https://fedoraproject.org/wiki/Changes/GNUToolchainF38
This document represents a proposed Change. As part of the Changes process, proposals are publicly announced in order to receive community feedback. This proposal will only be implemented if approved by the Fedora Engineering Steering Committee.
== Summary == Update the Fedora 38 GNU Toolchain to gcc 13.0, binutils 2.39, and glibc 2.37.
The existing gdb 12.1 will be used as-is.
The set of core GNU Toolchain packages for Fedora 38 are as follows:
* GNU C Compiler 13.0 ** Associated runtimes for C++ (libstdc++), Go (gccgo), OpenMP (gomp), Fortran (gfortran), D (phobos), Objective C/C++. * GNU Binary Utilities 2.39 * GNU C Library 2.37 * GNU Debugger 12.1 (immediately available in Fedora 37)
The gcc 13.0 change will be tracked in this top-level GNU Toolchain system-wide update.
The binutils 2.39 change will be tracked in this top-level GNU Toolchain system-wide update.
The glibc 2.37 change will be tracked in this top-level GNU Toolchain system-wide update.
== Owner == * Name: [[User:codonell|Carlos O'Donell]] * Email: carlos@redhat.com
== Detailed Description == The GNU Compiler Collection, GNU Binary Utilities, GNU C Library, and the GNU Debugger make up the core part of the GNU Toolchain and it is useful for our users to transition these components as a complete implementation when making a new release of Fedora.
The GNU Compiler Collection is expected to release version 13.0, after the Fedora 38 release. It will contain many new features, documented here: https://gcc.gnu.org/gcc-13/changes.html. The latest release candidate for gcc 13 will be included in Fedora 38 and will be updated when released.
The GNU Binutils version 2.39 was released before Fedora 38; and we have already been using this version of binutils in Fedora Rawhide successfully to build the distribution for the last 4 months. Given the present schedule for Fedora 38 we will continue to use Binutils 2.39.
The GNU C Library version 2.37 is expected to be release before Fedora 38; we have started closely tracking the glibc 2.37 development code in Fedora Rawhide and are addressing any issues as they arise. Given the present schedule Fedora 38 will branch after the release of glibc 2.37. However, the mass rebuild schedule means Fedora 38 will mass rebuild (if required) before the final release of glibc 2.37, but after the ABI is frozen.
The GNU Debugger version 12.1 was released before Fedora 38; and we plan to continue to use this version of the debugger.
== Benefit to Fedora == Stays up to date with latest features, improvements, security and bug fixes from gcc, glibc, binutils, and gdb upstream.
The goal is to track and transition to the latest components of the GNU Toolchain.
== Scope == * Proposal owners: Fedora Toolchain Team (gcc, glibc, binutils, gdb, ...) developers need to ensure that gcc, glibc, binutils, and gdb in rawhide are stable and ready for the Fedora 38 branch.
* Other developers: Given that glibc is backwards compatible and we have been testing the new glibc in rawhide it should make very little impact when updated, except for the occasional deprecation warnings and removal of legacy interfaces from public header files.
* Release engineering: A mass rebuild is strongly encouraged; [https://pagure.io/releng/issue/XX #XX] ** Filed after approval.
* Policies and guidelines: N/A (not needed for this Change) * Trademark approval: N/A (not needed for this Change) * Alignment with Objectives: N/A
== Upgrade/compatibility impact == The compiler, the static linker and the the library are backwards compatible with the previous version of Fedora.
Some source changes may be required for the gcc 13 update. Please refer to the latest changes here: https://gcc.gnu.org/gcc-13/changes.html
Any source level changes required for glibc 2.37 will be noted here: https://sourceware.org/glibc/wiki/Release/2.37#Packaging_Changes
== How To Test == The GNU Compiler Collection has its own testsuite which is run during the package build and examined by the gcc developers before being uploaded.
The GNU C Library has its own testsuite which is run during the package build and examined by the glibc developers before being uploaded. This test suite has over 6200 tests that run to verify the correct operation of the library. In the future we may also run the microbenchmark to look for performance regressions.
The GNU Binutils has its own testsuite which is run during the package build and examined by binutils developers before being uploaded. The regression testsuite is run to verify the correct operation of the static linker and attendant utilities.
The GNU Debugger has its own testsuite which is run during the package build and examined by gdb developers before being uploaded. The regression testsuite is run to verify the correct operation of the debugger.
