All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christophe Leroy <christophe.leroy@csgroup.eu>
To: Josh Poimboeuf <jpoimboe@redhat.com>,
	Jiri Kosina <jikos@kernel.org>, Miroslav Benes <mbenes@suse.cz>,
	Petr Mladek <pmladek@suse.com>,
	Joe Lawrence <joe.lawrence@redhat.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ingo Molnar <mingo@redhat.com>,
	"Naveen N . Rao" <naveen.n.rao@linux.vnet.ibm.com>
Cc: Christophe Leroy <christophe.leroy@csgroup.eu>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>,
	"live-patching@vger.kernel.org" <live-patching@vger.kernel.org>,
	Russell Currey <ruscur@russell.cc>
Subject: [PATCH v2 03/13] powerpc/module_32: Fix livepatching for RO modules
Date: Mon, 20 Dec 2021 16:38:09 +0000	[thread overview]
Message-ID: <d5697157cb7dba3927e19aa17c915a83bc550bb2.1640017960.git.christophe.leroy@csgroup.eu> (raw)
In-Reply-To: <cover.1640017960.git.christophe.leroy@csgroup.eu>

Livepatching a loaded module involves applying relocations through
apply_relocate_add(), which attempts to write to read-only memory when
CONFIG_STRICT_MODULE_RWX=y.

R_PPC_ADDR16_LO, R_PPC_ADDR16_HI, R_PPC_ADDR16_HA and R_PPC_REL24 are
the types generated by the kpatch-build userspace tool or klp-convert
kernel tree observed applying a relocation to a post-init module.

Use patch_instruction() to patch those relocations.

Commit 8734b41b3efe ("powerpc/module_64: Fix livepatching for
RO modules") did similar change in module_64.

Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Russell Currey <ruscur@russell.cc>
---
 arch/powerpc/kernel/module_32.c | 44 ++++++++++++++++++++++-----------
 1 file changed, 30 insertions(+), 14 deletions(-)

diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c
index a491ad481d85..a0432ef46967 100644
--- a/arch/powerpc/kernel/module_32.c
+++ b/arch/powerpc/kernel/module_32.c
@@ -18,6 +18,7 @@
 #include <linux/bug.h>
 #include <linux/sort.h>
 #include <asm/setup.h>
+#include <asm/code-patching.h>
 
 /* Count how many different relocations (different symbol, different
    addend) */
@@ -174,15 +175,25 @@ static uint32_t do_plt_call(void *location,
 		entry++;
 	}
 
-	entry->jump[0] = PPC_RAW_LIS(_R12, PPC_HA(val));
-	entry->jump[1] = PPC_RAW_ADDI(_R12, _R12, PPC_LO(val));
-	entry->jump[2] = PPC_RAW_MTCTR(_R12);
-	entry->jump[3] = PPC_RAW_BCTR();
+	if (patch_instruction(&entry->jump[0], ppc_inst(PPC_RAW_LIS(_R12, PPC_HA(val)))))
+		return 0;
+	if (patch_instruction(&entry->jump[1], ppc_inst(PPC_RAW_ADDI(_R12, _R12, PPC_LO(val)))))
+		return 0;
+	if (patch_instruction(&entry->jump[2], ppc_inst(PPC_RAW_MTCTR(_R12))))
+		return 0;
+	if (patch_instruction(&entry->jump[3], ppc_inst(PPC_RAW_BCTR())))
+		return 0;
 
 	pr_debug("Initialized plt for 0x%x at %p\n", val, entry);
 	return (uint32_t)entry;
 }
 
+static int patch_location_16(uint32_t *loc, u16 value)
+{
+	loc = PTR_ALIGN_DOWN(loc, sizeof(u32));
+	return patch_instruction(loc, ppc_inst((*loc & 0xffff0000) | value));
+}
+
 int apply_relocate_add(Elf32_Shdr *sechdrs,
 		       const char *strtab,
 		       unsigned int symindex,
@@ -216,37 +227,42 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 
 		case R_PPC_ADDR16_LO:
 			/* Low half of the symbol */
-			*(uint16_t *)location = value;
+			if (patch_location_16(location, PPC_LO(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_ADDR16_HI:
 			/* Higher half of the symbol */
-			*(uint16_t *)location = (value >> 16);
+			if (patch_location_16(location, PPC_HI(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_ADDR16_HA:
-			/* Sign-adjusted lower 16 bits: PPC ELF ABI says:
-			   (((x >> 16) + ((x & 0x8000) ? 1 : 0))) & 0xFFFF.
-			   This is the same, only sane.
-			 */
-			*(uint16_t *)location = (value + 0x8000) >> 16;
+			if (patch_location_16(location, PPC_HA(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_REL24:
 			if ((int)(value - (uint32_t)location) < -0x02000000
-			    || (int)(value - (uint32_t)location) >= 0x02000000)
+			    || (int)(value - (uint32_t)location) >= 0x02000000) {
 				value = do_plt_call(location, value,
 						    sechdrs, module);
+				if (!value)
+					return -EFAULT;
+			}
 
