From: "Ard Biesheuvel" <ard.biesheuvel@arm.com>
To: Pete Batard <pete@akeo.ie>, devel@edk2.groups.io
Cc: leif@nuviainc.com, andrey.warkentin@gmail.com
Subject: Re: [edk2-devel] [PATCH edk2-platforms 4/5] Platform/RaspberryPi3: query firmware for 16550 input clock at boot
Date: Wed, 6 May 2020 12:31:36 +0200 [thread overview]
Message-ID: <4f8a19c3-80d9-ad1a-4915-69fe13182197@arm.com> (raw)
In-Reply-To: <2144540f-1b10-de24-95fa-09a0b2601cbb@akeo.ie>
On 5/6/20 12:18 PM, Pete Batard wrote:
> One remark below:
>
> On 2020.05.05 19:11, Ard Biesheuvel wrote:
>> On 5/5/20 4:50 PM, Ard Biesheuvel wrote:
>>> Query the firmware for the clock rate that is used to drive the
>>> 16550 baud clock, so that we can program the correct baud rate.
>>>
>>> Co-authored-by: Pete Batard <pete@akeo.ie>
>>> Co-authored-by: Andrei Warkentin <andrey.warkentin@gmail.com>
>>> Co-authored-by: Ard Biesheuvel <ard.biesheuvel@arm.com>
>>> Signed-off-by: Pete Batard <pete@akeo.ie>
>>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@arm.com>
>>> ---
>>>
>>> Platform/RaspberryPi/Library/PlatformLib/AArch64/RaspberryPiHelper.S
>>> | 25 +++++++++++++++++++-
>>> 1 file changed, 24 insertions(+), 1 deletion(-)
>>>
>>> diff --git
>>> a/Platform/RaspberryPi/Library/PlatformLib/AArch64/RaspberryPiHelper.S b/Platform/RaspberryPi/Library/PlatformLib/AArch64/RaspberryPiHelper.S
>>>
>>> index 91dfe1bb981e..35580e4ed73a 100644
>>> ---
>>> a/Platform/RaspberryPi/Library/PlatformLib/AArch64/RaspberryPiHelper.S
>>> +++
>>> b/Platform/RaspberryPi/Library/PlatformLib/AArch64/RaspberryPiHelper.S
>>> @@ -3,7 +3,7 @@
>>> * Copyright (c) 2020, Andrei Warkentin <andrey.warkentin@gmail.com>
>>> * Copyright (c) 2019-2020, Pete Batard <pete@akeo.ie>
>>> * Copyright (c) 2016, Linaro Limited. All rights reserved.
>>> - * Copyright (c) 2011-2013, ARM Limited. All rights reserved.
>>> + * Copyright (c) 2011-2020, ARM Limited. All rights reserved.
>>> *
>>> * SPDX-License-Identifier: BSD-2-Clause-Patent
>>> *
>>> @@ -85,6 +85,14 @@ ASM_FUNC (ArmPlatformPeiBootAction)
>>> adr x2, mBoardRevision
>>> str w0, [x2]
>>> +#if (RPI_MODEL == 3)
>>
>> As noted by Pete off-list, doing this doesn't work unless we add
>> something like
>>
>> GCC:*_*_*_PP_FLAGS = -DRPI_MODEL=3
>>
>> to the [BuildOptions] in RPi3.dsc
>>
>>
>>
>>> + run .Lclkinfo_buffer
>>> +
>>> + ldr w0, .Lfrequency
>>> + adrp x2, _gPcd_BinaryPatch_PcdSerialClockRate
>>> + str w0, [x2, :lo12:_gPcd_BinaryPatch_PcdSerialClockRate]
>>> +#endif
>>> +
>
> Since we're modifying a patchable PCD here, shouldn't we add a:
>
> [PatchPcd]
> gEfiMdeModulePkgTokenSpaceGuid.PcdSerialClockRate
>
> section in PlatformLib.inf?
>
We should add it, but we can add it to the [Pcd] section.
