By: bakaneko (nyan.delete@this.hyan.wan), November 21, 2012 12:25 pm
Room: Moderated Discussions
Paul A. Clayton (paaronclayton.delete@this.gmail.com) on November 21, 2012 9:39 am wrote:
> bakaneko (nyan.delete@this.hyan.wan) on November 21, 2012 6:52 am wrote:
> [snip]
> > This way you don't need to add superfluous reads to protect
> > objects which span multiple cachelines, which is as annoying
> > as multiple unrelated objects in one cacheline.
>
> When would one need to guard a value that is not accessed as part of the critical
> section? Is such an artifact of certain languages (e.g., providing object-granular
> locks) or are there cases where such provides a useful guard?
Good question, I just assumed it is, but
yeah, you are right, it's probably something
which happens rarely and not a good idea.
Even as guard, it would just make other
transactions fail more often...
> bakaneko (nyan.delete@this.hyan.wan) on November 21, 2012 6:52 am wrote:
> [snip]
> > This way you don't need to add superfluous reads to protect
> > objects which span multiple cachelines, which is as annoying
> > as multiple unrelated objects in one cacheline.
>
> When would one need to guard a value that is not accessed as part of the critical
> section? Is such an artifact of certain languages (e.g., providing object-granular
> locks) or are there cases where such provides a useful guard?
Good question, I just assumed it is, but
yeah, you are right, it's probably something
which happens rarely and not a good idea.
Even as guard, it would just make other
transactions fail more often...
| Topic | Posted By | Date |
|---|---|---|
| Article: Haswell TM Alternatives | David Kanter | 08/21/12 10:17 PM |
| Article: Haswell TM Alternatives | Håkan Winbom | 08/22/12 12:52 AM |
| Article: Haswell TM Alternatives | David Kanter | 08/22/12 02:06 AM |
| Article: Haswell TM Alternatives | anon | 08/22/12 09:46 AM |
| Article: Haswell TM Alternatives | Linus Torvalds | 08/22/12 10:16 AM |
| Article: Haswell TM Alternatives | Doug S | 08/24/12 09:34 AM |
| AMD's ASF even more limited | Paul A. Clayton | 08/22/12 10:20 AM |
| AMD's ASF even more limited | Linus Torvalds | 08/22/12 10:41 AM |
| Compiler use of ll/sc? | Paul A. Clayton | 08/28/12 10:28 AM |
| Compiler use of ll/sc? | Linus Torvalds | 09/08/12 01:58 PM |
| Lock recognition? | Paul A. Clayton | 09/10/12 02:17 PM |
| Sorry, I was confused | Paul A. Clayton | 09/13/12 11:56 AM |
| Filter to detect store conflicts | Paul A. Clayton | 08/22/12 10:19 AM |
| Article: Haswell TM Alternatives | bakaneko | 08/22/12 03:02 PM |
| Article: Haswell TM Alternatives | David Kanter | 08/22/12 03:45 PM |
| Article: Haswell TM Alternatives | bakaneko | 08/22/12 10:56 PM |
| Cache line granularity? | Paul A. Clayton | 08/28/12 10:28 AM |
| Cache line granularity? | David Kanter | 08/31/12 09:13 AM |
| A looser definition might have advantages | Paul A. Clayton | 09/01/12 07:29 AM |
| Cache line granularity? | rwessel | 08/31/12 08:54 PM |
| Alpha load locked granularity | Paul A. Clayton | 09/01/12 07:29 AM |
| Alpha load locked granularity | anon | 09/02/12 06:23 PM |
| Alpha pages groups | Paul A. Clayton | 09/03/12 05:16 AM |
| An alternative implementation | Maynard Handley | 11/20/12 10:52 PM |
| An alternative implementation | bakaneko | 11/21/12 06:52 AM |
| Guarding unread values? | Paul A. Clayton | 11/21/12 09:39 AM |
| Guarding unread values? | bakaneko | 11/21/12 12:25 PM |
| TM granularity and versioning | Paul A. Clayton | 11/21/12 09:27 AM |
| TM granularity and versioning | Maynard Handley | 11/21/12 11:52 AM |
| Indeed, TM (and coherence) has devilish details (NT) | Paul A. Clayton | 11/21/12 11:56 AM |
