210dffe0d1
99 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Arnaldo Carvalho de Melo
|
1bc63a4cff |
fprintf: Fixup truncation possibility pointed out by gcc -O2
Addresses this warning:
[ 26%] Building C object CMakeFiles/dwarves.dir/dwarves_fprintf.c.o
/home/acme/git/pahole/dwarves_fprintf.c: In function ‘type__fprintf’:
/home/acme/git/pahole/dwarves_fprintf.c:709:47: error: ‘%s’ directive output may be truncated writing up to 257 bytes into a region of size 256 [-Werror=format-truncation=]
709 | snprintf(namebfptr, sizeof(namebfptr), "* %s", name);
| ^~
/home/acme/git/pahole/dwarves_fprintf.c:709:5: note: ‘snprintf’ output between 3 and 260 bytes into a destination of size 258
709 | snprintf(namebfptr, sizeof(namebfptr), "* %s", name);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
644466dce7 |
fprintf: Remove extraneous sizeof operator
That was affecting --suppress_aligned_attribute, detected using the 'btfdiff' utility, that uses pahole with DWARF and BTF and compares their outputs, and since we don't have this in BTF, we use --suppress_aligned_attribute with '-F dwarf'. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
a59459bb80 |
fprintf: Account inline type __aligned__ member types for spacing
Fixing things like that anonymous union offset/size comment
misalignment:
union {
refcount_t rcu_users; /* 2568 4 */
struct callback_head rcu __attribute__((__aligned__(8))); /* 2568 16 */
- } __attribute__((__aligned__(8))); /* 2568 16 */
+ } __attribute__((__aligned__(8))); /* 2568 16 */
struct pipe_inode_info * splice_pipe; /* 2584 8 */
struct page_frag task_frag; /* 2592 16 */
struct task_delay_info * delays; /* 2608 8 */
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
56547f133a |
fprintf: Fix alignment of class members that are structs/enums/unions
E.g. look at that 'completion' member in this struct:
struct cpu_stop_done {
atomic_t nr_todo; /* 0 4 */
int ret; /* 4 4 */
- struct completion completion; /* 8 32 */
+ struct completion completion; /* 8 32 */
/* size: 40, cachelines: 1, members: 3 */
/* last cacheline: 40 bytes */
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
ccf3eebfcd |
btf_loader: Add support for BTF_KIND_FUNC
Some changes to the fprintf routines were needed, as BTF has as the function type just a BTF_KIND_FUNC_PROTO, while DWARF has as the type for a function its return value type. With a function->btf flag this was overcome and all the other goodies in pfunct are present, for instance: $ pahole -JV examples/tcp.o | grep -w FUNC | head [4068] FUNC tcp_init type_id=4067 [4070] FUNC tcp_abort type_id=4069 [4072] FUNC tcp_done type_id=4071 [4074] FUNC tcp_md5_hash_key type_id=4073 [4076] FUNC tcp_md5_hash_skb_data type_id=4075 [4078] FUNC tcp_get_md5sig_pool type_id=4077 [4080] FUNC tcp_alloc_md5sig_pool type_id=4079 [4082] FUNC compat_tcp_getsockopt type_id=4081 [4084] FUNC tcp_getsockopt type_id=4083 [4086] FUNC tcp_get_timestamping_opt_stats type_id=4085 $ $ pfunct -F btf examples/tcp.o | head memset memcpy tcp_enter_memory_pressure tcp_leave_memory_pressure tcp_init_sock tcp_init_transfer tcp_poll tcp_ioctl tcp_splice_read sk_stream_alloc_skb $ $ pfunct --prototype -F btf examples/tcp.o | head void * memset(void * p, int c, __kernel_size_t size); void * memcpy(void * p, const void * q, __kernel_size_t size); void tcp_enter_memory_pressure(struct sock * sk); void tcp_leave_memory_pressure(struct sock * sk); void tcp_init_sock(struct sock * sk); void tcp_init_transfer(struct sock * sk, int bpf_op); __poll_t tcp_poll(struct file * file, struct socket * sock, poll_table * wait); int tcp_ioctl(struct sock * sk, int cmd, long unsigned int arg); ssize_t tcp_splice_read(struct socket * sock, loff_t * ppos, struct pipe_inode_info * pipe, size_t len, unsigned int flags); struct sk_buff * sk_stream_alloc_skb(struct sock * sk, int size, gfp_t gfp, bool force_schedule); $ Now to ask just for the 'struct sock' 'methods', i.e. functions that have as one of its arguments a pointer to the given 'class' name: $ pfunct --class sock -F btf examples/tcp.o | head tcp_abort tcp_done compat_tcp_getsockopt tcp_getsockopt tcp_get_info compat_tcp_setsockopt tcp_setsockopt tcp_disconnect tcp_write_queue_purge tcp_close $ Then ask for the prototypes, which requires -V, should have that fixed: $ pfunct -V --prototypes --class sock -F btf examples/tcp.o | head int tcp_abort(struct sock * sk, int err); void tcp_done(struct sock * sk); int compat_tcp_getsockopt(struct sock * sk, int level, int optname, char * optval, int * optlen); int tcp_getsockopt(struct sock * sk, int level, int optname, char * optval, int * optlen); void tcp_get_info(struct sock * sk, struct tcp_info * info); int compat_tcp_setsockopt(struct sock * sk, int level, int optname, char * optval, unsigned int optlen); int tcp_setsockopt(struct sock * sk, int level, int optname, char * optval, unsigned int optlen); int tcp_disconnect(struct sock * sk, int flags); void tcp_write_queue_purge(struct sock * sk); void tcp_close(struct sock * sk, long int timeout); $ Don't like prototypes with parm names, got you covered: $ pfunct --no_parm_names -V --prototypes --class sock -F btf examples/tcp.o | head int tcp_abort(struct sock *, int); void tcp_done(struct sock *); int compat_tcp_getsockopt(struct sock *, int, int, char *, int *); int tcp_getsockopt(struct sock *, int, int, char *, int *); void tcp_get_info(struct sock *, struct tcp_info *); int compat_tcp_setsockopt(struct sock *, int, int, char *, unsigned int); int tcp_setsockopt(struct sock *, int, int, char *, unsigned int); int tcp_disconnect(struct sock *, int); void tcp_write_queue_purge(struct sock *); void tcp_close(struct sock *, long int); $ Don't like long options and want just one function? $ pfunct -f tcp_setsockopt -F btf examples/tcp.o int tcp_setsockopt(struct sock * sk, int level, int optname, char * optval, unsigned int optlen); $ Want to generate compileable code for all of those functions, full with the necessary types, etc? $ pfunct -F btf --compile examples/tcp.o > a.c $ gcc -c -o a.o a.c $ pfunct -F dwarf --prototypes --class sock a.o | head pfunct: a.o: No debugging information found $ gcc -g -c -o a.o a.c $ pfunct -V -F dwarf --prototypes --class sock a.o | head void tcp_enter_memory_pressure(struct sock * sk); void tcp_leave_memory_pressure(struct sock * sk); void tcp_init_sock(struct sock * sk); void tcp_init_transfer(struct sock * sk, int bpf_op); int tcp_ioctl(struct sock * sk, int cmd, long unsigned int arg); struct sk_buff * sk_stream_alloc_skb(struct sock * sk, int size, gfp_t gfp, bool force_schedule); ssize_t do_tcp_sendpages(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendpage_locked(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendpage(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendmsg_locked(struct sock * sk, struct msghdr * msg, size_t size); $ Now lets go full circle and encode BTF for this a.o generated from source code generated from the original BTF info in that examples/tcp.o file: $ pahole -JV a.o | tail [465] FUNC_PROTO (anon) return=35 args=(392 hp, 393 skb, 5 header_len) [466] FUNC tcp_md5_hash_skb_data type_id=465 [467] FUNC_PROTO (anon) return=35 args=(392 hp, 394 key) [468] FUNC tcp_md5_hash_key type_id=467 [469] FUNC_PROTO (anon) return=0 args=(49 sk) [470] FUNC tcp_done type_id=469 [471] FUNC_PROTO (anon) return=35 args=(49 sk, 35 err) [472] FUNC tcp_abort type_id=471 [473] FUNC_PROTO (anon) return=0 args=(void) [474] FUNC tcp_init type_id=473 $ $ pfunct -F btf -V --prototypes --class=sock a.o | head void tcp_enter_memory_pressure(struct sock * sk); void tcp_leave_memory_pressure(struct sock * sk); void tcp_init_sock(struct sock * sk); void tcp_init_transfer(struct sock * sk, int bpf_op); int tcp_ioctl(struct sock * sk, int cmd, long unsigned int arg); struct sk_buff * sk_stream_alloc_skb(struct sock * sk, int size, gfp_t gfp, bool force_schedule); ssize_t do_tcp_sendpages(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendpage_locked(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendpage(struct sock * sk, struct page * page, int offset, size_t size, int flags); int tcp_sendmsg_locked(struct sock * sk, struct msghdr * msg, size_t size); $ Curious about the code generated by 'pfunct -F btf --compile examples/tcp.o? http://vger.kernel.org/~acme/pahole/pfunct-F-BTF--compile-examples-tcp.o.txt Cc: Alexei Starovoitov <ast@fb.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Yonghong Song <yhs@fb.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
3e8f09e304 |
pdwtags: Print DW_TAG_subroutine_type as well
So that we can see at least via pdwtags those tags, be it from DWARF of BTF. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
d5e01d10e5 |
fprintf: Set tconf.type_spacing earlier
As there are cases where we jump to the type not found label without having done the full copy of *conf to tconf, so at least the type_spacing needs to have been set. This addresses this coverity report entry: Error: UNINIT (CWE-457): [#def23] dwarves-1.13/dwarves_fprintf.c:600: var_decl: Declaring variable "tconf" without initializer. dwarves-1.13/dwarves_fprintf.c:779: uninit_use_in_call: Using uninitialized value "tconf.type_spacing" when calling "fprintf". # 777| return printed; # 778| out_type_not_found: # 779|-> printed = fprintf(fp, "%-*s %s", tconf.type_spacing, "<ERROR>", name); # 780| goto out; # 781| } Reported-by: William Cohen <wcohen@redhat.com> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
c6a9a0eb6a |
fprintf: Fix up decrementing recursivity level in type__fprintf()
In some exit paths we were accessing tconf before we had copied it from
conf, and we also were losing track of the original type that could have
been expanded and where we bumped the recursivity level member, so just
store that original type and if it is set, decrement its recursivity
level.
This addresses these coverity report entries:
Error: UNINIT (CWE-457): [#def21]
dwarves-1.13/dwarves_fprintf.c:600: var_decl: Declaring variable "tconf" without initializer.
dwarves-1.13/dwarves_fprintf.c:774: uninit_use: Using uninitialized value "tconf.expand_types".
