From: Ryan Roberts <ryan.roberts@arm.com>
To: Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>,
Pasha Tatashin <pasha.tatashin@soleen.com>,
Andrew Morton <akpm@linux-foundation.org>,
Uladzislau Rezki <urezki@gmail.com>,
Christoph Hellwig <hch@infradead.org>,
David Hildenbrand <david@redhat.com>,
"Matthew Wilcox (Oracle)" <willy@infradead.org>,
Mark Rutland <mark.rutland@arm.com>,
Anshuman Khandual <anshuman.khandual@arm.com>,
Alexandre Ghiti <alexghiti@rivosinc.com>,
Kevin Brodsky <kevin.brodsky@arm.com>
Cc: Ryan Roberts <ryan.roberts@arm.com>,
linux-arm-kernel@lists.infradead.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: [PATCH v3 06/11] arm64/mm: Hoist barriers out of set_ptes_anysz() loop
Date: Tue, 4 Mar 2025 15:04:36 +0000 [thread overview]
Message-ID: <20250304150444.3788920-7-ryan.roberts@arm.com> (raw)
In-Reply-To: <20250304150444.3788920-1-ryan.roberts@arm.com>
set_ptes_anysz() previously called __set_pte() for each PTE in the
range, which would conditionally issue a DSB and ISB to make the new PTE
value immediately visible to the table walker if the new PTE was valid
and for kernel space.
We can do better than this; let's hoist those barriers out of the loop
so that they are only issued once at the end of the loop. We then reduce
the cost by the number of PTEs in the range.
Signed-off-by: Ryan Roberts <ryan.roberts@arm.com>
---
arch/arm64/include/asm/pgtable.h | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index e255a36380dc..1898c3069c43 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -317,13 +317,11 @@ static inline void __set_pte_nosync(pte_t *ptep, pte_t pte)
WRITE_ONCE(*ptep, pte);
}
-static inline void __set_pte(pte_t *ptep, pte_t pte)
+static inline void __set_pte_complete(pte_t pte)
{
- __set_pte_nosync(ptep, pte);
-
/*
* Only if the new pte is valid and kernel, otherwise TLB maintenance
- * or update_mmu_cache() have the necessary barriers.
+ * has the necessary barriers.
*/
if (pte_valid_not_user(pte)) {
dsb(ishst);
@@ -331,6 +329,12 @@ static inline void __set_pte(pte_t *ptep, pte_t pte)
}
}
+static inline void __set_pte(pte_t *ptep, pte_t pte)
+{
+ __set_pte_nosync(ptep, pte);
+ __set_pte_complete(pte);
+}
+
static inline pte_t __ptep_get(pte_t *ptep)
{
return READ_ONCE(*ptep);
@@ -647,12 +651,14 @@ static inline void set_ptes_anysz(struct mm_struct *mm, pte_t *ptep, pte_t pte,
for (;;) {
__check_safe_pte_update(mm, ptep, pte);
- __set_pte(ptep, pte);
+ __set_pte_nosync(ptep, pte);
if (--nr == 0)
break;
ptep++;
pte = pte_advance_pfn(pte, stride);
}
+
+ __set_pte_complete(pte);
}
static inline void __set_ptes(struct mm_struct *mm,
--
2.43.0
next prev parent reply other threads:[~2025-03-04 15:05 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-04 15:04 [PATCH v3 00/11] Perf improvements for hugetlb and vmalloc on arm64 Ryan Roberts
2025-03-04 15:04 ` [PATCH v3 01/11] arm64: hugetlb: Cleanup huge_pte size discovery mechanisms Ryan Roberts
2025-04-03 20:46 ` Catalin Marinas
2025-04-04 3:03 ` Anshuman Khandual
2025-03-04 15:04 ` [PATCH v3 02/11] arm64: hugetlb: Refine tlb maintenance scope Ryan Roberts
2025-04-03 20:47 ` Catalin Marinas
2025-04-04 3:50 ` Anshuman Khandual
2025-03-04 15:04 ` [PATCH v3 03/11] mm/page_table_check: Batch-check pmds/puds just like ptes Ryan Roberts
2025-03-26 14:48 ` Pasha Tatashin
2025-03-26 14:54 ` Ryan Roberts
2025-04-03 20:46 ` Catalin Marinas
2025-03-04 15:04 ` [PATCH v3 04/11] arm64/mm: Refactor __set_ptes() and __ptep_get_and_clear() Ryan Roberts
2025-03-06 5:08 ` kernel test robot
2025-03-06 11:54 ` Ryan Roberts
2025-04-14 16:25 ` Catalin Marinas
2025-03-04 15:04 ` [PATCH v3 05/11] arm64: hugetlb: Use set_ptes_anysz() and ptep_get_and_clear_anysz() Ryan Roberts
2025-03-05 16:00 ` kernel test robot
2025-03-05 16:32 ` Ryan Roberts
2025-04-03 20:47 ` Catalin Marinas
2025-03-04 15:04 ` Ryan Roberts [this message]
2025-04-03 20:46 ` [PATCH v3 06/11] arm64/mm: Hoist barriers out of set_ptes_anysz() loop Catalin Marinas
2025-04-04 4:11 ` Anshuman Khandual
2025-03-04 15:04 ` [PATCH v3 07/11] mm/vmalloc: Warn on improper use of vunmap_range() Ryan Roberts
2025-03-27 13:05 ` Uladzislau Rezki
2025-03-04 15:04 ` [PATCH v3 08/11] mm/vmalloc: Gracefully unmap huge ptes Ryan Roberts
2025-03-04 15:04 ` [PATCH v3 09/11] arm64/mm: Support huge pte-mapped pages in vmap Ryan Roberts
2025-03-04 15:04 ` [PATCH v3 10/11] mm/vmalloc: Enter lazy mmu mode while manipulating vmalloc ptes Ryan Roberts
2025-03-27 13:06 ` Uladzislau Rezki
2025-04-03 20:47 ` Catalin Marinas
2025-04-04 4:54 ` Anshuman Khandual
2025-03-04 15:04 ` [PATCH v3 11/11] arm64/mm: Batch barriers when updating kernel mappings Ryan Roberts
2025-04-04 6:02 ` Anshuman Khandual
2025-04-14 17:38 ` Catalin Marinas
2025-04-14 18:28 ` Ryan Roberts
2025-04-15 10:51 ` Catalin Marinas
2025-04-15 17:28 ` Ryan Roberts
2025-03-27 12:16 ` [PATCH v3 00/11] Perf improvements for hugetlb and vmalloc on arm64 Uladzislau Rezki
2025-03-27 13:46 ` Ryan Roberts
2025-04-14 13:56 ` Ryan Roberts
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=20250304150444.3788920-7-ryan.roberts@arm.com \
--to=ryan.roberts@arm.com \
--cc=akpm@linux-foundation.org \
--cc=alexghiti@rivosinc.com \
--cc=anshuman.khandual@arm.com \
--cc=catalin.marinas@arm.com \
--cc=david@redhat.com \
--cc=hch@infradead.org \
--cc=kevin.brodsky@arm.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mark.rutland@arm.com \
--cc=pasha.tatashin@soleen.com \
--cc=urezki@gmail.com \
--cc=will@kernel.org \
--cc=willy@infradead.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