[PATCH] drm/hisilicon: Use pcim_enable_device()
tiantao (H)
tiantao6 at huawei.com
Tue Dec 22 00:38:34 UTC 2020
在 2020/12/22 6:02, Daniel Vetter 写道:
> On Mon, Dec 21, 2020 at 08:45:22AM +0800, Tian Tao wrote:
>> Using the managed function simplifies the error handling. After
>> unloading the driver, the PCI device should now get disabled as
>> well.
>>
>> Signed-off-by: Tian Tao <tiantao6 at hisilicon.com>
>> ---
>> drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 6 ++----
>> 1 file changed, 2 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> index 02f3bd1..7159018 100644
>> --- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> +++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c
>> @@ -329,7 +329,7 @@ static int hibmc_pci_probe(struct pci_dev *pdev,
>> dev->pdev = pdev;
>> pci_set_drvdata(pdev, dev);
>>
>> - ret = pci_enable_device(pdev);
>> + ret = pcim_enable_device(pdev);
>> if (ret) {
>> drm_err(dev, "failed to enable pci device: %d\n", ret);
>> goto err_free;
>> @@ -338,7 +338,7 @@ static int hibmc_pci_probe(struct pci_dev *pdev,
>> ret = hibmc_load(dev);
>> if (ret) {
>> drm_err(dev, "failed to load hibmc: %d\n", ret);
>> - goto err_disable;
>> + goto err_free;
>> }
>>
>> ret = drm_dev_register(dev, 0);
>> @@ -354,8 +354,6 @@ static int hibmc_pci_probe(struct pci_dev *pdev,
>>
>> err_unload:
>> hibmc_unload(dev);
>> -err_disable:
>> - pci_disable_device(pdev);
>> err_free:
>> drm_dev_put(dev);
> The drm_dev_put here and in hibmc_pci_remove is wrong, devm_drm_dev_alloc
> takes care of that already. I'm kinda suprised you don't have a refcount
> underrun already - do you test module unload with KASAN enabled?
Thanks for helping to review the code,and kindly giving me advice。
this problem have been fixed。
c855af2f9c5c60760fd1bed7889a81bc37d2591d drm/hisilicon: Fix use-after-free
> The pcim patch looks ok, so Reviewed-by: Daniel Vetter <daniel.vetter at ffwll.ch>
>
> Cheers, Daniel
>
>>
>> --
>> 2.7.4
>>
More information about the dri-devel
mailing list