# 772| }
# 773| out:
# 774|-> if (tconf.expand_types)
# 775| --type->recursivity_level;
# 776|
Error: FORWARD_NULL (CWE-476): [#def22]
dwarves-1.13/dwarves_fprintf.c:605: var_compare_op: Comparing "type" to null implies that "type" might be null.
dwarves-1.13/dwarves_fprintf.c:775: var_deref_op: Dereferencing null pointer "type".
# 773| out:
# 774| if (tconf.expand_types)
# 775|-> --type->recursivity_level;
# 776|
# 777| return printed;
Reported-by: William Cohen <wcohen@redhat.com>
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
f67c281f98 |
fprintf: Correct the type for the 'cacheline' variable, it should be uint32_t
[ 22%] Building C object CMakeFiles/dwarves.dir/dwarves_fprintf.c.o
/home/acme/git/pahole/dwarves_fprintf.c: In function ‘union__fprintf’:
/home/acme/git/pahole/dwarves_fprintf.c:963:34: error: pointer targets in assignment from ‘int *’ to ‘uint32_t *’ {aka ‘unsigned int *’} differ in signedness [-Werror=pointer-sign]
uconf.cachelinep = &cacheline;
^
cc1: all warnings being treated as errors
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
7b36fab5a8 |
fprintf: Add guard against unlikely overlapping copy
In |
||
|
Arnaldo Carvalho de Melo
|
e737976c09 |
fprintf: Do not scrub type when looking up its type
As we end up goto'ing to a place where we need that original type and
were having just a NULL pointer, oops, fix it.
Related to:
Error: NULL_RETURNS (CWE-476): [#def18]
dwarves-1.13/dwarves_fprintf.c:727: returned_null: "cu__type" returns "NULL" (checked 54 out of 62 times).
dwarves-1.13/dwarves_fprintf.c:727: var_assigned: Assigning: "type" = "NULL" return value from "cu__type".
dwarves-1.13/dwarves_fprintf.c:686: dereference: Dereferencing "type", which is known to be "NULL".
dwarves-1.13/codiff.c:137: example_assign: Example 1: Assigning: "old_type" = return value from "cu__type(old_cu, old->tag.type)".
dwarves-1.13/codiff.c:141: example_checked: Example 1 (cont.): "old_type" has its value checked in "old_type == NULL".
dwarves-1.13/ctracer.c:356: example_assign: Example 2: Assigning: "type" = return value from "cu__type(cu, tag->type)".
dwarves-1.13/ctracer.c:358: example_checked: Example 2 (cont.): "type" has its value checked in "type == NULL".
dwarves-1.13/dwarves.c:914: example_assign: Example 3: Assigning: "type" = return value from "cu__type(cu, tag->type)".
dwarves-1.13/dwarves.c:916: example_checked: Example 3 (cont.): "type" has its value checked in "type == NULL".
dwarves-1.13/dwarves.c:941: example_assign: Example 4: Assigning: "tag" = return value from "cu__type(cu, var->ip.tag.type)".
dwarves-1.13/dwarves.c:942: example_checked: Example 4 (cont.): "tag" has its value checked in "tag != NULL".
dwarves-1.13/dwarves_emit.c:139: example_assign: Example 5: Assigning: "ptr_type" = return value from "cu__type(cu, type->type)".
dwarves-1.13/dwarves_emit.c:141: example_checked: Example 5 (cont.): "ptr_type" has its value checked in "ptr_type == NULL".
# 684|
# 685| next_type:
# 686|-> switch (type->tag) {
# 687| case DW_TAG_pointer_type:
# 688|
Reported-by: William Cohen <wcohen@redhat.com>
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
e95dacb704 |
fprintf: Remove unused printf arg when printing enumerations
A cut'n'paste error, noticed by coverity:
Error: PRINTF_ARGS: [#def16]
dwarves-1.13/dwarves_fprintf.c:369: extra_argument: This argument was not used by the format string: "conf->suffix". [Note: The source code implementation of the function has been overridden by a builtin model.]
# 367| */
# 368| if (type->size / 8 != sizeof(int))
# 369|-> printed += fprintf(fp, " __attribute__((__packed__))", conf->suffix);
# 370|
# 371| if (conf->suffix)
Reported-by: William Cohen <wcohen@redhat.com>
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
b1412a88bb |
fprintf: Fixup handling classes with no members
Will Cohen reported this NULL pointer dereference while processing some
object linking with cuda:
#0 0x00007ffff7f91453 in __class__fprintf (class=0x522560, cu=0x40ff80, conf=0x7fffffffa930, fp=0x7ffff7ece780 <_IO_2_1_stdout_>)
at /home/acme/git/pahole/dwarves_fprintf.c:1624
#1 0x00007ffff7f92195 in tag__fprintf (tag=0x522560, cu=0x40ff80, conf=0x7fffffffa930, fp=0x7ffff7ece780 <_IO_2_1_stdout_>)
at /home/acme/git/pahole/dwarves_fprintf.c:1835
#2 0x00007ffff7f90b57 in __class__fprintf (class=0x5224c0, cu=0x40ff80, conf=0x7fffffffaaa0, fp=0x7ffff7ece780 <_IO_2_1_stdout_>)
at /home/acme/git/pahole/dwarves_fprintf.c:1406
#3 0x00007ffff7f92195 in tag__fprintf (tag=0x5224c0, cu=0x40ff80, conf=0x40a200 <conf>, fp=0x7ffff7ece780 <_IO_2_1_stdout_>)
at /home/acme/git/pahole/dwarves_fprintf.c:1835
#4 0x0000000000402d03 in class_formatter (class=0x5224c0, cu=0x40ff80, id=1257) at /home/acme/git/pahole/pahole.c:224
#5 0x0000000000403074 in print_classes (cu=0x40ff80) at /home/acme/git/pahole/pahole.c:319
#6 0x0000000000404bb2 in pahole_stealer (cu=0x40ff80, conf_load=0x40a240 <conf_load>) at /home/acme/git/pahole/pahole.c:1174
#7 0x00007ffff7f9ff73 in finalize_cu (cus=0x40b2b0, cu=0x40ff80, dcu=0x7fffffffacf0, conf=0x40a240 <conf_load>)
at /home/acme/git/pahole/dwarf_loader.c:2227
#8 0x00007ffff7f9ffac in finalize_cu_immediately (cus=0x40b2b0, cu=0x40ff80, dcu=0x7fffffffacf0, conf=0x40a240 <conf_load>)
at /home/acme/git/pahole/dwarf_loader.c:2236
#9 0x00007ffff7fa064c in cus__load_module (cus=0x40b2b0, conf=0x40a240 <conf_load>, mod=0x40d760, dw=0x40e980, elf=0x40b360,
filename=0x7fffffffd5e3 "examples/wcohen/02_Exercise.cuda") at /home/acme/git/pahole/dwarf_loader.c:2389
#10 0x00007ffff7fa0760 in cus__process_dwflmod (dwflmod=0x40d760, userdata=0x40d770, name=0x40d910 "examples/wcohen/02_Exercise.cuda",
base=4194304, arg=0x7fffffffcf10) at /home/acme/git/pahole/dwarf_loader.c:2434
#11 0x00007ffff7f32be1 in dwfl_getmodules () from /lib64/libdw.so.1
#12 0x00007ffff7fa0820 in cus__process_file (cus=0x40b2b0, conf=0x40a240 <conf_load>, fd=3,
filename=0x7fffffffd5e3 "examples/wcohen/02_Exercise.cuda") at /home/acme/git/pahole/dwarf_loader.c:2487
#13 0x00007ffff7fa089c in dwarf__load_file (cus=0x40b2b0, conf=0x40a240 <conf_load>, filename=0x7fffffffd5e3 "examples/wcohen/02_Exercise.cuda")
at /home/acme/git/pahole/dwarf_loader.c:2504
#14 0x00007ffff7f8b0dd in cus__load_file (cus=0x40b2b0, conf=0x40a240 <conf_load>, filename=0x7fffffffd5e3 "examples/wcohen/02_Exercise.cuda")
at /home/acme/git/pahole/dwarves.c:1745
#15 0x00007ffff7f8bc2a in cus__load_files (cus=0x40b2b0, conf=0x40a240 <conf_load>, filenames=0x7fffffffd150)
at /home/acme/git/pahole/dwarves.c:2109
#16 0x0000000000404ff0 in main (argc=2, argv=0x7fffffffd148) at /home/acme/git/pahole/pahole.c:1294
(gdb)
(gdb) p class__name(class, cu)
$6 = 0x5cbb85 "__nv_hdl_helper_trait<__nv_dl_tag<int (*)(int, char**), main, 1u>, void (main(int, char**)::__lambda0::*)(int, double&)const>"
(gdb) p class->type.nr_members
$7 = 0
(gdb) p last
$8 = (struct class_member *) 0x0
(gdb)
So, before checking for bitfield details, first check if there were
members.
Now, if we show all structs/classes in that object file and look for the
above data structure, we find it inside another:
$ pahole examples/wcohen/02_Exercise.cuda
<SNIP>
struct __nv_hdl_helper_trait_outer<false, false, int, Kokkos::View<double**>, Kokkos::View<double*>, Kokkos::View<double*> > {
struct __nv_hdl_helper_trait<__nv_dl_tag<int (*)(int, char**), main, 1u>, void (main(int, char**)::__lambda0::*)(int, double&)const> {
class __nv_hdl_wrapper_t<false, false, __nv_dl_tag<int (*)(int, char**), main, 1u>, void(int, double&), int, Kokkos::View<doubl get<main(int, char**)::__lambda0>(class __lambda0, int, class View<double**>, class View<double*>, class View<double*>);
/* size: 1, cachelines: 0, members: 0 */
/* padding: 1 */
/* last cacheline: 1 bytes */
};
/* size: 1, cachelines: 0, members: 0 */
/* padding: 1 */
/* last cacheline: 1 bytes */
};
<SNIP>
$
Reported-by: William Cohen <wcohen@redhat.com>
Fixes:
|
||
|
Timo Paulssen
|
3ed9a67967 |
fprintf: Avoid null dereference with NULL configs
Adding support for suppress_packed and leaving out bitfield paddings at the end of classes when no alignment info is available accidentally used conf instead of cconf, where conf can be NULL, which causes cconf to have the default fprintf config. Fixes: |
||
|
Arnaldo Carvalho de Melo
|
568dae4bd4 |
printf: Fixup printing "const" early with "const void"
When printing const more early we're not considering "const void *", fix
it.
