mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Anshuman Khandual <anshuman.khandual@arm.com>
To: linux-arm-kernel@lists.infradead.org, catalin.marinas@arm.com,
	will@kernel.org
Cc: Anshuman Khandual <anshuman.khandual@arm.com>,
	Ryan Roberts <ryan.roberts@arm.com>,
	Mark Rutland <mark.rutland@arm.com>,
	Lorenzo Stoakes <ljs@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	David Hildenbrand <david@kernel.org>,
	Mike Rapoport <rppt@kernel.org>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH V3 06/14] arm64/mm: Enable fixmap with 5 level page table
Date: Tue, 17 Nov 2026 12:36:05 +0530	[thread overview]
Message-ID: <20261117070613.1183243-7-anshuman.khandual@arm.com> (raw)
In-Reply-To: <20261117070613.1183243-1-anshuman.khandual@arm.com>

FEAT_D128 halves PTRS_PER_PXX thus shrinking the VA range coverage for each
pgtable level. So in order to preserve all existing VA range configurations
for various base page sizes, some pgtable configurations now need to become
5-level which were 4-level earlier.

Since fixmap builds and manipulates kernel page tables early on during boot
these mappings must also gain that additional level, which was not required
earlier.

Changes in the page table geometry can result in the P4D table covering the
fixmap range no longer being populated by other kernel initialization code.

Enable this scenario by allocating and populating a static P4D table during
early fixmap initialization, similar to the existing handling of other page
table levels.

Add a pgd_page_paddr() fallback stub when (PGTBALE_LEVELS <= 4) which helps
in intercepting any unintended usage.

Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Ryan Roberts <ryan.roberts@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
---
Changes in V3:

- Updated the commit message

 arch/arm64/include/asm/pgtable.h |  1 +
 arch/arm64/mm/fixmap.c           | 18 +++++++++++++++---
 2 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 737e1e42f1ec0..290aefcf0f0f1 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -1184,6 +1184,7 @@ static inline p4d_t *p4d_offset_kimg(pgd_t *pgdp, u64 addr)
 #else
 
 static inline bool pgtable_l5_enabled(void) { return false; }
+#define pgd_page_paddr(pgd)	({ BUILD_BUG(); 0; })
 
 #define p4d_index(addr)		(((addr) >> P4D_SHIFT) & (PTRS_PER_P4D - 1))
 
diff --git a/arch/arm64/mm/fixmap.c b/arch/arm64/mm/fixmap.c
index 2a0fe0b0f7710..5512521fef8fa 100644
--- a/arch/arm64/mm/fixmap.c
+++ b/arch/arm64/mm/fixmap.c
@@ -34,6 +34,7 @@ static_assert(NR_BM_PMD_TABLES == 1);
 static pte_t bm_pte[NR_BM_PTE_TABLES][PTRS_PER_PTE] __bss_pgtbl;
 static pmd_t bm_pmd[PTRS_PER_PMD] __bss_pgtbl __maybe_unused;
 static pud_t bm_pud[PTRS_PER_PUD] __bss_pgtbl __maybe_unused;
+static p4d_t bm_p4d[PTRS_PER_P4D] __bss_pgtbl __maybe_unused;
 
 static inline pte_t *fixmap_pte(unsigned long addr)
 {
@@ -95,6 +96,19 @@ static void __init early_fixmap_init_pud(p4d_t *p4dp, unsigned long addr,
 	early_fixmap_init_pmd(pudp, addr, end);
 }
 
+static void __init early_fixmap_init_p4d(pgd_t *pgdp, unsigned long addr,
+					 unsigned long end)
+{
+	pgd_t pgd = pgdp_get(pgdp);
+	p4d_t *p4dp;
+
+	if (pgd_none(pgd))
+		__pgd_populate(pgdp, __pa_symbol(bm_p4d),
+			       PGD_TYPE_TABLE | PGD_TABLE_AF);
+	p4dp = p4d_offset_kimg(pgdp, addr);
+	early_fixmap_init_pud(p4dp, addr, end);
+}
+
 /*
  * The p*d_populate functions call virt_to_phys implicitly so they can't be used
  * directly on kernel symbols (bm_p*d). This function is called too early to use
@@ -107,9 +121,7 @@ void __init early_fixmap_init(void)
 	unsigned long end = FIXADDR_TOP;
 
 	pgd_t *pgdp = pgd_offset_k(addr);
-	p4d_t *p4dp = p4d_offset_kimg(pgdp, addr);
-
-	early_fixmap_init_pud(p4dp, addr, end);
+	early_fixmap_init_p4d(pgdp, addr, end);
 }
 
 /*
-- 
2.43.0


  parent reply	other threads:[~2026-11-17  7:06 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-17  7:05 [PATCH V3 00/14] arm64/mm: Enable 128 bit page table entries Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 01/14] mm: Add read-write accessors for vm_page_prot Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 02/14] arm64/mm: Route all pgtable reads via ptval_get() Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 03/14] arm64/mm: Route all pgtable writes via ptval_set() Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 04/14] arm64/mm: Route all pgtable atomics to central helpers Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 05/14] arm64/mm: Override read-write accessors for vm_page_prot Anshuman Khandual
2026-11-17  7:06 ` Anshuman Khandual [this message]
2026-11-17  7:06 ` [PATCH V3 07/14] arm64/mm: Enable pgtable geometry for FEAT_D128 Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 08/14] arm64/mm: Enable pgtable descriptor " Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 09/14] arm64/mm: Enable 128 bit translation Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 10/14] arm64/mm: Add an abstraction level for tlbi_op Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 11/14] arm64/mm: Enable TLBIP instruction based TLB flush Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 12/14] arm64/mm: Relax KASAN shadow alignment for FEAT_D128 Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 13/14] arm64/mm: Make ARM64_D128 selectable Anshuman Khandual
2026-11-17  7:06 ` [PATCH V3 14/14] arm64/configs: Add d128.config Anshuman Khandual

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=20261117070613.1183243-7-anshuman.khandual@arm.com \
    --to=anshuman.khandual@arm.com \
    --cc=akpm@linux-foundation.org \
    --cc=catalin.marinas@arm.com \
    --cc=david@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=ljs@kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=rppt@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=will@kernel.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

all inboxes | Powered by JetHome®