[PATCH 1/2] powerpc/perf: Fix the threshold compare group constraint for power10

Athira Rajeev atrajeev at linux.vnet.ibm.com
Mon May 9 15:52:45 AEST 2022



> On 06-May-2022, at 11:40 AM, Kajol Jain <kjain at linux.ibm.com> wrote:
> 
> Thresh compare bits for a event is used to program thresh compare
> field in Monitor Mode Control Register A (MMCRA: 8-18 bits for power10).
> When scheduling events as a group, all events in that group should
> match value in threshold bits. Otherwise event open for the sibling
> events should fail. But in the current code, incase thresh compare bits are
> not valid, we are not failing in group_constraint function which can result
> in invalid group schduling.
> 
> Fix the issue by returning -1 incase event is threshold and threshold
> compare value is not valid in group_constraint function.
> 
> Patch also fixes the p10_thresh_cmp_val function to return -1,
> incase threshold bits are not valid and changes corresponding check in
> is_thresh_cmp_valid function to return false only when the thresh_cmp
> value is less then 0.
> 
> Thresh control bits in the event code is used to program thresh_ctl
> field in Monitor Mode Control Register A (MMCRA: 48-55). In below example,
> the scheduling of group events PM_MRK_INST_CMPL (3534401e0) and
> PM_THRESH_MET (34340101ec) is expected to fail as both event
> request different thresh control bits.
> 
> Result before the patch changes:
> 
> [command]# perf stat -e "{r35340401e0,r34340101ec}" sleep 1
> 
> Performance counter stats for 'sleep 1':
> 
>             8,482      r35340401e0
>                 0      r34340101ec
> 
>       1.001474838 seconds time elapsed
> 
>       0.001145000 seconds user
>       0.000000000 seconds sys
> 
> Result after the patch changes:
> 
> [command]# perf stat -e "{r35340401e0,r34340101ec}" sleep 1
> 
> Performance counter stats for 'sleep 1':
> 
>     <not counted>      r35340401e0
>   <not supported>      r34340101ec
> 
>       1.001499607 seconds time elapsed
> 
>       0.000204000 seconds user
>       0.000760000 seconds sys
> 
> Fixes: 82d2c16b350f7 ("powerpc/perf: Adds support for programming
> of Thresholding in P10")
> Signed-off-by: Kajol Jain <kjain at linux.ibm.com>

Reviewed-by: Athira Rajeev <atrajeev at linux.vnet.ibm.com>

Thanks
Athira
> ---
> arch/powerpc/perf/isa207-common.c | 9 +++++----
> 1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/powerpc/perf/isa207-common.c b/arch/powerpc/perf/isa207-common.c
> index a74d382ecbb7..013b06af6fe6 100644
> --- a/arch/powerpc/perf/isa207-common.c
> +++ b/arch/powerpc/perf/isa207-common.c
> @@ -108,7 +108,7 @@ static void mmcra_sdar_mode(u64 event, unsigned long *mmcra)
> 		*mmcra |= MMCRA_SDAR_MODE_TLB;
> }
> 
> -static u64 p10_thresh_cmp_val(u64 value)
> +static int p10_thresh_cmp_val(u64 value)
> {
> 	int exp = 0;
> 	u64 result = value;
> @@ -139,7 +139,7 @@ static u64 p10_thresh_cmp_val(u64 value)
> 		 * exponent is also zero.
> 		 */
> 		if (!(value & 0xC0) && exp)
> -			result = 0;
> +			result = -1;
> 		else
> 			result = (exp << 8) | value;
> 	}
> @@ -187,7 +187,7 @@ static bool is_thresh_cmp_valid(u64 event)
> 	unsigned int cmp, exp;
> 
> 	if (cpu_has_feature(CPU_FTR_ARCH_31))
> -		return p10_thresh_cmp_val(event) != 0;
> +		return p10_thresh_cmp_val(event) >= 0;
> 
> 	/*
> 	 * Check the mantissa upper two bits are not zero, unless the
> @@ -502,7 +502,8 @@ int isa207_get_constraint(u64 event, unsigned long *maskp, unsigned long *valp,
> 			value |= CNST_THRESH_CTL_SEL_VAL(event >> EVENT_THRESH_SHIFT);
> 			mask  |= p10_CNST_THRESH_CMP_MASK;
> 			value |= p10_CNST_THRESH_CMP_VAL(p10_thresh_cmp_val(event_config1));
> -		}
> +		} else if (event_is_threshold(event))
> +			return -1;
> 	} else if (cpu_has_feature(CPU_FTR_ARCH_300))  {
> 		if (event_is_threshold(event) && is_thresh_cmp_valid(event)) {
> 			mask  |= CNST_THRESH_MASK;
> -- 
> 2.31.1
> 



More information about the Linuxppc-dev mailing list