Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.programming.threads > #158 > unrolled thread
| Started by | J de Boyne Pollard <j.deboynepollard@tesco.net> |
|---|---|
| First post | 2011-05-17 02:21 -0700 |
| Last post | 2011-05-17 13:22 +0100 |
| Articles | 2 — 2 participants |
Back to article view | Back to comp.programming.threads
Re: What it means to be first. J de Boyne Pollard <j.deboynepollard@tesco.net> - 2011-05-17 02:21 -0700
Re: What it means to be first. Geoff Clare <geoff@clare.See-My-Signature.invalid> - 2011-05-17 13:22 +0100
| From | J de Boyne Pollard <j.deboynepollard@tesco.net> |
|---|---|
| Date | 2011-05-17 02:21 -0700 |
| Subject | Re: What it means to be first. |
| Message-ID | <9065fced-4b9b-44eb-85c0-2b56e09b7d31@glegroupsg2000goo.googlegroups.com> |
> 4. If the thread hadn't run on any of the selected processors
> before becoming a blocked thread, select the first processor
> that was found in the search.
Unless there's some ordering defined for a set of processors, that's not a very meaningful constraint, since it lacks a concrete definition of "first" that can be relied upon. ("first" by processor number? "first" by processor interrupt arbitration priority? "first" in order of address in kernel space of some per-processor data structure? "first" in some linked list that was scanned? "first" in processor bootstrap order? All are conceivable implementation strategies/side-effects.) Is there such an ordering given in what you read?
[toc] | [next] | [standalone]
| From | Geoff Clare <geoff@clare.See-My-Signature.invalid> |
|---|---|
| Date | 2011-05-17 13:22 +0100 |
| Message-ID | <is5aa8-cuo.ln1@leafnode-msgid.gclare.org.uk> |
| In reply to | #158 |
J de Boyne Pollard wrote:
>> 4. If the thread hadn't run on any of the selected processors
>> before becoming a blocked thread, select the first processor
>> that was found in the search.
>
> Unless there's some ordering defined for a set of processors,
> that's not a very meaningful constraint, since it lacks a concrete
> definition of "first" that can be relied upon. ("first" by processor
> number? "first" by processor interrupt arbitration priority? "first"
> in order of address in kernel space of some per-processor data
> structure? "first" in some linked list that was scanned? "first"
> in processor bootstrap order? All are conceivable implementation
> strategies/side-effects.) Is there such an ordering given in what you
> read?
I gave the URL for the document; you could have looked yourself :-)
Anyway, I have searched it for all occurrences of "processor" and
didn't find anything relevant.
--
Geoff Clare <netnews@gclare.org.uk>
[toc] | [prev] | [standalone]
Back to top | Article view | comp.programming.threads
csiph-web