Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH v2] vmalloc: Fix issues with flush flag
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <20190520200703.15997-1-rick.p.edgecombe@intel.com>
Date:   Mon, 20 May 2019 14:25:21 -0700
Cc:     linux-kernel@vger.kernel.org, peterz@infradead.org,
        sparclinux@vger.kernel.org, linux-mm@kvack.org,
        netdev@vger.kernel.org, dave.hansen@intel.com, namit@vmware.com,
        Meelis Roos <mroos@linux.ee>,
        "David S. Miller" <davem@davemloft.net>,
        Borislav Petkov <bp@alien8.de>,
        Andy Lutomirski <luto@kernel.org>,
        Ingo Molnar <mingo@redhat.com>
Content-Transfer-Encoding: quoted-printable
Message-Id: <28F28A46-C57B-483A-A5CB-8BEA06AF15F8@amacapital.net>
References: <20190520200703.15997-1-rick.p.edgecombe@intel.com>
To:     Rick Edgecombe <rick.p.edgecombe@intel.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On May 20, 2019, at 1:07 PM, Rick Edgecombe <rick.p.edgecombe@intel.com> w=
rote:
>=20
> Switch VM_FLUSH_RESET_PERMS to use a regular TLB flush intead of
> vm_unmap_aliases() and fix calculation of the direct map for the
> CONFIG_ARCH_HAS_SET_DIRECT_MAP case.
>=20
> Meelis Roos reported issues with the new VM_FLUSH_RESET_PERMS flag on a
> sparc machine. On investigation some issues were noticed:
>=20

Can you split this into a few (3?) patches, each fixing one issue?

> 1. The calculation of the direct map address range to flush was wrong.
> This could cause problems on x86 if a RO direct map alias ever got loaded
> into the TLB. This shouldn't normally happen, but it could cause the
> permissions to remain RO on the direct map alias, and then the page
> would return from the page allocator to some other component as RO and
> cause a crash.
>=20
> 2. Calling vm_unmap_alias() on vfree could potentially be a lot of work to=

> do on a free operation. Simply flushing the TLB instead of the whole
> vm_unmap_alias() operation makes the frees faster and pushes the heavy
> work to happen on allocation where it would be more expected.
> In addition to the extra work, vm_unmap_alias() takes some locks including=

> a long hold of vmap_purge_lock, which will make all other
> VM_FLUSH_RESET_PERMS vfrees wait while the purge operation happens.
>=20
> 3. page_address() can have locking on some configurations, so skip calling=

> this when possible to further speed this up.
>=20
> Fixes: 868b104d7379 ("mm/vmalloc: Add flag for freeing of special permsiss=
ions")
> Reported-by: Meelis Roos <mroos@linux.ee>
> Cc: Meelis Roos <mroos@linux.ee>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: "David S. Miller" <davem@davemloft.net>
> Cc: Dave Hansen <dave.hansen@intel.com>
> Cc: Borislav Petkov <bp@alien8.de>
> Cc: Andy Lutomirski <luto@kernel.org>
> Cc: Ingo Molnar <mingo@redhat.com>
> Cc: Nadav Amit <namit@vmware.com>
> Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
> ---
>=20
> Changes since v1:
> - Update commit message with more detail
> - Fix flush end range on !CONFIG_ARCH_HAS_SET_DIRECT_MAP case
>=20
> mm/vmalloc.c | 23 +++++++++++++----------
> 1 file changed, 13 insertions(+), 10 deletions(-)
>=20
> diff --git a/mm/vmalloc.c b/mm/vmalloc.c
> index c42872ed82ac..8d03427626dc 100644
> --- a/mm/vmalloc.c
> +++ b/mm/vmalloc.c
> @@ -2122,9 +2122,10 @@ static inline void set_area_direct_map(const struct=
 vm_struct *area,
> /* Handle removing and resetting vm mappings related to the vm_struct. */
> static void vm_remove_mappings(struct vm_struct *area, int deallocate_page=
s)
> {
> +    const bool has_set_direct =3D IS_ENABLED(CONFIG_ARCH_HAS_SET_DIRECT_M=
AP);
> +    const bool flush_reset =3D area->flags & VM_FLUSH_RESET_PERMS;
>    unsigned long addr =3D (unsigned long)area->addr;
> -    unsigned long start =3D ULONG_MAX, end =3D 0;
> -    int flush_reset =3D area->flags & VM_FLUSH_RESET_PERMS;
> +    unsigned long start =3D addr, end =3D addr + area->size;
>    int i;
>=20
>    /*
> @@ -2133,7 +2134,7 @@ static void vm_remove_mappings(struct vm_struct *are=
a, int deallocate_pages)
>     * This is concerned with resetting the direct map any an vm alias with=

>     * execute permissions, without leaving a RW+X window.
>     */
> -    if (flush_reset && !IS_ENABLED(CONFIG_ARCH_HAS_SET_DIRECT_MAP)) {
> +    if (flush_reset && !has_set_direct) {
>        set_memory_nx(addr, area->nr_pages);
>        set_memory_rw(addr, area->nr_pages);
>    }
> @@ -2146,22 +2147,24 @@ static void vm_remove_mappings(struct vm_struct *a=
rea, int deallocate_pages)
>=20
>    /*
>     * If not deallocating pages, just do the flush of the VM area and
> -     * return.
> +     * return. If the arch doesn't have set_direct_map_(), also skip the
> +     * below work.
>     */
> -    if (!deallocate_pages) {
> -        vm_unmap_aliases();
> +    if (!deallocate_pages || !has_set_direct) {
> +        flush_tlb_kernel_range(start, end);
>        return;
>    }
>=20
>    /*
>     * If execution gets here, flush the vm mapping and reset the direct
>     * map. Find the start and end range of the direct mappings to make sur=
e
> -     * the vm_unmap_aliases() flush includes the direct map.
> +     * the flush_tlb_kernel_range() includes the direct map.
>     */
>    for (i =3D 0; i < area->nr_pages; i++) {
> -        if (page_address(area->pages[i])) {
> +        addr =3D (unsigned long)page_address(area->pages[i]);
> +        if (addr) {
>            start =3D min(addr, start);
> -            end =3D max(addr, end);
> +            end =3D max(addr + PAGE_SIZE, end);
>        }
>    }
>=20
> @@ -2171,7 +2174,7 @@ static void vm_remove_mappings(struct vm_struct *are=
a, int deallocate_pages)
>     * reset the direct map permissions to the default.
>     */
>    set_area_direct_map(area, set_direct_map_invalid_noflush);
> -    _vm_unmap_aliases(start, end, 1);
> +    flush_tlb_kernel_range(start, end);
>    set_area_direct_map(area, set_direct_map_default_noflush);
> }
>=20
> --=20
> 2.20.1
>=20