perf annotate-data: Update data-type profiling quality (v2)
Namhyung Kim
maintainercover perf annotate-data: Update data-type profiling quality (v2)
Hello,
This is continuation of data type profiling series to improve its quality. I've also found some more bugs and room for improvements.
Acutally it depends on the a couple of fixes I sent out yesterday. But as they will be merged to perf-tools-next, I omitted them here.
There's no dramatic improvements at this moment, but I handled some corner cases which makes instruction tracking more flexible. One change is when it accesses (stores) a stack slot for the first time. As it doesn't save anything there yet, it doesn't come with a type info. In that case it can try to process the instruction with update_insn_state() and get the type information again.
Another update is when compiler generates the debug info inaccurately. A variable information is there but in some cases it was using another register to access the data and it resulted in no type information. So I tracked the (direct) register move and update the type info for the source registre as well.
The code is available at 'perf/data-profile-update-v2' branch in
git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
Thanks,
Namhyung
Namhyung Kim (4):
perf dwarf-aux: Handle bitfield members from pointer access perf annotate-data: Update debug messages
perf annotate-data: Update stack slot for the store perf annotate-data: Copy back variable types after move
tools/perf/arch/x86/annotate/instructions.c | 8 ++ tools/perf/util/annotate-data.c | 105 +++++++++++++++++--- tools/perf/util/annotate-data.h | 1 + tools/perf/util/dwarf-aux.c | 11 +- 4 files changed, 109 insertions(+), 16 deletions(-)
Signature · 1 line
2.46.0.184.g6999bdac58-goog
Namhyung Kim
maintainerPATCH 1/4 perf dwarf-aux: Handle bitfield members from pointer access
The __die_find_member_offset_cb() missed to handle bitfield members which don't have DW_AT_data_member_location. Like in adding member types in __add_member_cb() it should fallback to check the bit offset when it resolves the member type for an offset.
--- tools/perf/util/dwarf-aux.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
tools/perf/util/dwarf-aux.c +9 -2
@@ -1977,8 +1977,15 @@ static int __die_find_member_offset_cb(Dwarf_Die *die_mem, void *arg)return DIE_FIND_CB_SIBLING ;/* Unions might not have location */ -if (die_get_data_member_location (die_mem, &loc) <0 ) - loc =0 ; +if (die_get_data_member_location (die_mem, &loc) <0 ) { +Dwarf_Attribute attr; + +if (dwarf_attr_integrate (die_mem, DW_AT_data_bit_offset, &attr) && +dwarf_formudata (&attr, &loc) ==0 ) + loc /=8 ; +else + loc =0 ; + }if (offset == loc)return DIE_FIND_CB_END ;
Signature · 1 line
2.46.0.184.g6999bdac58-goog
Namhyung Kim
maintainerPATCH 2/4 perf annotate-data: Update debug messages
In reply to Namhyung KimIn check_matching_type(), it'd be easier to display the typename in question if it's available.
For example, check out the line starts with 'chk'.
-----------------------------------------------------------
find data type for 0x10(reg0) at cpuacct_charge+0x13
CU for kernel/sched/build_utility.c (die:0x137ee0b)
frame base: cfa=1 fbreg=7
scope: [3/3] (die:13d9632)
bb: [c - 13]
var [c] reg5 type='struct task_struct*' size=0x8 (die:0x1381230) mov [c] 0xdf8(reg5) -> reg0 type='struct css_set*' size=0x8 (die:0x1385c56)
chk [13] reg0 offset=0x10 ok=1 kind=1 (struct css_set*) : Good! <<<--- here
found by insn track: 0x10(reg0) type-offset=0x10
final result: type='struct css_set' size=0x250 (die:0x1385b0e)
Another example:
-----------------------------------------------------------
find data type for 0x8(reg0) at menu_select+0x279
CU for drivers/cpuidle/governors/menu.c (die:0x7b0fe79) frame base: cfa=1 fbreg=7
scope: [2/2] (die:7b11010)
bb: [273 - 277]
bb: [279 - 279]
chk [279] reg0 offset=0x8 ok=0 kind=0 cfa : no type information
scope: [1/2] (die:7b10cbc)
bb: [0 - 64]
...
mov [26a] imm=0xffffffff -> reg15
bb: [273 - 277]
bb: [279 - 279]
chk [279] reg0 offset=0x8 ok=1 kind=1 (long long unsigned int) : no/void pointer <<<--- here
final result: no/void pointer
Also change some places to print negative offsets properly.
