linux-kernel - Re: [PATCH] fbdev: vga16fb: Request memory region.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <6c565f4c-ef05-45f2-9a82-cbba4a11cc07@gmx.de>
Date: Mon, 27 Oct 2025 21:07:21 +0100
From: Helge Deller <deller@....de>
To: Javier Garcia <rampxxxx@...il.com>
Cc: linux-fbdev@...r.kernel.org, dri-devel@...ts.freedesktop.org,
 linux-kernel@...r.kernel.org, shuah@...nel.org
Subject: Re: [PATCH] fbdev: vga16fb: Request memory region.

On 10/27/25 19:02, Javier Garcia wrote:
> Hi,
> 
> Helge Deller, any comment on this patch?

Well, it doesn't hurt, as it will not abort (as before) even if the region is occupied.
In that sense, I don't see a problem to add it.

But you are missing to release the memory when the driver is released
via vga16fb_remove.

If you send an updated/fixed patch I will apply it to the fbdev tree
to get some testing.

Helge

  
> ---
> Javier Garcia
> 
> 
> On Thu, 16 Oct 2025 at 19:18, Javier Garcia <rampxxxx@...il.com> wrote:
>>
>> This patch reserve and release VGA memory region with `*_mem_region`
>> fn's.
>>
>> This align with Documentation/drm/todo.rst
>> "Request memory regions in all fbdev drivers"
>>
>> I've tested with kernel and qemu both 32bits.
>>
>> Signed-off-by: Javier Garcia <rampxxxx@...il.com>
>> ---
>>
>> When I've run the code always return -EBUSY which makes sense as
>> mem is already requested,`/proc/iomem` shows `000a0000-000bffff : Video RAM area`.
>>
>> I've seen that `cirrusfb` has this kind of code wrapped up with `#if 0`, and I
>> wonder if it makes sense to also wrap up with `#if 0`, at least , in
>> that way the code gets commented about expected behavior.
>>
>>
>>   drivers/video/fbdev/vga16fb.c | 7 +++++++
>>   1 file changed, 7 insertions(+)
>>
>> diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c
>> index eedab14c7d51..f506bf144a97 100644
>> --- a/drivers/video/fbdev/vga16fb.c
>> +++ b/drivers/video/fbdev/vga16fb.c
>> @@ -1319,6 +1319,11 @@ static int vga16fb_probe(struct platform_device *dev)
>>          if (ret)
>>                  return ret;
>>
>> +       if (!request_mem_region(vga16fb_fix.smem_start, vga16fb_fix.smem_len,
>> +                               "vga16b")) {
>> +               dev_err(&dev->dev,"vga16b: cannot reserve video memory at 0x%lx\n",
>> +                      vga16fb_fix.smem_start);
>> +       }
>>          printk(KERN_DEBUG "vga16fb: initializing\n");
>>          info = framebuffer_alloc(sizeof(struct vga16fb_par), &dev->dev);
>>
>> @@ -1398,6 +1403,8 @@ static int vga16fb_probe(struct platform_device *dev)
>>    err_ioremap:
>>          framebuffer_release(info);
>>    err_fb_alloc:
>> +       release_mem_region(vga16fb_fix.smem_start,
>> +                   vga16fb_fix.smem_len);
>>          return ret;
>>   }
>>
>> --
>> 2.50.1
>>