Skip to contentSkip to conversation

PID_MAX_LIMIT limit on ia32?

DISCUSSION Davy Durham 1 participant 1 message Nov 24, 2010

lore
Share & tools
lore
#0 lore raw

I have searched around without a definitive answer to my question.

PID_MAX_LIMIT is currently defined as:

/*
* A maximum of 4 million PIDs should be enough for a while.
* [NOTE: PID/TIDs are limited to 2^29 ~= 500+ million, see
futex.h.]

         */
        #define PID_MAX_LIMIT (CONFIG_BASE_SMALL ? PAGE_SIZE * 8 :

(sizeof(long) > 4 ? 4 * 1024 * 1024 : PID_MAX_DEFAULT))

And PID_MAX_DEFAULT is normally 32k.

I found a comment on the net:

“When recycling PID numbers, the kernel must manage a pidmap_array bitmap that denotes which are the PIDs currently assigned and which are the free ones. Because a page frame contains 32,768 bits, in 32-bit architectures the pidmap_array bitmap is stored in a single page. In 64-bit architectures, however, additional pages can be added to the bitmap when the kernel assigns a PID number too large for the current bitmap size. These pages are never released.”

Which may seem to indicate that PID_MAX_DEFAULT cannot be more than 1 memory page on 32bit archs.

Can the PID_MAX_LIMIT #define be increased safely on the ia32 arch which would cause pidmap_array to use more than one memory page? (say 4 pages or so).

I thought about just compiling it and testing things, but was unsure if I would run into problems long-term.

[Please CC me on the response as I'm not currently subscribed]

TIA,
Davy