Details
Diff Detail
Event Timeline
| ELF/Writer.cpp | ||
|---|---|---|
| 300–302 | There's assembly. It also makes layout a bit simpler, as you can assume all TLS sections are together, and there's no good reason to reject it. | |
| ELF/Writer.cpp | ||
|---|---|---|
| 300–302 | That's still a user code bug, no? I'm not convinced unless it is very prevalent issue that we had to take care of this specially. | |
Also please remove the "don't care" sections from the test:
.section v,"T",@nobits
.section u,"T"
section h,"aT",@nobits
.section f,"aT"
We only really care how we sort tls sections that are aw
| ELF/Writer.cpp | ||
|---|---|---|
| 327 | Move this past the comment // If we got here we know that both A and B are in the same PT_LOAD. | |
It sounds like a compiler bug if it emits .tbss or .tdata without the writable bit. Do we actually care about this case?