Thanks for pointing that out. I sent out a fix:
https://patchwork.ozlabs.org/patch/851073/
On Tue, Dec 19, 2017 at 02:13:38PM -0800, Yifeng Sun wrote:
> At the start of ovsdb_txn_commit_, if there is an error, it looks like that
> txn doesn't get freed.
> After that, it surely is freed, a
At the start of ovsdb_txn_commit_, if there is an error, it looks like that
txn doesn't get freed.
After that, it surely is freed, as you said.
error = for_each_txn_row(txn, determine_changes);
if (error) {
return OVSDB_WRAP_BUG("can't happen", error);
}
On Tue, Dec 19
Thank you for the review.
ovsdb_txn_commit() is at least meant to free its argument whether it
succeeds or not. Do you see a bug there?
I applied this to master.
On Tue, Dec 19, 2017 at 09:50:45AM -0800, Yifeng Sun wrote:
> It seems that txn is not freed when ovsdb_txn_commit returns an error.
It seems that txn is not freed when ovsdb_txn_commit returns an error.
Other than that, this patch looks good to me.
Thanks,
Yifeng
On Fri, Dec 15, 2017 at 11:01 AM, Ben Pfaff wrote:
> This member was only used in one particular code path, so this commit
> adds code to pass it around as a funct
This member was only used in one particular code path, so this commit
adds code to pass it around as a function parameter instead.
Signed-off-by: Ben Pfaff
---
ovsdb/ovsdb-server.c | 40
1 file changed, 16 insertions(+), 24 deletions(-)
diff --git a/ovsd