#define MAX_PATCH_LEN (255-1)
-static int __initdata_or_module debug_alternative;
+#define DA_ALL (~0)
+#define DA_ALT 0x01
+#define DA_RET 0x02
+#define DA_RETPOLINE 0x04
+#define DA_ENDBR 0x08
+#define DA_SMP 0x10
+
+static unsigned int __initdata_or_module debug_alternative;
static int __init debug_alt(char *str)
{
- debug_alternative = 1;
+ if (str && *str == '=')
+ str++;
+
+ if (!str || kstrtouint(str, 0, &debug_alternative))
+ debug_alternative = DA_ALL;
+
return 1;
}
__setup("debug-alternative", debug_alt);
}
__setup("noreplace-smp", setup_noreplace_smp);
-#define DPRINTK(fmt, args...) \
+#define DPRINTK(type, fmt, args...) \
do { \
- if (debug_alternative) \
+ if (debug_alternative & DA_##type) \
printk(KERN_DEBUG pr_fmt(fmt) "\n", ##args); \
} while (0)
-#define DUMP_BYTES(buf, len, fmt, args...) \
+#define DUMP_BYTES(type, buf, len, fmt, args...) \
do { \
- if (unlikely(debug_alternative)) { \
+ if (unlikely(debug_alternative & DA_##type)) { \
int j; \
\
if (!(len)) \
BYTES_NOP6,
BYTES_NOP7,
BYTES_NOP8,
+#ifdef CONFIG_64BIT
+ BYTES_NOP9,
+ BYTES_NOP10,
+ BYTES_NOP11,
+#endif
};
const unsigned char * const x86_nops[ASM_NOP_MAX+1] =
x86nops + 1 + 2 + 3 + 4 + 5,
x86nops + 1 + 2 + 3 + 4 + 5 + 6,
x86nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+#ifdef CONFIG_64BIT
+ x86nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8,
+ x86nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9,
+ x86nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8 + 9 + 10,
+#endif
};
-/* Use this to add nops to a buffer, then text_poke the whole buffer. */
-static void __init_or_module add_nops(void *insns, unsigned int len)
+/*
+ * Fill the buffer with a single effective instruction of size @len.
+ *
+ * In order not to issue an ORC stack depth tracking CFI entry (Call Frame Info)
+ * for every single-byte NOP, try to generate the maximally available NOP of
+ * size <= ASM_NOP_MAX such that only a single CFI entry is generated (vs one for
+ * each single-byte NOPs). If @len to fill out is > ASM_NOP_MAX, pad with INT3 and
+ * *jump* over instead of executing long and daft NOPs.
+ */
+static void __init_or_module add_nop(u8 *instr, unsigned int len)
{
- while (len > 0) {
- unsigned int noplen = len;
- if (noplen > ASM_NOP_MAX)
- noplen = ASM_NOP_MAX;
- memcpy(insns, x86_nops[noplen], noplen);
- insns += noplen;
- len -= noplen;
+ u8 *target = instr + len;
+
+ if (!len)
+ return;
+
+ if (len <= ASM_NOP_MAX) {
+ memcpy(instr, x86_nops[len], len);
+ return;
+ }
+
+ if (len < 128) {
+ __text_gen_insn(instr, JMP8_INSN_OPCODE, instr, target, JMP8_INSN_SIZE);
+ instr += JMP8_INSN_SIZE;
+ } else {
+ __text_gen_insn(instr, JMP32_INSN_OPCODE, instr, target, JMP32_INSN_SIZE);
+ instr += JMP32_INSN_SIZE;
}
+
+ for (;instr < target; instr++)
+ *instr = INT3_INSN_OPCODE;
}
extern s32 __retpoline_sites[], __retpoline_sites_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.
+ * Matches NOP and NOPL, not any of the other possible NOPs.
*/
-static inline bool is_jmp(const u8 opcode)
+static bool insn_is_nop(struct insn *insn)
{
- return opcode == 0xeb || opcode == 0xe9;
+ /* Anything NOP, but no REP NOP */
+ if (insn->opcode.bytes[0] == 0x90 &&
+ (!insn->prefixes.nbytes || insn->prefixes.bytes[0] != 0xF3))
+ return true;
+
+ /* NOPL */
+ if (insn->opcode.bytes[0] == 0x0F && insn->opcode.bytes[1] == 0x1F)
+ return true;
+
+ /* TODO: more nops */
+
+ return false;
}
-static void __init_or_module
-recompute_jump(struct alt_instr *a, u8 *orig_insn, u8 *repl_insn, u8 *insn_buff)
+/*
+ * Find the offset of the first non-NOP instruction starting at @offset
+ * but no further than @len.
+ */
+static int skip_nops(u8 *instr, int offset, int len)
{
- u8 *next_rip, *tgt_rip;
- s32 n_dspl, o_dspl;
- int repl_len;
+ struct insn insn;
- if (a->replacementlen != 5)
- return;
+ for (; offset < len; offset += insn.length) {
+ if (insn_decode_kernel(&insn, &instr[offset]))
+ break;
- o_dspl = *(s32 *)(insn_buff + 1);
+ if (!insn_is_nop(&insn))
+ break;
+ }
- /* 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;
+ return offset;
+}
- DPRINTK("target RIP: %px, new_displ: 0x%x", tgt_rip, n_dspl);
+/*
+ * Optimize a sequence of NOPs, possibly preceded by an unconditional jump
+ * to the end of the NOP sequence into a single NOP.
+ */
+static bool __init_or_module
+__optimize_nops(u8 *instr, size_t len, struct insn *insn, int *next, int *prev, int *target)
+{
+ int i = *next - insn->length;
- 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;
+ switch (insn->opcode.bytes[0]) {
+ case JMP8_INSN_OPCODE:
+ case JMP32_INSN_OPCODE:
+ *prev = i;
+ *target = *next + insn->immediate.value;
+ return false;
}
-two_byte_jmp:
- n_dspl -= 2;
+ if (insn_is_nop(insn)) {
+ int nop = i;
- insn_buff[0] = 0xeb;
- insn_buff[1] = (s8)n_dspl;
- add_nops(insn_buff + 2, 3);
+ *next = skip_nops(instr, *next, len);
+ if (*target && *next == *target)
+ nop = *prev;
- repl_len = 2;
- goto done;
+ add_nop(instr + nop, *next - nop);
+ DUMP_BYTES(ALT, instr, len, "%px: [%d:%d) optimized NOPs: ", instr, nop, *next);
+ return true;
+ }
-five_byte_jmp:
- n_dspl -= 5;
+ *target = 0;
+ return false;
+}
- insn_buff[0] = 0xe9;
- *(s32 *)&insn_buff[1] = n_dspl;
+/*
+ * "noinline" to cause control flow change and thus invalidate I$ and
+ * cause refetch after modification.
