[PATCH -next v7 5/6] arm64/mm: Enable ARCH_SUPPORTS_PAGE_TABLE_CHECK
Anders Roxell
anders.roxell at linaro.org
Sat May 28 09:41:55 PDT 2022
On Tue, 17 May 2022 at 16:33, Anders Roxell <anders.roxell at linaro.org> wrote:
>
> On Tue, 17 May 2022 at 16:02, Pasha Tatashin <pasha.tatashin at soleen.com> wrote:
> >
> > On Tue, May 17, 2022 at 9:54 AM Anders Roxell <anders.roxell at linaro.org> wrote:
> > >
> > > On 2022-05-07 11:01, Tong Tiangen wrote:
> > > > From: Kefeng Wang <wangkefeng.wang at huawei.com>
> > > >
> > > > As commit d283d422c6c4 ("x86: mm: add x86_64 support for page table check")
> > > > , enable ARCH_SUPPORTS_PAGE_TABLE_CHECK on arm64.
> > > >
> > > > Add additional page table check stubs for page table helpers, these stubs
> > > > can be used to check the existing page table entries.
> > > >
> > > > Signed-off-by: Kefeng Wang <wangkefeng.wang at huawei.com>
> > > > Signed-off-by: Tong Tiangen <tongtiangen at huawei.com>
> > > > Reviewed-by: Pasha Tatashin <pasha.tatashin at soleen.com>
> > >
> > > When building and booting an arm64 allmodconfig kernel on the next tree, branch next-20220516,
> > > see the following kernel oops when booting in QEMU [1]:
> > >
> > > T35] ------------[ cut here ]------------
> > > [ 578.695796][ T35] kernel BUG at mm/page_table_check.c:82!
> > > [ 578.697292][ T35] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
> > > [ 578.704318][ T35] Modules linked in:
> > > [ 578.705907][ T35] CPU: 0 PID: 35 Comm: khugepaged Tainted: G T 5.18.0-rc6-next-20220513 #1 893498a5d8159d9fb26e12492a93c07e83dd4b7f
> > > [ 578.711170][ T35] Hardware name: linux,dummy-virt (DT)
> > > [ 578.713315][ T35] pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > > [ 578.716398][ T35] pc : page_table_check_clear.constprop.0+0x1f4/0x280
> > > [ 578.719107][ T35] lr : page_table_check_clear.constprop.0+0x1cc/0x280
> > > [ 578.721781][ T35] sp : ffff80000f3778b0
> > > [ 578.723446][ T35] x29: ffff80000f3778b0 x28: ffff80000b891218 x27: ffff000012dd55f0
> > > [ 578.726667][ T35] x26: 0000000000000008 x25: ffff80000c38cd80 x24: 0000000000000000
> > > [ 578.729870][ T35] x23: ffff80000c38c9c0 x22: 0000000000000000 x21: 0000000000000200
> > > [ 578.733079][ T35] x20: ffff000007bae000 x19: ffff000007bae008 x18: 0000000000000000
> > > [ 578.736299][ T35] x17: 0000000000000000 x16: 0000000000000000 x15: 0000000000000000
> > > [ 578.739505][ T35] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
> > > [ 578.742735][ T35] x11: 0000000000000000 x10: 0000000000000000 x9 : 0000000000000000
> > > [ 578.745925][ T35] x8 : 0000000000000000 x7 : 0000000000000000 x6 : 0000000000000000
> > > [ 578.749145][ T35] x5 : 0000000000000000 x4 : 0000000000000000 x3 : ffff000007bae00c
> > > [ 578.752348][ T35] x2 : 0000000000000000 x1 : 0000000000000001 x0 : 00000000ffffffff
> > > [ 578.755556][ T35] Call trace:
> > > [ 578.756877][ T35] page_table_check_clear.constprop.0+0x1f4/0x280
> > > [ 578.759446][ T35] __page_table_check_pmd_clear+0xc4/0x140
> > > [ 578.761757][ T35] pmdp_collapse_flush+0xa4/0x1c0
> > > [ 578.763771][ T35] collapse_huge_page+0x4e4/0xb00
> > > [ 578.765778][ T35] khugepaged_scan_pmd+0xc18/0xd00
> > > [ 578.767840][ T35] khugepaged_scan_mm_slot+0x580/0x780
> > > [ 578.770018][ T35] khugepaged+0x2dc/0x400
> > > [ 578.771786][ T35] kthread+0x164/0x180
> > > [ 578.773430][ T35] ret_from_fork+0x10/0x20
> > > [ 578.775253][ T35] Code: 52800021 91001263 14000388 36f80040 (d4210000)
> > > [ 578.777990][ T35] ---[ end trace 0000000000000000 ]---
> > > [ 578.778021][ T35] Kernel panic - not syncing: Oops - BUG: Fatal exception
> > > [ 578.782934][ T35] Kernel Offset: disabled
> > > [ 578.784642][ T35] CPU features: 0x000,00100010,00001086
> > > [ 578.786848][ T35] Memory Limit: none
> > > [ 578.788433][ T35] ---[ end Kernel panic - not syncing: Oops - BUG: Fatal exception ]---
Now I see this oops on the mainline kernel too when I'm building and booting an
arm64 allmodconfig kernel, sha
9d004b2f4fea ("Merge tag 'cxl-for-5.19' of
git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl").
