* Re: [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:33 [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe() Dan Carpenter
@ 2021-06-21 12:41 ` Hans de Goede
2021-06-21 12:47 ` Dan Carpenter
2021-06-22 10:46 ` [PATCH v3] " Dan Carpenter
` (3 subsequent siblings)
4 siblings, 1 reply; 8+ messages in thread
From: Hans de Goede @ 2021-06-21 12:41 UTC (permalink / raw)
To: Dan Carpenter, Arend van Spriel, Matthias Brugger, Christophe JAILLET
Cc: Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
Hi,
On 6/21/21 2:33 PM, Dan Carpenter wrote:
> There are two bugs here:
> 1) devm_kzalloc() needs to be checked for allocation errors.
> 2) The loop was intended to be:
>
> Bad: for (i = 0; i < board_type[i]; i++) {
> Good: for (i = 0; i < len; i++) {
>
> Neither of these bugs are likely to cause an issue in practice but
> they're worth fixing. Also the code could be made simpler by using the
> devm_kstrdup() and strreplace() functions.
>
> Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
> Suggested-by: Johannes Berg <johannes@sipsolutions.net>
> Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> ---
> v2: Basically a completely different patch. :)
>
> .../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 ++++--------
> 1 file changed, 4 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> index a7554265f95f..8ca62cd0e8ac 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> @@ -24,20 +24,16 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
> /* Set board-type to the first string of the machine compatible prop */
> root = of_find_node_by_path("/");
> if (root) {
> - int i, len;
> char *board_type;
> const char *tmp;
>
> of_property_read_string_index(root, "compatible", 0, &tmp);
>
> /* get rid of '/' in the compatible string to be able to find the FW */
> - len = strlen(tmp) + 1;
> - board_type = devm_kzalloc(dev, len, GFP_KERNEL);
> - strscpy(board_type, tmp, len);
> - for (i = 0; i < board_type[i]; i++) {
> - if (board_type[i] == '/')
> - board_type[i] = '-';
> - }
> + board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
> + if (!board_type)
> + return;
The error handling here seems to be lacking, specifically
this error-exit path needs a of_node_put(root); call (or
a "goto put;" or some such)
Regards,
Hans
> + strreplace(board_type, '/', '-');
> settings->board_type = board_type;
>
> of_node_put(root);
>
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:41 ` Hans de Goede
@ 2021-06-21 12:47 ` Dan Carpenter
0 siblings, 0 replies; 8+ messages in thread
From: Dan Carpenter @ 2021-06-21 12:47 UTC (permalink / raw)
To: Hans de Goede
Cc: Arend van Spriel, Matthias Brugger, Christophe JAILLET,
Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
On Mon, Jun 21, 2021 at 02:41:03PM +0200, Hans de Goede wrote:
> Hi,
>
> On 6/21/21 2:33 PM, Dan Carpenter wrote:
> > There are two bugs here:
> > 1) devm_kzalloc() needs to be checked for allocation errors.
> > 2) The loop was intended to be:
> >
> > Bad: for (i = 0; i < board_type[i]; i++) {
> > Good: for (i = 0; i < len; i++) {
> >
> > Neither of these bugs are likely to cause an issue in practice but
> > they're worth fixing. Also the code could be made simpler by using the
> > devm_kstrdup() and strreplace() functions.
> >
> > Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
> > Suggested-by: Johannes Berg <johannes@sipsolutions.net>
> > Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> > ---
> > v2: Basically a completely different patch. :)
> >
> > .../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 ++++--------
> > 1 file changed, 4 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> > index a7554265f95f..8ca62cd0e8ac 100644
> > --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> > +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> > @@ -24,20 +24,16 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
> > /* Set board-type to the first string of the machine compatible prop */
> > root = of_find_node_by_path("/");
> > if (root) {
> > - int i, len;
> > char *board_type;
> > const char *tmp;
> >
> > of_property_read_string_index(root, "compatible", 0, &tmp);
> >
> > /* get rid of '/' in the compatible string to be able to find the FW */
> > - len = strlen(tmp) + 1;
> > - board_type = devm_kzalloc(dev, len, GFP_KERNEL);
> > - strscpy(board_type, tmp, len);
> > - for (i = 0; i < board_type[i]; i++) {
> > - if (board_type[i] == '/')
> > - board_type[i] = '-';
> > - }
> > + board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
> > + if (!board_type)
> > + return;
>
> The error handling here seems to be lacking, specifically
> this error-exit path needs a of_node_put(root); call (or
> a "goto put;" or some such)
Good catch. I will add a call to of_node_put().
