Re: current bash.git.devel segfaults on my code
On Wed, Oct 20, 2021 at 3:46 AM Alex fxmbsw7 Ratchev wrote: > > it betteted up, it still segfaults without valgrind but valgrind dorsnt > show so many entries anymore > there are 3 valgrind outputs, one without -- two with ine --arg ( the > valgrind supported .. ) and three the two --opts > > the first two i can paste, the third due to being long is attached The Valgrind reports that you shared do not exhibit the segmentation fault crash, so they aren't too helpful in diagnosing the crash.
Re: current bash.git.devel segfaults on my code
look, i do ./bash it segfaults, also when make valgrind, then i do valgrind ./bash it shows what i pasted the segfault didnt happen a month or two ago On Wed, Oct 20, 2021, 22:55 Eduardo Bustamante wrote: > On Wed, Oct 20, 2021 at 3:46 AM Alex fxmbsw7 Ratchev > wrote: > > > > it betteted up, it still segfaults without valgrind but valgrind dorsnt > > show so many entries anymore > > there are 3 valgrind outputs, one without -- two with ine --arg ( the > > valgrind supported .. ) and three the two --opts > > > > the first two i can paste, the third due to being long is attached > > The Valgrind reports that you shared do not exhibit the segmentation > fault crash, so they aren't too helpful in diagnosing the crash. >
Re: current bash.git.devel segfaults on my code
you got other cmd suggestions i can run to help debugging ? im very limited in bin.c debugging On Wed, Oct 20, 2021, 22:57 Alex fxmbsw7 Ratchev wrote: > look, i do ./bash it segfaults, also when make valgrind, then i do > valgrind ./bash it shows what i pasted > the segfault didnt happen a month or two ago > > On Wed, Oct 20, 2021, 22:55 Eduardo Bustamante wrote: > >> On Wed, Oct 20, 2021 at 3:46 AM Alex fxmbsw7 Ratchev >> wrote: >> > >> > it betteted up, it still segfaults without valgrind but valgrind dorsnt >> > show so many entries anymore >> > there are 3 valgrind outputs, one without -- two with ine --arg ( the >> > valgrind supported .. ) and three the two --opts >> > >> > the first two i can paste, the third due to being long is attached >> >> The Valgrind reports that you shared do not exhibit the segmentation >> fault crash, so they aren't too helpful in diagnosing the crash. >> >
Re: current bash.git.devel segfaults on my code
On Wed, Oct 20, 2021, at 5:07 PM, Alex fxmbsw7 Ratchev wrote: > you got other cmd suggestions i can run to help debugging ? im very limited > in bin.c debugging You could bisect the devel branch to determine whether your problem was caused by a change to bash and, if so, which commit is responsible. https://git-scm.com/book/en/v2/Git-Tools-Debugging-with-Git#_binary_search https://git-scm.com/docs/git-bisect -- vq
Re: current bash.git.devel segfaults on my code
ill try to see if those texts on the page make sense to me, some later, im no fit im sorry On Wed, Oct 20, 2021, 23:34 Lawrence Velázquez wrote: > On Wed, Oct 20, 2021, at 5:07 PM, Alex fxmbsw7 Ratchev wrote: > > you got other cmd suggestions i can run to help debugging ? im very > limited > > in bin.c debugging > > You could bisect the devel branch to determine whether your problem > was caused by a change to bash and, if so, which commit is responsible. > > https://git-scm.com/book/en/v2/Git-Tools-Debugging-with-Git#_binary_search > https://git-scm.com/docs/git-bisect > > -- > vq >
Re: current bash.git.devel segfaults on my code
specific bisect cmds / instructions here welcome On Wed, Oct 20, 2021, 23:58 Alex fxmbsw7 Ratchev wrote: > ill try to see if those texts on the page make sense to me, some later, im > no fit im sorry > > On Wed, Oct 20, 2021, 23:34 Lawrence Velázquez wrote: > >> On Wed, Oct 20, 2021, at 5:07 PM, Alex fxmbsw7 Ratchev wrote: >> > you got other cmd suggestions i can run to help debugging ? im very >> limited >> > in bin.c debugging >> >> You could bisect the devel branch to determine whether your problem >> was caused by a change to bash and, if so, which commit is responsible. >> >> https://git-scm.com/book/en/v2/Git-Tools-Debugging-with-Git#_binary_search >> https://git-scm.com/docs/git-bisect >> >> -- >> vq >> >
