Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1640057 > unrolled thread

[net-dsa-mv88e6xxx] question about potential use of uninitialized variable

Started by"Gustavo A. R. Silva" <garsilva@embeddedor.com>
First post2017-05-11 23:40 +0200
Last post2017-05-12 18:20 +0200
Articles 5 — 3 participants

Back to article view | Back to linux.kernel

Contents

  [net-dsa-mv88e6xxx] question about potential use of uninitialized  variable "Gustavo A. R. Silva" <garsilva@embeddedor.com> - 2017-05-11 23:40 +0200
    Re: [net-dsa-mv88e6xxx] question about potential use of  uninitialized variable Andrew Lunn <andrew@lunn.ch> - 2017-05-12 04:40 +0200
      Re: [net-dsa-mv88e6xxx] question about potential use of  uninitialized variable "Gustavo A. R. Silva" <garsilva@embeddedor.com> - 2017-05-12 04:50 +0200
        [PATCH] net: dsa: mv88e6xxx: add default case to switch "Gustavo A. R. Silva" <garsilva@embeddedor.com> - 2017-05-12 05:20 +0200
          Re: [PATCH] net: dsa: mv88e6xxx: add default case to switch David Miller <davem@davemloft.net> - 2017-05-12 18:20 +0200

#1640057 — [net-dsa-mv88e6xxx] question about potential use of uninitialized variable

From"Gustavo A. R. Silva" <garsilva@embeddedor.com>
Date2017-05-11 23:40 +0200
Subject[net-dsa-mv88e6xxx] question about potential use of uninitialized variable
Message-ID<tG3Yt-3hN-1@gated-at.bofh.it>
Hello everybody,

While looking into Coverity ID 1398130 I ran into the following piece  
of code at drivers/net/dsa/mv88e6xxx/chip.c:849:

  849static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_chip *chip,
  850                                            struct mv88e6xxx_hw_stat *s,
  851                                            int port, u16 bank1_select,
  852                                            u16 histogram)
  853{
  854        u32 low;
  855        u32 high = 0;
  856        u16 reg = 0;
  857        int err;
  858        u64 value;
  859
  860        switch (s->type) {
  861        case STATS_TYPE_PORT:
  862                err = mv88e6xxx_port_read(chip, port, s->reg, &reg);
  863                if (err)
  864                        return UINT64_MAX;
  865
  866                low = reg;
  867                if (s->sizeof_stat == 4) {
  868                        err = mv88e6xxx_port_read(chip, port,  
s->reg + 1, &reg);
  869                        if (err)
  870                                return UINT64_MAX;
  871                        high = reg;
  872                }
  873                break;
  874        case STATS_TYPE_BANK1:
  875                reg = bank1_select;
  876                /* fall through */
  877        case STATS_TYPE_BANK0:
  878                reg |= s->reg | histogram;
  879                mv88e6xxx_g1_stats_read(chip, reg, &low);
  880                if (s->sizeof_stat == 8)
  881                        mv88e6xxx_g1_stats_read(chip, reg + 1, &high);
  882        }
  883        value = (((u64)high) << 16) | low;
  884        return value;
  885}

My question here is if there is any chance for the execution path to  
directly jump from line 860 to line 883, hence ending up using the  
uninitialized variable _low_?

I'm trying to figure out if this is a false positive or something that  
needs to be fixed.

I'd really appreciate any comment on this.

Thank you!
--
Gustavo A. R. Silva

[toc] | [next] | [standalone]

#1640134 — Re: [net-dsa-mv88e6xxx] question about potential use of uninitialized variable

FromAndrew Lunn <andrew@lunn.ch>
Date2017-05-12 04:40 +0200
SubjectRe: [net-dsa-mv88e6xxx] question about potential use of uninitialized variable
Message-ID<tG8EN-6nn-1@gated-at.bofh.it>
In reply to#1640057
On Thu, May 11, 2017 at 04:35:37PM -0500, Gustavo A. R. Silva wrote:
> 
> Hello everybody,
> 
> While looking into Coverity ID 1398130 I ran into the following
> piece of code at drivers/net/dsa/mv88e6xxx/chip.c:849:
> 
>  849static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_chip *chip,
>  850                                            struct mv88e6xxx_hw_stat *s,
>  851                                            int port, u16 bank1_select,
>  852                                            u16 histogram)
>  853{
>  854        u32 low;
>  855        u32 high = 0;
>  856        u16 reg = 0;
>  857        int err;
>  858        u64 value;
>  859
>  860        switch (s->type) {
>  861        case STATS_TYPE_PORT:
>  862                err = mv88e6xxx_port_read(chip, port, s->reg, &reg);
>  863                if (err)
>  864                        return UINT64_MAX;
>  865
>  866                low = reg;
>  867                if (s->sizeof_stat == 4) {
>  868                        err = mv88e6xxx_port_read(chip, port,
> s->reg + 1, &reg);
>  869                        if (err)
>  870                                return UINT64_MAX;
>  871                        high = reg;
>  872                }
>  873                break;
>  874        case STATS_TYPE_BANK1:
>  875                reg = bank1_select;
>  876                /* fall through */
>  877        case STATS_TYPE_BANK0:
>  878                reg |= s->reg | histogram;
>  879                mv88e6xxx_g1_stats_read(chip, reg, &low);
>  880                if (s->sizeof_stat == 8)
>  881                        mv88e6xxx_g1_stats_read(chip, reg + 1, &high);
>  882        }
>  883        value = (((u64)high) << 16) | low;
>  884        return value;
>  885}
> 
> My question here is if there is any chance for the execution path to
> directly jump from line 860 to line 883, hence ending up using the
> uninitialized variable _low_?

