Re: Sphinx parallel build error: UnicodeEncodeError: 'latin-1' codec can't encode characters in position 18-20: ordinal not in range(256)

[Mauro Carvalho Chehab <mchehab@kernel.org>]

Em Sat, 8 May 2021 10:46:46 -0700
Randy Dunlap <rdunlap@infradead.org> escreveu:

> On 5/8/21 10:09 AM, Michal Suchánek wrote:
> > On Sat, May 08, 2021 at 08:55:11AM -0700, Randy Dunlap wrote:  
> >> Hi Mauro,
> >>
> >> On 5/8/21 7:41 AM, Mauro Carvalho Chehab wrote:  
> >>> Em Sat, 8 May 2021 12:41:57 +0200
> >>> Michal Suchánek <msuchanek@suse.de> escreveu:
> >>>  
> >>>> On Sat, May 08, 2021 at 11:22:05AM +0200, Mauro Carvalho Chehab wrote:  
> >>>>> Em Fri, 7 May 2021 08:39:24 +0200
> >>>>> Mauro Carvalho Chehab <mchehab@kernel.org> escreveu:
> >>>>>     
> >>>>>> Em Thu, 6 May 2021 14:21:01 -0700
> >>>>>> Randy Dunlap <rdunlap@infradead.org> escreveu:
> >>>>>>     
> >>>>>     
> >>>>>> I'll prepare a patch fixing it. Some care should be taken, however, as
> >>>>>> it has two places where UTF-8 chars should be used[2].    
> >>>>>
> >>>>> Ok, I did a small script in order to check what special chars we
> >>>>> currently have (next-20210507) at Documentation/ excluding the
> >>>>> translations.
> >>>>>
> >>>>> Based on my script results, we have those groups:
> >>>>>
> >>>>> 1. Latin accented characters:
> >>>>> 	- U+00c7 (LATIN CAPITAL LETTER C WITH CEDILLA) (Ç)
> >>>>> 	- U+00df (LATIN SMALL LETTER SHARP S) (ß)
> >>>>> 	- U+00e1 (LATIN SMALL LETTER A WITH ACUTE) (á)
> >>>>> 	- U+00e4 (LATIN SMALL LETTER A WITH DIAERESIS) (ä)
> >>>>> 	- U+00e6 (LATIN SMALL LETTER AE) (æ)
> >>>>> 	- U+00e7 (LATIN SMALL LETTER C WITH CEDILLA) (ç)
> >>>>> 	- U+00e9 (LATIN SMALL LETTER E WITH ACUTE) (é)
> >>>>> 	- U+00ea (LATIN SMALL LETTER E WITH CIRCUMFLEX) (ê)
> >>>>> 	- U+00eb (LATIN SMALL LETTER E WITH DIAERESIS) (ë)
> >>>>> 	- U+00f3 (LATIN SMALL LETTER O WITH ACUTE) (ó)
> >>>>> 	- U+00f4 (LATIN SMALL LETTER O WITH CIRCUMFLEX) (ô)
> >>>>> 	- U+00f6 (LATIN SMALL LETTER O WITH DIAERESIS) (ö)
> >>>>> 	- U+00f8 (LATIN SMALL LETTER O WITH STROKE) (ø)
> >>>>> 	- U+00fc (LATIN SMALL LETTER U WITH DIAERESIS) (ü)
> >>>>> 	- U+011f (LATIN SMALL LETTER G WITH BREVE) (ğ)
> >>>>> 	- U+0142 (LATIN SMALL LETTER L WITH STROKE) (ł)
> >>>>>
> >>>>> 2. symbols:
> >>>>> 	- U+00a9 (COPYRIGHT SIGN) (©)
> >>>>> 	- U+2122 (TRADE MARK SIGN) (™)
> >>>>> 	- U+00ae (REGISTERED SIGN) (®)
> >>>>> 	- U+00b0 (DEGREE SIGN) (°)
> >>>>> 	- U+00b1 (PLUS-MINUS SIGN) (±)
> >>>>> 	- U+00b2 (SUPERSCRIPT TWO) (²)
> >>>>> 	- U+00b5 (MICRO SIGN) (µ)
> >>>>> 	- U+00bd (VULGAR FRACTION ONE HALF) (½)
> >>>>> 	- U+2026 (HORIZONTAL ELLIPSIS) (…)
> >>>>>
> >>>>> 3. arrows:
