The driver should use the pci_resource_len() to get the actual length of
pci bar, and compare it with the expect value. If the bar size is too
small (such as a broken device), the driver should return an error.
Signed-off-by: Zheyu Ma <[email protected]>
---
sound/soc/sof/intel/pci-tng.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/sound/soc/sof/intel/pci-tng.c b/sound/soc/sof/intel/pci-tng.c
index 6efef225973f..7d502cc3ca80 100644
--- a/sound/soc/sof/intel/pci-tng.c
+++ b/sound/soc/sof/intel/pci-tng.c
@@ -75,7 +75,11 @@ static int tangier_pci_probe(struct snd_sof_dev *sdev)
/* LPE base */
base = pci_resource_start(pci, desc->resindex_lpe_base) - IRAM_OFFSET;
- size = PCI_BAR_SIZE;
+ size = pci_resource_len(pci, desc->resindex_lpe_base);
+ if (size < PCI_BAR_SIZE) {
+ dev_err(sdev->dev, "error: I/O region is too small.\n");
+ return -ENODEV;
+ }
dev_dbg(sdev->dev, "LPE PHY base at 0x%x size 0x%x", base, size);
sdev->bar[DSP_BAR] = devm_ioremap(sdev->dev, base, size);
--
2.25.1
On Tue, Apr 12, 2022 at 12:23 AM Pierre-Louis Bossart
<[email protected]> wrote:
>
>
>
> On 4/9/22 09:39, Zheyu Ma wrote:
> > The driver should use the pci_resource_len() to get the actual length of
> > pci bar, and compare it with the expect value. If the bar size is too
> > small (such as a broken device), the driver should return an error.
> >
> > Signed-off-by: Zheyu Ma <[email protected]>
> > ---
> > sound/soc/sof/intel/pci-tng.c | 6 +++++-
> > 1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/sound/soc/sof/intel/pci-tng.c b/sound/soc/sof/intel/pci-tng.c
> > index 6efef225973f..7d502cc3ca80 100644
> > --- a/sound/soc/sof/intel/pci-tng.c
> > +++ b/sound/soc/sof/intel/pci-tng.c
> > @@ -75,7 +75,11 @@ static int tangier_pci_probe(struct snd_sof_dev *sdev)
> >
> > /* LPE base */
> > base = pci_resource_start(pci, desc->resindex_lpe_base) - IRAM_OFFSET;
> > - size = PCI_BAR_SIZE;
> > + size = pci_resource_len(pci, desc->resindex_lpe_base);
> > + if (size < PCI_BAR_SIZE) {
> > + dev_err(sdev->dev, "error: I/O region is too small.\n");
> > + return -ENODEV;
> > + }
>
> May I ask how you found this issue?
Actually, I tested this driver via fuzzing in a simulated environment
and got a crash. Hence, I try to propose a patch and ask for the help
of maintainers to determine whether this is an issue.
Thanks,
Zheyu Ma
On 4/9/22 09:39, Zheyu Ma wrote:
> The driver should use the pci_resource_len() to get the actual length of
> pci bar, and compare it with the expect value. If the bar size is too
> small (such as a broken device), the driver should return an error.
>
> Signed-off-by: Zheyu Ma <[email protected]>
> ---
> sound/soc/sof/intel/pci-tng.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/sound/soc/sof/intel/pci-tng.c b/sound/soc/sof/intel/pci-tng.c
> index 6efef225973f..7d502cc3ca80 100644
> --- a/sound/soc/sof/intel/pci-tng.c
> +++ b/sound/soc/sof/intel/pci-tng.c
> @@ -75,7 +75,11 @@ static int tangier_pci_probe(struct snd_sof_dev *sdev)
>
> /* LPE base */
> base = pci_resource_start(pci, desc->resindex_lpe_base) - IRAM_OFFSET;
> - size = PCI_BAR_SIZE;
> + size = pci_resource_len(pci, desc->resindex_lpe_base);
> + if (size < PCI_BAR_SIZE) {
> + dev_err(sdev->dev, "error: I/O region is too small.\n");
> + return -ENODEV;
> + }
May I ask how you found this issue?
I am not clear on why there's a patch dedicated for a single device, but the same pattern in hda.c and in the HDaudio legacy driver exists.
>
> dev_dbg(sdev->dev, "LPE PHY base at 0x%x size 0x%x", base, size);
> sdev->bar[DSP_BAR] = devm_ioremap(sdev->dev, base, size);
On Sat, 9 Apr 2022 22:39:50 +0800, Zheyu Ma wrote:
> The driver should use the pci_resource_len() to get the actual length of
> pci bar, and compare it with the expect value. If the bar size is too
> small (such as a broken device), the driver should return an error.
>
>
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next
Thanks!
[1/1] ASoC: SOF: Intel: Check the bar size before remapping
commit: 5947b2726beb61fe7911580f239222ec9c4f6967
All being well this means that it will be integrated into the linux-next
tree (usually sometime in the next 24 hours) and sent to Linus during
the next merge window (or sooner if it is a bug fix), however if
problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing
and review of the tree, please engage with people reporting problems and
send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they
should be sent as incremental updates against current git, existing
patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying
to this mail.
Thanks,
Mark