From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Laight Date: Wed, 26 Aug 2020 09:56:56 +0000 Subject: RE: [PATCH 00/13] lib/generic-radix-tree: genradix bug fix and optimisations. Message-Id: <7894f0bc41764c0a8cb87f1482df7679@AcuMS.aculab.com> List-Id: References: <21289d79b0474811b21ed8478c465159@AcuMS.aculab.com> <20200825154127.GB2444@localhost.localdomain> In-Reply-To: <20200825154127.GB2444@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: 'Marcelo Ricardo Leitner' Cc: "linux-kernel@vger.kernel.org" , "'linux-sctp@vger.kernel.org'" , Eric Biggers , 'Catalin Marinas' , "'kent.overstreet@gmail.com'" , Andrew Morton , 'Neil Horman' From: 'Marcelo Ricardo Leitner' > Sent: 25 August 2020 16:41 > > On Tue, Aug 25, 2020 at 02:52:34PM +0000, David Laight wrote: > > The genradix code is used by SCTP for accessing per-stream data. > > This means there are quite a lot of lookups but the code wasn't > > really optimised at all. > > My test box is down for the moment and will bring it on later today or > tomorrow, so I can't test it yet. What should we expect as performance > gains here? I've just done quick test that shows ~1% improvement on an MTP3+M2PA 'double reflect' test using the 'test' userpart and a test program. So there is quite a lot of extra protocol stack in the way. The test is running ~200k very small data chunks/sec (tx and rx) on a poor little 2.4GHz 8 core atom C2758. However the 'baseline' kernel just locked up. (Reports 5.8.0+ so must be late in the merge for rc1). Annoyingly the Ubuntu 20.04 userspace won't show the kernel messages (only a graphical login) - so I can see any kernel ooops messages. I'm going to have to setup a serial console. That'll need a trip into the office. David - Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)