regards,
dan carpenter
^ permalink raw reply [flat|nested] 8+ messages in thread
* [PATCH v3] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:33 [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe() Dan Carpenter
2021-06-21 12:41 ` Hans de Goede
@ 2021-06-22 10:46 ` Dan Carpenter
2021-06-22 10:48 ` Hans de Goede
` (2 subsequent siblings)
4 siblings, 0 replies; 8+ messages in thread
From: Dan Carpenter @ 2021-06-22 10:46 UTC (permalink / raw)
To: Arend van Spriel, Matthias Brugger, Christophe JAILLET
Cc: Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, Hans deGoede, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
There are two bugs here:
1) devm_kzalloc() needs to be checked for allocation errors.
2) The loop was intended to be:
Bad: for (i = 0; i < board_type[i]; i++) {
Good: for (i = 0; i < len; i++) {
Neither of these bugs are likely to cause an issue in practice but
they're worth fixing. Also the code could be made simpler by using the
devm_kstrdup() and strreplace() functions.
Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
Suggested-by: Johannes Berg <johannes@sipsolutions.net>
Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reported-by: Hans deGoede <hdegoede@redhat.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
---
v2: Basically a completely different patch. :)
v3: Add missing of_node_put() caught by Hans de Goede
.../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 +++++-------
1 file changed, 5 insertions(+), 7 deletions(-)
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
index a7554265f95f..dee3d968e49b 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
@@ -24,20 +24,18 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
/* Set board-type to the first string of the machine compatible prop */
root = of_find_node_by_path("/");
if (root) {
- int i, len;
char *board_type;
const char *tmp;
of_property_read_string_index(root, "compatible", 0, &tmp);
/* get rid of '/' in the compatible string to be able to find the FW */
- len = strlen(tmp) + 1;
- board_type = devm_kzalloc(dev, len, GFP_KERNEL);
- strscpy(board_type, tmp, len);
- for (i = 0; i < board_type[i]; i++) {
- if (board_type[i] == '/')
- board_type[i] = '-';
+ board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
+ if (!board_type) {
+ of_node_put(root);
+ return;
}
+ strreplace(board_type, '/', '-');
settings->board_type = board_type;
of_node_put(root);
--
2.30.2
^ permalink raw reply related [flat|nested] 8+ messages in thread* Re: [PATCH v3] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:33 [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe() Dan Carpenter
2021-06-21 12:41 ` Hans de Goede
2021-06-22 10:46 ` [PATCH v3] " Dan Carpenter
@ 2021-06-22 10:48 ` Hans de Goede
2021-06-22 10:56 ` Matthias Brugger
2021-09-08 13:12 ` Dan Carpenter
4 siblings, 0 replies; 8+ messages in thread
From: Hans de Goede @ 2021-06-22 10:48 UTC (permalink / raw)
To: Dan Carpenter, Arend van Spriel, Matthias Brugger, Christophe JAILLET
Cc: Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
Hi,
On 6/22/21 12:46 PM, Dan Carpenter wrote:
> There are two bugs here:
> 1) devm_kzalloc() needs to be checked for allocation errors.
> 2) The loop was intended to be:
>
> Bad: for (i = 0; i < board_type[i]; i++) {
> Good: for (i = 0; i < len; i++) {
>
> Neither of these bugs are likely to cause an issue in practice but
> they're worth fixing. Also the code could be made simpler by using the
> devm_kstrdup() and strreplace() functions.
