From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id C0CF1E77199 for ; Tue, 7 Jan 2025 15:30:38 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:In-Reply-To:Content-Type: MIME-Version:References:Message-ID:Subject:Cc:To:From:Date:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=N/6XPJ85Rq8WiLNIrO3Lx2eRBJVsdpBUiKmI2/RgmY4=; b=f0RucrTnNxQPgSJCE/wDhdzO27 cYzAI3ohJb8Px3fIB5ZthkFG90g6kpEZ4zk8m0fHgriKTw95LRIh4k9B3hnQXKrS5MzcwvUwquXHX OaJ2ckvzgUwQjetEH342XCZVI0Rb4RuE7uIuLeXeHHyeTQivXu3x2m5Gk8Yu3y2RFJEicyLoU+Zfs ShK5AfgbNkBMrMznBRjpqQifjpdlwuePKhJGYxj+DL+97hLsHkhpvNSoEMWtWmcc0lSjf+w/Q/Ium DJcSBrgjL6V0cwGbHVMhj0jsQF81aER/6fg8oUjtZ9Utm/KZmC2hCVYJO+xD7E7uuO4vkTzEKCw0p rHq+ZDdg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1tVBXD-00000005RDZ-1Pew; Tue, 07 Jan 2025 15:30:23 +0000 Received: from pandora.armlinux.org.uk ([2001:4d48:ad52:32c8:5054:ff:fe00:142]) by bombadil.infradead.org with esmtps (Exim 4.98 #2 (Red Hat Linux)) id 1tVBVE-00000005QME-1Vgr for linux-arm-kernel@lists.infradead.org; Tue, 07 Jan 2025 15:28:24 +0000 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=armlinux.org.uk; s=pandora-2019; h=Sender:In-Reply-To:Content-Type: MIME-Version:References:Message-ID:Subject:Cc:To:From:Date:Reply-To: Content-Transfer-Encoding:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Id: List-Help:List-Unsubscribe:List-Subscribe:List-Post:List-Owner:List-Archive; bh=N/6XPJ85Rq8WiLNIrO3Lx2eRBJVsdpBUiKmI2/RgmY4=; b=jo2HXdS6rkT3QNwp/AyDHIzAPL akHlZDBZ1qktngD1Uxnj16POZ34pApA//rO/1ank552xVpXGqmZwy6oneJNcR/doSCoJdBm5laHJc rMxxuCWyvylF68mokRcNbgfOWSFH67fpyGZZTc41RcYF3B9JjxQkX3sN4ZnI48qaMGfv/wWPJsCK/ EqYOohemm6k2EU3qfSA/R6ukZJTVvzrkBq+YU/gPIWXc1A4I0+L5DQ/qgMFTjtDkk37ySGQnwNdjy gywclq5XLFX7WwGxO5dsCQlDN5KwPyBIj6BPyZTeosUGcG8v1jNRsWAZH80CMy+h4iX1D1dBSon0l Z7iSK0Xg==; Received: from shell.armlinux.org.uk ([fd8f:7570:feb6:1:5054:ff:fe00:4ec]:58202) by pandora.armlinux.org.uk with esmtpsa (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1tVBT7-0007gS-0d; Tue, 07 Jan 2025 15:26:09 +0000 Received: from linux by shell.armlinux.org.uk with local (Exim 4.96) (envelope-from ) id 1tVBT4-0005Ob-0f; Tue, 07 Jan 2025 15:26:06 +0000 Date: Tue, 7 Jan 2025 15:26:06 +0000 From: "Russell King (Oracle)" To: Simon Horman Cc: Andrew Lunn , Heiner Kallweit , Alexandre Torgue , Andrew Lunn , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Jose Abreu , linux-arm-kernel@lists.infradead.org, linux-stm32@st-md-mailman.stormreply.com, Maxime Coquelin , netdev@vger.kernel.org, Paolo Abeni Subject: Re: [PATCH net-next v2 03/17] net: stmmac: use correct type for tx_lpi_timer Message-ID: References: <20250107112851.GE33144@kernel.org> <20250107144103.GB5872@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20250107144103.GB5872@kernel.org> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20250107_072820_399343_DF7254AD X-CRM114-Status: GOOD ( 37.70 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org On Tue, Jan 07, 2025 at 02:41:03PM +0000, Simon Horman wrote: > On Tue, Jan 07, 2025 at 11:57:12AM +0000, Russell King (Oracle) wrote: > > On Tue, Jan 07, 2025 at 11:28:51AM +0000, Simon Horman wrote: > > > On Mon, Jan 06, 2025 at 12:24:58PM +0000, Russell King (Oracle) wrote: > > > > The ethtool interface uses u32 for tx_lpi_timer, and so does phylib. > > > > Use u32 to store this internally within stmmac rather than "int" > > > > which