Hi Bhupesh, On 10/29/2018 3:50 AM, Bhupesh Sharma wrote: > makedumpfile '--mem-usage' use-case is currently broken on x86_64 with > latest kernels (see a sample log below with 4.19 kernel): > > [root@hp-dl380pgen8-02-vm-15 ~]# makedumpfile -f --mem-usage /proc/kcore > The kernel version is not supported. > The makedumpfile operation may be incomplete. > <snip..> > set_kcore_vmcoreinfo: Can't get the offset of VMCOREINFO(/proc/kcore). > Success > > makedumpfile Failed. > > It seems that the issue is seen because KCORE_REMAP which was added > to recent kernels and appears as a PT_LOAD in '/proc/kcore'. I'm a bit late, but as for this problem: * Since kernel 4.19 itself doesn't have NUMBER(page_offset_base) in vmcoreinfo, we need to merge the patch I sent in order to support kernel 4.19. http://lists.infradead.org/pipermail/kexec/2018-October/021749.html * It looks like the following commit already removed KCORE_REMAP PT_LOAD from /proc/kcore at 4.20-rc1, and the problem doesn't occur with mainline kernel. commit bf904d2762ee6fc1e4acfcb0772bbfb4a27ad8a6 Author: Andy Lutomirski <luto@xxxxxxxxxx> Date: Mon Sep 3 15:59:44 2018 -0700 x86/pti/64: Remove the SYSCALL64 entry trampoline ... - /* - * The cpu_entry_area alias addresses are not in the kernel binary - * so they do not show up in /proc/kcore normally. This adds entries - * for them manually. - */ - kclist_add_remap(&per_cpu(kcore_entry_trampoline, cpu), - _entry_trampoline, - &get_cpu_entry_area(cpu)->entry_trampoline, PAGE_SIZE); But yes, kcore PT_LOADs seems changeable like this. Thanks, Kazu _______________________________________________ kexec mailing list kexec@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/kexec