== User Experience == Fedora developers as well as developers using the distribution will be able to use and develop using the new features offered by the updated components. Developers will need to enable specific compiler features as required e.g. '-mcpu=neoverse-v2'.
== Dependencies == All packages do not need to be rebuilt due to backwards compatibility. However, it is advantageous if a mass rebuild is performed during the Fedora 38 cycle. The mass rebuild would ensure all packages can be built with the newer compiler and core runtime.
== Contingency Plan ==
* Contingency mechanism glibc: If glibc 2.37 proves too disruptive to compiling the distribution we could revert to 2.36, but given that Rawhide has started tracking glibc 2.37, no show-stopper problems are expected. At this point we can still revert to upstream version 2.36 if insurmountable problems appear, but to do so may require a mass rebuild to remove new symbols from the ABI/API.
* Contingency mechanism binutils: If binutils 2.39 proves too distruptive to assembling and linking the distribution we could revert to 2.38, but given that Rawhide is using 2.39, no show-stopper problems are expected. At this point we can still revert if insurmountable problems appear, but to do so may require a mass rebuild if the defects involve generated binaries.
* Contingency mechanism for gcc: If gcc 13 proves too disruptive to compiling the distribution we could revert to gcc 12.
* Contingency mechanism for gdb: The gdb 12.1 release is already in all the Fedora releases and it would not be reverted. If any gcc, glibc or binutils changes cause gdb to fail then that would need to be reviewed on a case-by-case basis.
* Contingency deadline: Fedora mass rebuild on 2023-01-18. * Blocks release? ** Yes, upgrading to gcc 13.0 does block the release. ** Yes, upgrading to binutils 2.39 does block the release. ** Yes, upgrading to glibc 2.37 does block the release. ** No, upgrading to gdb 12.1 does block the release (already released).
== Documentation == The gcc manual contains the documentation for the release and doesn't need any more additional work.
The binutils manual contains the documentation for the release and doesn't need any more additional work.
The glibc manual contains the documentation for the release and doesn't need any more additional work.
The gdb manual contains the documentation for the release and doesn't need any more additional work.
== Release Notes == <!-- Use this text for GCC updates: -->
See https://gcc.gnu.org/gcc-13/changes.html for the GNU Compiler Collection version 13 release notes.
<!-- Use this text for GLIBC updates: --> The GNU C Library version 2.37 will be released at the beginning of February 2023. The current NEWS notes can be seen here as they are added: https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS;hb=HEAD
The GNU Binary Utilities version 2.39 was released August 2022. The current release notes will be sent to the developer mailing list: https://sourceware.org/pipermail/binutils/2022-August/122246.html.
On Fri, Dec 30, 2022 at 2:02 PM Ben Cotton bcotton@redhat.com wrote:
https://fedoraproject.org/wiki/Changes/GNUToolchainF38
This document represents a proposed Change. As part of the Changes process, proposals are publicly announced in order to receive community feedback. This proposal will only be implemented if approved by the Fedora Engineering Steering Committee.
== Summary == Update the Fedora 38 GNU Toolchain to gcc 13.0, binutils 2.39, and glibc 2.37.
The existing gdb 12.1 will be used as-is.
The set of core GNU Toolchain packages for Fedora 38 are as follows:
- GNU C Compiler 13.0
** Associated runtimes for C++ (libstdc++), Go (gccgo), OpenMP (gomp), Fortran (gfortran), D (phobos), Objective C/C++.
- GNU Binary Utilities 2.39
- GNU C Library 2.37
- GNU Debugger 12.1 (immediately available in Fedora 37)
The gcc 13.0 change will be tracked in this top-level GNU Toolchain system-wide update.
The binutils 2.39 change will be tracked in this top-level GNU Toolchain system-wide update.
The glibc 2.37 change will be tracked in this top-level GNU Toolchain system-wide update.
== Owner ==
- Name: [[User:codonell|Carlos O'Donell]]
- Email: carlos@redhat.com
== Detailed Description == The GNU Compiler Collection, GNU Binary Utilities, GNU C Library, and the GNU Debugger make up the core part of the GNU Toolchain and it is useful for our users to transition these components as a complete implementation when making a new release of Fedora.
The GNU Compiler Collection is expected to release version 13.0, after the Fedora 38 release. It will contain many new features, documented here: https://gcc.gnu.org/gcc-13/changes.html. The latest release candidate for gcc 13 will be included in Fedora 38 and will be updated when released.
