From: Yunhui Cui <cuiyunhui@bytedance.com>
To: akpm@linux-foundation.org, alex@ghiti.fr,
andrew+kernel@donnellan.id.au, andreyknvl@gmail.com,
anup@brainfault.org, aou@eecs.berkeley.edu, apopple@nvidia.com,
ardb@kernel.org, atish.patra@linux.dev,
baolin.wang@linux.alibaba.com, cuiyunhui@bytedance.com,
david@kernel.org, debug@rivosinc.com,
djordje.todorovic@htecgroup.com, dvyukov@google.com,
elver@google.com, glider@google.com, ilias.apalodimas@linaro.org,
junhui.liu@pigmoral.tech, kasan-dev@googlegroups.com,
kees@kernel.org, kevin.brodsky@arm.com,
kvm-riscv@lists.infradead.org, kvm@vger.kernel.org,
linux-efi@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-riscv@lists.infradead.org, liu.xuemei1@zte.com.cn,
ljs@kernel.org, namcao@linutronix.de, osalvador@suse.de,
palmer@dabbelt.com, pjw@kernel.org, rmclure@linux.ibm.com,
rostedt@goodmis.org, rppt@kernel.org, ryabinin.a.a@gmail.com,
surenb@google.com, vincenzo.frascino@arm.com,
vishal.moola@gmail.com, wangruikang@iscas.ac.cn,
zhangchunyan@iscas.ac.cn
Subject: [PATCH 4/7] riscv: hugetlb: switch NAPOT mappings to raw PTE helpers
Date: Tue, 21 Apr 2026 17:24:54 +0800 [thread overview]
Message-ID: <20260421092457.37649-5-cuiyunhui@bytedance.com> (raw)
In-Reply-To: <20260421092457.37649-1-cuiyunhui@bytedance.com>
Use raw PTE helpers in hugetlb code to operate directly on the
underlying PTE entries. This lets hugetlb manage NAPOT
folding/unfolding explicitly instead of going through Svnapot-aware
public wrappers.
Add explicit NAPOT unfolding in set_huge_pte_at() before replacing an
existing NAPOT mapping with non-NAPOT entries.
No functional change intended.
Signed-off-by: Yunhui Cui <cuiyunhui@bytedance.com>
---
arch/riscv/mm/hugetlbpage.c | 55 +++++++++++++++++++++++--------------
1 file changed, 35 insertions(+), 20 deletions(-)
diff --git a/arch/riscv/mm/hugetlbpage.c b/arch/riscv/mm/hugetlbpage.c
index a6d217112cf46..65a89b4fdad8b 100644
--- a/arch/riscv/mm/hugetlbpage.c
+++ b/arch/riscv/mm/hugetlbpage.c
@@ -7,7 +7,7 @@ pte_t huge_ptep_get(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
{
unsigned long pte_num;
int i;
- pte_t orig_pte = ptep_get(ptep);
+ pte_t orig_pte = __ptep_get(ptep);
if (!pte_present(orig_pte) || !pte_napot(orig_pte))
return orig_pte;
@@ -15,7 +15,7 @@ pte_t huge_ptep_get(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
pte_num = napot_pte_num(napot_cont_order(orig_pte));
for (i = 0; i < pte_num; i++, ptep++) {
- pte_t pte = ptep_get(ptep);
+ pte_t pte = __ptep_get(ptep);
if (pte_dirty(pte))
orig_pte = pte_mkdirty(orig_pte);
@@ -74,7 +74,7 @@ pte_t *huge_pte_alloc(struct mm_struct *mm,
out:
if (pte) {
- pte_t pteval = ptep_get_lockless(pte);
+ pte_t pteval = __ptep_get_lockless(pte);
WARN_ON_ONCE(pte_present(pteval) && !pte_huge(pteval));
}
@@ -153,12 +153,12 @@ static pte_t get_clear_contig(struct mm_struct *mm,
pte_t pte, tmp_pte;
bool present;
