From nobody Mon Jan 15 17:48:52 2024 X-Original-To: freebsd-stable@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4TDKPy704Kz57Byy for ; Mon, 15 Jan 2024 17:48:54 +0000 (UTC) (envelope-from scf@FreeBSD.org) Received: from smtp.freebsd.org (smtp.freebsd.org [IPv6:2610:1c1:1:606c::24b:4]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (4096 bits) client-digest SHA256) (Client CN "smtp.freebsd.org", Issuer "R3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4TDKPy6Sftz4CB7 for ; Mon, 15 Jan 2024 17:48:54 +0000 (UTC) (envelope-from scf@FreeBSD.org) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=freebsd.org; s=dkim; t=1705340934; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type; bh=hqAf3xCkhhXfdMFkfMebhxdxTTVAWt7tNEiqdW5lAow=; b=FEZwrKX5cEJ6MGlkVZytus/QGj1/yaOD/4QM146yRV980VsYQX84RYMwhTtQ6tfuK1uY+a yn+5guzlw5v9Uxs6Xk/qTeAZCq1iMnC4C0cAMyxV4Vh/4UpHZi55gigINsmx8NFImaMhCY nxPZJ49hWPXSu19hPWOgafkR8EtXGyqie3zNOVf6SeZ/hSoei9uuCtW9PSr8Bi802Kf5Km AKjGczncOi8YdMjuUzCPnA+RPIhbfRhPI2ev55HFa+mymVhEwgmfHT/8cXA6hHBj0CIRlP HLUUJkswVLIjG2Vn+gLHZjrPcN5fzDDs0KXG7zCEICHtVSop5fBIM0FkD2Hr5A== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=freebsd.org; s=dkim; t=1705340934; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type; bh=hqAf3xCkhhXfdMFkfMebhxdxTTVAWt7tNEiqdW5lAow=; b=O22QOTErYTXX8Dnt+1C/AajZO0Cw9PHG4VolKC1/vZFnPmTLg+HROBjOl1vKTdyLjroTwW SR+VHlfPYzBO2g6EmTAyZRKWqthqHhWqftrJ3NTmUj0PKhYtY0BeDayiK9ixEmqbtEq7C8 HT4fMk9NKUUi/6/YzR5Sa2PZY1+0kZRETgOAkQRse85Bnfl33+V93pDWN8k3D5Qu/OJqJG VOG8dqIdbLnQVQbq//4PURguJvmGTL9hlOLrHOUp/gx7IBV6XB7sIgnyRR67SuO11jNbIs sWWZ/1UgptI3as7DeoKj2gqjfhZmPe1tEcf46o/t/+01lumXsaNobvUvF9MJ8g== ARC-Authentication-Results: i=1; mx1.freebsd.org; none ARC-Seal: i=1; s=dkim; d=freebsd.org; t=1705340934; a=rsa-sha256; cv=none; b=RRGT9e50rMSjcFbOKbrJJFT1sLyaH+ky78B8Yqb0oyRJ4f6p/oMOKjpEPBsrPG8T6zXfua /6E4ZOoGbjOAuV2uvogRYYBxLnKtbvL9HCBf6lpScdAZxFfqKDo0ztu0jIjHp5XvKVpDOg zT0M7WcBHIjG5aFOf9Qw7eiW71MR9wpIO8sh1jKOOzxfRSELCczqPlo2O5FJ98Sc8zU3ky aMI4N+WzAqPTq2HaUaLG326enoqneZTMApSSvtTWsmzTZzkICmxJNVen+0PQJOCTHXLlUL R2YqiGt78MIeEXlGcgVOBV1+mztx7Voym5XkMUI4gHRy+CzkIdkjKoFyN/niTA== Received: from thor.farley.org (1609341-v107.1360-static.crmlinaa.metronetinc.net [104.254.222.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (Client did not present a certificate) (Authenticated sender: scf/mail) by smtp.freebsd.org (Postfix) with ESMTPSA id 4TDKPy4Xm8zLdY for ; Mon, 15 Jan 2024 17:48:54 +0000 (UTC) (envelope-from scf@FreeBSD.org) Date: Mon, 15 Jan 2024 12:48:52 -0500 (EST) From: "Sean C. Farley" To: freebsd-stable@FreeBSD.org Subject: hwpstate_intel epp performance oddities Message-ID: <370810c9-ce96-d597-3440-a85c1e4223f0@FreeBSD.org> List-Id: Production branch of FreeBSD source code List-Archive: https://lists.freebsd.org/archives/freebsd-stable List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-stable@freebsd.org X-BeenThere: freebsd-stable@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset=US-ASCII I recently assembled a new system (FreeBSD 14-STABLE) using an i7-14700K and am odd results when setting epp levels in the hwpstate driver. From the man page, I expected setting dev.hwpstate_intel.0.epp to change the performance level in some manner where 0 is highest and 100 is lowest with some in-between values, however, it does not quite obey that. Upon boot, all dev.hwpstate_intel.N.epp (28 cores) default to 50. Using sysutils/powermon and hearing the volume of the fans, running "stress-ng --matrix 0 --timeout 1m --metrics" shows a moderate run. When changing dev.hwpstate_intel.0.epp to any valid value, the system runs at full performance. Setting it to 0 actually increases the power draw from approximate 14W to close to 40W when the system is idle, so I avoid that value. This appears to be the only difference between all values from 0 to 100 after setting it once. However, there is another variable that can affect the system: dev.hwpstate_intel.16.epp. I have tested all the others with zero difference. Setting dev.hwpstate_intel.16.epp to any value lowers the performance back to moderate performance level regardless of how dev.hwpstate_intel.0.epp is set. All these results are with machdep.hwpstate_pkg_ctrl at its default of 1. Changing that did not seem to help. Toggling BIOS options to prevent the motherboard from "optimizing" the CPUs also did not help. Can anyone replicate or explain this? Sean -- scf@FreeBSD.org