| Message ID | 20201021235758.59993-1-saeed.mirzamohammadi@oracle.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/1] video: fbdev: fix divide error in fbcon_switch | expand |
Attached the Syzkaller C reproducer. > On Oct 21, 2020, at 4:57 PM, saeed.mirzamohammadi@oracle.com wrote: > > From: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > > This patch fixes the issue due to: > > [ 89.572883] divide_error: 0000 [#1] SMP KASAN PTI > [ 89.572897] CPU: 3 PID: 16083 Comm: repro Not tainted 5.9.0-rc7.20200930.rc1.allarch-19-g3e32d0d.syzk #5 > [ 89.572902] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 > [ 89.572934] RIP: 0010:cirrusfb_check_var+0x84/0x1260 > > The error happens when the pixels value is calculated before performing the sanity checks on bits_per_pixel. > A bits_per_pixel set to zero causes divide by zero error. > > This patch moves the calculation after the sanity check. > > Signed-off-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > Tested-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > --- > drivers/video/fbdev/cirrusfb.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c > index 15a9ee7cd734..a7749101b094 100644 > --- a/drivers/video/fbdev/cirrusfb.c > +++ b/drivers/video/fbdev/cirrusfb.c > @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > { > int yres; > /* memory size in pixels */ > - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; > + unsigned int pixels; > struct cirrusfb_info *cinfo = info->par; > > switch (var->bits_per_pixel) { > @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > return -EINVAL; > } > > + pixels = info->screen_size * 8 / var->bits_per_pixel; > if (var->xres_virtual < var->xres) > var->xres_virtual = var->xres; > /* use highest possible virtual resolution */ > -- > 2.27.0 >
Hi On 22.10.20 01:57, saeed.mirzamohammadi@oracle.com wrote: > From: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > > This patch fixes the issue due to: > > [ 89.572883] divide_error: 0000 [#1] SMP KASAN PTI > [ 89.572897] CPU: 3 PID: 16083 Comm: repro Not tainted 5.9.0-rc7.20200930.rc1.allarch-19-g3e32d0d.syzk #5 > [ 89.572902] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 > [ 89.572934] RIP: 0010:cirrusfb_check_var+0x84/0x1260 > > The error happens when the pixels value is calculated before performing the sanity checks on bits_per_pixel. > A bits_per_pixel set to zero causes divide by zero error. > > This patch moves the calculation after the sanity check. > > Signed-off-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > Tested-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> Looks good, thanks a lot. I'll add the patch to drm-misc-next Reviewed-by: Thomas Zimemrmann <tzimmermann@suse.de> Best regards Thomas > --- > drivers/video/fbdev/cirrusfb.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c > index 15a9ee7cd734..a7749101b094 100644 > --- a/drivers/video/fbdev/cirrusfb.c > +++ b/drivers/video/fbdev/cirrusfb.c > @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > { > int yres; > /* memory size in pixels */ > - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; > + unsigned int pixels; > struct cirrusfb_info *cinfo = info->par; > > switch (var->bits_per_pixel) { > @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > return -EINVAL; > } > > + pixels = info->screen_size * 8 / var->bits_per_pixel; > if (var->xres_virtual < var->xres) > var->xres_virtual = var->xres; > /* use highest possible virtual resolution */ >
Hi On 22.10.20 01:57, saeed.mirzamohammadi@oracle.com wrote: > From: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > > This patch fixes the issue due to: > > [ 89.572883] divide_error: 0000 [#1] SMP KASAN PTI > [ 89.572897] CPU: 3 PID: 16083 Comm: repro Not tainted 5.9.0-rc7.20200930.rc1.allarch-19-g3e32d0d.syzk #5 > [ 89.572902] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 > [ 89.572934] RIP: 0010:cirrusfb_check_var+0x84/0x1260 BTW, if you run qemu with cirrus, there's also a DRM driver named cirrus.ko. Might be a better choice than the old fbdev driver. If you just care about qemu, but not the actual graphics device, take a look at https://www.kraxel.org/blog/2014/10/qemu-using-cirrus-considered-harmful/ Anyway, thanks for your patch. Best regards Thomas > > The error happens when the pixels value is calculated before performing the sanity checks on bits_per_pixel. > A bits_per_pixel set to zero causes divide by zero error. > > This patch moves the calculation after the sanity check. > > Signed-off-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > Tested-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > --- > drivers/video/fbdev/cirrusfb.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c > index 15a9ee7cd734..a7749101b094 100644 > --- a/drivers/video/fbdev/cirrusfb.c > +++ b/drivers/video/fbdev/cirrusfb.c > @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > { > int yres; > /* memory size in pixels */ > - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; > + unsigned int pixels; > struct cirrusfb_info *cinfo = info->par; > > switch (var->bits_per_pixel) { > @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, > return -EINVAL; > } > > + pixels = info->screen_size * 8 / var->bits_per_pixel; > if (var->xres_virtual < var->xres) > var->xres_virtual = var->xres; > /* use highest possible virtual resolution */ >