>
> Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
> Suggested-by: Johannes Berg <johannes@sipsolutions.net>
> Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> Reported-by: Hans deGoede <hdegoede@redhat.com>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Thanks, patch looks good to me:
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Regards,
Hans
> ---
> v2: Basically a completely different patch. :)
> v3: Add missing of_node_put() caught by Hans de Goede
>
> .../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 +++++-------
> 1 file changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> index a7554265f95f..dee3d968e49b 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> @@ -24,20 +24,18 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
> /* Set board-type to the first string of the machine compatible prop */
> root = of_find_node_by_path("/");
> if (root) {
> - int i, len;
> char *board_type;
> const char *tmp;
>
> of_property_read_string_index(root, "compatible", 0, &tmp);
>
> /* get rid of '/' in the compatible string to be able to find the FW */
> - len = strlen(tmp) + 1;
> - board_type = devm_kzalloc(dev, len, GFP_KERNEL);
> - strscpy(board_type, tmp, len);
> - for (i = 0; i < board_type[i]; i++) {
> - if (board_type[i] == '/')
> - board_type[i] = '-';
> + board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
> + if (!board_type) {
> + of_node_put(root);
> + return;
> }
> + strreplace(board_type, '/', '-');
> settings->board_type = board_type;
>
> of_node_put(root);
>
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH v3] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:33 [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe() Dan Carpenter
` (2 preceding siblings ...)
2021-06-22 10:48 ` Hans de Goede
@ 2021-06-22 10:56 ` Matthias Brugger
2021-09-08 13:12 ` Dan Carpenter
4 siblings, 0 replies; 8+ messages in thread
From: Matthias Brugger @ 2021-06-22 10:56 UTC (permalink / raw)
To: Dan Carpenter, Arend van Spriel, Christophe JAILLET
Cc: Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, Hans deGoede, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
On 22/06/2021 12:46, Dan Carpenter wrote:
> There are two bugs here:
> 1) devm_kzalloc() needs to be checked for allocation errors.
> 2) The loop was intended to be:
>
> Bad: for (i = 0; i < board_type[i]; i++) {
> Good: for (i = 0; i < len; i++) {
>
> Neither of these bugs are likely to cause an issue in practice but
> they're worth fixing. Also the code could be made simpler by using the
> devm_kstrdup() and strreplace() functions.
>
> Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
> Suggested-by: Johannes Berg <johannes@sipsolutions.net>
> Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> Reported-by: Hans deGoede <hdegoede@redhat.com>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Thanks for taking care of this.
Reviewed-by: Matthias Brugger <mbrugger@suse.com>
> ---
> v2: Basically a completely different patch. :)
> v3: Add missing of_node_put() caught by Hans de Goede
>
> .../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 +++++-------
> 1 file changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> index a7554265f95f..dee3d968e49b 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> @@ -24,20 +24,18 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
> /* Set board-type to the first string of the machine compatible prop */
> root = of_find_node_by_path("/");
> if (root) {
> - int i, len;
> char *board_type;
> const char *tmp;
>
> of_property_read_string_index(root, "compatible", 0, &tmp);
>
> /* get rid of '/' in the compatible string to be able to find the FW */
> - len = strlen(tmp) + 1;
> - board_type = devm_kzalloc(dev, len, GFP_KERNEL);
> - strscpy(board_type, tmp, len);
> - for (i = 0; i < board_type[i]; i++) {
> - if (board_type[i] == '/')
> - board_type[i] = '-';
> + board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
> + if (!board_type) {
> + of_node_put(root);
> + return;
> }
> + strreplace(board_type, '/', '-');
> settings->board_type = board_type;
>
> of_node_put(root);
>
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH v3] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-06-21 12:33 [PATCH v2] brcmfmac: fix "board_type" in brcmf_of_probe() Dan Carpenter
` (3 preceding siblings ...)
