[Bug c/95818] New: wrong "used uninitialized" warning

[Bug c/95818] New: wrong "used uninitialized" warning

[ferruh.yigit at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

            Bug ID: 95818
           Summary: wrong "used uninitialized" warning
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ferruh.yigit at intel dot com
  Target Milestone: ---

gcc (GCC) 11.0.0 20200621 (experimental)

Sorry, I couldn't able to reproduce with test code, I will copy-paste the real
code that causes the warning hoping it helps.

Warning [1] and code that causes it [2], struct in question [3].

As you can see all the fields of the struct has been set before used, so not
sure why giving used uninitialized warning.



[1]
.../drivers/net/iavf/iavf_ethdev.c: In function ‘iavf_dev_link_update’:
.../drivers/net/iavf/iavf_ethdev.c:641:6: error: ‘new_link’ is used
uninitialized [-Werror=uninitialized]
  641 |  if (rte_atomic64_cmpset((uint64_t *)&dev->data->dev_link,
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  642 |     *(uint64_t *)&dev->data->dev_link,
      |     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  643 |     *(uint64_t *)&new_link) == 0)
      |     ~~~~~~~~~~~~~~~~~~~~~~~
.../drivers/net/iavf/iavf_ethdev.c:596:22: note: ‘new_link’ declared here
  596 |  struct rte_eth_link new_link;
      |                      ^~~~~~~~
cc1: all warnings being treated as error



[2]
iavf_dev_link_update(struct rte_eth_dev *dev,
                    __rte_unused int wait_to_complete)
{
        struct rte_eth_link new_link;
        struct iavf_info *vf = IAVF_DEV_PRIVATE_TO_VF(dev->data->dev_private);

        /* Only read status info stored in VF, and the info is updated
         *  when receive LINK_CHANGE evnet from PF by Virtchnnl.
         */
        switch (vf->link_speed) {
        case 10:
                new_link.link_speed = ETH_SPEED_NUM_10M;
                break;
        case 100:
                new_link.link_speed = ETH_SPEED_NUM_100M;
                break;
        case 1000:
                new_link.link_speed = ETH_SPEED_NUM_1G;
                break;
        case 10000:
                new_link.link_speed = ETH_SPEED_NUM_10G;
                break;
        case 20000:
                new_link.link_speed = ETH_SPEED_NUM_20G;
                break;
        case 25000:
                new_link.link_speed = ETH_SPEED_NUM_25G;
                break;
        case 40000:
                new_link.link_speed = ETH_SPEED_NUM_40G;
                break;
        case 50000:
                new_link.link_speed = ETH_SPEED_NUM_50G;
                break;
        case 100000:
                new_link.link_speed = ETH_SPEED_NUM_100G;
                break;
        default:
                new_link.link_speed = ETH_SPEED_NUM_NONE;
                break;
        }

        new_link.link_duplex = ETH_LINK_FULL_DUPLEX;
        new_link.link_status = vf->link_up ? ETH_LINK_UP :
                                             ETH_LINK_DOWN;
        new_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
                                ETH_LINK_SPEED_FIXED);

        if (rte_atomic64_cmpset((uint64_t *)&dev->data->dev_link,
                                *(uint64_t *)&dev->data->dev_link,
                                *(uint64_t *)&new_link) == 0)
                return -1;

        return 0;
}



[3]
struct rte_eth_link {
        uint32_t link_speed;        /**< ETH_SPEED_NUM_ */
        uint16_t link_duplex  : 1;  /**< ETH_LINK_[HALF/FULL]_DUPLEX */
        uint16_t link_autoneg : 1;  /**< ETH_LINK_[AUTONEG/FIXED] */
        uint16_t link_status  : 1;  /**< ETH_LINK_[DOWN/UP] */
} __rte_aligned(8);      /**< aligned for atomic64 read/write */

[Bug c/95818] wrong "used uninitialized" warning

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |24639
   Last reconfirmed|                            |2020-06-22
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |WAITING
           Keywords|                            |diagnostic

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
please provide complete preprocessed source


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=24639
[Bug 24639] [meta-bug] bug to track all Wuninitialized issues

[Bug c/95818] wrong "used uninitialized" warning

[ferruh.yigit at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #2 from Ferruh YIGIT <ferruh.yigit at intel dot com> ---
.i output [1] and .s output [2] below, please let me know if the request was
something else.