Before:
-----------------------------------------------------------
find data type for 0xffffff40(reg6) at __tcp_transmit_skb+0x58
After:
-----------------------------------------------------------
find data type for -0xc0(reg6) at __tcp_transmit_skb+0x58
--- tools/perf/util/annotate-data.c | 45 +++++++++++++++++++++++++-------- 1 file changed, 35 insertions(+), 10 deletions(-)
tools/perf/util/annotate-data.c +35 -10
@@ -960,9 +960,16 @@ static enum type_match_result check_matching_type(struct type_state *state,Dwarf_Word size;u32 insn_offset = dloc->ip - dloc->ms->sym->start;int reg = dloc->op->reg1; +int offset = dloc->op->offset; +const char *offset_sign ="" ; -pr_debug_dtp ("chk [%x] reg%d offset=%#x ok=%d kind=%d " , - insn_offset, reg, dloc->op->offset, +if (offset <0 ) { + offset = -offset; + offset_sign ="-" ; + } + +pr_debug_dtp ("chk [%x] reg%d offset=%s%#x ok=%d kind=%d " , + insn_offset, reg, offset_sign, offset, state->regs[reg].ok, state->regs[reg].kind);if (!state->regs[reg].ok)
@@ -970,6 +977,12 @@ static enum type_match_result check_matching_type(struct type_state *state,if (state->regs[reg].kind ==TSR_KIND_TYPE ) {Dwarf_Die sized_type; +struct strbuf sb; + +strbuf_init (&sb,32 ); +die_get_typename_from_type (&state->regs[reg].type, &sb); +pr_debug_dtp ("(%s)" , sb.buf); +strbuf_release (&sb);/* * Normal registers should hold a pointer (or array) to
@@ -1119,7 +1132,6 @@ static enum type_match_result check_matching_type(struct type_state *state, check_kernel:if (dso__kernel (map__dso (dloc->ms->map))) {u64 addr; -int offset;/* Direct this-cpu access like "%gs:0x34740" */ if (dloc->op->segment == INSN_SEG_X86_GS && dloc->op->imm &&
@@ -1271,6 +1283,13 @@ static enum type_match_result find_data_type_block(struct data_loc_info *dloc, cu_die, type_die);if (ret ==PERF_TMR_OK ) {char buf[64 ]; +int offset = dloc->op->offset; +const char *offset_sign ="" ; + +if (offset <0 ) { + offset = -offset; + offset_sign ="-" ; + }if (dloc->op->multi_regs) snprintf(buf, sizeof(buf), "reg%d, reg%d",
@@ -1278,8 +1297,8 @@ static enum type_match_result find_data_type_block(struct data_loc_info *dloc,else snprintf (buf,sizeof (buf),"reg%d" , dloc->op->reg1); -pr_debug_dtp ("found by insn track: %#x(%s) type-offset=%#x\n" , - dloc->op->offset, buf, dloc->type_offset); +pr_debug_dtp ("found by insn track: %s%#x(%s) type-offset=%#x\n" , + offset_sign, offset, buf, dloc->type_offset);break ; }
@@ -1302,7 +1321,7 @@ static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die)struct annotated_op_loc *loc = dloc->op;Dwarf_Die cu_die, var_die;Dwarf_Die *scopes =NULL ; -int reg, offset; +int reg, offset = loc->offset;int ret =-1 ;int i, nr_scopes;int fbreg =-1 ;
@@ -1312,6 +1331,7 @@ static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die)u64 pc;char buf[64 ];enum type_match_result result =PERF_TMR_UNKNOWN ; +const char *offset_sign ="" ;if (dloc->op->multi_regs)snprintf (buf,sizeof (buf),"reg%d, reg%d" , dloc->op->reg1, dloc->op->reg2);