2021-06-22 10:56 ` Matthias Brugger
@ 2021-09-08 13:12 ` Dan Carpenter
2021-09-09 9:42 ` Kalle Valo
4 siblings, 1 reply; 8+ messages in thread
From: Dan Carpenter @ 2021-09-08 13:12 UTC (permalink / raw)
To: Arend van Spriel, Matthias Brugger, Christophe JAILLET
Cc: Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Kalle Valo, Hans deGoede, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
This patch is still needed.
regards,
dan carpenter
On Tue, Jun 22, 2021 at 01:46:55PM +0300, Dan Carpenter wrote:
> There are two bugs here:
> 1) devm_kzalloc() needs to be checked for allocation errors.
> 2) The loop was intended to be:
>
> Bad: for (i = 0; i < board_type[i]; i++) {
> Good: for (i = 0; i < len; i++) {
>
> Neither of these bugs are likely to cause an issue in practice but
> they're worth fixing. Also the code could be made simpler by using the
> devm_kstrdup() and strreplace() functions.
>
> Fixes: 29e354ebeeec ("brcmfmac: Transform compatible string for FW loading")
> Suggested-by: Johannes Berg <johannes@sipsolutions.net>
> Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
> Reported-by: Hans deGoede <hdegoede@redhat.com>
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> ---
> v2: Basically a completely different patch. :)
> v3: Add missing of_node_put() caught by Hans de Goede
>
> .../net/wireless/broadcom/brcm80211/brcmfmac/of.c | 12 +++++-------
> 1 file changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> index a7554265f95f..dee3d968e49b 100644
> --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/of.c
> @@ -24,20 +24,18 @@ void brcmf_of_probe(struct device *dev, enum brcmf_bus_type bus_type,
> /* Set board-type to the first string of the machine compatible prop */
> root = of_find_node_by_path("/");
> if (root) {
> - int i, len;
> char *board_type;
> const char *tmp;
>
> of_property_read_string_index(root, "compatible", 0, &tmp);
>
> /* get rid of '/' in the compatible string to be able to find the FW */
> - len = strlen(tmp) + 1;
> - board_type = devm_kzalloc(dev, len, GFP_KERNEL);
> - strscpy(board_type, tmp, len);
> - for (i = 0; i < board_type[i]; i++) {
> - if (board_type[i] == '/')
> - board_type[i] = '-';
> + board_type = devm_kstrdup(dev, tmp, GFP_KERNEL);
> + if (!board_type) {
> + of_node_put(root);
> + return;
> }
> + strreplace(board_type, '/', '-');
> settings->board_type = board_type;
>
> of_node_put(root);
> --
> 2.30.2
^ permalink raw reply [flat|nested] 8+ messages in thread* Re: [PATCH v3] brcmfmac: fix "board_type" in brcmf_of_probe()
2021-09-08 13:12 ` Dan Carpenter
@ 2021-09-09 9:42 ` Kalle Valo
0 siblings, 0 replies; 8+ messages in thread
From: Kalle Valo @ 2021-09-09 9:42 UTC (permalink / raw)
To: Dan Carpenter
Cc: Arend van Spriel, Matthias Brugger, Christophe JAILLET,
Franky Lin, Hante Meuleman, Chi-hsien Lin, Wright Feng,
Chung-hsien Hsu, Hans deGoede, linux-wireless,
brcm80211-dev-list.pdl, SHA-cyfmac-dev-list, kernel-janitors
Dan Carpenter <dan.carpenter@oracle.com> writes:
> This patch is still needed.
I don't see v3 in patchwork[1], but I do see v2. I don't know what
happened, but please resend the patch as v4.
[1] https://patchwork.kernel.org/project/linux-wireless/list/?series=&submitter=37111&state=*&q=&archive=&delegate=
--
https://patchwork.kernel.org/project/linux-wireless/list/
https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
^ permalink raw reply [flat|nested] 8+ messages in thread