From: Jason Merrill <jason@redhat.com>
To: Thomas Neumann <thomas.neumann@in.tum.de>,
"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Cc: "H.J. Lu" <hjl.tools@gmail.com>, Jakub Jelinek <jakub@redhat.com>,
Tamar Christina <Tamar.Christina@arm.com>,
Jonathan Wakely <jwakely.gcc@gmail.com>,
Florian Weimer <fweimer@redhat.com>
Subject: Re: [PATCH] initialize fde objects lazily
Date: Fri, 16 Dec 2022 12:25:45 -0500 [thread overview]
Message-ID: <2f2a22a0-18ba-c4e1-e464-fae17fb2223f@redhat.com> (raw)
In-Reply-To: <7d18f085-ae46-138d-4f04-df5857b7b014@in.tum.de>
On 12/9/22 12:34, Thomas Neumann wrote:
> When registering an unwind frame with __register_frame_info_bases
> we currently initialize that fde object eagerly. This has the
> advantage that it is immutable afterwards and we can safely
> access it from multiple threads, but it has the disadvantage
> that we pay the initialization cost even if the application
> never throws an exception.
>
> This commit changes the logic to initialize the objects lazily.
> The objects themselves are inserted into the b-tree when
> registering the frame, but the sorted fde_vector is
> not constructed yet. Only on the first time that an
> exception tries to pass through the registered code the
> object is initialized. We notice that with a double checking,
> first doing a relaxed load of the sorted bit and then re-checking
> under a mutex when the object was not initialized yet.
>
> Note that the check must implicitly be safe concering a concurrent
> frame deregistration, as trying the deregister a frame that is
> on the unwinding path of a concurrent exception is inherently racy.
OK, thanks.
> libgcc/ChangeLog:
> * unwind-dw2-fde.c: Initialize fde object lazily when
> the first exception tries to pass through.
> ---
> libgcc/unwind-dw2-fde.c | 52 ++++++++++++++++++++++++++++++++---------
> 1 file changed, 41 insertions(+), 11 deletions(-)
>
> diff --git a/libgcc/unwind-dw2-fde.c b/libgcc/unwind-dw2-fde.c
> index 3c0cc654ec0..6f69c20ff4b 100644
> --- a/libgcc/unwind-dw2-fde.c
> +++ b/libgcc/unwind-dw2-fde.c
> @@ -63,8 +63,6 @@ release_registered_frames (void)
>
> static void
> get_pc_range (const struct object *ob, uintptr_type *range);
> -static void
> -init_object (struct object *ob);
>
> #else
> /* Without fast path frame deregistration must always succeed. */
> @@ -76,6 +74,7 @@ static const int in_shutdown = 0;
> by decreasing value of pc_begin. */
> static struct object *unseen_objects;
> static struct object *seen_objects;
> +#endif
>
> #ifdef __GTHREAD_MUTEX_INIT
> static __gthread_mutex_t object_mutex = __GTHREAD_MUTEX_INIT;
> @@ -103,7 +102,6 @@ init_object_mutex_once (void)
> static __gthread_mutex_t object_mutex;
> #endif
> #endif
> -#endif
>
> /* Called from crtbegin.o to register the unwind info for an object. */
>
> @@ -126,10 +124,7 @@ __register_frame_info_bases (const void *begin,
> struct object *ob,
> #endif
>
> #ifdef ATOMIC_FDE_FAST_PATH
> - // Initialize eagerly to avoid locking later
> - init_object (ob);
> -
> - // And register the frame
> + // Register the frame in the b-tree
> uintptr_type range[2];
> get_pc_range (ob, range);
> btree_insert (®istered_frames, range[0], range[1] - range[0], ob);
> @@ -180,10 +175,7 @@ __register_frame_info_table_bases (void *begin,
> struct object *ob,
> ob->s.b.encoding = DW_EH_PE_omit;
>
> #ifdef ATOMIC_FDE_FAST_PATH
> - // Initialize eagerly to avoid locking later
> - init_object (ob);
> -
> - // And register the frame
> + // Register the frame in the b-tree
> uintptr_type range[2];
> get_pc_range (ob, range);
