Re: increasing sysv semaphore limits
Jason Thorpe <[email protected]> Thu, 16 Jul 2026 08:23:21 -0700
| Newsgroups | gmane.os.netbsd.devel.kernel |
|---|---|
| Message-ID | <[email protected]> |
> On Jul 16, 2026, at 8:05=E2=80=AFAM, J=C3=B6rg Sonnenberger = <[email protected]> wrote: >=20 > On 7/16/26 12:23 AM, Jason Thorpe wrote: >>> On Jul 15, 2026, at 2:58=E2=80=AFPM, Mouse = <[email protected]> wrote: >>>=20 >>> Except doesn't that open you up to a DoS attack, running the kernel = out >>> of memory by creating a zillion semaphores? >> I mean, it=E2=80=99s really the same thing as with file descriptors. >=20 > Yes and no. SYSV semaphores are persistent and independent of both = file descriptors and processes. So even killing processes is not going = to return the memory used by them. While true, I mean it=E2=80=99s more akin to having dynamically-tunable = maximums / limits. -- thorpej