diff mbox series

mm: huge_memory: use !CONFIG_64BIT to relax huge page alignment on 32 bit machines

Message ID 20240712155855.1130330-1-yang@os.amperecomputing.com (mailing list archive)
State New
Headers show
Series mm: huge_memory: use !CONFIG_64BIT to relax huge page alignment on 32 bit machines | expand

Commit Message

Yang Shi July 12, 2024, 3:58 p.m. UTC
Yves-Alexis Perez reported commit 4ef9ad19e176 ("mm: huge_memory: don't
force huge page alignment on 32 bit") didn't work for x86_32 [1].  It is
because x86_32 uses CONFIG_X86_32 instead of CONFIG_32BIT.

!CONFIG_64BIT should cover all 32 bit machines.

[1] https://lore.kernel.org/linux-mm/CAHbLzkr1LwH3pcTgM+aGQ31ip2bKqiqEQ8=FQB+t2c3dhNKNHA@mail.gmail.com/

Fixes: 4ef9ad19e176 ("mm: huge_memory: don't force huge page alignment on 32 bit")
Reported-by: Yves-Alexis Perez <corsac@debian.org>
Tested-By: Yves-Alexis Perez <corsac@debian.org>
Cc: <stable@vger.kernel.org>    [6.8+]
Signed-off-by: Yang Shi <yang@os.amperecomputing.com>
---
 mm/huge_memory.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

David Hildenbrand July 13, 2024, 1:20 a.m. UTC | #1
On 12.07.24 17:58, Yang Shi wrote:
> Yves-Alexis Perez reported commit 4ef9ad19e176 ("mm: huge_memory: don't
> force huge page alignment on 32 bit") didn't work for x86_32 [1].  It is
> because x86_32 uses CONFIG_X86_32 instead of CONFIG_32BIT.
> 
> !CONFIG_64BIT should cover all 32 bit machines.
> 
> [1] https://lore.kernel.org/linux-mm/CAHbLzkr1LwH3pcTgM+aGQ31ip2bKqiqEQ8=FQB+t2c3dhNKNHA@mail.gmail.com/
> 
> Fixes: 4ef9ad19e176 ("mm: huge_memory: don't force huge page alignment on 32 bit")
> Reported-by: Yves-Alexis Perez <corsac@debian.org>
> Tested-By: Yves-Alexis Perez <corsac@debian.org>
> Cc: <stable@vger.kernel.org>    [6.8+]
> Signed-off-by: Yang Shi <yang@os.amperecomputing.com>
> ---
>   mm/huge_memory.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 2120f7478e55..64f00aedf9af 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -857,7 +857,7 @@ static unsigned long __thp_get_unmapped_area(struct file *filp,
>   	loff_t off_align = round_up(off, size);
>   	unsigned long len_pad, ret, off_sub;
>   
> -	if (IS_ENABLED(CONFIG_32BIT) || in_compat_syscall())
> +	if (!IS_ENABLED(CONFIG_64BIT) || in_compat_syscall())
>   		return 0;
>   
>   	if (off_end <= off_align || (off_end - off_align) < size)

Acked-by: David Hildenbrand <david@redhat.com>
Yves-Alexis Perez July 19, 2024, 9:19 a.m. UTC | #2
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Fri, 2024-07-12 at 08:58 -0700, Yang Shi wrote:
> Yves-Alexis Perez reported commit 4ef9ad19e176 ("mm: huge_memory: don't
> force huge page alignment on 32 bit") didn't work for x86_32 [1].  It is
> because x86_32 uses CONFIG_X86_32 instead of CONFIG_32BIT.
> 
> !CONFIG_64BIT should cover all 32 bit machines.

Hi,

I've noticed that the patch was integrated into the -mm tree and next/master.
It's not yet in the first half of the merge window for 6.11 but do you know if
it's scheduled for rc1?

Regards,
- -- 
Yves-Alexis
-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEE8vi34Qgfo83x35gF3rYcyPpXRFsFAmaaL5wACgkQ3rYcyPpX
RFtzIQf8Dn0wqm4ZDeAvoxX+xPTn5Qhu5T1tDZfnryL584DRroe5PyYr5uI+BL7W
oMfPc+cNYUmGRPc8qmbAhg5K18xlfmPfUDq4idWzLMjpUnmEZflWgrGFvRgJmTo5
Kq2MRxpLi38M9GTeqG3MKtWxouKilch9n1ukmQhV9H0DUtiiS3EhITt9X7PbJ7Zc
yvPtdRcdHXRXSl0x/0hztQy4xuKSJxGQULGoSL0HyS8R9OCj+0hDFgEz98YkSbYW
6pZ8+iZLc+OXEkjKfC3uwI/q0Dw3eZW1zIyPn7ahYteisBr+HfKwoGz7UNzw3aB4
/BB3VA+A7/ikJ9NIwjgSlGTmZso73A==
=1puG
-----END PGP SIGNATURE-----
Andrew Morton July 19, 2024, 8:46 p.m. UTC | #3
On Fri, 19 Jul 2024 11:19:24 +0200 Yves-Alexis Perez <corsac@debian.org> wrote:

> On Fri, 2024-07-12 at 08:58 -0700, Yang Shi wrote:
> > Yves-Alexis Perez reported commit 4ef9ad19e176 ("mm: huge_memory: don't
> > force huge page alignment on 32 bit") didn't work for x86_32 [1].  It is
> > because x86_32 uses CONFIG_X86_32 instead of CONFIG_32BIT.
> > 
> > !CONFIG_64BIT should cover all 32 bit machines.
> 
> Hi,
> 
> I've noticed that the patch was integrated into the -mm tree and next/master.
> It's not yet in the first half of the merge window for 6.11 but do you know if
> it's scheduled for rc1?

I put it in the mm-hotfixes branch since it's cc:stable.  I expect to 
send the hotfixes to Linus next week.
diff mbox series

Patch

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 2120f7478e55..64f00aedf9af 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -857,7 +857,7 @@  static unsigned long __thp_get_unmapped_area(struct file *filp,
 	loff_t off_align = round_up(off, size);
 	unsigned long len_pad, ret, off_sub;
 
-	if (IS_ENABLED(CONFIG_32BIT) || in_compat_syscall())
+	if (!IS_ENABLED(CONFIG_64BIT) || in_compat_syscall())
 		return 0;
 
 	if (off_end <= off_align || (off_end - off_align) < size)