From: Namhyung Kim <namhyung@kernel.org>
To: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Cc: acme@kernel.org, jolsa@kernel.org, adrian.hunter@intel.com,
irogers@google.com, segher@kernel.crashing.org,
christophe.leroy@csgroup.eu, linux-kernel@vger.kernel.org,
linux-perf-users@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
akanksha@linux.ibm.com, maddy@linux.ibm.com, kjain@linux.ibm.com,
disgoel@linux.vnet.ibm.com
Subject: Re: [PATCH V7 15/18] tools/perf: Add support to find global register variables using find_data_type_global_reg
Date: Wed, 17 Jul 2024 22:11:33 -0700 [thread overview]
Message-ID: <ZpikBQdVxVpy5ETw@google.com> (raw)
In-Reply-To: <20240713165529.59298-16-atrajeev@linux.vnet.ibm.com>
Hello,
On Sat, Jul 13, 2024 at 10:25:26PM +0530, Athira Rajeev wrote:
> There are cases where define a global register variable and associate it
> with a specified register. Example, in powerpc, two registers are
> defined to represent variable:
> 1. r13: represents local_paca
> register struct paca_struct *local_paca asm("r13");
>
> 2. r1: represents stack_pointer
> register void *__stack_pointer asm("r1");
>
> These regs are present in dwarf debug as DW_OP_reg as part of variables
> in the cu_die (compile unit). These are not present in die search done
> in the list of nested scopes since these are global register variables.
>
> Example for local_paca represented by r13:
>
> <<>>
> <1><18dc6b4>: Abbrev Number: 128 (DW_TAG_variable)
> <18dc6b6> DW_AT_name : (indirect string, offset: 0x3861): local_paca
> <18dc6ba> DW_AT_decl_file : 48
> <18dc6bb> DW_AT_decl_line : 36
> <18dc6bc> DW_AT_decl_column : 30
> <18dc6bd> DW_AT_type : <0x18dc6c3>
> <18dc6c1> DW_AT_external : 1
> <18dc6c1> DW_AT_location : 1 byte block: 5d (DW_OP_reg13 (r13))
>
> <1><18dc6c3>: Abbrev Number: 3 (DW_TAG_pointer_type)
> <18dc6c4> DW_AT_byte_size : 8
> <18dc6c4> DW_AT_type : <0x18dc353>
>
> Where DW_AT_type : <0x18dc6c3> further points to :
>
> <1><18dc6c3>: Abbrev Number: 3 (DW_TAG_pointer_type)
> <18dc6c4> DW_AT_byte_size : 8
> <18dc6c4> DW_AT_type : <0x18dc353>
>
> which belongs to:
>
> <1><18dc353>: Abbrev Number: 67 (DW_TAG_structure_type)
> <18dc354> DW_AT_name : (indirect string, offset: 0x56cd): paca_struct
> <18dc358> DW_AT_byte_size : 2944
> <18dc35a> DW_AT_alignment : 128
> <18dc35b> DW_AT_decl_file : 48
> <18dc35c> DW_AT_decl_line : 61
> <18dc35d> DW_AT_decl_column : 8
> <18dc35d> DW_AT_sibling : <0x18dc6b4>
> <<>>
>
> Similar is case with "r1".
>
> <<>>
> <1><18dd772>: Abbrev Number: 129 (DW_TAG_variable)
> <18dd774> DW_AT_name : (indirect string, offset: 0x11ba): current_stack_pointer
> <18dd778> DW_AT_decl_file : 51
> <18dd779> DW_AT_decl_line : 1468
> <18dd77b> DW_AT_decl_column : 24
> <18dd77c> DW_AT_type : <0x18da5cd>
> <18dd780> DW_AT_external : 1
> <18dd780> DW_AT_location : 1 byte block: 51 (DW_OP_reg1 (r1))
>
> where 18da5cd is:
>
> <1><18da5cd>: Abbrev Number: 47 (DW_TAG_base_type)
> <18da5ce> DW_AT_byte_size : 8
> <18da5cf> DW_AT_encoding : 7 (unsigned)
> <18da5d0> DW_AT_name : (indirect string, offset: 0x55c7): long unsigned int
> <<>>
>
> To identify data type for these two special cases, iterate over
> variables in the CU die (Compile Unit) and match it with the register.
