Re: Having memory leak issues with perl-c
[email protected] (Mark Murawski) Fri, 12 Aug 2022 19:17:14 -0400
| Newsgroups | perl.qa |
|---|---|
| Message-ID | <[email protected]> |
--------------H2Q5PUxY2WXV6PP15803El3y Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit On 8/12/22 18:26, Mark Murawski wrote: > On 8/12/22 18:11, Mark Murawski wrote: >> On 8/4/22 15:10, demerphq wrote: >>> On Thu, 4 Aug 2022 at 17:04, Mark Murawski >>> <[email protected]> wrote: >>> >>> On 8/4/22 02:50, demerphq wrote: >>>> On Thu, 4 Aug 2022 at 01:58, Mark Murawski >>>> <[email protected]> wrote: >>>> >>>> I'm still not getting something... if I want to fix the >>>> code-as-is and do this: >>>> >>>> FNsv = get_sv("main::_FN", GV_ADD); >>>> if (!FNsv) >>>> ereport(ERROR, >>>> (errcode(ERRCODE_EXTERNAL_ROUTINE_EXCEPTION), >>>> errmsg("couldn't fetch $_FN"))); >>>> >>>> save_item(FNsv); /* local $_FN */ >>>> >>>> >>>> I dont get the sequence here. You take the old value of >>>> $main::_FN and then you localize it after you fetch it? That >>>> seems weird. >>> >>> >>> You did not respond to this comment ^^ >> >> >> The reason for the save_item*( was because I was modeling this code >> on another section that uses save_item() to accomplish something >> similar (to send internal postgres details to some perl). I don't >> know why the original author uses save_item() for this purpose. >> > > Oh... I know why save_item is used. It's because this code can be > executed multiple times in the same perl process. So each one needs > it's own _FN > > After a serious amount of trial and error... I got it to stop crashing and I fixed the leak HV *hv; // hash SV *FNsv; // scalar reference to the hash SV *SVhv; ENTER; SAVETMPS; FNsv = get_sv("main::_FN", GV_ADD); save_item(FNsv); /* local $_FN */ hv = newHV(); // create new hash SVhv = newRV_noinc((SV *) hv); sv_setsv(FNsv, SVhv); hv_store_string(hv, "name", cstr2sv(desc->proname)); PUTBACK; FREETMPS; LEAVE; SvREFCNT_dec_current(SVhv); The fix I found was to put SvREFCNT_dec_current *after* PUTBACK/FREETMPS/LEAVE if you do the SvREFCNT_dec_current prior.. then it leaks If the trailing end of the code looks like this... you get a leak SvREFCNT_dec_current(SVhv); PUTBACK; FREETMPS; LEAVE; --------------H2Q5PUxY2WXV6PP15803El3y Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: 8bit <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> </head> <body> <div class="moz-cite-prefix">On 8/12/22 18:26, Mark Murawski wrote:<br> </div> <blockquote type="cite" cite="mid:[email protected]"> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> <div class="moz-cite-prefix">On 8/12/22 18:11, Mark Murawski wrote:<br> </div> <blockquote type="cite" cite="mid:[email protected]"> <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> <div class="moz-cite-prefix">On 8/4/22 15:10, demerphq wrote:<br> </div> <blockquote type="cite" cite="mid:CANgJU+W95nxb9HB_EEXUo3KDYqx-1xe14TZmJG_v-RBD=evZpw@mail.gmail.com"> <meta http-equiv="content-type" content="text/html; charset=UTF-8"> <div dir="ltr"> <div dir="ltr">On Thu, 4 Aug 2022 at 17:04, Mark Murawski <<a href="mailto:[email protected]" moz-do-not-send="true" class="moz-txt-link-freetext">[email protected]</a>> wrote:<br> </div> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <div> <div>On 8/4/22 02:50, demerphq wrote:<br> </div> <blockquote type="cite"> <div dir="ltr"> <div dir="ltr">On Thu, 4 Aug 2022 at 01:58, Mark Murawski <<a href="mailto:[email protected]" target="_blank" moz-do-not-send="true" class="moz-txt-link-freetext">[email protected]</a>> wrote:<br> </div> <div class="gmail_quote"> <blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <div> <div>I'm still not getting something... if I want to fix the code-as-is and do this:<br> </div> <br> FNsv = get_sv("main::_FN", GV_ADD);<br> if (!FNsv)<br> ereport(ERROR,<br> (errcode(ERRCODE_EXTERNAL_ROUTINE_EXCEPTION),<br> errmsg("couldn't fetch $_FN")));<br> <br> save_item(FNsv); /* local $_FN */<br> </div> </blockquote> <div><br> </div> <div>I dont get the sequence here. You take the old value of $main::_FN and then you localize it after you fetch it? That seems weird.</div> </div> </div> </blockquote> </div> </blockquote> <div><br> </div> <div>You did not respond to this comment ^^</div> </div> </div> </blockquote> <br> <br> The reason for the save_item*( was because I was modeling this code on another section that uses save_item() to accomplish something similar (to send internal postgres details to some perl). I don't know why the original author uses save_item() for this purpose.<br> <br> </blockquote> <br> Oh... I know why save_item is used. It's because this code can be executed multiple times in the same perl process. So each one needs it's own _FN<br> <br> <br> </blockquote> <br> <br> <br> After a serious amount of trial and error... I got it to stop crashing and I fixed the leak<br> <br> <br> HV *hv; // hash<br> SV *FNsv; // scalar reference to the hash<br> SV *SVhv;<br> <br> ENTER;<br> SAVETMPS;<br> <br> FNsv = get_sv("main::_FN", GV_ADD);<br> save_item(FNsv); /* local $_FN */<br> <br> hv = newHV(); // create new hash<br> SVhv = newRV_noinc((SV *) hv);<br> <br> sv_setsv(FNsv, SVhv);<br> hv_store_string(hv, "name", cstr2sv(desc->proname));<br> <br> PUTBACK;<br> FREETMPS;<br> LEAVE;<br> <br> SvREFCNT_dec_current(SVhv);<br> <br> <br> The fix I found was to put SvREFCNT_dec_current *after* PUTBACK/FREETMPS/LEAVE<br> <br> if you do the SvREFCNT_dec_current prior.. then it leaks<br> <br> <br> If the trailing end of the code looks like this... you get a leak<br> <br> <br> SvREFCNT_dec_current(SVhv);<br> <br> PUTBACK;<br> FREETMPS;<br> LEAVE;<br> <br> <br> </body> </html> --------------H2Q5PUxY2WXV6PP15803El3y--