On Mon, Jul 9, 2018 at 4:07 PM, Maxime Ripard <maxime.rip...@bootlin.com> wrote:
> On Fri, Jul 06, 2018 at 02:45:53PM +0200, Arnd Bergmann wrote:
>> Having DRM_SUN4I built-in but DRM_SUN8I_MIXER as a loadable module results in
>> a link error, as we try to access a symbol from the sun8i_tcon_top.ko module:
>>
>> ERROR: "sun8i_tcon_top_of_table" [drivers/gpu/drm/sun4i/sun8i-drm-hdmi.ko]
>> undefined!
>> ERROR: "sun8i_tcon_top_of_table" [drivers/gpu/drm/sun4i/sun4i-drm.ko]
>> undefined!
>>
>> This solves the problem by making DRM_SUN8I_MIXER a 'bool' symbol, building
>> the sun8i_tcon_top module the same way as the core sun4i-drm module whenever
>> DRM_SUN8I_MIXER is enabled, or not building it at all otherwise.
>>
>> Alternatively, we could always build sun8i_tcon_top.ko along with sun4-drm.ko
>> and detach it from the mixer module, I could not tell which way is more
>> appropriate here.
>
> If that's easily doable, then yeah, that would be the preferred option
> I guess. Jernej? Chen-Yu? Any opinion on this?
Yeah, that definitely works for me. Having TCON TOP being part of the core
sun4i-drm makes more sense. The TCON code will likely call into it later on
when Jernej adds the encoder muxing code.
I wonder if we shouldn't just build the whole display engine code, excluding
downstream encoders (HDMI, DSI, TV) into one module. That would also fix the
issue that DRM_SUN4I_BACKEND has to be built-in if DRM_SUN4I is built-in.
That might be overkill though, given that one day we should be able to get
rid of the frontend check.
Here's a list of the size of various parts of this driver, built for ARMv7:
Common stuff:
text data bss dec hex filename
5021 344 0 5365 14f5 drivers/gpu/drm/sun4i/sun4i_drv.o
1058 0 0 1058 422 drivers/gpu/drm/sun4i/sun4i_layer.o
192 4 0 196 c4
drivers/gpu/drm/sun4i/sun4i_framebuffer.o
1704 0 0 1704 6a8 drivers/gpu/drm/sun4i/sun4i_crtc.o
1221 0 0 1221 4c5 drivers/gpu/drm/sun4i/sun4i_dotclock.o
1192 28 0 1220 4c4 drivers/gpu/drm/sun4i/sun4i_lvds.o
1583 92 0 1675 68b drivers/gpu/drm/sun4i/sun4i_rgb.o
10088 248 0 10336 2860 drivers/gpu/drm/sun4i/sun4i_tcon.o
1690 104 0 1794 702 drivers/gpu/drm/sun4i/sun6i_drc.o
23749 820 0 24569 5ff9 (TOTALS)
DE 1.0:
text data bss dec hex filename
3596 248 0 3844 f04 drivers/gpu/drm/sun4i/sun4i_frontend.o
8735 248 0 8983 2317 drivers/gpu/drm/sun4i/sun4i_backend.o
12331 496 0 12827 321b (TOTALS)
DE 2.0:
text data bss dec hex filename
4040 248 0 4288 10c0 drivers/gpu/drm/sun4i/sun8i_mixer.o
2620 28 0 2648 a58 drivers/gpu/drm/sun4i/sun8i_ui_layer.o
1500 0 0 1500 5dc drivers/gpu/drm/sun4i/sun8i_ui_scaler.o
2780 28 0 2808 af8 drivers/gpu/drm/sun4i/sun8i_vi_layer.o
12612 0 0 12612 3144 drivers/gpu/drm/sun4i/sun8i_vi_scaler.o
367 0 0 367 16f drivers/gpu/drm/sun4i/sun8i_csc.o
23919 304 0 24223 5e9f (TOTALS)
TCON TOP:
text data bss dec hex filename
2623 104 0 2727 aa7 drivers/gpu/drm/sun4i/sun8i_tcon_top.o
2623 104 0 2727 aa7 (TOTALS)
DE 1.0 HDMI:
text data bss dec hex filename
913 0 0 913 391
drivers/gpu/drm/sun4i/sun4i_hdmi_ddc_clk.o
5543 104 0 5647 160f drivers/gpu/drm/sun4i/sun4i_hdmi_enc.o
2583 0 0 2583 a17 drivers/gpu/drm/sun4i/sun4i_hdmi_i2c.o
1326 0 0 1326 52e
drivers/gpu/drm/sun4i/sun4i_hdmi_tmds_clk.o
10365 104 0 10469 28e5 (TOTALS)
MIPI DSI:
text data bss dec hex filename
1990 144 0 2134 856 drivers/gpu/drm/sun4i/sun6i_mipi_dphy.o
5921 132 0 6053 17a5 drivers/gpu/drm/sun4i/sun6i_mipi_dsi.o
7911 276 0 8187 1ffb (TOTALS)
DW HDMI:
text data bss dec hex filename
1774 104 0 1878 756 drivers/gpu/drm/sun4i/sun8i_dw_hdmi.o
1189 0 0 1189 4a5
drivers/gpu/drm/sun4i/sun8i_hdmi_phy_clk.o
4877 144 0 5021 139d drivers/gpu/drm/sun4i/sun8i_hdmi_phy.o
7840 248 0 8088 1f98 (TOTALS)
So it seems better to keep various parts as separate modules.
The "data" column makes me wonder if we've constify-ed all possible data
structures yet.
Regards
ChenYu
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
