mwrap user+dev discussion/patches/pulls/bugs/help
 help / color / mirror / code / Atom feed
From: Eric Wong <e@80x24.org>
To: mwrap-public@80x24.org
Subject: [PATCH 17/19] disable memalign tracking by default
Date: Mon, 16 Jul 2018 21:19:31 +0000	[thread overview]
Message-ID: <20180716211933.5835-18-e@80x24.org> (raw)
In-Reply-To: <20180716211933.5835-1-e@80x24.org>

Users will still pay the cost in memory overhead, but this
gets rid of misleading statistics.  Tracking actual Ruby
object allocations is better left to Ruby itself.
---
 ext/mwrap/mwrap.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/ext/mwrap/mwrap.c b/ext/mwrap/mwrap.c
index 627bb63..2752598 100644
--- a/ext/mwrap/mwrap.c
+++ b/ext/mwrap/mwrap.c
@@ -23,6 +23,7 @@
 #include "jhash.h"
 
 static ID id_uminus;
+static unsigned int track_memalign;
 const char *rb_source_location_cstr(int *line); /* requires 2.6.0dev */
 extern int __attribute__((weak)) ruby_thread_has_gvl_p(void);
 extern void * __attribute__((weak)) ruby_current_execution_context_ptr;
@@ -92,6 +93,7 @@ lfht_new(void)
 __attribute__((constructor)) static void resolve_malloc(void)
 {
 	int err;
+	const char *opt;
 	++locating;
 
 #ifdef __FreeBSD__
@@ -135,6 +137,11 @@ __attribute__((constructor)) static void resolve_malloc(void)
 	if (err)
 		fprintf(stderr, "pthread_atfork failed: %s\n", strerror(err));
 	page_size = sysconf(_SC_PAGESIZE);
+	opt = getenv("MWRAP");
+	if (opt && (opt = strstr(opt, "memalign:"))) {
+		if (!sscanf(opt, "memalign:%u", &track_memalign))
+			fprintf(stderr, "not an unsigned int: %s\n", opt);
+	}
 	--locating;
 }
 
@@ -461,7 +468,7 @@ internal_memalign(void **pp, size_t alignment, size_t size, uintptr_t caller)
 		return ENOMEM;
 
 	/* assert(asize == (alignment + size + sizeof(struct alloc_hdr))); */
-	l = update_stats_rcu_lock(size, caller);
+	l = track_memalign ? update_stats_rcu_lock(size, caller) : 0;
 	real = real_malloc(asize);
 	if (real) {
 		void *p = hdr2ptr(real);
@@ -1008,9 +1015,16 @@ static VALUE src_loc_name(VALUE self)
  * * dump_fd: a writable FD to dump to
  * * dump_path: a path to dump to, the file is opened in O_APPEND mode
  * * dump_min: the minimum allocation size (total) to dump
+ * * memalign: use `1' to enable tracking the memalign family
  *
  * If both `dump_fd' and `dump_path' are specified, dump_path takes
  * precedence.
+ *
+ * Tracking the memalign family of functions is misleading for Ruby
+ * applications, as heap page allocations can happen anywhere a
+ * Ruby object is allocated, even in the coldest code paths.
+ * Furthermore, it is rarely-used outside of the Ruby object allocator.
+ * Thus tracking memalign functions is disabled by default.
  */
 void Init_mwrap(void)
 {
-- 
EW


  parent reply	other threads:[~2018-07-16 21:19 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-16 21:19 [PATCH 0/19] the heavy version of mwrap Eric Wong
2018-07-16 21:19 ` [PATCH 01/19] support per-allocation headers for per-alloc tracking Eric Wong
2018-07-16 21:19 ` [PATCH 02/19] mwrap: use malloc to do our own memalign Eric Wong
2018-07-16 21:19 ` [PATCH 03/19] hold RCU read lock to insert each allocation Eric Wong
2018-07-16 21:19 ` [PATCH 04/19] realloc: do not copy if allocation failed Eric Wong
2018-07-16 21:19 ` [PATCH 05/19] internal_memalign: do not assume real_malloc succeeds Eric Wong
2018-07-16 21:19 ` [PATCH 06/19] ensure ENOMEM is preserved in errno when appropriate Eric Wong
2018-07-16 21:19 ` [PATCH 07/19] memalign: check alignment on all public functions Eric Wong
2018-07-16 21:19 ` [PATCH 08/19] reduce stack usage from file names Eric Wong
2018-07-16 21:19 ` [PATCH 09/19] resolve real_malloc earlier for C++ programs Eric Wong
2018-07-16 21:19 ` [PATCH 10/19] allow analyzing live allocations via Mwrap[location] Eric Wong
2018-07-16 21:19 ` [PATCH 11/19] alias Mwrap.clear to Mwrap.reset Eric Wong
2018-07-16 21:19 ` [PATCH 12/19] implement accessors for SourceLocation Eric Wong
2018-07-16 21:19 ` [PATCH 13/19] mwrap_aref: quiet -Wshorten-64-to-32 warning Eric Wong
2018-07-16 21:19 ` [PATCH 14/19] fixes for FreeBSD 11.1 Eric Wong
2018-07-16 21:19 ` [PATCH 15/19] use memrchr to extract address under glibc Eric Wong
2018-07-16 21:19 ` [PATCH 16/19] do not track allocations for constructor and Init_ Eric Wong
2018-07-16 21:19 ` Eric Wong [this message]
2018-07-16 21:19 ` [PATCH 18/19] support Mwrap.quiet to temporarily disable allocation tracking Eric Wong
2018-07-16 21:19 ` [PATCH 19/19] mwrap_rack: Rack app to track live allocations Eric Wong

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

  List information: https://80x24.org/mwrap/

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

  git send-email \
    --in-reply-to=20180716211933.5835-18-e@80x24.org \
    --to=e@80x24.org \
    --cc=mwrap-public@80x24.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.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mwrap.git/

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).