[dpdk-dev] [PATCH v3] eal: make hugetlb initialization more robust

Hi Jianfeng, On 09/05/2016 11:48, Jianfeng Tan wrote: > /* find physical addresses and sockets for each hugepage */ > @@ -1172,8 +1255,9 @@ rte_eal_hugepage_init(void) > hp_offset += new_pages_count[i]; > #else > /* remap all hugepages */ > -

[dpdk-dev] [PATCH v3] eal: make hugetlb initialization more robust

Hi Sergio, > -Original Message- > From: Gonzalez Monroy, Sergio > Sent: Tuesday, May 10, 2016 4:55 PM > To: Tan, Jianfeng; dev at dpdk.org > Cc: david.marchand at 6wind.com; nhorman at tuxdriver.com > Subject: Re: [PATCH v3] eal: make hugetlb initialization more robust > > > Hi Jianfeng,

[dpdk-dev] [PATCH v3] eal: make hugetlb initialization more robust

This patch adds an option, --huge-trybest, to use a recover mechanism to the case that there are not so many hugepages (declared in sysfs), which can be used. It relys on a mem access to fault-in hugepages, and if fails with SIGBUS, recover to previously saved stack environment with siglongjmp().