On Wed, Dec 15, 2010 at 08:35:35AM -0500, Matthew Mondor wrote:
> On Tue, 14 Dec 2010 20:49:14 -0800
> Matt Thomas wrote:
>
> > I have a fairly large but mostly simple patch which changes the stats
> > collected in
> > uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits
On Thu, Dec 16, 2010 at 7:07 AM, David Holland wrote:
> On Tue, Dec 14, 2010 at 08:49:14PM -0800, Matt Thomas wrote:
> > I have a fairly large but mostly simple patch which changes the
> > stats collected in uvmexp for faults, intrs, softs, syscalls, and
> > traps from 32 bit to 64 bits and put
On Tue, Dec 14, 2010 at 08:49:14PM -0800, Matt Thomas wrote:
> I have a fairly large but mostly simple patch which changes the
> stats collected in uvmexp for faults, intrs, softs, syscalls, and
> traps from 32 bit to 64 bits and puts them in cpu_data (in
> cpu_info). This makes more accurate
On Dec 15, 2010, at 5:35 AM, Matthew Mondor wrote:
> On Tue, 14 Dec 2010 20:49:14 -0800
> Matt Thomas wrote:
>
>> I have a fairly large but mostly simple patch which changes the stats
>> collected in
>> uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits
>> and
>> puts
On Dec 15, 2010, at 11:16 AM, Jean-Yves Migeon wrote:
> On 15.12.2010 19:43, Matt Thomas wrote:
>>
>> On Dec 15, 2010, at 1:53 AM, Jean-Yves Migeon wrote:
>>
> On Wed, Dec 15, 2010 at 1:49 PM, Matt Thomas
> wrote:
>> The diffs are at http://www.netbsd.org/uvmexp-diff.txt
>>>
>>>
On 15.12.2010 19:43, Matt Thomas wrote:
>
> On Dec 15, 2010, at 1:53 AM, Jean-Yves Migeon wrote:
>
On Wed, Dec 15, 2010 at 1:49 PM, Matt Thomas wrote:
> The diffs are at http://www.netbsd.org/uvmexp-diff.txt
>>
>> Purely cosmetic comments:
>> - why are most of the cpu_nsoft count commen
Eduardo Horvath wrote:
> On Wed, 15 Dec 2010, Martin Husemann wrote:
>
> > I have one stupid question: why can't we leave the size of the counters
> > at 32bit on a per arch basis?
>
> Or use 64-bit counters but just update the lower 32-bits of them. Is
> there some danger that a 32-bit counte
On Dec 15, 2010, at 1:53 AM, Jean-Yves Migeon wrote:
>>> On Wed, Dec 15, 2010 at 1:49 PM, Matt Thomas wrote:
The diffs are at http://www.netbsd.org/uvmexp-diff.txt
>
> Purely cosmetic comments:
> - why are most of the cpu_nsoft count commented out?
Because we were counting ASTs as soft in
On Wed, 15 Dec 2010, Martin Husemann wrote:
> I have one stupid question: why can't we leave the size of the counters
> at 32bit on a per arch basis?
Or use 64-bit counters but just update the lower 32-bits of them. Is
there some danger that a 32-bit counter will overflow?
> At a quick glance
I have one stupid question: why can't we leave the size of the counters
at 32bit on a per arch basis?
At a quick glance the sparc code looked v9 only, so will need some work.
Martin
On Tue, Dec 14, 2010 at 08:49:14PM -0800, Matt Thomas wrote:
> I have a fairly large but mostly simple patch which changes the stats
> collected in
> uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits
> and
> puts them in cpu_data (in cpu_info). This makes more accurate
On Tue Dec 14 2010 at 20:49:14 -0800, Matt Thomas wrote:
> I have a fairly large but mostly simple patch which changes the stats
> collected in
> uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits
> and
> puts them in cpu_data (in cpu_info). This makes more accurate and
On Tue, 14 Dec 2010 20:49:14 -0800
Matt Thomas wrote:
> I have a fairly large but mostly simple patch which changes the stats
> collected in
> uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits
> and
> puts them in cpu_data (in cpu_info). This makes more accurate and a
I have a fairly large but mostly simple patch which changes the stats
collected in uvmexp for faults, intrs, softs, syscalls, and traps from
32 bit to 64 bits and puts them in cpu_data (in cpu_info). This makes
more accurate and a little cheaper to update on 64bit systems.
I hvaen't look
On Wed, Dec 15, 2010 at 1:49 PM, Matt Thomas
wrote:
The diffs are at http://www.netbsd.org/uvmexp-diff.txt
Purely cosmetic comments:
- why are most of the cpu_nsoft count commented out?
- in uvmexp_print(), just use PRIu64 instead of %llu (as you did in
other places)
--
Jean-Yves Migeon
jea
On Dec 14, 2010, at 10:11 PM, Masao Uebayashi wrote:
> On Wed, Dec 15, 2010 at 1:49 PM, Matt Thomas wrote:
>>
>> The diffs are at http://www.netbsd.org/uvmexp-diff.txt
>
> 404
Opps.
www.netbsd.org/~matt/uvmexp-diff.txt
I have a fairly large but mostly simple patch which changes the stats collected
in
uvmexp for faults, intrs, softs, syscalls, and traps from 32 bit to 64 bits and
puts them in cpu_data (in cpu_info). This makes more accurate and a little
cheaper
to update on 64bit systems.
I've had to modify so
17 matches
Mail list logo