> btree_insert (®istered_frames, range[0], range[1] - range[0], ob);
> @@ -892,7 +884,15 @@ init_object (struct object* ob)
> accu.linear->orig_data = ob->u.single;
> ob->u.sort = accu.linear;
>
> +#ifdef ATOMIC_FDE_FAST_PATH
> + // We must update the sorted bit with an atomic operation
> + struct object tmp;
> + tmp.s.b = ob->s.b;
> + tmp.s.b.sorted = 1;
> + __atomic_store (&(ob->s.b), &(tmp.s.b), __ATOMIC_SEQ_CST);
> +#else
> ob->s.b.sorted = 1;
> +#endif
> }
>
> #ifdef ATOMIC_FDE_FAST_PATH
> @@ -1130,6 +1130,21 @@ search_object (struct object* ob, void *pc)
> }
> }
>
> +#ifdef ATOMIC_FDE_FAST_PATH
> +
> +// Check if the object was already initialized
> +static inline bool
> +is_object_initialized (struct object *ob)
> +{
> + // We have to use relaxed atomics for the read, which
> + // is a bit involved as we read from a bitfield
> + struct object tmp;
> + __atomic_load (&(ob->s.b), &(tmp.s.b), __ATOMIC_RELAXED);
> + return tmp.s.b.sorted;
> +}
> +
> +#endif
> +
> const fde *
> _Unwind_Find_FDE (void *pc, struct dwarf_eh_bases *bases)
> {
> @@ -1141,6 +1156,21 @@ _Unwind_Find_FDE (void *pc, struct dwarf_eh_bases
> *bases)
> if (!ob)
> return NULL;
>
> + // Initialize the object lazily
> + if (!is_object_initialized (ob))
> + {
> + // Check again under mutex
> + init_object_mutex_once ();
> + __gthread_mutex_lock (&object_mutex);
> +
> + if (!ob->s.b.sorted)
> + {
> + init_object (ob);
> + }
> +
> + __gthread_mutex_unlock (&object_mutex);
> + }
> +
> f = search_object (ob, pc);
> #else
>
next prev parent reply other threads:[~2022-12-16 17:25 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-16 10:19 [PATCH v4] eliminate mutex in fast path of __register_frame Thomas Neumann
2022-09-16 14:49 ` Jason Merrill
2022-09-18 8:59 ` Dimitar Dimitrov
2022-09-18 9:20 ` Thomas Neumann
2022-09-18 10:02 ` Thomas Neumann
2022-09-19 13:46 ` Stephan Bergmann
2022-09-19 13:55 ` Thomas Neumann
2022-09-19 14:00 ` Stephan Bergmann
2022-09-19 15:33 ` Thomas Neumann
2022-09-20 5:39 ` Stephan Bergmann
2022-11-21 11:14 ` Tamar Christina
2022-11-21 11:22 ` Thomas Neumann
2022-11-21 11:48 ` Jakub Jelinek
2022-11-21 17:13 ` H.J. Lu
2022-11-22 0:31 ` Thomas Neumann
2022-11-22 8:20 ` Florian Weimer
2022-11-22 9:12 ` Thomas Neumann
2022-12-09 17:34 ` [PATCH] initialize fde objects lazily Thomas Neumann
2022-12-15 16:11 ` Tamar Christina
2022-12-16 17:25 ` Jason Merrill [this message]
2023-05-02 14:32 ` [PATCH] release the sorted FDE array when deregistering a frame [PR109685] Thomas Neumann
2023-05-10 10:49 ` [PATCH] fix radix sort on 32bit platforms [PR109670] Thomas Neumann
2023-08-10 11:33 ` [PATCH] preserve base pointer for __deregister_frame [PR110956] Thomas Neumann
2023-08-11 15:21 ` Jeff Law
2024-03-15 10:29 ` [PATCH] handle unwind tables that are embedded within unwinding code, [PR111731] Thomas Neumann
2024-03-20 8:25 ` Richard Biener
2024-03-22 13:35 ` Jeff Law
2024-03-22 13:36 ` Jeff Law
2024-03-22 14:43 ` Thomas Neumann
2022-11-22 8:00 ` [PATCH] speed up end_fde_sort using radix sort Thomas Neumann
2022-12-16 18:02 ` Jason Merrill
2022-11-21 11:49 ` [PATCH v4] eliminate mutex in fast path of __register_frame Tamar Christina
2022-11-21 11:53 ` Thomas Neumann
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=2f2a22a0-18ba-c4e1-e464-fae17fb2223f@redhat.com \
--to=jason@redhat.com \
--cc=Tamar.Christina@arm.com \
--cc=fweimer@redhat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=hjl.tools@gmail.com \
--cc=jakub@redhat.com \
--cc=jwakely.gcc@gmail.com \
--cc=thomas.neumann@in.tum.de \
/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).