could misinterpret large values. > > > > > > > > Since eee_timer is used to initialise priv->tx_lpi_timer, this also > > > > should be unsigned to avoid a negative number being interpreted as a > > > > very large positive number. > > > > > > > > Also correct "value" in dwmac4_set_eee_lpi_entry_timer() to use u32 > > > > rather than int, which is derived from tx_lpi_timer, even though > > > > masking with STMMAC_ET_MAX will truncate the sign bits. u32 is the > > > > value argument type for writel(). > > > > > > > > Signed-off-by: Russell King (Oracle) > > > > > > ... > > > > > > > diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c > > > > index 9a9169ca7cd2..b0ef439b715b 100644 > > > > --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c > > > > +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c > > > > @@ -111,8 +111,8 @@ static const u32 default_msg_level = (NETIF_MSG_DRV | NETIF_MSG_PROBE | > > > > NETIF_MSG_IFDOWN | NETIF_MSG_TIMER); > > > > > > > > #define STMMAC_DEFAULT_LPI_TIMER 1000 > > > > -static int eee_timer = STMMAC_DEFAULT_LPI_TIMER; > > > > -module_param(eee_timer, int, 0644); > > > > +static unsigned int eee_timer = STMMAC_DEFAULT_LPI_TIMER; > > > > +module_param(eee_timer, uint, 0644); > > > > MODULE_PARM_DESC(eee_timer, "LPI tx expiration time in msec"); > > > > #define STMMAC_LPI_T(x) (jiffies + usecs_to_jiffies(x)) > > > > > > > > > > Hi Russell, > > > > > > now that eee_timer is unsigned the following check in stmmac_verify_args() > > > can never be true. I guess it should be removed. > > > > > > if (eee_timer < 0) > > > eee_timer = STMMAC_DEFAULT_LPI_TIMER; > > > > > > > Thanks for finding that. The parameter description doesn't seem to > > detail whether this is intentional behaviour or not, or whether it is > > "because someone used int and we shouldn't have negative values here". > > > > I can't see why someone would pass a negative number for eee_timer > > given that it already defaults to STMMAC_DEFAULT_LPI_TIMER. > > > > So, I'm tempted to remove this. > > I'm not sure either. It did cross my mind that the check could be changed, > to disallow illegal values (if the range of legal values is not all > possible unsigned integer values). But it was just an idea without any > inspection of the code or thought about it's practicality. And my first > instinct was the same as yours: remove the check. My reasoning is as follows: In the existing code with the module paramter is a signed int, then it take a value up to INT_MAX. Provided sizeof(int) == sizeof(u32), then this can be reported through the ethtool API. However, ethtool can set the timer to U32_MAX which can exceed INT_MAX in this case. The driver doesn't stop this, and uses a software based timer for any delay greater than the capabilities of the hardware timer (if any.) So, through ethtool one can set the LPI delay to anything between 0 and U32_MAX, whereas through the module parameter it's between 0 and INT_MAX. values between INT_MIN and -1 inclusive result in the default being used. It is, of course, absurd to have a negative delay, or even a delay anywhere near INT_MAX or U32_MAX. I'll separate out the change to eee_timer so if necessary, that can be reverted without reverting the entire patch. Oh goodo, an extra patch for a patchset which already exceeds netdev's 15 patches... -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!