Re: [PATCH 05/15] x86_32: Provide consistent pt_regs

2019-06-11 Thread Peter Zijlstra
On Fri, Jun 07, 2019 at 03:32:15PM -0400, Josh Poimboeuf wrote: > On Wed, Jun 05, 2019 at 03:07:58PM +0200, Peter Zijlstra wrote: > > Currently pt_regs on x86_32 has an oddity in that kernel regs > > (!user_mode(regs)) are short two entries (esp/ss). This means that any > > code trying to use them

Re: [PATCH 05/15] x86_32: Provide consistent pt_regs

2019-06-07 Thread Josh Poimboeuf
On Wed, Jun 05, 2019 at 03:07:58PM +0200, Peter Zijlstra wrote: > Currently pt_regs on x86_32 has an oddity in that kernel regs > (!user_mode(regs)) are short two entries (esp/ss). This means that any > code trying to use them (typically: regs->sp) needs to jump through > some unfortunate hoops. >

Re: [PATCH 05/15] x86_32: Provide consistent pt_regs

2019-06-07 Thread Masami Hiramatsu
On Wed, 05 Jun 2019 15:07:58 +0200 Peter Zijlstra wrote: > Currently pt_regs on x86_32 has an oddity in that kernel regs > (!user_mode(regs)) are short two entries (esp/ss). This means that any > code trying to use them (typically: regs->sp) needs to jump through > some unfortunate hoops. > > Ch

[PATCH 05/15] x86_32: Provide consistent pt_regs

2019-06-05 Thread Peter Zijlstra
Currently pt_regs on x86_32 has an oddity in that kernel regs (!user_mode(regs)) are short two entries (esp/ss). This means that any code trying to use them (typically: regs->sp) needs to jump through some unfortunate hoops. Change the entry code to fix this up and create a full pt_regs frame. Th