@@ -1320,10 +1340,15 @@ static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die)else snprintf (buf,sizeof (buf),"reg%d" , dloc->op->reg1); +if (offset <0 ) { + offset = -offset; + offset_sign ="-" ; + } +pr_debug_dtp ("-----------------------------------------------------------\n" ); -pr_debug_dtp ("find data type for %#x(%s) at %s+%#" PRIx64"\n" , - dloc->op->offset, buf, dloc->ms->sym->name, - dloc->ip - dloc->ms->sym->start); +pr_debug_dtp ("find data type for %s%#x(%s) at %s+%#" PRIx64"\n" , + offset_sign, offset, buf, + dloc->ms->sym->name, dloc->ip - dloc->ms->sym->start);/* * IP is a relative instruction address from the start of the map, as
@@ -1453,8 +1478,8 @@ static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die) } out: +pr_debug_dtp ("final result: " );if (found) { -pr_debug_dtp ("final type:" );pr_debug_type_name (type_die,TSR_KIND_TYPE ); ret =0 ; }else {
Signature · 1 line
2.46.0.184.g6999bdac58-goog
Namhyung Kim
maintainerPATCH 3/4 perf annotate-data: Update stack slot for the store
In reply to Namhyung KimWhen checking the match variable at the target instruction, it might not have any information if it's a first write to a stack slot. In this case it could spill a register value into the stack so the type info is in the source operand.
But currently it's hard to get the operand from the checking function. Let's process the instruction and retry to get the type info from the stack if there's no information already.
This is an example of __tcp_transmit_skb(). The instructions are
<__tcp_transmit_skb>:
0: nopl 0x0(%rax, %rax, 1) 5: push %rbp 6: mov %rsp, %rbp 9: push %r15 b: push %r14 d: push %r13 f: push %r12 11: push %rbx 12: sub $0x98, %rsp 19: mov %r8d, -0xa8(%rbp)
...
It cannot find any variable at -0xa8(%rbp) at this point.
-----------------------------------------------------------
find data type for -0xa8(reg6) at __tcp_transmit_skb+0x19
CU for net/ipv4/tcp_output.c (die:0x817f543)
frame base: cfa=0 fbreg=6
scope: [1/1] (die:81aac3e)
bb: [0 - 19]
var [0] -0x98(stack) type='struct tcp_out_options' size=0x28 (die:0x81af3df) var [5] reg8 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg2 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg1 type='int' size=0x4 (die:0x818059e)
var [5] reg4 type='struct sk_buff*' size=0x8 (die:0x8181360) var [5] reg5 type='struct sock*' size=0x8 (die:0x8181a0c)
chk [19] reg6 offset=-0xa8 ok=0 kind=0 fbreg : no type information no type information
And it was able to find the type after processing the 'mov' instruction.
-----------------------------------------------------------
find data type for -0xa8(reg6) at __tcp_transmit_skb+0x19
CU for net/ipv4/tcp_output.c (die:0x817f543)
frame base: cfa=0 fbreg=6
scope: [1/1] (die:81aac3e)
bb: [0 - 19]
var [0] -0x98(stack) type='struct tcp_out_options' size=0x28 (die:0x81af3df) var [5] reg8 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg2 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg1 type='int' size=0x4 (die:0x818059e)
var [5] reg4 type='struct sk_buff*' size=0x8 (die:0x8181360) var [5] reg5 type='struct sock*' size=0x8 (die:0x8181a0c)
chk [19] reg6 offset=-0xa8 ok=0 kind=0 fbreg : retry <<<--- here
mov [19] reg8 -> -0xa8(stack) type='unsigned int' size=0x4 (die:0x8180ed6) chk [19] reg6 offset=-0xa8 ok=0 kind=0 fbreg : Good!
