All of lore.kernel.org
 help / color / mirror / Atom feed
From: Julien Grall <julien@xen.org>
To: Penny Zheng <Penny.Zheng@arm.com>, xen-devel@lists.xenproject.org
Cc: wei.chen@arm.com, Andrew Cooper <andrew.cooper3@citrix.com>,
	George Dunlap <george.dunlap@citrix.com>,
	Jan Beulich <jbeulich@suse.com>,
	Stefano Stabellini <sstabellini@kernel.org>, Wei Liu <wl@xen.org>
Subject: Re: [PATCH v4 6/6] xen: retrieve reserved pages on populate_physmap
Date: Mon, 16 May 2022 19:29:13 +0100	[thread overview]
Message-ID: <7af0d53c-6fdc-33a4-a09b-93831b4ee2c0@xen.org> (raw)
In-Reply-To: <20220510022733.2422581-7-Penny.Zheng@arm.com>

Hi Penny,

On 10/05/2022 03:27, Penny Zheng wrote:
> When static domain populates memory through populate_physmap on runtime,

Typo: s/when static/when a static/ or "when static domains populate"

s/on runtime/at runtime/

> other than allocating from heap, it shall retrieve reserved pages from

I am not sure to understand the part before the comma. But it doens't 
sound necessary so maybe drop it?

> resv_page_list to make sure that guest RAM is still restricted in statically
> configured memory regions. And this commit introduces a new helper
> acquire_reserved_page to make it work.
> 
> Signed-off-by: Penny Zheng <penny.zheng@arm.com>

[...]

> diff --git a/xen/common/page_alloc.c b/xen/common/page_alloc.c
> index 290526adaf..06e7037a28 100644
> --- a/xen/common/page_alloc.c
> +++ b/xen/common/page_alloc.c
> @@ -2740,8 +2740,8 @@ static struct page_info * __init acquire_staticmem_pages(mfn_t smfn,
>    * Acquire nr_mfns contiguous pages, starting at #smfn, of static memory,
>    * then assign them to one specific domain #d.
>    */
> -int __init acquire_domstatic_pages(struct domain *d, mfn_t smfn,
> -                                   unsigned int nr_mfns, unsigned int memflags)
> +int acquire_domstatic_pages(struct domain *d, mfn_t smfn, unsigned int nr_mfns,
> +                            unsigned int memflags)
>   {
>       struct page_info *pg;
>   
> @@ -2769,12 +2769,43 @@ int __init acquire_domstatic_pages(struct domain *d, mfn_t smfn,
>   
>       return 0;
>   }
> +
> +/*
> + * Acquire a page from reserved page list(resv_page_list), when populating
> + * memory for static domain on runtime.
> + */
> +mfn_t acquire_reserved_page(struct domain *d, unsigned int memflags)
> +{
> +    struct page_info *page;
> +    mfn_t smfn;
> +
> +    /* Acquire a page from reserved page list(resv_page_list). */
> +    page = page_list_remove_head(&d->resv_page_list);
Alloc/free of memory can happen concurrently. So access to rsv_page_list 
needs to be protected with a spinlock (mostly like d->page_alloc_lock).

> +    if ( unlikely(!page) )
> +        return INVALID_MFN;
> +
> +    smfn = page_to_mfn(page);
> +
> +    if ( acquire_domstatic_pages(d, smfn, 1, memflags) )

I am OK if we call acquire_domstatic_pages() for now. But long term, I 
think we should consider to optimize it because we know the page is 
valid and belong to the guest. So there are a lot of pointless work 
(checking mfn_valid(), scrubbing in the free part, cleaning the cache...).

> +    {
> +        page_list_add_tail(page, &d->resv_page_list);
> +        return INVALID_MFN;
> +    }
> +
> +    return smfn;
> +}
>   #else
>   void free_staticmem_pages(struct page_info *pg, unsigned long nr_mfns,
>                             bool need_scrub)
>   {
>       ASSERT_UNREACHABLE();
>   }
> +
> +mfn_t acquire_reserved_page(struct domain *d, unsigned int memflags)
> +{
> +    ASSERT_UNREACHABLE();
> +    return INVALID_MFN;
> +}
>   #endif
>   
>   /*
> diff --git a/xen/include/xen/domain.h b/xen/include/xen/domain.h
> index 35dc7143a4..c613afa57e 100644
> --- a/xen/include/xen/domain.h
> +++ b/xen/include/xen/domain.h
> @@ -38,6 +38,10 @@ void arch_get_domain_info(const struct domain *d,
>   #define CDF_staticmem            (1U << 2)
>   #endif
>   
> +#ifndef is_domain_using_staticmem
> +#define is_domain_using_staticmem(d) ((void)(d), false)
> +#endif
> +
>   /*
>    * Arch-specifics.
>    */
> diff --git a/xen/include/xen/mm.h b/xen/include/xen/mm.h
> index 9fd95deaec..74810e1f54 100644
> --- a/xen/include/xen/mm.h
> +++ b/xen/include/xen/mm.h
> @@ -92,6 +92,7 @@ void free_staticmem_pages(struct page_info *pg, unsigned long nr_mfns,
>   int acquire_domstatic_pages(struct domain *d, mfn_t smfn, unsigned int nr_mfns,
>                               unsigned int memflags);
>   #endif
> +mfn_t acquire_reserved_page(struct domain *d, unsigned int memflags);
>   
>   /* Map machine page range in Xen virtual address space. */
>   int map_pages_to_xen(

Cheers,

-- 
Julien Grall


  reply	other threads:[~2022-05-16 18:29 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-10  2:27 [PATCH V4 0/6] populate/unpopulate memory when domain on static Penny Zheng
2022-05-10  2:27 ` [PATCH v4 1/6] xen: do not free reserved memory into heap Penny Zheng
2022-05-16 18:01   ` Julien Grall
2022-05-17  8:21     ` Penny Zheng
2022-05-17  9:28       ` Julien Grall
2022-05-17 16:07         ` Jan Beulich
2022-05-18  6:06         ` Penny Zheng
2022-05-17 16:11   ` Jan Beulich
2022-05-18  2:29     ` Penny Zheng
2022-05-18  6:30       ` Jan Beulich
2022-05-10  2:27 ` [PATCH v4 2/6] xen: do not merge reserved pages in free_heap_pages() Penny Zheng
2022-05-16 17:36   ` Julien Grall
2022-05-10  2:27 ` [PATCH v4 3/6] xen: add field "flags" to cover all internal CDF_XXX Penny Zheng
2022-05-10  2:27 ` [PATCH v4 4/6] xen/arm: introduce CDF_staticmem Penny Zheng
2022-05-10  2:27 ` [PATCH v4 5/6] xen/arm: unpopulate memory when domain is static Penny Zheng
2022-05-10  2:27 ` [PATCH v4 6/6] xen: retrieve reserved pages on populate_physmap Penny Zheng
2022-05-16 18:29   ` Julien Grall [this message]
2022-05-17  6:24     ` Penny Zheng
2022-05-17  8:48       ` Julien Grall
2022-05-17 16:16   ` Jan Beulich

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=7af0d53c-6fdc-33a4-a09b-93831b4ee2c0@xen.org \
    --to=julien@xen.org \
    --cc=Penny.Zheng@arm.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=george.dunlap@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=sstabellini@kernel.org \
    --cc=wei.chen@arm.com \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.