From fb4835bc6516d3f592f6394e1168f3ebf6cdb776 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?M=C3=A1ty=C3=A1s=20Aradi?= Date: Fri, 20 Dec 2024 16:01:01 +0100 Subject: [PATCH] Update based on review comment --- docs/how-to/hip_runtime_api/asynchronous.rst | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/how-to/hip_runtime_api/asynchronous.rst b/docs/how-to/hip_runtime_api/asynchronous.rst index b68e743309..2015d31c39 100644 --- a/docs/how-to/hip_runtime_api/asynchronous.rst +++ b/docs/how-to/hip_runtime_api/asynchronous.rst @@ -115,7 +115,9 @@ or from the GPU concurrently with kernel execution. Applications can query this capability by checking the ``asyncEngineCount`` device property. Devices with an ``asyncEngineCount`` greater than zero support concurrent data transfers. Additionally, if host memory is involved in the copy, it should be page-locked -to ensure optimal performance. +to ensure optimal performance. Page-locking (or pinning) host memory increases +the bandwidth between the host and the device, reducing the overhead associated +with data transfers. For more details, visit :ref:`host_memory` page. Asynchronous memory operations -------------------------------------------------------------------------------