building and booting an arm64 allmodconfig kernel.
When I revert 42b2547137f5 ("arm64/mm: enable
ARCH_SUPPORTS_PAGE_TABLE_CHECK") I'm able to boot.
The kernel boots fine.
This is the config [1].
Building with tuxmake[2].
$ tuxmake --runtime podman --target-arch arm64 --toolchain gcc-11
--kconfig http://ix.io/3YPH
Cheers,
Anders
[1] http://ix.io/3YPH
> > >
> > > Bisected down to this patch, see the bisect log [2].
> > >
> > > When I revert this patch I don't see the issue anymore.
> >
> > Thank you for reporting this. I believe, this is the same problem that
> > Anshuman saw [1].
>
> Yes looks like the same issue, I missed that.
>
> > However, at that time he could not reproduce it
> > anymore. Can you please provide QEMU command line
>
> $ qemu-system-aarch64 --enable-kvm -cpu cortex-a53 -kernel
> Image-20220517-1.gz -serial stdio -monitor none -nographic -m 2G -M
> virt -fsdev local,id=root,path=/srv/kvm/tmp/stretch/arm64,security_model=none,writeout=immediate
> -device virtio-rng-pci -device
> virtio-9p-pci,fsdev=root,mount_tag=/dev/root -append "root=/dev/root
> rootfstype=9p rootflags=trans=virtio,msize=131072
> console=ttyAMA0,38400n8 earlycon=pl011,0x9000000 initcall_debug
> softlockup_panic=0 security=none kpti=no kfence.sample_interval=0"
> -object rng-random,id=rng0,filename=/dev/urandom -device
> virtio-rng-pci,rng=rng0
>
> >, QEMU version,
>
> $ qemu-system-aarch64 --version
> QEMU emulator version 5.2.0 (Debian 1:5.2+dfsg-11+deb11u1)
> Copyright (c) 2003-2020 Fabrice Bellard and the QEMU Project developers
>
> I'm running on an arm64 host.
>
> > kernel config
>
> Kernel config [1].
>
> I build the kernel with tuxmake [2] like this:
> $ tuxmake --runtime podman --target-arch arm64 --toolchain gcc-11
> --kconfig http://ix.io/3Y06
>
> >, and information about the base image you are using.
>
> Using a debian:stretch release when booting up QEMU.
