| 22 Dec 2025 |
 Randy Eckenrode | https://github.com/apple-oss-distributions/xnu/blob/f6217f891ac0bb64f3d375211650a4c1ff8ca1ea/libsyscall/wrappers/select-base.c#L169 | 16:58:24 |
| Randy Eckenrode | It appears the check may be happening prior to the syscall? | 17:00:49 |
 Ihar Hrachyshka | implies we would need our own libSystem or smth? :D | 17:01:09 |
| Randy Eckenrode | Darwin doesn’t support alternate libcs. There is no guarantee of syscall compatibility between kernel versions. | 17:01:45 |
| Randy Eckenrode | (You can try to do it. Go did. Someone got Musl working. If you encounter problems, you’ll be informed that you shouldn’t be doing that.) | 17:02:35 |
| Ihar Hrachyshka | guess we could ship a custom select in glib... with some kind of runtime check during build to avoid broken glib if they ever change the behavior... | 17:04:34 |
| Randy Eckenrode | https://github.com/apple-oss-distributions/xnu/blob/f6217f891ac0bb64f3d375211650a4c1ff8ca1ea/bsd/sys/_types/_fd_def.h#L37_L40 | 17:04:36 |
| Randy Eckenrode | What happens if you define your own FD_SETSIZE? | 17:04:52 |
| Ihar Hrachyshka | I assume that since libSystem is as it was compiled with the 1024 limit, it won't do much. but let me check. | 17:05:51 |
| Randy Eckenrode | The checks are inline functions in the header. | 17:06:44 |
| Randy Eckenrode | Or setting _DARWIN_UNLIMITED_SELECT with select? | 17:07:08 |
| Randy Eckenrode | * Or defining _DARWIN_UNLIMITED_SELECT when using select? | 17:07:19 |
| Randy Eckenrode | * Or defining _DARWIN_UNLIMITED_SELECT when using select or pselect? | 17:07:42 |
| Ihar Hrachyshka | I copied half of that file, commented out include for select.h, removed the check and now it works
=== Test 2: pselect() with FD >= FD_SETSIZE (DANGEROUS!) ===
This test intentionally demonstrates undefined behavior
Attempting to add FD 1024 to fd_set (FD_SETSIZE=1024)...
WARNING: This will cause undefined behavior!
Before FD_SET: fd_set memory looks normal
fd_set bytes before: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Calling FD_SET(1024, &dangerous_set)...
fd_set bytes after: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Memory corruption may have occurred!
Attempting pselect() with corrupted fd_set...
pselect() unexpectedly succeeded with 0 ready FDs
I don't know if "works" means more than "it returned 0" but...
| 17:11:29 |
