Hi Ulrich,
Thank you for the update. Please see my comments below. I am sending v5 version of the patch series.
>That last part I suspect still won't work correctly in all cases:
>+ /* l_value holds the link-time TP-relative offset. It is unique
>+ within a module, so matching against static_offset identifies
>+ exactly the right variable. */
>+ if ((CORE_ADDR)(int64_t) ldsym.l_value != static_offset)
>+ continue;
>Sure, the offset is unique within the module where the TLS variable
>is defined. But you are not actually guaranteed to be within that
>module at this point. In fact, more likely than not, "objfile” at
>this point is some *other* module that has a *reference* to the
>TLS variable in question - or actually some *other* TLS variable
>defined in another module with the same static offset …
Please check the
rs6000_aix_objfile_exports_symbol where name is matched to the owning library before
l_value is compared
>I'm not sure how to handle the R_TLSML case - I think if a module
>has that relocation, there should also be at least one instance
>of an R_TLS_LD to a symbol in the module, and then you could
>again compute the block offset.
This is implemented implicitly. The runtime TOC value for an R_TLS_LD slot in an initial-exec library already holds the loader-adjusted absolute TP-relative offset, so reading
it directly gives the right answer without needing a separate block-offset computation.
>However, I guess depending on compiler options it might also be
>the case that all references to TLS symbols in a given module
>happen to only use R_TLS_IE - in that case, you know the module
>must be in the initial-exec segment (i.e. module ID zero), and
>then again get the offset from the R_TLS_IE relocated value.
In v5 of the patch rs6000_aix_find_initial_exec_tls_offset
() only scans R_TLS and R_TLS_LD slots , it does not scan R_TLS_IE. But R_TLS_IE slots don't need scanning, the linker writes the final absolute TP-relative offset directly into the XCOFF symbol value at link time, so
offset
(the static symbol value GDB received) is already correct. When no R_TLS/R_TLS_LD slot is found, the code falls through to:
return tp + (CORE_ADDR)(int64_t) offset;
Let me know if I missed anything that we need.
Thanks and regards,
Aditya.