Message ID | 20231121000048.789613-1-kuba@kernel.org (mailing list archive) |
---|---|
Headers | show |
Series | net: page_pool: add netlink-based introspection | expand |
On Mon, 20 Nov 2023 16:00:33 -0800 Jakub Kicinski wrote: > net: page_pool: split the page_pool_params into fast and slow > net: page_pool: avoid touching slow on the fastpath To relieve some of the pain and suffering this series causes to our build tester I'm going to apply the first 2 patches already. I hope that's fine. They are pretty stand-alone and have broad acks/review tags.
Hello: This series was applied to netdev/net-next.git (main) by Jakub Kicinski <kuba@kernel.org>: On Mon, 20 Nov 2023 16:00:33 -0800 you wrote: > We recently started to deploy newer kernels / drivers at Meta, > making significant use of page pools for the first time. > We immediately run into page pool leaks both real and false positive > warnings. As Eric pointed out/predicted there's no guarantee that > applications will read / close their sockets so a page pool page > may be stuck in a socket (but not leaked) forever. This happens > a lot in our fleet. Most of these are obviously due to application > bugs but we should not be printing kernel warnings due to minor > application resource leaks. > > [...] Here is the summary with links: - [net-next,v2,01/15] net: page_pool: split the page_pool_params into fast and slow https://git.kernel.org/netdev/net-next/c/5027ec19f104 - [net-next,v2,02/15] net: page_pool: avoid touching slow on the fastpath https://git.kernel.org/netdev/net-next/c/2da0cac1e949 - [net-next,v2,03/15] net: page_pool: factor out uninit (no matching commit) - [net-next,v2,04/15] net: page_pool: id the page pools (no matching commit) - [net-next,v2,05/15] net: page_pool: record pools per netdev (no matching commit) - [net-next,v2,06/15] net: page_pool: stash the NAPI ID for easier access (no matching commit) - [net-next,v2,07/15] eth: link netdev to page_pools in drivers (no matching commit) - [net-next,v2,08/15] net: page_pool: add nlspec for basic access to page pools (no matching commit) - [net-next,v2,09/15] net: page_pool: implement GET in the netlink API (no matching commit) - [net-next,v2,10/15] net: page_pool: add netlink notifications for state changes (no matching commit) - [net-next,v2,11/15] net: page_pool: report amount of memory held by page pools (no matching commit) - [net-next,v2,12/15] net: page_pool: report when page pool was destroyed (no matching commit) - [net-next,v2,13/15] net: page_pool: expose page pool stats via netlink (no matching commit) - [net-next,v2,14/15] net: page_pool: mute the periodic warning for visible page pools (no matching commit) - [net-next,v2,15/15] tools: ynl: add sample for getting page-pool information (no matching commit) You are awesome, thank you!
On 11/22/23 02:31, Jakub Kicinski wrote: > On Mon, 20 Nov 2023 16:00:33 -0800 Jakub Kicinski wrote: >> net: page_pool: split the page_pool_params into fast and slow >> net: page_pool: avoid touching slow on the fastpath > > To relieve some of the pain and suffering this series causes to our > build tester I'm going to apply the first 2 patches already. I hope > that's fine. They are pretty stand-alone and have broad acks/review > tags. Fine by me to apply the first 2 patches. Thanks for communicating this as I did get confused seeing patchwork-bot reporting this was applied and seeing a V3 on the list. Keep up the good work. Overall I like this netlink-based introspection. I'll try to get some opinions from Cloudflare people how this can be integrated into existing monitoring systems. --Jesper