Re: [PATCH v2 06/26] qemu-keymap: properly check return from xkb_keymap_mod_get_index

[Alex Bennée]

Richard Henderson <richard.henderson@linaro.org> writes:

> On 6/26/23 23:59, Alex Bennée wrote:
>> We can return XKB_MOD_INVALID which rightly gets flagged by sanitisers
>> as an overly wide shift attempt.
>> Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
>> ---
>>   qemu-keymap.c | 24 ++++++++++++++++--------
>>   1 file changed, 16 insertions(+), 8 deletions(-)
>> diff --git a/qemu-keymap.c b/qemu-keymap.c
>> index 229866e004..8c80f7a4ed 100644
>> --- a/qemu-keymap.c
>> +++ b/qemu-keymap.c
>> @@ -140,6 +140,18 @@ static void usage(FILE *out)
>>               names.options ?: "-");
>>   }
>>   +static xkb_mod_mask_t get_mod(struct xkb_keymap *map, const char
>> *name)
>> +{
>> +    xkb_mod_index_t mod;
>> +    xkb_mod_mask_t mask = 0;
>> +
>> +    mod = xkb_keymap_mod_get_index(map, name);
>> +    if (mod != XKB_MOD_INVALID) {
>> +        mask = (1 << mod);
>> +    }
>> +    return mask;
>> +}
>
> You have yet to answer Peter's question -- asked twice -- about what
> changes in the keymaps with this. If nothing, should it in fact be an
> assert instead?

Ahh in the other thread. No change, it looks like AltGr just doesn't
exist for some keymaps:

  🕙21:20:36 alex@zen:qemu.git/builds/all  (399fc0c) (REBASING 2/22) [$!?] 
  ➜  diff -ub gb.before gb.after
  🕙21:20:43 alex@zen:qemu.git/builds/all  (399fc0c) (REBASING 2/22) [$!?] 
  ➜  diff -ub ara.before ara.after
  🕙21:20:50 alex@zen:qemu.git/builds/all  (399fc0c) (REBASING 2/22) [$!?] 
  ➜  ag "Alt" ara.after 
  21:#     9: Alt
  23:#    11: LAlt
  24:#    12: RAlt
  29:#    17: AltGr
  294:Alt_L 0x38
  1711:Alt_R 0xb8
  🕙21:22:14 alex@zen:qemu.git/builds/all  (399fc0c) (REBASING 2/22) [$!?] 
  ➜  ag "Alt" gb.after 
  21:#     9: Alt
  23:#    11: LAlt
  24:#    12: RAlt
  29:#    17: AltGr
  338:Alt_L 0x38
  1757:Alt_R 0xb8


-- 
Alex Bennée
Virtualisation Tech Lead @ Linaro