RFC on writel and writel_relaxed

[Sinan Kaya]

On 3/26/2018 8:11 AM, okaya at codeaurora.org wrote:
> On 2018-03-26 07:44, Will Deacon wrote:
>> Hi Ben,
>>
>> I don't seem to have the beginning of this thread, so please bounce it over
>> if you'd like me to look at it!
>>
> 
> https://www.spinics.net/lists/linux-rdma/msg62570.html
> 
> https://www.spinics.net/lists/linux-rdma/index.html#62666
> 

To add some more details on why we are looking at this now:

I posted several patches last week to remove duplicate barriers on ARM while
trying to make the code friendly with other architectures.

https://www.spinics.net/lists/netdev/msg491842.html
https://www.spinics.net/lists/linux-rdma/msg62434.html
https://www.spinics.net/lists/arm-kernel/msg642336.html

The conversation on this thread is interesting.

https://patchwork.kernel.org/patch/10288987/

1. I tried to replace wmb()+writel() with wmb()+writel_relaxed().
2. writel_relaxed() is equal to writel() at this moment for PPC.
3. Chelsio developers wanted to pull it into wmb()+__raw_writel() direction
to take advantage of the same optimization for PPC.
4. Dave informed us that behavior of __raw_write() is not identical on all
architectures.
5. We decided to go back to PPC and ask to implement writel_relaxed()
instead of coming up with writel_realy_relaxed() API.

> 
>> On Fri, Mar 23, 2018 at 11:16:08AM +1100, Benjamin Herrenschmidt wrote:
>>> On Thu, 2018-03-22 at 12:51 -0500, Sinan Kaya wrote:
>>> > On 3/22/2018 8:52 AM, Benjamin Herrenschmidt wrote:
>>> > > > > No, it's not sufficient.
>>> > >
>>> > > Just to clarify ... barrier() is just a compiler barrier, it means the
>>> > > compiler will generate things in the order they are written. This isn't
>>> > > sufficient on archs with an OO memory model, where an actual memory
>>> > > barrier instruction needs to be emited.
>>> >
>>> > Surprisingly, ARM64 GCC compiler generates a write barrier as
>>> > opposed to preventing code reordering.
>>
>> In context, this looks like a misunderstanding somewhere. barrier() is a
>> compiler barrier for us just like everybody else and we use the generic
>> implementation with the empty asm + memory clobber.
>>
> 
> True, I clarified it this weekend
> 
> https://www.spinics.net/lists/linux-rdma/msg62788.html
> 
> 
> 


-- 
Sinan Kaya
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.