Re: [ovs-dev] [PATCH ovn v1] controller: Fix issue with ct_commit encode.

> On 11 Jun 2024, at 00:07, Dumitru Ceara wrote: > > On 6/10/24 18:05, martin.kal...@canonical.com > wrote: >> On Mon, 2024-06-10 at 12:54 +, Naveen Yerramneni wrote: >>> Action length is getting set incorrectly during ct_commit encode >>> due to which

Re: [ovs-dev] [PATCH ovn v1] controller: Fix issue with ct_commit encode.

On 6/11/24 00:07, Dumitru Ceara wrote: > On 6/10/24 18:05, martin.kal...@canonical.com wrote: >> On Mon, 2024-06-10 at 12:54 +, Naveen Yerramneni wrote: >>> Action length is getting set incorrectly during ct_commit encode >>> due to which ct action is getting skipped  during phsycial flows >>>

Re: [ovs-dev] [PATCH ovn v1] controller: Fix issue with ct_commit encode.

On 6/10/24 18:05, martin.kal...@canonical.com wrote: > On Mon, 2024-06-10 at 12:54 +, Naveen Yerramneni wrote: >> Action length is getting set incorrectly during ct_commit encode >> due to which ct action is getting skipped  during phsycial flows >> creation. This issue is noticed only if ct_co

Re: [ovs-dev] [PATCH ovn v1] controller: Fix issue with ct_commit encode.

On Mon, 2024-06-10 at 12:54 +, Naveen Yerramneni wrote: > Action length is getting set incorrectly during ct_commit encode > due to which ct action is getting skipped  during phsycial flows > creation. This issue is noticed only if ct_commit is prefixed > with other actions. > > logical flow:

[ovs-dev] [PATCH ovn v1] controller: Fix issue with ct_commit encode.

Action length is getting set incorrectly during ct_commit encode due to which ct action is getting skipped during phsycial flows creation. This issue is noticed only if ct_commit is prefixed with other actions. logical flow: reg8[17] = 1; ct_commit { ct_mark.blocked = 1; }; without fix: encodes a