[dpdk-dev] [Bug 785] rte_pktmbuf_alloc fails with a segmentation fault but rte_pktmbuf_pool_create r

link管理

链接快照平台

输入网页链接，自动生成快照
标签化管理网页链接

相关文章推荐

憨厚的紫菜汤 · 申办认证须知 — 中华人民共和国驻奥克兰总领馆· 3 周前 ·

果断的可乐 · 偷偷藏不住免费阅读快看漫画(细雨飞花)小说全 ...· 3 月前 ·

豁达的帽子 · Home | Team Focus· 7 月前 ·

豪情万千的花卷 · android 判断开机向导是否完成 ...· 8 月前 ·

宽容的楼梯 · TensorRT教程（六）使用Python和 ...· 9 月前 ·

Mon Aug 9 15:23:57 CEST 2021

Previous message (by thread): [dpdk-dev] [PATCH v2] doc: fix CI typo warnings

Next message (by thread): [dpdk-dev] [PATCH] net/mlx5: fix matching on eCPRI

Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] Bug ID: 785 Summary: rte_pktmbuf_alloc fails with a segmentation fault but rte_pktmbuf_pool_create returns valid data Product: DPDK Version: 20.11 Hardware: All OS: All Status: UNCONFIRMED Severity: normal Priority: Normal Component: core Assignee: dev at dpdk.org Reporter: brydon_gibson at dell.com Target Milestone: --- (backtrace below) mbuf_pool is valid, and there is sufficient memory onboard, yet __mempool_generic_get (from rte_pktmbuf_alloc) fails because cache points to invalid memory. There doesn't seem to be any way of detecting that something is wrong because pool_create returns perfectly good data. Backtrace: Program received signal SIGSEGV, Segmentation fault. 0x0000000000402735 in __mempool_generic_get (cache=0x2f6e69622f77735f, n=1, obj_table=0x7fffffffe0d8, mp=0x7fffffffe6db) at /usr/local/include/rte_mempool.h:1417 1417 if (unlikely(cache == NULL || n >= cache->size)) (gdb) bt full #0 0x0000000000402735 in __mempool_generic_get (cache=0x2f6e69622f77735f, n=1, obj_table=0x7fffffffe0d8, mp=0x7fffffffe6db) at /usr/local/include/rte_mempool.h:1417 ret = 0 len = 4294960256 index = 32767 cache_objs = 0x0 #1 rte_mempool_generic_get (cache=0x2f6e69622f77735f, n=1, obj_table=0x7fffffffe0d8, mp=0x7fffffffe6db) at /usr/local/include/rte_mempool.h:1491 ret = 0 #2 rte_mempool_get_bulk (n=1, obj_table=0x7fffffffe0d8, mp=0x7fffffffe6db) at /usr/local/include/rte_mempool.h:1526 cache = 0x2f6e69622f77735f #3 rte_mempool_get (obj_p=0x7fffffffe0d8, mp=0x7fffffffe6db) at /usr/local/include/rte_mempool.h:1552 No locals. #4 rte_mbuf_raw_alloc (mp=0x7fffffffe6db) at /usr/local/include/rte_mbuf.h:590 m = 0x1656380 #5 0x0000000000402cd2 in rte_pktmbuf_alloc (mp=0x7fffffffe6db) at /usr/local/include/rte_mbuf.h:895 m = 0x7ffff70e5d4d <__GI_rewind+109> #6 0x000000000040798e in getAndMbufData (argstring=0x7a9935 "-f", einf=0x7fffffffe480) at cmdParser.c:73 fd = 0x1656380 fsize = 0 readSize = 5 data = 0x1651f20 "" argv = {0x4095f0 "progname", 0x7a9935 "-f", 0x7a9938 "test.txt", 0x0} argc = 3 opt = -1 myBuf = 1 total_mbufs = 0 bufToUse = 0x7fffffffe70a You are receiving this mail because: You are the assignee for the bug.

Previous message (by thread): [dpdk-dev] [PATCH v2] doc: fix CI typo warnings

Next message (by thread): [dpdk-dev] [PATCH] net/mlx5: fix matching on eCPRI

Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]