From: Ingo Molnar <mingo@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: Juergen Gross <jgross@suse.com>,
"H . Peter Anvin" <hpa@zytor.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Peter Zijlstra <peterz@infradead.org>,
Borislav Petkov <bp@alien8.de>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@kernel.org>
Subject: [PATCH 40/53] x86/alternatives: Rename 'text_poke_sync()' to 'smp_text_poke_sync_each_cpu()'
Date: Fri, 11 Apr 2025 07:40:52 +0200 [thread overview]
Message-ID: <20250411054105.2341982-41-mingo@kernel.org> (raw)
In-Reply-To: <20250411054105.2341982-1-mingo@kernel.org>
Unlike sync_core(), text_poke_sync() is a very heavy operation, as
it sends an IPI to every online CPU in the system and waits for
completion.
Reflect this in the name.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
arch/x86/include/asm/text-patching.h | 2 +-
arch/x86/kernel/alternative.c | 12 ++++++------
arch/x86/kernel/kprobes/core.c | 4 ++--
arch/x86/kernel/kprobes/opt.c | 4 ++--
arch/x86/kernel/module.c | 2 +-
5 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/arch/x86/include/asm/text-patching.h b/arch/x86/include/asm/text-patching.h
index f3c9b70afb0c..d9dbbe9d9667 100644
--- a/arch/x86/include/asm/text-patching.h
+++ b/arch/x86/include/asm/text-patching.h
@@ -32,7 +32,7 @@ extern void apply_relocation(u8 *buf, const u8 * const instr, size_t instrlen, u
* an inconsistent instruction while you patch.
*/
extern void *text_poke(void *addr, const void *opcode, size_t len);
-extern void text_poke_sync(void);
+extern void smp_text_poke_sync_each_cpu(void);
extern void *text_poke_kgdb(void *addr, const void *opcode, size_t len);
extern void *text_poke_copy(void *addr, const void *opcode, size_t len);
#define text_poke_copy text_poke_copy
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 556a82f576cd..e4c51d81a72f 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -2445,7 +2445,7 @@ static void do_sync_core(void *info)
sync_core();
}
-void text_poke_sync(void)
+void smp_text_poke_sync_each_cpu(void)
{
on_each_cpu(do_sync_core, NULL, 1);
}
@@ -2469,8 +2469,8 @@ struct smp_text_poke_loc {
#define TP_ARRAY_NR_ENTRIES_MAX (PAGE_SIZE / sizeof(struct smp_text_poke_loc))
static struct smp_text_poke_array {
- int nr_entries;
struct smp_text_poke_loc vec[TP_ARRAY_NR_ENTRIES_MAX];
+ int nr_entries;
} text_poke_array;
static DEFINE_PER_CPU(atomic_t, text_poke_array_refs);
@@ -2649,7 +2649,7 @@ static void smp_text_poke_batch_process(void)
text_poke(text_poke_addr(&text_poke_array.vec[i]), &int3, INT3_INSN_SIZE);
}
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
/*
* Second step: update all but the first byte of the patched range.
@@ -2711,7 +2711,7 @@ static void smp_text_poke_batch_process(void)
* not necessary and we'd be safe even without it. But
* better safe than sorry (plus there's not only Intel).
*/
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
}
/*
@@ -2732,13 +2732,13 @@ static void smp_text_poke_batch_process(void)
}
if (do_sync)
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
/*
* Remove and wait for refs to be zero.
*
* Notably, if after step-3 above the INT3 got removed, then the
- * text_poke_sync() will have serialized against any running INT3
+ * smp_text_poke_sync_each_cpu() will have serialized against any running INT3
* handlers and the below spin-wait will not happen.