> >>>>> 	- U+2191 (UPWARDS ARROW) (↑)
> >>>>> 	- U+2192 (RIGHTWARDS ARROW) (→)
> >>>>> 	- U+2193 (DOWNWARDS ARROW) (↓)
> >>>>> 	- U+2b0d (UP DOWN BLACK ARROW) (⬍)
> >>>>>
> >>>>> 4. box drawings:
> >>>>> 	- U+2500 (BOX DRAWINGS LIGHT HORIZONTAL) (─)
> >>>>> 	- U+2502 (BOX DRAWINGS LIGHT VERTICAL) (│)
> >>>>> 	- U+2514 (BOX DRAWINGS LIGHT UP AND RIGHT) (└)
> >>>>> 	- U+251c (BOX DRAWINGS LIGHT VERTICAL AND RIGHT) (├)
> >>>>>
> >>>>> 5. math symbols:
> >>>>> 	- U+00b7 (MIDDLE DOT) (·)
> >>>>> 	- U+00d7 (MULTIPLICATION SIGN) (×)
> >>>>> 	- U+2212 (MINUS SIGN) (−)
> >>>>> 	- U+2217 (ASTERISK OPERATOR) (∗)
> >>>>> 	- U+223c (TILDE OPERATOR) (∼)
> >>>>> 	- U+2264 (LESS-THAN OR EQUAL TO) (≤)
> >>>>> 	- U+2265 (GREATER-THAN OR EQUAL TO) (≥)
> >>>>> 	- U+27e8 (MATHEMATICAL LEFT ANGLE BRACKET) (⟨)
> >>>>> 	- U+27e9 (MATHEMATICAL RIGHT ANGLE BRACKET) (⟩)
> >>>>> 	- U+00ac (NOT SIGN) (¬)    
> >>>>  
> >>>  
> >>>>
> >>>> Use of ¬ is also very dubious in documentation (in fonts it is understandable):
> >>>> Documentation/ABI/obsolete/sysfs-kernel-fadump_registered:This ABI is renamed and moved to a new location /sys/kernel/fadump/registered.¬
> >>>> Documentation/ABI/obsolete/sysfs-kernel-fadump_release_mem:This ABI is renamed and moved to a new location /sys/kernel/fadump/release_mem.¬  
> >>>
> >>>  
> >>>> Documentation/powerpc/transactional_memory.rst:  if (MSR 29:31 ¬ = 0b010 | SRR1 29:31 ¬ = 0b000) then  
> >>>
> >>> Yeah, this should probably be better written as:
> >>>
> >>>   if (MSR 29:31 == 0b010 | SRR1 29:31 == 0b000) then  
> >>
> >> If the original with the 'NOT SIGN' was correct, then this
> >> version can't be correct. Or do you suspect that the "original"
> >> was corrupted somehow?  

No, I just misread the expression.

> > 
> > This does not make sense however you look at it. Using | between logical
> > expressions ...  
> 
> To my eyes/brain, it looks like classic (IBM) symbolic logic notation.
> In that context, I don't see anything wrong with it.

In this particular case, I would keep it as-is, with the UTF-8 char
on it. I mean, it might be converted to some other symbolic logic
notation, but "MSR 29:31" and "SRR1 29:31" aren't valid names on C.

> Yeah, I have been looking thru the arch/powerpc/ source code for this,
> but I haven't found it yet.

The title of the session says that it is part of "h/rfid mtmsrd quirk".

Searching for rfid:

	$ git grep -l rfid arch/powerpc/

Shows a lot of asm code. I guess that if the above quirk is still at
the Kernel, it is probably somewhere at the assembler part.

So, it sounds to me that converting it into C (or pseudo-C) won't
make it any better.

Thanks,
Mauro