linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Duyck <alexander.duyck@gmail.com>
To: Yunsheng Lin <linyunsheng@huawei.com>
Cc: davem@davemloft.net, kuba@kernel.org, pabeni@redhat.com,
	 netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	 Andrew Morton <akpm@linux-foundation.org>,
	linux-mm@kvack.org
Subject: Re: [PATCH net-next v9 10/13] mm: page_frag: introduce prepare/probe/commit API
Date: Sat, 29 Jun 2024 10:37:43 -0700	[thread overview]
Message-ID: <CAKgT0Ueg1u2S5LJuo0Ecs9dAPPDujtJ0GLcm8BTsfDx9LpJZVg@mail.gmail.com> (raw)
In-Reply-To: <38da183b-92ba-ce9d-5472-def199854563@huawei.com>

On Sat, Jun 29, 2024 at 4:15 AM Yunsheng Lin <linyunsheng@huawei.com> wrote:
>
> On 2024/6/29 6:35, Alexander H Duyck wrote:
> > On Tue, 2024-06-25 at 21:52 +0800, Yunsheng Lin wrote:
> >> There are many use cases that need minimum memory in order
> >> for forward progress, but more performant if more memory is
> >> available or need to probe the cache info to use any memory
> >> available for frag caoleasing reason.
> >>
> >> Currently skb_page_frag_refill() API is used to solve the
> >> above use cases, but caller needs to know about the internal
> >> detail and access the data field of 'struct page_frag' to
> >> meet the requirement of the above use cases and its
> >> implementation is similar to the one in mm subsystem.
> >>
> >> To unify those two page_frag implementations, introduce a
> >> prepare API to ensure minimum memory is satisfied and return
> >> how much the actual memory is available to the caller and a
> >> probe API to report the current available memory to caller
> >> without doing cache refilling. The caller needs to either call
> >> the commit API to report how much memory it actually uses, or
> >> not do so if deciding to not use any memory.
> >>
> >> As next patch is about to replace 'struct page_frag' with
> >> 'struct page_frag_cache' in linux/sched.h, which is included
> >> by the asm-offsets.s, using the virt_to_page() in the inline
> >> helper of page_frag_cache.h cause a "'vmemmap' undeclared"
> >> compiling error for asm-offsets.s, use a macro for probe API
> >> to avoid that compiling error.
> >>
> >> CC: Alexander Duyck <alexander.duyck@gmail.com>
> >> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
> >> ---
> >>  include/linux/page_frag_cache.h |  82 +++++++++++++++++++++++
> >>  mm/page_frag_cache.c            | 114 ++++++++++++++++++++++++++++++++
> >>  2 files changed, 196 insertions(+)
> >>
> >> diff --git a/include/linux/page_frag_cache.h b/include/linux/page_frag_cache.h
> >> index b33904d4494f..e95d44a36ec9 100644
> >> --- a/include/linux/page_frag_cache.h
> >> +++ b/include/linux/page_frag_cache.h
> >> @@ -4,6 +4,7 @@
> >>  #define _LINUX_PAGE_FRAG_CACHE_H
> >>
> >>  #include <linux/gfp_types.h>
> >> +#include <linux/mmdebug.h>
> >>
> >>  #define PAGE_FRAG_CACHE_MAX_SIZE    __ALIGN_MASK(32768, ~PAGE_MASK)
> >>  #define PAGE_FRAG_CACHE_MAX_ORDER   get_order(PAGE_FRAG_CACHE_MAX_SIZE)
> >> @@ -87,6 +88,9 @@ static inline unsigned int page_frag_cache_page_size(struct encoded_va *encoded_
> >>
> >>  void page_frag_cache_drain(struct page_frag_cache *nc);
> >>  void __page_frag_cache_drain(struct page *page, unsigned int count);
> >> +struct page *page_frag_alloc_pg(struct page_frag_cache *nc,
> >> +                            unsigned int *offset, unsigned int fragsz,
> >> +                            gfp_t gfp);
> >>  void *__page_frag_alloc_va_align(struct page_frag_cache *nc,
> >>                               unsigned int fragsz, gfp_t gfp_mask,
> >>                               unsigned int align_mask);
> >> @@ -99,12 +103,90 @@ static inline void *page_frag_alloc_va_align(struct page_frag_cache *nc,
> >>      return __page_frag_alloc_va_align(nc, fragsz, gfp_mask, -align);
> >>  }
> >>
> >> +static inline unsigned int page_frag_cache_page_offset(const struct page_frag_cache *nc)
> >> +{
> >> +    return page_frag_cache_page_size(nc->encoded_va) - nc->remaining;
> >> +}
> >> +
> >>  static inline void *page_frag_alloc_va(struct page_frag_cache *nc,