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
68f261d8df |
fprintf: Fix recursively printing named structs in --expand_types
In |
||
|
Arnaldo Carvalho de Melo
|
0fb727166a |
pfunct: Strip inlines in the code generated for --compile
If we have:
inline void process_adjtimex_modes(const struct __kernel_timex * txc, s32 * time_tai)
{
}
And any other struct receiving as a parameter pointers to 'struct
__kerne_timex', then the source file with the above inline, since it
doesn't have any inline expansion, i.e. 'pfunct --compile' generates
just empty function bodies, the types won't be included in the resulting
.o.
Since the original file has the expansions, type types will be there and
thus we will not be able to compare those types, so ask for any 'inline'
to be stripped, so that we keep those types and 'fullcircle' can do its
work.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
30526e2848 |
fprintf: Print relative offsets for inner pointer structs
I.e. before:
$ pahole -C parsed_partitions /home/acme/git/build/v5.1-rc4+/block/partitions/check.o > a.c ; head -29 a.c
struct parsed_partitions {
struct block_device * bdev; /* 0 8 */
char name[32]; /* 8 32 */
struct {
sector_t from; /* 40 8 */
sector_t size; /* 48 8 */
int flags; /* 56 4 */
bool has_info; /* 60 1 */
struct partition_meta_info info; /* 61 101 */
} * parts; /* 40 8 */
int next; /* 48 4 */
int limit; /* 52 4 */
bool access_beyond_eod; /* 56 1 */
/* XXX 7 bytes hole, try to pack */
/* --- cacheline 1 boundary (64 bytes) --- */
char * pp_buf; /* 64 8 */
/* size: 72, cachelines: 2, members: 7 */
/* sum members: 65, holes: 1, sum holes: 7 */
/* last cacheline: 8 bytes */
};
$
Now:
$ pahole -C parsed_partitions /home/acme/git/build/v5.1-rc4+/block/partitions/check.o
struct parsed_partitions {
struct block_device * bdev; /* 0 8 */
char name[32]; /* 8 32 */
struct {
sector_t from; /* 0 8 */
sector_t size; /* 8 8 */
int flags; /* 16 4 */
bool has_info; /* 20 1 */
struct partition_meta_info info; /* 21 101 */
} * parts; /* 40 8 */
int next; /* 48 4 */
int limit; /* 52 4 */
bool access_beyond_eod; /* 56 1 */
/* XXX 7 bytes hole, try to pack */
/* --- cacheline 1 boundary (64 bytes) --- */
char * pp_buf; /* 64 8 */
/* size: 72, cachelines: 2, members: 7 */
/* sum members: 65, holes: 1, sum holes: 7 */
/* last cacheline: 8 bytes */
};
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
cf459ca16f |
fprintf: Pretty print struct members that are pointers to nameless structs
I.e. to structs defined inside other structs, without a name:
Before:
$ pahole -C parsed_partitions /home/acme/git/build/v5.1-rc4+/block/partitions/check.o
struct parsed_partitions {
struct block_device * bdev; /* 0 8 */
char name[32]; /* 8 32 */
struct * parts; /* 40 8 */
int next; /* 48 4 */
int limit; /* 52 4 */
bool access_beyond_eod; /* 56 1 */
/* XXX 7 bytes hole, try to pack */
/* --- cacheline 1 boundary (64 bytes) --- */
char * pp_buf; /* 64 8 */
/* size: 72, cachelines: 2, members: 7 */
/* sum members: 65, holes: 1, sum holes: 7 */
/* last cacheline: 8 bytes */
};
$
After:
$ pahole -C parsed_partitions /home/acme/git/build/v5.1-rc4+/block/partitions/check.o
struct parsed_partitions {
struct block_device * bdev; /* 0 8 */
char name[32]; /* 8 32 */
struct {
sector_t from; /* 40 8 */
sector_t size; /* 48 8 */
int flags; /* 56 4 */
bool has_info; /* 60 1 */
struct partition_meta_info info; /* 61 101 */
} * parts; /* 40 8 */
int next; /* 48 4 */
int limit; /* 52 4 */
bool access_beyond_eod; /* 56 1 */
/* XXX 7 bytes hole, try to pack */
/* --- cacheline 1 boundary (64 bytes) --- */
char * pp_buf; /* 64 8 */
/* size: 72, cachelines: 2, members: 7 */
/* sum members: 65, holes: 1, sum holes: 7 */
/* last cacheline: 8 bytes */
};
$
Still need to align that offsets, leave this for later.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
2bcb01fc2f |
fprintf: Handle single zero sized array member structs
/home/acme/git/build/v5.1-rc4+/fs/proc/kcore.o
/tmp/fullcircle.Vnd2oz.c:788:29: error: flexible array member in a struct with no named members
char x[]; /* 0 0 */
Original:
include/linux/mmzone.h, line 109:
/*
* zone->lock and the zone lru_lock are two of the hottest locks in the kernel.
* So add a wild amount of padding here to ensure that they fall into separate
* cachelines. There are very few zone structures in the machine, so space
* consumption is not a concern here.
*/
#if defined(CONFIG_SMP)
struct zone_padding {
char x[0];
} ____cacheline_internodealigned_in_smp;
#define ZONE_PADDING(name) struct zone_padding name;
#else
#define ZONE_PADDING(name)
#endif
B0rken:
struct zone_padding {
char x[]; /* 0 0 */
/* size: 0, cachelines: 0, members: 1 */
} __attribute__((__aligned__(64)));
Fixed:
struct zone_padding {
char x[0]; /* 0 0 */
/* size: 0, cachelines: 0, members: 1 */
} __attribute__((__aligned__(64)));
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
0987266cd9 |
fprintf: Deal with zero sized arrays in the midle of a union
gcc dislikes invalid C:
/home/acme/git/build/v5.1-rc4+/fs/ceph/export.o
/tmp/fullcircle.CvAfpM.c:591:22: error: flexible array member in union
__u32 raw[]; /* 0 0 */
^~~
Before:
$ pahole -C fid /home/acme/git/build/v5.1-rc4+/fs/ceph/export.o
struct fid {
union {
struct {
u32 ino; /* 0 4 */
u32 gen; /* 4 4 */
u32 parent_ino; /* 8 4 */
u32 parent_gen; /* 12 4 */
} i32; /* 0 16 */
struct {
u32 block; /* 0 4 */
u16 partref; /* 4 2 */
u16 parent_partref; /* 6 2 */
u32 generation; /* 8 4 */
u32 parent_block; /* 12 4 */
u32 parent_generation; /* 16 4 */
} udf; /* 0 20 */
__u32 raw[]; /* 0 0 */
}; /* 0 20 */
/* size: 20, cachelines: 1, members: 1 */
/* last cacheline: 20 bytes */
};
$
After:
$ pahole -C fid /home/acme/git/build/v5.1-rc4+/fs/ceph/export.o
struct fid {
union {
struct {
u32 ino; /* 0 4 */
u32 gen; /* 4 4 */
u32 parent_ino; /* 8 4 */
u32 parent_gen; /* 12 4 */
} i32; /* 0 16 */
struct {
u32 block; /* 0 4 */
u16 partref; /* 4 2 */
u16 parent_partref; /* 6 2 */
u32 generation; /* 8 4 */
u32 parent_block; /* 12 4 */
u32 parent_generation; /* 16 4 */
} udf; /* 0 20 */
__u32 raw[0]; /* 0 0 */
}; /* 0 20 */
/* size: 20, cachelines: 1, members: 1 */
/* last cacheline: 20 bytes */
};
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
1101337a74 |
fprintf: Deal with zero sized arrays in the middle of a struct
Consider:
struct ipc64_perm {
__kernel_key_t key;
__kernel_uid32_t uid;
__kernel_gid32_t gid;
__kernel_uid32_t cuid;
__kernel_gid32_t cgid;
__kernel_mode_t mode;
/* pad if mode_t is u16: */
unsigned char __pad1[4 - sizeof(__kernel_mode_t)];
unsigned short seq;
unsigned short __pad2;
__kernel_ulong_t __unused1;
__kernel_ulong_t __unused2;
};
That is a roundabout way of using __attribute__(__aligned__(4)), but
should work nonetheless.
We were not putting the [0] in that zero sized array which ended up
making gcc complain with:
$ gcc -g -c shm.c
shm.c:199:29: error: flexible array member not at end of struct
unsigned char __pad1[]; /* 24 0 */
^~~~~~
$
Now this works, i.e. generates compilable source code out of the
type tags, be it from BTF or from DWARF, i.e. this is all from the
internal representation of such types, agnostic wrt the original type
format.
So, the full circle:
$ pahole -C ipc64_perm /home/acme/git/build/v5.1-rc4+/ipc/shm.o
struct ipc64_perm {
__kernel_key_t key; /* 0 4 */
__kernel_uid32_t uid; /* 4 4 */
__kernel_gid32_t gid; /* 8 4 */
__kernel_uid32_t cuid; /* 12 4 */
__kernel_gid32_t cgid; /* 16 4 */
__kernel_mode_t mode; /* 20 4 */
unsigned char __pad1[0]; /* 24 0 */
short unsigned int seq; /* 24 2 */
short unsigned int __pad2; /* 26 2 */
/* XXX 4 bytes hole, try to pack */
__kernel_ulong_t __unused1; /* 32 8 */
__kernel_ulong_t __unused2; /* 40 8 */
/* size: 48, cachelines: 1, members: 11 */
/* sum members: 44, holes: 1, sum holes: 4 */
/* last cacheline: 48 bytes */
};
$ pfunct --compile /home/acme/git/build/v5.1-rc4+/ipc/shm.o > shm.c
$ gcc -g -c shm.c
$ pahole -C ipc64_perm shm.o
struct ipc64_perm {
__kernel_key_t key; /* 0 4 */
__kernel_uid32_t uid; /* 4 4 */
__kernel_gid32_t gid; /* 8 4 */
__kernel_uid32_t cuid; /* 12 4 */
__kernel_gid32_t cgid; /* 16 4 */
__kernel_mode_t mode; /* 20 4 */
unsigned char __pad1[0]; /* 24 0 */
short unsigned int seq; /* 24 2 */
short unsigned int __pad2; /* 26 2 */
/* XXX 4 bytes hole, try to pack */
__kernel_ulong_t __unused1; /* 32 8 */
__kernel_ulong_t __unused2; /* 40 8 */
/* size: 48, cachelines: 1, members: 11 */
/* sum members: 44, holes: 1, sum holes: 4 */
/* last cacheline: 48 bytes */
};
$
And for a chuckle, the original source code with a bit of history about
struct layout worries:
include/uapi/asm-generic/ipcbuf.h:
/*
* The generic ipc64_perm structure:
* Note extra padding because this structure is passed back and forth
* between kernel and user space.
*
* ipc64_perm was originally meant to be architecture specific, but
* everyone just ended up making identical copies without specific
* optimizations, so we may just as well all use the same one.
