{"id":159,"date":"2026-09-01T20:18:18","date_gmt":"2026-09-01T12:18:18","guid":{"rendered":"https:\/\/www.xiaoxindong.cloud\/?p=159"},"modified":"2026-09-01T23:21:06","modified_gmt":"2026-09-01T15:21:06","slug":"xid%e5%85%b8%e5%9e%8b%e6%8a%a5%e9%94%99","status":"publish","type":"post","link":"https:\/\/www.xiaoxindong.cloud\/?p=159","title":{"rendered":"Xid\u5178\u578b\u62a5\u9519"},"content":{"rendered":"\n<h1 class=\"wp-block-heading\">NVIDIA GPU \u5178\u578b XID \u62a5\u9519<\/h1>\n\n\n\n<p class=\"wp-block-paragraph\">nvidia\u5b98\u65b9xid\u9519\u8bef\u624b\u518c\uff1a<a href=\"https:\/\/docs.nvidia.com\/deploy\/xid-errors\/latest\/analyzing-xid-catalog.html\">https:\/\/docs.nvidia.com\/deploy\/xid-errors\/latest\/analyzing-xid-catalog.html<\/a><\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<p class=\"wp-block-paragraph\">XID \u662f GPU \u786c\u4ef6\u5185\u90e8\u629b\u51fa\u7684\u9519\u8bef\uff0c\u8bb0\u5f55\u5728\u7cfb\u7edf dmesg \u5185\u6838\u65e5\u5fd7\uff0c<code>dmesg | grep -i xid<\/code>\u67e5\u770b\u3002 \u533a\u5206\uff1a<strong>\u5e94\u7528\u7a0b\u5e8f bug\u3001\u9a71\u52a8 bug\u3001\u771f\u5b9e\u786c\u4ef6\u6545\u969c<\/strong>\u3002<\/p>\n<\/blockquote>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>XID<\/th><th>\u540d\u79f0<\/th><th>\u73b0\u8c61 &amp; \u539f\u56e0<\/th><th>\u5904\u7406<\/th><\/tr><\/thead><tbody><tr><td><strong>13<\/strong><\/td><td>Graphics Engine Exception\u56fe\u5f62\u5f15\u64ce\u5f02\u5e38<\/td><td>\u6700\u5e38\u89c1\u3002kernel \u6267\u884c\u5f02\u5e38\uff1a\u6570\u7ec4\u8d8a\u754c\u3001\u9664 0\u3001\u975e\u6cd5\u6307\u4ee4\u3001\u7f16\u8bd1\u67b6\u6784\u4e0d\u5339\u914d\uff1b<strong>\u591a\u6570\u662f\u7a0b\u5e8f\u95ee\u9898\uff0c\u5c11\u6570\u786c\u4ef6 SM \u635f\u574f<\/strong><\/td><td>1\uff09\u66f4\u6362\u7a0b\u5e8f\u590d\u73b0\uff1b\u7528<code>compute\u2011sanitizer<\/code>\u8c03\u8bd5\uff1b2\uff09gpu\u2011burn \u957f\u65f6\u95f4\u70e4\u673a\uff0c\u5982\u679c\u4f9d\u7136\u62a5 13\uff0c\u6000\u7591\u786c\u4ef6 SM \u6545\u969c<\/td><\/tr><tr><td><strong>31<\/strong><\/td><td>MMU Error \u663e\u5b58\u9875\u9519\u8befNVIDIA<\/td><td>MMU \u5185\u5b58\u7ba1\u7406\u5355\u5143\u62a5\u975e\u6cd5\u5730\u5740\u8bbf\u95ee\uff1b\u663e\u5b58\u975e\u6cd5\u8bbf\u95ee\uff0cUVM \u9519\u8bef\uff1b\u53ef\u80fd\u7a0b\u5e8f\u3001\u9a71\u52a8\u3001\u663e\u5b58\u786c\u4ef6<\/td><td>1\uff09sanitizer \u67e5\u5185\u5b58\u8d8a\u754c\uff1b2\uff09\u53cd\u590d\u51fa\u73b0\u8dd1 Field\u2011Diag \u786c\u4ef6\u8bca\u65ad\uff0c\u6000\u7591\u663e\u5b58\u786c\u4ef6\u6545\u969c<\/td><\/tr><tr><td><strong>32<\/strong><\/td><td>PBDMA Error PCIE \u63a8\u9001\u7f13\u51b2\u533a\u9519\u8bef<\/td><td>PCIe \u603b\u7ebf\uff0cGPU\u2011\u9a71\u52a8\u901a\u4fe1\u635f\u574f\uff1bPCIe \u94fe\u8def\u4e0d\u7a33\u5b9a\u3001\u63d2\u69fd\u3001\u4fe1\u53f7\u95ee\u9898\uff0c<strong>\u548c\u7528\u6237\u7a0b\u5e8f\u65e0\u5173<\/strong>NVIDIA<\/td><td>\u68c0\u67e5 PCIe \u63d2\u69fd\u3001\u91cd\u65b0\u63d2\u62d4\u3001\u66f4\u6362\u63d2\u69fd\uff1b\u68c0\u67e5\u670d\u52a1\u5668\u4e3b\u677f\u56fa\u4ef6<\/td><\/tr><tr><td><strong>43<\/strong><\/td><td>Reset Channel Verif Error<\/td><td>GPU \u5185\u6838\u4efb\u52a1\u5d29\u6e83\uff0c\u5e94\u7528\u88ab\u7ec8\u6b62\uff1bGPU \u672c\u8eab\u4e0d\u4e00\u5b9a\u574f\uff1b\u5e38\u89c1\u4e8e\u8d85\u65f6\u3001\u6b7b\u9501\u3001\u975e\u6cd5\u6307\u4ee4\u3001\u8fc7\u70ed\u964d\u9891\u3001NCCL \u6b7b\u9501<\/td><td>\u91cd\u542f\u5e94\u7528\uff1b\u6392\u67e5\u6e29\u5ea6\u3001\u529f\u8017\uff1b\u770b\u662f\u5426\u4f34\u968f\u5176\u4ed6 XID<\/td><\/tr><tr><td><strong>45<\/strong><\/td><td>Preemptive Removal<\/td><td>\u4fdd\u62a4\u6027\u6e05\u7406\uff0c<strong>\u672c\u8eab\u4e0d\u662f\u6839\u6e90\u9519\u8bef<\/strong>\uff0c\u4f34\u968f\u522b\u7684 XID \u4e00\u8d77\u51fa\u73b0\uff1b\u4efb\u52a1\u88ab kill\u3001GPU \u91cd\u7f6e\u90fd\u4f1a\u89e6\u53d1\uff0c\u53ef\u5355\u72ec\u5ffd\u7565NVIDIA<\/td><td>\u53ea\u770b\u524d\u9762\u771f\u6b63\u62a5\u9519\u7684 XID\uff0c45 \u53ea\u662f\u526f\u4ea7\u54c1<\/td><\/tr><tr><td><strong>48<\/strong><\/td><td>Double\u2011Bit ECC DBE \u53cc\u6bd4\u7279\u4e0d\u53ef\u7ea0\u6b63 ECC \u9519\u8bef\ud83d\udd25<\/td><td><strong>\u4e25\u91cd\u786c\u4ef6\u62a5\u9519<\/strong>\uff0c\u663e\u5b58 \/ SRAM \u51fa\u73b0\u4e0d\u53ef\u4fee\u590d\u9519\u8bef\uff1b\u5fc5\u987b GPU \u91cd\u7f6e\u6216\u6574\u673a\u91cd\u542f\uff1b\u53cd\u590d\u51fa\u73b0\u8981 RMA \u6362\u5361Tencent Cl&#8230;<\/td><td>\u7acb\u523b\u8dd1\u786c\u4ef6\u8bca\u65ad\uff1b\u591a\u6b21\u51fa\u73b0\u76f4\u63a5\u5224\u5b9a\u663e\u5b58\u786c\u4ef6\u6545\u969c<\/td><\/tr><tr><td><strong>63<\/strong><\/td><td>INFOROM_DRAM_RETIREMENT_EVENT<\/td><td>ECC \u5355\u6bd4\u7279\u9519\u8bef\u8fc7\u591a\uff0cGPU \u6210\u529f\u628a\u574f\u9875 \/ \u574f\u884c\u505a\u9694\u79bb\u9000\u4f11\uff1b\u4fe1\u606f\u65e5\u5fd7\uff0c\u975e\u81f4\u547d<\/td><td>\u89c2\u5bdf ECC \u8ba1\u6570\uff1b\u6301\u7eed\u4e0a\u6da8\u9700\u8981\u5173\u6ce8\u786c\u4ef6\u5bff\u547d<\/td><\/tr><tr><td><strong>64<\/strong><\/td><td>INFOROM_DRAM_RETIREMENT_FAILURE<\/td><td>\u5c1d\u8bd5\u9694\u79bb\u574f\u663e\u5b58\u9875<strong>\u5931\u8d25<\/strong>\uff0c\u786c\u4ef6\u95ee\u9898\uff0c\u98ce\u9669\u9ad8<\/td><td>\u8dd1 Field\u2011Diag\uff0c\u5efa\u8bae\u51c6\u5907\u6362\u5361<\/td><\/tr><tr><td><strong>74<\/strong><\/td><td>NVLink Error<\/td><td>NVLink \u94fe\u8def\u9519\u8bef\uff0c\u591a\u5361\u673a\u5668\uff1b\u7ebf\u7f06\u3001NVSwitch\u3001GPU \u94fe\u8def\u786c\u4ef6\u6545\u969c<\/td><td>\u5355\u5361 V100 \u4e0d\u4f1a\u51fa\u73b0\uff1b\u591a\u5361\u770b<code>nvidia\u2011smi nvlink<\/code><\/td><\/tr><tr><td><strong>79<\/strong><\/td><td>GPU has fallen off the bus\ud83d\udd25<\/td><td><strong>\u6389\u5361\uff0cGPU \u4ece PCIe \u603b\u7ebf\u6d88\u5931<\/strong>\uff1bPCIe \u94fe\u8def\u65ad\u88c2\uff0c\u670d\u52a1\u5668\u8bc6\u522b\u4e0d\u5230 GPU\uff1b\u786c\u4ef6\u3001\u63d2\u69fd\u3001\u4f9b\u7535\u6545\u969cTencent Cl&#8230;<\/td><td>\u6574\u673a\u51b7\u91cd\u542f\uff1b\u4f9d\u7136\u590d\u73b0\uff0c\u786c\u4ef6\u6545\u969c\uff0c\u66f4\u6362 GPU \/ \u6392\u67e5\u4e3b\u677f\u63d2\u69fd<\/td><\/tr><tr><td><strong>92<\/strong><\/td><td>Excessive SBE interrupts<\/td><td>\u5927\u91cf\u5355\u6bd4\u7279\u53ef\u7ea0\u6b63 ECC \u62a5\u9519\uff1bSBE \u7d2f\u79ef\u8fc7\u9ad8\uff0c\u6f5c\u5728\u663e\u5b58\u9690\u60a3<\/td><td>\u6301\u7eed\u589e\u957f\u8981\u505a\u786c\u4ef6\u8bca\u65ad<\/td><\/tr><tr><td><strong>94<\/strong><\/td><td>Contained Error \u9694\u79bb\u9519\u8bef (A100+)<\/td><td>\u9519\u8bef\u88ab\u9650\u5236\u5728\u51fa\u9519\u8fdb\u7a0b\uff0c\u5176\u4ed6\u4e1a\u52a1\u4e0d\u53d7\u5f71\u54cd\uff1b\u4e00\u822c ECC \u663e\u5b58\u9519\u8bef\uff0cAmpere \u4ee5\u540e\u624d\u6709\uff0cV100 \u65e0\u6b64 XID<\/td><td>\u91cd\u542f\u5f02\u5e38\u8fdb\u7a0b\uff0c\u7a7a\u95f2\u65f6\u91cd\u7f6e GPU<\/td><\/tr><tr><td><strong>95<\/strong><\/td><td>Uncontained Error \u4e0d\u53ef\u9694\u79bb\u9519\u8bef (A100+)<\/td><td>\u9519\u8bef\u6269\u6563\uff0c\u6574\u5f20 GPU \u5168\u90e8\u6302\u6389\uff0c\u5fc5\u987b\u91cd\u7f6e GPU\uff0cV100 \u4e0d\u4f1a\u51fa\u73b0<\/td><td>GPU \u91cd\u7f6e\uff0c\u590d\u73b0\u5219\u786c\u4ef6\u6545\u969c<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\">\u6392\u67e5\u5b9e\u64cd\u547d\u4ee4\uff08V100 \u670d\u52a1\u5668\uff09<\/h2>\n\n\n\n<pre class=\"wp-block-code\"><code># \u67e5\u770bXID\u62a5\u9519\ndmesg -T | grep -i xid\n\n# \u67e5\u770bECC\u7edf\u8ba1\uff0cSBE\u5355\u6bd4\u7279\u3001DBE\u53cc\u6bd4\u7279\nnvidia-smi -q -d ECC\n\n# \u6536\u96c6nvidia\u5b8c\u6574bug\u62a5\u544a\uff0c\u63d0\u4ea4\u7ed9\u5382\u5546\nnvidia-bug-report.sh\n<\/code><\/pre>\n\n\n\n<h2 class=\"wp-block-heading\">\u7b80\u5355\u5224\u65ad\u53e3\u8bc0<\/h2>\n\n\n\n<ol class=\"wp-block-list\">\n<li><strong>\u53ea\u8dd1\u67d0\u4e2a\u4e1a\u52a1\u624d\u62a5 XID (13\/31)\uff0c\u6362\u522b\u7684\u7a0b\u5e8f\u6b63\u5e38 \u2192 \u5927\u6982\u7387\u8f6f\u4ef6 \/ \u7a0b\u5e8f bug<\/strong><\/li>\n\n\n\n<li><strong>\u8dd1 gpu\u2011burn \u70e4\u673a\u5c31\u62a5 XID\uff0c\u6362\u7a0b\u5e8f\u4f9d\u7136\u590d\u73b0 \u2192 \u9ad8\u5ea6\u6000\u7591\u786c\u4ef6<\/strong><\/li>\n\n\n\n<li><strong>\u51fa\u73b0 XID48 \/79 \u2192 \u4e25\u91cd\u786c\u4ef6\u6545\u969c\uff0c\u4f18\u5148\u786c\u4ef6\u8bca\u65ad<\/strong><\/li>\n\n\n\n<li><strong>XID45 \u6c38\u8fdc\u4e0d\u662f\u6839\u56e0\uff0c\u770b\u5b83\u524d\u9762\u7684 XID<\/strong><\/li>\n<\/ol>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n","protected":false},"excerpt":{"rendered":"<p>NVIDIA GPU \u5178\u578b XID \u62a5\u9519 nvidia\u5b98\u65b9xid\u9519\u8bef\u624b\u518c\uff1ahttps:\/\/docs.nvidi [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[],"class_list":["post-159","post","type-post","status-publish","format-standard","hentry","category-7"],"_links":{"self":[{"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/posts\/159","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=159"}],"version-history":[{"count":4,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/posts\/159\/revisions"}],"predecessor-version":[{"id":164,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=\/wp\/v2\/posts\/159\/revisions\/164"}],"wp:attachment":[{"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=159"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=159"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.xiaoxindong.cloud\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=159"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}