On 09.04.2026 09:52, Roger Pau Monné wrote: > On Wed, Apr 08, 2026 at 02:07:23PM +0200, Jan Beulich wrote: >> This core backing function is uniform; what varies across architectures >> are the attributes passed and hence the wrappers around it. Yet of course >> extra checking or special handling may be needed per arch, so introduce a >> suitable hook. >> >> Signed-off-by: Jan Beulich <[email protected]> > > Reviewed-by: Roger Pau Monné <[email protected]>
Thanks. > Just looked at the common and x86 bits, both LGTM. I'll transcribe this into Reviewed-by: Roger Pau Monné <[email protected]> # x86, common >> --- >> Declarations (or inline counterparts) aren't being moved around, to avoid >> the need to touch source files using the functions. Quite possibly they >> want to consistently go into xen/io.h and asm/io.h. >> >> Of course ioremap.c could also go into lib/. > > Maybe I'm missing the point, but what's the benefit for it to go into > lib/? Is there any realistic scenario where we might have a > functional hypervisor build that doesn't require ioremap? Well, the difference likely is between "in principle" vs "in practice". In principle a lot of code could really be regarded "library" like. In practice whether to put code in lib/ really matters for stuff which may only be needed in certain configurations. Jan