*
* Pad space is left for:
* - 32-bit mode_t on architectures that only had 16 bit
* - 32-bit seq
* - 2 miscellaneous 32-bit values
*/
struct ipc64_perm {
__kernel_key_t key;
__kernel_uid32_t uid;
__kernel_gid32_t gid;
__kernel_uid32_t cuid;
__kernel_gid32_t cgid;
__kernel_mode_t mode;
/* pad if mode_t is u16: */
unsigned char __pad1[4 - sizeof(__kernel_mode_t)];
unsigned short seq;
unsigned short __pad2;
__kernel_ulong_t __unused1;
__kernel_ulong_t __unused2;
};
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
13aa13eb0c |
fprintf: Don't reuse 'type' in multiple scopes in the same function
Its confusing, so rename some variables. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
6f0f9a8815 |
fprintf: Fixup multi-dimensional zero sized arrays const handling
Before:
$ pahole -C piix_map_db /home/acme/git/build/v5.1-rc4+/drivers/ata/ata_piix.o
struct piix_map_db {
const u32 mask; /* 0 4 */
const u16 port_enable; /* 4 2 */
/* XXX 2 bytes hole, try to pack */
const const int map[][4]; /* 8 0 */
/* size: 8, cachelines: 1, members: 3 */
/* sum members: 6, holes: 1, sum holes: 2 */
/* last cacheline: 8 bytes */
};
$
After:
$ pahole -C piix_map_db /home/acme/git/build/v5.1-rc4+/drivers/ata/ata_piix.o
struct piix_map_db {
const u32 mask; /* 0 4 */
const u16 port_enable; /* 4 2 */
/* XXX 2 bytes hole, try to pack */
const int map[][4]; /* 8 0 */
/* size: 8, cachelines: 1, members: 3 */
/* sum members: 6, holes: 1, sum holes: 2 */
/* last cacheline: 8 bytes */
};
$
The DWARF tag sequence:
<2><17e50>: Abbrev Number: 12 (DW_TAG_member)
<17e51> DW_AT_name : map
<17e55> DW_AT_decl_file : 1
<17e56> DW_AT_decl_line : 160
<17e57> DW_AT_decl_column : 12
<17e58> DW_AT_type : <0x17e78>
<17e5c> DW_AT_data_member_location: 8
<1><17e78>: Abbrev Number: 15 (DW_TAG_const_type)
<17e79> DW_AT_type : <0x17e63>
<1><17e63>: Abbrev Number: 11 (DW_TAG_array_type)
<17e64> DW_AT_type : <0xd8>
<1><d8>: Abbrev Number: 15 (DW_TAG_const_type)
<d9> DW_AT_type : <0xd1>
<1><d1>: Abbrev Number: 120 (DW_TAG_base_type)
<d2> DW_AT_byte_size : 4
<d3> DW_AT_encoding : 5 (signed)
<d4> DW_AT_name : int
const -> array -> const -> int
So just make the check be at least one dimension, if the number of
elements is zero, then drop the double const.
With this btfdiff for the allyesconfig ppc64 reference kernel we're
using is again clean.
$ pahole -F btf --sizes examples/vmlinux-aarch64 | wc -l
51023
$
> 50K types with output from BTF and DWARF matching.
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
9a4d719304 |
fprintf: Allow suppressing the inferred __attribute__((__packed__))
We use things like DW_AT_alignment, so not all of those attributes are inferred by formats like BTF that lack that info, allow suppressing the output and make btfdiff ask for both DWARF and BTF output to have this suppressed. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
ec935ee422 |
fprintf: Allow suppressing the output of force paddings at the end of structs
Things like 'struct timex' in the linux kernel led to this output: Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
986a3b58a8 |
fprintf: Only add bitfield forced paddings when alignment info available
I.e. BTF doesn't have DW_AT_alignment, so avoid emitting the bitfield
for padding at the end:
$ pahole -F btf -C rseq examples/tcp.o
struct rseq {
__u32 cpu_id_start; /* 0 4 */
__u32 cpu_id; /* 4 4 */
union {
__u64 ptr64; /* 8 8 */
__u64 ptr; /* 8 8 */
} rseq_cs; /* 8 8 */
__u32 flags; /* 16 4 */
/* Force padding: */
__u32 :32;
__u32 :32;
__u32 :32;
/* size: 32, cachelines: 1, members: 4 */
/* padding: 12 */
/* last cacheline: 32 bytes */
};
$
After:
$ pahole -F btf -C rseq examples/tcp.o
struct rseq {
__u32 cpu_id_start; /* 0 4 */
__u32 cpu_id; /* 4 4 */
union {
__u64 ptr64; /* 8 8 */
__u64 ptr; /* 8 8 */
} rseq_cs; /* 8 8 */
__u32 flags; /* 16 4 */
/* size: 32, cachelines: 1, members: 4 */
/* padding: 12 */
/* last cacheline: 32 bytes */
};
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
dc6b9437a3 |
emit: Handle structs with DW_AT_alignment=1 meaning __packed__
In the following struct the ceph_entity_addr entries all appear marked with a
__attribute__((__aligned__(8)), which, for the first two members of this type,
'peer_addr' and 'peer_addr_for_me', don't cause the regenerated struct to
differ in layout from the original layout put in place by the compiler as per
the original source code.
But the third member of this type, 'actual_peer_addr' ends up in a different
offset, even in a different cacheline, here is how it looks like in the code generated
from the original source code, at offset 568.
char in_banner[30]; /* 472 30 */
struct ceph_msg_connect out_connect; /* 502 33 */
/* --- cacheline 8 boundary (512 bytes) was 23 bytes ago --- */
struct ceph_msg_connect_reply in_reply; /* 535 26 */
struct ceph_entity_addr actual_peer_addr __attribute__((__aligned__(1))); /* 561 136 */
/* --- cacheline 10 boundary (640 bytes) was 57 bytes ago --- */
struct ceph_msg_header out_hdr; /* 697 53 */
/* XXX 2 bytes hole, try to pack */
/* --- cacheline 11 boundary (704 bytes) was 48 bytes ago --- */
And here is how it looks like when built from the regenerated source code, at
offset 568:
$ pfunct --compile /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o > ceph.c
$ gcc -g -c ceph.c
$ pahole -C ceph_connection ceph.o | head -46
struct ceph_connection {
void * private; /* 0 8 */
const struct ceph_connection_operations * ops; /* 8 8 */
struct ceph_messenger * msgr; /* 16 8 */
atomic_t sock_state; /* 24 4 */
/* XXX 4 bytes hole, try to pack */
struct socket * sock; /* 32 8 */
struct ceph_entity_addr peer_addr __attribute__((__aligned__(8))); /* 40 136 */
/* --- cacheline 2 boundary (128 bytes) was 48 bytes ago --- */
struct ceph_entity_addr peer_addr_for_me __attribute__((__aligned__(8))); /* 176 136 */
/* --- cacheline 4 boundary (256 bytes) was 56 bytes ago --- */
long unsigned int flags; /* 312 8 */
/* --- cacheline 5 boundary (320 bytes) --- */
long unsigned int state; /* 320 8 */
const char * error_msg; /* 328 8 */
struct ceph_entity_name peer_name; /* 336 9 */
/* XXX 7 bytes hole, try to pack */
u64 peer_features; /* 352 8 */
u32 connect_seq; /* 360 4 */
u32 peer_global_seq; /* 364 4 */
struct ceph_auth_handshake * auth; /* 368 8 */
int auth_retry; /* 376 4 */
/* XXX 4 bytes hole, try to pack */
/* --- cacheline 6 boundary (384 bytes) --- */
struct mutex mutex; /* 384 32 */
struct list_head out_queue; /* 416 16 */
struct list_head out_sent; /* 432 16 */
/* --- cacheline 7 boundary (448 bytes) --- */
u64 out_seq; /* 448 8 */
u64 in_seq; /* 456 8 */
u64 in_seq_acked; /* 464 8 */
char in_banner[30]; /* 472 30 */
struct ceph_msg_connect out_connect; /* 502 33 */
/* --- cacheline 8 boundary (512 bytes) was 23 bytes ago --- */
struct ceph_msg_connect_reply in_reply; /* 535 26 */
/* XXX 7 bytes hole, try to pack */
struct ceph_entity_addr actual_peer_addr __attribute__((__aligned__(8))); /* 568 136 */
/* --- cacheline 11 boundary (704 bytes) --- */
$
That happens because 'struct ceph_entity_addr' has that __attribute__
((__aligned__(8)) in the regenerated source code, above, now look at how it
gets regenerated:
$ pahole -C ceph_entity_addr ceph.o
struct ceph_entity_addr {
__le32 type; /* 0 4 */
__le32 nonce; /* 4 4 */
struct __kernel_sockaddr_storage in_addr __attribute__((__aligned__(8))); /* 8 128 */
/* size: 136, cachelines: 3, members: 3 */
/* forced alignments: 1 */
/* last cacheline: 8 bytes */
} __attribute__((__aligned__(8)));
$
While when looking at the original DWARF:
$ pahole -C ceph_entity_addr /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o
struct ceph_entity_addr {
__le32 type; /* 0 4 */
__le32 nonce; /* 4 4 */
struct __kernel_sockaddr_storage in_addr __attribute__((__aligned__(1))); /* 8 128 */
/* size: 136, cachelines: 3, members: 3 */
/* forced alignments: 1 */
/* last cacheline: 8 bytes */
} __attribute__((__aligned__(1)));
$
The confusion may further come from the fact that 'struct __kernel_sockaddr_storage' has,
in the regenerated source code, the __attribute__((__aligned__8)))
$ pahole -C __kernel_sockaddr_storage ceph.o
struct __kernel_sockaddr_storage {
__kernel_sa_family_t ss_family; /* 0 2 */
char __data[126]; /* 2 126 */
/* size: 128, cachelines: 2, members: 2 */
} __attribute__((__aligned__(8)));
$
Which is the same as in the original DWARF:
$ pahole -C __kernel_sockaddr_storage /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o
struct __kernel_sockaddr_storage {
__kernel_sa_family_t ss_family; /* 0 2 */
char __data[126]; /* 2 126 */
/* size: 128, cachelines: 2, members: 2 */
} __attribute__((__aligned__(8)));
$
Looking at the original original source code for 'struct ceph_entity_addr'
helps here, as it reads:
include/linux/ceph/msgr.h, line 63:
/*
* entity_addr -- network address
*/
struct ceph_entity_addr {
__le32 type;
__le32 nonce; /* unique id for process (e.g. pid) */
struct sockaddr_storage in_addr;
} __attribute__ ((packed));
So the original code has no __attribute__((__aligned__(1))), so, lets look at
what the compiler generates for 'struct ceph_entity_addr':
$ readelf -wi /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o | grep ceph_entity_addr -A7
<193a6> DW_AT_name : (indirect string, offset: 0x1586): ceph_entity_addr
<193aa> DW_AT_byte_size : 136
<193ab> DW_AT_alignment : 1
<193ac> DW_AT_decl_file : 296
<193ae> DW_AT_decl_line : 63
<193af> DW_AT_decl_column : 8
<193b0> DW_AT_sibling : <0x193e0>
<2><193b4>: Abbrev Number: 5 (DW_TAG_member)
$
So the natural alignment for 'struct ceph_entity_addr' ends up being the
natural alignment for 'struct __kernel_sockaddr_storage', which is 8, but
since 'struct ceph_entity_addr' was marked in the original source code as __packed__,
the compiler added the DW_AT_alignment: 1 to override that.
