All of lore.kernel.org
 help / color / mirror / Atom feed
From: Szabolcs Nagy <szabolcs.nagy@arm.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: libc-alpha@sourceware.org, Mark Rutland <mark.rutland@arm.com>,
	kernel-hardening@lists.openwall.com,
	Catalin Marinas <catalin.marinas@arm.com>,
	linux-kernel@vger.kernel.org,
	Jeremy Linton <jeremy.linton@arm.com>,
	Mark Brown <broonie@kernel.org>,
	Topi Miettinen <toiwoton@gmail.com>,
	Will Deacon <will@kernel.org>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v2 1/6] aarch64: Fix missing BTI protection from dependencies [BZ #26926]
Date: Fri, 11 Dec 2020 15:33:34 +0000	[thread overview]
Message-ID: <20201211153334.GF24625@arm.com> (raw)
In-Reply-To: <f077a6d5-082c-c3a5-ce07-71b87a70dc12@linaro.org>

[-- Attachment #1: Type: text/plain, Size: 789 bytes --]

The 12/10/2020 14:51, Adhemerval Zanella wrote:
> On 27/11/2020 10:19, Szabolcs Nagy via Libc-alpha wrote:
> > The _dl_open_check and _rtld_main_check hooks are not called on the
> > dependencies of a loaded module, so BTI protection was missed on
> > every module other than the main executable and directly dlopened
> > libraries.
> > 
> > The fix just iterates over dependencies to enable BTI.
> > 
> > Fixes bug 26926.
> 
> LGTM, modulus the argument name change.
> 
> I also think it would be better to add a testcase, for both DT_NEEDED
> and dlopen case.

thanks, i committed this with fixed argument name as attached.

i plan to do tests later once i understand the BTI semantics
(right now it's not clear if in the presence of some W^X
policy BTI may be silently ignored or not).

[-- Attachment #2: 0001-aarch64-Fix-missing-BTI-protection-from-dependencies.patch --]
[-- Type: text/x-diff, Size: 1446 bytes --]

From 72739c79f61989a76b7dd719f34fcfb7b8eadde9 Mon Sep 17 00:00:00 2001
From: Szabolcs Nagy <szabolcs.nagy@arm.com>
Date: Fri, 20 Nov 2020 15:27:06 +0000
Subject: [PATCH] aarch64: Fix missing BTI protection from dependencies [BZ #26926]

The _dl_open_check and _rtld_main_check hooks are not called on the
dependencies of a loaded module, so BTI protection was missed on
every module other than the main executable and directly dlopened
libraries.

The fix just iterates over dependencies to enable BTI.

Fixes bug 26926.
---
 sysdeps/aarch64/dl-bti.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/sysdeps/aarch64/dl-bti.c b/sysdeps/aarch64/dl-bti.c
index 196e462520..56c097210a 100644
--- a/sysdeps/aarch64/dl-bti.c
+++ b/sysdeps/aarch64/dl-bti.c
@@ -51,11 +51,24 @@ enable_bti (struct link_map *map, const char *program)
   return 0;
 }
 
-/* Enable BTI for L if required.  */
+/* Enable BTI for L and its dependencies.  */
 
 void
 _dl_bti_check (struct link_map *l, const char *program)
 {
-  if (GLRO(dl_aarch64_cpu_features).bti && l->l_mach.bti)
+  if (!GLRO(dl_aarch64_cpu_features).bti)
+    return;
+
+  if (l->l_mach.bti)
     enable_bti (l, program);
+
+  unsigned int i = l->l_searchlist.r_nlist;
+  while (i-- > 0)
+    {
+      struct link_map *dep = l->l_initfini[i];
+      if (dep->l_init_called)
+	continue;
+      if (dep->l_mach.bti)
+	enable_bti (dep, program);
+    }
 }
-- 
2.17.1


WARNING: multiple messages have this Message-ID (diff)
From: Szabolcs Nagy <szabolcs.nagy@arm.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>,
	libc-alpha@sourceware.org, kernel-hardening@lists.openwall.com,
	Catalin Marinas <catalin.marinas@arm.com>,
	linux-kernel@vger.kernel.org,
	Jeremy Linton <jeremy.linton@arm.com>,
	Mark Brown <broonie@kernel.org>,
	Topi Miettinen <toiwoton@gmail.com>,
	Will Deacon <will@kernel.org>,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v2 1/6] aarch64: Fix missing BTI protection from dependencies [BZ #26926]
Date: Fri, 11 Dec 2020 15:33:34 +0000	[thread overview]
Message-ID: <20201211153334.GF24625@arm.com> (raw)
In-Reply-To: <f077a6d5-082c-c3a5-ce07-71b87a70dc12@linaro.org>

[-- Attachment #1: Type: text/plain, Size: 789 bytes --]

The 12/10/2020 14:51, Adhemerval Zanella wrote:
> On 27/11/2020 10:19, Szabolcs Nagy via Libc-alpha wrote:
> > The _dl_open_check and _rtld_main_check hooks are not called on the
> > dependencies of a loaded module, so BTI protection was missed on
> > every module other than the main executable and directly dlopened
> > libraries.
> > 
> > The fix just iterates over dependencies to enable BTI.
> > 
> > Fixes bug 26926.
> 
> LGTM, modulus the argument name change.
> 
> I also think it would be better to add a testcase, for both DT_NEEDED
> and dlopen case.

thanks, i committed this with fixed argument name as attached.

i plan to do tests later once i understand the BTI semantics
(right now it's not clear if in the presence of some W^X
policy BTI may be silently ignored or not).

