[PATCHv6,00/26] HSM rearch series for TI K3 devices

Message ID	20210611084527.7048-1-kristo@kernel.org
Headers	show Return-Path: <u-boot-bounces@lists.denx.de> From: Tero Kristo <kristo@kernel.org> To: lokeshvutla@ti.com, trini@konsulko.com, u-boot@lists.denx.de Subject: [PATCHv6 00/26] HSM rearch series for TI K3 devices Date: Fri, 11 Jun 2021 11:45:01 +0300 Message-Id: <20210611084527.7048-1-kristo@kernel.org> Precedence: list Errors-To: u-boot-bounces@lists.denx.de Sender: "U-Boot" <u-boot-bounces@lists.denx.de>
Series	HSM rearch series for TI K3 devices \| expand [PATCHv6,00/26] HSM rearch series for TI K3 devices [PATCHv6,01/26] lib: rational: copy the rational fraction lib routines from Linux [PATCHv6,02/26] arm: mach-k3: introduce new config option for sysfw split [PATCHv6,03/26] remoteproc: k3-r5: remove sysfw PM calls if not supported [PATCHv6,04/26] common: fit: Update board_fit_image_post_process() to pass fit and node_offset [PATCHv6,05/26] clk: fixed_rate: add API for directly registering fixed rate clocks [PATCHv6,06/26] clk: fix clock tree dump to properly dump out every registered clock [PATCHv6,07/26] clk: do not attempt to fetch clock pointer with null device [PATCHv6,08/26] clk: add support for setting clk rate from cmdline [PATCHv6,09/26] clk: sci-clk: fix return value of set_rate [PATCHv6,10/26] clk: fix assigned-clocks to pass with deferring provider [PATCHv6,11/26] clk: fix set_rate to clean up cached rates for the hierarchy [PATCHv6,12/26] clk: add support for TI K3 SoC PLL [PATCHv6,13/26] clk: add support for TI K3 SoC clocks [PATCHv6,14/26] power: domain: Introduce driver for raw TI K3 PDs [PATCHv6,15/26] cmd: ti: pd: Add debug command for K3 power domains [PATCHv6,16/26] tools: k3_fit_atf: add DM binary to the FIT image [PATCHv6,17/26] arm: mach-k3: Add platform data for j721e and j7200 [PATCHv6,18/26] arm: mach-k3: add support for detecting firmware images from FIT [PATCHv6,19/26] arm: mach-k3: do board config for PM only if supported [PATCHv6,20/26] arm: mach-k3: common: Drop main r5 start [PATCHv6,21/26] arm: mach-k3: sysfw-loader: pass boardcfg to sciserver [PATCHv6,22/26] arm: mach-k3: j721e_init: Force early probe of clk-k3 driver [PATCHv6,23/26] configs: j721e_evm_r5: Enable raw access power management features [PATCHv6,24/26] configs: j7200_evm_r5: Enable raw access power management features [PATCHv6,25/26] board: ti: j72xx: README: update build instructions and image formats [PATCHv6,26/26] arm: dts: k3-j72xx: correct MCU timer1 frequency

Message ID

20210611084527.7048-1-kristo@kernel.org

Headers

From: Tero Kristo <kristo@kernel.org>
To: lokeshvutla@ti.com,
	trini@konsulko.com,
	u-boot@lists.denx.de
Subject: [PATCHv6 00/26] HSM rearch series for TI K3 devices
Date: Fri, 11 Jun 2021 11:45:01 +0300
Message-Id: <20210611084527.7048-1-kristo@kernel.org>
Precedence: list
Errors-To: u-boot-bounces@lists.denx.de
Sender: "U-Boot" <u-boot-bounces@lists.denx.de>

Series

HSM rearch series for TI K3 devices | expand

Message

Tero Kristo June 11, 2021, 8:45 a.m. UTC

Hello,

One more post, this time with the #ifdef hackery converted to use the
IS_ENABLED / CONFIG_IS_ENABLED macros, and also removed the "common.h"
include from k3-clk.h header. This version also contains fixes to any
build issues reported by Lokesh, and these are squashed in to relevant
patches.

