloat/fast_float"
-  GIT_TAG v8.2.0)
+  GIT_TAG v8.2.1)
 ```
 
 ## Using as single header
@@ -597,7 +597,7 @@ if desired as described in the command line help.
 
 You may directly download automatically generated single-header files:
 
-<https://github.com/fastfloat/fast_float/releases/download/v8.2.0/fast_float.h>
+<https://github.com/fastfloat/fast_float/releases/download/v8.2.1/fast_float.h>
 
 ## Benchmarking
 
diff --git a/include/fast_float/ascii_number.h b/include/fast_float/ascii_number.h
index c91aed3..286b7d4 100644
--- a/include/fast_float/ascii_number.h
+++ b/include/fast_float/ascii_number.h
@@ -586,6 +586,9 @@ parse_int_string(UC const *p, UC const *pend, T &value,
         str[j] = static_cast<uint8_t>(p[j]);
       }
       digits = bit_cast<uint32_t>(str);
+#if FASTFLOAT_IS_BIG_ENDIAN
+      digits = byteswap(digits);
+#endif
     }
 #else
     if (false) {
@@ -593,6 +596,9 @@ parse_int_string(UC const *p, UC const *pend, T &value,
 #endif
     else if (len >= 4) {
       std::memcpy(&digits, p, 4);
+#if FASTFLOAT_IS_BIG_ENDIAN
+      digits = byteswap(digits);
+#endif
     } else {
       uint32_t b0 = static_cast<uint8_t>(p[0]);
       uint32_t b1 = (len > 1) ? static_cast<uint8_t>(p[1]) : 0xFFu;
diff --git a/include/fast_float/float_common.h b/include/fast_float/float_common.h
index 2661c9f..6e51143 100644
--- a/include/fast_float/float_common.h
+++ b/include/fast_float/float_common.h
@@ -17,7 +17,7 @@
 
 #define FASTFLOAT_VERSION_MAJOR 8
 #define FASTFLOAT_VERSION_MINOR 2
-#define FASTFLOAT_VERSION_PATCH 0
+#define FASTFLOAT_VERSION_PATCH 1
 
 #define FASTFLOAT_STRINGIZE_IMPL(x) #x
 #define FASTFLOAT_STRINGIZE(x) FASTFLOAT_STRINGIZE_IMPL(x)