<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yestino - The Signal · Cut ProcessGroupNCCLErrorsTest</title><link>https://yestino.com/entities/cut-processgroupncclerrorstest-7bed61</link><description>Every event involving Cut ProcessGroupNCCLErrorsTest</description><language>en</language><atom:link href="https://yestino.com/entities/cut-processgroupncclerrorstest-7bed61/feed.xml" rel="self" type="application/rss+xml"/><item><title>trunk/2f2f675bc57fb8bde0bf9c21ed41dd8f44de6a45: [c10d] Cut ProcessGroupNCCLErrorsTest runtime by ~13 min (#199826)</title><link>https://yestino.com/events/trunk-2f2f675bc57fb8bde0bf9c21ed41dd8f44de6a45-c10d-cut-proc-8dc73a</link><guid isPermaLink="true">https://yestino.com/events/trunk-2f2f675bc57fb8bde0bf9c21ed41dd8f44de6a45-c10d-cut-proc-8dc73a</guid><pubDate>Thu, 08 Oct 2026 01:55:10 GMT</pubDate><description>ProcessGroupNCCLErrorsTest spends ~830s per CUDA distributed config in three tests: testNCCLErrorsNonBlocking (242s): the watchdog sleeps TORCH_NCCL_WAIT_TIMEO…
Sources: PyTorch Releases</description></item></channel></rss>