mirrors/Odin
1
0
Fork 0
mirror of https://github.com/odin-lang/Odin.git synced 2026-06-02 08:47:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

speed up path_to_fullpath on Linux/MacOS

Browse Source 
We did some profiling for #3343 and this seems to be the biggest
problem. `realpath` is expensive, and we are locking here for no reason
that I can think of.

This improves the "check procedure bodies" timing (of the linked issue)
from 2.4s to .4s on my machine.
This commit is contained in:
Laytan Laats
2024-03-28 18:44:21 +01:00
parent 308e9112f2
commit 63f30a8207
1 changed files with 13 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
src/build_settings.cpp
View File

@@ -1190,11 +1190,16 @@ gb_internal String path_to_fullpath(gbAllocator a, String s, bool *ok_) {
}
#elif defined(GB_SYSTEM_OSX) || defined(GB_SYSTEM_UNIX)
gb_internal String path_to_fullpath(gbAllocator a, String s, bool *ok_) {
static gb_thread_local StringMap<String> cache;
String* cached = string_map_get(&cache, s);
if (cached != nullptr) {
return copy_string(a, *cached);
}
char *p;
mutex_lock(&fullpath_mutex);
p = realpath(cast(char *)s.text, 0);
defer (free(p));
mutex_unlock(&fullpath_mutex);
if(p == nullptr) {
if (ok_) *ok_ = false;
@@ -1207,10 +1212,14 @@ gb_internal String path_to_fullpath(gbAllocator a, String s, bool *ok_) {
//
// I have opted for 2 because it is much simpler + we already return `ok = false` + further
// checks and processes will use the path and cause errors (which we want).
return copy_string(a, s);
String result = copy_string(a, s);
string_map_set(&cache, copy_string(permanent_allocator(), s), copy_string(permanent_allocator(), result));
return result;
}
if (ok_) *ok_ = true;
return copy_string(a, make_string_c(p));
String result = copy_string(a, make_string_c(p));
string_map_set(&cache, copy_string(permanent_allocator(), s), copy_string(permanent_allocator(), result));
return result;
}
#else
#error Implement system