Alternatively, we could have different .INFs for RPi3 and RPi4, but that
is really overkill IMO.
Making it patchable on both platforms just to patch it on one is also
unnecessary I think. The current approach will ensure that we catch any
issues at build time, without any major hacks,].
> Of course, if we do that, we can't keep PcdSerialClockRate fixed for
> RPi4, as the build process will complain about PCD mismatch.
>
> I also wouldn't mind a comment that explains how one arrives at figuring
> out that "_gPcd_BinaryPatch_PcdSerialClockRate" should be used to locate
> our address (and possibly the addition of :lo12:), because I don't think
> it's going to be that straightforward for people reading the code for
> the first time, though I fear that the explanation will boil down to "we
> need to do it this specific way for a gcc aarch64 relocation"...
>
We don't actually need the adrp/str pair with the lo12 here, I will
replace it with adr. (Just muscle memory)
>>> ret
>>> .align 4
>>> @@ -127,6 +135,21 @@ ASM_FUNC (ArmPlatformPeiBootAction)
>>> .long 0 // end tag
>>> .set .Lrevinfo_size, . - .Lrevinfo_buffer
>>> +#if (RPI_MODEL == 3)
>>> + .align 4
>>> +.Lclkinfo_buffer:
>>> + .long .Lclkinfo_size
>>> + .long 0x0
>>> + .long RPI_MBOX_GET_CLOCK_RATE
>>> + .long 8 // buf size
>>> + .long 4 // input len
>>> + .long 4 // clock id: 0x04 = Core/VPU
>>> +.Lfrequency:
>>> + .long 0 // frequency
>>> + .long 0 // end tag
>>> + .set .Lclkinfo_size, . - .Lclkinfo_buffer
>>> +#endif
>>> +
>>> //UINTN
>>> //ArmPlatformGetPrimaryCoreMpId (
>>> // VOID
>>>
>>
>>
>>
>>
>
> With addition of "GCC:*_*_*_PP_FLAGS = -DRPI_MODEL=3" in the .dsc:
> Reviewed-by: Pete Batard <pete@akeo.ie>
>
Thanks
next prev parent reply other threads:[~2020-05-06 10:31 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-05 14:50 [PATCH edk2-platforms 0/5] Platform/RaspberryPi: fix serialportlib dependency hell Ard Biesheuvel
2020-05-05 14:50 ` [PATCH edk2-platforms 1/5] Platform/RaspberryPi/DualSerialPortLib: split up to ease reuse Ard Biesheuvel
2020-05-06 10:18 ` Pete Batard
2020-05-05 14:50 ` [PATCH edk2-platforms 2/5] Platform/RaspberryPi: introduce DebugDualSerialPortLib Ard Biesheuvel
2020-05-06 10:18 ` Pete Batard
2020-05-05 14:50 ` [PATCH edk2-platforms 3/5] Platform/RaspberryPi: fix 16550 divisor calculation logic Ard Biesheuvel
2020-05-05 18:10 ` Ard Biesheuvel
2020-05-06 10:18 ` [edk2-devel] " Pete Batard
2020-05-06 10:25 ` Ard Biesheuvel
2020-05-05 14:50 ` [PATCH edk2-platforms 4/5] Platform/RaspberryPi3: query firmware for 16550 input clock at boot Ard Biesheuvel
2020-05-05 18:11 ` Ard Biesheuvel
2020-05-06 10:18 ` [edk2-devel] " Pete Batard
2020-05-06 10:31 ` Ard Biesheuvel [this message]
2020-05-06 10:38 ` Pete Batard
2020-05-05 14:50 ` [PATCH edk2-platforms 5/5] Platform/RaspberryPi: create DXE phase SerialPortLib version for RPi3 Ard Biesheuvel
2020-05-06 10:19 ` Pete Batard
2020-05-06 16:16 ` [PATCH edk2-platforms 0/5] Platform/RaspberryPi: fix serialportlib dependency hell Ard Biesheuvel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4f8a19c3-80d9-ad1a-4915-69fe13182197@arm.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox