public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Tom Tromey <tom@tromey.com>
To: gdb-patches@sourceware.org
Cc: Tom Tromey <tom@tromey.com>
Subject: [PATCH v3 10/33] Add batching parameter to parallel_for_each
Date: Sat,  4 Dec 2021 13:38:21 -0700	[thread overview]
Message-ID: <20211204203844.1188999-11-tom@tromey.com> (raw)
In-Reply-To: <20211204203844.1188999-1-tom@tromey.com>

parallel_for_each currently requires each thread to process at least
10 elements.  However, when indexing, it's fine for a thread to handle
just a single CU.  This patch parameterizes this, and updates the one
user.
---
 gdb/minsyms.c                          |  4 ++--
 gdb/unittests/parallel-for-selftests.c |  2 +-
 gdbsupport/parallel-for.h              | 18 ++++++++++++------
 3 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/gdb/minsyms.c b/gdb/minsyms.c
index 5f4cf54c663..7f2f23dd789 100644
--- a/gdb/minsyms.c
+++ b/gdb/minsyms.c
@@ -1462,8 +1462,8 @@ minimal_symbol_reader::install ()
       std::vector<computed_hash_values> hash_values (mcount);
 
       msymbols = m_objfile->per_bfd->msymbols.get ();
-      gdb::parallel_for_each
-	(&msymbols[0], &msymbols[mcount],
+      /* Arbitrarily require at least 10 elements in a thread.  */
+      gdb::parallel_for_each (10, &msymbols[0], &msymbols[mcount],
 	 [&] (minimal_symbol *start, minimal_symbol *end)
 	 {
 	   for (minimal_symbol *msym = start; msym < end; ++msym)
diff --git a/gdb/unittests/parallel-for-selftests.c b/gdb/unittests/parallel-for-selftests.c
index f149e72f0c7..c7a3780c743 100644
--- a/gdb/unittests/parallel-for-selftests.c
+++ b/gdb/unittests/parallel-for-selftests.c
@@ -52,7 +52,7 @@ test (int n_threads)
 #define NUMBER 10000
 
   std::atomic<int> counter (0);
-  gdb::parallel_for_each (0, NUMBER,
+  gdb::parallel_for_each (1, 0, NUMBER,
 			  [&] (int start, int end)
 			  {
 			    counter += end - start;
diff --git a/gdbsupport/parallel-for.h b/gdbsupport/parallel-for.h
index e4fcb4053b2..54027e69402 100644
--- a/gdbsupport/parallel-for.h
+++ b/gdbsupport/parallel-for.h
@@ -35,11 +35,16 @@ namespace gdb
 
    This approach was chosen over having the callback work on single
    items because it makes it simple for the caller to do
-   once-per-subrange initialization and destruction.  */
+   once-per-subrange initialization and destruction.
+
+   The parameter N says how batching ought to be done -- there will be
+   at least N elements processed per thread.  Setting N to 0 is not
+   allowed.  */
 
 template<class RandomIt, class RangeFunction>
 void
-parallel_for_each (RandomIt first, RandomIt last, RangeFunction callback)
+parallel_for_each (unsigned n, RandomIt first, RandomIt last,
+		   RangeFunction callback)
 {
 #if CXX_STD_THREAD
   /* So we can use a local array below.  */
@@ -52,10 +57,11 @@ parallel_for_each (RandomIt first, RandomIt last, RangeFunction callback)
   size_t n_elements = last - first;
   if (n_threads > 1)
     {
-      /* Arbitrarily require that there should be at least 10 elements
-	 in a thread.  */
-      if (n_elements / n_threads < 10)
-	n_threads = std::max (n_elements / 10, (size_t) 1);
+      /* Require that there should be at least N elements in a
+	 thread.  */
+      gdb_assert (n > 0);
+      if (n_elements / n_threads < n)
+	n_threads = std::max (n_elements / n, (size_t) 1);
       size_t elts_per_thread = n_elements / n_threads;
       n_actual_threads = n_threads - 1;
       for (int i = 0; i < n_actual_threads; ++i)
-- 
2.31.1


  parent reply	other threads:[~2021-12-04 20:39 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-04 20:38 [PATCH v3 00/33] Rewrite the DWARF "partial" reader Tom Tromey
2021-12-04 20:38 ` [PATCH v3 01/33] Split create_addrmap_from_aranges Tom Tromey
2021-12-04 20:38 ` [PATCH v3 02/33] Fix latent bug in read_addrmap_from_aranges Tom Tromey
2021-12-04 20:38 ` [PATCH v3 03/33] Add dwarf2_per_cu_data::addresses_seen Tom Tromey
2021-12-04 20:38 ` [PATCH v3 04/33] Refactor dwarf2_get_pc_bounds Tom Tromey
2021-12-04 20:38 ` [PATCH v3 05/33] Allow ada_decode not to decode operators Tom Tromey
2021-12-04 20:38 ` [PATCH v3 06/33] Let skip_one_die not skip children Tom Tromey
2021-12-04 20:38 ` [PATCH v3 07/33] Add name splitting Tom Tromey
2021-12-04 20:38 ` [PATCH v3 08/33] Add new overload of dwarf5_djb_hash Tom Tromey
2021-12-04 20:38 ` [PATCH v3 09/33] Refactor build_type_psymtabs_reader Tom Tromey
2021-12-04 20:38 ` Tom Tromey [this message]
2021-12-04 20:38 ` [PATCH v3 11/33] Return vector of results from parallel_for_each Tom Tromey
2022-03-29 17:36   ` Pedro Alves
2022-03-29 20:07     ` Tom Tromey
2021-12-04 20:38 ` [PATCH v3 12/33] Specialize std::hash for gdb_exception Tom Tromey
2021-12-04 20:38 ` [PATCH v3 13/33] Add "fullname" handling to file_and_directory Tom Tromey
2021-12-07 17:17   ` Tom Tromey
2021-12-04 20:38 ` [PATCH v3 14/33] Introduce DWARF abbrev cache Tom Tromey
2021-12-04 20:38 ` [PATCH v3 15/33] Statically examine abbrev properties Tom Tromey
2021-12-04 20:38 ` [PATCH v3 16/33] Update skip_one_die for new " Tom Tromey
2021-12-04 20:38 ` [PATCH v3 17/33] Introduce the new DWARF index class Tom Tromey
2021-12-04 20:38 ` [PATCH v3 18/33] The new DWARF indexer Tom Tromey
2022-03-29 18:04   ` Pedro Alves
2022-03-29 20:08     ` Tom Tromey
2021-12-04 20:38 ` [PATCH v3 19/33] Implement quick_symbol_functions for cooked DWARF index Tom Tromey
2021-12-04 20:38 ` [PATCH v3 20/33] Wire in the new DWARF indexer Tom Tromey
2021-12-04 20:38 ` [PATCH v3 21/33] Introduce thread-safe handling for complaints Tom Tromey
2021-12-04 20:38 ` [PATCH v3 22/33] Pre-read DWARF section data Tom Tromey
2021-12-04 20:38 ` [PATCH v3 23/33] Parallelize DWARF indexing Tom Tromey
2022-03-30 10:24   ` Pedro Alves
2022-03-30 20:23     ` Tom Tromey
2022-03-30 21:24       ` Tom Tromey
2021-12-04 20:38 ` [PATCH v3 24/33] "Finalize" the DWARF index in the background Tom Tromey
2021-12-04 20:38 ` [PATCH v3 25/33] Rename write_psymtabs_to_index Tom Tromey
2021-12-04 20:38 ` [PATCH v3 26/33] Change the key type in psym_index_map Tom Tromey
2021-12-04 20:38 ` [PATCH v3 27/33] Change parameters to write_address_map Tom Tromey
2021-12-04 20:38 ` [PATCH v3 28/33] Genericize addrmap handling in the DWARF index writer Tom Tromey
2021-12-04 20:38 ` [PATCH v3 29/33] Adapt .gdb_index writer to new DWARF scanner Tom Tromey
2021-12-04 20:38 ` [PATCH v3 30/33] Adapt .debug_names " Tom Tromey
2021-12-04 20:38 ` [PATCH v3 31/33] Enable the new DWARF indexer Tom Tromey
2021-12-04 20:38 ` [PATCH v3 32/33] Delete DWARF psymtab code Tom Tromey
2021-12-04 20:38 ` [PATCH v3 33/33] Remove dwarf2_per_cu_data::v Tom Tromey
2022-01-05 17:18 ` [PATCH v3 00/33] Rewrite the DWARF "partial" reader Tom Tromey
2022-03-24 16:18   ` Tom Tromey

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=20211204203844.1188999-11-tom@tromey.com \
    --to=tom@tromey.com \
    --cc=gdb-patches@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).