found by insn track: -0xa8(reg6) type-offset=0
final result: type='unsigned int' size=0x4 (die:0x8180ed6)
--- tools/perf/util/annotate-data.c | 29 +++++++++++++++++++++++++---- 1 file changed, 25 insertions(+), 4 deletions(-)
tools/perf/util/annotate-data.c +25 -4
@@ -955,19 +955,22 @@ static void setup_stack_canary(struct data_loc_info *dloc)static enum type_match_result check_matching_type (struct type_state *state,struct data_loc_info *dloc,Dwarf_Die *cu_die, +struct disasm_line *dl,Dwarf_Die *type_die) {Dwarf_Word size; -u32 insn_offset = dloc->ip - dloc->ms->sym->start; +u32 insn_offset = dl->al.offset;int reg = dloc->op->reg1;int offset = dloc->op->offset;const char *offset_sign ="" ; +bool retry = true;if (offset <0 ) { offset = -offset; offset_sign ="-" ; } +again:pr_debug_dtp ("chk [%x] reg%d offset=%s%#x ok=%d kind=%d " , insn_offset, reg, offset_sign, offset, state->regs[reg].ok, state->regs[reg].kind);
@@ -1079,8 +1082,17 @@ static enum type_match_result check_matching_type(struct type_state *state,pr_debug_dtp ("fbreg" ); stack =find_stack_state (state, dloc->type_offset); -if (stack ==NULL ) +if (stack ==NULL ) { +if (retry) { +pr_debug_dtp (" : retry\n" ); + retry = false; + +/* update type info it's the first store to the stack */ +update_insn_state (state, dloc, cu_die, dl); + goto again; + }return PERF_TMR_NO_TYPE ; + }if (stack->kind ==TSR_KIND_CANARY ) {setup_stack_canary (dloc);
@@ -1111,8 +1123,17 @@ static enum type_match_result check_matching_type(struct type_state *state,return PERF_TMR_NO_TYPE ; stack =find_stack_state (state, dloc->type_offset - fboff); -if (stack ==NULL ) +if (stack ==NULL ) { +if (retry) { +pr_debug_dtp (" : retry\n" ); + retry = false; + +/* update type info it's the first store to the stack */ +update_insn_state (state, dloc, cu_die, dl); + goto again; + }return PERF_TMR_NO_TYPE ; + }if (stack->kind ==TSR_KIND_CANARY ) {setup_stack_canary (dloc);
@@ -1202,7 +1223,7 @@ static enum type_match_result find_data_type_insn(struct data_loc_info *dloc,if (this_ip == dloc->ip) { ret =check_matching_type (&state, dloc, - cu_die, type_die); + cu_die, dl, type_die);pr_debug_dtp (" : %s\n" ,match_result_str (ret)); goto out; }
Signature · 1 line
2.46.0.184.g6999bdac58-goog
Namhyung Kim
maintainerPATCH 4/4 perf annotate-data: Copy back variable types after move
In reply to Namhyung KimIn some cases, compilers don't set the location expression in DWARF precisely. For instance, it may assign a variable to a register after copying it from a different register. Then it should use the register for the new type but still uses the old register. This makes hard to track the type information properly.
This is an example I found in __tcp_transmit_skb(). The first argument (sk) of this function is a pointer to sock and there's a variable (tp) for tcp_sock.
static int __tcp_transmit_skb(struct sock *sk, struct sk_buff *skb,
int clone_it, gfp_t gfp_mask, u32 rcv_nxt)
{
...
struct tcp_sock *tp;
BUG_ON(!skb || !tcp_skb_pcount(skb)); tp = tcp_sk(sk); prior_wstamp = tp->tcp_wstamp_ns; tp->tcp_wstamp_ns = max(tp->tcp_wstamp_ns, tp->tcp_clock_cache); ...
So it basically calls tcp_sk(sk) to get the tcp_sock pointer from sk. But it turned out to be the same value because tcp_sock embeds sock as the first member. The sk is located in reg5 (RDI) and tp is in reg3 (RBX). The offset of tcp_wstamp_ns is 0x748 and tcp_clock_cache is 0x750. So you need to use RBX (reg3) to access the fields in the tcp_sock. But the code used RDI (reg5) as it has the same value.
$ pahole --hex -C tcp_sock vmlinux | grep -e 748 -e 750 u64 tcp_wstamp_ns; /* 0x748 0x8 */ u64 tcp_clock_cache; /* 0x750 0x8 */
And this is the disassembly of the part of the function.