Hi Gustavo

It would require that s->type not have one of the listed case values.
Currently all members of mv88e6xxx_hw_stats due use expected values.
However, it would not hurt to add a

	 default:
		return UINT64_MAX;

Do you want to submit a patch?

   Thanks
	Andrew

[toc] | [prev] | [next] | [standalone]

#1640141 — Re: [net-dsa-mv88e6xxx] question about potential use of uninitialized variable

From"Gustavo A. R. Silva" <garsilva@embeddedor.com>
Date2017-05-12 04:50 +0200
SubjectRe: [net-dsa-mv88e6xxx] question about potential use of uninitialized variable
Message-ID<tG8Ot-6rP-3@gated-at.bofh.it>
In reply to#1640134
Hi Andrew,

Quoting Andrew Lunn <andrew@lunn.ch>:

> On Thu, May 11, 2017 at 04:35:37PM -0500, Gustavo A. R. Silva wrote:
>>
>> Hello everybody,
>>
>> While looking into Coverity ID 1398130 I ran into the following
>> piece of code at drivers/net/dsa/mv88e6xxx/chip.c:849:
>>
>>  849static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_chip *chip,
>>  850                                            struct mv88e6xxx_hw_stat *s,
>>  851                                            int port, u16 bank1_select,
>>  852                                            u16 histogram)
>>  853{
>>  854        u32 low;
>>  855        u32 high = 0;
>>  856        u16 reg = 0;
>>  857        int err;
>>  858        u64 value;
>>  859
>>  860        switch (s->type) {
>>  861        case STATS_TYPE_PORT:
>>  862                err = mv88e6xxx_port_read(chip, port, s->reg, &reg);
>>  863                if (err)
>>  864                        return UINT64_MAX;
>>  865
>>  866                low = reg;
>>  867                if (s->sizeof_stat == 4) {
>>  868                        err = mv88e6xxx_port_read(chip, port,
>> s->reg + 1, &reg);
>>  869                        if (err)
>>  870                                return UINT64_MAX;
>>  871                        high = reg;
>>  872                }
>>  873                break;
>>  874        case STATS_TYPE_BANK1:
>>  875                reg = bank1_select;
>>  876                /* fall through */
>>  877        case STATS_TYPE_BANK0:
>>  878                reg |= s->reg | histogram;
>>  879                mv88e6xxx_g1_stats_read(chip, reg, &low);
>>  880                if (s->sizeof_stat == 8)
>>  881                        mv88e6xxx_g1_stats_read(chip, reg + 1, &high);
>>  882        }
>>  883        value = (((u64)high) << 16) | low;
>>  884        return value;
>>  885}
>>
>> My question here is if there is any chance for the execution path to
>> directly jump from line 860 to line 883, hence ending up using the
>> uninitialized variable _low_?
>
> Hi Gustavo
>
> It would require that s->type not have one of the listed case values.
> Currently all members of mv88e6xxx_hw_stats due use expected values.
> However, it would not hurt to add a
>
> 	 default:
> 		return UINT64_MAX;
>
> Do you want to submit a patch?
>

Sure, I'll send it shortly.

Thanks for clarifying!
--
Gustavo A. R. Silva

[toc] | [prev] | [next] | [standalone]

#1640153 — [PATCH] net: dsa: mv88e6xxx: add default case to switch

From"Gustavo A. R. Silva" <garsilva@embeddedor.com>
Date2017-05-12 05:20 +0200
Subject[PATCH] net: dsa: mv88e6xxx: add default case to switch
Message-ID<tG9hw-6RZ-11@gated-at.bofh.it>
In reply to#1640141
Add default case to switch in order to avoid any chance of using an
uninitialized variable _low_, in case s->type does not match any of
the listed case values.

Addresses-Coverity-ID: 1398130
Suggested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>
---
 drivers/net/dsa/mv88e6xxx/chip.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index 03dc886..d39e210 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -879,6 +879,9 @@ static uint64_t _mv88e6xxx_get_ethtool_stat(struct mv88e6xxx_chip *chip,
 		mv88e6xxx_g1_stats_read(chip, reg, &low);
 		if (s->sizeof_stat == 8)
 			mv88e6xxx_g1_stats_read(chip, reg + 1, &high);
+		break;
+	default:
+		return UINT64_MAX;
 	}
 	value = (((u64)high) << 16) | low;
 	return value;
-- 
2.5.0

[toc] | [prev] | [next] | [standalone]

#1640567 — Re: [PATCH] net: dsa: mv88e6xxx: add default case to switch

FromDavid Miller <davem@davemloft.net>
Date2017-05-12 18:20 +0200
SubjectRe: [PATCH] net: dsa: mv88e6xxx: add default case to switch
Message-ID<tGlsm-76m-25@gated-at.bofh.it>
In reply to#1640153
From: "Gustavo A. R. Silva" <garsilva@embeddedor.com>
Date: Thu, 11 May 2017 22:11:29 -0500

> Add default case to switch in order to avoid any chance of using an
> uninitialized variable _low_, in case s->type does not match any of
> the listed case values.
> 
> Addresses-Coverity-ID: 1398130
> Suggested-by: Andrew Lunn <andrew@lunn.ch>
> Signed-off-by: Gustavo A. R. Silva <garsilva@embeddedor.com>

Applied, thanks.

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web