[SLOF] [PATCH v3 0/6] Accelerate the screen drawing functions
Greg Kurz
gkurz at linux.vnet.ibm.com
Fri Aug 14 20:15:48 AEST 2015
On Mon, 3 Aug 2015 23:06:34 +0200
Thomas Huth <thuth at redhat.com> wrote:
> While working on the cursor drawing issues, I've noticed that
> some other functions in the framebuffer code of SLOF could be
> accelerated quite a bit. The first two patches are rather
> cosmetic and do not contribute much to the speed improvement,
> but the following two patches speed up the drawing quite a lot!
>
Wow ! Your two series result in a VERY impressive speedup ! :)
Thanks Thomas !
--
Greg
> There are two core functions, MRMOVE and RFILL, which are used
> by the drawing functions to write to the framebuffer and which
> were not really adapted to the KVMPPC_H_LOGICAL_MEMOP of
> board-qemu yet (i.e. they were written with js2x in mind).
> So they both triggered a hypercall for every 8-byte access
> to the framebuffer instead of handling larger transfers at
> once. By implementing a proper accelerated version of these
> calls, the text drawing and screen-erasing can be speed-up
> quite a bit:
>
> - Fixing MRMOVE improves the character drawing by factor 2
> or more (the "benchmark" that I've published in my
> cover-letter of the cursor-improvement series earlier
> this week now only takes 0.3 s instead of 0.8 s!)
>
> - Fixing RFILL improves the screen erasing function by
> factor 10 or even more! The function "erase-screen"
> took approx. 1.3 s before the fix, and now only takes
> 0.1 s after the fix.
>
> Both fixes also speed up grub in graphical mode quite a bit,
> since grub apparently uses the firmware to erase the screen.
>
> v3:
> - Fixed stupid bug in RFILL for js2x (loop counter was not
> properly decremented)
>
> v2:
> - Cosmetic changes according to the review feedback from v1
> - Move the hunk to remove the RMOVE64 comment into a
> separate patch
> - Added a new patch to fix the 16-bit alignment check in
> _FASTMOVE and _FASTRMOVE
>
> Thomas Huth (6):
> terminal: Disable the terminal-write trace by default
> fbuffer: Precalculate line length in bytes
> fbuffer: Implement MRMOVE as an accelerated primitive
> fbuffer: Implement RFILL as an accelerated primitive
> Remove unused RMOVE64 stub
> Add missing half word access case to _FASTRMOVE and _FASTMOVE
>
> include/ppc970/cache.h | 50 +++++++++++++++++++++++++++++++++++++++++++----
> include/ppcp7/cache.h | 22 +++++++++++++++++----
> slof/fs/base.fs | 2 --
> slof/fs/fbuffer.fs | 25 +++++++++++++-----------
> slof/fs/rmove.fs | 53 --------------------------------------------------
> slof/fs/terminal.fs | 3 ++-
> slof/prim.code | 13 +++++++++++++
> slof/prim.in | 3 ++-
> 8 files changed, 95 insertions(+), 76 deletions(-)
> delete mode 100644 slof/fs/rmove.fs
>
More information about the SLOF
mailing list