From: Lars Persson <lars.persson@axis.com>
To: linux-mips@linux-mips.org
Cc: Lars Persson <larper@axis.com>
Subject: [PATCH] MIPS: Remove race window in page fault handling
Date: Sat, 31 May 2014 12:36:06 +0200 [thread overview]
Message-ID: <1401532566-22929-1-git-send-email-larper@axis.com> (raw)
Multicore MIPSes without I/D hardware coherency suffered from a race
condition in the page fault handler. The page table entry was published
before any pending lazy D-cache flush was committed, hence it allowed
execution of stale page cache data by other VPEs in the system.
Signed-off-by: Lars Persson <larper@axis.com>
---
arch/mips/include/asm/pgtable.h | 17 +++++++++++++++--
arch/mips/mm/cache.c | 10 ++++++++++
2 files changed, 25 insertions(+), 2 deletions(-)
diff --git a/arch/mips/include/asm/pgtable.h b/arch/mips/include/asm/pgtable.h
index 008324d..7b175e5 100644
--- a/arch/mips/include/asm/pgtable.h
+++ b/arch/mips/include/asm/pgtable.h
@@ -95,6 +95,9 @@ extern void paging_init(void);
#define pmd_page_vaddr(pmd) pmd_val(pmd)
+static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
+ pte_t *ptep, pte_t pteval);
+
#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)
#define pte_none(pte) (!(((pte).pte_low | (pte).pte_high) & ~_PAGE_GLOBAL))
@@ -118,7 +121,6 @@ static inline void set_pte(pte_t *ptep, pte_t pte)
}
}
}
-#define set_pte_at(mm, addr, ptep, pteval) set_pte(ptep, pteval)
static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
{
@@ -155,7 +157,6 @@ static inline void set_pte(pte_t *ptep, pte_t pteval)
}
#endif
}
-#define set_pte_at(mm, addr, ptep, pteval) set_pte(ptep, pteval)
static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
{
@@ -169,6 +170,18 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *pt
}
#endif
+extern void mips_flush_dcache_from_pte(pte_t pteval);
+
+static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
+ pte_t *ptep, pte_t pteval)
+{
+ /* Make code globally visible before publishing the page
+ table entry. */
+ if (addr < TASK_SIZE && pte_present(pteval))
+ mips_flush_dcache_from_pte(pteval);
+ set_pte(ptep, pteval);
+}
+
/*
* (pmds are folded into puds so this doesn't get actually called,
* but the define is needed for a generic inline function.)
diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c
index 9e67cde..1320afc 100644
--- a/arch/mips/mm/cache.c
+++ b/arch/mips/mm/cache.c
@@ -137,6 +137,16 @@ void __update_cache(struct vm_area_struct *vma, unsigned long address,
}
}
+void mips_flush_dcache_from_pte(pte_t pteval)
+{
+ unsigned long pfn = pte_pfn(pteval);
+ if (likely(pfn_valid(pfn))) {
+ struct page *page = pfn_to_page(pfn);
+ if (Page_dcache_dirty(page))
+ flush_dcache_page(page);
+ }
+}
+
unsigned long _page_cachable_default;
EXPORT_SYMBOL(_page_cachable_default);
--
1.7.10.4
next reply other threads:[~2014-05-31 10:36 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-31 10:36 Lars Persson [this message]
2014-06-02 16:47 ` [PATCH] MIPS: Remove race window in page fault handling David Daney
2014-06-03 10:29 ` Lars Persson
2014-12-03 13:37 ` Ralf Baechle
2014-12-04 5:55 ` Jayachandran C.
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=1401532566-22929-1-git-send-email-larper@axis.com \
--to=lars.persson@axis.com \
--cc=larper@axis.com \
--cc=linux-mips@linux-mips.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