Szabolcs Nagy

@jackschmidt can you bulid this code and run it on the kernel where you observed the issue and send us vdso.so: ``` #include #include int main() { void *p =...

@jackschmidt please explain what that patch is? are you saying you are using a broken modified kernel with non-upstream patches? so far only you could reproduce the bug you reported...

@jackschmidt what makes you think it affects other hw? i don't see the problem on a cortex a72 with same toolchain and 5.6 arm 32bit kernel, i'm building the rpi...

those macros sound like perprocessor api between compiler and libc which would be nice to document, but not necessary in a language spec. however ptraddr_t is a c language extension...

> NIT maybe the draft could be dropped now from the name of the PR. potential issues with current spec: - gcs store intrinsic is missing - const void* vs...

we will have to extend ifunc arg with hwcap3 soon so it is a good time to discuss extensions. i think pagesize is ok to add, pointer to auxv may...

this prevents plt hooking at runtime (not just lazy binding). e.g. glibc ld.so supports LD_PROFILE and LD_AUDIT even with bind-now (then plt got is not readonly). but i think there...

well a reasonable interpretation of "the relocation type is pc-relative" is that the relocation operation has "P" in it. the operation of R_AARCH64_ADR_GOT_PAGE is "Page(G(GDAT(S+A)))-Page(P)" so it is pc-relative and...

relaxing adrp+ldr to adrp+add is not valid for undef weak symbols, but i think you could relax to mov+nop (or mov+mov if the registers are different). (btw this is an...

ah yes, adrp can work in pde (assuming the adrp location is below 4G which is the case usually and in your example)