The heuristic in pahole, so far, took that __attribute__((__aligned__(1)))
literally:
$ pahole -C ceph_entity_addr /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o
struct ceph_entity_addr {
__le32 type; /* 0 4 */
__le32 nonce; /* 4 4 */
struct __kernel_sockaddr_storage in_addr __attribute__((__aligned__(1))); /* 8 128 */
/* size: 136, cachelines: 3, members: 3 */
/* forced alignments: 1 */
/* last cacheline: 8 bytes */
} __attribute__((__aligned__(1)));
$
which ends up making the regenerated source code (with the __aligned__((1))),
generate a different layout, the __aligned__((8)) in one of its members
overrode that __aligned__((1)).
Take this into account and when faced with a structure which natural alignment
is not one and that has a DW_AT_alignment:1 to mean it really is __packed__.
Doing that makes the regenerated source code match the original structure
layouts, i.e. after the patch we get:
$ pahole -C ceph_entity_addr /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o
struct ceph_entity_addr {
__le32 type; /* 0 4 */
__le32 nonce; /* 4 4 */
struct __kernel_sockaddr_storage in_addr __attribute__((__aligned__(1))); /* 8 128 */
/* size: 136, cachelines: 3, members: 3 */
/* forced alignments: 1 */
/* last cacheline: 8 bytes */
} __attribute__((__packed__));
$
And that member in 'struct ceph_connection', in the original, continues to read:
$ pahole -C ceph_connection /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o | grep -w actual_peer_addr -B4 -A6
char in_banner[30]; /* 472 30 */
struct ceph_msg_connect out_connect; /* 502 33 */
/* --- cacheline 8 boundary (512 bytes) was 23 bytes ago --- */
struct ceph_msg_connect_reply in_reply; /* 535 26 */
struct ceph_entity_addr actual_peer_addr __attribute__((__aligned__(1))); /* 561 136 */
/* --- cacheline 10 boundary (640 bytes) was 57 bytes ago --- */
struct ceph_msg_header out_hdr; /* 697 53 */
/* XXX 2 bytes hole, try to pack */
/* --- cacheline 11 boundary (704 bytes) was 48 bytes ago --- */
$
While in the regenerated DWARF from the regenerated source code reads:
$ pfunct --compile /home/acme/git/build/v5.1-rc4+/fs/ceph/super.o > ceph.c
$ gcc -g -c ceph.c
$ pahole -C ceph_connection ceph.o | grep -w actual_peer_addr -B4 -A6
char in_banner[30]; /* 472 30 */
struct ceph_msg_connect out_connect; /* 502 33 */
/* --- cacheline 8 boundary (512 bytes) was 23 bytes ago --- */
struct ceph_msg_connect_reply in_reply; /* 535 26 */
struct ceph_entity_addr actual_peer_addr __attribute__((__aligned__(1))); /* 561 136 */
/* --- cacheline 10 boundary (640 bytes) was 57 bytes ago --- */
struct ceph_msg_header out_hdr; /* 697 53 */
/* XXX 2 bytes hole, try to pack */
/* --- cacheline 11 boundary (704 bytes) was 48 bytes ago --- */
$
I.e. it now matches.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
f78633cfb9 |
core: Infer __packed__ for union struct members
I.e. check that all the structs that are embedded in a union have their natural
alignment satisfied by the size of the array they are contained in,
Before this change we ended up not marking union struct members that didn't had
natural alignment violations as __packed__ even tho they had to be to be in a
struct that didn't satisfied their natural alignment requirements, which would
violate them when said union was in an array, i.e. the second entry would have
the non __packed__ union struct member in a bad position.
E.g. Before:
$ pahole -C ceph_osd_op /home/acme/git/build/v5.1-rc4+/net/ceph/osd_client.o
struct ceph_osd_op {
__le16 op; /* 0 2 */
__le32 flags; /* 2 4 */
union {
struct {
__le64 offset; /* 6 8 */
__le64 length; /* 14 8 */
__le64 truncate_size; /* 22 8 */
__le32 truncate_seq; /* 30 4 */
} __attribute__((__packed__)) extent; /* 6 28 */
struct {
__le32 name_len; /* 6 4 */
__le32 value_len; /* 10 4 */
__u8 cmp_op; /* 14 1 */
__u8 cmp_mode; /* 15 1 */
} __attribute__((__packed__)) xattr; /* 6 10 */
struct {
__u8 class_len; /* 6 1 */
__u8 method_len; /* 7 1 */
__u8 argc; /* 8 1 */
__le32 indata_len; /* 9 4 */
} __attribute__((__packed__)) cls; /* 6 7 */
struct {
__le64 cookie; /* 6 8 */
__le64 count; /* 14 8 */
} pgls; /* 6 16 */
struct {
__le64 snapid; /* 6 8 */
} snap; /* 6 8 */
struct {
__le64 cookie; /* 6 8 */
__le64 ver; /* 14 8 */
__u8 op; /* 22 1 */
__le32 gen; /* 23 4 */
} __attribute__((__packed__)) watch; /* 6 21 */
struct {
__le64 cookie; /* 6 8 */
} notify; /* 6 8 */
struct {
__le64 offset; /* 6 8 */
__le64 length; /* 14 8 */
__le64 src_offset; /* 22 8 */
} clonerange; /* 6 24 */
struct {
__le64 expected_object_size; /* 6 8 */
__le64 expected_write_size; /* 14 8 */
} alloc_hint; /* 6 16 */
struct {
__le64 snapid; /* 6 8 */
__le64 src_version; /* 14 8 */
__u8 flags; /* 22 1 */
__le32 src_fadvise_flags; /* 23 4 */
} __attribute__((__packed__)) copy_from; /* 6 21 */
}; /* 6 28 */
__le32 payload_len; /* 34 4 */
/* size: 38, cachelines: 1, members: 4 */
/* last cacheline: 38 bytes */
} __attribute__((__packed__));
After:
$ pahole -C ceph_osd_op /home/acme/git/build/v5.1-rc4+/net/ceph/osd_client.o
struct ceph_osd_op {
__le16 op; /* 0 2 */
__le32 flags; /* 2 4 */
union {
struct {
__le64 offset; /* 6 8 */
__le64 length; /* 14 8 */
__le64 truncate_size; /* 22 8 */
__le32 truncate_seq; /* 30 4 */
} __attribute__((__packed__)) extent; /* 6 28 */
struct {
__le32 name_len; /* 6 4 */
__le32 value_len; /* 10 4 */
__u8 cmp_op; /* 14 1 */
__u8 cmp_mode; /* 15 1 */
} __attribute__((__packed__)) xattr; /* 6 10 */
struct {
__u8 class_len; /* 6 1 */
__u8 method_len; /* 7 1 */
__u8 argc; /* 8 1 */
__le32 indata_len; /* 9 4 */
} __attribute__((__packed__)) cls; /* 6 7 */
struct {
__le64 cookie; /* 6 8 */
__le64 count; /* 14 8 */
} pgls; /* 6 16 */
struct {
__le64 snapid; /* 6 8 */
} snap; /* 6 8 */
struct {
__le64 cookie; /* 6 8 */
__le64 ver; /* 14 8 */
__u8 op; /* 22 1 */
__le32 gen; /* 23 4 */
} __attribute__((__packed__)) watch; /* 6 21 */
struct {
__le64 cookie; /* 6 8 */
} notify; /* 6 8 */
struct {
__le64 offset; /* 6 8 */
__le64 length; /* 14 8 */
__le64 src_offset; /* 22 8 */
} clonerange; /* 6 24 */
struct {
__le64 expected_object_size; /* 6 8 */
__le64 expected_write_size; /* 14 8 */
} alloc_hint; /* 6 16 */
struct {
__le64 snapid; /* 6 8 */
__le64 src_version; /* 14 8 */
__u8 flags; /* 22 1 */
__le32 src_fadvise_flags; /* 23 4 */
} __attribute__((__packed__)) copy_from; /* 6 21 */
}; /* 6 28 */
__le32 payload_len; /* 34 4 */
/* size: 38, cachelines: 1, members: 4 */
/* last cacheline: 38 bytes */
} __attribute__((__packed__));
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
1bb4527220 |
fprintf: Fixup const pointers
Before:
$ pahole -C nft_ctx /home/acme/git/build/v5.1-rc4+/net/netfilter/nft_set_rbtree.o
struct nft_ctx {
struct net * net; /* 0 8 */
struct nft_table * table; /* 8 8 */
struct nft_chain * chain; /* 16 8 */
const const struct nlattr * * nla; /* 24 8 */
u32 portid; /* 32 4 */
u32 seq; /* 36 4 */
u8 family; /* 40 1 */
u8 level; /* 41 1 */
bool report; /* 42 1 */
/* size: 48, cachelines: 1, members: 9 */
/* padding: 5 */
/* last cacheline: 48 bytes */
};
$
Original:
struct nft_ctx {
struct net *net;
struct nft_table *table;
struct nft_chain *chain;
const struct nlattr * const *nla;
u32 portid;
u32 seq;
u8 family;
u8 level;
bool report;
};
DWARF tags:
<1><12c8a>: Abbrev Number: 12 (DW_TAG_structure_type)
<12c8b> DW_AT_name : (indirect string, offset: 0xcc6f): nlattr
<12c8f> DW_AT_byte_size : 4
<12c93> DW_AT_sibling : <0x12cb2>
<1><12cb2>: Abbrev Number: 17 (DW_TAG_const_type)
<12cb3> DW_AT_type : <0x12c8a>
<1><12cf9>: Abbrev Number: 4 (DW_TAG_pointer_type)
<12cfa> DW_AT_byte_size : 8
<12cfb> DW_AT_type : <0x12cb2>
<1><12cff>: Abbrev Number: 17 (DW_TAG_const_type)
<12d00> DW_AT_type : <0x12cf9>
<1><1d54b>: Abbrev Number: 4 (DW_TAG_pointer_type)
<1d54c> DW_AT_byte_size : 8
<1d54d> DW_AT_type : <0x12cff>
<2><1e52e>: Abbrev Number: 14 (DW_TAG_member)
<1e52f> DW_AT_name : nla
<1e536> DW_AT_type : <0x1d54b>
<1e53a> DW_AT_data_member_location: 24
Fixed now:
$ pahole -C nft_ctx /home/acme/git/build/v5.1-rc4+/net/netfilter/nft_set_rbtree.o
struct nft_ctx {
struct net * net; /* 0 8 */
struct nft_table * table; /* 8 8 */
struct nft_chain * chain; /* 16 8 */
const struct nlattr * const * nla; /* 24 8 */
u32 portid; /* 32 4 */
u32 seq; /* 36 4 */
u8 family; /* 40 1 */
u8 level; /* 41 1 */
bool report; /* 42 1 */
/* size: 48, cachelines: 1, members: 9 */
/* padding: 5 */
/* last cacheline: 48 bytes */
};
$
So, one more full circled:
$ fullcircle /home/acme/git/build/v5.1-rc4+/net/netfilter/nft_set_rbtree.o
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
85c9936963 |
fprintf: Do not add explicit padding when struct has __aligned__ attr
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
28a3bc7add |
fprintf: Support packed enums
Check if the size is different than sizeof(int), which should be good enough for now for both 64-bit and 32-bit targets. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
f77a442f09 |
fprintf: Do not print the __aligned__ attribute if asked
I.e. honour conf_fprintf.suppress_aligned_attribute, noticed with
btfdiff, as BTF doesn't carries the alignment attribute, so can't
regenerate it, we need to suppress it so as to compare the output of
DWARF with that of the equivalent BTF.
