On Thu, Mar 07, 2019 at 01:18:41PM -0500, Steven Rostedt wrote:
> On Thu, 7 Mar 2019 09:45:35 -0800
> Linus Torvalds wrote:
>
> > On Thu, Mar 7, 2019 at 9:38 AM Peter Zijlstra wrote:
> > >
> > > Also; it seems to me that something PT, or maybe even simply:
> > >
> > > perf -e branches -e
On Thu, 7 Mar 2019 09:45:35 -0800
Linus Torvalds wrote:
> On Thu, Mar 7, 2019 at 9:38 AM Peter Zijlstra wrote:
> >
> > Also; it seems to me that something PT, or maybe even simply:
> >
> > perf -e branches -e branch-misses
> >
> > would get you similar or sufficient information.
>
> Yeah,
On Thu, Mar 7, 2019 at 9:46 AM Josh Poimboeuf wrote:
>
> Is that possible to do in a C macro? Doesn't seem to work for me...
The meat of that macro could easily be done as a helper inline function.
But as mentioned, I think a better option would be to remove it
entirely, if at all possible.
On Thu, Mar 07, 2019 at 09:29:17AM -0800, h...@zytor.com wrote:
> On March 7, 2019 9:18:29 AM PST, Josh Poimboeuf wrote:
> >On Thu, Mar 07, 2019 at 09:04:36AM -0800, h...@zytor.com wrote:
> >> On March 7, 2019 8:47:05 AM PST, Josh Poimboeuf
> >wrote:
> >> >On Thu, Mar 07, 2019 at 02:13:12PM
On Thu, Mar 07, 2019 at 06:43:22PM +0100, Peter Zijlstra wrote:
> On Thu, Mar 07, 2019 at 10:47:05AM -0600, Josh Poimboeuf wrote:
>
> > This "fixes" it, and also seems to help -Os make much code:
> >
> > diff --git a/include/linux/compiler.h b/include/linux/compiler.h
> > index
On Thu, Mar 7, 2019 at 9:38 AM Peter Zijlstra wrote:
>
> Also; it seems to me that something PT, or maybe even simply:
>
> perf -e branches -e branch-misses
>
> would get you similar or sufficient information.
Yeah, I'm not really seeing a lot of upside to PROFILE_ALL_BRANCHES.
Particularly
On Thu, Mar 07, 2019 at 10:47:05AM -0600, Josh Poimboeuf wrote:
> This "fixes" it, and also seems to help -Os make much code:
>
> diff --git a/include/linux/compiler.h b/include/linux/compiler.h
> index 445348facea9..8de63db58fdd 100644
> --- a/include/linux/compiler.h
> +++
On Thu, Mar 07, 2019 at 11:17:09AM -0600, Josh Poimboeuf wrote:
> On Thu, Mar 07, 2019 at 09:00:49AM -0800, Linus Torvalds wrote:
> > But that macro really is the macro from hell regardless.
> >
> > Do people really use CONFIG_PROFILE_ALL_BRANCHES?
>
> IIRC, Steven runs it once a year or so...
On March 7, 2019 9:18:29 AM PST, Josh Poimboeuf wrote:
>On Thu, Mar 07, 2019 at 09:04:36AM -0800, h...@zytor.com wrote:
>> On March 7, 2019 8:47:05 AM PST, Josh Poimboeuf
>wrote:
>> >On Thu, Mar 07, 2019 at 02:13:12PM +0100, Peter Zijlstra wrote:
>> >> On Thu, Mar 07, 2019 at 01:55:26PM +0100,
On Thu, Mar 07, 2019 at 09:04:36AM -0800, h...@zytor.com wrote:
> On March 7, 2019 8:47:05 AM PST, Josh Poimboeuf wrote:
> >On Thu, Mar 07, 2019 at 02:13:12PM +0100, Peter Zijlstra wrote:
> >> On Thu, Mar 07, 2019 at 01:55:26PM +0100, Peter Zijlstra wrote:
> >> > On Thu, Mar 07, 2019 at
On Thu, Mar 07, 2019 at 09:00:49AM -0800, Linus Torvalds wrote:
> On Thu, Mar 7, 2019 at 8:47 AM Josh Poimboeuf wrote:
> >
> > This "fixes" it, and also seems to help -Os make much code:
>
> Yeah, considering that this __trace_if() macro from hell is doing an
> 'if()' on the result of that inner
On March 7, 2019 3:45:11 AM PST, Peter Zijlstra wrote:
>Teach objtool to validate the UACCESS (SMAP, PAN) rules with are
>currently
>unenforced and (therefore obviously) violated.
