Maxim Pugachev wrote:
> Currently two checks in free() function confirm the correctness of
> freedsize argument. I think that it's better to check that provided
> freedsize fall into the same bucket that was recorded in kmemusage
> struct: it covers both cases.
I don't know. This doesn't seem clearer to me. An error message with
bucket numbers isn't any more helpful.
>
>
> Index: sys/kern/kern_malloc.c
> ===================================================================
> RCS file: /cvs/src/sys/kern/kern_malloc.c,v
> retrieving revision 1.128
> diff -u -p -r1.128 kern_malloc.c
> --- sys/kern/kern_malloc.c 14 Mar 2015 03:38:50 -0000 1.128
> +++ sys/kern/kern_malloc.c 9 Dec 2015 17:54:30 -0000
> @@ -387,12 +387,10 @@ free(void *addr, int type, size_t freeds
> size = kup->ku_pagecnt << PAGE_SHIFT;
> s = splvm();
> #ifdef DIAGNOSTIC
> - if (freedsize != 0 && freedsize > size)
> - panic("free: size too large %zu > %ld (%p) type %s",
> - freedsize, size, addr, memname[type]);
> - if (freedsize != 0 && size > MINALLOCSIZE && freedsize < size / 2)
> - panic("free: size too small %zu < %ld / 2 (%p) type %s",
> - freedsize, size, addr, memname[type]);
> + if (freedsize != 0 && kup->ku_indx != BUCKETINDX(freedsize))
> + panic("free: size is wrong %zu (bucket %ld, should be
> %d) type %s",
> + freedsize, BUCKETINDX(freedsize),
> + kup->ku_indx, memname[type]);
> /*
> * Check for returns of data that do not point to the
> * begi
>