*
* IOW. unless the replacement instruction is INT3, this case goes
diff --git a/arch/x86/kernel/kprobes/core.c b/arch/x86/kernel/kprobes/core.c
index 09608fd93687..47cb8eb138ba 100644
--- a/arch/x86/kernel/kprobes/core.c
+++ b/arch/x86/kernel/kprobes/core.c
@@ -808,7 +808,7 @@ void arch_arm_kprobe(struct kprobe *p)
u8 int3 = INT3_INSN_OPCODE;
text_poke(p->addr, &int3, 1);
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
perf_event_text_poke(p->addr, &p->opcode, 1, &int3, 1);
}
@@ -818,7 +818,7 @@ void arch_disarm_kprobe(struct kprobe *p)
perf_event_text_poke(p->addr, &int3, 1, &p->opcode, 1);
text_poke(p->addr, &p->opcode, 1);
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
}
void arch_remove_kprobe(struct kprobe *p)
diff --git a/arch/x86/kernel/kprobes/opt.c b/arch/x86/kernel/kprobes/opt.c
index 9307a40f4983..0aabd4c4e2c4 100644
--- a/arch/x86/kernel/kprobes/opt.c
+++ b/arch/x86/kernel/kprobes/opt.c
@@ -513,11 +513,11 @@ void arch_unoptimize_kprobe(struct optimized_kprobe *op)
JMP32_INSN_SIZE - INT3_INSN_SIZE);
text_poke(addr, new, INT3_INSN_SIZE);
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
text_poke(addr + INT3_INSN_SIZE,
new + INT3_INSN_SIZE,
JMP32_INSN_SIZE - INT3_INSN_SIZE);
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
perf_event_text_poke(op->kp.addr, old, JMP32_INSN_SIZE, new, JMP32_INSN_SIZE);
}
diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
index a7998f351701..231d6326d1fd 100644
--- a/arch/x86/kernel/module.c
+++ b/arch/x86/kernel/module.c
@@ -206,7 +206,7 @@ static int write_relocate_add(Elf64_Shdr *sechdrs,
write, apply);
if (!early) {
- text_poke_sync();
+ smp_text_poke_sync_each_cpu();
mutex_unlock(&text_mutex);
}
--
2.45.2
next prev parent reply other threads:[~2025-04-11 5:42 UTC|newest]
Thread overview: 105+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-11 5:40 [PATCH -v3 00/53] Simplify, reorganize and clean up the x86 text-patching code (alternative.c) Ingo Molnar
2025-04-11 5:40 ` [PATCH 01/53] x86/alternatives: Improve code-patching scalability by removing false sharing in poke_int3_handler() Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Eric Dumazet
2025-04-11 5:40 ` [PATCH 02/53] x86/alternatives: Document the text_poke_bp_batch() synchronization rules a bit more Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Peter Zijlstra
2025-04-11 5:40 ` [PATCH 03/53] x86/alternatives: Rename 'struct bp_patching_desc' to 'struct int3_patching_desc' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] x86/alternatives: Rename 'struct bp_patching_desc' to 'struct text_poke_int3_vec' tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 04/53] x86/alternatives: Rename 'bp_refs' to 'int3_refs' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] x86/alternatives: Rename 'bp_refs' to 'text_poke_array_refs' tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 05/53] x86/alternatives: Rename 'text_poke_bp_batch()' to 'smp_text_poke_batch_process()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 06/53] x86/alternatives: Rename 'text_poke_bp()' to 'smp_text_poke_single()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 07/53] x86/alternatives: Rename 'poke_int3_handler()' to 'smp_text_poke_int3_handler()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 08/53] x86/alternatives: Rename 'poking_mm' to 'text_poke_mm' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 09/53] x86/alternatives: Rename 'poking_addr' to 'text_poke_mm_addr' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 10/53] x86/alternatives: Rename 'bp_desc' to 'int3_desc' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 11/53] x86/alternatives: Remove duplicate 'text_poke_early()' prototype Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 12/53] x86/alternatives: Update comments in int3_emulate_push() Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 13/53] x86/alternatives: Remove the confusing, inaccurate & unnecessary 'temp_mm_state_t' abstraction Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 14/53] x86/alternatives: Rename 'text_poke_flush()' to 'smp_text_poke_batch_flush()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 15/53] x86/alternatives: Rename 'text_poke_finish()' to 'smp_text_poke_batch_finish()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 16/53] x86/alternatives: Rename 'text_poke_queue()' to 'smp_text_poke_batch_add()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 17/53] x86/alternatives: Rename 'text_poke_loc_init()' to 'text_poke_int3_loc_init()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 18/53] x86/alternatives: Rename 'struct text_poke_loc' to 'struct smp_text_poke_loc' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 19/53] x86/alternatives: Rename 'struct int3_patching_desc' to 'struct text_poke_int3_vec' Ingo Molnar
