From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794825757; b=dM+IICKunmr2WZiSvV2CIIDyuP2y6PzVsN9xKVMNziXkgg7HgsqX13xKhv4L2TESLgej INfXAHRnDTXUcgicUw38xOyo/ixnh94TXAUKKCAT9Y1ejrW8KLjy44Kr707NKAcOCyXIY 9xlIMRgH+jg7jbqKxopgGbjJVlOaajmotf8/7onXKXAnLhP64TxmQz+Is0MvCSh6cyQFu 6bFmYFZyCxMslUdVbtYqW6RYeg8O4z+1yB/vU13kJ3gJ59rj5gHhYwrP9ZZNmTuNwV6gm ifrWvXAEa7JYaU4RkZmip06AFz3k4/FfmBLDkwbLR+UzBDR3PmzI51yRVLvBicdD0Jg== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794825757; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:Content-Transfer-Encoding; bh=Vvrt3ksep/uByZ7+Whx51+tzPNsOAV7Zt47AvUyFumo=; b=1Ceaw+AJ0k7NI3QuyJobr4pH1mxHjwneYdERwKJxSBxUAKfX92YAbQui3lBAwuBXPqoX +OJRl15C9DsHzDlsx7KyrIdJ3zg3j277O1+OwFMzZ99Lm15+PVw/4n8fnsi40TOX2ckVe buvXEdXfxtO+3zv2zGh6b/LKaQov17bYlEPfqeiDAVtVXCl3UTZGDLH6h8Yu2ES0iXePK hGz0ruyEuHNRs927KP1NZVWafMV78GeNRsBLQQ7hvds9Yoe2XdoTPQZ04wdIJ+yKZwif+ 30w5GVhpUeZJvY0Xsd+I0OUUKtHX+7BOTnTy/1akr5ai8A4Mkm/8UTG5MXMz4DU0+Gw== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=KH4TvZ1y; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8EE5D606A323; Mon, 16 Nov 2026 10:42:37 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=KH4TvZ1y Received: by smtp.kernel.org (Postfix) with ESMTPSA id 1779E1F000C4; Mon, 16 Nov 2026 10:42:34 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794825757; bh=Vvrt3ksep/uByZ7+Whx51+tzPNsOAV7Zt47AvUyFumo=; h=From:To:Cc:Subject:Date:In-Reply-To:References; b=KH4TvZ1y7FqJS9NZhglXOeRoWbic1VZhv9sft0QtxWHoN9oqww9kMSLfO8wRbdgEe eOvyfoZ3tgELJ4YAXT6sdieD52RFUS0BUFVvTBNdlztndeiHAFNz0BEpCZ/thog9xJ eWwuNy0PFzyaAFuKNhuuOwdJWeIUwfX+2MlnVCCZUr4d6c7eZECN3EB6fgditnh+EO 9VDGPfWvHG0JG9QLSozn95r9XnToTGr62Uw1+EzaWAGtVwsoNAaewEyV57WM7FXauT fTuadLl4MDjRwJo9lRzCe7we/Y4JkpMn4HR6vWetdCiNKZQXmFjctx21CtmjkEugpD snjWzHNzLkOlw== From: Imre Kaloz To: netdev-bot+sashiko@kernel.org Cc: Thomas Bogendoerfer , netdev@vger.kernel.org, Andrew Lunn , "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH net-next 1/2] net: sgi: ioc3-eth: apply the DMA attributes only to 64-bit addresses Date: Mon, 16 Nov 2026 11:41:12 +0100 Message-ID: <20261116104113.5946-1-kaloz@kernel.org> X-Mailer: git-send-email 2.47.3 In-Reply-To: <179473759964.2199510.1259617691806814840@kernel.org> References: <20261111095138.1757-2-kaloz@kernel.org> <179473759964.2199510.1259617691806814840@kernel.org> X-Mailing-List: netdev@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: 8bit > Can the code currently in the tree ever reach this branch? No. On IP27 and IP30 every IOC3 DMA address has a nonzero target widget in bits 63:60. The branch is for the IP35 series, where an IOC3 behind an XBridge or PIC uses the 32-bit window. v2 says so in the commit message: https://lore.kernel.org/all/20261116102115.27574-1-kaloz@kernel.org/ > Does the window path rely on platform code setting the D32 bits in > b_device for the IOC3 slot? It does, and nothing sets them today. On a Fuel both XBridges leave PREF, PRECISE, COH and BARRIER clear in every b_device register, including the IOC3 slot (0x180010f8). That belongs to the bridge code in the IP35 series, which will set the attributes for slots that use the window. ioc3-eth itself needs no change for it. > Should the ERBAR value be decided from the RX descriptor ring > address? The IP35 series picks the window per device, not per address: an IOC3 behind an XBridge or PIC gets a 32-bit bus_dma_limit and SWIOTLB bounces anything above it. The ring and every buffer of one IOC3 therefore use the same form. Thanks, Imre