Re: 80286 protected mode

Liste des GroupesRevenir à c arch 
Sujet : Re: 80286 protected mode
De : terje.mathisen (at) *nospam* tmsw.no (Terje Mathisen)
Groupes : comp.arch
Date : 13. Oct 2024, 20:21:11
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <veh6j8$q71j$1@dont-email.me>
References : 1 2 3 4 5 6 7 8 9 10 11 12
User-Agent : Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Firefox/91.0 SeaMonkey/2.53.19
David Brown wrote:
On 10/10/2024 20:38, MitchAlsup1 wrote:
On Thu, 10 Oct 2024 6:31:52 +0000, David Brown wrote:
>
On 09/10/2024 23:37, MitchAlsup1 wrote:
On Wed, 9 Oct 2024 20:22:16 +0000, David Brown wrote:
>
On 09/10/2024 20:10, Thomas Koenig wrote:
David Brown <david.brown@hesbynett.no> schrieb:
>
When would you ever /need/ to compare pointers to different objects?
For almost all C programmers, the answer is "never".
>
Sometimes, it is handy to encode certain conditions in pointers,
rather than having only a valid pointer or NULL.  A compiler,
for example, might want to store the fact that an error occurred
while parsing a subexpression as a special pointer constant.
>
Compilers often have the unfair advantage, though, that they can
rely on what application programmers cannot, their implementation
details.  (Some do not, such as f2c).
>
Standard library authors have the same superpowers, so that they can
implement an efficient memmove() even though a pure standard C
programmer cannot (other than by simply calling the standard library
memmove() function!).
>
This is more a symptom of bad ISA design/evolution than of libc
writers needing superpowers.
>
No, it is not.  It has absolutely /nothing/ to do with the ISA.
>
For example, if ISA contains an MM instruction which is the
embodiment of memmove() then absolutely no heroics are needed
of desired in the libc call.
>
 The existence of a dedicated assembly instruction does not let you write an efficient memmove() in standard C.  That's why I said there was no connection between the two concepts.
 For some targets, it can be helpful to write memmove() in assembly or using inline assembly, rather than in non-portable C (which is the common case).
 
Thus, it IS a symptom of ISA evolution that one has to rewrite
memmove() every time wider SIMD registers are available.
 It is not that simple.
 There can often be trade-offs between the speed of memmove() and memcpy() on large transfers, and the overhead in setting things up that is proportionally more costly for small transfers.  Often that can be eliminated when the compiler optimises the functions inline - when the compiler knows the size of the move/copy, it can optimise directly.
What you are missing here David is the fact that Mitch's MM is a single instruction which does the entire memmove() operation, and has the inside knowledge about cache (residency at level x? width in bytes)/memory ranges/access rights/etc needed to do so in a very close to optimal manner, for both short and long transfers.
I.e. totally removing the need for compiler tricks or wide register operations.
Also apropos the compiler library issue:
You start by teaching the compiler about the MM instruction, and to recognize common patterns (just as most compilers already do today), and then the memmove() calls will usually be inlined.
Terje
--
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

