lkml.org 
[lkml]   [2009]   [May]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC patch 15/20] LTTng Kernel Trace Thread Flag SH
On Sat, May 09, 2009 at 12:22:24PM -0400, Mathieu Desnoyers wrote:
> Index: linux-2.6-lttng/arch/sh/include/asm/thread_info.h
> ===================================================================
> --- linux-2.6-lttng.orig/arch/sh/include/asm/thread_info.h 2009-03-15 15:57:04.000000000 -0400
> +++ linux-2.6-lttng/arch/sh/include/asm/thread_info.h 2009-03-15 15:57:17.000000000 -0400
> @@ -116,6 +116,7 @@ extern void free_thread_info(struct thre
> #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */
> #define TIF_SECCOMP 6 /* secure computing */
> #define TIF_NOTIFY_RESUME 7 /* callback before returning to user */
> +#define TIF_KERNEL_TRACE 8 /* kernel trace active */
> #define TIF_USEDFPU 16 /* FPU was used by this task this quantum (SMP) */
> #define TIF_POLLING_NRFLAG 17 /* true if poll_idle() is polling TIF_NEED_RESCHED */
> #define TIF_MEMDIE 18
> @@ -129,6 +130,7 @@ extern void free_thread_info(struct thre
> #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
> #define _TIF_SECCOMP (1 << TIF_SECCOMP)
> #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME)
> +#define _TIF_KERNEL_TRACE (1 << TIF_KERNEL_TRACE)
> #define _TIF_USEDFPU (1 << TIF_USEDFPU)
> #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG)
> #define _TIF_FREEZE (1 << TIF_FREEZE)
> @@ -141,17 +143,19 @@ extern void free_thread_info(struct thre
>
> /* work to do in syscall trace */
> #define _TIF_WORK_SYSCALL_MASK (_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP | \
> - _TIF_SYSCALL_AUDIT | _TIF_SECCOMP)
> + _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | \
> + _TIF_KERNEL_TRACE)
>
> /* work to do on any return to u-space */
> #define _TIF_ALLWORK_MASK (_TIF_SYSCALL_TRACE | _TIF_SIGPENDING | \
> _TIF_NEED_RESCHED | _TIF_SYSCALL_AUDIT | \
> _TIF_SINGLESTEP | _TIF_RESTORE_SIGMASK | \
> - _TIF_NOTIFY_RESUME)
> + _TIF_NOTIFY_RESUME | _TIF_KERNEL_TRACE)
>
> /* work to do on interrupt/exception return */
> #define _TIF_WORK_MASK (_TIF_ALLWORK_MASK & ~(_TIF_SYSCALL_TRACE | \
> - _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP))
> + _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP | \
> + _TIF_KERNEL_TRACE))
>
> #endif /* __KERNEL__ */
>
I think you missed the comment above this hunk in the code..

This will blow up immediately, _TIF_ALLWORK_MASK must presently fit
within a byte, as it just happens to right now. If this can take the
place of _TIF_SYSCALL_TRACE in the future, then that bit position can be
used instead, otherwise the assembly code will have to be rewritten to
load a larger value, which means we lose the ability to load the mask as
an immediate without resorting to shifting and masking :-(

Other platforms have similar constraints, have you verified that this is
not a problem on any of the other platforms?

I'll of course rewrite the assembly if we can't avoid it.


\
 
 \ /
  Last update: 2009-05-09 19:23    [W:1.724 / U:0.004 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site