[tcpdump-workers] Re: tcpdump and pcapng with comments
Flex is not getting built. We have the binary checked in to our repository. "which flex" gives the same path : tools/arm-5.3-uclibc-1.0.12/usr/bin/flex On Mon, Apr 7, 2025 at 2:19 PM Guy Harris wrote: > On Apr 6, 2025, at 11:19 PM, Mahesh V > wrote: > > > configure:7921: checking for flex > > configure:7937: found tools/arm-5.3-uclibc-1.0.12/usr/bin/flex > > configure:7948: result: flex > > configure:7986: flex conftest.l > > flex: fatal internal error, exec of > /var/tmp/tool-chains/buildroot-2016.02-5.3-arm/crosstools-arm-gcc-5.3-linux-4.1-uclibc-1.0.12-binutils-2.25-NPTL/usr/bin/m4 > failed > > > > which m4 > > # which m4 > > tools/arm-5.3-uclibc-1.0.12/usr/bin/m4 > > "which" gave you a *relative* path? > > But that path appears to differ from the path that flex tried to use. > > What does "which flex" print? > > > m4 exists and I have set M4PATH env variable as well; however, it seems > to pick it up from where flex was built before. > > Why was flex built at all? Why not just rely on the system flex or the > cross-build flex that comes with the OS? ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
[tcpdump-workers] Re: tcpdump and pcapng with comments
--- Begin Message --- On Mon, 7 Apr 2025 11:49:49 +0530 Mahesh V wrote: > m4 exists and I have set M4PATH env variable as well In Autoconf "M4" is the optional environment variable with a custom path to m4. -- Denis Ovsienko --- End Message --- ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
[tcpdump-workers] Re: tcpdump and pcapng with comments
I did not run autoconf. Only thing that I did was run the following 1) configure 2) make Is it possible to set M4 using some other means? On Mon, Apr 7, 2025 at 1:40 PM Denis Ovsienko via tcpdump-workers < tcpdump-workers@lists.tcpdump.org> wrote: > > > > -- Forwarded message -- > From: Denis Ovsienko > To: tcpdump-workers@lists.tcpdump.org > Cc: > Bcc: > Date: Mon, 7 Apr 2025 09:09:57 +0100 > Subject: [tcpdump-workers] Re: tcpdump and pcapng with comments > On Mon, 7 Apr 2025 11:49:49 +0530 > Mahesh V wrote: > > > m4 exists and I have set M4PATH env variable as well > > In Autoconf "M4" is the optional environment variable with a custom path > to m4. > > -- > Denis Ovsienko > > > > -- Forwarded message -- > From: Denis Ovsienko via tcpdump-workers < > tcpdump-workers@lists.tcpdump.org> > To: tcpdump-workers@lists.tcpdump.org > Cc: > Bcc: > Date: Mon, 7 Apr 2025 09:09:57 +0100 > Subject: [tcpdump-workers] Re: tcpdump and pcapng with comments > ___ > tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org > To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
[tcpdump-workers] Re: tcpdump and pcapng with comments
On Apr 7, 2025, at 2:11 AM, Mahesh V wrote: > Flex is not getting built. We have the binary checked in to our repository. What process produced that binary? Is it a binary you compiled, is it a copy of a binary from the OS, or is it from somewhere else? ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
[tcpdump-workers] Re: tcpdump and pcapng with comments
On Apr 6, 2025, at 11:19 PM, Mahesh V wrote: > configure:7921: checking for flex > configure:7937: found tools/arm-5.3-uclibc-1.0.12/usr/bin/flex > configure:7948: result: flex > configure:7986: flex conftest.l > flex: fatal internal error, exec of > /var/tmp/tool-chains/buildroot-2016.02-5.3-arm/crosstools-arm-gcc-5.3-linux-4.1-uclibc-1.0.12-binutils-2.25-NPTL/usr/bin/m4 > failed > > which m4 > # which m4 > tools/arm-5.3-uclibc-1.0.12/usr/bin/m4 "which" gave you a *relative* path? But that path appears to differ from the path that flex tried to use. What does "which flex" print? > m4 exists and I have set M4PATH env variable as well; however, it seems to > pick it up from where flex was built before. Why was flex built at all? Why not just rely on the system flex or the cross-build flex that comes with the OS? ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
[tcpdump-workers] Re: tcpdump and pcapng with comments
--- Begin Message --- On 06/04/2025 23:28, Michael Richardson wrote: > > Mahesh V wrote: > > I added some code (modified) tcpdump to write the pcapng file. > > while configuring/compiling the source code I get this error > > This is a cross compilation for ARM platform > > Well, likely the resulting flex test can't be run, since it's cross-compiled. > > I suggest *NOT* doing cross-compilation while you are working on new code. +1 --- End Message --- ___ tcpdump-workers mailing list -- tcpdump-workers@lists.tcpdump.org To unsubscribe send an email to tcpdump-workers-le...@lists.tcpdump.org %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