-Tero

Comments

Lokesh Vutla June 11, 2021, 11:08 a.m. UTC | #1

Hi Tero,

On 11/06/21 2:15 pm, Tero Kristo wrote:
> Hello,
> 
> One more post, this time with the #ifdef hackery converted to use the
> IS_ENABLED / CONFIG_IS_ENABLED macros, and also removed the "common.h"
> include from k3-clk.h header. This version also contains fixes to any
> build issues reported by Lokesh, and these are squashed in to relevant
> patches.

Can you see if the below warnings can be fixed?

hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch
-------------------------------------------------------------------
WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#26: FILE: arch/arm/mach-k3/common.c:31:
+#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#35: FILE: arch/arm/mach-k3/common.c:40:
+#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)

WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#55: FILE: arch/arm/mach-k3/common.c:131:
+#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#124: FILE: arch/arm/mach-k3/common.c:264:
+#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)

WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#128: FILE: arch/arm/mach-k3/common.c:268:
+#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#150: FILE: arch/arm/mach-k3/common.c:290:
+#if IS_ENABLED(CONFIG_TI_SECURE_DEVICE)

total: 0 errors, 6 warnings, 0 checks, 144 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.

hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch has style
problems, please review.
-------------------------------------------------------------------
hsm/0019-arm-mach-k3-do-board-config-for-PM-only-if-supported.patch
-------------------------------------------------------------------
WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
#24: FILE: arch/arm/mach-k3/sysfw-loader.c:162:
+#if !CONFIG_IS_ENABLED(K3_DM_FW)

total: 0 errors, 1 warnings, 0 checks, 13 lines checked

NOTE: For some of the reported defects, checkpatch may be able to
      mechanically convert to the typical style using --fix or --fix-inplace.


Thanks and regards,
Lokesh

Tero Kristo June 11, 2021, 11:18 a.m. UTC | #2

On 11/06/2021 14:08, Lokesh Vutla wrote:
> Hi Tero,
> 
> On 11/06/21 2:15 pm, Tero Kristo wrote:
>> Hello,
>>
>> One more post, this time with the #ifdef hackery converted to use the
>> IS_ENABLED / CONFIG_IS_ENABLED macros, and also removed the "common.h"
>> include from k3-clk.h header. This version also contains fixes to any
>> build issues reported by Lokesh, and these are squashed in to relevant
>> patches.
> 
> Can you see if the below warnings can be fixed?
> 
> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch
> -------------------------------------------------------------------
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #26: FILE: arch/arm/mach-k3/common.c:31:
> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

This is static data, can't be fixed. Unless we want to compile it in always?

> 
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #35: FILE: arch/arm/mach-k3/common.c:40:
> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)

Same, static data.

> 
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #55: FILE: arch/arm/mach-k3/common.c:131:
> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

This is actually just an old macro I changed from #ifdef to IS_ENABLED. 
Fixing the whole file from the existing #ifdef:s should be outside the 
scope of this series.

> 
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #124: FILE: arch/arm/mach-k3/common.c:264:
> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)

This code addresses the static data defined before, changing this will 
break compilation; unless we compile the data always in.

> 
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #128: FILE: arch/arm/mach-k3/common.c:268:
> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)

Same as above.

> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #150: FILE: arch/arm/mach-k3/common.c:290:
> +#if IS_ENABLED(CONFIG_TI_SECURE_DEVICE)

This can't be changed, the code it addresses is only linked in with the 
config, causing a linker time failure if this is fixed.

