| Message ID | 1557851610-5602-1-git-send-email-david.marchand@redhat.com |
|---|---|
| Headers | show |
| Series | Quicker pmd threads reloads | expand |
David this patch set looks fine by me, guess a none-RFC patch would be next? Acked-by: Eelco Chaudron <echaudro@redhat.com> On 14 May 2019, at 18:33, David Marchand wrote: > We have been testing the rebalance code in different situations while > having traffic going through OVS. > Those tests have shown that part of the observed packets losses is due > to > some time wasted in signaling/waiting for the pmd threads to reload > their > polling configurations. > > This RFC series is an attempt at getting pmd threads reloads quicker > and > more deterministic. > > Example of number of cycles spent by a pmd between two polling > configurations (in cycles minimum/average/maximum of 1000 changes): > - d58b59c17c70: 126822/312103/756580 > - patch1: 113658/296157/741688 > - patch2: 49198/167206/466108 > - patch3: 13032/120730/341163 > - patch4: 12803/112964/323455 > - patch5: 13633/ 20373/ 47410 > > Changelog since v1: > - added numbers per patch in cover letter > - added memory ordering for explicit synchronisations between threads > in patch 1 and patch 2 > > -- > David Marchand > > David Marchand (5): > dpif-netdev: Convert exit latch to flag. > dpif-netdev: Trigger parallel pmd reloads. > dpif-netdev: Do not sleep when swapping queues. > dpif-netdev: Only reload static tx qid when needed. > dpif-netdev: Catch reloads faster. > > lib/dpif-netdev.c | 131 > +++++++++++++++++++++++++++++++++++++++++------------- > 1 file changed, 100 insertions(+), 31 deletions(-) > > -- > 1.8.3.1
On Mon, May 20, 2019 at 11:26 AM Eelco Chaudron <echaudro@redhat.com> wrote: > David this patch set looks fine by me, guess a none-RFC patch would be > next? > > Acked-by: Eelco Chaudron <echaudro@redhat.com> > Yes, just waiting for more comments, if any :-). Thanks for the review, Eelco.
On 14/05/2019 17:33, David Marchand wrote: > We have been testing the rebalance code in different situations while > having traffic going through OVS. > Those tests have shown that part of the observed packets losses is due to > some time wasted in signaling/waiting for the pmd threads to reload their > polling configurations. > > This RFC series is an attempt at getting pmd threads reloads quicker and > more deterministic. > > Example of number of cycles spent by a pmd between two polling > configurations (in cycles minimum/average/maximum of 1000 changes): > - d58b59c17c70: 126822/312103/756580 > - patch1: 113658/296157/741688 > - patch2: 49198/167206/466108 > - patch3: 13032/120730/341163 > - patch4: 12803/112964/323455 > - patch5: 13633/ 20373/ 47410 > > Changelog since v1: > - added numbers per patch in cover letter > - added memory ordering for explicit synchronisations between threads > in patch 1 and patch 2 > Aside from the couple of very minor comments, the series LGTM