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
ea583dac52 |
fprintf: Print zero sized flat arrays as [], not [0]
To match the case when we really have just one dimension, so the --flat-arrays should show for zero sized arrays, [], not [0]: Noticed with btfdiff. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
f909f13dd7 |
fprintf: Fixup handling of unnamed bitfields
We were only handling holes inside bitfields as a request to change the
byte_offset, which is not the case when instead of 'int foo:0;' we have
'int foo:6;' to ask for a explicit 6 bit hole inside a bitfield, like
in:
Before this patch:
$ pahole -F btf -C kvm_mmu_page_role /home/acme/git/build/v5.1-rc4+/arch/x86/kvm/hyperv.o
union kvm_mmu_page_role {
u32 word; /* 0 4 */
struct {
unsigned int level:4; /* 0: 0 4 */
unsigned int gpte_is_8_bytes:1; /* 0: 4 4 */
unsigned int quadrant:2; /* 0: 5 4 */
unsigned int direct:1; /* 0: 7 4 */
unsigned int access:3; /* 0: 8 4 */
unsigned int invalid:1; /* 0:11 4 */
unsigned int nxe:1; /* 0:12 4 */
unsigned int cr0_wp:1; /* 0:13 4 */
unsigned int smep_andnot_wp:1; /* 0:14 4 */
unsigned int smap_andnot_wp:1; /* 0:15 4 */
unsigned int ad_disabled:1; /* 0:16 4 */
unsigned int guest_mode:1; /* 0:17 4 */
/* XXX 6 bits hole, try to pack */
/* Force alignment to the next boundary: */
unsigned int :0;
unsigned int smm:8; /* 0:24 4 */
}; /* 0 4 */
};
$
After:
$ pahole -F btf -C kvm_mmu_page_role /home/acme/git/build/v5.1-rc4+/arch/x86/kvm/hyperv.o
union kvm_mmu_page_role {
u32 word; /* 0 4 */
struct {
unsigned int level:4; /* 0: 0 4 */
unsigned int gpte_is_8_bytes:1; /* 0: 4 4 */
unsigned int quadrant:2; /* 0: 5 4 */
unsigned int direct:1; /* 0: 7 4 */
unsigned int access:3; /* 0: 8 4 */
unsigned int invalid:1; /* 0:11 4 */
unsigned int nxe:1; /* 0:12 4 */
unsigned int cr0_wp:1; /* 0:13 4 */
unsigned int smep_andnot_wp:1; /* 0:14 4 */
unsigned int smap_andnot_wp:1; /* 0:15 4 */
unsigned int ad_disabled:1; /* 0:16 4 */
unsigned int guest_mode:1; /* 0:17 4 */
/* XXX 6 bits hole, try to pack */
unsigned int :6;
unsigned int smm:8; /* 0:24 4 */
}; /* 0 4 */
};
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Andrii Nakryiko <andriin@fb.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Yonghong Song <yhs@fb.com>
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
3247a777dc |
core: Infer if a struct is packed by the offsets/natural alignments
As DWARF (nor BTF) provides explicit attributes, we need to look at the
natural alignments, a byte is always alignted, etc.
This probably fails with things like __attribute__(__aligned(power-of-two)),
but with it most of the kernel data structures are full circled, i.e.
'pfunct --compile' regenerates source code from debug info that when
compiled generats debug info that end up matching the original sources.
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
const char * uptr;
int refcnt;
};
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
const char * uptr; /* 8 8 */
int refcnt; /* 16 4 */
/* size: 24, cachelines: 1, members: 3 */
/* padding: 4 */
/* last cacheline: 24 bytes */
};
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
const char * uptr;
int refcnt;
} __packed;
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
const char * uptr; /* 8 8 */
int refcnt; /* 16 4 */
/* size: 20, cachelines: 1, members: 3 */
/* last cacheline: 20 bytes */
} __attribute__((__packed__));
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
int refcnt;
const char * uptr;
};
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
int refcnt; /* 8 4 */
/* XXX 4 bytes hole, try to pack */
const char * uptr; /* 16 8 */
/* size: 24, cachelines: 1, members: 3 */
/* sum members: 20, holes: 1, sum holes: 4 */
/* last cacheline: 24 bytes */
};
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
int refcnt;
const char * uptr;
} __packed;
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
int refcnt; /* 8 4 */
const char * uptr; /* 12 8 */
/* size: 20, cachelines: 1, members: 3 */
/* last cacheline: 20 bytes */
} __attribute__((__packed__));
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
const char * uptr;
unsigned char refcnt;
};
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
const char * uptr; /* 8 8 */
unsigned char refcnt; /* 16 1 */
/* size: 24, cachelines: 1, members: 3 */
/* padding: 7 */
/* last cacheline: 24 bytes */
};
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
const char * uptr;
unsigned char refcnt;
} __packed;
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
const char * uptr; /* 8 8 */
unsigned char refcnt; /* 16 1 */
/* size: 17, cachelines: 1, members: 3 */
/* last cacheline: 17 bytes */
} __attribute__((__packed__));
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
unsigned char refcnt;
const char * uptr;
};
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
unsigned char refcnt; /* 8 1 */
/* XXX 7 bytes hole, try to pack */
const char * uptr; /* 16 8 */
/* size: 24, cachelines: 1, members: 3 */
/* sum members: 17, holes: 1, sum holes: 7 */
/* last cacheline: 24 bytes */
};
$ cat a.c
#define __packed __attribute__((__packed__))
struct filename {
const char * name;
unsigned char refcnt;
const char * uptr;
} __packed;
void m(struct filename *f) {}
$ gcc -g -c a.c
$ pahole a.o
struct filename {
const char * name; /* 0 8 */
unsigned char refcnt; /* 8 1 */
const char * uptr; /* 9 8 */
/* size: 17, cachelines: 1, members: 3 */
/* last cacheline: 17 bytes */
} __attribute__((__packed__));
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
13e5b9fc00 |
fprintf: Add unnamed bitfield padding at the end to rebuild original type
Just like the 'struct timex' in the linux kernel UAPI, that is now
correctly reconstructed as:
$ pahole -IC timex /home/acme/git/build/v5.0-rc2+/kernel/time/posix-clock.o | tail -32
__kernel_long_t ppsfreq; /* 96 8 */
__kernel_long_t jitter; /* 104 8 */
int shift; /* 112 4 */
/* XXX 4 bytes hole, try to pack */
__kernel_long_t stabil; /* 120 8 */
/* --- cacheline 2 boundary (128 bytes) --- */
__kernel_long_t jitcnt; /* 128 8 */
__kernel_long_t calcnt; /* 136 8 */
__kernel_long_t errcnt; /* 144 8 */
__kernel_long_t stbcnt; /* 152 8 */
int tai; /* 160 4 */
/* Force padding: */
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
int :32;
/* size: 208, cachelines: 4, members: 20 */
/* sum members: 152, holes: 3, sum holes: 12 */
/* padding: 44 */
/* last cacheline: 16 bytes */
};
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
ccd67bdb20 |
fprintf: Print "const" for class members more early, in type__fprintf()
We want to reach array__fprintf() from here, with the class_member
name, as __tag__name() isn't handling arrays properly.
I.e. to print an array when we have its name we can't use __tag__name().
