In D52440#1198791, @imp wrote:

So this change is a bit larger than I like to review. But the moved code was simple (and I know I have comment about the moved code, and that maybe the NVMe driver isn't good about doing it true). In the future, though it would make it easiest if you had the 'move it' commits separate from the rest. The place I want a comment could get one later too, if that's easier to do.

Looking at the recovery algorithm, it does seem quite similar to the NVMe one. The only concern I have here is that we'll need to track it in two places. I've been tweaking it from time to time in the nvme driver, but I don't know how applicable future tweaks will be. I guess this is more of a "be aware" in the future comment.

Add a comment

update commit log

Fix queue contruct sequence

I expected the timeout handling to be in the CAM, but it seems the CAM only performs a retry and does not execute cmd abort and controller reset. Therefore, I implemented the cmd abort and controller reset within the ufshci driver. Is my understanding correct?

Added some comments for reviewers.

fix indent

add a newline at end of file

In D51893#1193102, @imp wrote:

One question, and one minor problem... Otherwise, I think this is fine. It's a feature that's enable on attach and then everything about it happens automatically, right?

rebase to the head

Add error return

Add error log

Could you please submit this patch? Thanks!

Could you please submit this patch? Thanks!

Thank you!

In D50933#1172976, @imp wrote:

In D50933#1172975, @j_yoon.choi_samsung.com wrote:

In D50933#1172970, @imp wrote:

committeed

I had hoped the test box I ordered to test the UFS stuff would have arrived by now since I wanted to test boot this first... but it's held up in customs, it seems, so after yet another delay I'll push this an the man page in.

Would you mind sharing which test box you ordered?
I think it would be easier to troubleshoot if we have the same device.

Minisforum S100-US Intel® N100 / 8GB RAM+256GB SSD / US is what I ordered.
https://store.minisforum.com/products/minisforum-s100?_pos=1&_psq=Minisforum+S100-US&_ss=e&_v=1.0
though it's showing out of stock at the moment... I ordered mine about a month and a half ago now and had to wait until now to get it (now the tracking is telling me it will be here any day... Monday the tracking info made it sound like it was weeks away).

In D50933#1172970, @imp wrote:

committeed

I had hoped the test box I ordered to test the UFS stuff would have arrived by now since I wanted to test boot this first... but it's held up in customs, it seems, so after yet another delay I'll push this an the man page in.

In D51029#1172968, @imp wrote:

Committed...

Could you please commit this patch?

Could you please commit this patch?

Apply review comment

Apply review comment

initialisation -> initialization

Add ufshci to Makefile

Use full name of driver

In D50370#1160086, @imp wrote:

In D50370#1160085, @j_yoon.choi_samsung.com wrote:

In D50370#1159626, @j_yoon.choi_samsung.com wrote:

In D50370#1159535, @imp wrote:

I think this is ready to commit (or at least to start to try)... What do you think?

Yes, I think this is ready to commit!

I don't understand the commit process. Is there anything I need to do to commit?

Since you don't yet have commit permissions, I'll need to do that. I'll download it to my tree, make sure it compiles everywhere (or is disabled) and then push it into the repo. I just wanted to make sure that you didn't have anything else planned before I started that process...

In D50370#1159626, @j_yoon.choi_samsung.com wrote:

In D50370#1159535, @imp wrote:

I think this is ready to commit (or at least to start to try)... What do you think?

Yes, I think this is ready to commit!

In D50370#1159535, @imp wrote:

I think this is ready to commit (or at least to start to try)... What do you think?

Change UIC power mode ready time from 500ms to 2000ms

Apply review comments.

Apply review comments

Thank you for the kind review.
Your advice was a great help in preparing the patch. :)

Apply review comment, Remove unnecessary functions

Rename functions

Fix a uic command bug

I measured the key performance metrics on the real device and confirmed that the results meet my goals.
For Write, if WriteBooster is applied, the bandwitch will increase.
Here are the results I measured on each device.

Fix a memory leak bug

Remove debug code

This patch is now ready for full review.
Thank you for your patience.

Apply clang-format

The UFSHCI driver now works on the Samsung Galaxy Book S and Lenovo Duet3.

Check the completion notification register when processing request completion

Number of bugs found during testing on Lenovo duet3

Fix a bug in interrupt enable register, Add Lenovo Duet3 11ian device id

Refactoring quirks

Move the quirk check routine to PCI attach.

The latest version of the UFS spec is paid, but you can download version 2.1 for free. Except for the MCQs, 4.1 and 2.1 are almost identical.
Please refer to them in your review.

Apply review comments

Thanks for the kind review! Most comments have been applied.
I'll mark the remaining ones as done after applying the review.
Thanks again!

Apply clang-format

In D50370#1151862, @emaste wrote:

We've updated style(9) to permit // comments but don't want styles mixed within a file/component. The //s here look like mostly temporary/debug code that won't be committed.

Comments starting with // may be used in both C and C++.  Styles should
not be mixed within a file or component.

There's also a file for clang-format which should produce acceptable results.

We can focus on architectural / high level approach review initially and do a style(9) pass once ready to commit.

In D50370#1151827, @jrtc27 wrote:

Obligatory style review:

• I see commented out code; either delete it or use #if(def) to gate it with an internal debug define
• FreeBSD uses /* C-style comments */ not // C++-style comments but I see a few instances of the latter
• Continuation lines aren't being consistently indented (I'm struggling to figure out what rule you've followed, even); they should be indented by four spaces relative to the block-level indent
• For combinations of bitwise operations etc where the operator precedence can be unintuitive we generally include redundant parentheses to avoid the need to think about whether it's correct without them (both for reading and writing)

You may find tools/build/checkstyle9.pl helpful, it can probably tell you about some of these, and maybe more I've not spotted. As an adaptation of QEMU's style checker it's not perfect though, so if it looks to be producing unexpected output then feel free to ignore those suggestions.

In D50370#1151769, @adrian wrote:

first pass - this on the surface looks like a well formatted / written driver. no magic numbers, consistent indenting, commented code, handling errors, thanks!

Add manual completion, Add comments to req_queue

Update UFS/UFSHCI version to 4.1, Add a Intel Lakefield quirk

Fix typo, update data_direction in UTRD, Add delay for HCE enable

I’d like to request an early review.
The items listed below are still incomplete, but I plan to finish them and update the patch while the review is in progress.