Re: current bash.git.devel segfaults on my code
to tell u honestly, im in no state to read freely happily new docs if u dont specificly help with the cmds, i cannot succeed much On Wed, Oct 20, 2021, 23:59 Alex fxmbsw7 Ratchev wrote: > specific bisect cmds / instructions here welcome > > On Wed, Oct 20, 2021, 23:58 Alex fxmbsw7 Ratchev > wrote: > >> ill try to see if those texts on the page make sense to me, some later, >> im no fit im sorry >> >> On Wed, Oct 20, 2021, 23:34 Lawrence Velázquez wrote: >> >>> On Wed, Oct 20, 2021, at 5:07 PM, Alex fxmbsw7 Ratchev wrote: >>> > you got other cmd suggestions i can run to help debugging ? im very >>> limited >>> > in bin.c debugging >>> >>> You could bisect the devel branch to determine whether your problem >>> was caused by a change to bash and, if so, which commit is responsible. >>> >>> >>> https://git-scm.com/book/en/v2/Git-Tools-Debugging-with-Git#_binary_search >>> https://git-scm.com/docs/git-bisect >>> >>> -- >>> vq >>> >>
Re: current bash.git.devel segfaults on my code
On 10/19/21 7:47 PM, Alex fxmbsw7 Ratchev wrote: any reply to this ? There's nothing helpful there, and nothing that might point to any problem, much less a solution. -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
Re: current bash.git.devel segfaults on my code
On 10/20/21 4:57 PM, Alex fxmbsw7 Ratchev wrote: look, i do ./bash it segfaults, also when make valgrind, then i do valgrind ./bash it shows what i pasted the segfault didnt happen a month or two ago The way you present this makes it impossible for anyone but you to debug it. -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
Re: current bash.git.devel segfaults on my code
i see no good .. again such a stuckcase On Thu, Oct 21, 2021, 02:21 Chet Ramey wrote: > On 10/19/21 7:47 PM, Alex fxmbsw7 Ratchev wrote: > > any reply to this ? > > There's nothing helpful there, and nothing that might point to any problem, > much less a solution. > > -- > ``The lyf so short, the craft so long to lerne.'' - Chaucer > ``Ars longa, vita brevis'' - Hippocrates > Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/ >
Re: current bash.git.devel segfaults on my code
On 10/20/21 8:24 PM, Alex fxmbsw7 Ratchev wrote: erm sorry what The issue, if any, is obviously triggered by something in your environment. You can run bash --noprofile --norc to verify that. No one else has access to everything in your environment. I could not reproduce the crash with the parts of your startup scripts you posted. You are the only one who can reproduce this crash. There is little anyone can do to help you at this point without you taking the lead to isolate the crash to either a specific piece of shell code or a specific change in bash. -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
Re: current bash.git.devel segfaults on my code
i see makes sense i was hoping for finished cmds to test / save debugging info , rather than for example this git docs i can say it for anyone, ..imho 99 % to all docs are.. not for me like how to check bash devel version and version history to mark as good / bad .. is one small question about that all On Thu, Oct 21, 2021, 02:24 Chet Ramey wrote: > On 10/20/21 8:22 PM, Alex fxmbsw7 Ratchev wrote: > > i see > > > > no good > > > > .. > > > > again such a stuckcase > > If you're the only one who can reproduce it, you're going to have to take > the lead in debugging it. > > -- > ``The lyf so short, the craft so long to lerne.'' - Chaucer > ``Ars longa, vita brevis'' - Hippocrates > Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/ >
Re: current bash.git.devel segfaults on my code
erm sorry what On Thu, Oct 21, 2021, 02:22 Chet Ramey wrote: > On 10/20/21 4:57 PM, Alex fxmbsw7 Ratchev wrote: > > look, i do ./bash it segfaults, also when make valgrind, then i do > valgrind > > ./bash it shows what i pasted > > the segfault didnt happen a month or two ago > > The way you present this makes it impossible for anyone but you to debug > it. > > -- > ``The lyf so short, the craft so long to lerne.'' - Chaucer > ``Ars longa, vita brevis'' - Hippocrates > Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/ >
