| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
net/oenetcls: remove HISI_L0 dependency and L0 allocator usage hulk inclusion category: cleanup bugzilla: https://atomgit.com/openeuler/kernel/issues/9230 -------------------------------- oenetcls currently depends on HISI_L0 and uses custom L0 allocation/free helpers, which introduces platform-specific module dependencies. Remove the HISI_L0 Kconfig dependency, drop oenetcls_l0.c and related interfaces, and switch remaining allocation/free sites to generic kernel memory APIs (kzalloc/kcalloc/kfree). Also update build targets and init/exit paths accordingly. This decouples oenetcls from HISI_L0 and keeps behavior consistent on platforms without L0 support. Signed-off-by: Li Xiasong <lixiasong1@huawei.com> | 1 个月前 | |
net/oenetcls: remove HISI_L0 dependency and L0 allocator usage hulk inclusion category: cleanup bugzilla: https://atomgit.com/openeuler/kernel/issues/9230 -------------------------------- oenetcls currently depends on HISI_L0 and uses custom L0 allocation/free helpers, which introduces platform-specific module dependencies. Remove the HISI_L0 Kconfig dependency, drop oenetcls_l0.c and related interfaces, and switch remaining allocation/free sites to generic kernel memory APIs (kzalloc/kcalloc/kfree). Also update build targets and init/exit paths accordingly. This decouples oenetcls from HISI_L0 and keeps behavior consistent on platforms without L0 support. Signed-off-by: Li Xiasong <lixiasong1@huawei.com> | 1 个月前 | |
net/oenetcls: introduce oenetcls for network optimization hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/ICBFCS CVE: NA -------------------------------- This introduces a kind of network optimization method named oenetcls. It can configure the ntuple rule, and bind interrupt to the netdev queue automatically. Signed-off-by: Yue Haibing <yuehaibing@huawei.com> Signed-off-by: Wang Liang <wangliang74@huawei.com> Signed-off-by: Liu Jian <liujian56@huawei.com> Signed-off-by: yuelg <yuelg@chinaunicom.cn> | 1 年前 | |
net/oenetcls: introduce oenetcls for network optimization hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/ICBFCS CVE: NA -------------------------------- This introduces a kind of network optimization method named oenetcls. It can configure the ntuple rule, and bind interrupt to the netdev queue automatically. Signed-off-by: Yue Haibing <yuehaibing@huawei.com> Signed-off-by: Wang Liang <wangliang74@huawei.com> Signed-off-by: Liu Jian <liujian56@huawei.com> Signed-off-by: yuelg <yuelg@chinaunicom.cn> | 1 年前 | |
| 1 个月前 | ||
net/oenetcls: fix RPS CPU index scaling bias hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9102 -------------------------------- get_rps_cpu() uses reciprocal_scale(hash, cpus_num - 1) when selecting RPS CPU indices for both NUMA and cluster policies. This excludes the last bucket from selection and introduces a persistent distribution bias. Use the full candidate count in reciprocal_scale() so all candidate CPUs can be selected uniformly, and keep boundary checks for invalid counts. Fixes: 763b3255b12d ("net/oenetcls: Add rps policy switch for phy NIC") Signed-off-by: Li Xiasong <lixiasong1@huawei.com> | 1 个月前 | |
| 1 个月前 | ||
| 1 个月前 |
| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
| 1 个月前 | ||
| 1 个月前 | ||
| 1 年前 | ||
| 1 年前 | ||
| 1 个月前 | ||
| 1 个月前 | ||
| 1 个月前 | ||
| 1 个月前 |