Re: [PATCH] qlge: refactor qlge_change_rx_buffers() in qlge_main.c

[Pavel Skripkin <paskripkin@gmail.com>]

Hi, Adam!

Looks like you have missed all maintainers in your CC list :) Please, 
use ./script/get_mainainer.pl to get list of people who are responsible 
for reviewing your code

On 12/28/21 16:57, Adam Kandur wrote:
> Greetings. As I understand, qlge_change_rx_buffers() waits 4 times for
> an outstanding reset to complete. And if it cannot, reports about
> timeout. I assume this can be written using for loop instead of while,
> wich looks nicer for me. Hope my guess was right.
> 

Commit message should only explain why you do this change. You can leave 
your unrelated thoughts/questions below ---, since this part won't be 
visible in gitlog

> Signed-off-by: Adam Kandur <sys.arch.adam@gmail.com>
> ---

<--- here

>   drivers/staging/qlge/qlge_main.c | 15 ++++++---------
>   1 file changed, 6 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/staging/qlge/qlge_main.c b/drivers/staging/qlge/qlge_main.c
> index 9873bb2a9ee4..f36e43a85038 100644
> --- a/drivers/staging/qlge/qlge_main.c
> +++ b/drivers/staging/qlge/qlge_main.c
> @@ -4012,19 +4012,16 @@ static int qlge_change_rx_buffers(struct qlge_adapter *qdev)
>   
>   	/* Wait for an outstanding reset to complete. */
>   	if (!test_bit(QL_ADAPTER_UP, &qdev->flags)) {

Aren't there 5 checks actually? It seems unlikely that this bit will be 
set after 3-4 instructions


> -		int i = 4;
> -
> -		while (--i && !test_bit(QL_ADAPTER_UP, &qdev->flags)) {
> +		for (int i = 4; !test_bit(QL_ADAPTER_UP, &qdev->flags); i--) {

Have you compiled this patch? Compiler should have warn you about mixed 
declaration and code.



With regards,
Pavel Skripkin