Date Sujet#  Auteur
16 Apr 24 * Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)237Lawrence D'Oliveiro
16 Apr 24 `* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)236David Brown
16 Apr 24  +- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1MitchAlsup1
26 May 24  +- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1MitchAlsup1
1 Oct 24  `* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)233MitchAlsup1
1 Oct 24   `* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)232Thomas Koenig
1 Oct 24    +* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)225MitchAlsup1
2 Oct 24    i+* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)223Brett
3 Oct 24    ii`* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)222Lawrence D'Oliveiro
3 Oct 24    ii +- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1Brett
3 Oct 24    ii +- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1Anton Ertl
3 Oct 24    ii `* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)219David Brown
3 Oct 24    ii  `* Byte ordering (was: Whether something is RISC or not)218Anton Ertl
3 Oct 24    ii   +- Re: Byte ordering (was: Whether something is RISC or not)1David Brown
4 Oct 24    ii   +* Re: Byte ordering (was: Whether something is RISC or not)215Lawrence D'Oliveiro
4 Oct 24    ii   i+- Re: Byte ordering1Lynn Wheeler
4 Oct 24    ii   i+* Re: Byte ordering (was: Whether something is RISC or not)211David Brown
4 Oct 24    ii   ii`* Re: Byte ordering (was: Whether something is RISC or not)210Anton Ertl
4 Oct 24    ii   ii +* Re: Byte ordering5BGB
5 Oct 24    ii   ii i`* Re: Byte ordering4MitchAlsup1
5 Oct 24    ii   ii i +* Re: Byte ordering2BGB
5 Oct 24    ii   ii i i`- Re: Byte ordering1Lawrence D'Oliveiro
5 Oct 24    ii   ii i `- Re: Byte ordering1Lawrence D'Oliveiro
5 Oct 24    ii   ii +* Re: Byte ordering (was: Whether something is RISC or not)13Lawrence D'Oliveiro
5 Oct 24    ii   ii i`* Re: Byte ordering (was: Whether something is RISC or not)12Brett
5 Oct 24    ii   ii i `* Re: Byte ordering (was: Whether something is RISC or not)11Anton Ertl
5 Oct 24    ii   ii i  `* Re: Byte ordering (was: Whether something is RISC or not)10Michael S
6 Oct 24    ii   ii i   +- Re: Byte ordering1Terje Mathisen
6 Oct 24    ii   ii i   `* Re: Byte ordering (was: Whether something is RISC or not)8Brett
7 Oct 24    ii   ii i    `* Re: Byte ordering (was: Whether something is RISC or not)7Lawrence D'Oliveiro
7 Oct 24    ii   ii i     `* Re: Byte ordering (was: Whether something is RISC or not)6Brett
7 Oct 24    ii   ii i      `* Re: Byte ordering (was: Whether something is RISC or not)5Michael S
7 Oct 24    ii   ii i       +* Re: Byte ordering2Stefan Monnier
7 Oct 24    ii   ii i       i`- Re: Byte ordering1Michael S
7 Oct 24    ii   ii i       `* Re: Byte ordering (was: Whether something is RISC or not)2Lawrence D'Oliveiro
8 Oct 24    ii   ii i        `- Re: Byte ordering1Terje Mathisen
6 Oct 24    ii   ii `* Re: Byte ordering191David Brown
6 Oct 24    ii   ii  `* Re: Byte ordering190Anton Ertl
6 Oct 24    ii   ii   `* Re: Byte ordering189John Dallman
7 Oct 24    ii   ii    +* Re: Byte ordering20Lawrence D'Oliveiro
8 Oct 24    ii   ii    i`* Re: Byte ordering19John Dallman
9 Oct 24    ii   ii    i +- VMS/NT memory management (was: Byte ordering)1Stefan Monnier
15 Oct 24    ii   ii    i +* Re: Byte ordering2Lawrence D'Oliveiro
15 Oct 24    ii   ii    i i`- Re: Byte ordering1MitchAlsup1
15 Oct 24    ii   ii    i `* Re: Byte ordering15Lawrence D'Oliveiro
15 Oct 24    ii   ii    i  +* Re: Byte ordering3Michael S
15 Oct 24    ii   ii    i  i+- Re: Byte ordering1John Dallman
18 Oct 24    ii   ii    i  i`- Re: Byte ordering1Lawrence D'Oliveiro
15 Oct 24    ii   ii    i  +* Re: Byte ordering9John Dallman
16 Oct 24    ii   ii    i  i+* Re: Byte ordering7George Neuner
16 Oct 24    ii   ii    i  ii`* Re: Byte ordering6Terje Mathisen
16 Oct 24    ii   ii    i  ii `* Re: Byte ordering5David Brown
17 Oct 24    ii   ii    i  ii  +* Re: Byte ordering2George Neuner
17 Oct 24    ii   ii    i  ii  i`- Re: Byte ordering1David Brown
17 Oct 24    ii   ii    i  ii  `* Re: clouds, not Byte ordering2John Levine
17 Oct 24    ii   ii    i  ii   `- Re: clouds, not Byte ordering1David Brown
18 Oct 24    ii   ii    i  i`- Re: Byte ordering1Lawrence D'Oliveiro
16 Oct 24    ii   ii    i  `* Re: Byte ordering2Paul A. Clayton
