On Mon, Mar 2, 2015 at 2:04 PM, Tom Zanussi wrote:
> On Mon, 2015-03-02 at 11:29 -0800, Alexei Starovoitov wrote:
>> On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
>> wrote:
>> >> >
>> >> > The idea would be that instead of getting your individually kmalloc'ed
>> >> > elements on-demand from
On Mon, 2015-03-02 at 11:29 -0800, Alexei Starovoitov wrote:
> On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
> wrote:
> >> >
> >> > The idea would be that instead of getting your individually kmalloc'ed
> >> > elements on-demand from kmalloc while in the handler, you'd get them
> >> > from a pool
On Mon, Mar 2, 2015 at 11:33 AM, Steven Rostedt wrote:
> On Mon, 2 Mar 2015 11:24:04 -0800
> Alexei Starovoitov wrote:
>
>> well, percentage of tracepoints called from NMI is tiny
>> comparing to the rest, so assuming nmi context
>> everywhere is very inefficient.
>> Like we can use
On Mon, 2 Mar 2015 11:24:04 -0800
Alexei Starovoitov wrote:
> well, percentage of tracepoints called from NMI is tiny
> comparing to the rest, so assuming nmi context
> everywhere is very inefficient.
> Like we can use pre-allocated pool of map entries when
> tracepoint is called from NMI, but
On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
wrote:
>> >
>> > The idea would be that instead of getting your individually kmalloc'ed
>> > elements on-demand from kmalloc while in the handler, you'd get them
>> > from a pool you've pre-allocated when you set up the table. This could
>> > be from
On Mon, Mar 2, 2015 at 10:43 AM, Steven Rostedt wrote:
> On Mon, 2 Mar 2015 10:12:32 -0800
> Alexei Starovoitov wrote:
>
>> I'm not sure what would be the meaning of hash map that has all
>> elements pre-allocated...
>> As I'm reading your cover letter, I agree, we need to find a way
>> to call
On Mon, 2015-03-02 at 10:40 -0800, Alexei Starovoitov wrote:
> On Mon, Mar 2, 2015 at 10:25 AM, Tom Zanussi
> wrote:
> > On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
> >> On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
> >> wrote:
> >> > On Mon, 2015-03-02 at 09:58 -0800, Alexei
On Mon, 2 Mar 2015 10:12:32 -0800
Alexei Starovoitov wrote:
> I'm not sure what would be the meaning of hash map that has all
> elements pre-allocated...
> As I'm reading your cover letter, I agree, we need to find a way
> to call kmalloc_notrace-like from tracepoints.
> Not sure that patch 8
On Mon, Mar 2, 2015 at 10:25 AM, Tom Zanussi
wrote:
> On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
>> On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
>> wrote:
>> > On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
>> >> On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi
>> >>
On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
> On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
> wrote:
> > On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
> >> On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi
> >> wrote:
> >> > On Mon, 2015-03-02 at 11:37 -0500, Steven
On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
wrote:
> On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
>> On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi
>> wrote:
>> > On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
>> >> On Mon, 2 Mar 2015 10:01:00 -0600
>> >> Tom Zanussi
On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
> On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi
> wrote:
> > On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
> >> On Mon, 2 Mar 2015 10:01:00 -0600
> >> Tom Zanussi wrote:
> >>
> >> > Add a gfp flag that allows kmalloc() et al
On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi wrote:
> On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
>> On Mon, 2 Mar 2015 10:01:00 -0600
>> Tom Zanussi wrote:
>>
>> > Add a gfp flag that allows kmalloc() et al to be used in tracing
>> > functions.
>> >
>> > The problem with using
On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
> On Mon, 2 Mar 2015 10:01:00 -0600
> Tom Zanussi wrote:
>
> > Add a gfp flag that allows kmalloc() et al to be used in tracing
> > functions.
> >
> > The problem with using kmalloc for tracing is that the tracing
> > subsystem should be
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi wrote:
> Add a gfp flag that allows kmalloc() et al to be used in tracing
> functions.
>
> The problem with using kmalloc for tracing is that the tracing
> subsystem should be able to trace kmalloc itself, which it can't do
> directly because of
On Mon, 2015-03-02 at 11:29 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
The idea would be that instead of getting your individually kmalloc'ed
elements on-demand from kmalloc while in the handler, you'd get them
On Mon, Mar 2, 2015 at 2:04 PM, Tom Zanussi tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 11:29 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
The idea would be that instead of getting your individually
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi tom.zanu...@linux.intel.com wrote:
Add a gfp flag that allows kmalloc() et al to be used in tracing
functions.
The problem with using kmalloc for tracing is that the tracing
subsystem should be able to trace kmalloc itself, which it can't do
On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi tom.zanu...@linux.intel.com wrote:
Add a gfp flag that allows kmalloc() et al to be used in tracing
functions.
The problem with using kmalloc for tracing is that the tracing
On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi tom.zanu...@linux.intel.com
wrote:
On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
On Mon, 2
On Mon, Mar 2, 2015 at 10:25 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
On Mon,
On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi tom.zanu...@linux.intel.com wrote:
Add a gfp flag that allows kmalloc() et al to be used in tracing
On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi tom.zanu...@linux.intel.com
wrote:
On
On Mon, 2 Mar 2015 10:12:32 -0800
Alexei Starovoitov a...@plumgrid.com wrote:
I'm not sure what would be the meaning of hash map that has all
elements pre-allocated...
As I'm reading your cover letter, I agree, we need to find a way
to call kmalloc_notrace-like from tracepoints.
Not sure
On Mon, 2015-03-02 at 09:58 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 8:46 AM, Tom Zanussi tom.zanu...@linux.intel.com
wrote:
On Mon, 2015-03-02 at 11:37 -0500, Steven Rostedt wrote:
On Mon, 2 Mar 2015 10:01:00 -0600
Tom Zanussi tom.zanu...@linux.intel.com wrote:
Add a
On Mon, 2015-03-02 at 10:40 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:25 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On Mon, 2015-03-02 at 10:12 -0800, Alexei Starovoitov wrote:
On Mon, Mar 2, 2015 at 10:03 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
On
On Mon, Mar 2, 2015 at 10:54 AM, Tom Zanussi
tom.zanu...@linux.intel.com wrote:
The idea would be that instead of getting your individually kmalloc'ed
elements on-demand from kmalloc while in the handler, you'd get them
from a pool you've pre-allocated when you set up the table. This
On Mon, 2 Mar 2015 11:24:04 -0800
Alexei Starovoitov a...@plumgrid.com wrote:
well, percentage of tracepoints called from NMI is tiny
comparing to the rest, so assuming nmi context
everywhere is very inefficient.
Like we can use pre-allocated pool of map entries when
tracepoint is called
On Mon, Mar 2, 2015 at 11:33 AM, Steven Rostedt rost...@goodmis.org wrote:
On Mon, 2 Mar 2015 11:24:04 -0800
Alexei Starovoitov a...@plumgrid.com wrote:
well, percentage of tracepoints called from NMI is tiny
comparing to the rest, so assuming nmi context
everywhere is very inefficient.
On Mon, Mar 2, 2015 at 10:43 AM, Steven Rostedt rost...@goodmis.org wrote:
On Mon, 2 Mar 2015 10:12:32 -0800
Alexei Starovoitov a...@plumgrid.com wrote:
I'm not sure what would be the meaning of hash map that has all
elements pre-allocated...
As I'm reading your cover letter, I agree, we
30 matches
Mail list logo
