git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH 3/8] xread_nonblock: add functionality to read from fds without blocking

From
Stefan Beller <sbeller@google.com>
Date
Dec 15, 2015, 00:09 UTC
Message-ID
<CAGZ79kZGjCy-o=2hO22=4=n2JqUsEG+dqOZFP4Hhf5E72B-_JA@mail.gmail.com>
In-Reply-To
<20151214235736.GA26133@sigill.intra.peff.net>
On Mon, Dec 14, 2015 at 3:57 PM, Jeff King <peff@peff.net> wrote:
Show 17 quoted lines
> On Mon, Dec 14, 2015 at 03:15:29PM -0800, Junio C Hamano wrote:
>
>> -- >8 --
>> From: Stefan Beller <sbeller@google.com>
>> Date: Mon, 14 Dec 2015 11:37:13 -0800
>> Subject: [PATCH] xread_nonblock: add functionality to read from fds without blocking
>>
>> Provide a wrapper to read(), similar to xread(), that restarts on
>> EINTR but not EAGAIN (or EWOULDBLOCK). This enables the caller to
>> handle polling itself, possibly polling multiple sockets or performing
>> some other action.
>
> This makes me wonder why we restart xread() on EAGAIN in the first
> place.
>
> On EINTR, sure; signals can come and we want to keep going. But if do
> not have non-blocking descriptors, it should never happen, right?
right.
>
> Are we trying to protect ourselves against somebody _else_ giving us a
> non-blocking descriptor? In that case we'll quietly spin and waste CPU.
> Which isn't great, but perhaps better than returning an error.

Yes. This sounds like a good reasoning for 2/8 (add in the poll, so we are more polite), though.

This patch is a prerequisite for 4/8, which explicitly doesn't want to loop but a quick return. Maybe we could even drop this patch and just use `read` as is in 4/8. Looking from a higher level perspective, we don't care about strbuf_read_nonblocking to return after a signal without retry.

>
> -Peff
Previous: Jeff KingNext: Jeff King
Message 14 of 30 in “Rerolling sb/submodule-parallel-fetch for the time after 2.7”
  1. 0/8 Rerolling sb/submodule-parallel-fetch for the time after 2.7Stefan Beller, Dec 14, 2015
  2. 1/8 submodule.c: write "Fetching submodule <foo>" to stderrStefan Beller, Dec 14, 2015
  3. 2/8 xread: poll on non blocking fdsStefan Beller, Dec 14, 2015
  4. Eric SunshineDec 14, 2015
  5. Stefan BellerDec 14, 2015
  6. Junio C HamanoDec 14, 2015
  7. Stefan BellerDec 14, 2015
  8. 3/8 xread_nonblock: add functionality to read from fds without blockingStefan Beller, Dec 14, 2015
  9. Junio C HamanoDec 14, 2015
  10. Eric SunshineDec 14, 2015
  11. Eric SunshineDec 14, 2015
  12. Junio C HamanoDec 14, 2015
  13. Jeff KingDec 14, 2015
  14. Stefan BellerDec 15, 2015
  15. Jeff KingDec 15, 2015
  16. Stefan BellerDec 15, 2015
  17. Jeff KingDec 15, 2015
  18. Johannes SixtDec 15, 2015
  19. Junio C HamanoDec 15, 2015
  20. 4/8 strbuf: add strbuf_read_once to read without blockingStefan Beller, Dec 14, 2015
  21. Eric SunshineDec 14, 2015
  22. Stefan BellerDec 14, 2015
  23. 5/8 sigchain: add command to pop all common signalsStefan Beller, Dec 14, 2015
  24. 6/8 run-command: add an asynchronous parallel child processorStefan Beller, Dec 14, 2015
  25. Johannes SixtDec 14, 2015
  26. Stefan BellerDec 14, 2015
  27. 7/8 fetch_populated_submodules: use new parallel job processingStefan Beller, Dec 14, 2015
  28. 8/8 submodules: allow parallel fetching, add tests and documentationStefan Beller, Dec 14, 2015
  29. Johannes SixtDec 14, 2015
  30. Junio C HamanoDec 14, 2015

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.