Drop Pascal GPU support. (#14630)

Closes #14629.

Authors:
  - Bradley Dice (https://github.com/bdice)

Approvers:
  - Matthew Roeschke (https://github.com/mroeschke)
  - Yunsong Wang (https://github.com/PointKernel)
  - Vukasin Milovanovic (https://github.com/vuule)
  - Nghia Truong (https://github.com/ttnghia)

URL: #14630

CONTRIBUTING.md

@@ -79,7 +79,7 @@ CUDA/GPU:
 
 * CUDA 11.5+
 * NVIDIA driver 450.80.02+
-* Pascal architecture or better
+* Volta architecture or better (Compute Capability >=7.0)
 
 You can obtain CUDA from
 [https://developer.nvidia.com/cuda-downloads](https://developer.nvidia.com/cuda-downloads).

README.md

@@ -62,7 +62,7 @@ print(tips_df.groupby("size").tip_percentage.mean())
 
 * CUDA 11.2+
 * NVIDIA driver 450.80.02+
-* Pascal architecture or better (Compute Capability >=6.0)
+* Volta architecture or better (Compute Capability >=7.0)
 
 ### Conda
 

cpp/src/io/comp/nvcomp_adapter.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2022-2023, NVIDIA CORPORATION.
+ * Copyright (c) 2022-2024, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -48,11 +48,6 @@
 #define NVCOMP_ZSTD_DECOMP_IS_STABLE(MAJOR, MINOR, PATCH) \
   (MAJOR > 2 or (MAJOR == 2 and MINOR > 3) or (MAJOR == 2 and MINOR == 3 and PATCH >= 2))
 
-// Issue https://github.com/NVIDIA/spark-rapids/issues/6614 impacts nvCOMP 2.4.0 ZSTD decompression
-// on compute 6.x
-#define NVCOMP_ZSTD_IS_DISABLED_ON_PASCAL(MAJOR, MINOR, PATCH) \
-  (MAJOR == 2 and MINOR == 4 and PATCH == 0)
-
 namespace cudf::io::nvcomp {
 
 // Dispatcher for nvcompBatched<format>DecompressGetTempSizeEx
@@ -551,11 +546,6 @@ std::optional<std::string> is_zstd_decomp_disabled(feature_status_parameters con
            "`LIBCUDF_NVCOMP_POLICY` environment variable.";
   }
 
-  if (NVCOMP_ZSTD_IS_DISABLED_ON_PASCAL(
-        params.lib_major_version, params.lib_minor_version, params.lib_patch_version) and
-      params.compute_capability_major == 6) {
-    return "Zstandard decompression is disabled on Pascal GPUs";
-  }
   return std::nullopt;
 }
 

cpp/tests/io/comp/decomp_test.cpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2023, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2024, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -213,8 +213,6 @@ TEST_F(NvcompConfigTest, Decompression)
   EXPECT_TRUE(decomp_disabled(compression_type::ZSTD, {2, 2, 0, true, true, 7}));
   // stable integrations enabled required
   EXPECT_TRUE(decomp_disabled(compression_type::ZSTD, {2, 4, 0, false, false, 7}));
-  // 2.4.0 disabled on Pascal
-  EXPECT_TRUE(decomp_disabled(compression_type::ZSTD, {2, 4, 0, true, true, 6}));
 
   EXPECT_FALSE(decomp_disabled(compression_type::SNAPPY, {2, 4, 0, true, true, 7}));
   EXPECT_FALSE(decomp_disabled(compression_type::SNAPPY, {2, 3, 0, false, true, 7}));

python/cudf/cudf/utils/gpu_utils.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2024, NVIDIA CORPORATION.
 
 
 def validate_setup():
@@ -70,23 +70,24 @@ def validate_setup():
             cudaDeviceAttr.cudaDevAttrComputeCapabilityMajor, 0
         )
 
-        if major_version < 6:
-            # A GPU with NVIDIA Pascal™ architecture or newer is required.
+        if major_version < 7:
+            # A GPU with NVIDIA Volta™ architecture or newer is required.
             # Reference: https://developer.nvidia.com/cuda-gpus
-            # Hardware Generation	Compute Capability
-            #    Ampere	                8.x
-            #    Turing	                7.5
-            #    Volta	                7.0, 7.2
-            #    Pascal	                6.x
+            # Hardware Generation       Compute Capability
+            #    Hopper                 9.x
+            #    Ampere                 8.x
+            #    Turing                 7.5
+            #    Volta                  7.0, 7.2
+            #    Pascal                 6.x
             #    Maxwell                5.x
-            #    Kepler	                3.x
-            #    Fermi	                2.x
+            #    Kepler                 3.x
+            #    Fermi                  2.x
             device_name = deviceGetName(0)
             minor_version = getDeviceAttribute(
                 cudaDeviceAttr.cudaDevAttrComputeCapabilityMinor, 0
             )
             warnings.warn(
-                "A GPU with NVIDIA Pascal™ (Compute Capability 6.0) "
+                "A GPU with NVIDIA Volta™ (Compute Capability 7.0) "
                 "or newer architecture is required.\n"
                 f"Detected GPU 0: {device_name}\n"
                 f"Detected Compute Capability: {major_version}.{minor_version}"

python/custreamz/README.md

@@ -14,8 +14,8 @@ from custreamz import kafka
 
 # Full list of configurations can be found at: https://github.com/edenhill/librdkafka/blob/master/CONFIGURATION.md
 kafka_configs = {
-	"metadata.broker.list": "localhost:9092",
-	"group.id": "custreamz-client",
+    "metadata.broker.list": "localhost:9092",
+    "group.id": "custreamz-client",
 }
 
 # Create a reusable Kafka Consumer client; "datasource"
@@ -50,7 +50,7 @@ Please see the [Demo Docker Repository](https://hub.docker.com/r/rapidsai/rapids
 
 * CUDA 11.0+
 * NVIDIA driver 450.80.02+
-* Pascal architecture or better (Compute Capability >=6.0)
+* Volta architecture or better (Compute Capability >=7.0)
 
 ### Conda