Thanks, adding stable. > On Oct 22, 2020, at 12:34 AM, Thomas Zimmermann <tzimmermann@suse.de> wrote: > > Hi > > On 22.10.20 01:57, saeed.mirzamohammadi@oracle.com <mailto:saeed.mirzamohammadi@oracle.com> wrote: >> From: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com <mailto:saeed.mirzamohammadi@oracle.com>> >> >> This patch fixes the issue due to: >> >> [ 89.572883] divide_error: 0000 [#1] SMP KASAN PTI >> [ 89.572897] CPU: 3 PID: 16083 Comm: repro Not tainted 5.9.0-rc7.20200930.rc1.allarch-19-g3e32d0d.syzk #5 >> [ 89.572902] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 >> [ 89.572934] RIP: 0010:cirrusfb_check_var+0x84/0x1260 > > BTW, if you run qemu with cirrus, there's also a DRM driver named > cirrus.ko. Might be a better choice than the old fbdev driver. If you > just care about qemu, but not the actual graphics device, take a look at > > https://urldefense.com/v3/__https://www.kraxel.org/blog/2014/10/qemu-using-cirrus-considered-harmful/__;!!GqivPVa7Brio!LmgeM-pVBVH80uVELF1P1nBGAbAlhvnxKKE_ZrEc9d76AznvAAgP1FAp3_zNa2frKaIUZteK$ <https://urldefense.com/v3/__https://www.kraxel.org/blog/2014/10/qemu-using-cirrus-considered-harmful/__;!!GqivPVa7Brio!LmgeM-pVBVH80uVELF1P1nBGAbAlhvnxKKE_ZrEc9d76AznvAAgP1FAp3_zNa2frKaIUZteK$> > > Anyway, thanks for your patch. > > Best regards > Thomas > >> >> The error happens when the pixels value is calculated before performing the sanity checks on bits_per_pixel. >> A bits_per_pixel set to zero causes divide by zero error. >> >> This patch moves the calculation after the sanity check. >> >> Signed-off-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> Tested-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> --- >> drivers/video/fbdev/cirrusfb.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c >> index 15a9ee7cd734..a7749101b094 100644 >> --- a/drivers/video/fbdev/cirrusfb.c >> +++ b/drivers/video/fbdev/cirrusfb.c >> @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, >> { >> int yres; >> /* memory size in pixels */ >> - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; >> + unsigned int pixels; >> struct cirrusfb_info *cinfo = info->par; >> >> switch (var->bits_per_pixel) { >> @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, >> return -EINVAL; >> } >> >> + pixels = info->screen_size * 8 / var->bits_per_pixel; >> if (var->xres_virtual < var->xres) >> var->xres_virtual = var->xres; >> /* use highest possible virtual resolution */ >> > > -- > Thomas Zimmermann > Graphics Driver Developer > SUSE Software Solutions Germany GmbH > Maxfeldstr. 5, 90409 Nürnberg, Germany > (HRB 36809, AG Nürnberg) > Geschäftsführer: Felix Imendörffer
Thanks, adding stable. Saeed > On Oct 22, 2020, at 12:34 AM, Thomas Zimmermann <tzimmermann@suse.de> wrote: > > Hi > > On 22.10.20 01:57, saeed.mirzamohammadi@oracle.com wrote: >> From: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> >> This patch fixes the issue due to: >> >> [ 89.572883] divide_error: 0000 [#1] SMP KASAN PTI >> [ 89.572897] CPU: 3 PID: 16083 Comm: repro Not tainted 5.9.0-rc7.20200930.rc1.allarch-19-g3e32d0d.syzk #5 >> [ 89.572902] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.5.1 01/01/2011 >> [ 89.572934] RIP: 0010:cirrusfb_check_var+0x84/0x1260 > > BTW, if you run qemu with cirrus, there's also a DRM driver named > cirrus.ko. Might be a better choice than the old fbdev driver. If you > just care about qemu, but not the actual graphics device, take a look at > > https://urldefense.com/v3/__https://www.kraxel.org/blog/2014/10/qemu-using-cirrus-considered-harmful/__;!!GqivPVa7Brio!LmgeM-pVBVH80uVELF1P1nBGAbAlhvnxKKE_ZrEc9d76AznvAAgP1FAp3_zNa2frKaIUZteK$ > > Anyway, thanks for your patch. > > Best regards > Thomas > >> >> The error happens when the pixels value is calculated before performing the sanity checks on bits_per_pixel. >> A bits_per_pixel set to zero causes divide by zero error. >> >> This patch moves the calculation after the sanity check. >> >> Signed-off-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> Tested-by: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> --- >> drivers/video/fbdev/cirrusfb.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c >> index 15a9ee7cd734..a7749101b094 100644 >> --- a/drivers/video/fbdev/cirrusfb.c >> +++ b/drivers/video/fbdev/cirrusfb.c >> @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, >> { >> int yres; >> /* memory size in pixels */ >> - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; >> + unsigned int pixels; >> struct cirrusfb_info *cinfo = info->par; >> >> switch (var->bits_per_pixel) { >> @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, >> return -EINVAL; >> } >> >> + pixels = info->screen_size * 8 / var->bits_per_pixel; >> if (var->xres_virtual < var->xres) >> var->xres_virtual = var->xres; >> /* use highest possible virtual resolution */ >> > > -- > Thomas Zimmermann > Graphics Driver Developer > SUSE Software Solutions Germany GmbH > Maxfeldstr. 5, 90409 Nürnberg, Germany > (HRB 36809, AG Nürnberg) > Geschäftsführer: Felix Imendörffer
On Mon, Oct 26, 2020 at 10:00:11AM -0700, Saeed Mirzamohammadi wrote:
> Thanks, adding stable.