> >>                                     unsigned int fragsz, gfp_t gfp_mask)
> >>  {
> >>      return __page_frag_alloc_va_align(nc, fragsz, gfp_mask, ~0u);
> >>  }
> >>
> >> +void *page_frag_alloc_va_prepare(struct page_frag_cache *nc, unsigned int *fragsz,
> >> +                             gfp_t gfp);
> >> +
> >> +static inline void *page_frag_alloc_va_prepare_align(struct page_frag_cache *nc,
> >> +                                                 unsigned int *fragsz,
> >> +                                                 gfp_t gfp,
> >> +                                                 unsigned int align)
> >> +{
> >> +    WARN_ON_ONCE(!is_power_of_2(align) || align > PAGE_SIZE);
> >> +    nc->remaining = nc->remaining & -align;
> >> +    return page_frag_alloc_va_prepare(nc, fragsz, gfp);
> >> +}
> >> +
> >> +struct page *page_frag_alloc_pg_prepare(struct page_frag_cache *nc,
> >> +                                    unsigned int *offset,
> >> +                                    unsigned int *fragsz, gfp_t gfp);
> >> +
> >> +struct page *page_frag_alloc_prepare(struct page_frag_cache *nc,
> >> +                                 unsigned int *offset,
> >> +                                 unsigned int *fragsz,
> >> +                                 void **va, gfp_t gfp);
> >> +
> >> +static inline struct encoded_va *__page_frag_alloc_probe(struct page_frag_cache *nc,
> >> +                                                     unsigned int *offset,
> >> +                                                     unsigned int *fragsz,
> >> +                                                     void **va)
> >> +{
> >> +    struct encoded_va *encoded_va;
> >> +
> >> +    *fragsz = nc->remaining;
> >> +    encoded_va = nc->encoded_va;
> >> +    *offset = page_frag_cache_page_size(encoded_va) - *fragsz;
> >> +    *va = encoded_page_address(encoded_va) + *offset;
> >> +
> >> +    return encoded_va;
> >> +}
> >> +
> >> +#define page_frag_alloc_probe(nc, offset, fragsz, va)                       \
> >> +({                                                                  \
> >> +    struct page *__page = NULL;                                     \
> >> +                                                                    \
> >> +    VM_BUG_ON(!*(fragsz));                                          \
> >> +    if (likely((nc)->remaining >= *(fragsz)))                       \
> >> +            __page = virt_to_page(__page_frag_alloc_probe(nc,       \
> >> +                                                          offset,   \
> >> +                                                          fragsz,   \
> >> +                                                          va));     \
> >> +                                                                    \
> >> +    __page;                                                         \
> >> +})
> >> +
> >
> > Why is this a macro instead of just being an inline? Are you trying to
> > avoid having to include a header due to the virt_to_page?
>
> Yes, you are right.
> I tried including different headers for virt_to_page(), and it did not
> work for arch/x86/kernel/asm-offsets.s, which has included linux/sched.h,
> and linux/sched.h need 'struct page_frag_cache' for 'struct task_struct'
> after this patchset, including page_frag_cache.h for sched.h causes the
> below compiler error:
>
>   CC      arch/x86/kernel/asm-offsets.s
> In file included from ./arch/x86/include/asm/page.h:89,
>                  from ./arch/x86/include/asm/thread_info.h:12,
>                  from ./include/linux/thread_info.h:60,
>                  from ./include/linux/spinlock.h:60,
>                  from ./include/linux/swait.h:7,
>                  from ./include/linux/completion.h:12,
>                  from ./include/linux/crypto.h:15,
>                  from arch/x86/kernel/asm-offsets.c:9:
> ./include/linux/page_frag_cache.h: In function ‘page_frag_alloc_align’:
> ./include/asm-generic/memory_model.h:37:34: error: ‘vmemmap’ undeclared (first use in this function); did you mean ‘mem_map’?
