On Fri, 24 Sep 2004 07:27:18 +1200, Juha Saarinen <juhasaarinen_at_gmail.com> wrote: > On Thu, 23 Sep 2004 05:26:20 -0700, David G. Lawrence > <dg_at_dglawrence.com> wrote: >> I feel compelled to respond since you mentioned me above and since I >> wrote most of the code involved... :-) >> The main issue with increasing the size of ARG_MAX is that it will >> result >> in more kernel virtual memory being reserved for temporary storage of >> the >> args. This used to be a much larger problem when KVM was scarce, but >> less >> of a problem now with 1GB or more of KVM. The args temporary space is >> allocated out of exec_map (a submap of kernel_map), which is sized to be >> about 16 * ARG_MAX. The '16' is to allow up to 16 processes to >> simultaneously >> exec until additional execs are blocked waiting for KVM to become >> available. Anyway, increasing ARG_MAX to 256K (roughly 4MB of KVM) >> should >> be okay on most systems. Can't it be made dependend on kern.maxusers which is dependend on the max amount of memory available? So people with low memory aren't wasting a lot of memory? Ronald. -- Ronald Klop Amsterdam, The NetherlandsReceived on Thu Sep 23 2004 - 17:37:51 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:13 UTC