Why? What are we supposed to do with this?
Hi Greg,
Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8. I’ll be more explicit.
Commit: cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3 ("video: fbdev: fix divide error in fbcon_pswitch”)
Thanks,
Saeed
> On Oct 26, 2020, at 11:22 PM, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Mon, Oct 26, 2020 at 10:00:11AM -0700, Saeed Mirzamohammadi wrote:
>> Thanks, adding stable.
>
> Why? What are we supposed to do with this?
Hi Greg,
Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8. I’ll be more explicit.
Commit: cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3 ("video: fbdev: fix divide error in fbcon_pswitch”)
Thanks,
Saeed
> On Oct 26, 2020, at 11:22 PM, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Mon, Oct 26, 2020 at 10:00:11AM -0700, Saeed Mirzamohammadi wrote:
>> Thanks, adding stable.
>
> Why? What are we supposed to do with this?
Hi Greg,
Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8.
commit cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3
Author: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com>
Date: Wed Oct 21 16:57:58 2020 -0700
video: fbdev: fix divide error in fbcon_switch
Thanks,
Saeed
> On Oct 26, 2020, at 11:22 PM, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Mon, Oct 26, 2020 at 10:00:11AM -0700, Saeed Mirzamohammadi wrote:
>> Thanks, adding stable.
>
> Why? What are we supposed to do with this?
On Tue, Oct 27, 2020 at 10:12:49AM -0700, Saeed Mirzamohammadi wrote: > Hi Greg, > > Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8. > commit cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3 > Author: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > Date: Wed Oct 21 16:57:58 2020 -0700 > > video: fbdev: fix divide error in fbcon_switch I do not see that commit in Linus's tree, do you? confused, greg k-h
On Thu, 29 Oct 2020, Greg KH <gregkh@linuxfoundation.org> wrote: > On Tue, Oct 27, 2020 at 10:12:49AM -0700, Saeed Mirzamohammadi wrote: >> Hi Greg, >> >> Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8. >> commit cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3 >> Author: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> >> Date: Wed Oct 21 16:57:58 2020 -0700 >> >> video: fbdev: fix divide error in fbcon_switch > > I do not see that commit in Linus's tree, do you? It's in drm-misc-next, IIUC heading for Linus' tree in the next merge window in 6-8 weeks. Which is to say this should probably have been applied to drm-misc-fixes branch heading for v5.10-rcX, with a Cc: stable tag, to begin with. BR, Jani.
On Thu, Oct 29, 2020 at 01:13:01PM +0200, Jani Nikula wrote: > On Thu, 29 Oct 2020, Greg KH <gregkh@linuxfoundation.org> wrote: > > On Tue, Oct 27, 2020 at 10:12:49AM -0700, Saeed Mirzamohammadi wrote: > >> Hi Greg, > >> > >> Sorry for the confusion. I’m requesting stable maintainers to cherry-pick this patch into stable 5.4 and 5.8. > >> commit cc07057c7c88fb8eff3b1991131ded0f0bcfa7e3 > >> Author: Saeed Mirzamohammadi <saeed.mirzamohammadi@oracle.com> > >> Date: Wed Oct 21 16:57:58 2020 -0700 > >> > >> video: fbdev: fix divide error in fbcon_switch > > > > I do not see that commit in Linus's tree, do you? > > It's in drm-misc-next, IIUC heading for Linus' tree in the next merge > window in 6-8 weeks. Which is to say this should probably have been > applied to drm-misc-fixes branch heading for v5.10-rcX, with a Cc: > stable tag, to begin with. Ok, nothing I can do with this now, please email stable@vger.kernel.org when it hits Linus's tree and we can take it then. Saeed, please read: https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html for how to do this properly. thanks, greg k-h
diff --git a/drivers/video/fbdev/cirrusfb.c b/drivers/video/fbdev/cirrusfb.c index 15a9ee7cd734..a7749101b094 100644 --- a/drivers/video/fbdev/cirrusfb.c +++ b/drivers/video/fbdev/cirrusfb.c @@ -531,7 +531,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, { int yres; /* memory size in pixels */ - unsigned pixels = info->screen_size * 8 / var->bits_per_pixel; + unsigned int pixels; struct cirrusfb_info *cinfo = info->par; switch (var->bits_per_pixel) { @@ -573,6 +573,7 @@ static int cirrusfb_check_var(struct fb_var_screeninfo *var, return -EINVAL; } + pixels = info->screen_size * 8 / var->bits_per_pixel; if (var->xres_virtual < var->xres) var->xres_virtual = var->xres; /* use highest possible virtual resolution */