+ */
+static void __init_or_module noinline optimize_nops(u8 *instr, size_t len)
+{
+ int prev, target = 0;
- repl_len = 5;
+ for (int next, i = 0; i < len; i = next) {
+ struct insn insn;
-done:
+ if (insn_decode_kernel(&insn, &instr[i]))
+ return;
- DPRINTK("final displ: 0x%08x, JMP 0x%lx",
- n_dspl, (unsigned long)orig_insn + n_dspl + repl_len);
+ next = i + insn.length;
+
+ __optimize_nops(instr, len, &insn, &next, &prev, &target);
+ }
}
/*
- * optimize_nops_range() - Optimize a sequence of single byte NOPs (0x90)
+ * In this context, "source" is where the instructions are placed in the
+ * section .altinstr_replacement, for example during kernel build by the
+ * toolchain.
+ * "Destination" is where the instructions are being patched in by this
+ * machinery.
+ *
+ * The source offset is:
+ *
+ * src_imm = target - src_next_ip (1)
*
- * @instr: instruction byte stream
- * @instrlen: length of the above
- * @off: offset within @instr where the first NOP has been detected
+ * and the target offset is:
*
- * Return: number of NOPs found (and replaced).
+ * dst_imm = target - dst_next_ip (2)
+ *
+ * so rework (1) as an expression for target like:
+ *
+ * target = src_imm + src_next_ip (1a)
+ *
+ * and substitute in (2) to get:
+ *
+ * dst_imm = (src_imm + src_next_ip) - dst_next_ip (3)
+ *
+ * Now, since the instruction stream is 'identical' at src and dst (it
+ * is being copied after all) it can be stated that:
+ *
+ * src_next_ip = src + ip_offset
+ * dst_next_ip = dst + ip_offset (4)
+ *
+ * Substitute (4) in (3) and observe ip_offset being cancelled out to
+ * obtain:
+ *
+ * dst_imm = src_imm + (src + ip_offset) - (dst + ip_offset)
+ * = src_imm + src - dst + ip_offset - ip_offset
+ * = src_imm + src - dst (5)
+ *
+ * IOW, only the relative displacement of the code block matters.
*/
-static __always_inline int optimize_nops_range(u8 *instr, u8 instrlen, int off)
-{
- unsigned long flags;
- int i = off, nnops;
- while (i < instrlen) {
- if (instr[i] != 0x90)
- break;
+#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)
+
- i++;
+static __always_inline
+void 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();
}
+}
- nnops = i - off;
+static __always_inline
+bool need_reloc(unsigned long offset, u8 *src, size_t src_len)
+{
+ u8 *target = src + offset;
+ /*
+ * If the target is inside the patched block, it's relative to the
+ * block itself and does not need relocation.
+ */
+ return (target < src || target > src + src_len);
+}
- if (nnops <= 1)
- return nnops;
+static void __init_or_module noinline
+apply_relocation(u8 *buf, size_t len, u8 *dest, u8 *src, size_t src_len)
+{
+ int prev, target = 0;
- local_irq_save(flags);
- add_nops(instr + off, nnops);
- local_irq_restore(flags);
+ for (int next, i = 0; i < len; i = next) {
+ struct insn insn;
- DUMP_BYTES(instr, instrlen, "%px: [%d:%d) optimized NOPs: ", instr, off, i);
+ if (WARN_ON_ONCE(insn_decode_kernel(&insn, &buf[i])))
+ return;
- return nnops;
-}
+ next = i + insn.length;
-/*
- * "noinline" to cause control flow change and thus invalidate I$ and
- * cause refetch after modification.
- */
-static void __init_or_module noinline optimize_nops(u8 *instr, size_t len)
-{
- struct insn insn;
- int i = 0;
+ if (__optimize_nops(buf, len, &insn, &next, &prev, &target))
+ continue;
- /*
- * Jump over the non-NOP insns and optimize single-byte NOPs into bigger
- * ones.
- */
- 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;
- /*
- * 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;
+ fallthrough; /* Jcc.d32 */
+ case 0x70 ... 0x7f: /* Jcc.d8 */
+ case JMP8_INSN_OPCODE:
+ case JMP32_INSN_OPCODE:
+ case CALL_INSN_OPCODE:
+ if (need_reloc(next + insn.immediate.value, src, src_len)) {
+ apply_reloc(insn.immediate.nbytes,
+ buf + i + insn_offset_immediate(&insn),
+ src - dest);
+ }
+
+ /*
+ * Where possible, convert JMP.d32 into JMP.d8.
+ */
+ if (insn.opcode.bytes[0] == JMP32_INSN_OPCODE) {
+ s32 imm = insn.immediate.value;
+ imm += src - dest;
+ imm += JMP32_INSN_SIZE - JMP8_INSN_SIZE;
+ if ((imm >> 31) == (imm >> 7)) {
+ buf[i+0] = JMP8_INSN_OPCODE;
+ buf[i+1] = (s8)imm;
+
+ memset(&buf[i+2], INT3_INSN_OPCODE, insn.length - 2);
+ }
+ }
+ break;
+ }
- if (i >= len)
- return;
+ if (insn_rip_relative(&insn)) {
+ if (need_reloc(next + insn.displacement.value, src, src_len)) {
+ apply_reloc(insn.displacement.nbytes,
+ buf + i + insn_offset_displacement(&insn),
+ src - dest);
+ }
+ }
}
}
u8 *instr, *replacement;
u8 insn_buff[MAX_PATCH_LEN];
- DPRINTK("alt table %px, -> %px", start, end);
+ DPRINTK(ALT, "alt table %px, -> %px", start, end);
/*
* The scan order should be from start to end. A later scanned
* alternative code can overwrite previously scanned alternative code.
