std::bad_alloc in low memory conditions

[sndive]

Do linux kernels 2.4 and 2.6 have some stash of memory set aside to
handle paging in std::bad_alloc
or ... exception handling code when new or new[] throws that exception?
How is that handled? Does gcc generate some report to the elf loader
that the largest exception
handling section is x kb long or what?

 

[Victor Bazarov]

Do linux kernels 2.4 and 2.6 have some stash of memory set aside to
handle paging in std::bad_alloc
or ... exception handling code when new or new[] throws that
exception? How is that handled? Does gcc generate some report to the
elf loader that the largest exception
handling section is x kb long or what?

This is OT in comp.lang.c++ since your question is platform-specific.
Just thought you might want to know...

 

[sndive]

Do linux kernels 2.4 and 2.6 have some stash of memory set aside to
handle paging in std::bad_alloc
or ... exception handling code when new or new[] throws that
exception? How is that handled? Does gcc generate some report to the
elf loader that the largest exception
handling section is x kb long or what?

This is OT in comp.lang.c++ since your question is platform-specific.
Just thought you might want to know...

I think this is a very comp.lang.c++ answer.

 

[Frederick Gotham]

:

I think this is a very comp.lang.c++ answer.

You obviously haven't been to comp.lang.c...

 

[Roland Pibinger]

... exception handling code when new or new[] throws that exception?
How is that handled?

OOM is usually handled with a new handler. See also:
http://www.gotw.ca/publications/mill16.htm

Best wishes,
Roland Pibinger