[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [XEN PATCH v1 2/3] ocaml/libs: Fill build failure due to unused variable in ocaml macro


  • To: Javi Merino <javi.merino@xxxxxxxxx>, xen-devel@xxxxxxxxxxxxxxxxxxxx
  • From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
  • Date: Thu, 17 Oct 2024 19:08:21 +0100
  • Autocrypt: addr=andrew.cooper3@xxxxxxxxxx; keydata= xsFNBFLhNn8BEADVhE+Hb8i0GV6mihnnr/uiQQdPF8kUoFzCOPXkf7jQ5sLYeJa0cQi6Penp VtiFYznTairnVsN5J+ujSTIb+OlMSJUWV4opS7WVNnxHbFTPYZVQ3erv7NKc2iVizCRZ2Kxn srM1oPXWRic8BIAdYOKOloF2300SL/bIpeD+x7h3w9B/qez7nOin5NzkxgFoaUeIal12pXSR Q354FKFoy6Vh96gc4VRqte3jw8mPuJQpfws+Pb+swvSf/i1q1+1I4jsRQQh2m6OTADHIqg2E ofTYAEh7R5HfPx0EXoEDMdRjOeKn8+vvkAwhviWXTHlG3R1QkbE5M/oywnZ83udJmi+lxjJ5 YhQ5IzomvJ16H0Bq+TLyVLO/VRksp1VR9HxCzItLNCS8PdpYYz5TC204ViycobYU65WMpzWe LFAGn8jSS25XIpqv0Y9k87dLbctKKA14Ifw2kq5OIVu2FuX+3i446JOa2vpCI9GcjCzi3oHV e00bzYiHMIl0FICrNJU0Kjho8pdo0m2uxkn6SYEpogAy9pnatUlO+erL4LqFUO7GXSdBRbw5 gNt25XTLdSFuZtMxkY3tq8MFss5QnjhehCVPEpE6y9ZjI4XB8ad1G4oBHVGK5LMsvg22PfMJ ISWFSHoF/B5+lHkCKWkFxZ0gZn33ju5n6/FOdEx4B8cMJt+cWwARAQABzSlBbmRyZXcgQ29v cGVyIDxhbmRyZXcuY29vcGVyM0BjaXRyaXguY29tPsLBegQTAQgAJAIbAwULCQgHAwUVCgkI CwUWAgMBAAIeAQIXgAUCWKD95wIZAQAKCRBlw/kGpdefoHbdD/9AIoR3k6fKl+RFiFpyAhvO 59ttDFI7nIAnlYngev2XUR3acFElJATHSDO0ju+hqWqAb8kVijXLops0gOfqt3VPZq9cuHlh IMDquatGLzAadfFx2eQYIYT+FYuMoPZy/aTUazmJIDVxP7L383grjIkn+7tAv+qeDfE+txL4 SAm1UHNvmdfgL2/lcmL3xRh7sub3nJilM93RWX1Pe5LBSDXO45uzCGEdst6uSlzYR/MEr+5Z JQQ32JV64zwvf/aKaagSQSQMYNX9JFgfZ3TKWC1KJQbX5ssoX/5hNLqxMcZV3TN7kU8I3kjK mPec9+1nECOjjJSO/h4P0sBZyIUGfguwzhEeGf4sMCuSEM4xjCnwiBwftR17sr0spYcOpqET ZGcAmyYcNjy6CYadNCnfR40vhhWuCfNCBzWnUW0lFoo12wb0YnzoOLjvfD6OL3JjIUJNOmJy RCsJ5IA/Iz33RhSVRmROu+TztwuThClw63g7+hoyewv7BemKyuU6FTVhjjW+XUWmS/FzknSi dAG+insr0746cTPpSkGl3KAXeWDGJzve7/SBBfyznWCMGaf8E2P1oOdIZRxHgWj0zNr1+ooF /PzgLPiCI4OMUttTlEKChgbUTQ+5o0P080JojqfXwbPAyumbaYcQNiH1/xYbJdOFSiBv9rpt TQTBLzDKXok86M7BTQRS4TZ/ARAAkgqudHsp+hd82UVkvgnlqZjzz2vyrYfz7bkPtXaGb9H4 Rfo7mQsEQavEBdWWjbga6eMnDqtu+FC+qeTGYebToxEyp2lKDSoAsvt8w82tIlP/EbmRbDVn 7bhjBlfRcFjVYw8uVDPptT0TV47vpoCVkTwcyb6OltJrvg/QzV9f07DJswuda1JH3/qvYu0p vjPnYvCq4NsqY2XSdAJ02HrdYPFtNyPEntu1n1KK+gJrstjtw7KsZ4ygXYrsm/oCBiVW/OgU g/XIlGErkrxe4vQvJyVwg6YH653YTX5hLLUEL1NS4TCo47RP+wi6y+TnuAL36UtK/uFyEuPy wwrDVcC4cIFhYSfsO0BumEI65yu7a8aHbGfq2lW251UcoU48Z27ZUUZd2Dr6O/n8poQHbaTd 6bJJSjzGGHZVbRP9UQ3lkmkmc0+XCHmj5WhwNNYjgbbmML7y0fsJT5RgvefAIFfHBg7fTY/i kBEimoUsTEQz+N4hbKwo1hULfVxDJStE4sbPhjbsPCrlXf6W9CxSyQ0qmZ2bXsLQYRj2xqd1 bpA+1o1j2N4/au1R/uSiUFjewJdT/LX1EklKDcQwpk06Af/N7VZtSfEJeRV04unbsKVXWZAk uAJyDDKN99ziC0Wz5kcPyVD1HNf8bgaqGDzrv3TfYjwqayRFcMf7xJaL9xXedMcAEQEAAcLB XwQYAQgACQUCUuE2fwIbDAAKCRBlw/kGpdefoG4XEACD1Qf/er8EA7g23HMxYWd3FXHThrVQ HgiGdk5Yh632vjOm9L4sd/GCEACVQKjsu98e8o3ysitFlznEns5EAAXEbITrgKWXDDUWGYxd pnjj2u+GkVdsOAGk0kxczX6s+VRBhpbBI2PWnOsRJgU2n10PZ3mZD4Xu9kU2IXYmuW+e5KCA vTArRUdCrAtIa1k01sPipPPw6dfxx2e5asy21YOytzxuWFfJTGnVxZZSCyLUO83sh6OZhJkk b9rxL9wPmpN/t2IPaEKoAc0FTQZS36wAMOXkBh24PQ9gaLJvfPKpNzGD8XWR5HHF0NLIJhgg 4ZlEXQ2fVp3XrtocHqhu4UZR4koCijgB8sB7Tb0GCpwK+C4UePdFLfhKyRdSXuvY3AHJd4CP 4JzW0Bzq/WXY3XMOzUTYApGQpnUpdOmuQSfpV9MQO+/jo7r6yPbxT7CwRS5dcQPzUiuHLK9i nvjREdh84qycnx0/6dDroYhp0DFv4udxuAvt1h4wGwTPRQZerSm4xaYegEFusyhbZrI0U9tJ B8WrhBLXDiYlyJT6zOV2yZFuW47VrLsjYnHwn27hmxTC/7tvG3euCklmkn9Sl9IAKFu29RSo d5bD8kMSCYsTqtTfT6W4A3qHGvIDta3ptLYpIAOD2sY3GYq2nf3Bbzx81wZK14JdDDHUX2Rs 6+ahAA==
  • Cc: Christian Lindig <christian.lindig@xxxxxxxxxx>, David Scott <dave@xxxxxxxxxx>, Anthony PERARD <anthony.perard@xxxxxxxxxx>, Andrii Sultanov <andrii.sultanov@xxxxxxxxx>
  • Delivery-date: Thu, 17 Oct 2024 18:08:37 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>

On 17/10/2024 6:47 pm, Andrew Cooper wrote:
> On 17/10/2024 5:20 pm, Javi Merino wrote:
>> On Fedora 40, the build fails with:
>>
>>     gcc -m64 -DBUILD_ID -fno-strict-aliasing -std=gnu99 -Wall 
>> -Wstrict-prototypes -Wno-unused-but-set-variable -Wno-unused-local-typedefs  
>>  -Werror -O2 -fomit-frame-pointer 
>> -D__XEN_INTERFACE_VERSION__=__XEN_LATEST_INTERFACE_VERSION__ -MMD -MP -MF 
>> .domain_getinfo_stubs_v1.o.d -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE   
>> -fPIC -I/usr/lib64/ocaml -I 
>> /build/tools/ocaml/libs/xsd_glue/domain_getinfo_plugin_v1/../../../libs -I 
>> /build/tools/ocaml/libs/xsd_glue/domain_getinfo_plugin_v1/../../../libs/xsd_glue
>>  
>> -I/build/tools/ocaml/libs/xsd_glue/domain_getinfo_plugin_v1/../../../../../tools/include
>>  -D__XEN_TOOLS__ 
>> -I/build/tools/ocaml/libs\/xsd_glue/domain_getinfo_plugin_v1/../../../../../tools/include
>>   -c -o domain_getinfo_stubs_v1.o domain_getinfo_stubs_v1.c
>>     In file included from domain_getinfo_stubs_v1.c:10:
>>     domain_getinfo_stubs_v1.c: In function 'xsd_glue_failwith':
>>     /usr/lib64/ocaml/caml/memory.h:275:29: error: unused variable 
>> 'caml__frame' [-Werror=unused-variable]
>>       275 |   struct caml__roots_block *caml__frame = *caml_local_roots_ptr
>>           |                             ^~~~~~~~~~~
>>     domain_getinfo_stubs_v1.c:48:9: note: in expansion of macro 'CAMLparam0'
>>       48 |         CAMLparam0();
>>          |         ^~~~~~~~~~
>>     cc1: all warnings being treated as errors
>>
>> The CAMLparam0 macro is defined in /usr/lib64/ocaml/caml/memory.h:255 as:
>>
>>     #define CAMLparam0()                                                    \
>>       struct caml__roots_block** caml_local_roots_ptr =                     \
>>         (DO_CHECK_CAML_STATE ? Caml_check_caml_state() : (void)0,           \
>>          &CAML_LOCAL_ROOTS);                                                \
>>       struct caml__roots_block *caml__frame = *caml_local_roots_ptr
>>
>> We can't modify the macro.  Mark the xsd_glue_failwith() function with
>> ignore "-Wunused-variable" to prevent gcc from failing the build due
>> to the unused variable.
>>
>> Fixes: a6576011a4d2 ("ocaml/libs: Implement a dynamically-loaded plugin for 
>> Xenctrl.domain_getinfo")
>> Signed-off-by: Javi Merino <javi.merino@xxxxxxxxx>
> That's horrible...
>
> I note the Ocaml manual even says:
>
> "Some C compilers give bogus warnings about unused variables
> caml__dummy_xxx at each use of CAMLparam and CAMLlocal. You should
> ignore them."  which a brave claim to make...
>
>
> The problem with pragma gcc is that we support Clang too.
>
> Having had a play, this works too
>
> @@ -45,7 +45,9 @@ static struct custom_operations xsd_glue_xenctrl_ops = {
>  static void Noreturn xsd_glue_failwith(
>         xc_interface *xch, const char *func, unsigned int line)
>  {
> +#define caml__frame __attribute__((unused)) caml__frame
>         CAMLparam0();
> +#undef caml__frame
>         CAMLlocal1(msg);
>         const xc_error *error = xch ? xc_get_last_error(xch) : NULL;
>         char *str = NULL;
>
> and is rather more selective.
>
>
> However, looking through other bits of memory.h, there's this gem:
>
> #define CAMLnoreturn ((void) caml__frame)
>
> which has existed since db3745919 "suppression des warnings "unused
> variable" de gcc" in 2004.
>
> So, I think this is a better fix:
>
> @@ -69,6 +69,7 @@ static void Noreturn xsd_glue_failwith(
>         free(str);
>  
>         caml_raise_with_arg(*caml_named_value("xsg.error_v1"), msg);
> +       CAMLnoreturn;
>  }
>  #define xsd_glue_failwith(xch) xsd_glue_failwith(xch, __func__, __LINE__)
>  
> and F40 seem happy with the result.

And https://github.com/ocaml/ocaml/pull/13561 to update Ocaml's
documentation on the matter.

~Andrew



 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.