In D61209#1403514, @manu wrote:

Revision (and commit ?) title is wrong :)

I tested the patch with my amdgpu GPU and it works fine.

I’m sorry for the delay to accept the patch. Again, I’m fairly sure I already reviewed it, but apparently don’t know how to click on buttons…

Thank you for the patch! I missed it when I updated other ports because I thought this one was tracking the drm-kmod master branch.

I’m sorry, I was sure I reviewed this patch with the other one in the stack…

A concurrent patch was committed two days ago; see D57444.

The patch looks good to me. Just one comment about a return value of linuxkpi_devm_memremap.

Address feedback from @bz.

I don’t have a better suggestion to make w.r.t. to the best location for this code.

I guess Phabricator ate the commit title because it does not mean much now :-)

Address concern from @bz

Address feedback from @kib

Sure, I will do that! I will just move the comment closer to the macro if you don’t mind. Thank you for the patch :-)

Abandoning this revision. D57465 is the same patch.

I have one small comment left.

Add __unused after unused arguent in linuxkpi_hrtimer_dummy_timeout() as suggested by @bz.

In drm-kmod, pci_map_rom() was defined as a macro inside C files, after all #include, so such a check won’t prevent a build failure in this specific case. I guess there are no other consumers of this API. Do you think it’s still useful to add it?

Fix new file license as suggested by @bz.

Fix build failure

Address feedback from @bz.

Based on the feedback from @bz:

• Fix return type, it should have been unsigned long.
• Improve comment to explain the source of the code.

Remove the #if LINUXKPI_VERSION ... check as suggested by @bz.

Indeed, perhaps we don’t neet it.

I’m not sure anything uses it from the Ports tree as all applications and librairies will certainly be designed around sysfs.

In D58765#1349262, @emaste wrote:

Seems like the right thing to do. Does linux hlist have any concurrency support?

Add a TODO.

Add comment to explain kthread_run_worker(), as suggested by @emaste.

Fix issue reported by @emaste

I didn't work on the implementation of dma-buf for FreeBSD. We have this BSD-licensed implementation in drm-kmod right now and we can surely move that to base. I find it receives changes infrequently in Linux, so hopefully, that shouldn't be something that slows down or interferes with making a given version of drm-kmod compatible with multiple versions of FreeBSD.