* - feature not present but ALT_FLAG_NOT is set to mean,
* patch if feature is *NOT* present.
*/
- if (!boot_cpu_has(a->cpuid) == !(a->flags & ALT_FLAG_NOT))
- goto next;
+ if (!boot_cpu_has(a->cpuid) == !(a->flags & ALT_FLAG_NOT)) {
+ optimize_nops(instr, a->instrlen);
+ continue;
+ }
- DPRINTK("feat: %s%d*32+%d, old: (%pS (%px) len: %d), repl: (%px, len: %d)",
+ DPRINTK(ALT, "feat: %s%d*32+%d, old: (%pS (%px) len: %d), repl: (%px, len: %d)",
(a->flags & ALT_FLAG_NOT) ? "!" : "",
a->cpuid >> 5,
a->cpuid & 0x1f,
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;
- DUMP_BYTES(insn_buff, insn_buff_sz, "%px: final_insn: ", instr);
+ apply_relocation(insn_buff, a->instrlen, instr, replacement, a->replacementlen);
- text_poke_early(instr, insn_buff, insn_buff_sz);
+ DUMP_BYTES(ALT, instr, a->instrlen, "%px: old_insn: ", instr);
+ DUMP_BYTES(ALT, replacement, a->replacementlen, "%px: rpl_insn: ", replacement);
+ DUMP_BYTES(ALT, insn_buff, insn_buff_sz, "%px: final_insn: ", instr);
-next:
- optimize_nops(instr, a->instrlen);
+ text_poke_early(instr, insn_buff, insn_buff_sz);
}
}
continue;
}
- DPRINTK("retpoline at: %pS (%px) len: %d to: %pS",
+ DPRINTK(RETPOLINE, "retpoline at: %pS (%px) len: %d to: %pS",
addr, addr, insn.length,
addr + insn.length + insn.immediate.value);
len = patch_retpoline(addr, &insn, bytes);
if (len == insn.length) {
optimize_nops(bytes, len);
- DUMP_BYTES(((u8*)addr), len, "%px: orig: ", addr);
- DUMP_BYTES(((u8*)bytes), len, "%px: repl: ", addr);
+ DUMP_BYTES(RETPOLINE, ((u8*)addr), len, "%px: orig: ", addr);
+ DUMP_BYTES(RETPOLINE, ((u8*)bytes), len, "%px: repl: ", addr);
text_poke_early(addr, bytes, len);
}
}
{
int i = 0;
+ /* Patch the custom return thunks... */
if (cpu_feature_enabled(X86_FEATURE_RETHUNK)) {
- if (x86_return_thunk == __x86_return_thunk)
- return -1;
-
i = JMP32_INSN_SIZE;
__text_gen_insn(bytes, JMP32_INSN_OPCODE, addr, x86_return_thunk, i);
} else {
+ /* ... or patch them out if not needed. */
bytes[i++] = RET_INSN_OPCODE;
}
{
s32 *s;
+ /*
+ * Do not patch out the default return thunks if those needed are the
+ * ones generated by the compiler.
+ */
+ if (cpu_feature_enabled(X86_FEATURE_RETHUNK) &&
+ (x86_return_thunk == __x86_return_thunk))
+ return;
+
for (s = start; s < end; s++) {
void *dest = NULL, *addr = (void *)s + *s;
struct insn insn;
addr, dest, 5, addr))
continue;
- DPRINTK("return thunk at: %pS (%px) len: %d to: %pS",
+ DPRINTK(RET, "return thunk at: %pS (%px) len: %d to: %pS",
addr, addr, insn.length,
addr + insn.length + insn.immediate.value);
len = patch_return(addr, &insn, bytes);
if (len == insn.length) {
- DUMP_BYTES(((u8*)addr), len, "%px: orig: ", addr);
- DUMP_BYTES(((u8*)bytes), len, "%px: repl: ", addr);
+ DUMP_BYTES(RET, ((u8*)addr), len, "%px: orig: ", addr);
+ DUMP_BYTES(RET, ((u8*)bytes), len, "%px: repl: ", addr);
text_poke_early(addr, bytes, len);
}
}
#ifdef CONFIG_X86_KERNEL_IBT
-static void poison_endbr(void *addr, bool warn)
+static void __init_or_module poison_endbr(void *addr, bool warn)
{
u32 endbr, poison = gen_endbr_poison();
return;
}
- DPRINTK("ENDBR at: %pS (%px)", addr, addr);
+ DPRINTK(ENDBR, "ENDBR at: %pS (%px)", addr, addr);
/*
* When we have IBT, the lack of ENDBR will trigger #CP
*/
- DUMP_BYTES(((u8*)addr), 4, "%px: orig: ", addr);
- DUMP_BYTES(((u8*)&poison), 4, "%px: repl: ", addr);
+ DUMP_BYTES(ENDBR, ((u8*)addr), 4, "%px: orig: ", addr);
+ DUMP_BYTES(ENDBR, ((u8*)&poison), 4, "%px: repl: ", addr);
text_poke_early(addr, &poison, 4);
}
smp->locks_end = locks_end;
smp->text = text;
smp->text_end = text_end;
- DPRINTK("locks %p -> %p, text %p -> %p, name %s\n",
+ DPRINTK(SMP, "locks %p -> %p, text %p -> %p, name %s\n",
smp->locks, smp->locks_end,
smp->text, smp->text_end, smp->name);
#endif /* CONFIG_SMP */
#ifdef CONFIG_PARAVIRT
+
+/* Use this to add nops to a buffer, then text_poke the whole buffer. */
+static void __init_or_module add_nops(void *insns, unsigned int len)
+{
+ while (len > 0) {
+ unsigned int noplen = len;
+ if (noplen > ASM_NOP_MAX)
+ noplen = ASM_NOP_MAX;
+ memcpy(insns, x86_nops[noplen], noplen);
+ insns += noplen;
+ len -= noplen;
+ }
+}
+
void __init_or_module apply_paravirt(struct paravirt_patch_site *start,
struct paravirt_patch_site *end)
{
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();
restart_nmi();
alternatives_patched = 1;
+
+ alt_reloc_selftest();
}
/**
{
struct bp_patching_desc *desc = &bp_desc;
- if (!arch_atomic_inc_not_zero(&desc->refs))
+ if (!raw_atomic_inc_not_zero(&desc->refs))
return NULL;
return desc;
struct bp_patching_desc *desc = &bp_desc;
smp_mb__before_atomic();
- arch_atomic_dec(&desc->refs);
+ raw_atomic_dec(&desc->refs);
}
static __always_inline void *text_poke_addr(struct text_poke_loc *tp)
atomic_set_release(&bp_desc.refs, 1);
/*
+ * Function tracing can enable thousands of places that need to be
+ * updated. This can take quite some time, and with full kernel debugging
+ * enabled, this could cause the softlockup watchdog to trigger.
