Address review comments

Signed-off-by: Chuck Ketcham <cketcham@nvidia.com>

Author: cketcham2333

docs/sphinx/examples/qec/realtime_decoding_demo/CMakeLists.txt

@@ -26,8 +26,8 @@ cmake_minimum_required(VERSION 3.23)
 # The lowered-kernel binary links the realtime dispatch archive (relocatable
 # CUDA device code), so the project enables CUDA and needs an architecture for
 # the device-link step. Default to 80 (A100) -- an architecture the shipped
-# SDK's dispatch archive includes; override to match your GPU, e.g.
-# -DCMAKE_CUDA_ARCHITECTURES=90 for Hopper / GB200.
+# SDK's dispatch archive includes; override to match your GPU:
+# -DCMAKE_CUDA_ARCHITECTURES=90 for Hopper, 100 for Blackwell (e.g. GB200).
 if(NOT DEFINED CMAKE_CUDA_ARCHITECTURES)
   set(CMAKE_CUDA_ARCHITECTURES 80)
 endif()

docs/sphinx/examples/qec/realtime_decoding_demo/README.md

@@ -33,8 +33,9 @@ In a CUDA-QX container `CUDAQ_INSTALL_DIR` defaults to `/usr/local/cudaq` (or
 installed. If the realtime libraries live in a separate prefix, add
 `-DCUDAQ_REALTIME_DIR=<realtime prefix>`. The lowered kernel links the realtime
 dispatch archive (relocatable CUDA device code), so the build needs a CUDA
-toolchain; override the device-link architecture with
-`-DCMAKE_CUDA_ARCHITECTURES=90` (default `80`) to match your GPU.
+toolchain; the device-link architecture defaults to `80` (Ampere) — override
+with `-DCMAKE_CUDA_ARCHITECTURES=90` for Hopper or
+`-DCMAKE_CUDA_ARCHITECTURES=100` for Blackwell (e.g. GB200).
 
 ## Run
 

docs/sphinx/examples_rst/qec/examples.rst

@@ -12,7 +12,7 @@ Examples that illustrate how to use CUDA-QX for application development are avai
       Decoding From Stim DEM Text <stim_dem_decoder.rst>
       Decoders <decoders.rst>
       Real-Time Decoding <realtime_decoding.rst>
-      Realtime Decoding from an FPGA or QPU Kernel <realtime_decoding_demo.rst>
+      Realtime Decoding with CUDA-Q Decoding Server <realtime_decoding_demo.rst>
       AI Predecoder with CUDA-Q Realtime <realtime_predecoder_pymatching.rst>
       AI Predecoder with CUDA-Q Realtime (with FPGA Data Injection) <realtime_predecoder_fpga.rst>
       Relay BP Decoding with CUDA-Q Realtime <realtime_relay_bp.rst>

docs/sphinx/examples_rst/qec/realtime_decoding_demo.rst

@@ -1,5 +1,5 @@
-Realtime Decoding: Driving the Decoding Server from an FPGA or a QPU Kernel
-===========================================================================
+Realtime Decoding with CUDA-Q Decoding Server
+=============================================
 
 .. note::
 
@@ -64,8 +64,9 @@ In a CUDA-QX container ``CUDAQ_INSTALL_DIR`` defaults to ``/usr/local/cudaq``.
 If the realtime libraries are in a separate prefix, add
 ``-DCUDAQ_REALTIME_DIR=<realtime prefix>``. The lowered kernel links the
 realtime dispatch archive (relocatable CUDA device code), so the build needs a
-CUDA toolchain; the device-link architecture defaults to ``80`` — override with
-``-DCMAKE_CUDA_ARCHITECTURES=90`` to match a Hopper / GB200 GPU.
+CUDA toolchain; the device-link architecture defaults to ``80`` (Ampere) —
+override with ``-DCMAKE_CUDA_ARCHITECTURES=90`` for Hopper or
+``-DCMAKE_CUDA_ARCHITECTURES=100`` for Blackwell (e.g. GB200).
 
 
 Running