18 Oct 24    ii   ii    i   `- Re: Microkernels & Capabilities (was Re: Byte ordering)1Lawrence D'Oliveiro
7 Oct 24    ii   ii    `* 80286 protected mode168Anton Ertl
7 Oct 24    ii   ii     +* Re: 80286 protected mode5Lars Poulsen
7 Oct 24    ii   ii     i`* Re: 80286 protected mode4Terje Mathisen
7 Oct 24    ii   ii     i +- Re: 80286 protected mode1Michael S
7 Oct 24    ii   ii     i `* Re: 80286 protected mode2Lawrence D'Oliveiro
8 Oct 24    ii   ii     i  `- Re: 80286 protected mode1Terje Mathisen
7 Oct 24    ii   ii     +* Re: 80286 protected mode3Brett
7 Oct 24    ii   ii     i`* Re: 80286 protected mode2Michael S
7 Oct 24    ii   ii     i `- Re: 80286 protected mode1Brett
7 Oct 24    ii   ii     +- Re: 80286 protected mode1Lawrence D'Oliveiro
8 Oct 24    ii   ii     +* Re: 80286 protected mode152MitchAlsup1
8 Oct 24    ii   ii     i+* Re: 80286 protected mode4Lawrence D'Oliveiro
8 Oct 24    ii   ii     ii`* Re: 80286 protected mode3MitchAlsup1
9 Oct 24    ii   ii     ii +- Re: 80286 protected mode1David Brown
15 Oct 24    ii   ii     ii `- Re: 80286 protected mode1Lawrence D'Oliveiro
8 Oct 24    ii   ii     i`* Re: 80286 protected mode147Anton Ertl
8 Oct 24    ii   ii     i +- Re: 80286 protected mode1Robert Finch
9 Oct 24    ii   ii     i `* Re: 80286 protected mode145David Brown
9 Oct 24    ii   ii     i  +* Re: 80286 protected mode79MitchAlsup1
9 Oct 24    ii   ii     i  i`* Re: 80286 protected mode78David Brown
9 Oct 24    ii   ii     i  i `* Re: 80286 protected mode77Stephen Fuld
10 Oct 24    ii   ii     i  i  +* Re: 80286 protected mode2MitchAlsup1
10 Oct 24    ii   ii     i  i  i`- Re: 80286 protected mode1David Brown
10 Oct 24    ii   ii     i  i  +- Re: 80286 protected mode1David Brown
11 Oct 24    ii   ii     i  i  `* Re: 80286 protected mode73Tim Rentsch
15 Oct 24    ii   ii     i  i   `* Re: 80286 protected mode72Stefan Monnier
15 Oct 24    ii   ii     i  i    +* Re: 80286 protected mode30MitchAlsup1
16 Oct 24    ii   ii     i  i    i+* Re: 80286 protected mode25MitchAlsup1
16 Oct 24    ii   ii     i  i    ii+* Re: C and turtles, 80286 protected mode13John Levine
16 Oct 24    ii   ii     i  i    iii+* Re: C and turtles, 80286 protected mode7MitchAlsup1
16 Oct 24    ii   ii     i  i    iiii`* Re: C and turtles, 80286 protected mode6John Levine
17 Oct 24    ii   ii     i  i    iiii `* Re: C and turtles, 80286 protected mode5Thomas Koenig
20 Oct 24    ii   ii     i  i    iiii  `* Re: C and turtles, 80286 protected mode4Lawrence D'Oliveiro
20 Oct 24    ii   ii     i  i    iiii   `* Re: C and turtles, 80286 protected mode3George Neuner
22 Oct 24    ii   ii     i  i    iiii    `* Re: C and turtles, 80286 protected mode2Tim Rentsch
22 Oct 24    ii   ii     i  i    iiii     `- Re: C and turtles, 80286 protected mode1George Neuner
16 Oct 24    ii   ii     i  i    iii+- Re: C and turtles, 80286 protected mode1David Brown
16 Oct 24    ii   ii     i  i    iii`* Re: C and turtles, 80286 protected mode4Paul A. Clayton
17 Oct 24    ii   ii     i  i    iii +- Re: C and turtles, 80286 protected mode1David Brown
20 Oct 24    ii   ii     i  i    iii `* Re: C and turtles, 80286 protected mode2Lawrence D'Oliveiro
20 Oct 24    ii   ii     i  i    iii  `- Re: C and turtles, 80286 protected mode1Paul A. Clayton
16 Oct 24    ii   ii     i  i    ii+* Re: 80286 protected mode7Thomas Koenig
17 Oct 24    ii   ii     i  i    ii+* Re: 80286 protected mode3George Neuner
17 Oct 24    ii   ii     i  i    ii`- Re: 80286 protected mode1Tim Rentsch
16 Oct 24    ii   ii     i  i    i+* Re: 80286 protected mode3David Brown
17 Oct 24    ii   ii     i  i    i`- Re: 80286 protected mode1Tim Rentsch
16 Oct 24    ii   ii     i  i    `* Re: 80286 protected mode41David Brown
9 Oct 24    ii   ii     i  +* Re: 80286 protected mode51Thomas Koenig
13 Oct 24    ii   ii     i  `* Re: 80286 protected mode14Anton Ertl
8 Oct 24    ii   ii     `* Re: 80286 protected mode6John Levine
6 Oct 24    ii   i`* Re: Byte ordering (was: Whether something is RISC or not)2Michael S
4 Oct 24    ii   `- Re: Byte ordering (was: Whether something is RISC or not)1John Dallman
2 Oct 24    i`- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1Thomas Koenig
2 Oct 24    +* Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)5David Schultz
3 Oct 24    `- Re: Whether something is RISC or not (Re: PDP-8 theology, not Concertina II Progress)1Lawrence D'Oliveiro

Haut de la page

Les messages affichés proviennent d'usenet.

NewsPortal