+ * This function gets called every 256 entries added to be patched.
+ * Call cond_resched() here to make sure that other tasks can get scheduled
+ * while processing all the functions being patched.
+ */
+ cond_resched();
+
+ /*
* Corresponding read barrier in int3 notifier for making sure the
* nr_entries and handler are correctly ordered wrt. patching.
*/
if (!timeout)
return ret;
- arch_atomic_add(*no_way_out, &global_nwo);
+ raw_atomic_add(*no_way_out, &global_nwo);
/*
* Rely on the implied barrier below, such that global_nwo
* is updated before mce_callin.
*/
- order = arch_atomic_inc_return(&mce_callin);
+ order = raw_atomic_inc_return(&mce_callin);
arch_cpumask_clear_cpu(smp_processor_id(), &mce_missing_cpus);
/* Enable instrumentation around calls to external facilities */
/*
* Wait for everyone.
*/
- while (arch_atomic_read(&mce_callin) != num_online_cpus()) {
+ while (raw_atomic_read(&mce_callin) != num_online_cpus()) {
if (mce_timed_out(&timeout,
"Timeout: Not all CPUs entered broadcast exception handler")) {
- arch_atomic_set(&global_nwo, 0);
+ raw_atomic_set(&global_nwo, 0);
goto out;
}
ndelay(SPINUNIT);
/*
* Monarch: Starts executing now, the others wait.
*/
- arch_atomic_set(&mce_executing, 1);
+ raw_atomic_set(&mce_executing, 1);
} else {
/*
* Subject: Now start the scanning loop one by one in
* This way when there are any shared banks it will be
* only seen by one CPU before cleared, avoiding duplicates.
*/
- while (arch_atomic_read(&mce_executing) < order) {
+ while (raw_atomic_read(&mce_executing) < order) {
if (mce_timed_out(&timeout,
"Timeout: Subject CPUs unable to finish machine check processing")) {
- arch_atomic_set(&global_nwo, 0);
+ raw_atomic_set(&global_nwo, 0);
goto out;
}
ndelay(SPINUNIT);
/*
* Cache the global no_way_out state.
*/
- *no_way_out = arch_atomic_read(&global_nwo);
+ *no_way_out = raw_atomic_read(&global_nwo);
ret = order;
/* If this triggers there is no way to recover. Die hard. */
BUG_ON(!on_thread_stack() || !user_mode(regs));
- if (kill_current_task)
+ if (!mce_usable_address(&m))
queue_task_work(&m, msg, kill_me_now);
else
queue_task_work(&m, msg, kill_me_maybe);
#endif
MSR_IA32_TSC, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA,
MSR_IA32_FEAT_CTL, MSR_IA32_BNDCFGS, MSR_TSC_AUX,
- MSR_IA32_SPEC_CTRL,
+ MSR_IA32_SPEC_CTRL, MSR_IA32_TSX_CTRL,
MSR_IA32_RTIT_CTL, MSR_IA32_RTIT_STATUS, MSR_IA32_RTIT_CR3_MATCH,
MSR_IA32_RTIT_OUTPUT_BASE, MSR_IA32_RTIT_OUTPUT_MASK,
MSR_IA32_RTIT_ADDR0_A, MSR_IA32_RTIT_ADDR0_B,
static inline u64 vgettsc(struct pvclock_clock *clock, u64 *tsc_timestamp,
int *mode)
{
- long v;
u64 tsc_pg_val;
+ long v;
switch (clock->vclock_mode) {
case VDSO_CLOCKMODE_HVCLOCK:
- tsc_pg_val = hv_read_tsc_page_tsc(hv_get_tsc_page(),
- tsc_timestamp);
- if (tsc_pg_val != U64_MAX) {
+ if (hv_read_tsc_page_tsc(hv_get_tsc_page(),
+ tsc_timestamp, &tsc_pg_val)) {
/* TSC page valid */
*mode = VDSO_CLOCKMODE_HVCLOCK;
v = (tsc_pg_val - clock->cycle_last) &
if (!kvm_cpu_cap_has(X86_FEATURE_XFD))
return;
break;
+ case MSR_IA32_TSX_CTRL:
+ if (!(kvm_get_arch_capabilities() & ARCH_CAP_TSX_CTRL_MSR))
+ return;
+ break;
default:
break;
}
exit_fastpath = EXIT_FASTPATH_EXIT_HANDLED;
break;
}
+
+ /* Note, VM-Exits that go down the "slow" path are accounted below. */
+ ++vcpu->stat.exits;
}
/*
bool noinstr kvm_arch_has_assigned_device(struct kvm *kvm)
{
- return arch_atomic_read(&kvm->arch.assigned_device_count);
+ return raw_atomic_read(&kvm->arch.assigned_device_count);
}
EXPORT_SYMBOL_GPL(kvm_arch_has_assigned_device);
(MMIO_STATUS_EVT_OVERFLOW_INT_MASK | \
MMIO_STATUS_EVT_INT_MASK | \
MMIO_STATUS_PPR_INT_MASK | \
+ MMIO_STATUS_GALOG_OVERFLOW_MASK | \
MMIO_STATUS_GALOG_INT_MASK)
irqreturn_t amd_iommu_int_thread(int irq, void *data)
}
#ifdef CONFIG_IRQ_REMAP
- if (status & MMIO_STATUS_GALOG_INT_MASK) {
+ if (status & (MMIO_STATUS_GALOG_INT_MASK |
+ MMIO_STATUS_GALOG_OVERFLOW_MASK)) {
pr_devel("Processing IOMMU GA Log\n");
iommu_poll_ga_log(iommu);
}
+
+ if (status & MMIO_STATUS_GALOG_OVERFLOW_MASK) {
+ pr_info_ratelimited("IOMMU GA Log overflow\n");
+ amd_iommu_restart_ga_log(iommu);
+ }
#endif
if (status & MMIO_STATUS_EVT_OVERFLOW_INT_MASK) {
{
struct io_pgtable_ops *pgtbl_ops;
struct protection_domain *domain;
- int pgtable = amd_iommu_pgtable;
+ int pgtable;
int mode = DEFAULT_PGTABLE_LEVEL;
int ret;
- domain = kzalloc(sizeof(*domain), GFP_KERNEL);
- if (!domain)
- return NULL;
-
/*
* Force IOMMU v1 page table when iommu=pt and
* when allocating domain for pass-through devices.
mode = PAGE_MODE_NONE;
} else if (type == IOMMU_DOMAIN_UNMANAGED) {
pgtable = AMD_IOMMU_V1;
+ } else if (type == IOMMU_DOMAIN_DMA || type == IOMMU_DOMAIN_DMA_FQ) {
+ pgtable = amd_iommu_pgtable;
+ } else {
+ return NULL;
}
+ domain = kzalloc(sizeof(*domain), GFP_KERNEL);
+ if (!domain)
+ return NULL;
+
switch (pgtable) {
case AMD_IOMMU_V1:
ret = protection_domain_init_v1(domain, mode);
return NULL;
}
+static inline u64 dma_max_address(void)
+{
+ if (amd_iommu_pgtable == AMD_IOMMU_V1)
+ return ~0ULL;
+
+ /* V2 with 4/5 level page table */
+ return ((1ULL << PM_LEVEL_SHIFT(amd_iommu_gpt_level)) - 1);
+}
+
static struct iommu_domain *amd_iommu_domain_alloc(unsigned type)
{
struct protection_domain *domain;
return NULL;
domain->domain.geometry.aperture_start = 0;
- domain->domain.geometry.aperture_end = ~0ULL;
+ domain->domain.geometry.aperture_end = dma_max_address();
domain->domain.geometry.force_aperture = true;
return &domain->domain;
unsigned long flags;
spin_lock_irqsave(&dom->lock, flags);
- domain_flush_pages(dom, gather->start, gather->end - gather->start, 1);
+ domain_flush_pages(dom, gather->start, gather->end - gather->start + 1, 1);
amd_iommu_domain_flush_complete(dom);
spin_unlock_irqrestore(&dom->lock, flags);
}
static int modify_irte_ga(struct amd_iommu *iommu, u16 devid, int index,
struct irte_ga *irte, struct amd_ir_data *data)
{
- bool ret;
struct irq_remap_table *table;
- unsigned long flags;
struct irte_ga *entry;
+ unsigned long flags;
+ u128 old;
table = get_irq_table(iommu, devid);
if (!table)
entry = (struct irte_ga *)table->table;
entry = &entry[index];
- ret = cmpxchg_double(&entry->lo.val, &entry->hi.val,
- entry->lo.val, entry->hi.val,
- irte->lo.val, irte->hi.val);
/*
* We use cmpxchg16 to atomically update the 128-bit IRTE,
* and it cannot be updated by the hardware or other processors
* behind us, so the return value of cmpxchg16 should be the
* same as the old value.
*/
- WARN_ON(!ret);
+ old = entry->irte;
+ WARN_ON(!try_cmpxchg128(&entry->irte, &old, irte->irte));
if (data)
data->ref = entry;
struct irte_ga *entry = (struct irte_ga *) ir_data->entry;
u64 valid;
- if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) ||
- !entry || entry->lo.fields_vapic.guest_mode)
+ if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) || !entry)
return 0;
valid = entry->lo.fields_vapic.valid;
}
}
+ #define cmp_int(l, r) ((l > r) - (l < r))
+
+ #ifdef CONFIG_PROVE_LOCKING
+ static int btree_lock_cmp_fn(const struct lockdep_map *_a,
+ const struct lockdep_map *_b)
+ {
+ const struct btree *a = container_of(_a, struct btree, lock.dep_map);
+ const struct btree *b = container_of(_b, struct btree, lock.dep_map);
+
+ return -cmp_int(a->level, b->level) ?: bkey_cmp(&a->key, &b->key);
+ }
+
+ static void btree_lock_print_fn(const struct lockdep_map *map)
+ {
+ const struct btree *b = container_of(map, struct btree, lock.dep_map);
+
+ printk(KERN_CONT " l=%u %llu:%llu", b->level,
+ KEY_INODE(&b->key), KEY_OFFSET(&b->key));
+ }
+ #endif
+
static struct btree *mca_bucket_alloc(struct cache_set *c,
struct bkey *k, gfp_t gfp)
{
return NULL;
init_rwsem(&b->lock);
- lockdep_set_novalidate_class(&b->lock);
+ lock_set_cmp_fn(&b->lock, btree_lock_cmp_fn, btree_lock_print_fn);
mutex_init(&b->write_lock);
lockdep_set_novalidate_class(&b->write_lock);
INIT_LIST_HEAD(&b->list);
* cannibalize_bucket() will take. This means every time we unlock the root of
* the btree, we need to release this lock if we have it held.