Imho, I am not too convinced about the checkpatch tool complaining about 
these issues. :)

-Tero

> 
> total: 0 errors, 6 warnings, 0 checks, 144 lines checked
> 
> NOTE: For some of the reported defects, checkpatch may be able to
>        mechanically convert to the typical style using --fix or --fix-inplace.
> 
> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch has style
> problems, please review.
> -------------------------------------------------------------------
> hsm/0019-arm-mach-k3-do-board-config-for-PM-only-if-supported.patch
> -------------------------------------------------------------------
> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where possible
> #24: FILE: arch/arm/mach-k3/sysfw-loader.c:162:
> +#if !CONFIG_IS_ENABLED(K3_DM_FW)
> 
> total: 0 errors, 1 warnings, 0 checks, 13 lines checked
> 
> NOTE: For some of the reported defects, checkpatch may be able to
>        mechanically convert to the typical style using --fix or --fix-inplace.
> 
> 
> Thanks and regards,
> Lokesh
>

Lokesh Vutla June 11, 2021, 1:43 p.m. UTC | #3

Tom,

On 11/06/21 4:48 pm, Tero Kristo wrote:
> On 11/06/2021 14:08, Lokesh Vutla wrote:
>> Hi Tero,
>>
>> On 11/06/21 2:15 pm, Tero Kristo wrote:
>>> Hello,
>>>
>>> One more post, this time with the #ifdef hackery converted to use the
>>> IS_ENABLED / CONFIG_IS_ENABLED macros, and also removed the "common.h"
>>> include from k3-clk.h header. This version also contains fixes to any
>>> build issues reported by Lokesh, and these are squashed in to relevant
>>> patches.
>>
>> Can you see if the below warnings can be fixed?
>>
>> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch
>> -------------------------------------------------------------------
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #26: FILE: arch/arm/mach-k3/common.c:31:
>> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> 
> This is static data, can't be fixed. Unless we want to compile it in always?
> 
>>
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #35: FILE: arch/arm/mach-k3/common.c:40:
>> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)
> 
> Same, static data.
> 
>>
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #55: FILE: arch/arm/mach-k3/common.c:131:
>> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> 
> This is actually just an old macro I changed from #ifdef to IS_ENABLED. Fixing
> the whole file from the existing #ifdef:s should be outside the scope of this
> series.
> 
>>
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #124: FILE: arch/arm/mach-k3/common.c:264:
>> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)
> 
> This code addresses the static data defined before, changing this will break
> compilation; unless we compile the data always in.
> 
>>
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #128: FILE: arch/arm/mach-k3/common.c:268:
>> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> 
> Same as above.
> 
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #150: FILE: arch/arm/mach-k3/common.c:290:
>> +#if IS_ENABLED(CONFIG_TI_SECURE_DEVICE)
> 
> This can't be changed, the code it addresses is only linked in with the config,
> causing a linker time failure if this is fixed.
> 
> Imho, I am not too convinced about the checkpatch tool complaining about these
> issues. :)
> 
> -Tero
> 
>>
>> total: 0 errors, 6 warnings, 0 checks, 144 lines checked
>>
>> NOTE: For some of the reported defects, checkpatch may be able to
>>        mechanically convert to the typical style using --fix or --fix-inplace.
>>
>> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch has style
>> problems, please review.
>> -------------------------------------------------------------------
>> hsm/0019-arm-mach-k3-do-board-config-for-PM-only-if-supported.patch
>> -------------------------------------------------------------------
>> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
>> possible
>> #24: FILE: arch/arm/mach-k3/sysfw-loader.c:162:
>> +#if !CONFIG_IS_ENABLED(K3_DM_FW)
>>
>> total: 0 errors, 1 warnings, 0 checks, 13 lines checked
>>
>> NOTE: For some of the reported defects, checkpatch may be able to
>>        mechanically convert to the typical style using --fix or --fix-inplace.

I assume you are okay with this?

