glyphless

glyphless

[Eli Zaretskii]

    /* Sub-stretch for type == GLYPHLESS_GLYPH.  */
    struct
    {
      /* Value is an enum of the type glyphless_display_method.  */
      unsigned method : 2;
      /* 1 iff this glyph is for a character of no font. */
      unsigned for_no_font : 1;
      /* Length of acronym or hexadecimal code string (at most 8).  */
      unsigned len : 4;
      /* Character to display.  Actually we need only 22 bits.  */
      unsigned ch : 22; <<<<<<<<<<<<<<<<<<<<<<<<<<<<<
    } glyphless;

I though Andreas said that using 25 here, so that the entire member
fits in 32 bits, would be better (performance-wise, I presume)?

Re: glyphless

[Andreas Schwab]

Eli Zaretskii <eliz@gnu.org> writes:

> I though Andreas said that using 25 here, so that the entire member
> fits in 32 bits, would be better (performance-wise, I presume)?

I guess the difference, if any, will be lost in the noise (glyphless
chars are rare anyway).

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: glyphless

[Jason Rumney]

Eli Zaretskii <eliz@gnu.org> writes:

>     /* Sub-stretch for type == GLYPHLESS_GLYPH.  */
>     struct
>     {
>       /* Value is an enum of the type glyphless_display_method.  */
>       unsigned method : 2;
>       /* 1 iff this glyph is for a character of no font. */
>       unsigned for_no_font : 1;
>       /* Length of acronym or hexadecimal code string (at most 8).  */
>       unsigned len : 4;
>       /* Character to display.  Actually we need only 22 bits.  */
>       unsigned ch : 22; <<<<<<<<<<<<<<<<<<<<<<<<<<<<<
>     } glyphless;
>
> I though Andreas said that using 25 here, so that the entire member
> fits in 32 bits, would be better (performance-wise, I presume)?

Probably better is to define an extra member "padding" or "spare", to
make it clear that the ch portion is only 22 bits.