*/
-static void bch_cannibalize_unlock(struct cache_set *c)
+void bch_cannibalize_unlock(struct cache_set *c)
{
spin_lock(&c->btree_cannibalize_lock);
if (c->btree_cache_alloc_lock == current) {
struct btree *parent)
{
BKEY_PADDED(key) k;
- struct btree *b = ERR_PTR(-EAGAIN);
+ struct btree *b;
mutex_lock(&c->bucket_lock);
retry:
+ /* return ERR_PTR(-EAGAIN) when it fails */
+ b = ERR_PTR(-EAGAIN);
if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, wait))
goto err;
{
struct btree *n = bch_btree_node_alloc(b->c, op, b->level, b->parent);
- if (!IS_ERR_OR_NULL(n)) {
+ if (!IS_ERR(n)) {
mutex_lock(&n->write_lock);
bch_btree_sort_into(&b->keys, &n->keys, &b->c->sort);
bkey_copy_key(&n->key, &b->key);
memset(new_nodes, 0, sizeof(new_nodes));
closure_init_stack(&cl);
- while (nodes < GC_MERGE_NODES && !IS_ERR_OR_NULL(r[nodes].b))
+ while (nodes < GC_MERGE_NODES && !IS_ERR(r[nodes].b))
keys += r[nodes++].keys;
blocks = btree_default_blocks(b->c) * 2 / 3;
for (i = 0; i < nodes; i++) {
new_nodes[i] = btree_node_alloc_replacement(r[i].b, NULL);
- if (IS_ERR_OR_NULL(new_nodes[i]))
+ if (IS_ERR(new_nodes[i]))
goto out_nocoalesce;
}
bch_keylist_free(&keylist);
for (i = 0; i < nodes; i++)
- if (!IS_ERR_OR_NULL(new_nodes[i])) {
+ if (!IS_ERR(new_nodes[i])) {
btree_node_free(new_nodes[i]);
rw_unlock(true, new_nodes[i]);
}
if (should_rewrite) {
n = btree_node_alloc_replacement(b, NULL);
- if (!IS_ERR_OR_NULL(n)) {
+ if (!IS_ERR(n)) {
bch_btree_node_write_sync(n);
bch_btree_set_root(n);
c->gc_stats.nodes++;
bch_btree_op_init(&op, 0);
ret = bcache_btree(check_recurse, p, c->root, &op);
+ /*
+ * The op may be added to cache_set's btree_cache_wait
+ * in mca_cannibalize(), must ensure it is removed from
+ * the list and release btree_cache_alloc_lock before
+ * free op memory.
+ * Otherwise, the btree_cache_wait will be damaged.
+ */
+ bch_cannibalize_unlock(c);
+ finish_wait(&c->btree_cache_wait, &(&op)->wait);
if (ret)
goto out;
}
static inline void rw_lock(bool w, struct btree *b, int level)
{
- w ? down_write_nested(&b->lock, level + 1)
- : down_read_nested(&b->lock, level + 1);
+ w ? down_write(&b->lock)
+ : down_read(&b->lock);
if (w)
b->seq++;
}
void bch_moving_gc(struct cache_set *c);
int bch_btree_check(struct cache_set *c);
void bch_initial_mark_key(struct cache_set *c, int level, struct bkey *k);
+void bch_cannibalize_unlock(struct cache_set *c);
static inline void wake_up_gc(struct cache_set *c)
{
#define lockdep_repin_lock(l,c) lock_repin_lock(&(l)->dep_map, (c))
#define lockdep_unpin_lock(l,c) lock_unpin_lock(&(l)->dep_map, (c))
+/*
+ * Must use lock_map_aquire_try() with override maps to avoid
+ * lockdep thinking they participate in the block chain.
+ */
+#define DEFINE_WAIT_OVERRIDE_MAP(_name, _wait_type) \
+ struct lockdep_map _name = { \
+ .name = #_name "-wait-type-override", \
+ .wait_type_inner = _wait_type, \
+ .lock_type = LD_LOCK_WAIT_OVERRIDE, }
+
#else /* !CONFIG_LOCKDEP */
static inline void lockdep_init_task(struct task_struct *task)
#define lockdep_repin_lock(l, c) do { (void)(l); (void)(c); } while (0)
#define lockdep_unpin_lock(l, c) do { (void)(l); (void)(c); } while (0)
+#define DEFINE_WAIT_OVERRIDE_MAP(_name, _wait_type) \
+ struct lockdep_map __maybe_unused _name = {}
+
#endif /* !LOCKDEP */
+ #ifdef CONFIG_PROVE_LOCKING
+ void lockdep_set_lock_cmp_fn(struct lockdep_map *, lock_cmp_fn, lock_print_fn);
+
+ #define lock_set_cmp_fn(lock, ...) lockdep_set_lock_cmp_fn(&(lock)->dep_map, __VA_ARGS__)
+ #else
+ #define lock_set_cmp_fn(lock, ...) do { } while (0)
+ #endif
+
enum xhlock_context_t {
XHLOCK_HARD,
XHLOCK_SOFT,
#define rwsem_release(l, i) lock_release(l, i)
#define lock_map_acquire(l) lock_acquire_exclusive(l, 0, 0, NULL, _THIS_IP_)
+#define lock_map_acquire_try(l) lock_acquire_exclusive(l, 0, 1, NULL, _THIS_IP_)
#define lock_map_acquire_read(l) lock_acquire_shared_recursive(l, 0, 0, NULL, _THIS_IP_)
#define lock_map_acquire_tryread(l) lock_acquire_shared_recursive(l, 0, 1, NULL, _THIS_IP_)
#define lock_map_release(l) lock_release(l, _THIS_IP_)
enum lockdep_lock_type {
LD_LOCK_NORMAL = 0, /* normal, catch all */
LD_LOCK_PERCPU, /* percpu */
+ LD_LOCK_WAIT_OVERRIDE, /* annotation */
LD_LOCK_MAX,
};
#define LOCKSTAT_POINTS 4
+ struct lockdep_map;
+ typedef int (*lock_cmp_fn)(const struct lockdep_map *a,
+ const struct lockdep_map *b);
+ typedef void (*lock_print_fn)(const struct lockdep_map *map);
+
/*
* The lock-class itself. The order of the structure members matters.
* reinit_class() zeroes the key member and all subsequent members.
struct list_head locks_after, locks_before;
const struct lockdep_subclass_key *key;
+ lock_cmp_fn cmp_fn;
+ lock_print_fn print_fn;
+
unsigned int subclass;
unsigned int dep_gen_id;
usage[i] = '\0';
}
- static void __print_lock_name(struct lock_class *class)
+ static void __print_lock_name(struct held_lock *hlock, struct lock_class *class)
{
char str[KSYM_NAME_LEN];
const char *name;
printk(KERN_CONT "#%d", class->name_version);
if (class->subclass)
printk(KERN_CONT "/%d", class->subclass);
+ if (hlock && class->print_fn)
+ class->print_fn(hlock->instance);
}
}
- static void print_lock_name(struct lock_class *class)
+ static void print_lock_name(struct held_lock *hlock, struct lock_class *class)
{
char usage[LOCK_USAGE_CHARS];
get_usage_chars(class, usage);
printk(KERN_CONT " (");
- __print_lock_name(class);
+ __print_lock_name(hlock, class);
printk(KERN_CONT "){%s}-{%d:%d}", usage,
class->wait_type_outer ?: class->wait_type_inner,
class->wait_type_inner);
}
printk(KERN_CONT "%px", hlock->instance);
- print_lock_name(lock);
+ print_lock_name(hlock, lock);
printk(KERN_CONT ", at: %pS\n", (void *)hlock->acquire_ip);
}
if (debug_locks_silent)
return;
printk("\n-> #%u", depth);
- print_lock_name(target->class);
+ print_lock_name(NULL, target->class);
printk(KERN_CONT ":\n");
print_lock_trace(target->trace, 6);
}
*/
if (parent != source) {
printk("Chain exists of:\n ");
- __print_lock_name(source);
+ __print_lock_name(src, source);
printk(KERN_CONT " --> ");
- __print_lock_name(parent);
+ __print_lock_name(NULL, parent);
printk(KERN_CONT " --> ");
- __print_lock_name(target);
+ __print_lock_name(tgt, target);
printk(KERN_CONT "\n\n");
}
printk(" rlock(");
else
printk(" lock(");
- __print_lock_name(target);
+ __print_lock_name(tgt, target);
printk(KERN_CONT ");\n");
printk(" lock(");
- __print_lock_name(parent);
+ __print_lock_name(NULL, parent);
printk(KERN_CONT ");\n");
printk(" lock(");
- __print_lock_name(target);
+ __print_lock_name(tgt, target);
printk(KERN_CONT ");\n");
if (src_read != 0)
printk(" rlock(");
printk(" sync(");
else
printk(" lock(");
- __print_lock_name(source);
+ __print_lock_name(src, source);
printk(KERN_CONT ");\n");
printk("\n *** DEADLOCK ***\n\n");
}
return ret;
}
+ static void print_deadlock_bug(struct task_struct *, struct held_lock *, struct held_lock *);
+
/*
* Prove that the dependency graph starting at <src> can not
* lead to <target>. If it can, there is a circle when adding
*trace = save_trace();
}
- print_circular_bug(&src_entry, target_entry, src, target);
+ if (src->class_idx == target->class_idx)
+ print_deadlock_bug(current, src, target);
+ else
+ print_circular_bug(&src_entry, target_entry, src, target);
}
return ret;
static inline bool usage_skip(struct lock_list *entry, void *mask)
{
+ if (entry->class->lock_type == LD_LOCK_NORMAL)
+ return false;
+
/*
* Skip local_lock() for irq inversion detection.
*
* As a result, we will skip local_lock(), when we search for irq
* inversion bugs.
*/
- if (entry->class->lock_type == LD_LOCK_PERCPU) {
- if (DEBUG_LOCKS_WARN_ON(entry->class->wait_type_inner < LD_WAIT_CONFIG))
- return false;
+ if (entry->class->lock_type == LD_LOCK_PERCPU &&
+ DEBUG_LOCKS_WARN_ON(entry->class->wait_type_inner < LD_WAIT_CONFIG))
+ return false;
- return true;
- }
+ /*
+ * Skip WAIT_OVERRIDE for irq inversion detection -- it's not actually
+ * a lock and only used to override the wait_type.
+ */
- return false;
+ return true;
}
/*
int bit;
printk("%*s->", depth, "");
- print_lock_name(class);
+ print_lock_name(NULL, class);
#ifdef CONFIG_DEBUG_LOCKDEP
printk(KERN_CONT " ops: %lu", debug_class_ops_read(class));
#endif
*/
if (middle_class != unsafe_class) {
printk("Chain exists of:\n ");
- __print_lock_name(safe_class);
+ __print_lock_name(NULL, safe_class);
printk(KERN_CONT " --> ");
- __print_lock_name(middle_class);
+ __print_lock_name(NULL, middle_class);
printk(KERN_CONT " --> ");
- __print_lock_name(unsafe_class);
+ __print_lock_name(NULL, unsafe_class);
printk(KERN_CONT "\n\n");
}
printk(" CPU0 CPU1\n");
printk(" ---- ----\n");
printk(" lock(");
- __print_lock_name(unsafe_class);
+ __print_lock_name(NULL, unsafe_class);
printk(KERN_CONT ");\n");
printk(" local_irq_disable();\n");
printk(" lock(");
- __print_lock_name(safe_class);
+ __print_lock_name(NULL, safe_class);
printk(KERN_CONT ");\n");
printk(" lock(");
- __print_lock_name(middle_class);
+ __print_lock_name(NULL, middle_class);
printk(KERN_CONT ");\n");
printk(" <Interrupt>\n");
printk(" lock(");
- __print_lock_name(safe_class);
+ __print_lock_name(NULL, safe_class);
printk(KERN_CONT ");\n");
printk("\n *** DEADLOCK ***\n\n");
}
pr_warn("\nand this task is already holding:\n");
print_lock(prev);
pr_warn("which would create a new lock dependency:\n");
- print_lock_name(hlock_class(prev));
+ print_lock_name(prev, hlock_class(prev));
pr_cont(" ->");
- print_lock_name(hlock_class(next));
+ print_lock_name(next, hlock_class(next));
pr_cont("\n");
pr_warn("\nbut this new dependency connects a %s-irq-safe lock:\n",
irqclass);
- print_lock_name(backwards_entry->class);
+ print_lock_name(NULL, backwards_entry->class);
pr_warn("\n... which became %s-irq-safe at:\n", irqclass);
print_lock_trace(backwards_entry->class->usage_traces[bit1], 1);
pr_warn("\nto a %s-irq-unsafe lock:\n", irqclass);
- print_lock_name(forwards_entry->class);
+ print_lock_name(NULL, forwards_entry->class);
pr_warn("\n... which became %s-irq-unsafe at:\n", irqclass);
pr_warn("...");
printk(" CPU0\n");
printk(" ----\n");
printk(" lock(");
- __print_lock_name(prev);
+ __print_lock_name(prv, prev);
printk(KERN_CONT ");\n");
printk(" lock(");
- __print_lock_name(next);
+ __print_lock_name(nxt, next);
printk(KERN_CONT ");\n");
printk("\n *** DEADLOCK ***\n\n");
printk(" May be due to missing lock nesting notation\n\n");
print_deadlock_bug(struct task_struct *curr, struct held_lock *prev,
struct held_lock *next)
{
+ struct lock_class *class = hlock_class(prev);
+
if (!debug_locks_off_graph_unlock() || debug_locks_silent)
return;
pr_warn("\nbut task is already holding lock:\n");
print_lock(prev);
+ if (class->cmp_fn) {
+ pr_warn("and the lock comparison function returns %i:\n",
+ class->cmp_fn(prev->instance, next->instance));
+ }
+
pr_warn("\nother info that might help us debug this:\n");
print_deadlock_scenario(next, prev);
lockdep_print_held_locks(curr);
static int
check_deadlock(struct task_struct *curr, struct held_lock *next)
{
+ struct lock_class *class;
struct held_lock *prev;
struct held_lock *nest = NULL;
int i;
if ((next->read == 2) && prev->read)
continue;
+ class = hlock_class(prev);
+
+ if (class->cmp_fn &&
+ class->cmp_fn(prev->instance, next->instance) < 0)
+ continue;
+
/*
* We're holding the nest_lock, which serializes this lock's
* nesting behaviour.
return 2;
}
+ if (prev->class_idx == next->class_idx) {
+ struct lock_class *class = hlock_class(prev);
+
+ if (class->cmp_fn &&
+ class->cmp_fn(prev->instance, next->instance) < 0)
+ return 2;
+ }
+
/*
* Prove that the new <prev> -> <next> dependency would not
* create a circular dependency in the graph. (We do this by
hlock_id = chain_hlocks[chain->base + i];
chain_key = print_chain_key_iteration(hlock_id, chain_key);
- print_lock_name(lock_classes + chain_hlock_class_idx(hlock_id));
+ print_lock_name(NULL, lock_classes + chain_hlock_class_idx(hlock_id));
printk("\n");
}
}
printk(" CPU0\n");
printk(" ----\n");
printk(" lock(");
- __print_lock_name(class);
+ __print_lock_name(lock, class);
printk(KERN_CONT ");\n");
printk(" <Interrupt>\n");
printk(" lock(");
- __print_lock_name(class);
+ __print_lock_name(lock, class);
printk(KERN_CONT ");\n");
printk("\n *** DEADLOCK ***\n\n");
}
pr_warn("but this lock took another, %s-unsafe lock in the past:\n", irqclass);
else
pr_warn("but this lock was taken by another, %s-safe lock in the past:\n", irqclass);
- print_lock_name(other->class);
+ print_lock_name(NULL, other->class);
pr_warn("\n\nand interrupts could create inverse lock ordering between them.\n\n");
pr_warn("\nother info that might help us debug this:\n");
for (; depth < curr->lockdep_depth; depth++) {
struct held_lock *prev = curr->held_locks + depth;
- u8 prev_inner = hlock_class(prev)->wait_type_inner;
+ struct lock_class *class = hlock_class(prev);
+ u8 prev_inner = class->wait_type_inner;
if (prev_inner) {
/*
* Also due to trylocks.
*/
curr_inner = min(curr_inner, prev_inner);
+
+ /*
+ * Allow override for annotations -- this is typically
+ * only valid/needed for code that only exists when
+ * CONFIG_PREEMPT_RT=n.
+ */
+ if (unlikely(class->lock_type == LD_LOCK_WAIT_OVERRIDE))
+ curr_inner = prev_inner;
}
}
struct lock_class_key __lockdep_no_validate__;
EXPORT_SYMBOL_GPL(__lockdep_no_validate__);
+ #ifdef CONFIG_PROVE_LOCKING
+ void lockdep_set_lock_cmp_fn(struct lockdep_map *lock, lock_cmp_fn cmp_fn,
+ lock_print_fn print_fn)
+ {
+ struct lock_class *class = lock->class_cache[0];
+ unsigned long flags;
+
+ raw_local_irq_save(flags);
+ lockdep_recursion_inc();
+
+ if (!class)
+ class = register_lock_class(lock, 0, 0);
+
+ if (class) {
+ WARN_ON(class->cmp_fn && class->cmp_fn != cmp_fn);
+ WARN_ON(class->print_fn && class->print_fn != print_fn);
+
+ class->cmp_fn = cmp_fn;
+ class->print_fn = print_fn;
+ }
+
+ lockdep_recursion_finish();
+ raw_local_irq_restore(flags);
+ }
+ EXPORT_SYMBOL_GPL(lockdep_set_lock_cmp_fn);
+ #endif
+
static void
print_lock_nested_lock_not_held(struct task_struct *curr,
struct held_lock *hlock)
s64 delta;
again:
- now = sched_clock();
+ now = sched_clock_noinstr();
delta = now - scd->tick_raw;
if (unlikely(delta < 0))
delta = 0;
clock = wrap_max(clock, min_clock);
clock = wrap_min(clock, max_clock);
- if (!arch_try_cmpxchg64(&scd->clock, &old_clock, clock))
+ if (!raw_try_cmpxchg64(&scd->clock, &old_clock, clock))
goto again;
return clock;
}
-noinstr u64 local_clock(void)
+noinstr u64 local_clock_noinstr(void)
{
u64 clock;
if (static_branch_likely(&__sched_clock_stable))
- return sched_clock() + __sched_clock_offset;
+ return sched_clock_noinstr() + __sched_clock_offset;
if (!static_branch_likely(&sched_clock_running))
- return sched_clock();
+ return sched_clock_noinstr();
- preempt_disable_notrace();
clock = sched_clock_local(this_scd());
- preempt_enable_notrace();
return clock;
}
+
+u64 local_clock(void)
+{
+ u64 now;
+ preempt_disable_notrace();
+ now = local_clock_noinstr();
+ preempt_enable_notrace();
+ return now;
+}
EXPORT_SYMBOL_GPL(local_clock);
static notrace u64 sched_clock_remote(struct sched_clock_data *scd)
echo 2.25.0
;;
gcc)
- echo 5.1.0
+ if [ "$SRCARCH" = parisc ]; then
+ echo 11.0.0
+ else
+ echo 5.1.0
+ fi
;;
llvm)
if [ "$SRCARCH" = s390 ]; then
fi
;;
rustc)
- echo 1.62.0
+ echo 1.68.2
;;
bindgen)
echo 0.56.0