Re: [PATCH 1/2] tty: pl011: Work around QDF2400 E44 stuck BUSY bit

From	Christopher Covington <cov@codeaurora.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH 1/2] tty: pl011: Work around QDF2400 E44 stuck BUSY bit
Date	2017-02-15 00:50 +0100
Message-ID	<taV18-8sF-17@gated-at.bofh.it> (permalink)
References	<t8oM1-1D5-3@gated-at.bofh.it> <t8rJT-3H5-5@gated-at.bofh.it> <t8IUq-67n-21@gated-at.bofh.it> <t8Jx8-6AK-15@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 02/08/2017 06:04 PM, Timur Tabi wrote:
> On 02/08/2017 04:22 PM, Christopher Covington wrote:
>>>> >> -    while (pl011_read(uap, REG_FR) & uap->vendor->fr_busy)
>>>> >> +    while ((pl011_read(uap, REG_FR) ^ uap->vendor->inv_fr)
>>>> >> +                        & uap->vendor->fr_busy)
>>> >
>>> > I really think the XOR logic needs to be documented wherever it's
>>> > used. It's just too confusing.
>> References such as the following for basic bit operations like
>> setting, clearing, and toggling are easy enough to come by.
>>
>> http://stackoverflow.com/questions/47981/how-do-you-set-clear-and-toggle-a-single-bit-in-c-c
> 
> I know what XOR does. I was thinking a more high-level comment would
> be appropriate. With E44, we want to ignore the BUSY bit and use TXFE
> instead. So was hoping to a comment that says that:
> 
> /* Normally, we poll until BUSY=0, but E44 says we should poll until
> TXFE=1 instead. So with E44, we set fr_busy to TXFE, but we have to
> invert it. */

Oh, yeah, that'd be good. I'll include better comments in v2.

Cheers,
Cov

-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code
Aurora Forum, a Linux Foundation Collaborative Project.

Back to linux.kernel | Previous | Next | Find similar | Unroll thread

Thread

Re: [PATCH 1/2] tty: pl011: Work around QDF2400 E44 stuck BUSY bit Christopher Covington <cov@codeaurora.org> - 2017-02-15 00:50 +0100

csiph-web