> Host system debian:bullseye
>
> Cheers,
> Anders
> [1] http://ix.io/3Y06
> [2] https://tuxmake.org/install-pypi/
>
> >
> > Thank you,
> > Pasha
> >
> > [1] https://lore.kernel.org/all/1c314feb-cd78-2bb3-462e-4ea3cefe122e@arm.com/
> >
> >
> > >
> > > Cheers,
> > > Anders
> > > [1] https://people.linaro.org/~anders.roxell/output-next-20220513.log
> > > [2] http://ix.io/3XZB
> > >
> > > > ---
> > > > arch/arm64/Kconfig | 1 +
> > > > arch/arm64/include/asm/pgtable.h | 61 ++++++++++++++++++++++++++++----
> > > > 2 files changed, 56 insertions(+), 6 deletions(-)
> > > >
> > > > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > > > index 272c75af9302..3055fb5b3fb4 100644
> > > > --- a/arch/arm64/Kconfig
> > > > +++ b/arch/arm64/Kconfig
> > > > @@ -92,6 +92,7 @@ config ARM64
> > > > select ARCH_SUPPORTS_ATOMIC_RMW
> > > > select ARCH_SUPPORTS_INT128 if CC_HAS_INT128
> > > > select ARCH_SUPPORTS_NUMA_BALANCING
> > > > + select ARCH_SUPPORTS_PAGE_TABLE_CHECK
> > > > select ARCH_WANT_COMPAT_IPC_PARSE_VERSION if COMPAT
> > > > select ARCH_WANT_DEFAULT_BPF_JIT
> > > > select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT
> > > > diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
> > > > index 8ebf1cec5d90..4e61cde27f9f 100644
> > > > --- a/arch/arm64/include/asm/pgtable.h
> > > > +++ b/arch/arm64/include/asm/pgtable.h
> > > > @@ -33,6 +33,7 @@
> > > > #include <linux/mmdebug.h>
> > > > #include <linux/mm_types.h>
> > > > #include <linux/sched.h>
> > > > +#include <linux/page_table_check.h>
> > > >
> > > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > > > #define __HAVE_ARCH_FLUSH_PMD_TLB_RANGE
> > > > @@ -96,6 +97,7 @@ static inline pteval_t __phys_to_pte_val(phys_addr_t phys)
> > > > #define pte_young(pte) (!!(pte_val(pte) & PTE_AF))
> > > > #define pte_special(pte) (!!(pte_val(pte) & PTE_SPECIAL))
> > > > #define pte_write(pte) (!!(pte_val(pte) & PTE_WRITE))
> > > > +#define pte_user(pte) (!!(pte_val(pte) & PTE_USER))
> > > > #define pte_user_exec(pte) (!(pte_val(pte) & PTE_UXN))
> > > > #define pte_cont(pte) (!!(pte_val(pte) & PTE_CONT))
> > > > #define pte_devmap(pte) (!!(pte_val(pte) & PTE_DEVMAP))
> > > > @@ -312,8 +314,8 @@ static inline void __check_racy_pte_update(struct mm_struct *mm, pte_t *ptep,
> > > > __func__, pte_val(old_pte), pte_val(pte));
> > > > }
> > > >
> > > > -static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
> > > > - pte_t *ptep, pte_t pte)
> > > > +static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
> > > > + pte_t *ptep, pte_t pte)
> > > > {
> > > > if (pte_present(pte) && pte_user_exec(pte) && !pte_special(pte))
> > > > __sync_icache_dcache(pte);
> > > > @@ -343,6 +345,13 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
> > > > set_pte(ptep, pte);
> > > > }
> > > >
> > > > +static inline void set_pte_at(struct mm_struct *mm, unsigned long addr,
> > > > + pte_t *ptep, pte_t pte)
> > > > +{
> > > > + page_table_check_pte_set(mm, addr, ptep, pte);
> > > > + return __set_pte_at(mm, addr, ptep, pte);
> > > > +}
> > > > +
> > > > /*
> > > > * Huge pte definitions.
> > > > */
> > > > @@ -454,6 +463,8 @@ static inline int pmd_trans_huge(pmd_t pmd)
> > > > #define pmd_dirty(pmd) pte_dirty(pmd_pte(pmd))
> > > > #define pmd_young(pmd) pte_young(pmd_pte(pmd))
> > > > #define pmd_valid(pmd) pte_valid(pmd_pte(pmd))
> > > > +#define pmd_user(pmd) pte_user(pmd_pte(pmd))
> > > > +#define pmd_user_exec(pmd) pte_user_exec(pmd_pte(pmd))
> > > > #define pmd_cont(pmd) pte_cont(pmd_pte(pmd))
> > > > #define pmd_wrprotect(pmd) pte_pmd(pte_wrprotect(pmd_pte(pmd)))
> > > > #define pmd_mkold(pmd) pte_pmd(pte_mkold(pmd_pte(pmd)))
