Re: Rendering Latency


One more thing -- this might be useful:

WGL_OML_sync_control()If one can render multiple (> 2) frames per vertical
sync rate, one could "dirty" old frames before they get swapped to update
them with newer data before a swap is mandated, maybe.jbw- Hide quoted text -

Mmm. That looks interesting. However, the only stuff I can find on
that capability is an opengl document last edited in 2001, so I'm
guessing it may obsolete by now, unless you know any better?. Very
little other useful references on the web (as far as I can see).

Dave
.

Relevant Pages

  • Re: Bug 12309 - Large I/O operations result in poor interactive performance and high iowait time
    ... Disabling swap makes the terminal launch much faster while copying; ... No single stall shows ... this reclaim is called from zone reclaim ... If we find a dirty page and its queue is not congested, ...
    (Linux-Kernel)
  • Re: [PATCH] remove throttle_vm_writeout()
    ... Anonymous pages -> swap ... writeback count increases, dirty balancing will hold back file ... then we get an undefined amount of space extra. ...
    (Linux-Kernel)
  • Re: [PATCH 00/28] Swap over NFS -v16
    ... Anonymous pages are dirty by definition (except the zero page, ... there is the swap cache. ... NR_FREE_PAGES are the pages in the page allocators free lists. ... allocation and freeing, and that wouldn't work well with slub. ...
    (Linux-Kernel)
  • Re: [PATCH] remove throttle_vm_writeout()
    ... Direct reclaim does not _increase_ the number of dirty pages in the ... Anonymous pages -> swap ... writeback count increases, dirty balancing will hold back file ... does case 2 need rate limiting, ...
    (Linux-Kernel)
  • Re: question on symbol exports
    ... > specified by the caller, store the addresses of any dirty pages, then ... > this without the additional exports, I'd love to hear about it. ... more than no swap, you must also make sure you have no ... send the line "unsubscribe linux-kernel" in ...
    (Linux-Kernel)