On Wed, Aug 10, 2005 at 03:51:21PM -0700, john liu wrote: > Thanks Dan, > > I guess the reason is that the flag for 8192 cache is > reset to 0x269, note 0x9 not 0xf. 0x9 does not have > the redzone buf set. > > Other caches having 0x22f as the flag.
0x269 indicates KMF_FIREWALL (0x40); there is no need for a redzone, since the buffer is followed by an unmapped page, and the buffer will be unmapped the moment it is freed. Because of this, there is no need for ::kmem_verify to do any checking on the cache. Cheers, - jonathan > john > > > --- Dan Price <dp at eng.sun.com> wrote: > > > On Wed 10 Aug 2005 at 03:00PM, Qiang Liu wrote: > > > > 0000030000065b88::kmem_verify > > > Summary for cache 'kmem_alloc_2688' > > > clean > > > > 0000030000068288::kmem_verify > > > mdb: cache 30000068288 (kmem_alloc_8192) does not > > have redzone > > > checking enabled > > > > And I'm not sure what the answer is for the > > kmem_alloc_8192 cache. > > > > -dp > > > > -- > > Daniel Price - Solaris Kernel Engineering - > > dp at eng.sun.com - blogs.sun.com/dp > > > > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > mdb-discuss mailing list > mdb-discuss at opensolaris.org -- Jonathan Adams, Solaris Kernel Development