 			/* Only replace bits 2 through 26 */
 			pr_debug("REL24 value = %08X. location = %08X\n",
 			       value, (uint32_t)location);
 			pr_debug("Location before: %08X.\n",
 			       *(uint32_t *)location);
-			*(uint32_t *)location
-				= (*(uint32_t *)location & ~0x03fffffc)
+			value = (*(uint32_t *)location & ~0x03fffffc)
 				| ((value - (uint32_t)location)
 				   & 0x03fffffc);
+
+			if (patch_instruction(location, ppc_inst(value)))
+				return -EFAULT;
+
 			pr_debug("Location after: %08X.\n",
 			       *(uint32_t *)location);
 			pr_debug("ie. jump to %08X+%08X = %08X\n",
-- 
2.33.1

WARNING: multiple messages have this Message-ID (diff)
From: Christophe Leroy <christophe.leroy@csgroup.eu>
To: Josh Poimboeuf <jpoimboe@redhat.com>,
	Jiri Kosina <jikos@kernel.org>, Miroslav Benes <mbenes@suse.cz>,
	Petr Mladek <pmladek@suse.com>,
	Joe Lawrence <joe.lawrence@redhat.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ingo Molnar <mingo@redhat.com>,
	"Naveen N . Rao" <naveen.n.rao@linux.vnet.ibm.com>
Cc: "live-patching@vger.kernel.org" <live-patching@vger.kernel.org>,
	"linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: [PATCH v2 03/13] powerpc/module_32: Fix livepatching for RO modules
Date: Mon, 20 Dec 2021 16:38:09 +0000	[thread overview]
Message-ID: <d5697157cb7dba3927e19aa17c915a83bc550bb2.1640017960.git.christophe.leroy@csgroup.eu> (raw)
In-Reply-To: <cover.1640017960.git.christophe.leroy@csgroup.eu>

Livepatching a loaded module involves applying relocations through
apply_relocate_add(), which attempts to write to read-only memory when
CONFIG_STRICT_MODULE_RWX=y.

R_PPC_ADDR16_LO, R_PPC_ADDR16_HI, R_PPC_ADDR16_HA and R_PPC_REL24 are
the types generated by the kpatch-build userspace tool or klp-convert
kernel tree observed applying a relocation to a post-init module.

Use patch_instruction() to patch those relocations.

Commit 8734b41b3efe ("powerpc/module_64: Fix livepatching for
RO modules") did similar change in module_64.

Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: Russell Currey <ruscur@russell.cc>
---
 arch/powerpc/kernel/module_32.c | 44 ++++++++++++++++++++++-----------
 1 file changed, 30 insertions(+), 14 deletions(-)

diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c
index a491ad481d85..a0432ef46967 100644
--- a/arch/powerpc/kernel/module_32.c
+++ b/arch/powerpc/kernel/module_32.c
@@ -18,6 +18,7 @@
 #include <linux/bug.h>
 #include <linux/sort.h>
 #include <asm/setup.h>
+#include <asm/code-patching.h>
 
 /* Count how many different relocations (different symbol, different
    addend) */
@@ -174,15 +175,25 @@ static uint32_t do_plt_call(void *location,
 		entry++;
 	}
 
-	entry->jump[0] = PPC_RAW_LIS(_R12, PPC_HA(val));
-	entry->jump[1] = PPC_RAW_ADDI(_R12, _R12, PPC_LO(val));
-	entry->jump[2] = PPC_RAW_MTCTR(_R12);
-	entry->jump[3] = PPC_RAW_BCTR();
+	if (patch_instruction(&entry->jump[0], ppc_inst(PPC_RAW_LIS(_R12, PPC_HA(val)))))
+		return 0;
+	if (patch_instruction(&entry->jump[1], ppc_inst(PPC_RAW_ADDI(_R12, _R12, PPC_LO(val)))))
+		return 0;
+	if (patch_instruction(&entry->jump[2], ppc_inst(PPC_RAW_MTCTR(_R12))))
+		return 0;
+	if (patch_instruction(&entry->jump[3], ppc_inst(PPC_RAW_BCTR())))
+		return 0;
 
 	pr_debug("Initialized plt for 0x%x at %p\n", val, entry);
 	return (uint32_t)entry;
 }
 
