From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by gabe.freedesktop.org (Postfix) with ESMTPS id 7E8186E841 for ; Tue, 31 Mar 2020 14:30:44 +0000 (UTC) References: <20200331114746.81898-1-lionel.g.landwerlin@intel.com> <158566466722.5852.13316064706500516988@build.alporthouse.com> From: Lionel Landwerlin Message-ID: Date: Tue, 31 Mar 2020 17:30:41 +0300 MIME-Version: 1.0 In-Reply-To: <158566466722.5852.13316064706500516988@build.alporthouse.com> Content-Language: en-US Subject: Re: [igt-dev] [PATCH i-g-t] tests/perf: add tests for triggered OA reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii"; Format="flowed" Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" To: Chris Wilson , igt-dev@lists.freedesktop.org List-ID: On 31/03/2020 17:24, Chris Wilson wrote: > Quoting Lionel Landwerlin (2020-03-31 12:47:46) >> By whitelisting a couple of registers we can allow an application > By whitelisting, through the i915_perf interface, a couple of... > > I was looking for how you found a way to set RING_NONPRIV. :) Yeah, I'll give a try at whitelisting when OA is enabled. I have another patch where I whitelist 3 ranges of OA perf counter registers (total 72dwords) on Gen12. > >> batch to trigger OA reports in the OA buffer by switching back & forth >> an inverter on the condition logic. > Test looks reasonable, but the specific mechanics of OA I can't comment Yeah, I could enhance by adding stalls and verify the deltas, but I've checked in Mesa already and came up with piglit tests : https://gitlab.freedesktop.org/mesa/piglit/-/merge_requests/248 -Lionel > on. > -Chris _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev