From: Peter Zijlstra <peterz@infradead.org>
To: x86@kernel.org, Borislav Petkov <bp@alien8.de>
Cc: Masami Hiramatsu <mhiramat@kernel.org>,
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>,
Andrew Cooper <Andrew.Cooper3@citrix.com>,
Josh Poimboeuf <jpoimboe@redhat.com>,
linux-kernel@vger.kernel.org
Subject: [RFC] x86/alternative: Support relocations in alternatives
Date: Wed, 1 Feb 2023 15:10:33 +0100 [thread overview]
Message-ID: <Y9py2a5Xw0xbB8ou@hirez.programming.kicks-ass.net> (raw)
(sorry for the repost, forgot lkml)
A little while ago someone (Kirill) ran into the whole 'alternatives don't
do relocations nonsense' again and I got annoyed enough to actually look
at the code.
Since the whole alternative machinery already fully decodes the
instructions it is simple enough to adjust immediates and displacement
when needed. Specifically, the immediates for IP modifying instructions
(JMP, CALL, Jcc) and the displacement for RIP-relative instructions.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
arch/x86/kernel/alternative.c | 199 ++++++++++++++++++++++-----------------
tools/objtool/arch/x86/special.c | 8 +-
2 files changed, 112 insertions(+), 95 deletions(-)
diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c
index 7d8c3cbde368..f9a579fe47b0 100644
--- a/arch/x86/kernel/alternative.c
+++ b/arch/x86/kernel/alternative.c
@@ -122,71 +122,6 @@ extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
extern s32 __smp_locks[], __smp_locks_end[];
void text_poke_early(void *addr, const void *opcode, size_t len);
-/*
- * Are we looking at a near JMP with a 1 or 4-byte displacement.
- */
-static inline bool is_jmp(const u8 opcode)
-{
- return opcode == 0xeb || opcode == 0xe9;
-}
-
-static void __init_or_module
-recompute_jump(struct alt_instr *a, u8 *orig_insn, u8 *repl_insn, u8 *insn_buff)
-{
- u8 *next_rip, *tgt_rip;
- s32 n_dspl, o_dspl;
- int repl_len;
-
- if (a->replacementlen != 5)
- return;
-
- o_dspl = *(s32 *)(insn_buff + 1);
-
- /* next_rip of the replacement JMP */
- next_rip = repl_insn + a->replacementlen;
- /* target rip of the replacement JMP */
- tgt_rip = next_rip + o_dspl;
- n_dspl = tgt_rip - orig_insn;
-
- DPRINTK("target RIP: %px, new_displ: 0x%x", tgt_rip, n_dspl);
-
- if (tgt_rip - orig_insn >= 0) {
- if (n_dspl - 2 <= 127)
- goto two_byte_jmp;
- else
- goto five_byte_jmp;
- /* negative offset */
- } else {
- if (((n_dspl - 2) & 0xff) == (n_dspl - 2))
- goto two_byte_jmp;
- else
- goto five_byte_jmp;
- }
-
-two_byte_jmp:
- n_dspl -= 2;
-
- insn_buff[0] = 0xeb;
- insn_buff[1] = (s8)n_dspl;
- add_nops(insn_buff + 2, 3);
-
- repl_len = 2;
- goto done;
-
-five_byte_jmp:
- n_dspl -= 5;
-
- insn_buff[0] = 0xe9;
- *(s32 *)&insn_buff[1] = n_dspl;
-
- repl_len = 5;
-
-done:
-
- DPRINTK("final displ: 0x%08x, JMP 0x%lx",
- n_dspl, (unsigned long)orig_insn + n_dspl + repl_len);
-}
-
/*
* optimize_nops_range() - Optimize a sequence of single byte NOPs (0x90)
*
@@ -253,6 +188,78 @@ static void __init_or_module noinline optimize_nops(u8 *instr, size_t len)
}
}
+#define apply_reloc_n(n_, p_, d_) \
+ do { \
+ s32 v = *(s##n_ *)(p_); \
+ v += (d_); \
+ BUG_ON((v >> 31) != (v >> (n_-1))); \
+ *(s##n_ *)(p_) = (s##n_)v; \
+ } while (0)
+
+
+static void __always_inline
+apply_reloc(int n, void *ptr, uintptr_t diff)
+{
+ switch (n) {
+ case 1: apply_reloc_n(8, ptr, diff); break;
+ case 2: apply_reloc_n(16, ptr, diff); break;
+ case 4: apply_reloc_n(32, ptr, diff); break;
+ default: BUG();
+ }
+}
+
+static void __init_or_module noinline
+apply_relocation(u8 *instr, size_t len, u8 *dest, u8 *src, size_t src_len)
+{
+ struct insn insn;
+ int i = 0;
+
+ for (;;) {
+ if (insn_decode_kernel(&insn, &instr[i]))
+ return;
+
+ switch (insn.opcode.bytes[0]) {
+ case 0x0f:
+ if (insn.opcode.bytes[1] < 0x80 ||
+ insn.opcode.bytes[1] > 0x8f)
+ break;
+
+ fallthrough; /* Jcc.d32 */
+ case 0x70 ... 0x7f: /* Jcc.d8 */
+ case JMP8_INSN_OPCODE:
+ case JMP32_INSN_OPCODE:
+ case CALL_INSN_OPCODE:
+ u8 *target = src + i + insn.length + insn.immediate.value;
+ if (target < src || target > src + src_len) {
+ apply_reloc(insn.immediate.nbytes,
+ instr + i + insn_offset_immediate(&insn),
+ src - dest);
+ }
+ }
+
+ if (insn_rip_relative(&insn)) {
+ u8 *target = src + i + insn.length + insn.displacement.value;
+ if (target < src || target > src + src_len) {
+ apply_reloc(insn.displacement.nbytes,
+ instr + i + insn_offset_displacement(&insn),
+ src - dest);
+ }
+ }
+
+ /*
+ * See if this and any potentially following NOPs can be
+ * optimized.
+ */
+ if (insn.length == 1 && insn.opcode.bytes[0] == 0x90)
+ i += optimize_nops_range(instr, len, i);
+ else
+ i += insn.length;
+
+ if (i >= len)
+ return;
+ }
+}
+
/*
* Replace instructions with better alternatives for this CPU type. This runs
* before SMP is initialized to avoid SMP problems with self modifying code.
@@ -296,8 +303,10 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
* - feature not present but ALTINSTR_FLAG_INV is set to mean,
* patch if feature is *NOT* present.
*/
- if (!boot_cpu_has(feature) == !(a->cpuid & ALTINSTR_FLAG_INV))
- goto next;
+ if (!boot_cpu_has(feature) == !(a->cpuid & ALTINSTR_FLAG_INV)) {
+ optimize_nops(instr, a->instrlen);
+ continue;
+ }
DPRINTK("feat: %s%d*32+%d, old: (%pS (%px) len: %d), repl: (%px, len: %d)",
(a->cpuid & ALTINSTR_FLAG_INV) ? "!" : "",
@@ -306,37 +315,20 @@ void __init_or_module noinline apply_alternatives(struct alt_instr *start,
instr, instr, a->instrlen,
replacement, a->replacementlen);
- DUMP_BYTES(instr, a->instrlen, "%px: old_insn: ", instr);
- DUMP_BYTES(replacement, a->replacementlen, "%px: rpl_insn: ", replacement);
memcpy(insn_buff, replacement, a->replacementlen);
insn_buff_sz = a->replacementlen;
- /*
- * 0xe8 is a relative jump; fix the offset.
- *
- * Instruction length is checked before the opcode to avoid
- * accessing uninitialized bytes for zero-length replacements.