> If the variable is a base type, ie die_get_real_type will return NULL
I'm not sure why die_get_real_type() returned NULL. The variable has
the type attribute and the function run the loop only if the type is
either const, restrict, volatile, shared or typedef. So I think it
should return the base_type DIE.
> here, set offset to zero. With the changes, data type for "paca_struct"
> and "long unsigned int" for r1 is identified.
>
> Snippet from ./perf report -s type,type_off
>
> 12.85% long unsigned int long unsigned int +0 (no field)
> 4.68% struct paca_struct struct paca_struct +2312 (__current)
> 4.57% struct paca_struct struct paca_struct +2354 (irq_soft_mask)
>
> Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
> ---
> tools/perf/util/annotate-data.c | 42 ++++++++++++++++++++++++++++
> tools/perf/util/annotate.c | 8 ++++++
> tools/perf/util/annotate.h | 1 +
> tools/perf/util/include/dwarf-regs.h | 1 +
> 4 files changed, 52 insertions(+)
>
> diff --git a/tools/perf/util/annotate-data.c b/tools/perf/util/annotate-data.c
> index 734acdd8c4b7..a5b4429ede57 100644
> --- a/tools/perf/util/annotate-data.c
> +++ b/tools/perf/util/annotate-data.c
> @@ -1170,6 +1170,42 @@ static int find_data_type_block(struct data_loc_info *dloc,
> return ret;
> }
>
> +/*
> + * Handle cases where define a global register variable and
> + * associate it with a specified register. These regs are
> + * present in dwarf debug as DW_OP_reg as part of variables
> + * in the cu_die (compile unit). Iterate over variables in the
> + * cu_die and match with reg to identify data type die.
> + */
> +static int find_data_type_global_reg(struct data_loc_info *dloc, int reg, Dwarf_Die *cu_die,
> + Dwarf_Die *type_die)
> +{
> + Dwarf_Die vr_die;
> + int ret = -1;
> + struct die_var_type *var_types, *vt = NULL;
> +
> + die_collect_vars(cu_die, &vt);
> + for (var_types = vt; var_types; var_types = var_types->next) {
> + if (var_types->reg != reg)
> + continue;
> + if (dwarf_offdie(dloc->di->dbg, var_types->die_off, &vr_die)) {
> + if (die_get_real_type(&vr_die, type_die) == NULL) {
Ok, I think I know the reason. You don't need to call die_get_real_type()
here as var_types already has the type, not the variable. Usually we
want a pointer type for a variable. For the final result, you want the
target type of the pointer though.
> + dloc->type_offset = 0;
> + dwarf_offdie(dloc->di->dbg, var_types->die_off, type_die);
> + }
> + pr_debug_type_name(type_die, TSR_KIND_TYPE);
> + ret = 0;
> + pr_debug_dtp("found by CU for %s (die:%#lx)\n",
I think it's better to say found by global register %d.
> + dwarf_diename(type_die), (long)dwarf_dieoffset(type_die));
> + break;
> + }
> + }
> +
> + delete_var_types(vt);
> +
> + return ret;
> +}
> +
> /* The result will be saved in @type_die */
> static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die)
> {
> @@ -1217,6 +1253,12 @@ static int find_data_type_die(struct data_loc_info *dloc, Dwarf_Die *type_die)
> pr_debug_dtp("CU for %s (die:%#lx)\n",
> dwarf_diename(&cu_die), (long)dwarf_dieoffset(&cu_die));
>
> + if (loc->reg_type == DWARF_REG_GLOBAL) {
> + ret = find_data_type_global_reg(dloc, reg, &cu_die, type_die);
> + if (!ret)
> + goto out;
> + }
> +
> if (reg == DWARF_REG_PC) {
> if (get_global_var_type(&cu_die, dloc, dloc->ip, dloc->var_addr,
> &offset, type_die)) {
> diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
> index ce99db291c5e..8db2f32700aa 100644
> --- a/tools/perf/util/annotate.c
> +++ b/tools/perf/util/annotate.c
> @@ -2425,6 +2425,14 @@ struct annotated_data_type *hist_entry__get_data_type(struct hist_entry *he)
> op_loc->reg1 = DWARF_REG_PC;
> }
>
> + /* Global reg variable 13 and 1
> + * assign to DWARF_REG_GLOBAL
> + */
> + if (arch__is(arch, "powerpc")) {
> + if ((op_loc->reg1 == 13) || (op_loc->reg1 == 1))
> + op_loc->reg_type = DWARF_REG_GLOBAL;
> + }
> +
> mem_type = find_data_type(&dloc);
>
> if (mem_type == NULL && is_stack_canary(arch, op_loc)) {
> diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h
> index 9ba772f46270..ad69842a8ebc 100644
> --- a/tools/perf/util/annotate.h
> +++ b/tools/perf/util/annotate.h
> @@ -475,6 +475,7 @@ struct annotated_op_loc {
> bool mem_ref;
> bool multi_regs;
> bool imm;
> + int reg_type;
Just bool global_reg would be enough unless you plan to add more
register types.
