re: port-arm/50563: pool allocator corruption due to __MUTEX_PRIVATE

[christos%zoulas.com@localhost (Christos Zoulas)]

The following reply was made to PR port-arm/50563; it has been noted by GNATS.

From: christos%zoulas.com@localhost (Christos Zoulas)
To: matthew green <mrg%eterna.com.au@localhost>
Cc: gnats-bugs%NetBSD.org@localhost, port-arm-maintainer%netbsd.org@localhost, 
	gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost
Subject: re: port-arm/50563: pool allocator corruption due to __MUTEX_PRIVATE
Date: Wed, 16 Dec 2015 08:24:38 -0500

 On Dec 16,  8:34pm, mrg%eterna.com.au@localhost (matthew green) wrote:
 -- Subject: re: port-arm/50563: pool allocator corruption due to __MUTEX_PRIV
 
 | > Yes, if the types of ipl_cookie_t or __cpu_simple_lock_t are not defined
 | > to be 8 bits. Is that the case? We should add a:
 | > 
 | > CTASSERT(sizeof(uintptr_t) == sizeof(struct kmutex));
 | > 
 | > in the header file to enforce that.
 | 
 | that doesn't seem sane to me.  or even valid.  alpha has a uint32 and
 | a uintprt_t.
 | 
 | struct kmutex is MD-defined.  i don't know of a good way to check
 | this without a helper.  we have to compile machine/mutex.h with and
 | without __MUTEX_PRIVATE defined, and compare their sizes.  it seems
 | way to ugly to try to do this in a single C file (you'd have to
 | #undef all the sys/arch/foo/include/mutex.h's #ifdef _FOO_MUTEX_H,
 | but it would work i think...)
 
 That CTASSERT goes in the MD file.
 
 christos