From: Pan Bian <[email protected]>
Variable ret is reset in the loop, and its value will be 0 during the
second and after repeat of the loop. If pci_alloc_consistent() returns a
NULL pointer then, it will leaves with return value 0. 0 means no error,
which is contrary to the fact. This patches fixes the bug, explicitly
assigning "-ENOMEM" to return variable ret on the path that the call to
pci_alloc_consistent() fails.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=188941
Signed-off-by: Pan Bian <[email protected]>
---
drivers/scsi/be2iscsi/be_main.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c
index d9239c2..b6c5791 100644
--- a/drivers/scsi/be2iscsi/be_main.c
+++ b/drivers/scsi/be2iscsi/be_main.c
@@ -3113,8 +3113,10 @@ static int beiscsi_create_cqs(struct beiscsi_hba *phba,
cq_vaddress = pci_alloc_consistent(phba->pcidev,
num_cq_pages * PAGE_SIZE,
&paddr);
- if (!cq_vaddress)
+ if (!cq_vaddress) {
+ ret = -ENOMEM;
goto create_cq_error;
+ }
ret = be_fill_queue(cq, phba->params.num_cq_entries,
sizeof(struct sol_cqe), cq_vaddress);
--
1.9.1
> -----Original Message-----
> From: Pan Bian [mailto:[email protected]]
> Sent: Sunday, December 04, 2016 10:52 AM
> To: Subbu Seetharaman; Ketan Mukadam; Jitendra Bhivare; James E.J.
> Bottomley; Martin K. Petersen; [email protected]
> Cc: [email protected]; Pan Bian
> Subject: [PATCH 1/2] scsi: be2iscsi: set errno on error path
>
> From: Pan Bian <[email protected]>
>
> Variable ret is reset in the loop, and its value will be 0 during the
second and
> after repeat of the loop. If pci_alloc_consistent() returns a NULL
pointer then, it
> will leaves with return value 0. 0 means no error, which is contrary to
the fact.
> This patches fixes the bug, explicitly assigning "-ENOMEM" to return
variable ret
> on the path that the call to
> pci_alloc_consistent() fails.
>
> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=188941
>
> Signed-off-by: Pan Bian <[email protected]>
> ---
> drivers/scsi/be2iscsi/be_main.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/scsi/be2iscsi/be_main.c
b/drivers/scsi/be2iscsi/be_main.c
> index d9239c2..b6c5791 100644
> --- a/drivers/scsi/be2iscsi/be_main.c
> +++ b/drivers/scsi/be2iscsi/be_main.c
> @@ -3113,8 +3113,10 @@ static int beiscsi_create_cqs(struct beiscsi_hba
> *phba,
> cq_vaddress = pci_alloc_consistent(phba->pcidev,
> num_cq_pages *
PAGE_SIZE,
> &paddr);
> - if (!cq_vaddress)
> + if (!cq_vaddress) {
> + ret = -ENOMEM;
> goto create_cq_error;
> + }
>
> ret = be_fill_queue(cq, phba->params.num_cq_entries,
> sizeof(struct sol_cqe), cq_vaddress);
> --
> 1.9.1
[JB] Thanks.
Reviewed-by: Jitendra Bhivare <[email protected]>
>>>>> "Pan" == Pan Bian <[email protected]> writes:
Pan> Variable ret is reset in the loop, and its value will be 0 during
Pan> the second and after repeat of the loop. If pci_alloc_consistent()
Pan> returns a NULL pointer then, it will leaves with return value 0. 0
Pan> means no error, which is contrary to the fact. This patches fixes
Pan> the bug, explicitly assigning "-ENOMEM" to return variable ret on
Pan> the path that the call to pci_alloc_consistent() fails.
Applied patches 1 and 2 to 4.10/scsi-queue.
--
Martin K. Petersen Oracle Linux Engineering