- */
- if (a->replacementlen == 5 && *insn_buff == 0xe8) {
- *(s32 *)(insn_buff + 1) += replacement - instr;
- DPRINTK("Fix CALL offset: 0x%x, CALL 0x%lx",
- *(s32 *)(insn_buff + 1),
- (unsigned long)instr + *(s32 *)(insn_buff + 1) + 5);
- }
-
- if (a->replacementlen && is_jmp(replacement[0]))
- recompute_jump(a, instr, replacement, insn_buff);
-
for (; insn_buff_sz < a->instrlen; insn_buff_sz++)
insn_buff[insn_buff_sz] = 0x90;
+ apply_relocation(insn_buff, a->instrlen, instr, replacement, a->replacementlen);
+
+ DUMP_BYTES(instr, a->instrlen, "%px: old_insn: ", instr);
+ DUMP_BYTES(replacement, a->replacementlen, "%px: rpl_insn: ", replacement);
DUMP_BYTES(insn_buff, insn_buff_sz, "%px: final_insn: ", instr);
text_poke_early(instr, insn_buff, insn_buff_sz);
-
-next:
- optimize_nops(instr, a->instrlen);
}
}
@@ -1334,6 +1326,35 @@ static noinline void __init int3_selftest(void)
unregister_die_notifier(&int3_exception_nb);
}
+static __initdata int __alt_reloc_selftest_addr;
+
+__visible noinline void __init __alt_reloc_selftest(void *arg)
+{
+ WARN_ON(arg != &__alt_reloc_selftest_addr);
+}
+
+static noinline void __init alt_reloc_selftest(void)
+{
+ /*
+ * Tests apply_relocation().
+ *
+ * This has a relative immediate (CALL) in a place other than the first
+ * instruction and additionally on x86_64 we get a RIP-relative LEA:
+ *
+ * lea 0x0(%rip),%rdi # 5d0: R_X86_64_PC32 .init.data+0x5566c
+ * call +0 # 5d5: R_X86_64_PLT32 __alt_reloc_selftest-0x4
+ *
+ * Getting this wrong will either crash and burn or tickle the WARN
+ * above.
+ */
+ asm_inline volatile (
+ ALTERNATIVE("", "lea %[mem], %%" _ASM_ARG1 "; call __alt_reloc_selftest;", X86_FEATURE_ALWAYS)
+ : /* output */
+ : [mem] "m" (__alt_reloc_selftest_addr)
+ : _ASM_ARG1
+ );
+}
+
void __init alternative_instructions(void)
{
int3_selftest();
@@ -1421,6 +1442,8 @@ void __init alternative_instructions(void)
restart_nmi();
alternatives_patched = 1;
+
+ alt_reloc_selftest();
}
/**
diff --git a/tools/objtool/arch/x86/special.c b/tools/objtool/arch/x86/special.c
index 7c97b7391279..799ad6bb72e5 100644
--- a/tools/objtool/arch/x86/special.c
+++ b/tools/objtool/arch/x86/special.c
@@ -42,13 +42,7 @@ bool arch_support_alt_relocation(struct special_alt *special_alt,
struct instruction *insn,
struct reloc *reloc)
{
- /*
- * The x86 alternatives code adjusts the offsets only when it
- * encounters a branch instruction at the very beginning of the
- * replacement group.
- */
- return insn->offset == special_alt->new_off &&
- (insn->type == INSN_CALL || is_jump(insn));
+ return true;
}
/*
next reply other threads:[~2023-02-01 14:10 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-01 14:10 Peter Zijlstra [this message]
2023-02-01 17:02 ` [RFC] x86/alternative: Support relocations in alternatives Kirill A. Shutemov
2023-02-03 13:13 ` Borislav Petkov
2023-02-03 15:04 ` Peter Zijlstra
2023-02-03 15:12 ` Peter Zijlstra
2023-02-03 16:04 ` Borislav Petkov
2023-02-03 16:46 ` Borislav Petkov
2023-02-03 23:25 ` Borislav Petkov
2023-02-06 14:29 ` Borislav Petkov
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=Y9py2a5Xw0xbB8ou@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=Andrew.Cooper3@citrix.com \
--cc=bp@alien8.de \
--cc=jpoimboe@redhat.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhiramat@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox