05:18 randomarchitect: nobody would actually develop such number system where the spectrum can not be used for compression or entropy removal, it's just that i kept doing mistakes due to your terror, these mistakes are already removed, i am able to compile a set and access the members of the set from the hash, as i am not interested to deal with your maniacs, one more time you come to my premise to terror
05:18 randomarchitect: me, and you suffer throughout the remainder of your pathetic deluded lives!!!
07:35 tzimmermann: javierm, ping i indent to merge the sysfb series. ok?
11:26 jani: tursulin: 9779489a31d7 ("hrtimers: Delete hrtimer_init()") breaks 5a99350794fe ("drm/sched: Add scheduler unit testing infrastructure and some basic tests")
11:28 jani: tzimmermann: drm-misc-next needs a backmerge to get -rc1 with the hrtimer_init -> hrtimer_setup change, and a fix to the scheduler kunit test
11:28 jani: drm-tip build is broken with the two coming in from separate branches
11:44 tzimmermann: jani, ok. but i just did build drm-tip
11:44 jani: tzimmermann: you'll need the right kconfig
11:45 jani: CONFIG_DRM_SCHED_KUNIT_TEST
11:45 tzimmermann: jani, give me a bit to do the backmerge
11:45 jani: tzimmermann: no rush from my part, I can just disable the config
12:12 javierm: tzimmermann: sure, is OK for me
12:43 jfalempe: tzimmermann: when I build drm-misc/drm-misc-next with llvm, I got this error: drivers/gpu/drm/sysfb/simpledrm.c:247:33: error: unused function 'simpledrm_device_of_dev' [-Werror,-Wunused-function]
12:43 jfalempe: 247 | static struct simpledrm_device *simpledrm_device_of_dev(struct drm_device *dev)
12:43 jfalempe: | ^~~~~~~~~~~~~~~~~~~~~~~
12:44 tzimmermann: jfalempe, could be. i rarely build with with clang
12:44 tzimmermann: care to send a patch?
12:44 jfalempe: I think the function is always defined but used only if some flags are set.
12:44 tzimmermann: it's likely the result of the recent rework
12:45 jfalempe: tzimmermann: Yes, I can fix it and send a patch.
12:46 tzimmermann: jfalempe, thanks. the func needs to go into this condition: https://elixir.bootlin.com/linux/v6.13.7/source/drivers/gpu/drm/tiny/simpledrm.c#L266 AFAICT
12:48 tzimmermann: jfalempe, fixes tag would be: Fixes: 177dfbdb7e67 ("drm/sysfb: Merge primary-plane functions")
12:48 jfalempe: hum I think it should be: #if defined CONFIG_OF && (defined CONFIG_COMMON_CLK || defined CONFIG_COMMON_CLK)
12:49 jfalempe: stupid copy/paste its #if defined CONFIG_OF && (defined CONFIG_COMMON_CLK || defined CONFIG_REGULATOR)
13:19 tzimmermann: jfalempe, that #if has worked for a long time
13:27 jfalempe: tzimmermann: when this code was in tiny/simpledrm.c, simpledrm_device_of_dev() was used in simpledrm_primary_plane_helper_atomic_check() which doesn't depends on #if
13:27 tzimmermann: jfalempe, it used to be used throughout the driver. but most of this code is now in drm_sysfb_helper.c (in the same directory)
13:29 javierm: jfalempe: that syntax seems off to me. Shouldn't be #if defined(CONFIG_OF) && (defined(CONFIG_COMMON_CLK) || defined(CONFIG_REGULATOR)) ?
13:29 javierm: at least that's the convention AFAICT and used in most places
13:29 tzimmermann: jfalempe, let me send a patch. i think i've also just found a bug in that code.
13:30 jfalempe: tzimmermann: I already sent one, but feel free to ignore it.
13:30 tzimmermann: maybe the upcast isn't even correct in several places
13:31 jfalempe: javierm: I think the parenthesis are not strictly required, so I use the same convention as the rest of the file.
13:35 javierm: jfalempe: yeah, no strictly required but:
13:35 javierm: git grep "#if defined " -- drivers/gpu/drm/ | wc -l
13:35 javierm: 22
13:35 javierm: git grep "#if defined(" -- drivers/gpu/drm/ | wc -l
13:35 javierm: 465
13:36 javierm: jfalempe: anyways, I didn't notice that the rest of the file was using that so I guess that makes sense to make it consistent
13:36 jfalempe: javierm: ok there is a clear winner ;)
13:36 javierm: jfalempe: yeah, and the ratio is similar if you look at the kernel as a whole
13:54 tzimmermann: jfalempe, javierm, patches are out. the upcast can be removed entirely
13:55 jfalempe: tzimmermann: agreed, that's better :)
13:56 javierm: indeed :)
13:59 javierm: tzimmermann: both patches look good to me
13:59 tzimmermann: thanks