linux-kernel - Re: [PATCH 1/9] mm: add generic adaptive large memory allocation APIs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTinLT5g5SKjqmQlS2kxvvMq1gsi1jPDgOKTnrT-q@mail.gmail.com>
Date:	Thu, 13 May 2010 22:08:26 +0800
From:	Changli Gao <xiaosuo@...il.com>
To:	Peter Zijlstra <peterz@...radead.org>
Cc:	akpm@...ux-foundation.org, Hoang-Nam Nguyen <hnguyen@...ibm.com>,
	Christoph Raisch <raisch@...ibm.com>,
	Roland Dreier <rolandd@...co.com>,
	Sean Hefty <sean.hefty@...el.com>,
	Hal Rosenstock <hal.rosenstock@...il.com>,
	Divy Le Ray <divy@...lsio.com>,
	"James E.J. Bottomley" <James.Bottomley@...e.de>,
	"Theodore Ts'o" <tytso@....edu>, Andreas Dilger <adilger@....com>,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Paul Menage <menage@...gle.com>,
	Li Zefan <lizf@...fujitsu.com>, linux-rdma@...r.kernel.org,
	linux-kernel@...r.kernel.org, netdev@...r.kernel.org,
	linux-scsi@...r.kernel.org, linux-ext4@...r.kernel.org,
	linux-fsdevel@...r.kernel.org, linux-mm@...ck.org,
	containers@...ts.linux-foundation.org,
	Eric Dumazet <eric.dumazet@...il.com>,
	Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
Subject: Re: [PATCH 1/9] mm: add generic adaptive large memory allocation APIs

On Thu, May 13, 2010 at 9:20 PM, Peter Zijlstra <peterz@...radead.org> wrote:
> On Thu, 2010-05-13 at 17:51 +0800, Changli Gao wrote:
>> +void *__kvmalloc(size_t size, gfp_t flags)
>> +{
>> +       void *ptr;
>> +
>> +       if (size < PAGE_SIZE)
>> +               return kmalloc(size, GFP_KERNEL | flags);
>> +       size = PAGE_ALIGN(size);
>> +       if (is_power_of_2(size))
>> +               ptr = (void *)__get_free_pages(GFP_KERNEL | flags |
>> +                                              __GFP_NOWARN, get_order(size));
>> +       else
>> +               ptr = alloc_pages_exact(size, GFP_KERNEL | flags |
>> +                                             __GFP_NOWARN);
>> +       if (ptr != NULL) {
>> +               virt_to_head_page(ptr)->private = size;
>> +               return ptr;
>> +       }
>> +
>> +       ptr = vmalloc(size);
>> +       if (ptr != NULL && (flags & __GFP_ZERO))
>> +               memset(ptr, 0, size);
>> +
>> +       return ptr;
>> +}
>> +EXPORT_SYMBOL(__kvmalloc);
>
> So if I do kvmalloc(size, GFP_ATOMIC) I get GFP_KERNEL|GFP_ATOMIC, which
> is not a recommended variation because one should not mix __GFP_WAIT and
> __GFP_HIGH.

__kvmalloc() is only for internal use(kvmalloc, kvcalloc, and
kvzalloc), and the only value of flags is __GFP_ZERO. How about
replacing flags with a bool variable zero?

void *__kvmalloc(size_t size, bool zero);

 Or check the value of flags in the front of __kvmalloc().

BUG_ON((flags & (~__GFP_ZERO)) != 0);

>
> So I would simply drop the gfp argument to avoid confusion.
>
>> +void __kvfree(void *ptr, bool inatomic)
>> +{
>> +       if (unlikely(ZERO_OR_NULL_PTR(ptr)))
>> +               return;
>> +       if (is_vmalloc_addr(ptr)) {
>> +               if (inatomic) {
>> +                       struct work_struct *work;
>> +
>> +                       work = ptr;
>> +                       BUILD_BUG_ON(sizeof(struct work_struct) > PAGE_SIZE);
>> +                       INIT_WORK(work, kvfree_work);
>> +                       schedule_work(work);
>> +               } else {
>> +                       vfree(ptr);
>> +               }
>> +       } else {
>> +               struct page *page;
>> +
>> +               page = virt_to_head_page(ptr);
>> +               if (PageSlab(page) || PageCompound(page))
>> +                       kfree(ptr);
>> +               else if (is_power_of_2(page->private))
>> +                       free_pages((unsigned long)ptr,
>> +                                  get_order(page->private));
>> +               else
>> +                       free_pages_exact(ptr, page->private);
>> +       }
>> +}
>> +EXPORT_SYMBOL(__kvfree);
>
> NAK, I really utterly dislike that inatomic argument. The alloc side
> doesn't function in atomic context either. Please keep the thing
> symmetric in that regards.
>

There are some users, who release memory in atomic context. for
example: fs/file.c: fdmem.

-- 
Regards，
Changli Gao(xiaosuo@...il.com)