More breaking of userspace

Posted May 18, 2024 21:05 UTC (Sat) by roc (subscriber, #30627)
In reply to: More breaking of userspace by vbabka
Parent article: White paper: Vendor Kernels, Bugs and Stability

I changed rr to use MAP_FIXED. I don't think it's worth making a kernel change just for old versions of rr, the impact on rr users is actually minimal in practice.

But I do think it would make sense to honour the hint if possible. If that's what the user wants, why not give it to them?

More breaking of userspace

Posted May 23, 2024 7:46 UTC (Thu) by taladar (subscriber, #68407) [Link]

I would suggest the opposite. The only implementation I would never want to see is one treating a hint as a guarantee. If you guarantee the behavior specify it as a guarantee, don't make it implicit like that.