Bug#972665: ICE building firefox-esr/firefox in unstable on armhf
On Thu, Oct 22, 2020 at 11:12:14AM +0200, Matthias Klose wrote: > On 10/22/20 9:57 AM, Mike Hommey wrote: > >> The code in question hasn't changed in a long time, so the same code > >> effectively built fine with older versions of gcc. The last successful > >> builds of firefox-esr and firefox were using gcc-10_10.2.0-9. > >> > >> I see in the gcc-10 changelog there is "On armel, armhf, configure with > >> --enable-checking=yes,extra,rtl", this could be related. > > no, that was added to track down that kind of issue. Do you see the example > failing every time, or is it just random? every time. Mike
Bug#972665: ICE building firefox-esr/firefox in unstable on armhf
On 10/22/20 9:57 AM, Mike Hommey wrote: >> The code in question hasn't changed in a long time, so the same code >> effectively built fine with older versions of gcc. The last successful >> builds of firefox-esr and firefox were using gcc-10_10.2.0-9. >> >> I see in the gcc-10 changelog there is "On armel, armhf, configure with >> --enable-checking=yes,extra,rtl", this could be related. no, that was added to track down that kind of issue. Do you see the example failing every time, or is it just random? Matthias
Bug#972665: ICE building firefox-esr/firefox in unstable on armhf
And _of course_ I forgot to attach the file. On Thu, Oct 22, 2020 at 04:50:58PM +0900, Mike Hommey wrote: > Package: gcc-10 > Version: 10.2.0-15 > Severity: serious > > See the full buildd log in > https://buildd.debian.org/status/fetch.php?pkg=firefox-esr&arch=armhf&ver=78.4.0esr-2&stamp=1603260510&raw=0 > > The preprocessed source that fails is attached as intrapred_neon.i.gz. > > This is reproducible on armhf with: >gcc -o intrapred_neon.o -c -O1 -mfpu=neon intrapred_neon.i > > The shortened ICE message looks like: > > during RTL pass: reload > /home/glandium/firefox-esr-78.4.0esr/third_party/aom/aom_dsp/arm/intrapred_neon.c: > In function 'aom_highbd_dc_predictor_4x4_neon': > /home/glandium/firefox-esr-78.4.0esr/third_party/aom/aom_dsp/arm/intrapred_neon.c:589:188: > internal compiler error: in decompose_automod_address, at rtlanal.c:6298 > 589 | intra_pred_square(dc); > | > >^ > 0x6cda45 decompose_automod_address > ../../src/gcc/rtlanal.c:6298 > 0x6cda45 decompose_address(address_info*, rtx_def**, machine_mode, unsigned > char, rtx_code) > ../../src/gcc/rtlanal.c:6463 > 0x6cdb19 decompose_mem_address(address_info*, rtx_def*) > ../../src/gcc/rtlanal.c:6486 > 0x59a67b process_address_1 > ../../src/gcc/lra-constraints.c:3374 > 0x59c4cd process_address > ../../src/gcc/lra-constraints.c:3648 > 0x59c4cd curr_insn_transform > ../../src/gcc/lra-constraints.c:3963 > 0x5a088f lra_constraints(bool) > ../../src/gcc/lra-constraints.c:5036 > 0x58e7a5 lra(_IO_FILE*) > ../../src/gcc/lra.c:2443 > 0x54f331 do_reload > ../../src/gcc/ira.c:5527 > 0x54f331 execute > ../../src/gcc/ira.c:5713 > Please submit a full bug report, > with preprocessed source if appropriate. > Please include the complete backtrace with any bug report. > See for instructions. > > > The code in question hasn't changed in a long time, so the same code > effectively built fine with older versions of gcc. The last successful > builds of firefox-esr and firefox were using gcc-10_10.2.0-9. > > I see in the gcc-10 changelog there is "On armel, armhf, configure with > --enable-checking=yes,extra,rtl", this could be related. > > Mike intrapred_neon.i.gz Description: application/gzip
Bug#972665: ICE building firefox-esr/firefox in unstable on armhf
Package: gcc-10 Version: 10.2.0-15 Severity: serious See the full buildd log in https://buildd.debian.org/status/fetch.php?pkg=firefox-esr&arch=armhf&ver=78.4.0esr-2&stamp=1603260510&raw=0 The preprocessed source that fails is attached as intrapred_neon.i.gz. This is reproducible on armhf with: gcc -o intrapred_neon.o -c -O1 -mfpu=neon intrapred_neon.i The shortened ICE message looks like: during RTL pass: reload /home/glandium/firefox-esr-78.4.0esr/third_party/aom/aom_dsp/arm/intrapred_neon.c: In function 'aom_highbd_dc_predictor_4x4_neon': /home/glandium/firefox-esr-78.4.0esr/third_party/aom/aom_dsp/arm/intrapred_neon.c:589:188: internal compiler error: in decompose_automod_address, at rtlanal.c:6298 589 | intra_pred_square(dc); | ^ 0x6cda45 decompose_automod_address ../../src/gcc/rtlanal.c:6298 0x6cda45 decompose_address(address_info*, rtx_def**, machine_mode, unsigned char, rtx_code) ../../src/gcc/rtlanal.c:6463 0x6cdb19 decompose_mem_address(address_info*, rtx_def*) ../../src/gcc/rtlanal.c:6486 0x59a67b process_address_1 ../../src/gcc/lra-constraints.c:3374 0x59c4cd process_address ../../src/gcc/lra-constraints.c:3648 0x59c4cd curr_insn_transform ../../src/gcc/lra-constraints.c:3963 0x5a088f lra_constraints(bool) ../../src/gcc/lra-constraints.c:5036 0x58e7a5 lra(_IO_FILE*) ../../src/gcc/lra.c:2443 0x54f331 do_reload ../../src/gcc/ira.c:5527 0x54f331 execute ../../src/gcc/ira.c:5713 Please submit a full bug report, with preprocessed source if appropriate. Please include the complete backtrace with any bug report. See for instructions. The code in question hasn't changed in a long time, so the same code effectively built fine with older versions of gcc. The last successful builds of firefox-esr and firefox were using gcc-10_10.2.0-9. I see in the gcc-10 changelog there is "On armel, armhf, configure with --enable-checking=yes,extra,rtl", this could be related. Mike