This also stops printing 0 for zero sized arrays and trows away the
extra DW_TAG_const_type that comes with zero sized arrays, where we
have:
class_member type: DW_TAG_const_type 1
DW_TAG_const_type 1: DW_TAG_array_type 2
DW_TAG_array_type 2: 0 entries, type: DW_TAG_const_type 3
DW_TAG_const_type 3: real type of the zero sized array
For instance, after this patch we get a sane reconstruction of this
type:
$ pahole -C filename /home/acme/git/build/v5.0-rc2+/ipc/mqueue.o
struct filename {
const char * name; /* 0 8 */
const char * uptr; /* 8 8 */
int refcnt; /* 16 4 */
/* XXX 4 bytes hole, try to pack */
struct audit_names * aname; /* 24 8 */
const char iname[]; /* 32 0 */
/* size: 32, cachelines: 1, members: 5 */
/* sum members: 28, holes: 1, sum holes: 4 */
/* last cacheline: 32 bytes */
};
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
b42d77b0bb |
fprintf: Print __attribute__((__aligned__(N))) for structs/classes
For instance:
$ pahole -C kern_ipc_perm /home/acme/git/build/v5.0-rc2+/ipc/util.o
struct kern_ipc_perm {
spinlock_t lock; /* 0 4 */
bool deleted; /* 4 1 */
/* XXX 3 bytes hole, try to pack */
int id; /* 8 4 */
key_t key; /* 12 4 */
kuid_t uid; /* 16 4 */
kgid_t gid; /* 20 4 */
kuid_t cuid; /* 24 4 */
kgid_t cgid; /* 28 4 */
umode_t mode; /* 32 2 */
/* XXX 6 bytes hole, try to pack */
long unsigned int seq; /* 40 8 */
void * security; /* 48 8 */
struct rhash_head khtnode; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
struct callback_head rcu __attribute__((__aligned__(8))); /* 64 16 */
refcount_t refcount; /* 80 4 */
/* size: 128, cachelines: 2, members: 14 */
/* sum members: 75, holes: 2, sum holes: 9 */
/* padding: 44 */
/* forced alignments: 1 */
} __attribute__((__aligned__(64)));
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
a104eb1ea1 |
fprintf: Notice explicit bitfield alignment modifications
I.e. when we find that the last member has a bit_hole, i.e. it is part
of a bitfield, and the current field has a bitfield_size, i.e. it _also_
is part of a bitfield, the only explanation is that they were
artificially put in different base types, i.e. like in these fields
in the linux kernel 'struct task_struct', here reconstructed by pahole:
$ pahole -C task_struct ~/git/build/v5.1-rc2+/kernel/sched/core.o | grep :0 -B9 -A12
unsigned int personality; /* 1128 4 */
unsigned int sched_reset_on_fork:1; /* 1132: 0 4 */
unsigned int sched_contributes_to_load:1; /* 1132: 1 4 */
unsigned int sched_migrated:1; /* 1132: 2 4 */
unsigned int sched_remote_wakeup:1; /* 1132: 3 4 */
/* XXX 28 bits hole, try to pack */
/* Force alignment to the next boundary: */
unsigned int :0;
unsigned int in_execve:1; /* 1136: 0 4 */
unsigned int in_iowait:1; /* 1136: 1 4 */
unsigned int restore_sigmask:1; /* 1136: 2 4 */
unsigned int in_user_fault:1; /* 1136: 3 4 */
unsigned int no_cgroup_migration:1; /* 1136: 4 4 */
unsigned int use_memdelay:1; /* 1136: 5 4 */
/* XXX 26 bits hole, try to pack */
/* XXX 4 bytes hole, try to pack */
long unsigned int atomic_flags; /* 1144 8 */
$
This matches the original definition in the original kernel sources, and
further more, the following sequence proves that with this and DW_AT_alignment,
we can go full circle, i.e.:
1. from an object file reconstruct the source code for all the types that
appears in function signatures, if pointers, them they will be fully defined,
not just forward declared:
$ pfunct --compile=sched_change_group ~/git/build/v5.1-rc2+/kernel/sched/core.o | egrep -w 'sched_change_group|task_struct {' -B10 -A5
/* --- cacheline 3 boundary (192 bytes) --- */
struct fpu fpu __attribute__((__aligned__(64))); /* 192 4160 */
/* size: 4352, cachelines: 68, members: 21 */
/* sum members: 4316, holes: 2, sum holes: 32 */
/* sum bitfield members: 2 bits, bit holes: 1, sum bit holes: 30 bits */
/* forced alignments: 1, forced holes: 1, sum forced holes: 28 */
};
struct task_struct {
struct thread_info thread_info; /* 0 16 */
/* XXX last struct has 4 bytes of padding */
volatile long int state; /* 16 8 */
--
/* --- cacheline 104 boundary (6656 bytes) --- */
struct thread_struct thread __attribute__((__aligned__(64))); /* 6656 4352 */
/* size: 11008, cachelines: 172, members: 207 */
/* sum members: 10902, holes: 16, sum holes: 98 */
/* sum bitfield members: 10 bits, bit holes: 2, sum bit holes: 54 bits */
/* paddings: 3, sum paddings: 14 */
/* forced alignments: 6, forced holes: 1, sum forced holes: 40 */
};
void sched_change_group(struct task_struct * tsk, int type)
{
}
$
2. Build the regenerated skeleton function + its types:
$ pfunct --compile=sched_change_group ~/git/build/v5.1-rc2+/kernel/sched/core.o > sched_change_group.c
$ gcc -g -c sched_change_group.c
$ file sched_change_group.o
sched_change_group.o: ELF 64-bit LSB relocatable, x86-64, version 1 (SYSV), with debug_info, not stripped
$
3. Now lets see if the original 'struct task_struct' printed by pahole, matches
the the output printed by pahole for the DWARF info generated for the regenerated
'struct task_struct' source code in sched_change_group.c:
$ pahole -C task_struct sched_change_group.o | tail
/* --- cacheline 104 boundary (6656 bytes) --- */
struct thread_struct thread __attribute__((__aligned__(64))); /* 6656 4352 */
/* size: 11008, cachelines: 172, members: 207 */
/* sum members: 10902, holes: 16, sum holes: 98 */
/* sum bitfield members: 10 bits, bit holes: 2, sum bit holes: 54 bits */
/* paddings: 3, sum paddings: 14 */
/* forced alignments: 6, forced holes: 1, sum forced holes: 40 */
};
$ pahole -C task_struct ~/git/build/v5.1-rc2+/kernel/sched/core.o | tail
/* --- cacheline 104 boundary (6656 bytes) --- */
struct thread_struct thread __attribute__((__aligned__(64))); /* 6656 4352 */
/* size: 11008, cachelines: 172, members: 207 */
/* sum members: 10902, holes: 16, sum holes: 98 */
/* sum bitfield members: 10 bits, bit holes: 2, sum bit holes: 54 bits */
/* paddings: 3, sum paddings: 14 */
/* forced alignments: 6, forced holes: 1, sum forced holes: 40 */
};
$
Furthermore:
$ pahole -C task_struct ~/git/build/v5.1-rc2+/kernel/sched/core.o > /tmp/original
$ pahole -C task_struct sched_change_group.o > /tmp/regenerated
$ diff -u /tmp/original /tmp/regenerated
$
So one of the most complex data structures in the Linux kernel seems to be under control,
and it uses zero sized unnamed bitfields and __attribute__((aligned(N))), a DWARF5 goodie,
time to go tag v1.13!
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Andrii Nakryiko <andriin@fb.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Yonghong Song <yhs@fb.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
45ad545944 |
tag: Introduce tag__is_pointer()
For the usual idiom to ask if a tag is a pointer, removing a bit of DWARFism and shortening the operation. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
ce6f393bc9 |
fprintf: Fixup the printing of const parameters
The last problem with 'pfunct --compile' at least for tcp.o:
Before:
$ pfunct --compile examples/tcp.o > tcp.pahole.c
$ gcc -c tcp.pahole.c -g
tcp.pahole.c:1808:48: error: unknown type name ‘u8const’; did you mean ‘const’?
inline void tcp_set_ca_state(struct sock * sk, u8const ca_state)
^~~~~~~
const
tcp.pahole.c:5346:56: error: unknown type name ‘intconst’; did you mean ‘const’?
inline void skb_set_tail_pointer(struct sk_buff * skb, intconst offset)
^~~~~~~~
const
tcp.pahole.c:5914:37: error: unknown type name ‘gfp_tconst’; did you mean ‘gfp_t’?
inline bool gfpflags_allow_blocking(gfp_tconst gfp_flags)
^~~~~~~~~~
gfp_t
tcp.pahole.c:5926:24: error: unknown type name ‘ktime_tconst’; did you mean ‘ktime_t’?
inline s64 ktime_to_ns(ktime_tconst kt)
^~~~~~~~~~~~
ktime_t
tcp.pahole.c:5939:54: warning: ‘struct timespec64const’ declared inside parameter list will not be visible outside of this definition or declaration
inline struct timespec timespec64_to_timespec(struct timespec64const ts64)
^~~~~~~~~~~~~~~
tcp.pahole.c:5939:70: error: parameter 1 (‘ts64’) has incomplete type
inline struct timespec timespec64_to_timespec(struct timespec64const ts64)
~~~~~~~~~~~~~~~~~~~~~~~^~~~
$
After:
$ pfunct --compile examples/tcp.o > tcp.pahole.c
$ gcc -c tcp.pahole.c -g
Because:
$ grep -A2 tcp_set_ca_state tcp.pahole.c
inline void tcp_set_ca_state(struct sock * sk, const u8 ca_state)
{
}
$
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
a7d9c58cb8 |
fprintf: Add missing closing parens to the align attribute
Noticed while trying to use pfunct's -b option, that will show a
function prototype + the types it uses in its function signature, i.e.:
$ pfunct -b -f tcp_sendmsg tcp.o
typedef long long unsigned int __u64;
typedef __u64 __addrpair;
typedef unsigned int __u32;
typedef __u32 __be32;
typedef short unsigned int __u16;
typedef __u32 __portpair;
typedef __u16 __be16;
struct hlist_node {
struct hlist_node * next; /* 0 8 */
struct hlist_node * * pprev; /* 8 8 */
/* size: 16, cachelines: 1, members: 2 */
/* last cacheline: 16 bytes */
};
<SNIP tons of types>
struct sock {
struct sock_common __sk_common; /* 0 136 */
/* --- cacheline 2 boundary (128 bytes) was 8 bytes ago --- */
socket_lock_t sk_lock; /* 136 32 */
atomic_t sk_drops; /* 168 4 */
/* --- cacheline 10 boundary (640 bytes) --- */
<SNIP the rest of the 'struct sock' members>
struct sock_cgroup_data sk_cgrp_data; /* 640 8 */
struct mem_cgroup * sk_memcg; /* 648 8 */
void (*sk_state_change)(struct sock *); /* 656 8 */
void (*sk_data_ready)(struct sock *); /* 664 8 */
void (*sk_write_space)(struct sock *); /* 672 8 */
void (*sk_error_report)(struct sock *); /* 680 8 */
int (*sk_backlog_rcv)(struct sock *, struct sk_buff *); /* 688 8 */
void (*sk_destruct)(struct sock *); /* 696 8 */
/* --- cacheline 11 boundary (704 bytes) --- */
struct sock_reuseport * sk_reuseport_cb; /* 704 8 */
struct callback_head sk_rcu __attribute__((__aligned__(8))); /* 712 16 */
/* size: 728, cachelines: 12, members: 84 */
/* sum members: 715, holes: 4, sum holes: 8 */
/* sum bitfield members: 40 bits (5 bytes) */
/* paddings: 1, sum paddings: 4 */
/* forced alignments: 1 */
/* last cacheline: 24 bytes */
};
<SNIP some more types>
struct kiocb;
struct msghdr {
void * msg_name; /* 0 8 */
int msg_namelen; /* 8 4 */
/* XXX 4 bytes hole, try to pack */
struct iov_iter msg_iter; /* 16 40 */
void * msg_control; /* 56 8 */
/* --- cacheline 1 boundary (64 bytes) --- */
__kernel_size_t msg_controllen; /* 64 8 */
unsigned int msg_flags; /* 72 4 */
/* XXX 4 bytes hole, try to pack */
struct kiocb * msg_iocb; /* 80 8 */
/* size: 88, cachelines: 2, members: 7 */
/* sum members: 80, holes: 2, sum holes: 8 */
/* last cacheline: 24 bytes */
};
typedef __kernel_size_t size_t;
int tcp_sendmsg(struct sock * sk, struct msghdr * msg, size_t size);
$
So if we then redirect the output to a file and if we make it a empty
function instead of a prototype, i.e. if we make the last line above to
become this:
int tcp_sendmsg(struct sock * sk, struct msghdr * msg, size_t size) {}
then build with gcc -g to have it build as a .o with DWARF info, then we
should be able to see if the struct rebuilt from DWARF matches the
original struct used to generate the DWARF, going full circle:
$ pfunct -b -f tcp_sendmsg tcp.o > tcp_sendmsg_types.c
$ gcc -c tcp_sendmsg_types.c -g
$ file tcp_sendmsg_types.o
tcp_sendmsg_types.o: ELF 64-bit LSB relocatable, x86-64, version 1 (SYSV), with debug_info, not stripped
$ pahole -E -C sock tcp_sendmsg_types.o > tcp_sendmsg_types.o.pahole
$ pahole -E -C sock tcp.o > tcp.o.pahole
$ diff -u tcp_sendmsg_types.o.pahole tcp.o.pahole
$ wc -l tcp_sendmsg_types.o.pahole
420 tcp_sendmsg_types.o.pahole
$
So all the types that come from sock are expanded and all its details
are reconstructed in the same way for both cases.
