| From: | Ranier Vilela <ranier(dot)vf(at)gmail(dot)com> |
|---|---|
| To: | Zhihong Yu <zyu(at)yugabyte(dot)com> |
| Cc: | Michael Paquier <michael(at)paquier(dot)xyz>, Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Possible dereference null return (src/backend/replication/logical/reorderbuffer.c) |
| Date: | 2021-02-13 20:35:05 |
| Message-ID: | CAEudQAovXyN4FwiYyKBE=pT3tu6s2Jva-Mh6pE26Xy_oXF3ycQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Em sáb., 13 de fev. de 2021 às 01:07, Zhihong Yu <zyu(at)yugabyte(dot)com>
escreveu:
> Hi,
> How about the following patch ?
>
> ReorderBufferSetBaseSnapshot() can return a bool to indicate whether the
> base snapshot is set up.
>
> For the call by SnapBuildCommitTxn(), it seems xid is top transaction. So
> the return value doesn't need to be checked.
>
IMO anything else is better than PANIC.
Anyway, if all fails, reporting an error can contribute to checking where.
Attached a patch suggestion v2.
1. SnapBuildProcessChange returns a result of ReorderBufferSetBaseSnapshot,
so the caller can act accordingly.
2. SnapBuildCommitTxn can't ignore a result
from ReorderBufferSetBaseSnapshot, even if it never fails.
regards,
Ranier Vilela
| Attachment | Content-Type | Size |
|---|---|---|
| reorderbuffer.patch | application/octet-stream | 3.2 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ranier Vilela | 2021-02-13 20:37:32 | Re: pg_cryptohash_final possible out-of-bounds access (per Coverity) |
| Previous Message | Patrick Handja | 2021-02-13 19:54:39 | How to get Relation tuples in C function |