Re: [patch] required signals

Albert D. Cahalan (acahalan@cs.uml.edu)
Sat, 20 Feb 1999 00:01:30 -0500 (EST)


Linus writes:
> On Fri, 19 Feb 1999, Albert D. Cahalan wrote:
>
>> Sure it is, not even counting libc 5. The kernel has code that is
>> signal-specific, so it must define the numbering. The kernel itself
>> uses the numbers in many places. Disagreement over allocation would
>> be terrible.
>
> There is no disagreement: the kernel specifies the numbers,
> and user space lives with them.
>
> The numbers the kernel doesn't care about might as well be defined by
> the library.

Unfortunately the kernel _does_ care about them. Every signal has a
default behavior determined by the kernel, and that behavior matters.

So you don't think the patch is important... but is is bad?
Proper default behavior can not be defined without defining
what each signal maps to.

Benefits:

kernel --> developers will know what is used for what
libc 5 --> gets signals that are needed
libc 6 --> developers get an assurance of stability

Please apply this patch. I can produce another with the SIGEMT
and SIGSYS signals reversed if you prefer.

begin 664 signals.gz
M'XL(`*L"Q#8"`^V776^B4!"&K^%73+)W(L7#E\I-[8=N3:D:P79W[VC!+5D$
M`Y)T__T.H)1S6C#-)G4WP0LB<^:=>1T.C^#ZZS6(,R>-(0I<+Y;\\"E(74]R
MDHWHQ!LI\7^&3G#V#($?IB^UR[PHBD<J<)/8AXGW"$2&GFHHLJ&J0(;#(2\(
MPI'RE%8S9-W0]MK1"$1%[?9!P*,.HQ$/7UQO[8<>6-.O4]/D.%"IF+V\6&!0
MHX(7ETL;@SHO5M7S(E9-O%Q9&.M3L<EBC+$!%;LM6A<&U=R@JN"1,;B89EJB
MT&;,Y1T&&=OC/*A1#BW[=F+:'-%YH1K]CAX)[?OJQKS&(&W\:C[#7TAHYY8]
MQ_F0PKFF=XD*@J9UR8#U_FUB_>`XF1[DO9V[E^GNB^5\@JE]ROS#='9UP\D#
M9N:8-Z1"BWDVRGR-%Z4.M6;.+;L0="1:]+#D.*57QC+]:K:RQM><0JAI'6R`
MU($P"L5DYX2N$[M=>$QW\!1M-E$(NP@NK6O`%;!^)_>`W80W%M&&\-8"6[<H
MD%4RLST/Q1ZO5LR\CN_P=RGDPZ9X@6WG!+Z3>`EK.=_<Y<Y'T3;:IH$3LWG[
M+7;85IB(MWFM]<,E+&?"%,93]NKM4W$Y2SYG![&_:(>1U/7G@<]F93][B0?)
M<Y1B4ACMX-'#:86)CV3RW/PKSF67P#J.-I`F7AS@F,X@JU%UMK3OIC-.D7FW
M"9"^,M";"$FMUR"2RGF'D;TF1M:+-8/T2G$+R<^"9+^%Y"=!$EI(?A22IT#D
M1A_\:D(DM5Z#2"KG'40.FA!9+T9$Z@89MHAL$=D^1_X-(J'\[.^`\OP8,5]E
MN->@*CO.3Z#:YG2L:?N6I[2L(Y6R<[9/<:U?^^`\C_H\!6>WVZ<FS%:7:RA;
M36$AJQ@]N0FRM5I\5Y<-17MEK*QE",/CO\I81=O_"?QWC%7[.6-5O65L^Z[^
0Z8^A<.K'T#]H`.1`RQ0`````
`
end

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/