2021-03-08 10:23:27

by Jia-Ju Bai

Subject: [PATCH] block: rsxx: fix error return code of rsxx_pci_probe()

Some error handling segments of rsxx_pci_probe() do not return error code,
so add error code for these segments.

Fixes: 8722ff8cdbfa ("block: IBM RamSan 70/80 device driver")
Reported-by: TOTE Robot <[email protected]>
Signed-off-by: Jia-Ju Bai <[email protected]>
---
drivers/block/rsxx/core.c | 18 +++++++++++++++---
1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
index 63f549889f87..6b3b9b31a3e8 100644
--- a/drivers/block/rsxx/core.c
+++ b/drivers/block/rsxx/core.c
@@ -760,13 +760,17 @@ static int rsxx_pci_probe(struct pci_dev *dev,
pci_set_drvdata(dev, card);

st = ida_alloc(&rsxx_disk_ida, GFP_KERNEL);
- if (st < 0)
+ if (st < 0) {
+ st = -ENOMEM;
goto failed_ida_get;
+ }
card->disk_id = st;

st = pci_enable_device(dev);
- if (st)
+ if (st) {
+ st = -EIO;
goto failed_enable;
+ }

pci_set_master(dev);

@@ -774,6 +778,7 @@ static int rsxx_pci_probe(struct pci_dev *dev,
if (st) {
dev_err(CARD_TO_DEV(card),
"No usable DMA configuration,aborting\n");
+ st = -EIO;
goto failed_dma_mask;
}

@@ -781,6 +786,7 @@ static int rsxx_pci_probe(struct pci_dev *dev,
if (st) {
dev_err(CARD_TO_DEV(card),
"Failed to request memory region\n");
+ st = -EIO;
goto failed_request_regions;
}

@@ -817,6 +823,7 @@ static int rsxx_pci_probe(struct pci_dev *dev,
if (st) {
dev_err(CARD_TO_DEV(card),
"Failed requesting IRQ%d\n", dev->irq);
+ st = -EINVAL;
goto failed_irq;
}

@@ -824,6 +831,7 @@ static int rsxx_pci_probe(struct pci_dev *dev,
st = rsxx_creg_setup(card);
if (st) {
dev_err(CARD_TO_DEV(card), "Failed to setup creg interface.\n");
+ st = -EINVAL;
goto failed_creg_setup;
}

@@ -862,6 +870,7 @@ static int rsxx_pci_probe(struct pci_dev *dev,
if (st) {
dev_info(CARD_TO_DEV(card),
"Failed to setup DMA engine\n");
+ st = -EINVAL;
goto failed_dma_setup;
}

@@ -869,14 +878,17 @@ static int rsxx_pci_probe(struct pci_dev *dev,
card->event_wq = create_singlethread_workqueue(DRIVER_NAME"_event");
if (!card->event_wq) {
dev_err(CARD_TO_DEV(card), "Failed card event setup.\n");
+ st = -ENOMEM;
goto failed_event_handler;
}

INIT_WORK(&card->event_work, card_event_handler);

st = rsxx_setup_dev(card);
- if (st)
+ if (st) {
+ st = -EINVAL;
goto failed_create_dev;
+ }

rsxx_get_card_state(card, &card->state);

--
2.17.1

2021-03-09 21:01:37

by Jens Axboe

Subject: Re: [PATCH] block: rsxx: fix error return code of rsxx_pci_probe()

On 3/8/21 3:05 AM, Jia-Ju Bai wrote:
> Some error handling segments of rsxx_pci_probe() do not return error code,
> so add error code for these segments.
>
> Fixes: 8722ff8cdbfa ("block: IBM RamSan 70/80 device driver")
> Reported-by: TOTE Robot <[email protected]>
> Signed-off-by: Jia-Ju Bai <[email protected]>
> ---
> drivers/block/rsxx/core.c | 18 +++++++++++++++---
> 1 file changed, 15 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
> index 63f549889f87..6b3b9b31a3e8 100644
> --- a/drivers/block/rsxx/core.c
> +++ b/drivers/block/rsxx/core.c
> @@ -760,13 +760,17 @@ static int rsxx_pci_probe(struct pci_dev *dev,
> pci_set_drvdata(dev, card);
>
> st = ida_alloc(&rsxx_disk_ida, GFP_KERNEL);
> - if (st < 0)
> + if (st < 0) {
> + st = -ENOMEM;
> goto failed_ida_get;
> + }
> card->disk_id = st;
>
> st = pci_enable_device(dev);
> - if (st)
> + if (st) {
> + st = -EIO;
> goto failed_enable;
> + }
>
> pci_set_master(dev);

Maybe there are some valid parts to the patch, but the two above at
least make no sense - we're returning the error here as passed from
ida_alloc or pci_enable_device, why are you overriding them?

--
Jens Axboe

2021-03-10 01:18:42

by Jia-Ju Bai

Subject: Re: [PATCH] block: rsxx: fix error return code of rsxx_pci_probe()

On 2021/3/10 4:59, Jens Axboe wrote:
> On 3/8/21 3:05 AM, Jia-Ju Bai wrote:
>> Some error handling segments of rsxx_pci_probe() do not return error code,
>> so add error code for these segments.
>>
>> Fixes: 8722ff8cdbfa ("block: IBM RamSan 70/80 device driver")
>> Reported-by: TOTE Robot <[email protected]>
>> Signed-off-by: Jia-Ju Bai <[email protected]>
>> ---
>> drivers/block/rsxx/core.c | 18 +++++++++++++++---
>> 1 file changed, 15 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/block/rsxx/core.c b/drivers/block/rsxx/core.c
>> index 63f549889f87..6b3b9b31a3e8 100644
>> --- a/drivers/block/rsxx/core.c
>> +++ b/drivers/block/rsxx/core.c
>> @@ -760,13 +760,17 @@ static int rsxx_pci_probe(struct pci_dev *dev,
>> pci_set_drvdata(dev, card);
>>
>> st = ida_alloc(&rsxx_disk_ida, GFP_KERNEL);
>> - if (st < 0)
>> + if (st < 0) {
>> + st = -ENOMEM;
>> goto failed_ida_get;
>> + }
>> card->disk_id = st;
>>
>> st = pci_enable_device(dev);
>> - if (st)
>> + if (st) {
>> + st = -EIO;
>> goto failed_enable;
>> + }
>>
>> pci_set_master(dev);
> Maybe there are some valid parts to the patch, but the two above at
> least make no sense - we're returning the error here as passed from
> ida_alloc or pci_enable_device, why are you overriding them?
>

Ah, sorry for these incorrect parts...
I will send a new patch.


Best wishes,
Jia-Ju Bai