$ pahole -C sock tcp.o | tail
struct sock_reuseport * sk_reuseport_cb; /* 704 8 */
struct callback_head sk_rcu __attribute__((__aligned__(8))); /* 712 16 */
/* size: 728, cachelines: 12, members: 84 */
/* sum members: 715, holes: 4, sum holes: 8 */
/* sum bitfield members: 40 bits (5 bytes) */
/* paddings: 1, sum paddings: 4 */
/* forced alignments: 1 */
/* last cacheline: 24 bytes */
};
$ pahole -C sock tcp_sendmsg_types.o | tail
struct sock_reuseport * sk_reuseport_cb; /* 704 8 */
struct callback_head sk_rcu __attribute__((__aligned__(8))); /* 712 16 */
/* size: 728, cachelines: 12, members: 84 */
/* sum members: 715, holes: 4, sum holes: 8 */
/* sum bitfield members: 40 bits (5 bytes) */
/* paddings: 1, sum paddings: 4 */
/* forced alignments: 1 */
/* last cacheline: 24 bytes */
};
$
Reported-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andrii Nakryiko <andriin@fb.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
b95961db69 |
fprintf: Show statistics about holes due to forced alignments
$ pahole -C task_struct | tail
/* --- cacheline 104 boundary (6656 bytes) --- */
struct thread_struct thread __attribute__((__aligned__(64)); /* 6656 4352 */
/* size: 11008, cachelines: 172, members: 207 */
/* sum members: 10902, holes: 16, sum holes: 98 */
/* sum bitfield members: 10 bits, bit holes: 2, sum bit holes: 54 bits */
/* paddings: 3, sum paddings: 14 */
/* forced alignments: 6, forced holes: 1, sum forced holes: 40 */
};
$ pahole -C inet_timewait_death_row
struct inet_timewait_death_row {
atomic_t tw_count; /* 0 4 */
/* XXX 60 bytes hole, try to pack */
/* --- cacheline 1 boundary (64 bytes) --- */
struct inet_hashinfo * hashinfo __attribute__((__aligned__(64)); /* 64 8 */
int sysctl_max_tw_buckets; /* 72 4 */
/* size: 128, cachelines: 2, members: 3 */
/* sum members: 16, holes: 1, sum holes: 60 */
/* padding: 52 */
/* forced alignments: 1, forced holes: 1, sum forced holes: 60 */
};
$
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Andrii Nakryiko <andriin@fb.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Wielaard <mark@klomp.org>
Cc: Yonghong Song <yhs@fb.com>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
ec772f21f6 |
fprintf: Show the number of forced alignments in a class
For instance, for task_struct: $ pahole -C task_struct | tail /* --- cacheline 104 boundary (6656 bytes) --- */ struct thread_struct thread __attribute__((__aligned__(64)); /* 6656 4352 */ /* size: 11008, cachelines: 172, members: 207 */ /* sum members: 10902, holes: 16, sum holes: 98 */ /* sum bitfield members: 10 bits, bit holes: 2, sum bit holes: 54 bits */ /* paddings: 3, sum paddings: 14 */ /* forced alignments: 6 */ }; $ Cc: Alexei Starovoitov <ast@fb.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Wielaard <mark@klomp.org> Cc: Yonghong Song <yhs@fb.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
6cd6a6bd87 |
dwarves_fprintf: Allow suppressing the __attribute__((__aligned__(N))
So that we can use it in things like btfdiff. Cc: Alexei Starovoitov <ast@fb.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Mark Wielaard <mark@klomp.org> Cc: Yonghong Song <yhs@fb.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
f31ea292e3 |
dwarf_loader: Store the DW_AT_alignment if available
DWARF got a DW_AT_alignment as described in: http://dwarfstd.org/ShowIssue.php?issue=140528.1 This appeared first in DWARF5: http://dwarfstd.org/doc/DWARF5.pdf In: ---------------------------------------------------------------------- Chapter 2. General Description 2.24 Alignment A debugging information entry may have a DW_AT_alignment attribute whose value of class constant is a positive, non-zero, integer describing the alignment of the entity. For example, an alignment attribute whose value is 8 indicates that the entity to which it applies occurs at an address that is a multiple of eight (not a multiple of 8 or 256) ---------------------------------------------------------------------- Use it on a struct present in the running kernel, i.e. not specifying which ELF file to look for the DWARF info to use: $ pahole -C inet_timewait_death_row struct inet_timewait_death_row { atomic_t tw_count; /* 0 4 */ /* XXX 60 bytes hole, try to pack */ /* --- cacheline 1 boundary (64 bytes) --- */ struct inet_hashinfo * hashinfo __attribute__((__aligned__(64)); /* 64 8 */ int sysctl_max_tw_buckets; /* 72 4 */ /* size: 128, cachelines: 2, members: 3 */ /* sum members: 16, holes: 1, sum holes: 60 */ /* padding: 52 */ }; $ Now to do some tweaking to get that "__attribute..." part nicely, hum, aligned in the pahole output :-) BTW: the original struct is in the kernel sources: include/net/netns/ipv4.h struct inet_timewait_death_row { atomic_t tw_count; struct inet_hashinfo *hashinfo ____cacheline_aligned_in_smp; int sysctl_max_tw_buckets; }; Reported-by: Mark Wielaard <mark@klomp.org> Cc: Alexei Starovoitov <ast@fb.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
c002873c44 |
dwarves_fprintf: Move invariant printing of ; to outside if block
Will facilitate printing something just before the ; Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Andrii Nakryiko
|
78c110a7ea |
dwarves: Revert semantics of member bit/byte hole
pahole --reorganize heavily depends on member's bit_hole and hole fields to denote bit/byte holes *after* member. Previous commit "dwarves: use bit sizes and bit/byte hole info in __class__fprintf" changed its meaning to bit/byte hole *before* member to accomodate possible bit/byte holes at the beginning of a struct. This change broke reorganization algorithm, though, which is quite involved and isn't trivially modifiable to accomodate new semantics. This patch reverts the meaning of bit_hole and hole, but also introduces per class pre_bit_hole/pre_hole to record initial bit/byte hole of a struct. This allows to fix reorg code more easily and still handle initial holes cases, if at the expense of being not as elegant. Committer testing: $ time pahole -F btf --packable vmlinux | sort -nr -k4 | head bts_ctx 12288 8192 4096 swsusp_info 4096 432 3664 vc_data 792 496 296 pci_dev 2488 2320 168 rcu_state 3392 3240 152 ptr_ring 192 40 152 xdp_sock 960 840 120 zone 1664 1552 112 rcu_data 576 472 104 rcu_node 576 480 96 real 0m0.038s user 0m0.029s sys 0m0.017s $ Signed-off-by: Andrii Nakryiko <andriin@fb.com> Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexei Starovoitov <ast@fb.com> Cc: Yonghong Song <yhs@fb.com> Cc: dwarves@vger.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Andrii Nakryiko
|
b56fed297e |
dwarves_fprintf: Count bitfield member sizes separately
Counting field sizes only in bits causes confusion and lots of differing
output, when compared to previous logic. This commit changes logic so
that it counts bit size of bitfield fields separately from byte size of
non-bitfield fields. In the end, if there were bit holes, this bit size
is emitted explicitly. This makes output for struct/unions not using
bitfields identical, while also preserving correctness (and data
completeness) for cases with bitfields and bit holes.
Example (-before/+after):
struct cfg80211_pmsr_request_peer {
u8 addr[6]; /* 0 6 */
/* XXX 2 bytes hole, try to pack */
struct cfg80211_chan_def chandef; /* 8 24 */
/* XXX last struct has 4 bytes of padding */
u8 report_ap_tsf:1; /* 32: 0 1 */
/* XXX 7 bits hole, try to pack */
/* XXX 3 bytes hole, try to pack */
struct cfg80211_pmsr_ftm_request_peer ftm; /* 36 12 */
/* XXX last struct has 1 byte of padding */
/* size: 48, cachelines: 1, members: 4 */
- /* sum members: 43, holes: 2, sum holes: 5 */
- /* bit holes: 1, sum bit holes: 7 bits */
+ /* sum members: 42, holes: 2, sum holes: 5 */
+ /* sum bitfield members: 1 bits, bit holes: 1, sum bit holes: 7 bits */
/* paddings: 2, sum paddings: 5 */
/* last cacheline: 48 bytes */
};
For cases where there is only byte or bit hole, we still emit total byte and
bit sizes of all members as to not mislead user:
struct sched_dl_entity {
... <snip ...
unsigned int dl_non_contending:1; /* 84: 3 4 */
unsigned int dl_overrun:1; /* 84: 4 4 */
/* XXX 27 bits hole, try to pack */
struct hrtimer dl_timer; /* 88 64 */
/* XXX last struct has 5 bytes of padding */
/* --- cacheline 2 boundary (128 bytes) was 24 bytes ago --- */
struct hrtimer inactive_timer; /* 152 64 */
/* XXX last struct has 5 bytes of padding */
/* size: 216, cachelines: 4, members: 16 */
- /* bit holes: 1, sum bit holes: 27 bits */
+ /* sum members: 212 */
+ /* sum bitfield members: 5 bits, bit holes: 1, sum bit holes: 27 bits */
/* paddings: 2, sum paddings: 10 */
/* last cacheline: 24 bytes */
};
For structs with tightly packed bitfield, we emit total number of bits and also
convert them to bytes. E.g., for struct sock output :
struct sock {
... <snip ...
/* size: 720, cachelines: 12, members: 84 */
- /* sum members: 712, holes: 4, sum holes: 8 */
+ /* sum members: 707, holes: 4, sum holes: 8 */
+ /* sum bitfield members: 40 bits (5 bytes) */
/* paddings: 1, sum paddings: 4 */
/* last cacheline: 16 bytes */
};
Suggested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Cc: Alexei Starovoitov <ast@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Cc: dwarves@vger.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|