drivers/net/ethernet/8390/ax88796.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
From: Stanislaw Gruszka <stf_xl@wp.pl>
Commit 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra
read clock cycle") added extra 'quirk' field to struct eeprom_93cx6,
which change how the data is read.
Some existing users of eeprom_93cx6, including ax88796 driver, allocate
the structure on the stack without initialization of all the fields.
As a result, the added quirk field has undefined value, what can
randomly cause reading wrong data from the EEPROM.
Fix by using designated initialization.
Fixes: 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra read clock cycle")
Cc: stable@kernel.org # v6.13+
Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
---
drivers/net/ethernet/8390/ax88796.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c
index e1695d0fbd8b..5ee1fbafdcfc 100644
--- a/drivers/net/ethernet/8390/ax88796.c
+++ b/drivers/net/ethernet/8390/ax88796.c
@@ -722,12 +722,12 @@ static int ax_init_dev(struct net_device *dev)
#ifdef CONFIG_AX88796_93CX6
if (ax->plat->flags & AXFLG_HAS_93CX6) {
unsigned char mac_addr[ETH_ALEN];
- struct eeprom_93cx6 eeprom;
-
- eeprom.data = ei_local;
- eeprom.register_read = ax_eeprom_register_read;
- eeprom.register_write = ax_eeprom_register_write;
- eeprom.width = PCI_EEPROM_WIDTH_93C56;
+ struct eeprom_93cx6 eeprom = {
+ .data = ei_local,
+ .register_read = ax_eeprom_register_read,
+ .register_write = ax_eeprom_register_write,
+ .width = PCI_EEPROM_WIDTH_93C56,
+ };
eeprom_93cx6_multiread(&eeprom, 0,
(__le16 __force *)mac_addr,
--
2.50.1
On 24/07/2026 09:34, stf_xl@wp.pl wrote:
> From: Stanislaw Gruszka <stf_xl@wp.pl>
>
> Commit 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra
> read clock cycle") added extra 'quirk' field to struct eeprom_93cx6,
> which change how the data is read.
>
> Some existing users of eeprom_93cx6, including ax88796 driver, allocate
> the structure on the stack without initialization of all the fields.
> As a result, the added quirk field has undefined value, what can
> randomly cause reading wrong data from the EEPROM.
>
> Fix by using designated initialization.
>
> Fixes: 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra read clock cycle")
> Cc: stable@kernel.org # v6.13+
> Signed-off-by: Stanislaw Gruszka <stf_xl@wp.pl>
> ---
> drivers/net/ethernet/8390/ax88796.c | 12 ++++++------
> 1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c
> index e1695d0fbd8b..5ee1fbafdcfc 100644
> --- a/drivers/net/ethernet/8390/ax88796.c
> +++ b/drivers/net/ethernet/8390/ax88796.c
> @@ -722,12 +722,12 @@ static int ax_init_dev(struct net_device *dev)
> #ifdef CONFIG_AX88796_93CX6
> if (ax->plat->flags & AXFLG_HAS_93CX6) {
> unsigned char mac_addr[ETH_ALEN];
> - struct eeprom_93cx6 eeprom;
> -
> - eeprom.data = ei_local;
> - eeprom.register_read = ax_eeprom_register_read;
> - eeprom.register_write = ax_eeprom_register_write;
> - eeprom.width = PCI_EEPROM_WIDTH_93C56;
> + struct eeprom_93cx6 eeprom = {
> + .data = ei_local,
> + .register_read = ax_eeprom_register_read,
> + .register_write = ax_eeprom_register_write,
> + .width = PCI_EEPROM_WIDTH_93C56,
> + };
>
> eeprom_93cx6_multiread(&eeprom, 0,
> (__le16 __force *)mac_addr,
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
On Fri, Jul 24, 2026 at 10:34:11AM +0200, stf_xl@wp.pl wrote:
> From: Stanislaw Gruszka <stf_xl@wp.pl>
>
> Commit 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra
> read clock cycle") added extra 'quirk' field to struct eeprom_93cx6,
> which change how the data is read.
>
> Some existing users of eeprom_93cx6, including ax88796 driver, allocate
> the structure on the stack without initialization of all the fields.
> As a result, the added quirk field has undefined value, what can
> randomly cause reading wrong data from the EEPROM.
>
> Fix by using designated initialization.
The problem with this solution is that the next time another member is
added, it breaks again.
struct eeprom_93cx6 eeprom = {0};
Will initialise all fields to 0, and then you can use the existing
code to set those fields as needed.
Andrew
---
pw-bot: cr
On Fri, Jul 24, 2026 at 02:57:33PM +0200, Andrew Lunn wrote:
> On Fri, Jul 24, 2026 at 10:34:11AM +0200, stf_xl@wp.pl wrote:
> > From: Stanislaw Gruszka <stf_xl@wp.pl>
> >
> > Commit 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra
> > read clock cycle") added extra 'quirk' field to struct eeprom_93cx6,
> > which change how the data is read.
> >
> > Some existing users of eeprom_93cx6, including ax88796 driver, allocate
> > the structure on the stack without initialization of all the fields.
> > As a result, the added quirk field has undefined value, what can
> > randomly cause reading wrong data from the EEPROM.
> >
> > Fix by using designated initialization.
>
> The problem with this solution is that the next time another member is
> added, it breaks again.
I did not initialized 'quirks' explicitly in the patch.
> struct eeprom_93cx6 eeprom = {0};
>
> Will initialise all fields to 0, and then you can use the existing
> code to set those fields as needed.
This form:
struct eeprom_93cx6 eeprom = {
.data = ei_local,
.register_read = ax_eeprom_register_read,
.register_write = ax_eeprom_register_write,
.width = PCI_EEPROM_WIDTH_93C56,
};
will also initialize _all_ remaining fields to zero.
I believe it is guaranteed by C standard.
Regards
Stanislaw
On Fri, 24 Jul 2026 17:04:03 +0200
Stanislaw Gruszka <stf_xl@wp.pl> wrote:
> On Fri, Jul 24, 2026 at 02:57:33PM +0200, Andrew Lunn wrote:
> > On Fri, Jul 24, 2026 at 10:34:11AM +0200, stf_xl@wp.pl wrote:
> > > From: Stanislaw Gruszka <stf_xl@wp.pl>
> > >
> > > Commit 7738a7ab9d12 ("misc: eeprom: eeprom_93cx6: Add quirk for extra
> > > read clock cycle") added extra 'quirk' field to struct eeprom_93cx6,
> > > which change how the data is read.
> > >
> > > Some existing users of eeprom_93cx6, including ax88796 driver, allocate
> > > the structure on the stack without initialization of all the fields.
> > > As a result, the added quirk field has undefined value, what can
> > > randomly cause reading wrong data from the EEPROM.
> > >
> > > Fix by using designated initialization.
> >
> > The problem with this solution is that the next time another member is
> > added, it breaks again.
>
> I did not initialized 'quirks' explicitly in the patch.
>
> > struct eeprom_93cx6 eeprom = {0};
> >
> > Will initialise all fields to 0, and then you can use the existing
> > code to set those fields as needed.
>
> This form:
> struct eeprom_93cx6 eeprom = {
> .data = ei_local,
> .register_read = ax_eeprom_register_read,
> .register_write = ax_eeprom_register_write,
> .width = PCI_EEPROM_WIDTH_93C56,
> };
>
> will also initialize _all_ remaining fields to zero.
>
> I believe it is guaranteed by C standard.
Unless there are unions or padding...
David
>
> Regards
> Stanislaw
>
© 2016 - 2026 Red Hat, Inc.