+static int patch_location_16(uint32_t *loc, u16 value)
+{
+	loc = PTR_ALIGN_DOWN(loc, sizeof(u32));
+	return patch_instruction(loc, ppc_inst((*loc & 0xffff0000) | value));
+}
+
 int apply_relocate_add(Elf32_Shdr *sechdrs,
 		       const char *strtab,
 		       unsigned int symindex,
@@ -216,37 +227,42 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 
 		case R_PPC_ADDR16_LO:
 			/* Low half of the symbol */
-			*(uint16_t *)location = value;
+			if (patch_location_16(location, PPC_LO(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_ADDR16_HI:
 			/* Higher half of the symbol */
-			*(uint16_t *)location = (value >> 16);
+			if (patch_location_16(location, PPC_HI(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_ADDR16_HA:
-			/* Sign-adjusted lower 16 bits: PPC ELF ABI says:
-			   (((x >> 16) + ((x & 0x8000) ? 1 : 0))) & 0xFFFF.
-			   This is the same, only sane.
-			 */
-			*(uint16_t *)location = (value + 0x8000) >> 16;
+			if (patch_location_16(location, PPC_HA(value)))
+				return -EFAULT;
 			break;
 
 		case R_PPC_REL24:
 			if ((int)(value - (uint32_t)location) < -0x02000000
-			    || (int)(value - (uint32_t)location) >= 0x02000000)
+			    || (int)(value - (uint32_t)location) >= 0x02000000) {
 				value = do_plt_call(location, value,
 						    sechdrs, module);
+				if (!value)
+					return -EFAULT;
+			}
 
 			/* Only replace bits 2 through 26 */
 			pr_debug("REL24 value = %08X. location = %08X\n",
 			       value, (uint32_t)location);
 			pr_debug("Location before: %08X.\n",
 			       *(uint32_t *)location);
-			*(uint32_t *)location
-				= (*(uint32_t *)location & ~0x03fffffc)
+			value = (*(uint32_t *)location & ~0x03fffffc)
 				| ((value - (uint32_t)location)
 				   & 0x03fffffc);
+
+			if (patch_instruction(location, ppc_inst(value)))
+				return -EFAULT;
+
 			pr_debug("Location after: %08X.\n",
 			       *(uint32_t *)location);
 			pr_debug("ie. jump to %08X+%08X = %08X\n",
-- 
2.33.1

  parent reply	other threads:[~2021-12-20 16:38 UTC|newest]

