lkml.org 
[lkml]   [2016]   [Feb]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: [PATCH V4 4/6] perf/amd/iommu: Introduce get_iommu_bnk_cnt_evt_idx
    On Thu, Feb 11, 2016 at 04:15:25PM +0700, Suravee Suthikulpanit wrote:
    > Introduce a helper function to calculate bit-index for assigning
    > performance counter assignment.
    >
    > Signed-off-by: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>
    > ---
    > arch/x86/events/amd/iommu.c | 19 ++++++++++++++-----
    > 1 file changed, 14 insertions(+), 5 deletions(-)
    >
    > diff --git a/arch/x86/events/amd/iommu.c b/arch/x86/events/amd/iommu.c
    > index debf22d..812eff2 100644
    > --- a/arch/x86/events/amd/iommu.c
    > +++ b/arch/x86/events/amd/iommu.c
    > @@ -145,18 +145,27 @@ static struct attribute_group amd_iommu_cpumask_group = {
    >
    > /*---------------------------------------------*/
    >
    > +static inline
    > +int get_iommu_bnk_cnt_evt_idx(struct perf_amd_iommu *perf_iommu,
    > + int iommu_index, int bank_index,
    > + int cntr_index)

    Align args at the opening brace.

    > +{
    > + int cntrs_per_iommu = perf_iommu->max_banks * perf_iommu->max_counters;
    > + int index = (perf_iommu->max_counters * bank_index) + cntr_index;
    > +
    > + return (cntrs_per_iommu * iommu_index) + index;
    > +}
    > +
    > static int get_next_avail_iommu_bnk_cntr(struct perf_amd_iommu *perf_iommu)
    > {
    > unsigned long flags;
    > int shift, bank, cntr, retval;
    > - int max_banks = perf_iommu->max_banks;
    > - int max_cntrs = perf_iommu->max_counters;
    >
    > raw_spin_lock_irqsave(&perf_iommu->lock, flags);
    >
    > - for (bank = 0, shift = 0; bank < max_banks; bank++) {
    > - for (cntr = 0; cntr < max_cntrs; cntr++) {
    > - shift = bank + (bank*3) + cntr;
    > + for (bank = 0, shift = 0; bank < perf_iommu->max_banks; bank++) {

    Please init that shift variable above in the function entry - it
    confuses unnecessarily here as if it had anything to do with the loop
    stride.

    > + for (cntr = 0; cntr < perf_iommu->max_counters; cntr++) {
    > + shift = get_iommu_bnk_cnt_evt_idx(perf_iommu, 0, bank, cntr);

    \n here

    > if (perf_iommu->cntr_assign_mask & (1ULL<<shift)) {
    > continue;
    > } else {
    > --
    > 2.5.0
    >
    >

    --
    Regards/Gruss,
    Boris.

    ECO tip #101: Trim your mails when you reply.

    \
     
     \ /
      Last update: 2016-02-18 13:01    [W:4.662 / U:0.196 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site