This issue tracker has been migrated to GitHub ,
and is currently read-only.
For more information,
see the GitHub FAQs in the Python's Developer Guide.
| Author | gregory.p.smith |
|---|---|
| Recipients | anthonybaxter, christian.heimes, georg.brandl, gregory.p.smith, jafo, janixia, jhylton, jyasskin, loewis, pitrou, tim.peters |
| Date | 2008年04月06日.03:07:14 |
| SpamBayes Score | 0.44891295 |
| Marked as misclassified | No |
| Message-id | <1207451236.5.0.96370591865.issue815646@psf.upfronthosting.co.za> |
| In-reply-to |
| Content | |
|---|---|
I'm reviewing this patch now and plan to commit it after some testing. A couple comments: I'd rename your sts variables to status. Also FYI: Your use of volatile on the int unlocked_count member of PyFileObject does not do what you think it does and isn't needed here anyways. Access to the variable is always protected by the GIL unlocking and locking of which should cause an implicit memory barrier guaranteeing that all other CPUs in the system will see the same value stored in the structure in memory. The C volatile keyword on the other hand does not guarantee this. volatile is useful for memory mapped IO but it makes no guarantees about cache coherent access between multiple CPUs. (the atomic types in the recent C++ standards are meant for that) Both of the above are trivial changes, no need for another patch. |
|
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2008年04月06日 03:07:16 | gregory.p.smith | set | spambayes_score: 0.448913 -> 0.44891295 recipients: + gregory.p.smith, tim.peters, loewis, jhylton, anthonybaxter, georg.brandl, jafo, janixia, pitrou, christian.heimes, jyasskin |
| 2008年04月06日 03:07:16 | gregory.p.smith | set | spambayes_score: 0.448913 -> 0.448913 messageid: <1207451236.5.0.96370591865.issue815646@psf.upfronthosting.co.za> |
| 2008年04月06日 03:07:15 | gregory.p.smith | link | issue815646 messages |
| 2008年04月06日 03:07:14 | gregory.p.smith | create | |