On Fri, 2015-09-25 at 17:53 +0800, He Chen wrote:
> > Quoting the relevant bits of code for clarity:
> > libxl_psr_cbm_type type = LIBXL_PSR_CBM_TYPE_L3_CBM;
> > ...
> > case 'd':
> > type = LIBXL_PSR_CBM_TYPE_L3_DATA;
> > opt_data = 1;
> > break;
> > case '
> Quoting the relevant bits of code for clarity:
> libxl_psr_cbm_type type = LIBXL_PSR_CBM_TYPE_L3_CBM;
> ...
> case 'd':
> type = LIBXL_PSR_CBM_TYPE_L3_DATA;
> opt_data = 1;
> break;
> case 'c':
> type = LIBXL_PSR_CBM_TYPE_L3_CODE;
> opt_cod
On Fri, 2015-09-25 at 17:04 +0800, He Chen wrote:
> On Thu, Sep 24, 2015 at 12:22:47PM +0100, Ian Campbell wrote:
> > On Thu, 2015-09-24 at 12:07 +0100, Ian Campbell wrote:
> > > @@ -8517,8 +8535,19 @@ int main_psr_cat_cbm_set(int argc, char
> > > **argv)
> > > > libxl_string_list_dispose(
On Fri, 2015-09-25 at 16:43 +0800, He Chen wrote:
> On Thu, Sep 24, 2015 at 12:07:27PM +0100, Ian Campbell wrote:
> > On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
> > > @@ -8410,20 +8415,29 @@ static void
> > > psr_cat_print_one_domain_cbm(uint32_t
> > > domid, uint32_t socketid)
> > > pr
On Thu, Sep 24, 2015 at 12:22:47PM +0100, Ian Campbell wrote:
> On Thu, 2015-09-24 at 12:07 +0100, Ian Campbell wrote:
> > @@ -8517,8 +8535,19 @@ int main_psr_cat_cbm_set(int argc, char **argv)
> > > libxl_string_list_dispose(&socket_list);
> > > free(value);
> > > break;
On Thu, Sep 24, 2015 at 12:07:27PM +0100, Ian Campbell wrote:
> On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
> > @@ -8410,20 +8415,29 @@ static void psr_cat_print_one_domain_cbm(uint32_t
> > domid, uint32_t socketid)
> > printf("%5d%25s", domid, domain_name);
> > free(domain_name);
>
On Thu, 2015-09-24 at 06:20 -0600, Jan Beulich wrote:
> > > > On 24.09.15 at 14:07, wrote:
> > On Thu, 2015-09-24 at 05:50 -0600, Jan Beulich wrote:
> > > a specific error type
> > > to me seems to go too far. Surely out of the several dozen
> > > possibilities a handful of not-so-common ones can
>>> On 24.09.15 at 14:07, wrote:
> On Thu, 2015-09-24 at 05:50 -0600, Jan Beulich wrote:
>> a specific error type
>> to me seems to go too far. Surely out of the several dozen
>> possibilities a handful of not-so-common ones can be picked?
>
> I was thinking in particular EINVAL was not in the n
On Thu, 2015-09-24 at 05:50 -0600, Jan Beulich wrote:
> > > > On 24.09.15 at 13:00, wrote:
> > On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
> > > diff --git a/tools/libxl/libxl_psr.c b/tools/libxl/libxl_psr.c
> > > index 3378239..62963cf 100644
> > > --- a/tools/libxl/libxl_psr.c
> > > +++ b/
>>> On 24.09.15 at 13:00, wrote:
> On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
>> diff --git a/tools/libxl/libxl_psr.c b/tools/libxl/libxl_psr.c
>> index 3378239..62963cf 100644
>> --- a/tools/libxl/libxl_psr.c
>> +++ b/tools/libxl/libxl_psr.c
>> @@ -87,6 +87,9 @@ static void libxl__psr_cat_
On Thu, 2015-09-24 at 12:07 +0100, Ian Campbell wrote:
> @@ -8517,8 +8535,19 @@ int main_psr_cat_cbm_set(int argc, char **argv)
> > libxl_string_list_dispose(&socket_list);
> > free(value);
> > break;
> > +case 'd':
> > +type = LIBXL_PSR_CBM_TYPE_L3_DATA;
> >
On 24/09/15 11:57, Ian Campbell wrote:
On Thu, 2015-09-17 at 11:38 +0100, Andrew Cooper wrote:
@@ -320,6 +333,8 @@ int xc_psr_cat_get_l3_info(xc_interface *xch,
uint32_t socket,
{
*cos_max = sysctl.u.psr_cat_op.u.l3_info.cos_max;
*cbm_len = sysctl.u.psr_cat_op.u.l3_inf
On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
> @@ -8410,20 +8415,29 @@ static void psr_cat_print_one_domain_cbm(uint32_t
> domid, uint32_t socketid)
> printf("%5d%25s", domid, domain_name);
> free(domain_name);
>
> -if (!libxl_psr_cat_get_cbm(ctx, domid, LIBXL_PSR_CBM_TYPE_L3_C
On Thu, 2015-09-17 at 17:35 +0800, He Chen wrote:
> diff --git a/tools/libxl/libxl_psr.c b/tools/libxl/libxl_psr.c
> index 3378239..62963cf 100644
> --- a/tools/libxl/libxl_psr.c
> +++ b/tools/libxl/libxl_psr.c
> @@ -87,6 +87,9 @@ static void libxl__psr_cat_log_err_msg(libxl__gc *gc,
> int err)
>
On Thu, 2015-09-17 at 11:38 +0100, Andrew Cooper wrote:
> @@ -320,6 +333,8 @@ int xc_psr_cat_get_l3_info(xc_interface *xch,
> > uint32_t socket,
> > {
> > *cos_max = sysctl.u.psr_cat_op.u.l3_info.cos_max;
> > *cbm_len = sysctl.u.psr_cat_op.u.l3_info.cbm_len;
> > +*cdp
On Thu, 2015-09-17 at 11:38 +0100, Andrew Cooper wrote:
> On 17/09/15 10:35, He Chen wrote:
> > @@ -2798,7 +2798,9 @@ enum xc_psr_cmt_type {
> > typedef enum xc_psr_cmt_type xc_psr_cmt_type;
> >
> > enum xc_psr_cat_type {
> > -XC_PSR_CAT_L3_CBM = 1,
> > +XC_PSR_CAT_L3_CBM = 1,
> > +
On 17/09/15 10:35, He Chen wrote:
> @@ -2798,7 +2798,9 @@ enum xc_psr_cmt_type {
> typedef enum xc_psr_cmt_type xc_psr_cmt_type;
>
> enum xc_psr_cat_type {
> -XC_PSR_CAT_L3_CBM = 1,
> +XC_PSR_CAT_L3_CBM = 1,
> +XC_PSR_CAT_L3_CODE = 2,
> +XC_PSR_CAT_L3_DATA = 3,
> };
No need f
This is the xl/xc changes to support Intel Code/Data Prioritization.
CAT xl commands to set/get CBMs are extended to support CDP.
Signed-off-by: He Chen
---
tools/libxc/include/xenctrl.h | 7 +--
tools/libxc/xc_psr.c | 17 ++-
tools/libxl/libxl.h | 7 +++
18 matches
Mail list logo