<__tcp_transmit_skb>:
...
44: mov %rdi, %rbx 47: mov 0x748(%rdi), %rsi 4e: mov 0x750(%rdi), %rax 55: cmp %rax, %rsi
Because compiler put the debug info to RBX, it only knows RDI is a pointer to sock and accessing those two fields resulted in error due to offset being beyond the type size.
-----------------------------------------------------------
find data type for 0x748(reg5) at __tcp_transmit_skb+0x63
CU for net/ipv4/tcp_output.c (die:0x817f543)
frame base: cfa=0 fbreg=6
scope: [1/1] (die:81aac3e)
bb: [0 - 30]
var [0] -0x98(stack) type='struct tcp_out_options' size=0x28 (die:0x81af3df) var [5] reg8 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg2 type='unsigned int' size=0x4 (die:0x8180ed6)
var [5] reg1 type='int' size=0x4 (die:0x818059e)
var [5] reg4 type='struct sk_buff*' size=0x8 (die:0x8181360)
var [5] reg5 type='struct sock*' size=0x8 (die:0x8181a0c) <<<--- the first argument ('sk' at %RDI)
mov [19] reg8 -> -0xa8(stack) type='unsigned int' size=0x4 (die:0x8180ed6) mov [20] stack canary -> reg0
mov [29] reg0 -> -0x30(stack) stack canary
bb: [36 - 3e]
mov [36] reg4 -> reg15 type='struct sk_buff*' size=0x8 (die:0x8181360)
bb: [44 - 63]
mov [44] reg5 -> reg3 type='struct sock*' size=0x8 (die:0x8181a0c) <<<--- calling tcp_sk()
var [47] reg3 type='struct tcp_sock*' size=0x8 (die:0x819eead) <<<--- new variable ('tp' at %RBX)
var [4e] reg4 type='unsigned long long' size=0x8 (die:0x8180edd) mov [58] reg4 -> -0xc0(stack) type='unsigned long long' size=0x8 (die:0x8180edd)
chk [63] reg5 offset=0x748 ok=1 kind=1 (struct sock*) : offset bigger than size <<<--- access with old variable
final result: offset bigger than size
While it's a fault in the compiler, we could work around this issue by using the type of new variable when it's copied directly. So I've added copied_from field in the register state to track those direct register to register copies. After that new register gets a new type and the old register still has the same type, it'll update (copy it back) the type of the old register.
For example, if we can update type of reg5 at __tcp_transmit_skb+0x47, we can find the target type of the instruction at 0x63 like below:
-----------------------------------------------------------
find data type for 0x748(reg5) at __tcp_transmit_skb+0x63
...
bb: [44 - 63]
mov [44] reg5 -> reg3 type='struct sock*' size=0x8 (die:0x8181a0c) var [47] reg3 type='struct tcp_sock*' size=0x8 (die:0x819eead)
var [47] copyback reg5 type='struct tcp_sock*' size=0x8 (die:0x819eead) <<<--- here
mov [47] 0x748(reg5) -> reg4 type='unsigned long long' size=0x8 (die:0x8180edd) mov [4e] 0x750(reg5) -> reg0 type='unsigned long long' size=0x8 (die:0x8180edd) mov [58] reg4 -> -0xc0(stack) type='unsigned long long' size=0x8 (die:0x8180edd)
chk [63] reg5 offset=0x748 ok=1 kind=1 (struct tcp_sock*) : Good! <<<--- new type
found by insn track: 0x748(reg5) type-offset=0x748
final result: type='struct tcp_sock' size=0xa98 (die:0x819eeb2)
--- tools/perf/arch/x86/annotate/instructions.c | 8 ++++++ tools/perf/util/annotate-data.c | 31 +++++++++++++++++++++ tools/perf/util/annotate-data.h | 1 + 3 files changed, 40 insertions(+)
tools/perf/arch/x86/annotate/instructions.c +8 -0
@@ -267,6 +267,7 @@ static void update_insn_state_x86(struct type_state *state,return ; tsr = &state->regs[dst->reg1]; + tsr->copied_from =-1 ;if (src->imm) imm_value = src->offset;
