2022-03-27 00:51:12

by Larry Finger

Subject: Re: [PATCH 21/22] rtw89: Replace comments with C99 initializers

On 3/26/22 11:59, Benjamin Stürz wrote:
> This replaces comments with C99's designated
> initializers because the kernel supports them now.
>
> Signed-off-by: Benjamin Stürz <[email protected]>
> ---
> drivers/net/wireless/realtek/rtw89/coex.c | 40 +++++++++++------------
> 1 file changed, 20 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
> index 684583955511..3c83a0bfb120 100644
> --- a/drivers/net/wireless/realtek/rtw89/coex.c
> +++ b/drivers/net/wireless/realtek/rtw89/coex.c
> @@ -97,26 +97,26 @@ static const struct rtw89_btc_fbtc_slot s_def[] = {
> };
>
> static const u32 cxtbl[] = {
> - 0xffffffff, /* 0 */
> - 0xaaaaaaaa, /* 1 */
> - 0x55555555, /* 2 */
> - 0x66555555, /* 3 */
> - 0x66556655, /* 4 */
> - 0x5a5a5a5a, /* 5 */
> - 0x5a5a5aaa, /* 6 */
> - 0xaa5a5a5a, /* 7 */
> - 0x6a5a5a5a, /* 8 */
> - 0x6a5a5aaa, /* 9 */
> - 0x6a5a6a5a, /* 10 */
> - 0x6a5a6aaa, /* 11 */
> - 0x6afa5afa, /* 12 */
> - 0xaaaa5aaa, /* 13 */
> - 0xaaffffaa, /* 14 */
> - 0xaa5555aa, /* 15 */
> - 0xfafafafa, /* 16 */
> - 0xffffddff, /* 17 */
> - 0xdaffdaff, /* 18 */
> - 0xfafadafa /* 19 */
> + [0] = 0xffffffff,
> + [1] = 0xaaaaaaaa,
> + [2] = 0x55555555,
> + [3] = 0x66555555,
> + [4] = 0x66556655,
> + [5] = 0x5a5a5a5a,
> + [6] = 0x5a5a5aaa,
> + [7] = 0xaa5a5a5a,
> + [8] = 0x6a5a5a5a,
> + [9] = 0x6a5a5aaa,
> + [10] = 0x6a5a6a5a,
> + [11] = 0x6a5a6aaa,
> + [12] = 0x6afa5afa,
> + [13] = 0xaaaa5aaa,
> + [14] = 0xaaffffaa,
> + [15] = 0xaa5555aa,
> + [16] = 0xfafafafa,
> + [17] = 0xffffddff,
> + [18] = 0xdaffdaff,
> + [19] = 0xfafadafa
> };
>
> struct rtw89_btc_btf_tlv {


Is this change really necessary? Yes, the entries must be ordered; however, the
comment carries that information at very few extra characters. To me, this patch
looks like unneeded source churn. One other concern is that this driver is
backported to older kernels and older compilers by several distros. Will this
change require adding extra conditional statements to the source used in these
applications?

Larry

2022-03-28 09:52:54

by Kalle Valo

Subject: Re: [PATCH 21/22] rtw89: Replace comments with C99 initializers

Larry Finger <[email protected]> writes:

> On 3/26/22 11:59, Benjamin Stürz wrote:
>> This replaces comments with C99's designated
>> initializers because the kernel supports them now.
>>
>> Signed-off-by: Benjamin Stürz <[email protected]>
>> ---
>> drivers/net/wireless/realtek/rtw89/coex.c | 40 +++++++++++------------
>> 1 file changed, 20 insertions(+), 20 deletions(-)
>>
>> diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
>> index 684583955511..3c83a0bfb120 100644
>> --- a/drivers/net/wireless/realtek/rtw89/coex.c
>> +++ b/drivers/net/wireless/realtek/rtw89/coex.c
>> @@ -97,26 +97,26 @@ static const struct rtw89_btc_fbtc_slot s_def[] = {
>> };
>> static const u32 cxtbl[] = {
>> - 0xffffffff, /* 0 */
>> - 0xaaaaaaaa, /* 1 */
>> - 0x55555555, /* 2 */
>> - 0x66555555, /* 3 */
>> - 0x66556655, /* 4 */
>> - 0x5a5a5a5a, /* 5 */
>> - 0x5a5a5aaa, /* 6 */
>> - 0xaa5a5a5a, /* 7 */
>> - 0x6a5a5a5a, /* 8 */
>> - 0x6a5a5aaa, /* 9 */
>> - 0x6a5a6a5a, /* 10 */
>> - 0x6a5a6aaa, /* 11 */
>> - 0x6afa5afa, /* 12 */
>> - 0xaaaa5aaa, /* 13 */
>> - 0xaaffffaa, /* 14 */
>> - 0xaa5555aa, /* 15 */
>> - 0xfafafafa, /* 16 */
>> - 0xffffddff, /* 17 */
>> - 0xdaffdaff, /* 18 */
>> - 0xfafadafa /* 19 */
>> + [0] = 0xffffffff,
>> + [1] = 0xaaaaaaaa,
>> + [2] = 0x55555555,
>> + [3] = 0x66555555,
>> + [4] = 0x66556655,
>> + [5] = 0x5a5a5a5a,
>> + [6] = 0x5a5a5aaa,
>> + [7] = 0xaa5a5a5a,
>> + [8] = 0x6a5a5a5a,
>> + [9] = 0x6a5a5aaa,
>> + [10] = 0x6a5a6a5a,
>> + [11] = 0x6a5a6aaa,
>> + [12] = 0x6afa5afa,
>> + [13] = 0xaaaa5aaa,
>> + [14] = 0xaaffffaa,
>> + [15] = 0xaa5555aa,
>> + [16] = 0xfafafafa,
>> + [17] = 0xffffddff,
>> + [18] = 0xdaffdaff,
>> + [19] = 0xfafadafa
>> };
>> struct rtw89_btc_btf_tlv {
>
>
> Is this change really necessary? Yes, the entries must be ordered;
> however, the comment carries that information at very few extra
> characters. To me, this patch looks like unneeded source churn.

