From: Peter Zijlstra <peterz@infradead.org>
To: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Frederic Weisbecker <frederic@kernel.org>,
LKML <linux-kernel@vger.kernel.org>, Mel Gorman <mgorman@suse.de>,
Michal Hocko <mhocko@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
"Paul E . McKenney" <paulmck@kernel.org>,
Ingo Molnar <mingo@redhat.com>, Michal Hocko <mhocko@suse.com>,
rostedt@goodmis.org, jbaron@akamai.com, ardb@kernel.org
Subject: Re: [RFC PATCH 6/8] preempt/dynamic: Provide preempt_schedule[_notrace]() static calls
Date: Wed, 27 Jan 2021 12:27:09 +0100 [thread overview]
Message-ID: <YBFODfNZCjA9s0IP@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <YBEuy6zlBcV8gLvY@hirez.programming.kicks-ass.net>
On Wed, Jan 27, 2021 at 10:13:47AM +0100, Peter Zijlstra wrote:
> On Tue, Jan 26, 2021 at 05:57:30PM -0600, Josh Poimboeuf wrote:
> > Well, I hate it, but I'm not sure I have any better ideas. It should be
> > possible to use kallsyms, instead of the rb-tree/register nonsense. Not
> > sure about the performance impact though. Might be a good reason to
> > speed up kallsyms!
>
> Oh right, let me see if I can make that work.
Something like so compiles.. but it does make the whole thing depend on
KALLSYMS_ALL, which is somewhat yuck.
Also, kallsyms_lookup_name() is horrible, but not trivial to fix because
of that compression scheme used.
---
--- a/kernel/static_call.c
+++ b/kernel/static_call.c
@@ -8,6 +8,7 @@
#include <linux/module.h>
#include <linux/cpu.h>
#include <linux/processor.h>
+#include <linux/kallsyms.h>
#include <asm/sections.h>
extern struct static_call_site __start_static_call_sites[],
@@ -325,8 +326,66 @@ static int __static_call_mod_text_reserv
static int static_call_add_module(struct module *mod)
{
- return __static_call_init(mod, mod->static_call_sites,
- mod->static_call_sites + mod->num_static_call_sites);
+ struct static_call_site *start = mod->static_call_sites;
+ struct static_call_site *stop = start + mod->num_static_call_sites;
+ struct static_call_site *site;
+
+ struct {
+ unsigned long tramp;
+ unsigned long key;
+ } cache[8] = { { 0, 0}, };
+ int idx = 0;
+
+ for (site = start; site != stop; site++) {
+ unsigned long key, addr = (unsigned long)static_call_key(site);
+ unsigned long sym_size, sym_offset;
+ char sym_name[KSYM_NAME_LEN];
+ const char *name;
+ int i;
+
+ if (!kernel_text_address(addr))
+ continue;
+
+ /*
+ * Gotta fix up the keys that point to the trampoline.
+ */
+
+ /* Simple cache to avoid kallsyms */
+ for (i = 0; i < ARRAY_SIZE(cache); i++) {
+ if (cache[i].tramp == addr) {
+ key = cache[i].key;
+ goto got_key;
+ }
+ }
+
+ name = kallsyms_lookup(addr, &sym_size, &sym_offset, NULL, sym_name);
+ if (!name)
+ goto fail;
+
+ if (name != sym_name)
+ strcpy(sym_name, name);
+ memcpy(sym_name, STATIC_CALL_KEY_PREFIX_STR, STATIC_CALL_KEY_PREFIX_LEN);
+ key = kallsyms_lookup_name(sym_name);
+ if (!key)
+ goto fail;
+
+ /* Remember for next time.. */
+ cache[idx].tramp = addr;
+ cache[idx].key = key;
+ idx++;
+ idx %= ARRAY_SIZE(cache);
+
+got_key:
+ site->key = (key - (unsigned long)&site->key) |
+ (site->key & STATIC_CALL_SITE_FLAGS);
+ }
+
+ return __static_call_init(mod, start, stop);
+
+fail:
+ pr_warn("Failed to fixup __raw_static_call() usage at: %ps\n",
+ static_call_addr(site));
+ return -EINVAL;
}
static void static_call_del_module(struct module *mod)
next prev parent reply other threads:[~2021-01-27 11:39 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-18 14:12 [RFC PATCH 0/7] preempt: Tune preemption flavour on boot v4 Frederic Weisbecker
2021-01-18 14:12 ` [RFC PATCH 1/8] static_call/x86: Add __static_call_return0() Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra
