Re: [RFC PATCH V2 5/7] x86/entry: don't shift stack on #DB

From: Peter Zijlstra
Date: Tue May 26 2020 - 05:11:10 EST

Next message: Catalin Marinas: "Re: [PATCH] mm/z3fold: silence kmemleak false positives of slots"
Previous message: Oleksij Rempel: "[PATCH ethtool v1] netlink: add master/slave configuration support"
In reply to: Lai Jiangshan: "[RFC PATCH V2 5/7] x86/entry: don't shift stack on #DB"
Next in thread: Lai Jiangshan: "[RFC PATCH V2 6/7] x86/entry: is_debug_stack() don't check of DB1 stack"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

On Tue, May 26, 2020 at 01:42:19AM +0000, Lai Jiangshan wrote:
> debug_enter() will disable #DB, there should be no recursive #DB.

should being the operative word.

I have similar patches but was still debating what to do about kgdb, I
think we'll just have to mark it BROKEN for now.

Next message: Catalin Marinas: "Re: [PATCH] mm/z3fold: silence kmemleak false positives of slots"
Previous message: Oleksij Rempel: "[PATCH ethtool v1] netlink: add master/slave configuration support"
In reply to: Lai Jiangshan: "[RFC PATCH V2 5/7] x86/entry: don't shift stack on #DB"
Next in thread: Lai Jiangshan: "[RFC PATCH V2 6/7] x86/entry: is_debug_stack() don't check of DB1 stack"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]