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

Groups > linux.kernel > #1585964 > unrolled thread

(none)

Started by"Pranay Kr. Srivastava" <pranjas@gmail.com>
First post2017-02-22 08:30 +0100
Last post2017-02-24 18:40 +0100
Articles 4 — 3 participants

Back to article view | Back to linux.kernel

Contents

  (none) "Pranay Kr. Srivastava" <pranjas@gmail.com> - 2017-02-22 08:30 +0100
    [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info "Pranay Kr. Srivastava" <pranjas@gmail.com> - 2017-02-22 08:30 +0100
      Re: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info Samuel Thibault <samuel.thibault@ens-lyon.org> - 2017-02-22 11:30 +0100
      Re: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info Greg KH <gregkh@linuxfoundation.org> - 2017-02-24 18:40 +0100

#1585964 — (none)

From"Pranay Kr. Srivastava" <pranjas@gmail.com>
Date2017-02-22 08:30 +0100
Subject(none)
Message-ID<tdzx7-6xa-3@gated-at.bofh.it>
Sending both patches as series instead. Made the required changes as
suggessted in earlier versions and fixed the warnings reported from
kbuild test robot.

[toc] | [next] | [standalone]

#1585967 — [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info

From"Pranay Kr. Srivastava" <pranjas@gmail.com>
Date2017-02-22 08:30 +0100
Subject[PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info
Message-ID<tdzx8-6xa-19@gated-at.bofh.it>
In reply to#1585964
This patch makes spk_set_key_info return -EINVAL
in case of failure instead of returning 4 different
values for the type of error that occurred.

Print the offending values instead as debug message.

Signed-off-by: Pranay Kr. Srivastava <pranjas@gmail.com>
---
 drivers/staging/speakup/main.c | 27 +++++++++++++++++++--------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/drivers/staging/speakup/main.c b/drivers/staging/speakup/main.c
index c2f70ef..6667cf2 100644
--- a/drivers/staging/speakup/main.c
+++ b/drivers/staging/speakup/main.c
@@ -1216,13 +1216,19 @@ int spk_set_key_info(const u_char *key_info, u_char *k_buffer)
 	u_char ch, version, num_keys;
 
 	version = *cp++;
-	if (version != KEY_MAP_VER)
-		return -1;
+	if (version != KEY_MAP_VER) {
+		pr_debug("version found %d should be %d\n",
+			 version, KEY_MAP_VER);
+		return -EINVAL;
+	}
 	num_keys = *cp;
 	states = (int)cp[1];
 	key_data_len = (states + 1) * (num_keys + 1);
-	if (key_data_len + SHIFT_TBL_SIZE + 4 >= sizeof(spk_key_buf))
-		return -2;
+	if (key_data_len + SHIFT_TBL_SIZE + 4 >= sizeof(spk_key_buf)) {
+		pr_debug("too many key_infos (%d over %lu)\n",
+			 key_data_len + SHIFT_TBL_SIZE + 4, sizeof(spk_key_buf));
+		return -EINVAL;
+	}
 	memset(k_buffer, 0, SHIFT_TBL_SIZE);
 	memset(spk_our_keys, 0, sizeof(spk_our_keys));
 	spk_shift_table = k_buffer;
@@ -1233,14 +1239,19 @@ int spk_set_key_info(const u_char *key_info, u_char *k_buffer)
 	cp1 += 2;		/* now pointing at shift states */
 	for (i = 1; i <= states; i++) {
 		ch = *cp1++;
-		if (ch >= SHIFT_TBL_SIZE)
-			return -3;
+		if (ch >= SHIFT_TBL_SIZE) {
+			pr_debug("(%d) not valid shift state (max_allowed = %d)\n", ch,
+				 SHIFT_TBL_SIZE);
+			return -EINVAL;
+		}
 		spk_shift_table[ch] = i;
 	}
 	keymap_flags = *cp1++;
 	while ((ch = *cp1)) {
-		if (ch >= MAX_KEY)
-			return -4;
+		if (ch >= MAX_KEY) {
+			pr_debug("(%d), not valid key, (max_allowed = %d)\n", ch, MAX_KEY);
+			return -EINVAL;
+		}
 		spk_our_keys[ch] = cp1;
 		cp1 += states + 1;
 	}
-- 
2.10.2