- pte = ptep_get_and_clear(mm, addr, ptep);
+ pte = __ptep_get_and_clear(mm, addr, ptep);
present = pte_present(pte);
while (--ncontig) {
ptep++;
addr += PAGE_SIZE;
- tmp_pte = ptep_get_and_clear(mm, addr, ptep);
+ tmp_pte = __ptep_get_and_clear(mm, addr, ptep);
if (present) {
if (pte_dirty(tmp_pte))
pte = pte_mkdirty(pte);
@@ -210,7 +210,7 @@ static void clear_flush(struct mm_struct *mm,
unsigned long i, saddr = addr;
for (i = 0; i < ncontig; i++, addr += pgsize, ptep++)
- ptep_get_and_clear(mm, addr, ptep);
+ __ptep_get_and_clear(mm, addr, ptep);
flush_tlb_range(&vma, saddr, addr);
}
@@ -250,25 +250,40 @@ void set_huge_pte_at(struct mm_struct *mm,
unsigned long sz)
{
size_t pgsize;
+ pte_t orig_pte;
+ pte_t pteval;
int i, pte_num;
pte_num = num_contig_ptes_from_size(sz, &pgsize);
if (!pte_present(pte)) {
- for (i = 0; i < pte_num; i++, ptep++, addr += pgsize)
- set_ptes(mm, addr, ptep, pte, 1);
+ for (i = 0; i < pte_num; i++, ptep++, addr += pgsize) {
+ pteval = pte_mknonnapot(pte, addr);
+ orig_pte = __ptep_get(ptep);
+
+ if (pte_present_napot(orig_pte))
+ __napotpte_try_unfold(mm, addr, ptep, orig_pte);
+
+ __set_ptes(mm, addr, ptep, pteval, 1);
+ }
return;
}
if (!pte_napot(pte)) {
- set_ptes(mm, addr, ptep, pte, 1);
+ pteval = pte_mknonnapot(pte, addr);
+ orig_pte = __ptep_get(ptep);
+
+ if (pte_present_napot(orig_pte))
+ __napotpte_try_unfold(mm, addr, ptep, orig_pte);
+
+ __set_ptes(mm, addr, ptep, pteval, 1);
return;
}
clear_flush(mm, addr, ptep, pgsize, pte_num);
for (i = 0; i < pte_num; i++, ptep++, addr += pgsize)
- set_pte_at(mm, addr, ptep, pte);
+ __set_ptes(mm, addr, ptep, pte, 1);
}
int huge_ptep_set_access_flags(struct vm_area_struct *vma,
@@ -283,7 +298,7 @@ int huge_ptep_set_access_flags(struct vm_area_struct *vma,
int i, pte_num;
if (!pte_napot(pte))
- return ptep_set_access_flags(vma, addr, ptep, pte, dirty);
+ return __ptep_set_access_flags(vma, addr, ptep, pte, dirty);
order = napot_cont_order(pte);
pte_num = napot_pte_num(order);
@@ -307,11 +322,11 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm,
pte_t *ptep, unsigned long sz)
{
size_t pgsize;
- pte_t orig_pte = ptep_get(ptep);
+ pte_t orig_pte = __ptep_get(ptep);
int pte_num;
if (!pte_napot(orig_pte))
- return ptep_get_and_clear(mm, addr, ptep);
+ return __ptep_get_and_clear(mm, addr, ptep);
pte_num = num_contig_ptes_from_size(sz, &pgsize);
@@ -322,13 +337,13 @@ void huge_ptep_set_wrprotect(struct mm_struct *mm,
unsigned long addr,
pte_t *ptep)
{
- pte_t pte = ptep_get(ptep);
+ pte_t pte = __ptep_get(ptep);
unsigned long order;
pte_t orig_pte;
int i, pte_num;
if (!pte_napot(pte)) {
- ptep_set_wrprotect(mm, addr, ptep);
+ __ptep_set_wrprotect(mm, addr, ptep);
return;
}
@@ -347,11 +362,11 @@ pte_t huge_ptep_clear_flush(struct vm_area_struct *vma,
unsigned long addr,
pte_t *ptep)
{
- pte_t pte = ptep_get(ptep);
+ pte_t pte = __ptep_get(ptep);
int pte_num;
if (!pte_napot(pte))