2025-04-11 5:40 ` [PATCH 20/53] x86/alternatives: Rename 'int3_desc' to 'int3_vec' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 21/53] x86/alternatives: Add text_mutex) assert to smp_text_poke_batch_flush() Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 22/53] x86/alternatives: Use non-inverted logic instead of 'tp_order_fail()' Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 23/53] x86/alternatives: Remove the 'addr == NULL means forced-flush' hack from smp_text_poke_batch_finish()/smp_text_poke_batch_flush()/text_poke_addr_ordered() Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 24/53] x86/alternatives: Simplify smp_text_poke_single() by using tp_vec and existing APIs Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 25/53] x86/alternatives: Assert that smp_text_poke_int3_handler() can only ever handle 'tp_vec[]' based requests Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 26/53] x86/alternatives: Assert input parameters in smp_text_poke_batch_process() Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 27/53] x86/alternatives: Introduce 'struct smp_text_poke_array' and move tp_vec and tp_vec_nr to it Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 28/53] x86/alternatives: Remove the tp_vec indirection Ingo Molnar
2025-04-11 10:02 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 29/53] x86/alternatives: Rename 'try_get_desc()' to 'try_get_text_poke_array()' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 30/53] x86/alternatives: Rename 'put_desc()' to 'put_text_poke_array()' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 31/53] x86/alternatives: Simplify try_get_text_poke_array() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 32/53] x86/alternatives: Simplify smp_text_poke_int3_handler() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 33/53] x86/alternatives: Simplify smp_text_poke_batch_process() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 34/53] x86/alternatives: Rename 'int3_refs' to 'text_poke_array_refs' Ingo Molnar
2025-04-11 5:40 ` [PATCH 35/53] x86/alternatives: Move the text_poke_array manipulation into text_poke_int3_loc_init() and rename it to __smp_text_poke_batch_add() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 36/53] x86/alternatives: Remove the mixed-patching restriction on smp_text_poke_single() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 37/53] x86/alternatives: Document 'smp_text_poke_single()' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 38/53] x86/alternatives: Add documentation for smp_text_poke_batch_add() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 39/53] x86/alternatives: Move text_poke_array completion from smp_text_poke_batch_finish() and smp_text_poke_batch_flush() to smp_text_poke_batch_process() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` Ingo Molnar [this message]
2025-04-11 10:01 ` [tip: x86/alternatives] x86/alternatives: Rename 'text_poke_sync()' to 'smp_text_poke_sync_each_cpu()' tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 41/53] x86/alternatives: Simplify text_poke_addr_ordered() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 42/53] x86/alternatives: Constify text_poke_addr() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 43/53] x86/alternatives: Simplify and clean up patch_cmp() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 44/53] x86/alternatives: Standardize on 'tpl' local variable names for 'struct smp_text_poke_loc *' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 45/53] x86/alternatives: Rename 'TP_ARRAY_NR_ENTRIES_MAX' to 'TEXT_POKE_ARRAY_MAX' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 46/53] x86/alternatives: Rename 'POKE_MAX_OPCODE_SIZE' to 'TEXT_POKE_MAX_OPCODE_SIZE' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:40 ` [PATCH 47/53] x86/alternatives: Simplify the #include section Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 48/53] x86/alternatives: Move declarations of vmlinux.lds.S defined section symbols to <asm/alternative.h> Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 49/53] x86/alternatives: Remove 'smp_text_poke_batch_flush()' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 50/53] x86/alternatives: Update the comments in smp_text_poke_batch_process() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 51/53] x86/alternatives: Rename 'apply_relocation()' to 'text_poke_apply_relocation()' Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 52/53] x86/alternatives: Add comment about noinstr expectations Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Ingo Molnar
2025-04-11 5:41 ` [PATCH 53/53] x86/alternatives: Make smp_text_poke_batch_process() subsume smp_text_poke_batch_finish() Ingo Molnar
2025-04-11 10:01 ` [tip: x86/alternatives] " tip-bot2 for Nikolay Borisov
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=20250411054105.2341982-41-mingo@kernel.org \
--to=mingo@kernel.org \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=jgross@suse.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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