> > > > @@ -501,8 +512,19 @@ static inline pmd_t pmd_mkdevmap(pmd_t pmd)
> > > > #define pud_pfn(pud) ((__pud_to_phys(pud) & PUD_MASK) >> PAGE_SHIFT)
> > > > #define pfn_pud(pfn,prot) __pud(__phys_to_pud_val((phys_addr_t)(pfn) << PAGE_SHIFT) | pgprot_val(prot))
> > > >
> > > > -#define set_pmd_at(mm, addr, pmdp, pmd) set_pte_at(mm, addr, (pte_t *)pmdp, pmd_pte(pmd))
> > > > -#define set_pud_at(mm, addr, pudp, pud) set_pte_at(mm, addr, (pte_t *)pudp, pud_pte(pud))
> > > > +static inline void set_pmd_at(struct mm_struct *mm, unsigned long addr,
> > > > + pmd_t *pmdp, pmd_t pmd)
> > > > +{
> > > > + page_table_check_pmd_set(mm, addr, pmdp, pmd);
> > > > + return __set_pte_at(mm, addr, (pte_t *)pmdp, pmd_pte(pmd));
> > > > +}
> > > > +
> > > > +static inline void set_pud_at(struct mm_struct *mm, unsigned long addr,
> > > > + pud_t *pudp, pud_t pud)
> > > > +{
> > > > + page_table_check_pud_set(mm, addr, pudp, pud);
> > > > + return __set_pte_at(mm, addr, (pte_t *)pudp, pud_pte(pud));
> > > > +}
> > > >
> > > > #define __p4d_to_phys(p4d) __pte_to_phys(p4d_pte(p4d))
> > > > #define __phys_to_p4d_val(phys) __phys_to_pte_val(phys)
> > > > @@ -643,6 +665,24 @@ static inline unsigned long pmd_page_vaddr(pmd_t pmd)
> > > > #define pud_present(pud) pte_present(pud_pte(pud))
> > > > #define pud_leaf(pud) (pud_present(pud) && !pud_table(pud))
> > > > #define pud_valid(pud) pte_valid(pud_pte(pud))
> > > > +#define pud_user(pud) pte_user(pud_pte(pud))
> > > > +
> > > > +#ifdef CONFIG_PAGE_TABLE_CHECK
> > > > +static inline bool pte_user_accessible_page(pte_t pte)
> > > > +{
> > > > + return pte_present(pte) && (pte_user(pte) || pte_user_exec(pte));
> > > > +}
> > > > +
> > > > +static inline bool pmd_user_accessible_page(pmd_t pmd)
> > > > +{
> > > > + return pmd_present(pmd) && (pmd_user(pmd) || pmd_user_exec(pmd));
> > > > +}
> > > > +
> > > > +static inline bool pud_user_accessible_page(pud_t pud)
> > > > +{
> > > > + return pud_present(pud) && pud_user(pud);
> > > > +}
> > > > +#endif
> > > >
> > > > static inline void set_pud(pud_t *pudp, pud_t pud)
> > > > {
> > > > @@ -876,7 +916,11 @@ static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma,
> > > > static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
> > > > unsigned long address, pte_t *ptep)
> > > > {
> > > > - return __pte(xchg_relaxed(&pte_val(*ptep), 0));
> > > > + pte_t pte = __pte(xchg_relaxed(&pte_val(*ptep), 0));
> > > > +
> > > > + page_table_check_pte_clear(mm, address, pte);
> > > > +
> > > > + return pte;
> > > > }
> > > >
> > > > #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> > > > @@ -884,7 +928,11 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm,
> > > > static inline pmd_t pmdp_huge_get_and_clear(struct mm_struct *mm,
> > > > unsigned long address, pmd_t *pmdp)
> > > > {
> > > > - return pte_pmd(ptep_get_and_clear(mm, address, (pte_t *)pmdp));
> > > > + pmd_t pmd = __pmd(xchg_relaxed(&pmd_val(*pmdp), 0));
> > > > +
> > > > + page_table_check_pmd_clear(mm, address, pmd);
> > > > +
> > > > + return pmd;
> > > > }
> > > > #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
> > > >
> > > > @@ -918,6 +966,7 @@ static inline void pmdp_set_wrprotect(struct mm_struct *mm,
> > > > static inline pmd_t pmdp_establish(struct vm_area_struct *vma,
> > > > unsigned long address, pmd_t *pmdp, pmd_t pmd)
> > > > {
> > > > + page_table_check_pmd_set(vma->vm_mm, address, pmdp, pmd);
> > > > return __pmd(xchg_relaxed(&pmd_val(*pmdp), pmd_val(pmd)));
> > > > }
> > > > #endif
> > > > --
> > > > 2.25.1
> > > >
> > >
> > > --
> > > Anders Roxell
> > > anders.roxell at linaro.org
> > > M: +46 708 22 71 05 | IRC: roxell
More information about the linux-riscv
mailing list