Last minute fix...
AR5K_CPCORN is changed to AR5K_CPCOVF in reg list update so we have to update debug.c in order to compile
Changes-licensed-under: ISC
Signed-off-by: Nick Kossifidis <[email protected]>
---
diff --git a/drivers/net/wireless/ath5k/debug.c b/drivers/net/wireless/ath5k/debug.c
index 41d5fa3..6fa6c8e 100644
--- a/drivers/net/wireless/ath5k/debug.c
+++ b/drivers/net/wireless/ath5k/debug.c
@@ -129,7 +129,7 @@ static struct reg regs[] = {
REG_STRUCT_INIT(AR5K_CPC1),
REG_STRUCT_INIT(AR5K_CPC2),
REG_STRUCT_INIT(AR5K_CPC3),
- REG_STRUCT_INIT(AR5K_CPCORN),
+ REG_STRUCT_INIT(AR5K_CPCOVF),
REG_STRUCT_INIT(AR5K_RESET_CTL),
REG_STRUCT_INIT(AR5K_SLEEP_CTL),
REG_STRUCT_INIT(AR5K_INTPEND),
Why the hell the mailer cropped Nick's address away...
On 07/20/2008 05:52 AM, Nick Kossifidis wrote:
> Last minute fix...
>
> AR5K_CPCORN is changed to AR5K_CPCOVF in reg list update so we have to update debug.c in order to compile
>
> Changes-licensed-under: ISC
> Signed-off-by: Nick Kossifidis <[email protected]>
>
> ---
> diff --git a/drivers/net/wireless/ath5k/debug.c b/drivers/net/wireless/ath5k/debug.c
> index 41d5fa3..6fa6c8e 100644
> --- a/drivers/net/wireless/ath5k/debug.c
> +++ b/drivers/net/wireless/ath5k/debug.c
> @@ -129,7 +129,7 @@ static struct reg regs[] = {
> REG_STRUCT_INIT(AR5K_CPC1),
> REG_STRUCT_INIT(AR5K_CPC2),
> REG_STRUCT_INIT(AR5K_CPC3),
> - REG_STRUCT_INIT(AR5K_CPCORN),
> + REG_STRUCT_INIT(AR5K_CPCOVF),
> REG_STRUCT_INIT(AR5K_RESET_CTL),
> REG_STRUCT_INIT(AR5K_SLEEP_CTL),
> REG_STRUCT_INIT(AR5K_INTPEND),
Please amend this to the patch 1 to not break bisectability.
2008/7/22 Jiri Slaby <[email protected]>:
> On 07/20/2008 05:52 AM, Nick Kossifidis wrote:
>> Last minute fix...
>>
>> AR5K_CPCORN is changed to AR5K_CPCOVF in reg list update so we have to update debug.c in order to compile
>>
>> Changes-licensed-under: ISC
>> Signed-off-by: Nick Kossifidis <[email protected]>
>>
>> ---
>> diff --git a/drivers/net/wireless/ath5k/debug.c b/drivers/net/wireless/ath5k/debug.c
>> index 41d5fa3..6fa6c8e 100644
>> --- a/drivers/net/wireless/ath5k/debug.c
>> +++ b/drivers/net/wireless/ath5k/debug.c
>> @@ -129,7 +129,7 @@ static struct reg regs[] = {
>> REG_STRUCT_INIT(AR5K_CPC1),
>> REG_STRUCT_INIT(AR5K_CPC2),
>> REG_STRUCT_INIT(AR5K_CPC3),
>> - REG_STRUCT_INIT(AR5K_CPCORN),
>> + REG_STRUCT_INIT(AR5K_CPCOVF),
>> REG_STRUCT_INIT(AR5K_RESET_CTL),
>> REG_STRUCT_INIT(AR5K_SLEEP_CTL),
>> REG_STRUCT_INIT(AR5K_INTPEND),
>
> Please amend this to the patch 1 to not break bisectability.
I see your point but this series is meant to be applied as a whole, i
didn't notice this bug until i tested the code with debugging enabled
(which is off by default), that's why it's "last minute fix". Is it
really that much needed to repost the first patch ? I mean the patch
series works and doesn't introduce any problems...
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick
On 07/20/2008 05:52 AM, Nick Kossifidis wrote:
> Last minute fix...
>
> AR5K_CPCORN is changed to AR5K_CPCOVF in reg list update so we have to update debug.c in order to compile
>
> Changes-licensed-under: ISC
> Signed-off-by: Nick Kossifidis <[email protected]>
>
> ---
> diff --git a/drivers/net/wireless/ath5k/debug.c b/drivers/net/wireless/ath5k/debug.c
> index 41d5fa3..6fa6c8e 100644
> --- a/drivers/net/wireless/ath5k/debug.c
> +++ b/drivers/net/wireless/ath5k/debug.c
> @@ -129,7 +129,7 @@ static struct reg regs[] = {
> REG_STRUCT_INIT(AR5K_CPC1),
> REG_STRUCT_INIT(AR5K_CPC2),
> REG_STRUCT_INIT(AR5K_CPC3),
> - REG_STRUCT_INIT(AR5K_CPCORN),
> + REG_STRUCT_INIT(AR5K_CPCOVF),
> REG_STRUCT_INIT(AR5K_RESET_CTL),
> REG_STRUCT_INIT(AR5K_SLEEP_CTL),
> REG_STRUCT_INIT(AR5K_INTPEND),
Please amend this to the patch 1 to not break bisectability.
On Mon, 2008-07-28 at 16:04 +0300, Nick Kossifidis wrote:
> > Please amend this to the patch 1 to not break bisectability.
>
> I see your point but this series is meant to be applied as a whole, i
> didn't notice this bug until i tested the code with debugging enabled
> (which is off by default), that's why it's "last minute fix". Is it
> really that much needed to repost the first patch ? I mean the patch
> series works and doesn't introduce any problems...
The reason is that git-bisect divides the number of commits in half and
checks out the middle commit. That commit make be a broken one.
Hitting one bug while hunting another is not nice.
Besides, nobody wants to apply a known bad patch.
It would be great if you repost fixed patches. If you are not using
stgit, please try it. I would make it easy to return to older patches
and change them. "stg edit -d" even allows editing the patch.
--
Regards,
Pavel Roskin
2008/7/28 Pavel Roskin <[email protected]>:
> On Mon, 2008-07-28 at 16:04 +0300, Nick Kossifidis wrote:
>
>> > Please amend this to the patch 1 to not break bisectability.
>>
>> I see your point but this series is meant to be applied as a whole, i
>> didn't notice this bug until i tested the code with debugging enabled
>> (which is off by default), that's why it's "last minute fix". Is it
>> really that much needed to repost the first patch ? I mean the patch
>> series works and doesn't introduce any problems...
>
> The reason is that git-bisect divides the number of commits in half and
> checks out the middle commit. That commit make be a broken one.
> Hitting one bug while hunting another is not nice.
>
> Besides, nobody wants to apply a known bad patch.
>
> It would be great if you repost fixed patches. If you are not using
> stgit, please try it. I would make it easy to return to older patches
> and change them. "stg edit -d" even allows editing the patch.
>
> --
> Regards,
> Pavel Roskin
>
O.K. then i'll repost the first patch asap (i'm still in Ottawa :P)...
--
GPG ID: 0xD21DB2DB
As you read this post global entropy rises. Have Fun ;-)
Nick