public inbox for elfutils@sourceware.org
 help / color / mirror / Atom feed
From: Aaron Merey <amerey@redhat.com>
To: elfutils-devel@sourceware.org
Cc: Aaron Merey <amerey@redhat.com>
Subject: [PATCH] debuginfod: Use auto-sized connection pool when -C is not given with arg
Date: Fri,  2 Sep 2022 21:23:32 -0400	[thread overview]
Message-ID: <20220903012332.1456366-1-amerey@redhat.com> (raw)

Since commit 4b42d9ad, libmicrohttpd's epoll event loop is used when
available in which case we must disable its setting for spawning a thread
per request.  This contradicts the debuginfod doc's description of '-C',
which indicates that if this command line option is not given then the
thread pool size is unbounded.

Fix this by using an auto-sized thread pool when '-C' is not given, just
as we do when it's given with no argument. Update the doc's description
of '-C'.

Also use a fixed-size pool even if epoll is not supported. The unbounded
pool config cannot be considered entirely reliable as it appears to cause
random fails in the run-debuginfod-webapi-concurrency test.

Signed-off-by: Aaron Merey <amerey@redhat.com>

---
 debuginfod/ChangeLog      |  7 +++++++
 debuginfod/debuginfod.cxx | 22 ++++++++++------------
 doc/ChangeLog             |  4 ++++
 doc/debuginfod.8          | 10 ++++------
 4 files changed, 25 insertions(+), 18 deletions(-)

diff --git a/debuginfod/ChangeLog b/debuginfod/ChangeLog
index da58e9a8..c692a389 100644
--- a/debuginfod/ChangeLog
+++ b/debuginfod/ChangeLog
@@ -1,3 +1,10 @@
+2022-09-02  Aaron Merey  <amerey@redhat.com>
+
+	* debuginfod.cxx (parse_opt): If '-C' is given with no arg, do not
+	update connection_pool since it will be done at a later point.
+	(main): Use auto-sized connection_pool if '-C' isn't given with an
+	arg. Do not use MHD_USE_THREAD_PER_CONNECTION.
+
 2022-08-17  Martin Liska  <mliska@suse.cz>
 
 	* debuginfod.cxx (handle_buildid): Update HTTP statistics only if
diff --git a/debuginfod/debuginfod.cxx b/debuginfod/debuginfod.cxx
index 3e2dd9ef..8680c048 100644
--- a/debuginfod/debuginfod.cxx
+++ b/debuginfod/debuginfod.cxx
@@ -575,8 +575,6 @@ parse_opt (int key, char *arg,
           if (connection_pool < 2)
             argp_failure(state, 1, EINVAL, "-C NUM minimum 2");
         }
-      else // arg not given
-        connection_pool = std::thread::hardware_concurrency() * 2 ?: 2;
       break;
     case 'I':
       // NB: no problem with unconditional free here - an earlier failed regcomp would exit program
@@ -3937,6 +3935,11 @@ main (int argc, char *argv[])
         }
     }
 
+  /* If '-C' wasn't given or was given with no arg, pick a reasonable default
+     for the number of worker threads.  */
+  if (connection_pool == 0)
+    connection_pool = std::thread::hardware_concurrency() * 2 ?: 2;
+
   /* Note that MHD_USE_EPOLL and MHD_USE_THREAD_PER_CONNECTION don't
      work together.  */
   unsigned int use_epoll = 0;
@@ -3944,12 +3947,11 @@ main (int argc, char *argv[])
   use_epoll = MHD_USE_EPOLL;
 #endif
 
-  unsigned int mhd_flags = ((connection_pool || use_epoll
-			     ? 0 : MHD_USE_THREAD_PER_CONNECTION)
+  unsigned int mhd_flags = (
 #if MHD_VERSION >= 0x00095300
-			    | MHD_USE_INTERNAL_POLLING_THREAD
+			    MHD_USE_INTERNAL_POLLING_THREAD
 #else
-			    | MHD_USE_SELECT_INTERNALLY
+			    MHD_USE_SELECT_INTERNALLY
 #endif
 			    | MHD_USE_DUAL_STACK
 			    | use_epoll
@@ -3964,12 +3966,8 @@ main (int argc, char *argv[])
 				      handler_cb, NULL, /* handler callback */
 				      MHD_OPTION_EXTERNAL_LOGGER,
 				      error_cb, NULL,
-				      (connection_pool
-				       ? MHD_OPTION_THREAD_POOL_SIZE
-				       : MHD_OPTION_END),
-				      (connection_pool
-				       ? (int)connection_pool
-				       : MHD_OPTION_END),
+				      MHD_OPTION_THREAD_POOL_SIZE,
+				      (int)connection_pool,
 				      MHD_OPTION_END);
 
   MHD_Daemon *d4 = NULL;
diff --git a/doc/ChangeLog b/doc/ChangeLog
index ceec2467..c446e99e 100644
--- a/doc/ChangeLog
+++ b/doc/ChangeLog
@@ -1,3 +1,7 @@
+2022-09-02  Aaron Merey  <amerey@redhat.com>
+
+	* debuginfod.8 (-C): Update description.
+
 2022-06-03  Michael Trapp <michael.trapp@sap.com>
 
 	* debuginfod.8 (--disable-source-scan): Document.
diff --git a/doc/debuginfod.8 b/doc/debuginfod.8
index 50fce7f5..7c1dc3dd 100644
--- a/doc/debuginfod.8
+++ b/doc/debuginfod.8
@@ -212,15 +212,13 @@ following table summarizes the interpretaton of this option and its
 optional NUM parameter.
 .TS
 l l.
-no option	clone new thread for every request, no fixed pool
-\-C	use a fixed thread pool sized automatically
+no option, \-C	use a fixed thread pool sized automatically
 \-C=NUM	use a fixed thread pool sized NUM, minimum 2
 .TE
 
-The first mode is useful for friendly bursty traffic.  The second mode
-is a simple and safe configuration based on the number of processors.
-The third mode is suitable for tuned load-limiting configurations
-facing unruly traffic.
+The first mode is a simple and safe configuration based on the number
+of processors. The second mode is suitable for tuned load-limiting
+configurations facing unruly traffic.
 
 .TP
 .B "\-L"
-- 
2.37.1


             reply	other threads:[~2022-09-03  1:23 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-03  1:23 Aaron Merey [this message]
2022-09-05 17:18 ` Frank Ch. Eigler

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220903012332.1456366-1-amerey@redhat.com \
    --to=amerey@redhat.com \
    --cc=elfutils-devel@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).