[clang] [llvm] [MC] Make UseAssemblerInfoForParsing mostly true (PR #91082)

[via cfe-commits]

gulfemsavrun wrote:

> > > @gulfemsavrun Commit [245491a](https://github.com/llvm/llvm-project/commit/245491a9f384e4c53421196533c2a2b693efaf8d) (`[MC] Disable MCAssembler based constant folding for DwarfDebug`; independent of this patch) should have fixed the compile time regression. Might be worth testing the commit before and at [245491a](https://github.com/llvm/llvm-project/commit/245491a9f384e4c53421196533c2a2b693efaf8d) to see the time decrease.
> > 
> > 
> > @MaskRay I already checked with that patch, and we still have the same issue. We see this issue in llvm ToT. For example, this is a run from today: [luci-milo.appspot.com/ui/p/fuchsia/builders/prod/clang-mac-x64/b8746512510687347537/overview](https://luci-milo.appspot.com/ui/p/fuchsia/builders/prod/clang-mac-x64/b8746512510687347537/overview)
> 
> This change exposed a latent performance issue (primarily due to -O1 or above + -g) in https://reviews.llvm.org/D153096 , which should be properly fixed by [b06e736](https://github.com/llvm/llvm-project/commit/b06e736982a3568fe2bcea8688550f9e393b7450) (just landed). Can you verify?
> 
> [245491a](https://github.com/llvm/llvm-project/commit/245491a9f384e4c53421196533c2a2b693efaf8d) was a workaround, but I believe it effectively addressed the performance issue, and I am unable to detect compile time regression. I am curious if you can find a source file that demonstrates a slowdown with [245491a](https://github.com/llvm/llvm-project/commit/245491a9f384e4c53421196533c2a2b693efaf8d) applied. The `AttemptToFoldSymbolOffsetDifference` function is likely related.
> 
> (The aforementioned commits modify different files and are independent.)

We build with Full LTO, and linking takes much longer with this patch. For example,  building `bin/lldb-server` takes almost 2.5x longer time now.
1) Before
```
bin/lldb-server: 9 min 18 seconds
```
2) After
```
bin/lldb-server: 22 min 24 seconds
```

Is there a specific source file or binary that might be interesting to look at so I can provide more information?



https://github.com/llvm/llvm-project/pull/91082