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=1795127429; b=Nd1aQ13cmX2pNgGAd49PbBH8AfhrWaNGgofdBjC9lU5qZkFzP+iusZsnv2hW5aqpacgN BU8meWUeWDOZ54nNWt14LTFt6w4HD24e9VjOOBDgQ4PCz2obZEiYFPzeoGGRAfAl5h7DW NRpFktZTG4AnbuHx84wTOFrVciEMlA/Yc0VLFC+aqO4KNabbhf3aV3LRyJDkB4GTV98P7 N2TM0vjPTn5U5lEJ1zcmpX508e1u+G6DZZayHWkQ6WEyGDP7UZQytX7hlAXZlWgdwozHJ htlNp7oqQ4IeUweROz+LBSN/qP2FL0uX5tWagkl4kkmSESh9+y+x4F5C0C/T3QDUImQ== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1795127429; h=Received:DKIM-Signature:Date:From:To:Cc:Subject:Message-ID: In-Reply-To:References:MIME-Version:Content-Type: Content-Transfer-Encoding; bh=Z1dTghTaKtz+6XlsUqAv2sd97175TAUzyOoX368JJ0c=; b=kwTwFpKMJQT8zIy9WLsbJYD7x4CHRKXs6keEL5/TNsMrNar+wX6vIQZiZz/Z7EP7S54I zV7Pr7NR66O7cwpUjZUZVrmyc56sEFou2iaRDjurczMaSMS5X9OY42NjBGWYiVMEObpn/ DNpoIi/1ByA7Q7ZHAZRqyoY9tHEFJJo7zOtmfAPLD5s4NfejlK4bJ34P7/MwBiPw8R4pv ykUGKwm19yD/1ZMiK9aeAk3jH7UrKEUlYLmSuI1WsJA5zNk7OKzZyVKvbGK4XZ7vXVjvR RJ1q21K/a8Q+c2APLG7xj0zeiUmzSmZNe9/w7amto6JLsCxXXDZLnVlgiw96GFlhL8A== 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=iTYVSRqu; 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 4AC5D606A320 for ; Thu, 19 Nov 2026 22:30:29 +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=iTYVSRqu Received: by smtp.kernel.org (Postfix) with ESMTPSA id A13F11F000C0; Thu, 19 Nov 2026 22:30:28 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1795127428; bh=Z1dTghTaKtz+6XlsUqAv2sd97175TAUzyOoX368JJ0c=; h=Date:From:To:Cc:Subject:In-Reply-To:References; b=iTYVSRqu5EeNmI5Knm9QQnCRzjBZp7JUhdRnBegEMRQFk0cR/YxFCRGMJZCIL2Bok /RVLjzfCvnywHoDruNrWXWP/k9BoRlnOWiKKjHaZvbK8f1h+OoC+QDWml8oD2z144H 7L4v1+HX1FU9COI9DVEDAcKctOR26DRYubmWJHY/0WyXHb334j8U+hTIineuXHC5EU 0xs7egjxj29aqHEnVNpKKTdziKQ6ZgvXhN7uW4n8PGSnVx6wj/9ci4Y5nMAsHgZsJ3 KJ9MT5hx5HItWORH4prqTLr5qaOHhY2wxmYyHBL+r7qq1aJuJgWFgWlOx9ZQE7p4Q9 C3xI5GE76aWkQ== Date: Thu, 19 Nov 2026 14:30:27 -0800 From: Jakub Kicinski To: Michael Chan Cc: davem@davemloft.net, netdev@vger.kernel.org, edumazet@google.com, pabeni@redhat.com, andrew+netdev@lunn.ch, horms@kernel.org, pavan.chebbi@broadcom.com Subject: Re: [PATCH net-next 6/6] eth: bnxt_en: restart Rx queues when their IRQ moves to another NUMA node Message-ID: <20261119143027.6d8ca735@kernel.org> In-Reply-To: References: <20261118023905.3050661-1-kuba@kernel.org> <20261118023905.3050661-7-kuba@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-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit On Wed, 18 Nov 2026 17:24:12 -0800 Michael Chan wrote: > queue restart already has the code to do HWRM_RING_FREE of the CP ring > to program the new TPH tag. We can add the missing logic to > re-allocate the CP ring pages from the proper NUMA. Since we are > paying the heavy price for a ring restart, should we include the CP > ring to make it more complete? I coded this up. AI review thing popped up the problem that re-allocating CP rings causes loss of IRQ moderation settings. This seems net-worthy to me :S Attaching a memory provider on a machine with TPH will lead to loss of coalescing settings. Quite unpleasant for the -irq settings in particular. Can I send out the coal fixes for net and this series without the CP realloc? I'd follow up with the CP realloc after the net->net-next merge which we just missed :(