[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <573aeaa2-f99b-47fb-b9ac-1439f601e7e7@amd.com>
Date: Wed, 30 Apr 2025 23:15:52 +0530
From: Ravi Bangoria <ravi.bangoria@....com>
To: Namhyung Kim <namhyung@...nel.org>
Cc: Ingo Molnar <mingo@...hat.com>, Arnaldo Carvalho de Melo
<acme@...nel.org>, Peter Zijlstra <peterz@...radead.org>,
Joe Mario <jmario@...hat.com>, Stephane Eranian <eranian@...gle.com>,
Jiri Olsa <jolsa@...nel.org>, Ian Rogers <irogers@...gle.com>,
Kan Liang <kan.liang@...ux.intel.com>, linux-kernel@...r.kernel.org,
linux-perf-users@...r.kernel.org, Santosh Shukla <santosh.shukla@....com>,
Ananth Narayan <ananth.narayan@....com>, Sandipan Das
<sandipan.das@....com>, Ravi Bangoria <ravi.bangoria@....com>
Subject: Re: [PATCH v4 1/4] perf amd ibs: Add Load Latency bits in raw dump
Hi Namhyung,
>> + # perf record -e ibs_op/ldlat=128/ -c 100000 -a
>> +
>> + Supported load latency threshold values are 128 to 2048 (both inclusive).
>
> What happens if user gives an out of range value?
Kernel returns error.
>> static void pr_ibs_op_ctl(union ibs_op_ctl reg)
>> {
>> char l3_miss_only[sizeof(" L3MissOnly _")] = "";
>> + char ldlat[sizeof(" LdLatThrsh __ LdLatEn _")] = "";
>
> Shouldn't it reserve 4 characters for the threshold since it can be up
> to 2048?
This function dumps HW register content. IBS_OP_CTL[LdLatThrsh] is a
4 bit field which should be programmed as:
(actual threshold / 128) - 1
Valid values for LdLatThrsh are 0 to 15.
>> if (zen4_ibs_extensions)
>> snprintf(l3_miss_only, sizeof(l3_miss_only), " L3MissOnly %d", reg.l3_miss_only);
>>
>> - printf("ibs_op_ctl:\t%016llx MaxCnt %9d%s En %d Val %d CntCtl %d=%s CurCnt %9d\n",
>> + if (ldlat_cap) {
>> + snprintf(ldlat, sizeof(ldlat), " LdLatThrsh %2d LdLatEn %d",
>
> Here, it would be %4d.
Since the valid values for LdLatThrsh are 0 to 15, two characters are
sufficient.
Thanks,
Ravi
Powered by blists - more mailing lists