From: Gerald Britton Date: Tue, 14 Feb 2006 15:19:04 +0000 (-0500) Subject: [PATCH] x86: fix oprofile kernel callgraph regression X-Git-Tag: v2.6.16-rc4~98 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=303794400992b907b7cac0d91788603636c7e0fe;p=platform%2Fkernel%2Flinux-3.10.git [PATCH] x86: fix oprofile kernel callgraph regression Fix x86 oprofile regression introduced by: commit c34d1b4d165c67b966bca4aba026443d7ff161eb [PATCH] mm: kill check_user_page_readable That commit reorganized tests for the userspace stack walking moving all those tests into dump_backtrace(), however, dump_backtrace() was used for both userspace and kernel stalk walking. The result is typically no recorded callgraph information for kernel samples. Revive the original function as dump_kernel_backtrace() and rename the other to dump_user_backtrace() to avoid future confusion. Signed-off-by: Gerald Britton Apology-from: Hugh Dickins Signed-off-by: Linus Torvalds --- diff --git a/arch/i386/oprofile/backtrace.c b/arch/i386/oprofile/backtrace.c index acc1813..c049ce4 100644 --- a/arch/i386/oprofile/backtrace.c +++ b/arch/i386/oprofile/backtrace.c @@ -20,7 +20,20 @@ struct frame_head { } __attribute__((packed)); static struct frame_head * -dump_backtrace(struct frame_head * head) +dump_kernel_backtrace(struct frame_head * head) +{ + oprofile_add_trace(head->ret); + + /* frame pointers should strictly progress back up the stack + * (towards higher addresses) */ + if (head >= head->ebp) + return NULL; + + return head->ebp; +} + +static struct frame_head * +dump_user_backtrace(struct frame_head * head) { struct frame_head bufhead[2]; @@ -105,10 +118,10 @@ x86_backtrace(struct pt_regs * const regs, unsigned int depth) if (!user_mode_vm(regs)) { while (depth-- && valid_kernel_stack(head, regs)) - head = dump_backtrace(head); + head = dump_kernel_backtrace(head); return; } while (depth-- && head) - head = dump_backtrace(head); + head = dump_user_backtrace(head); }