- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Is there a way to report the flags the compiler uses when -xHost is used ???
Reason:
When compiling/running on a AMD Zen3 processor the flag -xHost gives 20% better performance than using -march=znver3. Now I need to perform the compilation on an older CPU but still target the Zen3 for execution. How do I enable the same flags as using -xHost on the Zen3 but from a different CPU?
Using GCC this can be accomplished using:
> gcc -march=native -Q --help=target
Edit:
I should mention I am using the IFX compiler (not ifort).
Link Copied
0 Replies

Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page