Re: current bash.git.devel segfaults on my code
On 10/20/21 8:22 PM, Alex fxmbsw7 Ratchev wrote: i see no good .. again such a stuckcase If you're the only one who can reproduce it, you're going to have to take the lead in debugging it. -- ``The lyf so short, the craft so long to lerne.'' - Chaucer ``Ars longa, vita brevis'' - Hippocrates Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/
Re: current bash.git.devel segfaults on my code
erm im not sure.. it is my rc that bashlinker script i wrote in mails topic 'segfaults on >my< code' in specific on alike arr=( "$path"$glob ) i cannot test much else, i code this only, other bash scripts of others i dunno and till they can code dont wanna know so i can only code on my stuff it used to work new bash compilations broke it as u may have noticed i posted a few weirdnesses and fails already with.. no results to me i can post the tgz of my code here, u can . script it and maybe just maybe it will segf for u too but probably not tho a test its worth else what can i do, tell me status: new bash devel x times new, new sw maybe a bit beta, but official then my pro code, i code pro since 20 years ans it segfaults, where it uses to run, .. not much changes there were ao i got new sw and segf, .. .. can u at least test the tgz ? if that works for you what can i do besides i cant debug it cause i lack.c downgrade ? maybe no good solution the git bad good thing might show something .. On Thu, Oct 21, 2021, 02:27 Chet Ramey wrote: > On 10/20/21 8:24 PM, Alex fxmbsw7 Ratchev wrote: > > erm sorry what > > The issue, if any, is obviously triggered by something in your environment. > You can run bash --noprofile --norc to verify that. > > No one else has access to everything in your environment. I could not > reproduce the crash with the parts of your startup scripts you posted. > > You are the only one who can reproduce this crash. There is little anyone > can do to help you at this point without you taking the lead to isolate > the crash to either a specific piece of shell code or a specific change > in bash. > > -- > ``The lyf so short, the craft so long to lerne.'' - Chaucer > ``Ars longa, vita brevis'' - Hippocrates > Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/ >
Re: current bash.git.devel segfaults on my code
as for presentation, and me making it impossible to view ( greycat and others on irc spitted on me 'u code bad obfuscating only' where i dont at all as for presentation as said i tried as always make it in one text objective of few subjects .. On Thu, Oct 21, 2021, 02:34 Alex fxmbsw7 Ratchev wrote: > erm im not sure.. it is my rc that bashlinker script > i wrote in mails topic 'segfaults on >my< code' > in specific on alike arr=( "$path"$glob ) > > i cannot test much else, i code this only, other bash scripts of others i > dunno and till they can code dont wanna know > so i can only code on my stuff > > it used to work > new bash compilations broke it > as u may have noticed i posted a few weirdnesses and fails already > with.. no results to me > > i can post the tgz of my code here, u can . script it and maybe just maybe > it will segf for u too > but probably not > tho a test its worth > > else what can i do, tell me > status: new bash devel x times new, new sw maybe a bit beta, but official > then my pro code, i code pro since 20 years > ans it segfaults, where it uses to run, .. not much changes there were > > ao i got new sw and segf, .. .. > > can u at least test the tgz ? > > if that works for you > what can i do > besides i cant debug it cause i lack.c > > downgrade ? > maybe > no good solution > > the git bad good thing might show something > > .. > > On Thu, Oct 21, 2021, 02:27 Chet Ramey wrote: > >> On 10/20/21 8:24 PM, Alex fxmbsw7 Ratchev wrote: >> > erm sorry what >> >> The issue, if any, is obviously triggered by something in your >> environment. >> You can run bash --noprofile --norc to verify that. >> >> No one else has access to everything in your environment. I could not >> reproduce the crash with the parts of your startup scripts you posted. >> >> You are the only one who can reproduce this crash. There is little anyone >> can do to help you at this point without you taking the lead to isolate >> the crash to either a specific piece of shell code or a specific change >> in bash. >> >> -- >> ``The lyf so short, the craft so long to lerne.'' - Chaucer >> ``Ars longa, vita brevis'' - Hippocrates >> Chet Ramey, UTech, CWRUc...@case.eduhttp://tiswww.cwru.edu/~chet/ >> >
