29/02/2024 18:38, Stephen Hemminger: > On Thu, 29 Feb 2024 17:58:13 +0100 > Morten Brørup <m...@smartsharesystems.com> wrote: > > > > From: Stephen Hemminger [mailto:step...@networkplumber.org] > > > Sent: Thursday, 29 February 2024 17.45 > > > > > > On Thu, 29 Feb 2024 16:42:56 +0100 > > > Thomas Monjalon <tho...@monjalon.net> wrote: > > > > > > > +/* Link modes sorted with index as defined in ethtool. > > > > + * Values are speed in Mbps with LSB indicating duplex. > > > > + * > > > > + * The ethtool bits definition should not change as it is a kernel > > > API. > > > > + * Using raw numbers directly avoids checking API availability > > > > + * and allows to compile with new bits included even on an old > > > kernel. > > > > + * > > > > + * The array below is built from bit definitions with this shell > > > command: > > > > + * sed -rn 's;.*(ETHTOOL_LINK_MODE_)([0-9]+)([0-9a-zA-Z_]*).*= > > > *([0-9]*).*;'\ > > > > + * '[\4] = \2, /\* \1\2\3 *\/;p' > > > /usr/include/linux/ethtool.h | > > > > + * awk '/_Half_/{$3=$3+1","}1' > > > > + */ > > > > +static uint32_t link_modes[] = { > > > > > > Make it const please.
Yes > > > You could add meson rule to generate it and then use non-numeric tags. > > > > However you do it, make sure it cross builds. The kernel/ethtool on the > > target system may differ from the one on the build system. > > > > If the build system is older, the speed table will be smaller. And the code > should just print "Unknown" > If the build system is newer, then the table will be larger than kernel ever > returns which is Ok. There is no benefit in having a smaller table. That's why I prefer using numeric indices with the best coverage possible.
