Hello Al and Matthew! On 20-05-22 18:18:56, Al Viro wrote: [...] > > I think the patch is actually right, despite the shockingly bad changelog. > > He's removed the duplicate 'O_NDELAY' and reformatted the lines. > > So he has; my apologies - the obvious false duplicate in there would be > O_NDELAY vs. O_NONBLOCK and I'd misread the patch. > > Commit message should've been along the lines of "O_NDELAY occurs twice > in definition of VALID_OPEN_FLAGS; get rid of the duplicate", possibly > along with "Note: O_NONBLOCK in the same expression is *not* another > duplicate - on sparc O_NONBLOCK != O_NDELAY (done that way for ABI > compatibility with Solaris back when sparc port began)". Apologies. The v2 is coming and will have proper commit message along with subject that explains what the intended change is, etc. Krzysztof