Thanks,
Namhyung
> };
>
> enum annotated_insn_ops {
> diff --git a/tools/perf/util/include/dwarf-regs.h b/tools/perf/util/include/dwarf-regs.h
> index 75b28dcc8317..a5e8e9498683 100644
> --- a/tools/perf/util/include/dwarf-regs.h
> +++ b/tools/perf/util/include/dwarf-regs.h
> @@ -5,6 +5,7 @@
>
> #define DWARF_REG_PC 0xd3af9c /* random number */
> #define DWARF_REG_FB 0xd3affb /* random number */
> +#define DWARF_REG_GLOBAL 0xd3affc /* random number */
>
> #ifdef HAVE_DWARF_SUPPORT
> const char *get_arch_regstr(unsigned int n);
> --
> 2.43.0
>
next prev parent reply other threads:[~2024-07-18 5:11 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-13 16:55 [PATCH V7 00/18] Add data type profiling support for powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 01/18] tools/perf: Move the data structures related to register type to header file Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 02/18] tools/perf: Add "update_insn_state" callback function to handle arch specific instruction tracking Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 03/18] tools/perf: Update TYPE_STATE_MAX_REGS to include max of regs in powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 04/18] tools/perf: Add disasm_line__parse to parse raw instruction for powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 05/18] tools/perf: Add support to capture and parse raw instruction in powerpc using dso__data_read_offset utility Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 06/18] tools/perf: Update parameters for reg extract functions to use raw instruction on powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 07/18] tools/perf: Add parse function for memory instructions in powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 08/18] tools/perf: Add support to identify memory instructions of opcode 31 " Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 09/18] tools/perf: Add some of the arithmetic instructions to support instruction tracking " Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 10/18] tools/perf: Add more instructions for instruction tracking Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 11/18] tools/perf: Update instruction tracking for powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 12/18] tools/perf: Make capstone_init non-static so that it can be used during symbol disassemble Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 13/18] tools/perf: Use capstone_init and remove open_capstone_handle from disasm.c Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 14/18] tools/perf: Add support to use libcapstone in powerpc Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 15/18] tools/perf: Add support to find global register variables using find_data_type_global_reg Athira Rajeev
2024-07-18 5:11 ` Namhyung Kim [this message]
2024-07-13 16:55 ` [PATCH V7 16/18] tools/perf: Add support for global_die to capture name of variable in case of register defined variable Athira Rajeev
2024-07-18 5:25 ` Namhyung Kim
2024-07-13 16:55 ` [PATCH V7 17/18] tools/perf: Update data_type_cmp and sort__typeoff_sort function to include var_name in comparison Athira Rajeev
2024-07-13 16:55 ` [PATCH V7 18/18] tools/perf: Set instruction name to be used with insn-stat when using raw instruction Athira Rajeev
2024-07-16 14:18 ` [PATCH V7 00/18] Add data type profiling support for powerpc kajoljain
2024-07-18 5:34 ` Namhyung Kim
2024-07-18 6:11 ` Athira Rajeev
2024-07-18 6:43 ` Namhyung Kim
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=ZpikBQdVxVpy5ETw@google.com \
--to=namhyung@kernel.org \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=akanksha@linux.ibm.com \
--cc=atrajeev@linux.vnet.ibm.com \
--cc=christophe.leroy@csgroup.eu \
--cc=disgoel@linux.vnet.ibm.com \
--cc=irogers@google.com \
--cc=jolsa@kernel.org \
--cc=kjain@linux.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=maddy@linux.ibm.com \
--cc=segher@kernel.crashing.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;
as well as URLs for NNTP newsgroup(s).