- return ptep_clear_flush(vma, addr, ptep);
+ return __ptep_clear_flush(vma, addr, ptep);
pte_num = napot_pte_num(napot_cont_order(pte));
@@ -364,18 +379,18 @@ void huge_pte_clear(struct mm_struct *mm,
unsigned long sz)
{
size_t pgsize;
- pte_t pte = ptep_get(ptep);
+ pte_t pte = __ptep_get(ptep);
int i, pte_num;
if (!pte_napot(pte)) {
- pte_clear(mm, addr, ptep);
+ __pte_clear(mm, addr, ptep);
return;
}
pte_num = num_contig_ptes_from_size(sz, &pgsize);
for (i = 0; i < pte_num; i++, addr += pgsize, ptep++)
- pte_clear(mm, addr, ptep);
+ __pte_clear(mm, addr, ptep);
}
static bool is_napot_size(unsigned long size)
--
2.39.5
next prev parent reply other threads:[~2026-04-21 9:25 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-21 9:24 [PATCH v1 0/7] riscv: add Svnapot-based contiguous PTE support Yunhui Cui
2026-04-21 9:24 ` [PATCH 1/7] riscv: mm: split raw and public PTE helpers Yunhui Cui
2026-04-21 9:24 ` [PATCH 2/7] riscv/kvm: use raw PTE helpers for G-stage leaf PTEs Yunhui Cui
2026-04-21 9:24 ` [PATCH 3/7] riscv: mm: add Svnapot-aware contiguous PTE wrappers Yunhui Cui
2026-04-21 9:24 ` Yunhui Cui [this message]
2026-04-21 9:24 ` [PATCH 5/7] riscv: add contiguous PTE range clearing helpers Yunhui Cui
2026-04-21 9:24 ` [PATCH 6/7] riscv: batch write-protect contiguous PTE ranges Yunhui Cui
2026-04-21 9:24 ` [PATCH 7/7] riscv: add Svnapot-aware pte_batch_hint support Yunhui Cui
2026-04-21 11:28 ` [PATCH v1 0/7] riscv: add Svnapot-based contiguous PTE support David Hildenbrand (Arm)
2026-07-15 11:36 ` [External] " yunhui cui
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=20260421092457.37649-5-cuiyunhui@bytedance.com \
--to=cuiyunhui@bytedance.com \
--cc=akpm@linux-foundation.org \
--cc=alex@ghiti.fr \
--cc=andrew+kernel@donnellan.id.au \
--cc=andreyknvl@gmail.com \
--cc=anup@brainfault.org \
--cc=aou@eecs.berkeley.edu \
--cc=apopple@nvidia.com \
--cc=ardb@kernel.org \
--cc=atish.patra@linux.dev \
--cc=baolin.wang@linux.alibaba.com \
--cc=david@kernel.org \
--cc=debug@rivosinc.com \
--cc=djordje.todorovic@htecgroup.com \
--cc=dvyukov@google.com \
--cc=elver@google.com \
--cc=glider@google.com \
--cc=ilias.apalodimas@linaro.org \
--cc=junhui.liu@pigmoral.tech \
--cc=kasan-dev@googlegroups.com \
--cc=kees@kernel.org \
--cc=kevin.brodsky@arm.com \
--cc=kvm-riscv@lists.infradead.org \
--cc=kvm@vger.kernel.org \
--cc=linux-efi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=liu.xuemei1@zte.com.cn \
--cc=ljs@kernel.org \
--cc=namcao@linutronix.de \
--cc=osalvador@suse.de \
--cc=palmer@dabbelt.com \
--cc=pjw@kernel.org \
--cc=rmclure@linux.ibm.com \
--cc=rostedt@goodmis.org \
--cc=rppt@kernel.org \
--cc=ryabinin.a.a@gmail.com \
--cc=surenb@google.com \
--cc=vincenzo.frascino@arm.com \
--cc=vishal.moola@gmail.com \
--cc=wangruikang@iscas.ac.cn \
--cc=zhangchunyan@iscas.ac.cn \
/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®