On Fri, Mar 17, 2023 at 05:08:47PM +0100, Thomas Zimmermann wrote: > Hi > > Am 17.03.23 um 16:11 schrieb Maxime Ripard: > > On Fri, Mar 17, 2023 at 01:51:42PM +0100, Thomas Zimmermann wrote: > > > Hi > > > > > > Am 17.03.23 um 13:39 schrieb Javier Martinez Canillas: > > > > Thomas Zimmermann <tzimmermann@xxxxxxx> writes: > > > > > > > > > Consolidate all handling of CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM by > > > > > making the module parameter optional in drm_fb_helper.c. > > > > > > > > > > Without the config option, modules can set smem_start in struct > > > > > fb_info for internal usage, but not export if to userspace. The > > > > > address can only be exported by enabling the option and setting > > > > > the module parameter. Also update the comment. > > > > > > > > > > > > > I was going to ask at what point we could just get rid of this Kconfig > > > > symbol since it already depends on EXPERT anyways so most distros will > > > > not enable it. > > > > > > > > But I looked then and noticed that it was added just a few years ago in > > > > commit 4be9bd10e22d "(drm/fb_helper: Allow leaking fbdev smem_start"), > > > > so it seems that people still need that :( > > > > > > I don't even know which userspace drivers need this symbol. Probably > > > something on Android. > > > > At least the Mali (utgard) user-space stack uses it, including on > > "regular" distributions. > > Does this use the lima kernel driver? No, I meant the ARM proprietary blobs
