[PATCH v5.2 12/12] linux/termios: regression test for termios speed functions

[H. Peter Anvin]

On 6/2/25 05:37, Adhemerval Zanella Netto wrote:
> 
> On 30/05/25 17:22, H. Peter Anvin wrote:
>> On May 30, 2025 6:32:22 AM PDT, Adhemerval Zanella Netto <adhemerval.zanella@linaro.org> wrote:
>>>
>>>
>>> On 27/05/25 19:08, H. Peter Anvin wrote:
>>>> +
>>>> +#define NUM_TEST_SPEEDS 41
>>>> +static const speed_t test_speeds [NUM_TEST_SPEEDS] =
>>>> +{
>>>> +  0, 50, 75, 110, 134, 150, 200, 300, 600, 1200, 1800, 2400,
>>>> +  4800, 7200, 9600, 14400, 19200, 28800, 33600, 38400, 57600,
>>>> +  76800, 115200, 153600, 230400, 307200, 460800, 500000,
>>>> +  576000, 614400, 921600, 1000000, 1152000, 1500000, 2000000,
>>>> +  2500000, 3000000, 3500000, 4000000, 5000000, 10000000
>>>> +};
>>>
>>
>> Could you send me an strace?
> 
> Here it is https://termbin.com/unp4

Thanks!

I think I found the problem: the definition of CIBAUD on Alpha was
completely broken (to the extent that I have no idea how it got to be
that way.) I added a series of #ifs to speed.c to error out if CBAUD,
CIBAUD and IBSHIFT are inconsistent or violate the assumptions that I
used to optimize speed.c.

I will post a v6 shortly.

	-hpa