[lkml]   [2011]   [Oct]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
Subject[RFC] should VM_BUG_ON(cond) really evaluate cond
In commit 4e60c86bd9e (gcc-4.6: mm: fix unused but set warnings)
Andi forced VM_BUG_ON(cond) to evaluate cond, even if CONFIG_DEBUG_VM is
not set :

#define VM_BUG_ON(cond) BUG_ON(cond)
#define VM_BUG_ON(cond) do { (void)(cond); } while (0)

As a side effect, get_page()/put_page_testzero() are performing more bus
transactions on contended cache line on some workloads (tcp on loopback
for example, where a page is acting as a shared buffer)

0,05 : ffffffff815e4775: je ffffffff815e4970 <tcp_sendmsg+0xc80>
0,05 : ffffffff815e477b: mov 0x1c(%r9),%eax // useless
3,32 : ffffffff815e477f: mov (%r9),%rax // useless
0,51 : ffffffff815e4782: lock incl 0x1c(%r9)
3,87 : ffffffff815e4787: mov (%r9),%rax
0,00 : ffffffff815e478a: test $0x80,%ah
0,00 : ffffffff815e478d: jne ffffffff815e49f2 <tcp_sendmsg+0xd02>

Of course, we have to understand why

(void) (atomic_read(&some_atomic) == 1);

generates asm code...

mov some_atomic,%eax

Ah yes, this is because of the volatile...

static inline int atomic_read(const atomic_t *v)
return (*(volatile int *)&(v)->counter);

So maybe a fix would be to introduce an atomic_read_stable() variant ?

static inline int atomic_read_stable(const atomic_t *v)
return v->counter;

Thanks !

 \ /
  Last update: 2011-10-28 03:21    [W:0.048 / U:5.076 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site