Thread overview: 84+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-20 16:37 [PATCH v2 00/13] Implement livepatch on PPC32 and more Christophe Leroy
2021-12-20 16:37 ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 01/13] livepatch: Fix build failure on 32 bits processors Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-22 13:47   ` Miroslav Benes
2021-12-22 13:47     ` Miroslav Benes
2022-01-04 19:35   ` Joe Lawrence
2022-01-04 19:35     ` Joe Lawrence
2021-12-20 16:38 ` [PATCH v2 02/13] tracing: Fix selftest config check for function graph start up test Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2022-02-24 13:43   ` Christophe Leroy
2022-02-24 13:43     ` Christophe Leroy
2022-02-24 14:53     ` Steven Rostedt
2022-02-24 14:53       ` Steven Rostedt
2022-02-24 15:13       ` Christophe Leroy
2022-02-24 15:13         ` Christophe Leroy
2022-02-24 15:17         ` Steven Rostedt
2022-02-24 15:17           ` Steven Rostedt
2022-02-25  2:42       ` Michael Ellerman
2022-02-25  2:42         ` Michael Ellerman
2021-12-20 16:38 ` Christophe Leroy [this message]
2021-12-20 16:38   ` [PATCH v2 03/13] powerpc/module_32: Fix livepatching for RO modules Christophe Leroy
2022-01-04 19:44   ` Joe Lawrence
2022-01-04 19:44     ` Joe Lawrence
2021-12-20 16:38 ` [PATCH v2 04/13] powerpc/ftrace: Add support for livepatch to PPC32 Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-22 14:00   ` Miroslav Benes
2021-12-22 14:00     ` Miroslav Benes
2021-12-20 16:38 ` [PATCH v2 05/13] powerpc/ftrace: Don't save again LR in ftrace_regs_caller() on PPC32 Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 06/13] powerpc/ftrace: Simplify PPC32's return_to_handler() Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 07/13] powerpc/ftrace: Prepare PPC32's ftrace_caller() for CONFIG_DYNAMIC_FTRACE_WITH_ARGS Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 08/13] powerpc/ftrace: Prepare PPC64's " Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2022-02-14 15:19   ` Naveen N. Rao
2022-02-14 15:19     ` Naveen N. Rao
2021-12-20 16:38 ` [PATCH v2 09/13] powerpc/ftrace: Implement CONFIG_DYNAMIC_FTRACE_WITH_ARGS Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-22 14:19   ` Miroslav Benes
2021-12-22 14:19     ` Miroslav Benes
2021-12-22 14:19     ` Miroslav Benes
2022-02-14 15:25   ` Naveen N. Rao
2022-02-14 15:25     ` Naveen N. Rao
2022-02-15  8:00     ` Christophe Leroy
2022-02-15  8:00       ` Christophe Leroy
2022-02-15 11:05       ` Michael Ellerman
2022-02-15 11:05         ` Michael Ellerman
2022-02-15 13:36         ` Naveen N. Rao
2022-02-15 13:36           ` Naveen N. Rao
2022-02-15 14:28           ` Christophe Leroy
2022-02-15 14:28             ` Christophe Leroy
2022-02-15 14:51             ` Christophe Leroy
2022-02-15 14:51               ` Christophe Leroy
2022-02-15 16:25               ` Naveen N. Rao
2022-02-15 16:25                 ` Naveen N. Rao
2022-02-16 13:04                 ` Heiko Carstens
2022-02-16 13:04                   ` Heiko Carstens
2022-02-16 13:27                   ` Sven Schnelle
2022-02-16 13:27                     ` Sven Schnelle
2022-02-15 14:38           ` Steven Rostedt
2022-02-15 14:38             ` Steven Rostedt
2022-02-15 16:26             ` Naveen N. Rao
2022-02-15 16:26               ` Naveen N. Rao
2021-12-20 16:38 ` [PATCH v2 10/13] powerpc/ftrace: Refactor ftrace_{en/dis}able_ftrace_graph_caller Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 11/13] powerpc/ftrace: directly call of function graph tracer by ftrace caller Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2022-02-14 17:24   ` Naveen N. Rao
2022-02-14 17:24     ` Naveen N. Rao
2022-02-14 19:03     ` Steven Rostedt
2022-02-14 19:03       ` Steven Rostedt
2021-12-20 16:38 ` [PATCH v2 12/13] powerpc/ftrace: Prepare ftrace_64_mprofile.S for reuse by PPC32 Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2022-02-14 17:51   ` Naveen N. Rao
2022-02-14 17:51     ` Naveen N. Rao
2022-02-15  8:33     ` Christophe Leroy
2022-02-15  8:33       ` Christophe Leroy
2021-12-20 16:38 ` [PATCH v2 13/13] powerpc/ftrace: Remove ftrace_32.S Christophe Leroy
2021-12-20 16:38   ` Christophe Leroy
2022-02-11  7:41   ` [PATCH] Fixup for next-test 3a1a8f078670 ("powerpc/ftrace: Remove ftrace_32.S") Christophe Leroy
2022-02-11  7:41     ` Christophe Leroy
2022-02-16 12:26 ` [PATCH v2 00/13] Implement livepatch on PPC32 and more Michael Ellerman

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=d5697157cb7dba3927e19aa17c915a83bc550bb2.1640017960.git.christophe.leroy@csgroup.eu \
    --to=christophe.leroy@csgroup.eu \
    --cc=jikos@kernel.org \
    --cc=joe.lawrence@redhat.com \
    --cc=jpoimboe@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=live-patching@vger.kernel.org \
    --cc=mbenes@suse.cz \
    --cc=mingo@redhat.com \
    --cc=naveen.n.rao@linux.vnet.ibm.com \
    --cc=pmladek@suse.com \
    --cc=rostedt@goodmis.org \
    --cc=ruscur@russell.cc \
    /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.