Thanks. I noticed it. I should change my testing to include building aarch64 image.
On Wed, Sep 5, 2018 at 11:47 Nadav Har'El <n...@scylladb.com> wrote: > > On Wed, Sep 5, 2018 at 6:43 PM, Waldek Kozaczuk <jwkozac...@gmail.com> > wrote: > >> >> >> On Tue, Sep 4, 2018 at 2:36 AM Nadav Har'El <n...@scylladb.com> wrote: >> >> -all: $(out)/loader.img links >>>> +all: $(out)/loader.img links $(out)/loader.bin >>>> >>> >>> This is not a huge issue, but now every "make" will make yet another >>> copy (or even two, in boot32.o and loader.bin) of the kernel. >>> I'm not sure what to suggest instead, though. >>> >>> One option is to just not put loader.bin in "all", and ask a user to >>> manually "make" it (probably should add a comment on this target explaining >>> what it is). >>> Another option is to have a separate script, like we have >>> scripts/convert, which "converts" loader.img into loader.bin. >>> I guess we can consider these options later. Or leave it like it is now. >>> >> I will try to eventually address some way, either by following your >> suggestion or making lzloader handle multiboot. >> > > Thanks. By the way as you probably already noticed, I had to undo the > unconditional build of loader.bin under "all" > for the aarch64 architecture, because there we don't have boot32.S so we > can't build loader.bin, and this broke > the aarch64 build. > > For x64, I left the automatic building of loader.bin in "all". > > -- You received this message because you are subscribed to the Google Groups "OSv Development" group. To unsubscribe from this group and stop receiving emails from it, send an email to osv-dev+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
