Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread William Lee Irwin III
Anton Blanchard <[EMAIL PROTECTED]> writes: >> It is possible for one task to end up "owning" an mm from another - we >> have seen this with the procfs code when process 1 accesses >> /proc/pid/cmdline of process 2 while it is exiting. Process 2 exits >> but does not tear its mm down. Later on pro

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Andi Kleen
Anton Blanchard <[EMAIL PROTECTED]> writes: > > It is possible for one task to end up "owning" an mm from another - we > have seen this with the procfs code when process 1 accesses > /proc/pid/cmdline of process 2 while it is exiting. Process 2 exits > but does not tear its mm down. Later on proce

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Nick Piggin
[Hi, please cc Andi on 4 level page tables stuff too] Anton Blanchard wrote: Hi, The 4 level pagetable code changed the exit_mmap code to rely on TASK_SIZE. On some architectures (eg ppc64 and ia64), this is a per task property and bad things can happen in certain circumstances when using it. It is

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Anton Blanchard
Hi, > I don't think Anton can count. (and for some reason I seem to be missing > his mail at the moment.) > > include/asm-arm/pgtable.h:#define FIRST_USER_PGD_NR 1 > > there's two. FIRST_USER_PGD_NR was created specifically for ARM because > many of our CPUs place their hardware vector t

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Russell King
On Tue, Jan 25, 2005 at 11:02:10PM +, Ian Molton wrote: > Anton Blanchard wrote: > > > As an aside, all architectures except one define FIRST_USER_PGD_NR as 0: > > > > include/asm-arm26/pgtable.h:#define FIRST_USER_PGD_NR 1 I don't think Anton can count. (and for some reason I seem

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Ian Molton
Anton Blanchard wrote: > As an aside, all architectures except one define FIRST_USER_PGD_NR as 0: > > include/asm-arm26/pgtable.h:#define FIRST_USER_PGD_NR 1 All processes on arm26 must map the same page 0 as its where the SWI vector table goes. The vector table is located at address 0, and

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Chris Wedgwood
On Wed, Jan 26, 2005 at 03:57:45AM +1100, Anton Blanchard wrote: > > > #ifndef MM_VM_SIZE i didn't read carefull, so no it won't warn still, i don't like that being there but that can wait i guess - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a mess

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Anton Blanchard
> I don't really like it in general when we do > > #define MAYBE_CONST_LOOKING_THING (some_func()) > > I would almost rather see something like task_size(current) used. I dont particularly like it, but it would be better for that to be a separate cleanup patch. I want to maximise my chang

Re: [PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Chris Wedgwood
On Wed, Jan 26, 2005 at 01:22:10AM +1100, Anton Blanchard wrote: > The 4 level pagetable code changed the exit_mmap code to rely on > TASK_SIZE. On some architectures (eg ppc64 and ia64), this is a per > task property and bad things can happen in certain circumstances > when using it. I don't rea

[PATCH] Use MM_VM_SIZE in exit_mmap

2005-01-25 Thread Anton Blanchard
Hi, The 4 level pagetable code changed the exit_mmap code to rely on TASK_SIZE. On some architectures (eg ppc64 and ia64), this is a per task property and bad things can happen in certain circumstances when using it. It is possible for one task to end up "owning" an mm from another - we have see