All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rick Edgecombe <rick.p.edgecombe@intel.com>
To: akpm@linux-foundation.org, jeyu@kernel.org, bpf@vger.kernel.org,
	ast@kernel.org, daniel@iogearbox.net, luto@kernel.org,
	dave.hansen@linux.intel.com, peterz@infradead.org,
	x86@kernel.org, rppt@kernel.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org, dan.j.williams@intel.com
Cc: elena.reshetova@intel.com, ira.weiny@intel.com,
	Rick Edgecombe <rick.p.edgecombe@intel.com>
Subject: [PATCH RFC 08/10] jump_label: Handle module writable address
Date: Fri, 20 Nov 2020 12:24:24 -0800	[thread overview]
Message-ID: <20201120202426.18009-9-rick.p.edgecombe@intel.com> (raw)
In-Reply-To: <20201120202426.18009-1-rick.p.edgecombe@intel.com>

Since modules can have a separate writable address during loading,
do the nop application at the writable address.

As long as info is on hand about if the operations is happening during
a module load, don't do a full text_poke() when writing data to a
writable address.

Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
---
 arch/x86/kernel/jump_label.c | 18 ++++++++++++++++--
 kernel/jump_label.c          |  2 +-
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c
index 5ba8477c2cb7..7a50148a63dd 100644
--- a/arch/x86/kernel/jump_label.c
+++ b/arch/x86/kernel/jump_label.c
@@ -63,6 +63,18 @@ static inline void __jump_label_transform(struct jump_entry *entry,
 					  int init)
 {
 	const void *opcode = __jump_label_set_jump_code(entry, type, init);
+	unsigned long addr = jump_entry_code(entry);
+	struct module *mod = __module_address(addr);
+	bool mod_writable = false;
+
+	if (mod) {
+		struct perm_allocation *alloc = module_get_allocation(mod, addr);
+
+		if (perm_is_writable(alloc)) {
+			addr = perm_writable_addr(alloc, addr);
+			mod_writable = true;
+		}
+	}
 
 	/*
 	 * As long as only a single processor is running and the code is still
@@ -74,9 +86,11 @@ static inline void __jump_label_transform(struct jump_entry *entry,
 	 * At the time the change is being done, just ignore whether we
 	 * are doing nop -> jump or jump -> nop transition, and assume
 	 * always nop being the 'currently valid' instruction
+	 *
+	 * If this is a module being loaded, text_poke_early can also be used.
 	 */
-	if (init || system_state == SYSTEM_BOOTING) {
-		text_poke_early((void *)jump_entry_code(entry), opcode,
+	if (init || system_state == SYSTEM_BOOTING || mod_writable) {
+		text_poke_early((void *)addr, opcode,
 				JUMP_LABEL_NOP_SIZE);
 		return;
 	}
diff --git a/kernel/jump_label.c b/kernel/jump_label.c
index 015ef903ce8c..3919e78fce12 100644
--- a/kernel/jump_label.c
+++ b/kernel/jump_label.c
@@ -595,7 +595,7 @@ static void __jump_label_mod_update(struct static_key *key)
  */
 void jump_label_apply_nops(struct module *mod)
 {
-	struct jump_entry *iter_start = mod->jump_entries;
+	struct jump_entry *iter_start = module_adjust_writable_addr(mod->jump_entries);
 	struct jump_entry *iter_stop = iter_start + mod->num_jump_entries;
 	struct jump_entry *iter;
 
-- 
2.20.1


  parent reply	other threads:[~2020-11-20 20:31 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-20 20:24 [PATCH RFC 00/10] New permission vmalloc interface Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 01/10] vmalloc: Add basic perm alloc implementation Rick Edgecombe
2020-11-22  4:10   ` Andy Lutomirski
2020-11-22  4:10     ` Andy Lutomirski
2020-11-23  0:01     ` Edgecombe, Rick P
2020-11-24 10:16       ` Christoph Hellwig
2020-11-24 20:00         ` Edgecombe, Rick P
2020-11-23  9:00   ` Christoph Hellwig
2020-11-23 20:44     ` Edgecombe, Rick P
2020-11-24 10:19       ` hch
2020-11-24 19:59         ` Edgecombe, Rick P
2020-12-04 23:24   ` Sean Christopherson
2020-12-07 23:55     ` Edgecombe, Rick P
2020-11-20 20:24 ` [PATCH RFC 02/10] bpf: Use perm_alloc() for BPF JIT filters Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 03/10] module: Use perm_alloc() for modules Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 04/10] module: Support separate writable allocation Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 05/10] x86/modules: Use real perm_allocations Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 06/10] x86/alternatives: Handle perm_allocs for modules Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 07/10] x86/unwind: Unwind orc at module writable address Rick Edgecombe
2020-11-20 20:24 ` Rick Edgecombe [this message]
2020-11-20 20:24 ` [PATCH RFC 09/10] ftrace: Use " Rick Edgecombe
2020-11-20 20:24 ` [PATCH RFC 10/10] vmalloc: Add perm_alloc x86 implementation Rick Edgecombe
2020-11-22 15:29   ` [vmalloc] 377647beed: WARNING:at_arch/x86/kernel/ftrace.c:#ftrace_verify_code kernel test robot

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=20201120202426.18009-9-rick.p.edgecombe@intel.com \
    --to=rick.p.edgecombe@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=dan.j.williams@intel.com \
    --cc=daniel@iogearbox.net \
    --cc=dave.hansen@linux.intel.com \
    --cc=elena.reshetova@intel.com \
    --cc=ira.weiny@intel.com \
    --cc=jeyu@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rppt@kernel.org \
    --cc=x86@kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.