[tip:x86/process] x86_64: Record stack pointer before task execution begins

From: tip-bot for Siddhesh Poyarekar
Date: Mon Feb 27 2012 - 01:18:08 EST


Commit-ID: 42dfc43ee5999ac64284476ea0ac6c937587cf2b
Gitweb: http://git.kernel.org/tip/42dfc43ee5999ac64284476ea0ac6c937587cf2b
Author: Siddhesh Poyarekar <siddhesh.poyarekar@xxxxxxxxx>
AuthorDate: Sun, 26 Feb 2012 21:47:55 +0530
Committer: H. Peter Anvin <hpa@xxxxxxxxx>
CommitDate: Sun, 26 Feb 2012 12:59:04 -0800

x86_64: Record stack pointer before task execution begins

task->thread.usersp is unusable immediately after a binary is exec()'d
until it undergoes a context switch cycle. The start_thread() function
called during execve() saves the stack pointer into pt_regs and into
old_rsp, but fails to record it into task->thread.usersp.

Because of this, KSTK_ESP(task) returns an incorrect value for a
64-bit program until the task is switched out and back in since
switch_to swaps %rsp values in and out into task->thread.usersp.

Signed-off-by: Siddhesh Poyarekar <siddhesh.poyarekar@xxxxxxxxx>
Link: http://lkml.kernel.org/r/1330273075-2949-1-git-send-email-siddhesh.poyarekar@xxxxxxxxx
Signed-off-by: H. Peter Anvin <hpa@xxxxxxxxx>
---
arch/x86/kernel/process_64.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 1fd94bc..eb54dd0 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -341,6 +341,7 @@ start_thread_common(struct pt_regs *regs, unsigned long new_ip,
loadsegment(es, _ds);
loadsegment(ds, _ds);
load_gs_index(0);
+ current->thread.usersp = new_sp;
regs->ip = new_ip;
regs->sp = new_sp;
percpu_write(old_rsp, new_sp);
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/