[1]
int
iavf_dev_link_update(struct rte_eth_dev *dev,
      __attribute__((__unused__)) int wait_to_complete)
{
 struct rte_eth_link new_link;
 struct iavf_info *vf = (&((struct iavf_adapter *)dev->data->dev_private)->vf);




 switch (vf->link_speed) {
 case 10:
  new_link.link_speed = 10;
  break;
 case 100:
  new_link.link_speed = 100;
  break;
 case 1000:
  new_link.link_speed = 1000;
  break;
 case 10000:
  new_link.link_speed = 10000;
  break;
 case 20000:
  new_link.link_speed = 20000;
  break;
 case 25000:
  new_link.link_speed = 25000;
  break;
 case 40000:
  new_link.link_speed = 40000;
  break;
 case 50000:
  new_link.link_speed = 50000;
  break;
 case 100000:
  new_link.link_speed = 100000;
  break;
 default:
  new_link.link_speed = 0;
  break;
 }

 new_link.link_duplex = 1;
 new_link.link_status = vf->link_up ? 1 :
          0;
 new_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
    (1 << 0));

 if (rte_atomic64_cmpset((uint64_t *)&dev->data->dev_link,
    *(uint64_t *)&dev->data->dev_link,
    *(uint64_t *)&new_link) == 0)
  return -1;

 return 0;
}



[2]
        .globl  iavf_dev_link_update
        .type   iavf_dev_link_update, @function iavf_dev_link_update:
.LFB6034:
        .cfi_startproc
        movq    24(%rdi), %rcx
        movq    96(%rcx), %rsi
        movl    476(%rsi), %edx
        cmpl    $20000, %edx
        je      .L5
        jbe     .L28
        cmpl    $50000, %edx
        je      .L5
        jbe     .L29
        cmpl    $100000, %edx
        movl    $0, %eax
        cmovne  %rax, %rdx
.L5:
        movl    112(%rcx), %eax
        movzbl  473(%rsi), %esi
        notl    %eax
        andl    $1, %eax
        sall    $2, %esi
        leal    1(%rax,%rax), %eax
        orl     %esi, %eax
        andl    $7, %eax
        salq    $32, %rax
        orq     %rax, %rdx
        movq    104(%rcx), %rax
#APP
# 33 "/root/development/dpdk-next-net/build/include/rte_atomic_64.h" 1
        lock ; cmpxchgq %rdx, 104(%rcx);sete %al;
# 0 "" 2
#NO_APP
        cmpb    $1, %al
        sbbl    %eax, %eax
        ret
        .p2align 4,,10
        .p2align 3
.L29:
        cmpl    $25000, %edx
        je      .L5
        cmpl    $40000, %edx
        je      .L5
.L11:
        xorl    %edx, %edx
        jmp     .L5
        .p2align 4,,10
        .p2align 3
.L28:
        cmpl    $1000, %edx
        je      .L5
        jbe     .L30
        cmpl    $10000, %edx
        movl    $0, %eax
        cmovne  %rax, %rdx
        jmp     .L5
        .p2align 4,,10
        .p2align 3
.L30:
        cmpl    $10, %edx
        je      .L5
        cmpl    $100, %edx
        je      .L5
        jmp     .L11
        .cfi_endproc
.LFE6034:
        .size   iavf_dev_link_update, .-iavf_dev_link_update
        .p2align 4

[Bug c/95818] wrong "used uninitialized" warning

[glisse at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #3 from Marc Glisse <glisse at gcc dot gnu.org> ---
Richard said "complete", that is the whole .i file, not just one random
function. If we cannot reproduce the issue by copying your code and compiling
it, we can't do anything about your report.

[Bug c/95818] wrong "used uninitialized" warning

[ferruh.yigit at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #4 from Ferruh YIGIT <ferruh.yigit at intel dot com> ---
Created attachment 48772
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48772&action=edit
preprocessed source

[Bug c/95818] wrong "used uninitialized" warning

[ferruh.yigit at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #5 from Ferruh YIGIT <ferruh.yigit at intel dot com> ---
Got it, 'iavf_ethdev.i' attached.

I am able to reproduce with command [1] and got the output [2] using .i file.

[1]
/usr/local/gcc-latest/bin/gcc -m64 -pthread -march=native -O3 -Wall -Werror
-Wno-strict-aliasing -v -save-temps -o iavf_ethdev.o -x cpp-output
iavf_ethdev.i


[2]
Using built-in specs.
COLLECT_GCC=/usr/local/gcc-latest/bin/gcc
COLLECT_LTO_WRAPPER=/usr/local/gcc-latest/libexec/gcc/x86_64-pc-linux-gnu/11.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-latest/configure --prefix=/usr/local/gcc-latest :
(reconfigured) ../gcc-latest/configure --prefix=/usr/local/gcc-latest
--enable-languages=c,c++,fortran,lto,objc --no-create --no-recursion
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 11.0.0 20200621 (experimental) (GCC) 
COLLECT_GCC_OPTIONS='-m64' '-pthread' '-march=native' '-O3' '-Wall' '-Werror'
'-Wno-strict-aliasing' '-v' '-save-temps' '-o' 'iavf_ethdev.o' '-dumpdir'
'iavf_ethdev.o-'
 /usr/local/gcc-latest/libexec/gcc/x86_64-pc-linux-gnu/11.0.0/cc1
-fpreprocessed iavf_ethdev.i -march=skylake-avx512 -mmmx -mno-3dnow -msse
-msse2 -msse3 -mssse3 -mno-sse4a -mcx16 -msahf -mmovbe -maes -mno-sha -mpclmul
-mpopcnt -mabm -mno-lwp -mfma -mno-fma4 -mno-xop -mbmi -mno-sgx -mbmi2
-mno-pconfig -mno-wbnoinvd -mno-tbm -mavx -mavx2 -msse4.2 -msse4.1 -mlzcnt
-mrtm -mhle -mrdrnd -mf16c -mfsgsbase -mrdseed -mprfchw -madx -mfxsr -mxsave
-mxsaveopt -mavx512f -mno-avx512er -mavx512cd -mno-avx512pf -mno-prefetchwt1
-mclflushopt -mxsavec -mxsaves -mavx512dq -mavx512bw -mavx512vl -mno-avx512ifma
-mno-avx512vbmi -mno-avx5124fmaps -mno-avx5124vnniw -mclwb -mno-mwaitx
-mno-clzero -mpku -mno-rdpid -mno-gfni -mno-shstk -mno-avx512vbmi2
-mno-avx512vnni -mno-vaes -mno-vpclmulqdq -mno-avx512bitalg
-mno-avx512vpopcntdq -mno-movdiri -mno-movdir64b -mno-waitpkg -mno-cldemote
-mno-ptwrite -mno-avx512bf16 -mno-enqcmd -mno-avx512vp2intersect -mno-serialize
-mno-tsxldtrk --param l1-cache-size=32 --param l1-cache-line-size=64 --param
l2-cache-size=33792 -mtune=skylake-avx512 -quiet -dumpdir iavf_ethdev.o-
-dumpbase iavf_ethdev.i -dumpbase-ext .i -m64 -O3 -Wall -Werror
-Wno-strict-aliasing -version -o iavf_ethdev.o-iavf_ethdev.s
GNU C17 (GCC) version 11.0.0 20200621 (experimental) (x86_64-pc-linux-gnu)
        compiled by GNU C version 11.0.0 20200621 (experimental), GMP version
6.1.2, MPFR version 3.1.6-p2, MPC version 1.1.0, isl version none
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
GNU C17 (GCC) version 11.0.0 20200621 (experimental) (x86_64-pc-linux-gnu)
        compiled by GNU C version 11.0.0 20200621 (experimental), GMP version
6.1.2, MPFR version 3.1.6-p2, MPC version 1.1.0, isl version none
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 6a7f176d54144e44e3ed475e12ec1e06
.../net/iavf/iavf_ethdev.c: In function ‘iavf_dev_link_update’:
.../net/iavf/iavf_ethdev.c:641:6: error: ‘new_link’ is used uninitialized
[-Werror=uninitialized]
  641 |  if (rte_atomic64_cmpset((uint64_t *)&dev->data->dev_link,
      |      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  642 |     *(uint64_t *)&dev->data->dev_link,
      |     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  643 |     *(uint64_t *)&new_link) == 0)
      |     ~~~~~~~~~~~~~~~~~~~~~~~
.../net/iavf/iavf_ethdev.c:596:22: note: ‘new_link’ declared here
  596 |  struct rte_eth_link new_link;
      |                      ^~~~~~~~
cc1: all warnings being treated as errors

[Bug middle-end/95818] wrong "used uninitialized" warning

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |middle-end
         Resolution|---                         |INVALID
                 CC|                            |msebor at gcc dot gnu.org
             Status|WAITING                     |RESOLVED

--- Comment #6 from Martin Sebor <msebor at gcc dot gnu.org> ---
This instance of the warning looks familiar but I can't reproduce it with my
build of the kernel.  The code in the test case isn't valid (it triggers both
-Wstrict-aliasing and -Wuninitialized), and I don't think it's safe either. 
Only the leading 48 bits of the struct are initialized (by GCC; the program
initializes just 35 bits) but the access is to the full 64 bits.  Clearing the
whole struct by calling memset avoids the -Wuninitialized.

A small test case that reproduces both warnings is below.

$ cat z.c && gcc -O2 -S -Wall -fdump-tree-uninit=/dev/stdout z.c
struct S
{
  int a;
  short i: 1;
};

void f (long);

void g (void)
{
  struct S s;
  s.a = 0;
  s.i = 1;
  f (*(long*)&s);
}
z.c: In function ‘g’:
z.c:14:7: warning: dereferencing type-punned pointer will break strict-aliasing
rules [-Wstrict-aliasing]
   14 |   f (*(long*)&s);
      |       ^~~~~~~~~
z.c:14:3: warning: ‘s’ is used uninitialized [-Wuninitialized]
   14 |   f (*(long*)&s);
      |   ^~~~~~~~~~~~~~
z.c:11:12: note: ‘s’ declared here
   11 |   struct S s;
      |            ^

;; Function g (g, funcdef_no=0, decl_uid=1937, cgraph_uid=1, symbol_order=0)

g ()
{
  struct S s;
  long int _1;

  <bb 2> [local count: 1073741824]:
  s.a = 0;
  s.i = -1;
  _1 = MEM[(long int *)&s];
  f (_1); [tail call]
  s ={v} {CLOBBER};
  return;

}

[Bug middle-end/95818] wrong "used uninitialized" warning

[ferruh.yigit at intel dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #7 from Ferruh YIGIT <ferruh.yigit at intel dot com> ---
Thanks Martin, explanation and sample clarified the issue for me.
Sorry for the inconvenience caused.

[Bug middle-end/95818] wrong "used uninitialized" warning

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95818

--- Comment #8 from Martin Sebor <msebor at gcc dot gnu.org> ---
No problem (and no inconvenience either).