>
>UACCESS sections should be small; we want to limit the amount of code
>that can
>touch userspace. Furthermore,
On March 7, 2019 8:47:05 AM PST, Josh Poimboeuf wrote:
>On Thu, Mar 07, 2019 at 02:13:12PM +0100, Peter Zijlstra wrote:
>> On Thu, Mar 07, 2019 at 01:55:26PM +0100, Peter Zijlstra wrote:
>> > On Thu, Mar 07, 2019 at 01:03:17PM +0100, Peter Zijlstra wrote:
>>
>>
>> > > 01be 20d3: 31 c0
On Thu, Mar 7, 2019 at 8:47 AM Josh Poimboeuf wrote:
>
> This "fixes" it, and also seems to help -Os make much code:
Yeah, considering that this __trace_if() macro from hell is doing an
'if()' on the result of that inner thing, it makes sense to *not* use
that "looks simpler and shorter" array
On Thu, Mar 07, 2019 at 10:47:05AM -0600, Josh Poimboeuf wrote:
> On Thu, Mar 07, 2019 at 02:13:12PM +0100, Peter Zijlstra wrote:
> > On Thu, Mar 07, 2019 at 01:55:26PM +0100, Peter Zijlstra wrote:
> > > On Thu, Mar 07, 2019 at 01:03:17PM +0100, Peter Zijlstra wrote:
> >
> >
> > > > 01be
On Thu, Mar 07, 2019 at 02:13:12PM +0100, Peter Zijlstra wrote:
> On Thu, Mar 07, 2019 at 01:55:26PM +0100, Peter Zijlstra wrote:
> > On Thu, Mar 07, 2019 at 01:03:17PM +0100, Peter Zijlstra wrote:
>
>
> > > 01be 20d3: 31 c0 xor%eax,%eax
> > > 01c0 20d5: 4c 39 eb
On Thu, Mar 7, 2019 at 4:03 AM Peter Zijlstra wrote:
>
> Take for instance this one (.config attached); it has both
> CONFIG_PROFILE_ALL_BRANCHES=y and CONFIG_TRACE_BRANCH_PROFILING=y
> and it compiles:
How about just turning off SMAP checking for the really odd cases?
At some point it's not
On Thu, Mar 07, 2019 at 01:55:26PM +0100, Peter Zijlstra wrote:
> On Thu, Mar 07, 2019 at 01:03:17PM +0100, Peter Zijlstra wrote:
> > 01be 20d3: 31 c0 xor%eax,%eax
> > 01c0 20d5: 4c 39 ebcmp%r13,%rbx
> > 01c3 20d8: 77 08
On Thu, Mar 07, 2019 at 01:03:17PM +0100, Peter Zijlstra wrote:
> 01c3 20d8:77 08 ja 20e2
> <__do_sys_waitid+0x1cd>
taken:
randconfig-build/kernel/exit.o: warning: objtool: __do_sys_waitid()+0x1c3:
(branch)
> e0f5 alternative("",
On Thu, Mar 07, 2019 at 12:45:11PM +0100, Peter Zijlstra wrote:
> I've only got a few randconfig
> failures left (GCC-8) that I'm not quite understanding.
Take for instance this one (.config attached); it has both
CONFIG_PROFILE_ALL_BRANCHES=y and CONFIG_TRACE_BRANCH_PROFILING=y
and it compiles:
Teach objtool to validate the UACCESS (SMAP, PAN) rules with are currently
unenforced and (therefore obviously) violated.
UACCESS sections should be small; we want to limit the amount of code that can
touch userspace. Furthermore, UACCESS state isn't scheduled, this means that
anything that
21 matches
Mail list logo