| Message ID | 4C3D9782.1010801@t-online.de |
|---|---|
| State | New |
| Headers | show |
On 07/14/2010 12:54 PM, Bernd Schmidt wrote: > PR middle-end/44738 > * tree-ssa.c (warn_uninit): Avoid emitting an unnecessary message. > > PR middle-end/44738 > * c-c++-common/uninit-17.c: Correct expected error. Ping. Bernd
On Mon, Jul 19, 2010 at 2:50 PM, Bernd Schmidt <bernds@codesourcery.com> wrote: > On 07/14/2010 12:54 PM, Bernd Schmidt wrote: >> PR middle-end/44738 >> * tree-ssa.c (warn_uninit): Avoid emitting an unnecessary message. >> >> PR middle-end/44738 >> * c-c++-common/uninit-17.c: Correct expected error. > > Ping. Ok. (btw, your threading does not work - I had to lookup the original patch in the archives. Please make sure pings are replies to your original message) Thanks, Richard. > > Bernd >
On 07/19/2010 02:54 PM, Richard Guenther wrote: > On Mon, Jul 19, 2010 at 2:50 PM, Bernd Schmidt <bernds@codesourcery.com> wrote: >> On 07/14/2010 12:54 PM, Bernd Schmidt wrote: >>> PR middle-end/44738 >>> * tree-ssa.c (warn_uninit): Avoid emitting an unnecessary message. >>> >>> PR middle-end/44738 >>> * c-c++-common/uninit-17.c: Correct expected error. >> >> Ping. > > Ok. (btw, your threading does not work - I had to lookup the original > patch in the archives. Please make sure pings are replies to your > original message) It was. Any other known reason why this would get messed up? Have you seen this for other pings from me? Bernd
On Mon, Jul 19, 2010 at 2:55 PM, Bernd Schmidt <bernds@codesourcery.com> wrote: > On 07/19/2010 02:54 PM, Richard Guenther wrote: >> On Mon, Jul 19, 2010 at 2:50 PM, Bernd Schmidt <bernds@codesourcery.com> wrote: >>> On 07/14/2010 12:54 PM, Bernd Schmidt wrote: >>>> PR middle-end/44738 >>>> * tree-ssa.c (warn_uninit): Avoid emitting an unnecessary message. >>>> >>>> PR middle-end/44738 >>>> * c-c++-common/uninit-17.c: Correct expected error. >>> >>> Ping. >> >> Ok. (btw, your threading does not work - I had to lookup the original >> patch in the archives. Please make sure pings are replies to your >> original message) > > It was. Any other known reason why this would get messed up? Have you > seen this for other pings from me? I don't remember but will watch now. Maybe it is Google Mail that is basing threading purely on the subject (and so is confused by your Ping: prefix) instead of message-ids. Richard. > > Bernd >
On 10-07-19 08:59 , Richard Guenther wrote: > On Mon, Jul 19, 2010 at 2:55 PM, Bernd Schmidt<bernds@codesourcery.com> wrote: >> >> It was. Any other known reason why this would get messed up? Have you >> seen this for other pings from me? > > I don't remember but will watch now. Maybe it is Google Mail that is > basing threading purely on the subject (and so is confused by your > Ping: prefix) instead of message-ids. That's likely, gmail will do that. It threads by subject, instead of message-id. Diego.
Index: testsuite/c-c++-common/uninit-17.c =================================================================== --- testsuite/c-c++-common/uninit-17.c (revision 162146) +++ testsuite/c-c++-common/uninit-17.c (working copy) @@ -9,9 +9,9 @@ static void bar(int a, int *ptr) { do { - int b; /* { dg-message "note: 'b' was declared here" } */ + int b; /* { dg-warning "is used uninitialized" } */ if (b < 40) { - ptr[0] = b; /* { dg-warning "may be used uninitialized" } */ + ptr[0] = b; } b += 1; ptr++; Index: tree-ssa.c =================================================================== --- tree-ssa.c (revision 162146) +++ tree-ssa.c (working copy) @@ -1631,6 +1631,8 @@ warn_uninit (tree t, const char *gmsgid, { TREE_NO_WARNING (var) = 1; + if (location == DECL_SOURCE_LOCATION (var)) + return; if (xloc.file != floc.file || xloc.line < floc.line || xloc.line > LOCATION_LINE (cfun->function_end_locus))