From: Jan Beulich <jbeulich@suse.com>
To: Hongyan Xia <hx242@xen.org>
Cc: xen-devel@lists.xenproject.org,
"Roger Pau Monné" <roger.pau@citrix.com>,
julien@xen.org, "Wei Liu" <wl@xen.org>,
"Andrew Cooper" <andrew.cooper3@citrix.com>
Subject: Re: [PATCH v2 1/5] x86/shim: map and unmap page tables in replace_va_mapping
Date: Thu, 9 Apr 2020 11:42:02 +0200 [thread overview]
Message-ID: <ddbad9f5-307e-7b1d-0cc7-cd7ed684f680@suse.com> (raw)
In-Reply-To: <7638095024ec3379a8d9ddadfe47e36da168e4dd.1586352238.git.hongyxia@amazon.com>
On 08.04.2020 15:36, Hongyan Xia wrote:
> --- a/xen/arch/x86/pv/shim.c
> +++ b/xen/arch/x86/pv/shim.c
> @@ -168,16 +168,17 @@ const struct platform_bad_page *__init pv_shim_reserved_pages(unsigned int *size
> static void __init replace_va_mapping(struct domain *d, l4_pgentry_t *l4start,
> unsigned long va, mfn_t mfn)
> {
> - l4_pgentry_t *pl4e = l4start + l4_table_offset(va);
> - l3_pgentry_t *pl3e = l4e_to_l3e(*pl4e) + l3_table_offset(va);
> - l2_pgentry_t *pl2e = l3e_to_l2e(*pl3e) + l2_table_offset(va);
> - l1_pgentry_t *pl1e = l2e_to_l1e(*pl2e) + l1_table_offset(va);
> + l4_pgentry_t l4e = l4start[l4_table_offset(va)];
> + l3_pgentry_t l3e = l3e_from_l4e(l4e, l3_table_offset(va));
> + l2_pgentry_t l2e = l2e_from_l3e(l3e, l2_table_offset(va));
> + l1_pgentry_t *pl1e = map_l1t_from_l2e(l2e) + l1_table_offset(va);
> struct page_info *page = mfn_to_page(l1e_get_mfn(*pl1e));
>
> put_page_and_type(page);
>
> *pl1e = l1e_from_mfn(mfn, (!is_pv_32bit_domain(d) ? L1_PROT
> : COMPAT_L1_PROT));
> + UNMAP_DOMAIN_PAGE(pl1e);
> }
As said before, here and below I think it should be unmap_domain_page().
> --- a/xen/include/asm-x86/page.h
> +++ b/xen/include/asm-x86/page.h
> @@ -196,6 +196,19 @@ static inline l4_pgentry_t l4e_from_paddr(paddr_t pa, unsigned int flags)
> #define map_l2t_from_l3e(x) (l2_pgentry_t *)map_domain_page(l3e_get_mfn(x))
> #define map_l3t_from_l4e(x) (l3_pgentry_t *)map_domain_page(l4e_get_mfn(x))
>
> +/* Unlike lYe_to_lXe(), lXe_from_lYe() do not rely on the direct map. */
> +#define l2e_from_l3e(l3e, offset) ({ \
> + const l2_pgentry_t *l2t = map_l2t_from_l3e(l3e); \
> + l2_pgentry_t l2e = l2t[offset]; \
> + UNMAP_DOMAIN_PAGE(l2t); \
> + l2e; })
> +
> +#define l3e_from_l4e(l4e, offset) ({ \
> + const l3_pgentry_t *l3t = map_l3t_from_l4e(l4e); \
> + l3_pgentry_t l3e = l3t[offset]; \
> + UNMAP_DOMAIN_PAGE(l3t); \
> + l3e; })
I think l1e_from_l2e() should be introduced at the same time, even
if for now it's unused. I also think, like we do elsewhere, that
macro-local variables would better have _ suffixes, to avoid
possible variable aliasing issues.
Jan
next prev parent reply other threads:[~2020-04-09 9:42 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-08 13:36 [PATCH v2 0/5] use new API for Xen page tables Hongyan Xia
2020-04-08 13:36 ` [PATCH v2 1/5] x86/shim: map and unmap page tables in replace_va_mapping Hongyan Xia
2020-04-09 9:42 ` Jan Beulich [this message]
2020-04-14 16:53 ` Hongyan Xia
2020-04-15 6:13 ` Jan Beulich
2020-04-08 13:36 ` [PATCH v2 2/5] x86_64/mm: map and unmap page tables in m2p_mapped Hongyan Xia
2020-04-15 7:54 ` Jan Beulich
2020-04-08 13:36 ` [PATCH v2 3/5] x86_64/mm: map and unmap page tables in share_hotadd_m2p_table Hongyan Xia
2020-04-15 7:55 ` Jan Beulich
2020-04-08 13:36 ` [PATCH v2 4/5] x86_64/mm: map and unmap page tables in destroy_compat_m2p_mapping Hongyan Xia
2020-04-15 8:21 ` [PATCH 0/2] x86: high compat r/o M2P table handling adjustments Jan Beulich
2020-04-15 8:23 ` [PATCH 1/2] x86: drop unnecessary page table walking in compat r/o M2P handling Jan Beulich
2020-04-15 9:59 ` Hongyan Xia
2020-04-15 10:34 ` Jan Beulich
2020-04-15 10:50 ` Hongyan Xia
2020-04-15 11:16 ` Wei Liu
2020-04-15 8:23 ` [PATCH 2/2] x86: drop high compat r/o M2P table address range Jan Beulich
2020-04-27 19:52 ` Wei Liu
2020-04-28 6:14 ` [PATCH 0/2] x86: high compat r/o M2P table handling adjustments Jan Beulich
2020-04-15 8:25 ` [PATCH v2 4/5] x86_64/mm: map and unmap page tables in destroy_compat_m2p_mapping Jan Beulich
2020-04-08 13:36 ` [PATCH v2 5/5] x86_64/mm: map and unmap page tables in destroy_m2p_mapping Hongyan Xia
2020-04-15 8:32 ` 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=ddbad9f5-307e-7b1d-0cc7-cd7ed684f680@suse.com \
--to=jbeulich@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=hx242@xen.org \
--cc=julien@xen.org \
--cc=roger.pau@citrix.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).