On Wed, May 08, 2024 at 10:57:47AM -0700, Edward Liaw wrote:
On Wed, May 8, 2024 at 7:00 AM Jakub Kicinski kuba@kernel.org wrote:
On Wed, 8 May 2024 10:55:05 +0800 Tao Su wrote:
Back to commit 38c957f07038, I don't see any advantage in using LINE_MAX. Can we use a fixed value instead of LINE_MAX? E.g., 1024, 2048. Then we just need to revert commit 809216233555.
SGTM, FWIW. The print is printing a test summary line, printing more than 1k seems rather unreasonable. Other facilities, like TH_LOG(), should be used for displaying longer info.
I also submitted some patches to fix the _GNU_SOURCE issues here: https://lore.kernel.org/linux-kselftest/20240507214254.2787305-1-edliaw@goog...
I'm fine with this approach. It's aligned to what Sean suggested there, since it's causing a lot of troubles for the release cycle.
Thanks, I will submit patches with this approach soon.