net: axienet: fix a signedness bug in probe
Subsystems (5)
PATCH 1/1 net: axienet: fix a signedness bug in probe
The "lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.
--- drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1 -1
@@ -1762,7 +1762,7 @@ static int axienet_probe(struct platform_device *pdev) } }else { lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) { ret = -EINVAL ; goto free_netdev; }
Signature · 1 line
2.20.1
Hi, Dan
drivers/net/ethernet/xilinx/xilinx_axienet_main.c static int axienet_probe(struct platform_device *pdev) · line 1765
Quoted 6 lines · Dan Carpenter
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1
-1lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) {
This (almost) exact code appears in a lot of different drivers too, so maybe it'd be nice to review them all and apply the same cast if needed?
Best regards
Signature · 1 line
Alvaro G. M.
David Miller
maintainerFrom: "Alvaro G. M" <[email protected]>
Date: Wed, 25 Sep 2019 13:05:43 +0200
Quoted 6 lines
Hi, Dan
Quoted 3 lines · Dan Carpenter
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.
Quoted 6 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c static int axienet_probe(struct platform_device *pdev) · line 1765
Quoted 3 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1
-1lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) {This (almost) exact code appears in a lot of different drivers too, so maybe it'd be nice to review them all and apply the same cast if needed?
Or make the thing an int if negative values are never valid 32-bit phy_mode values anyways.
Quoted 10 lines
-----Original Message-----
From: Dan Carpenter <[email protected]>
Sent: Wednesday, September 25, 2019 4:29 PM
To: Radhey Shyam Pandey <[email protected]>; Alvaro G. M <[email protected]>
Cc: David S. Miller <[email protected]>; Michal Simek <[email protected]>; Russell King <[email protected]>; [email protected]; [email protected]
Subject: [PATCH net] net: axienet: fix a signedness bug in probe
Quoted 6 lines
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.Fixes ee06b1728b95 ("net: axienet: add support for standard phy-modebinding")
Signed-off-by Dan Carpenter <[email protected]>
Quoted 21 lines
--- drivers/net/ethernet/xilinx/xilinx_axienet_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1
-1@@ -1762,7 +1762,7 @@ static int axienet_probe(struct platform_device *pdev) } }else { lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) { ret = -EINVAL ; goto free_netdev;}
--
2.20.1
Quoted 16 lines · David Miller
From: "Alvaro G. M" <[email protected]>
Date: Wed, 25 Sep 2019 13:05:43 +0200Quoted 6 lines
Hi, Dan
Quoted 3 lines · Dan Carpenter
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.Quoted 6 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c static int axienet_probe(struct platform_device *pdev) · line 1765
Quoted 3 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1
-1lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) {This (almost) exact code appears in a lot of different drivers too, so maybe it'd be nice to review them all and apply the same cast if needed?
Quoted 2 lines
Or make the thing an int if negative values are never valid 32-bit
phy_modevalues anyways.
Maybe we should change the API
int of_get_phy_mode(struct device_node *np, phy_interface_t *phy_mode);
Separate the error from the value we are getting.
Andrew
Quoted 6 lines · Alvaro G. M
Hi, Dan
Quoted 3 lines · Dan Carpenter
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.
Quoted 7 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c static int axienet_probe(struct platform_device *pdev) · line 1765
Quoted 3 lines
drivers/net/ethernet/xilinx/xilinx_axienet_main.c +1
-1lp->phy_mode =of_get_phy_mode (pdev->dev.of_node); -if (lp->phy_mode <0 ) { +if ((int )lp->phy_mode <0 ) {This (almost) exact code appears in a lot of different drivers too, so maybe it'd be nice to review them all and apply the same cast if needed?
This is a new warning in Smatch. I did send patches for the whole kernel. We won't get these bugs in the future because people run Smatch on the kernel and will find the bugs. All the bugs were from 2017 or later which suggests that someone cleared these out two years ago but soon the 0-day bot will warn about issues so they will get fixed quicker.
I'm sort of out of it today...
The get_phy_mode() function seem like they lend themselves to creating these bugs. The ->phy_mode variables tend to be declared in the driver so it would require quite a few patches to make them all int and I'm not sure that's more beautiful. Andrew Lunn's idea to update the API would probably be a good idea.
I'm going back to bed for now and I'll think about this some more.
regards,
dan carpenter
Quoted 5 lines
The
get_phy_mode()function seem like they lend themselves to creating these bugs. The ->phy_mode variables tend to be declared in the driver so it would require quite a few patches to make them all int and I'm not sure that's more beautiful. Andrew Lunn's idea to update the API would probably be a good idea.
Hi Dan
I started on it. Once net-next has opened, and 0-day has compile tested my changes, i will post the code for review.
Andrew
From: Dan Carpenter <[email protected]>
Date: Wed, 25 Sep 2019 13:59:11 +0300
Quoted 5 lines
The "
lp->phy_mode" is an enum but in this context GCC treats it as an unsigned int so the error handling is never triggered.Fixes ee06b1728b95 ("net: axienet: add support for standard phy-mode binding")Signed-off-by Dan Carpenter <[email protected]>
Applied.