Thanks and regards,
Lokesh

>>
>>
>> Thanks and regards,
>> Lokesh
>>
>

Tom Rini June 11, 2021, 2:13 p.m. UTC | #4

On Fri, Jun 11, 2021 at 07:13:19PM +0530, Lokesh Vutla wrote:
> Tom,
> 
> On 11/06/21 4:48 pm, Tero Kristo wrote:
> > On 11/06/2021 14:08, Lokesh Vutla wrote:
> >> Hi Tero,
> >>
> >> On 11/06/21 2:15 pm, Tero Kristo wrote:
> >>> Hello,
> >>>
> >>> One more post, this time with the #ifdef hackery converted to use the
> >>> IS_ENABLED / CONFIG_IS_ENABLED macros, and also removed the "common.h"
> >>> include from k3-clk.h header. This version also contains fixes to any
> >>> build issues reported by Lokesh, and these are squashed in to relevant
> >>> patches.
> >>
> >> Can you see if the below warnings can be fixed?
> >>
> >> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch
> >> -------------------------------------------------------------------
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #26: FILE: arch/arm/mach-k3/common.c:31:
> >> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> > 
> > This is static data, can't be fixed. Unless we want to compile it in always?
> > 
> >>
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #35: FILE: arch/arm/mach-k3/common.c:40:
> >> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)
> > 
> > Same, static data.
> > 
> >>
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #55: FILE: arch/arm/mach-k3/common.c:131:
> >> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> > 
> > This is actually just an old macro I changed from #ifdef to IS_ENABLED. Fixing
> > the whole file from the existing #ifdef:s should be outside the scope of this
> > series.
> > 
> >>
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #124: FILE: arch/arm/mach-k3/common.c:264:
> >> +#if CONFIG_IS_ENABLED(FIT_IMAGE_POST_PROCESS)
> > 
> > This code addresses the static data defined before, changing this will break
> > compilation; unless we compile the data always in.
> > 
> >>
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #128: FILE: arch/arm/mach-k3/common.c:268:
> >> +#if IS_ENABLED(CONFIG_SYS_K3_SPL_ATF)
> > 
> > Same as above.
> > 
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #150: FILE: arch/arm/mach-k3/common.c:290:
> >> +#if IS_ENABLED(CONFIG_TI_SECURE_DEVICE)
> > 
> > This can't be changed, the code it addresses is only linked in with the config,
> > causing a linker time failure if this is fixed.
> > 
> > Imho, I am not too convinced about the checkpatch tool complaining about these
> > issues. :)
> > 
> > -Tero
> > 
> >>
> >> total: 0 errors, 6 warnings, 0 checks, 144 lines checked
> >>
> >> NOTE: For some of the reported defects, checkpatch may be able to
> >>        mechanically convert to the typical style using --fix or --fix-inplace.
> >>
> >> hsm/0018-arm-mach-k3-add-support-for-detecting-firmware-image.patch has style
> >> problems, please review.
> >> -------------------------------------------------------------------
> >> hsm/0019-arm-mach-k3-do-board-config-for-PM-only-if-supported.patch
> >> -------------------------------------------------------------------
> >> WARNING: Use 'if (IS_ENABLED(CONFIG...))' instead of '#if or #ifdef' where
> >> possible
> >> #24: FILE: arch/arm/mach-k3/sysfw-loader.c:162:
> >> +#if !CONFIG_IS_ENABLED(K3_DM_FW)
> >>
> >> total: 0 errors, 1 warnings, 0 checks, 13 lines checked
> >>
> >> NOTE: For some of the reported defects, checkpatch may be able to
> >>        mechanically convert to the typical style using --fix or --fix-inplace.
> 
> I assume you are okay with this?

Well, if all the cases where we should have been using
IS_ENABLED/CONFIG_IS_ENABLED but weren't have been fixed, and we aren't
adding <common.h> to more files, then yes, everything else that can be
done as a later clean-up can just be done as a later clean-up, and
indeed if it's not possible to use an if (...) it's not possible to use
an if :)