@@ -326,6 +327,8 @@ static void update_insn_state_x86(struct type_state *state,return ; tsr = &state->regs[dst->reg1]; + tsr->copied_from =-1 ; +if (dso__kernel (map__dso (dloc->ms->map)) && src->segment ==INSN_SEG_X86_GS && src->imm) {u64 ip = dloc->ms->sym->start + dl->al.offset;
@@ -386,6 +389,10 @@ static void update_insn_state_x86(struct type_state *state, tsr->imm_value = state->regs[src->reg1].imm_value; tsr->ok = true; +/* To copy back the variable type later (hopefully) */ +if (tsr->kind ==TSR_KIND_TYPE ) + tsr->copied_from = src->reg1; +pr_debug_dtp ("mov [%x] reg%d -> reg%d" , insn_offset, src->reg1, dst->reg1);pr_debug_type_name (&tsr->type, tsr->kind);
@@ -398,6 +405,7 @@ static void update_insn_state_x86(struct type_state *state,return ; tsr = &state->regs[dst->reg1]; + tsr->copied_from =-1 ; retry:/* Check stack variables with offset */
tools/perf/util/annotate-data.c +31 -0
@@ -774,6 +774,11 @@ bool get_global_var_type(Dwarf_Die *cu_die, struct data_loc_info *dloc,return true; } +static bool die_is_same (Dwarf_Die *die_a,Dwarf_Die *die_b) +{ +return (die_a->cu == die_b->cu) && (die_a->addr == die_b->addr); +} +/** * update_var_state - Update type state using given variables * @state: type state table
@@ -825,6 +830,7 @@ static void update_var_state(struct type_state *state, struct data_loc_info *dlopr_debug_type_name (&mem_die,TSR_KIND_TYPE ); }else if (has_reg_type (state, var->reg) && var->offset ==0 ) {struct type_state_reg *reg; +Dwarf_Die orig_type; reg = &state->regs[var->reg];
@@ -832,6 +838,8 @@ static void update_var_state(struct type_state *state, struct data_loc_info *dlo !is_better_type(®->type, &mem_die))continue ; + orig_type = reg->type; + reg->type = mem_die; reg->kind =TSR_KIND_TYPE ; reg->ok = true;
@@ -839,6 +847,29 @@ static void update_var_state(struct type_state *state, struct data_loc_info *dlopr_debug_dtp ("var [%" PRIx64"] reg%d" , insn_offset, var->reg);pr_debug_type_name (&mem_die,TSR_KIND_TYPE ); + +/* +* If this register is directly copied from another and it gets a +* better type, also update the type of the source register. This +* is usually the case of container_of() macro with offset of 0. +*/ +if (has_reg_type (state, reg->copied_from)) { +struct type_state_reg *copy_reg; + + copy_reg = &state->regs[reg->copied_from]; + +/* TODO: check if type is compatible or embedded */ +if (!copy_reg->ok || (copy_reg->kind !=TSR_KIND_TYPE ) || + !die_is_same (©_reg->type, &orig_type) || + !is_better_type (©_reg->type, &mem_die)) +continue ; + + copy_reg->type = mem_die; + +pr_debug_dtp ("var [%" PRIx64"] copyback reg%d" , + insn_offset, reg->copied_from); +pr_debug_type_name (&mem_die,TSR_KIND_TYPE ); + } } } }
tools/perf/util/annotate-data.h +1 -0
@@ -176,6 +176,7 @@ struct type_state_reg {bool ok;bool caller_saved;u8 kind; +u8 copied_from; };/* Type information in a stack location, dynamically allocated */
Signature · 1 line
2.46.0.184.g6999bdac58-goog
Quoted 7 lines · Namhyung Kim
Hello,
This is continuation of data type profiling series to improve its quality. I've also found some more bugs and room for improvements.
Acutally it depends on the a couple of fixes I sent out yesterday. But as they will be merged to perf-tools-next, I omitted them here.
Thanks, applied to perf-tools-next,
- Arnaldo