2021-01-18 14:12 ` [RFC PATCH 2/8] static_call: Provide DEFINE_STATIC_CALL_RET0() Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Frederic Weisbecker
2021-02-17 13:17 ` tip-bot2 for Frederic Weisbecker
2021-01-18 14:12 ` [RFC PATCH 3/8] static_call: Pull some static_call declarations to the type headers Frederic Weisbecker
2021-01-19 10:46 ` Jürgen Groß
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra
2021-01-18 14:12 ` [RFC PATCH 4/8] preempt: Introduce CONFIG_PREEMPT_DYNAMIC Frederic Weisbecker
2021-01-22 16:53 ` Peter Zijlstra
2021-01-28 12:17 ` Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Michal Hocko
2021-02-17 13:17 ` tip-bot2 for Michal Hocko
2021-01-18 14:12 ` [RFC PATCH 5/8] preempt/dynamic: Provide cond_resched() and might_resched() static calls Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra (Intel)
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra (Intel)
2021-01-18 14:12 ` [RFC PATCH 6/8] preempt/dynamic: Provide preempt_schedule[_notrace]() " Frederic Weisbecker
2021-01-21 21:58 ` Peter Zijlstra
2021-01-21 22:25 ` Peter Zijlstra
2021-01-22 16:52 ` Peter Zijlstra
2021-01-22 16:57 ` Ard Biesheuvel
2021-01-22 17:08 ` Peter Zijlstra
2021-02-08 12:00 ` [tip: sched/core] sched: Add /debug/sched_preempt tip-bot2 for Peter Zijlstra
2021-02-09 15:45 ` tip-bot2 for Peter Zijlstra
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra
2021-01-25 23:40 ` [RFC PATCH 6/8] preempt/dynamic: Provide preempt_schedule[_notrace]() static calls Josh Poimboeuf
2021-01-26 9:24 ` Peter Zijlstra
2021-01-26 23:57 ` Josh Poimboeuf
2021-01-27 9:13 ` Peter Zijlstra
2021-01-27 11:27 ` Peter Zijlstra [this message]
2021-01-27 15:59 ` Josh Poimboeuf
2021-01-27 16:19 ` Peter Zijlstra
2021-01-27 16:33 ` Josh Poimboeuf
2021-01-27 18:44 ` Peter Zijlstra
2021-01-27 19:00 ` Josh Poimboeuf
2021-01-27 19:02 ` Josh Poimboeuf
2021-01-27 23:18 ` Josh Poimboeuf
2021-02-03 14:04 ` Peter Zijlstra
2021-02-05 15:30 ` Peter Zijlstra
2021-02-06 2:31 ` Josh Poimboeuf
2021-02-06 9:03 ` Peter Zijlstra
2021-02-05 15:22 ` Peter Zijlstra
2021-02-08 12:00 ` [tip: sched/core] static_call: Allow module use without exposing static_call_key tip-bot2 for Josh Poimboeuf
2021-02-09 15:45 ` tip-bot2 for Josh Poimboeuf
2021-02-17 13:17 ` tip-bot2 for Josh Poimboeuf
2021-02-08 12:00 ` [tip: sched/core] preempt/dynamic: Provide preempt_schedule[_notrace]() static calls tip-bot2 for Peter Zijlstra (Intel)
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra (Intel)
2021-01-18 14:12 ` [RFC PATCH 7/8] preempt/dynamic: Provide irqentry_exit_cond_resched() static call Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra (Intel)
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra (Intel)
2021-01-18 14:12 ` [RFC PATCH 8/8] preempt/dynamic: Support dynamic preempt with preempt= boot option Frederic Weisbecker
2021-02-08 12:00 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra (Intel)
2021-02-09 15:45 ` tip-bot2 for Peter Zijlstra (Intel)
2021-02-17 13:17 ` tip-bot2 for Peter Zijlstra (Intel)
2021-01-21 21:22 ` [RFC PATCH 0/7] preempt: Tune preemption flavour on boot v4 Peter Zijlstra
2021-01-22 15:02 ` Paul E. McKenney
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=YBFODfNZCjA9s0IP@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=ardb@kernel.org \
--cc=frederic@kernel.org \
--cc=jbaron@akamai.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mhocko@kernel.org \
--cc=mhocko@suse.com \
--cc=mingo@redhat.com \
--cc=paulmck@kernel.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.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