Announcement

Collapse
No announcement yet.

AMD Drops TGSI Support From Their RadeonSI OpenGL Driver

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Veerappan
    replied
    Originally posted by sandy8925 View Post

    Is ACO usable for RadeonSi? AFAIK it's only used for RADV.
    I think the person you're replying to was hoping that ACO is expanded to being used in radeonsi instead of the LLVM backend. I guess it could happen, but I guess I'd personally rather that compile/linking performance improvements be worked on for LLVM to make it faster, since I also use the LLVM backend for clover state tracker stuff in mesa.

    Leave a comment:


  • Guest
    Guest replied
    Originally posted by puleglot View Post
    That's not very impressive. Still waiting for ACO in radeonsi. =P
    Is ACO usable for RadeonSi? AFAIK it's only used for RADV.

    Leave a comment:


  • puleglot
    replied
    That's not very impressive. Still waiting for ACO in radeonsi. =P

    Leave a comment:


  • Kayden
    replied
    Originally posted by Veerappan View Post

    I was going to ask basically the same question, but for the video-decoding state trackers. Hopefully the TGSI->NIR pass doesn't have any oddball corner cases that we haven't stumbled upon yet.
    Yeah, that's another reason for TGSI->NIR. If there are any, we'll have to fix them :)

    Leave a comment:


  • Veerappan
    replied
    Originally posted by Kayden View Post

    Nine should continue to work—radeonsi will translate the TGSI supplied by nine to NIR via tgsi_to_nir, like iris does. So everything is still supported, but now everything goes through NIR and there's one fewer path in the driver.

    (I wondered this myself, so I looked at the patches earlier.)
    I was going to ask basically the same question, but for the video-decoding state trackers. Hopefully the TGSI->NIR pass doesn't have any oddball corner cases that we haven't stumbled upon yet.

    Leave a comment:


  • R41N3R
    replied
    I believe this is the right decision, the NIR environment variable caused some confusion in the past, so it's good to drop TGSI and to focus on just one path.

    Leave a comment:


  • geearf
    replied
    Originally posted by Kayden View Post

    Nine should continue to work—radeonsi will translate the TGSI supplied by nine to NIR via tgsi_to_nir, like iris does. So everything is still supported, but now everything goes through NIR and there's one fewer path in the driver.

    (I wondered this myself, so I looked at the patches earlier.)
    Awesome, thank you for looking it up and answering!

    Leave a comment:


  • Kayden
    replied
    Originally posted by geearf View Post
    Is there an impact to certain state trackers, like nine?
    Nine should continue to work—radeonsi will translate the TGSI supplied by nine to NIR via tgsi_to_nir, like iris does. So everything is still supported, but now everything goes through NIR and there's one fewer path in the driver.

    (I wondered this myself, so I looked at the patches earlier.)

    Leave a comment:


  • geearf
    replied
    Is there an impact to certain state trackers, like nine?

    Leave a comment:


  • Guest
    Guest replied
    Code removal feels so nice

    Leave a comment:

Working...
X