Re: current bash.git.devel segfaults on my code
fix the alias bug to a release as u said and when that is im willig to skip all devel for release On Thu, Oct 21, 2021, 02:37 Alex fxmbsw7 Ratchev wrote: > as for presentation, and me making it impossible to view ( greycat and > others on irc spitted on me 'u code bad obfuscating only' where i dont at > all > > as for presentation as said i tried as always make it in one text > objective of few subjects > > .. > > On Thu, Oct 21, 2021, 02:34 Alex fxmbsw7 Ratchev > wrote: > >> erm im not sure.. it is my rc that bashlinker script >> i wrote in mails topic 'segfaults on >my< code' >> in specific on alike arr=( "$path"$glob ) >> >> i cannot test much else, i code this only, other bash scripts of others i >> dunno and till they can code dont wanna know >> so i can only code on my stuff >> >> it used to work >> new bash compilations broke it >> as u may have noticed i posted a few weirdnesses and fails already >> with.. no results to me >> >> i can post the tgz of my code here, u can . script it and maybe just >> maybe it will segf for u too >> but probably not >> tho a test its worth >> >> else what can i do, tell me >> status: new bash devel x times new, new sw maybe a bit beta, but official >> then my pro code, i code pro since 20 years >> ans it segfaults, where it uses to run, .. not much changes there were >> >> ao i got new sw and segf, .. .. >> >> can u at least test the tgz ? >> >> if that works for you >> what can i do >> besides i cant debug it cause i lack.c >> >> downgrade ? >> maybe >> no good solution >> >> the git bad good thing might show something >> >> .. >> >> On Thu, Oct 21, 2021, 02:27 Chet Ramey wrote: >> >>> On 10/20/21 8:24 PM, Alex fxmbsw7 Ratchev wrote: >>> > erm sorry what >>> >>> The issue, if any, is obviously triggered by something in your >>> environment. >>> You can run bash --noprofile --norc to verify that. >>> >>> No one else has access to everything in your environment. I could not >>> reproduce the crash with the parts of your startup scripts you posted. >>> >>> You are the only one who can reproduce this crash. There is little anyone >>> can do to help you at this point without you taking the lead to isolate >>> the crash to either a specific piece of shell code or a specific change >>> in bash. >>> >>> -- >>> ``The lyf so short, the craft so long to lerne.'' - Chaucer >>> ``Ars longa, vita brevis'' - Hippocrates >>> Chet Ramey, UTech, CWRUc...@case.edu >>> http://tiswww.cwru.edu/~chet/ >>> >>
Re: current bash.git.devel segfaults on my code
On Wed, Oct 20, 2021, at 7:43 PM, Alex fxmbsw7 Ratchev wrote: > to tell u honestly, im in no state to read freely happily new docs I'll be honest as well: If you're not going to put in effort beyond running commands that are spoon-fed to you, then I'm not going to put in effort beyond dumping URLs into emails. Of course, others who don't mind this asymmetry are free to provide further assistance. > if u dont specificly help with the cmds, i cannot succeed much That's unfortunate. -- vq
Re: current bash.git.devel segfaults on my code
u're as friendly as the psychiatry ppl fare well On Thu, Oct 21, 2021, 02:57 Lawrence Velázquez wrote: > On Wed, Oct 20, 2021, at 7:43 PM, Alex fxmbsw7 Ratchev wrote: > > to tell u honestly, im in no state to read freely happily new docs > > I'll be honest as well: If you're not going to put in effort beyond > running commands that are spoon-fed to you, then I'm not going to > put in effort beyond dumping URLs into emails. > > Of course, others who don't mind this asymmetry are free to provide > further assistance. > > > if u dont specificly help with the cmds, i cannot succeed much > > That's unfortunate. > > -- > vq >