[-- Attachment #2: 0001-aarch64-Fix-missing-BTI-protection-from-dependencies.patch --]
[-- Type: text/x-diff, Size: 1446 bytes --]

From 72739c79f61989a76b7dd719f34fcfb7b8eadde9 Mon Sep 17 00:00:00 2001
From: Szabolcs Nagy <szabolcs.nagy@arm.com>
Date: Fri, 20 Nov 2020 15:27:06 +0000
Subject: [PATCH] aarch64: Fix missing BTI protection from dependencies [BZ #26926]

The _dl_open_check and _rtld_main_check hooks are not called on the
dependencies of a loaded module, so BTI protection was missed on
every module other than the main executable and directly dlopened
libraries.

The fix just iterates over dependencies to enable BTI.

Fixes bug 26926.
---
 sysdeps/aarch64/dl-bti.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)

diff --git a/sysdeps/aarch64/dl-bti.c b/sysdeps/aarch64/dl-bti.c
index 196e462520..56c097210a 100644
--- a/sysdeps/aarch64/dl-bti.c
+++ b/sysdeps/aarch64/dl-bti.c
@@ -51,11 +51,24 @@ enable_bti (struct link_map *map, const char *program)
   return 0;
 }
 
-/* Enable BTI for L if required.  */
+/* Enable BTI for L and its dependencies.  */
 
 void
 _dl_bti_check (struct link_map *l, const char *program)
 {
-  if (GLRO(dl_aarch64_cpu_features).bti && l->l_mach.bti)
+  if (!GLRO(dl_aarch64_cpu_features).bti)
+    return;
+
+  if (l->l_mach.bti)
     enable_bti (l, program);
+
+  unsigned int i = l->l_searchlist.r_nlist;
+  while (i-- > 0)
+    {
+      struct link_map *dep = l->l_initfini[i];
+      if (dep->l_init_called)
+	continue;
+      if (dep->l_mach.bti)
+	enable_bti (dep, program);
+    }
 }
-- 
2.17.1


[-- Attachment #3: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply	other threads:[~2020-12-11 16:32 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-27 13:19 [PATCH v2 0/6] aarch64: avoid mprotect(PROT_BTI|PROT_EXEC) [BZ #26831] Szabolcs Nagy
2020-11-27 13:19 ` Szabolcs Nagy
2020-11-27 13:19 ` [PATCH v2 1/6] aarch64: Fix missing BTI protection from dependencies [BZ #26926] Szabolcs Nagy
2020-11-27 13:19   ` Szabolcs Nagy
2020-12-10 17:51   ` Adhemerval Zanella
2020-12-10 17:51     ` Adhemerval Zanella
2020-12-11 15:33     ` Szabolcs Nagy [this message]
2020-12-11 15:33       ` Szabolcs Nagy
2020-11-27 13:20 ` [PATCH v2 2/6] elf: lose is closely tied to _dl_map_object_from_fd Szabolcs Nagy
2020-11-27 13:20   ` Szabolcs Nagy
2020-11-27 13:20 ` [PATCH v2 3/6] elf: Fix failure handling in _dl_map_object_from_fd Szabolcs Nagy
2020-11-27 13:20   ` Szabolcs Nagy
2020-12-10 18:25   ` Adhemerval Zanella
2020-12-10 18:25     ` Adhemerval Zanella
2020-12-11  9:32     ` Szabolcs Nagy
2020-12-11  9:32       ` Szabolcs Nagy
2020-11-27 13:20 ` [PATCH v2 4/6] elf: Move note processing after l_phdr is updated Szabolcs Nagy
2020-11-27 13:20   ` Szabolcs Nagy
2020-11-27 13:21 ` [PATCH v2 5/6] elf: Pass the fd to note processing Szabolcs Nagy
2020-11-27 13:21   ` Szabolcs Nagy
2020-12-10 18:35   ` Adhemerval Zanella
2020-12-10 18:35     ` Adhemerval Zanella
2020-11-27 13:21 ` [PATCH v2 6/6] aarch64: Use mmap to add PROT_BTI instead of mprotect [BZ #26831] Szabolcs Nagy
2020-11-27 13:21   ` Szabolcs Nagy
2020-12-02  8:55   ` [PATCH v3 1/2] aarch64: align address for BTI protection [BZ #26988] Szabolcs Nagy
2020-12-02  8:55     ` Szabolcs Nagy
2020-12-10 18:49     ` Adhemerval Zanella
2020-12-10 18:49       ` Adhemerval Zanella
2020-12-02  8:55   ` [PATCH v3 2/2] aarch64: Use mmap to add PROT_BTI instead of mprotect [BZ #26831] Szabolcs Nagy
2020-12-02  8:55     ` Szabolcs Nagy
2020-12-10 19:12     ` Adhemerval Zanella
2020-12-10 19:12       ` Adhemerval Zanella
2020-11-30 15:56 ` [PATCH v2 0/6] aarch64: avoid mprotect(PROT_BTI|PROT_EXEC) " Szabolcs Nagy
2020-11-30 15:56   ` Szabolcs Nagy
2020-12-03 17:30 ` Catalin Marinas
2020-12-03 17:30   ` Catalin Marinas
2020-12-07 20:03   ` Szabolcs Nagy
2020-12-07 20:03     ` Szabolcs Nagy
2020-12-11 17:46     ` Catalin Marinas
2020-12-11 17:46       ` Catalin Marinas

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=20201211153334.GF24625@arm.com \
    --to=szabolcs.nagy@arm.com \
    --cc=adhemerval.zanella@linaro.org \
    --cc=broonie@kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=jeremy.linton@arm.com \
    --cc=kernel-hardening@lists.openwall.com \
    --cc=libc-alpha@sourceware.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=toiwoton@gmail.com \
    --cc=will@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.