* [PATCH] ARM: pxa: Fix error handling in viper_tpm_init
@ 2021-12-27 12:51 Miaoqian Lin
2021-12-27 15:34 ` Daniel Mack
0 siblings, 1 reply; 2+ messages in thread
From: Miaoqian Lin @ 2021-12-27 12:51 UTC (permalink / raw)
Cc: linmq006, Daniel Mack, Haojian Zhuang, Robert Jarzmik,
Russell King, Marc Zyngier, linux-arm-kernel, linux-kernel
tpm_device get by platform_device_alloc(), which calls device_initialize.
Call platform_device_put(tpm_device) instead of kfree(tmp_device) to
prevent memory leak.
Fixes: 352699a ("Basic support for the Arcom/Eurotech Viper SBC.")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
---
arch/arm/mach-pxa/viper.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/arch/arm/mach-pxa/viper.c b/arch/arm/mach-pxa/viper.c
index 3aa34e9a15d3..9a0f184dbecb 100644
--- a/arch/arm/mach-pxa/viper.c
+++ b/arch/arm/mach-pxa/viper.c
@@ -815,11 +815,11 @@ static void __init viper_tpm_init(void)
sizeof(i2c_tpm_data))) {
if (platform_device_add(tpm_device)) {
errstr = "register TPM i2c bus";
- goto error_free_tpm;
+ goto exit_put_device;
}
} else {
errstr = "allocate TPM i2c bus data";
- goto error_free_tpm;
+ goto exit_put_device;
}
} else {
errstr = "allocate TPM i2c device";
@@ -827,9 +827,8 @@ static void __init viper_tpm_init(void)
}
return;
-
-error_free_tpm:
- kfree(tpm_device);
+exit_put_device:
+ platform_device_put(tpm_device);
error_tpm:
pr_err("viper: Couldn't %s, giving up\n", errstr);
}
--
2.17.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] ARM: pxa: Fix error handling in viper_tpm_init
2021-12-27 12:51 [PATCH] ARM: pxa: Fix error handling in viper_tpm_init Miaoqian Lin
@ 2021-12-27 15:34 ` Daniel Mack
0 siblings, 0 replies; 2+ messages in thread
From: Daniel Mack @ 2021-12-27 15:34 UTC (permalink / raw)
To: Miaoqian Lin
Cc: Haojian Zhuang, Robert Jarzmik, Russell King, Marc Zyngier,
linux-arm-kernel, linux-kernel
Hi,
On 27/12/2021 13:51, Miaoqian Lin wrote:
> tpm_device get by platform_device_alloc(), which calls device_initialize.
> Call platform_device_put(tpm_device) instead of kfree(tmp_device) to
> prevent memory leak.
While this seems correct, I don't think we should make any effort to
maintain legacy board support files but rather move to DTS instead.
Robert and I laid the foundation and ported a few machines a while ago,
but I was reluctant to touch the board support code I can't test on real
hardware.
As you seem to have access to a Viper board, could you have a look at
the existing device-tree ported PXA machines and see if you can follow
suit here please?
Thanks,
Daniel
>
> Fixes: 352699a ("Basic support for the Arcom/Eurotech Viper SBC.")
> Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
> ---
> arch/arm/mach-pxa/viper.c | 9 ++++-----
> 1 file changed, 4 insertions(+), 5 deletions(-)
>
> diff --git a/arch/arm/mach-pxa/viper.c b/arch/arm/mach-pxa/viper.c
> index 3aa34e9a15d3..9a0f184dbecb 100644
> --- a/arch/arm/mach-pxa/viper.c
> +++ b/arch/arm/mach-pxa/viper.c
> @@ -815,11 +815,11 @@ static void __init viper_tpm_init(void)
> sizeof(i2c_tpm_data))) {
> if (platform_device_add(tpm_device)) {
> errstr = "register TPM i2c bus";
> - goto error_free_tpm;
> + goto exit_put_device;
> }
> } else {
> errstr = "allocate TPM i2c bus data";
> - goto error_free_tpm;
> + goto exit_put_device;
> }
> } else {
> errstr = "allocate TPM i2c device";
> @@ -827,9 +827,8 @@ static void __init viper_tpm_init(void)
> }
>
> return;
> -
> -error_free_tpm:
> - kfree(tpm_device);
> +exit_put_device:
> + platform_device_put(tpm_device);
> error_tpm:
> pr_err("viper: Couldn't %s, giving up\n", errstr);
> }
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-12-27 15:46 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-27 12:51 [PATCH] ARM: pxa: Fix error handling in viper_tpm_init Miaoqian Lin
2021-12-27 15:34 ` Daniel Mack
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).