fixed tcp_mem resource leak when sending 304

[Micr0mega]

Our server's kernel would unexpectedly report

TCP: out of memory -- consider tuning tcp_mem

after a while, while the RAM itself was fine.
After a bit of digging, and a process of elimination of what had changed over the past few days, we suspected a program must be keeping TCP connections open, thus leaking tcp_mem. We tracked it down to being this line, and we have been running a container with this fix in production all day, without any significant raises in tcp_mem allocation (monitoring /proc/net/sockstat).

[googlebot]

We found a Contributor License Agreement for you (the sender of this pull request), but were unable to find agreements for all the commit author(s) or Co-authors. If you authored these, maybe you used a different email address in the git commits than was used to sign the CLA (login here to double check)? If these were authored by someone else, then they will need to sign a CLA as well, and confirm that they're okay with these being contributed to Google.
In order to pass this check, please resolve this problem and have the pull request author add another comment and the bot will run again. If the bot doesn't comment, it means it doesn't think anything has changed.

[Micr0mega]

@googlebot please try again

[googlebot]

CLAs look good, thanks!

[willnorris]

ooh, good catch. Thanks for the patch!