>    37 | #define __pfn_to_page(pfn)      (vmemmap + (pfn))
>       |                                  ^~~~~~~
> ./include/asm-generic/memory_model.h:65:21: note: in expansion of macro ‘__pfn_to_page’
>    65 | #define pfn_to_page __pfn_to_page
>       |                     ^~~~~~~~~~~~~
> ./arch/x86/include/asm/page.h:68:33: note: in expansion of macro ‘pfn_to_page’
>    68 | #define virt_to_page(kaddr)     pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
>       |                                 ^~~~~~~~~~~
> ./include/linux/page_frag_cache.h:151:16: note: in expansion of macro ‘virt_to_page’
>   151 |         return virt_to_page(va);
>       |                ^~~~~~~~~~~~
> ./include/asm-generic/memory_model.h:37:34: note: each undeclared identifier is reported only once for each function it appears in
>    37 | #define __pfn_to_page(pfn)      (vmemmap + (pfn))
>       |                                  ^~~~~~~
> ./include/asm-generic/memory_model.h:65:21: note: in expansion of macro ‘__pfn_to_page’
>    65 | #define pfn_to_page __pfn_to_page
>       |                     ^~~~~~~~~~~~~
> ./arch/x86/include/asm/page.h:68:33: note: in expansion of macro ‘pfn_to_page’
>    68 | #define virt_to_page(kaddr)     pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
>       |                                 ^~~~~~~~~~~
> ./include/linux/page_frag_cache.h:151:16: note: in expansion of macro ‘virt_to_page’
>   151 |         return virt_to_page(va);
>
>

I am pretty sure you just need to add:
#include <asm/page.h>


  reply	other threads:[~2024-06-29 17:38 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20240625135216.47007-1-linyunsheng@huawei.com>
2024-06-25 13:52 ` [PATCH net-next v9 01/13] mm: page_frag: add a test module for page_frag Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 02/13] mm: move the page fragment allocator from page_alloc into its own file Yunsheng Lin
2024-07-01 23:10   ` Alexander H Duyck
2024-07-02 12:27     ` Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 03/13] mm: page_frag: use initial zero offset for page_frag_alloc_align() Yunsheng Lin
2024-07-01 23:27   ` Alexander H Duyck
2024-07-02 12:28     ` Yunsheng Lin
2024-07-02 16:00       ` Alexander Duyck
2024-07-03 11:25         ` Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 04/13] mm: page_frag: add '_va' suffix to page_frag API Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 05/13] mm: page_frag: avoid caller accessing 'page_frag_cache' directly Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 06/13] mm: page_frag: reuse existing space for 'size' and 'pfmemalloc' Yunsheng Lin
2024-07-02  0:08   ` Alexander H Duyck
2024-07-02 12:35     ` Yunsheng Lin
2024-07-02 14:55       ` Alexander H Duyck
2024-07-03 12:33         ` Yunsheng Lin
2024-07-10 15:28           ` Alexander H Duyck
2024-07-11  8:16             ` Yunsheng Lin
2024-07-11 16:49               ` Alexander Duyck
2024-07-12  8:42                 ` Yunsheng Lin
2024-07-12 16:55                   ` Alexander Duyck
2024-07-13  5:20                     ` Yunsheng Lin
2024-07-13 16:55                       ` Alexander Duyck
     [not found]                         ` <12ff13d9-1f3d-4c1b-a972-2efb6f247e31@gmail.com>
2024-07-15 17:55                           ` Alexander Duyck
2024-07-16 12:58                             ` Yunsheng Lin
2024-07-17 12:31                               ` Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 07/13] mm: page_frag: some minor refactoring before adding new API Yunsheng Lin
2024-07-02 15:30   ` Alexander H Duyck
2024-07-03 12:36     ` Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 08/13] mm: page_frag: use __alloc_pages() to replace alloc_pages_node() Yunsheng Lin
2024-06-25 13:52 ` [PATCH net-next v9 10/13] mm: page_frag: introduce prepare/probe/commit API Yunsheng Lin
2024-06-28 22:35   ` Alexander H Duyck
2024-06-29 11:15     ` Yunsheng Lin
2024-06-29 17:37       ` Alexander Duyck [this message]
     [not found]         ` <0a80e362-1eb7-40b0-b1b9-07ec5a6506ea@gmail.com>
2024-06-30 14:35           ` Alexander Duyck
2024-06-30 15:05             ` Yunsheng Lin
2024-07-03 12:40               ` Yunsheng Lin
2024-07-07 17:12                 ` Alexander Duyck
2024-07-08 10:58                   ` Yunsheng Lin
2024-07-08 14:30                     ` Alexander Duyck
2024-07-09  6:57                       ` Yunsheng Lin
2024-07-09 13:40                         ` Alexander Duyck
2024-06-25 13:52 ` [PATCH net-next v9 12/13] mm: page_frag: update documentation for page_frag Yunsheng Lin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAKgT0Ueg1u2S5LJuo0Ecs9dAPPDujtJ0GLcm8BTsfDx9LpJZVg@mail.gmail.com \
    --to=alexander.duyck@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=davem@davemloft.net \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linyunsheng@huawei.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox