how to solve "Warning: Inverse CDF calculation did not converge for p"

I have checked already an answer for this problem here (https://it.mathworks.com/matlabcentral/answers/263138-warning-inverse-cdf-calculation-did-not-converge-for-p) but is not clear for me exaclty how to solve it.

Basically I have these 2 lines of code giving the same error with different values of p.

x1 = ksdensity(x,xx,'function','icdf'); % error: Warning: Inverse CDF calculation did not converge for p = 0.0035454.
y1 = ksdensity(y,yy,'function','icdf'); % error: Warning: Inverse CDF calculation did not converge for p = 0.99161.

I have checked the help and documentation but I don't get it.

What exaclty is p?

I have checked x1 and y1 and I noticed that the scale is very big.

If I do:

max(x1) % 6.22e+07
max(y1) % 6.79e+07

such values are non-sense.

Can anybody help?

[Torsten]

p is the value where Matlab tries to evaluate the inverse cumulative density function.

This will be especially difficult at the boundaries of the ICDF, i.e. at 0 and at 1.

[Luigi Simeone]

Many many thanks Torsen. Actually when I plot x1 and y1, such high values (6e+7) are achieved not exactly in 1 or 0 but very close to them. Can you please suggest me how to solve it? it is not an error but a Warning however, I cannot ignore it as such values are dramatically large. The post I have linked suggest as a second option to remove these values. Do you know how to do it? thanks in advance

[Torsten]

Most probably by removing the smallest and largest value from your raw data.