One small benefit I see is to avoid the comment index being wrong and
there would be no way to catch that. But otherwise I don't have any
opinion about this.

--
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

2022-03-28 16:18:38

by David Laight

Subject: RE: [PATCH 21/22] rtw89: Replace comments with C99 initializers

From: Kalle Valo
> Sent: 28 March 2022 10:29
>
> Larry Finger <[email protected]> writes:
>
> > On 3/26/22 11:59, Benjamin Stürz wrote:
> >> This replaces comments with C99's designated
> >> initializers because the kernel supports them now.
> >>
> >> Signed-off-by: Benjamin Stürz <[email protected]>
> >> ---
> >> drivers/net/wireless/realtek/rtw89/coex.c | 40 +++++++++++------------
> >> 1 file changed, 20 insertions(+), 20 deletions(-)
> >>
> >> diff --git a/drivers/net/wireless/realtek/rtw89/coex.c b/drivers/net/wireless/realtek/rtw89/coex.c
> >> index 684583955511..3c83a0bfb120 100644
> >> --- a/drivers/net/wireless/realtek/rtw89/coex.c
> >> +++ b/drivers/net/wireless/realtek/rtw89/coex.c
> >> @@ -97,26 +97,26 @@ static const struct rtw89_btc_fbtc_slot s_def[] = {
> >> };
> >> static const u32 cxtbl[] = {
> >> - 0xffffffff, /* 0 */
> >> - 0xaaaaaaaa, /* 1 */
> >> - 0x55555555, /* 2 */
> >> - 0x66555555, /* 3 */
> >> - 0x66556655, /* 4 */
> >> - 0x5a5a5a5a, /* 5 */
> >> - 0x5a5a5aaa, /* 6 */
> >> - 0xaa5a5a5a, /* 7 */
> >> - 0x6a5a5a5a, /* 8 */
> >> - 0x6a5a5aaa, /* 9 */
> >> - 0x6a5a6a5a, /* 10 */
> >> - 0x6a5a6aaa, /* 11 */
> >> - 0x6afa5afa, /* 12 */
> >> - 0xaaaa5aaa, /* 13 */
> >> - 0xaaffffaa, /* 14 */
> >> - 0xaa5555aa, /* 15 */
> >> - 0xfafafafa, /* 16 */
> >> - 0xffffddff, /* 17 */
> >> - 0xdaffdaff, /* 18 */
> >> - 0xfafadafa /* 19 */
> >> + [0] = 0xffffffff,
> >> + [1] = 0xaaaaaaaa,
> >> + [2] = 0x55555555,
> >> + [3] = 0x66555555,
> >> + [4] = 0x66556655,
> >> + [5] = 0x5a5a5a5a,
> >> + [6] = 0x5a5a5aaa,
> >> + [7] = 0xaa5a5a5a,
> >> + [8] = 0x6a5a5a5a,
> >> + [9] = 0x6a5a5aaa,
> >> + [10] = 0x6a5a6a5a,
> >> + [11] = 0x6a5a6aaa,
> >> + [12] = 0x6afa5afa,
> >> + [13] = 0xaaaa5aaa,
> >> + [14] = 0xaaffffaa,
> >> + [15] = 0xaa5555aa,
> >> + [16] = 0xfafafafa,
> >> + [17] = 0xffffddff,
> >> + [18] = 0xdaffdaff,
> >> + [19] = 0xfafadafa
> >> };
> >> struct rtw89_btc_btf_tlv {
> >
> >
> > Is this change really necessary? Yes, the entries must be ordered;
> > however, the comment carries that information at very few extra
> > characters. To me, this patch looks like unneeded source churn.
>
> One small benefit I see is to avoid the comment index being wrong and
> there would be no way to catch that. But otherwise I don't have any
> opinion about this.

If the [nn] are wrong the effect is probably worse.
You really don't want a gap!

Doesn't seem worth using C99 initialisers unless they are
#defines or enum values.

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)