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

#1586047 — Re: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info

FromSamuel Thibault <samuel.thibault@ens-lyon.org>
Date2017-02-22 11:30 +0100
SubjectRe: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info
Message-ID<tdClj-aH-17@gated-at.bofh.it>
In reply to#1585967
Pranay Kr. Srivastava, on mer. 22 févr. 2017 12:50:26 +0530, wrote:
> This patch makes spk_set_key_info return -EINVAL
> in case of failure instead of returning 4 different
> values for the type of error that occurred.
> 
> Print the offending values instead as debug message.
> 
> Signed-off-by: Pranay Kr. Srivastava <pranjas@gmail.com>

Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>

> ---
>  drivers/staging/speakup/main.c | 27 +++++++++++++++++++--------
>  1 file changed, 19 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/staging/speakup/main.c b/drivers/staging/speakup/main.c
> index c2f70ef..6667cf2 100644
> --- a/drivers/staging/speakup/main.c
> +++ b/drivers/staging/speakup/main.c
> @@ -1216,13 +1216,19 @@ int spk_set_key_info(const u_char *key_info, u_char *k_buffer)
>  	u_char ch, version, num_keys;
>  
>  	version = *cp++;
> -	if (version != KEY_MAP_VER)
> -		return -1;
> +	if (version != KEY_MAP_VER) {
> +		pr_debug("version found %d should be %d\n",
> +			 version, KEY_MAP_VER);
> +		return -EINVAL;
> +	}
>  	num_keys = *cp;
>  	states = (int)cp[1];
>  	key_data_len = (states + 1) * (num_keys + 1);
> -	if (key_data_len + SHIFT_TBL_SIZE + 4 >= sizeof(spk_key_buf))
> -		return -2;
> +	if (key_data_len + SHIFT_TBL_SIZE + 4 >= sizeof(spk_key_buf)) {
> +		pr_debug("too many key_infos (%d over %lu)\n",
> +			 key_data_len + SHIFT_TBL_SIZE + 4, sizeof(spk_key_buf));
> +		return -EINVAL;
> +	}
>  	memset(k_buffer, 0, SHIFT_TBL_SIZE);
>  	memset(spk_our_keys, 0, sizeof(spk_our_keys));
>  	spk_shift_table = k_buffer;
> @@ -1233,14 +1239,19 @@ int spk_set_key_info(const u_char *key_info, u_char *k_buffer)
>  	cp1 += 2;		/* now pointing at shift states */
>  	for (i = 1; i <= states; i++) {
>  		ch = *cp1++;
> -		if (ch >= SHIFT_TBL_SIZE)
> -			return -3;
> +		if (ch >= SHIFT_TBL_SIZE) {
> +			pr_debug("(%d) not valid shift state (max_allowed = %d)\n", ch,
> +				 SHIFT_TBL_SIZE);
> +			return -EINVAL;
> +		}
>  		spk_shift_table[ch] = i;
>  	}
>  	keymap_flags = *cp1++;
>  	while ((ch = *cp1)) {
> -		if (ch >= MAX_KEY)
> -			return -4;
> +		if (ch >= MAX_KEY) {
> +			pr_debug("(%d), not valid key, (max_allowed = %d)\n", ch, MAX_KEY);
> +			return -EINVAL;
> +		}
>  		spk_our_keys[ch] = cp1;
>  		cp1 += states + 1;
>  	}
> -- 
> 2.10.2
> 

-- 
Samuel
<b> j'en ai parlé à xavier, il n'est pas interdit qu'il le change un jour
 -+- #sos - a le bras long vers le chameau -+-

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

#1587826 — Re: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info

FromGreg KH <gregkh@linuxfoundation.org>
Date2017-02-24 18:40 +0100
SubjectRe: [PATCH-SPEAKUP 1/2] return same error value from spk_set_key_info
Message-ID<tes0x-3rI-7@gated-at.bofh.it>
In reply to#1585967
On Wed, Feb 22, 2017 at 12:50:26PM +0530, Pranay Kr. Srivastava wrote:
> This patch makes spk_set_key_info return -EINVAL
> in case of failure instead of returning 4 different
> values for the type of error that occurred.
> 
> Print the offending values instead as debug message.
> 
> Signed-off-by: Pranay Kr. Srivastava <pranjas@gmail.com>
> ---
>  drivers/staging/speakup/main.c | 27 +++++++++++++++++++--------
>  1 file changed, 19 insertions(+), 8 deletions(-)

Please fix the build errors and resend this series.

thanks,

greg k-h

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web