From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: passt.top; dmarc=pass (p=quarantine dis=none) header.from=maxchernoff.ca Authentication-Results: passt.top; dkim=pass (2048-bit key; secure) header.d=maxchernoff.ca header.i=@maxchernoff.ca header.a=rsa-sha256 header.s=key1 header.b=rqdB+ALD; dkim-atps=neutral Received: from out-189.mta0.migadu.com (out-189.mta0.migadu.com [91.218.175.189]) by passt.top (Postfix) with ESMTPS id 814D85A0271 for ; Mon, 08 Dec 2025 09:17:19 +0100 (CET) Message-ID: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=maxchernoff.ca; s=key1; t=1765181838; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=FEALokkCMlcmSyNyr314VyUaRGRb3YaENTwgbMkLO94=; b=rqdB+ALDF4n5EmGSQE50JcBdxeNbdg3oNUDgzBvowhV4qBaZqx3phJ2xF9R/yRM2f688cy urrnqDZiLmJaYvDa/QeQjb6enUUlNPp9ykOlzCiDe/jJZYnfoCwmcutZ7RU5WIyj9lIJDI ZPGef5hiMa2/S84OhD+Q6aUoDQwcv6BBEYKVJA+hKc+mxQKdg9OGTvKW+8uW3Dg0Am61O1 WThjxjcbxGIA5ubZE4iE6znWkp0Rmj7j6ebfxy93NR5N4KTJylQGImZisVQqCNZ+xHH6VF 8ZoseNruOViv7JI/EIkPrtx/tyB8LIjUVEEWij92HEn5AiS7e25QTZjErBwJ1A== Subject: Re: [PATCH v3 00/10] tcp: Fix throughput issues with non-local peers X-Report-Abuse: Please report any abuse attempt to abuse@migadu.com and include these headers. From: Max Chernoff To: Stefano Brivio , passt-dev@passt.top In-Reply-To: <20251208072024.3884137-1-sbrivio@redhat.com> References: <20251208072024.3884137-1-sbrivio@redhat.com> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Date: Mon, 08 Dec 2025 01:11:56 -0700 MIME-Version: 1.0 X-Migadu-Flow: FLOW_OUT Message-ID-Hash: JI3YPWGXHCGU2PHIYYJGPM347ZUU37A5 X-Message-ID-Hash: JI3YPWGXHCGU2PHIYYJGPM347ZUU37A5 X-MailFrom: git@maxchernoff.ca X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency; loop; banned-address; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header CC: David Gibson X-Mailman-Version: 3.3.8 Precedence: list List-Id: Development discussion and patches for passt Archived-At: Archived-At: List-Archive: List-Archive: List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: Hi Stefano, On Mon, 2025-12-08 at 08:20 +0100, Stefano Brivio wrote: > With this series, I'm getting the expected (wirespeed) throughput for > transfers between peers with varying non-local RTTs: I checked > different guests bridged on the same machine (~500 us) and hosts with > increasing distance using iperf3, as well as HTTP transfers only for > some hosts I have control over (500 us and 5 ms case). > > With increasing RTTs, I can finally see the throughput converging to > the available bandwidth reasonably fast: Thanks for the patch, but this unfortunately seems to make things worse in my testing (with curl/https). Using my benchmarking script from the earlier thread, with a 10MB file size and a 30s timeout: $ pasta --version # Using the stock pasta in F43 pasta 0^20250919.g623dbf6-1.fc43.x86_64 $ ./pasta-upload-test.sh network ping_time wmem_max rmem_max tcp_notsent_low= at tcp_congestion_control default_qdisc download_time uploa= d_time host 50ms custom custom custom custom custom 1.508751 = 1.656876 pasta 50ms custom custom custom custom custom 1.548367 = 2.184099 host 170ms custom custom custom custom custom 9.313611 = 3.055348 pasta 170ms custom custom custom custom custom 13.300405 = 25.046154 $ sudo dnf install $ pasta --version pasta 0^20251208.g5943ea4-1.fc43.x86_64 $ ./pasta-upload-test.sh network ping_time wmem_max rmem_max tcp_notsent_low= at tcp_congestion_control default_qdisc download_time uploa= d_time host 50ms custom custom custom custom custom 1.490700 = 1.666525 pasta 50ms custom custom custom custom custom 1.474725 = 30.000000 host 170ms custom custom custom custom custom 9.618929 = 3.221314 pasta 170ms custom custom custom custom custom 10.475894 = 30.000000 $ sudo dnf downgrade pasta $ pasta --version # Back to the stock pasta in F43 pasta 0^20250919.g623dbf6-1.fc43.x86_64 $ ./pasta-upload-test.sh network ping_time wmem_max rmem_max tcp_notsent_low= at tcp_congestion_control default_qdisc download_time uploa= d_time host 50ms custom custom custom custom custom 1.407653 = 1.686541 pasta 50ms custom custom custom custom custom 1.558330 = 2.481097 host 170ms custom custom custom custom custom 8.951508 = 3.191743 pasta 170ms custom custom custom custom custom 9.891349 = 30.000000 $ sudo dnf install $ pasta --version # Try the patched version again in case the last tes= t was an anomaly pasta 0^20251208.g5943ea4-1.fc43.x86_64 $ ./pasta-upload-test.sh network ping_time wmem_max rmem_max tcp_notsent_low= at tcp_congestion_control default_qdisc download_time uploa= d_time host 50ms custom custom custom custom custom 1.450695 = 1.689421 pasta 50ms custom custom custom custom custom 1.605941 = 30.000000 host 170ms custom custom custom custom custom 5.610433 = 3.034058 pasta 170ms custom custom custom custom custom 5.544638 = 30.000000 Thanks, -- Max