hssi: suggest hssi{loopback,stats} when eth interface is absent

OFS / opae-sdk

Open Programmable Acceleration Engine

https://ofs.github.io

BSD 3-Clause "New" or "Revised" License

251 stars 84 forks source link

hssi: suggest hssi{loopback,stats} when eth interface is absent #3117

Closed pcolberg closed 4 months ago

pcolberg commented 4 months ago

While the D5005 FPGA card employs a separate kernel driver (s10hssi) to configure loopback and provide statistics via a network device, newer FPGA cards expose a UIO device used with hssiloopback and hssistats.

Closes: https://hsdes.intel.com/appstore/article/#/22019839240

Cc: @badanur

coveralls commented 4 months ago

Pull Request Test Coverage Report for Build 8653758794

Details

0 of 0 changed or added relevant lines in 0 files are covered.
No unchanged relevant lines lost coverage.
Overall coverage remained the same at 64.673%

Totals
Change from base Build 8653071263:	0.0%
Covered Lines:	15823
Relevant Lines:	24466

💛 - Coveralls

fpgamatt commented 4 months ago

This change is helpful because it adds information to a weird error message, but can't we do better? Can't the hssi command figure out if the hssi in question is d5005 or a uio? That way no weird message would need to be printed out in the first place.

pcolberg commented 4 months ago

This change is helpful because it adds information to a weird error message, but can't we do better? Can't the hssi command figure out if the hssi in question is d5005 or a uio? That way no weird message would need to be printed out in the first place.

Thanks @fpgamatt, my thought as well; why was this functionality not integrated into the hssi command in the first place? That requires a bit more investigation; I have opened https://github.com/OFS/opae-sdk/issues/3118.