Bug report

Callingsys.intern() on a string that's not interned acquires the interpreter-wideinterned_mutex. This can lead to scaling bottlenecks either due to explicitsys.intern() calls or indirectly becausePyObject_SetAttr calls_PyUnicode_InternMortal.

We should avoid the lock acquisition if there is an interned copy already present in theinterned_dict.

For example:

_ATTR_PREFIX="bench"@register_benchmarkdefsetattr_non_interned():prefix=_ATTR_PREFIXobj=MyObject()for_inrange(1000*WORK_SCALE):setattr(obj,f"{prefix}_a",None)setattr(obj,f"{prefix}_b",None)setattr(obj,f"{prefix}_c",None)

This came up when looking at Django scaling bottlenecks.

Linked PRs

• gh-146041: Avoid lock in sys.intern() for already interned strings #146072
• [3.14] gh-146041: Avoid lock in sys.intern() for already interned strings (gh-146072) #146390