From mboxrd@z Thu Jan 1 00:00:00 1970 Subject: Re: [PATCH] net/drivers: fec: Update fec driver for xenomai 3 + linux kernel 5.10 and add I.MX8 support References: <20211126080247.3392-1-jean-baptiste.tredez@ba-healthcare.com> <9c940d0d-8be1-9a37-4c7c-060a00716ed1@siemens.com> <58af3705-b52e-7b51-1d63-f30ac56f8f11@siemens.com> From: Jean-Baptiste Tredez Message-ID: Date: Tue, 7 Dec 2021 18:39:08 +0100 In-Reply-To: <58af3705-b52e-7b51-1d63-f30ac56f8f11@siemens.com> Content-Type: text/plain; charset="utf-8"; format="flowed" Content-Transfer-Encoding: 8bit Content-Language: en-US MIME-Version: 1.0 List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Florian Bezdeka , Jan Kiszka , xenomai@xenomai.org Le 03/12/2021 à 12:29, Florian Bezdeka a écrit : > On 03.12.21 12:12, Jan Kiszka via Xenomai wrote: >> On 01.12.21 20:04, Jan Kiszka via Xenomai wrote: >>> On 26.11.21 09:02, Jean-Baptiste Trédez via Xenomai wrote: >>>> Current fec driver does not build on xenomai 3 and on recent kernel (ex : 5.10). Fec driver was completely rewritten on mainline kernel. >>>> This work remove old fec driver, port driver from mainline linux 5.10 to xenomai 3.x and add I.MX8 support. >>>> >>>> Tested on i.MX8Q target >>>> >>> Thanks for the contribution! I assume you also tested this driver with >>> debug options on to catch any potential locking issues (sorry, TL;DR - >>> with "R" = "review")? The driver was mostly tested with user application and latency check. Some tests were done with debug options. Did you think of a particular debug option ? >>> Can we also make add this driver to our CI build (.gitlab-ci.yml)? Which >>> options need to be on in addition to make it compile? >>> >> It turns out that all needed switches were already set - and... the >> build fails for 5.4: >> >> https://gitlab.com/Xenomai/xenomai-hacker-space/-/jobs/1846388391 > Taken from the job log: > > drivers/xenomai/net/drivers/freescale/fec_main.c:737:10: error: void > value not ignored as it ought to be > 737 | hdr_len = tso_start(skb, &tso); > > I guess there is more work to do. The return type of tso_start for > example changed in Linux 5.9 from void to int. So we have to deal with > different Linux versions. > I can send a new patch to fix build errors for 5.4. >> Could you have a look and fix this up? >> >> Also, it looks like we do not build for 4.19 >> (https://gitlab.com/Xenomai/xenomai-hacker-space/-/jobs/1846388392). If >> that kernel had no support for the target SoC, then this is fine, though. >> >> Thanks, >> Jan >>