The two modified points are relevant. We bump the timeout for SET_FUNC_STATE
to ensure that cmd won't return failure when hw is busy. Otherwise hw
may stomp
host memory if we free memory regardless of the return value of
SET_FUNC_STATE.
I will mention the timeout changes in the commit message. And the bug
exists since
the first commit, not introduced. Thanks for your review.
On 2020/5/9 5:24, Jakub Kicinski wrote:
On Thu, 7 May 2020 18:22:27 +0000 Luo bin wrote:
if some function in ndo_stop interface returns failure because of
hardware fault, must go on excuting rest steps rather than return
failure directly, otherwise will cause memory leak
Signed-off-by: Luo bin <luob...@huawei.com>
The code looks good, but would it make sense to split this patch into
two? First one that ignores the return values on close path with these
fixes tags:
Fixes: e2585ea77538 ("net-next/hinic: Add Rx handler")
Fixes: c4d06d2d208a ("net-next/hinic: Add Rx mode and link event handler")
And a separate patch which bumps the timeout for SET_FUNC_STATE? Right
now you don't even mention the timeout changes in the commit message.
.
