[lkml]   [2020]   [Jan]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH 05/16] bus: mhi: core: Add support for ringing channel/event ring doorbells
On 1/23/2020 5:00 AM, Manivannan Sadhasivam wrote:
> Hi Arnd,
> On Thu, Jan 23, 2020 at 12:39:06PM +0100, Arnd Bergmann wrote:
>> On Thu, Jan 23, 2020 at 12:19 PM Manivannan Sadhasivam
>> <> wrote:
>>> +int __must_check mhi_read_reg(struct mhi_controller *mhi_cntrl,
>>> + void __iomem *base, u32 offset, u32 *out)
>>> +{
>>> + u32 tmp = readl_relaxed(base + offset);
>> ....
>>> +void mhi_write_reg(struct mhi_controller *mhi_cntrl, void __iomem *base,
>>> + u32 offset, u32 val)
>>> +{
>>> + writel_relaxed(val, base + offset);
>> Please avoid using _relaxed accessors by default, and use the regular
>> ones instead. There are a number of things that can go wrong with
>> the relaxed version, so ideally each caller should have a comment
>> explaining why this instance is safe without the barriers and why it
>> matters to not have it.
>> If there are performance critical callers of mhi_read_reg/mhi_write_reg,
>> you could add mhi_read_reg_relaxed/mhi_write_reg_relaxed for those
>> and apply the same rules there.
>> Usually most mmio accesses are only needed for reconfiguration or
>> other slow paths.
> Fair point. I'll defer to readl/writel APIs and I also need to add
> le32_to_cpu/cpu_to_le32 to them.

I would expect we would be using these in the "hot" path.

I'm a bit confused, I thought the convention was to put a comment why a
barrier was necessary, now we should be putting a comment why a barrier
is not necessary?

Jeffrey Hugo
Qualcomm Technologies, Inc. is a member of the
Code Aurora Forum, a Linux Foundation Collaborative Project.

 \ /
  Last update: 2020-01-23 15:45    [W:0.118 / U:23.960 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site