The GNU Binutils version 2.39 was released before Fedora 38; and we have already been using this version of binutils in Fedora Rawhide successfully to build the distribution for the last 4 months. Given the present schedule for Fedora 38 we will continue to use Binutils 2.39.
The GNU C Library version 2.37 is expected to be release before Fedora 38; we have started closely tracking the glibc 2.37 development code in Fedora Rawhide and are addressing any issues as they arise. Given the present schedule Fedora 38 will branch after the release of glibc 2.37. However, the mass rebuild schedule means Fedora 38 will mass rebuild (if required) before the final release of glibc 2.37, but after the ABI is frozen.
The GNU Debugger version 12.1 was released before Fedora 38; and we plan to continue to use this version of the debugger.
== Benefit to Fedora == Stays up to date with latest features, improvements, security and bug fixes from gcc, glibc, binutils, and gdb upstream.
The goal is to track and transition to the latest components of the GNU Toolchain.
== Scope ==
- Proposal owners: Fedora Toolchain Team (gcc, glibc, binutils, gdb,
...) developers need to ensure that gcc, glibc, binutils, and gdb in rawhide are stable and ready for the Fedora 38 branch.
- Other developers: Given that glibc is backwards compatible and we
have been testing the new glibc in rawhide it should make very little impact when updated, except for the occasional deprecation warnings and removal of legacy interfaces from public header files.
- Release engineering: A mass rebuild is strongly encouraged;
[https://pagure.io/releng/issue/XX #XX] ** Filed after approval.
- Policies and guidelines: N/A (not needed for this Change)
- Trademark approval: N/A (not needed for this Change)
- Alignment with Objectives: N/A
== Upgrade/compatibility impact == The compiler, the static linker and the the library are backwards compatible with the previous version of Fedora.
Some source changes may be required for the gcc 13 update. Please refer to the latest changes here: https://gcc.gnu.org/gcc-13/changes.html
Any source level changes required for glibc 2.37 will be noted here: https://sourceware.org/glibc/wiki/Release/2.37#Packaging_Changes
== How To Test == The GNU Compiler Collection has its own testsuite which is run during the package build and examined by the gcc developers before being uploaded.
The GNU C Library has its own testsuite which is run during the package build and examined by the glibc developers before being uploaded. This test suite has over 6200 tests that run to verify the correct operation of the library. In the future we may also run the microbenchmark to look for performance regressions.
The GNU Binutils has its own testsuite which is run during the package build and examined by binutils developers before being uploaded. The regression testsuite is run to verify the correct operation of the static linker and attendant utilities.
The GNU Debugger has its own testsuite which is run during the package build and examined by gdb developers before being uploaded. The regression testsuite is run to verify the correct operation of the debugger.
== User Experience == Fedora developers as well as developers using the distribution will be able to use and develop using the new features offered by the updated components. Developers will need to enable specific compiler features as required e.g. '-mcpu=neoverse-v2'.
== Dependencies == All packages do not need to be rebuilt due to backwards compatibility. However, it is advantageous if a mass rebuild is performed during the Fedora 38 cycle. The mass rebuild would ensure all packages can be built with the newer compiler and core runtime.
== Contingency Plan ==
- Contingency mechanism glibc: If glibc 2.37 proves too disruptive to
compiling the distribution we could revert to 2.36, but given that Rawhide has started tracking glibc 2.37, no show-stopper problems are expected. At this point we can still revert to upstream version 2.36 if insurmountable problems appear, but to do so may require a mass rebuild to remove new symbols from the ABI/API.
- Contingency mechanism binutils: If binutils 2.39 proves too
distruptive to assembling and linking the distribution we could revert to 2.38, but given that Rawhide is using 2.39, no show-stopper problems are expected. At this point we can still revert if insurmountable problems appear, but to do so may require a mass rebuild if the defects involve generated binaries.
- Contingency mechanism for gcc: If gcc 13 proves too disruptive to
compiling the distribution we could revert to gcc 12.
- Contingency mechanism for gdb: The gdb 12.1 release is already in
all the Fedora releases and it would not be reverted. If any gcc, glibc or binutils changes cause gdb to fail then that would need to be reviewed on a case-by-case basis.
- Contingency deadline: Fedora mass rebuild on 2023-01-18.
- Blocks release?
** Yes, upgrading to gcc 13.0 does block the release. ** Yes, upgrading to binutils 2.39 does block the release. ** Yes, upgrading to glibc 2.37 does block the release. ** No, upgrading to gdb 12.1 does block the release (already released).
== Documentation == The gcc manual contains the documentation for the release and doesn't need any more additional work.
The binutils manual contains the documentation for the release and doesn't need any more additional work.
The glibc manual contains the documentation for the release and doesn't need any more additional work.
The gdb manual contains the documentation for the release and doesn't need any more additional work.
== Release Notes ==
<!-- Use this text for GCC updates: -->
See https://gcc.gnu.org/gcc-13/changes.html for the GNU Compiler Collection version 13 release notes.
<!-- Use this text for GLIBC updates: -->
The GNU C Library version 2.37 will be released at the beginning of February 2023. The current NEWS notes can be seen here as they are added: https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS;hb=HEAD
The GNU Binary Utilities version 2.39 was released August 2022. The current release notes will be sent to the developer mailing list: https://sourceware.org/pipermail/binutils/2022-August/122246.html.
Can we please have gcc-rs also built (even though it's experimental)?
-- 真実はいつも一つ!/ Always, there's only one truth!
On 12/30/22 14:12, Neal Gompa wrote:
On Fri, Dec 30, 2022 at 2:02 PM Ben Cotton bcotton@redhat.com wrote:
https://fedoraproject.org/wiki/Changes/GNUToolchainF38
This document represents a proposed Change. As part of the Changes process, proposals are publicly announced in order to receive community feedback. This proposal will only be implemented if approved by the Fedora Engineering Steering Committee.
== Summary == Update the Fedora 38 GNU Toolchain to gcc 13.0, binutils 2.39, and glibc 2.37.
The existing gdb 12.1 will be used as-is.
The set of core GNU Toolchain packages for Fedora 38 are as follows:
- GNU C Compiler 13.0
** Associated runtimes for C++ (libstdc++), Go (gccgo), OpenMP (gomp), Fortran (gfortran), D (phobos), Objective C/C++.
- GNU Binary Utilities 2.39
- GNU C Library 2.37
- GNU Debugger 12.1 (immediately available in Fedora 37)
The gcc 13.0 change will be tracked in this top-level GNU Toolchain system-wide update.
The binutils 2.39 change will be tracked in this top-level GNU Toolchain system-wide update.
The glibc 2.37 change will be tracked in this top-level GNU Toolchain system-wide update.
== Owner ==
- Name: [[User:codonell|Carlos O'Donell]]
- Email: carlos@redhat.com
== Detailed Description == The GNU Compiler Collection, GNU Binary Utilities, GNU C Library, and the GNU Debugger make up the core part of the GNU Toolchain and it is useful for our users to transition these components as a complete implementation when making a new release of Fedora.
The GNU Compiler Collection is expected to release version 13.0, after the Fedora 38 release. It will contain many new features, documented here: https://gcc.gnu.org/gcc-13/changes.html. The latest release candidate for gcc 13 will be included in Fedora 38 and will be updated when released.
The GNU Binutils version 2.39 was released before Fedora 38; and we have already been using this version of binutils in Fedora Rawhide successfully to build the distribution for the last 4 months. Given the present schedule for Fedora 38 we will continue to use Binutils 2.39.
The GNU C Library version 2.37 is expected to be release before Fedora 38; we have started closely tracking the glibc 2.37 development code in Fedora Rawhide and are addressing any issues as they arise. Given the present schedule Fedora 38 will branch after the release of glibc 2.37. However, the mass rebuild schedule means Fedora 38 will mass rebuild (if required) before the final release of glibc 2.37, but after the ABI is frozen.
The GNU Debugger version 12.1 was released before Fedora 38; and we plan to continue to use this version of the debugger.
== Benefit to Fedora == Stays up to date with latest features, improvements, security and bug fixes from gcc, glibc, binutils, and gdb upstream.
The goal is to track and transition to the latest components of the GNU Toolchain.
== Scope ==
- Proposal owners: Fedora Toolchain Team (gcc, glibc, binutils, gdb,
...) developers need to ensure that gcc, glibc, binutils, and gdb in rawhide are stable and ready for the Fedora 38 branch.
- Other developers: Given that glibc is backwards compatible and we
have been testing the new glibc in rawhide it should make very little impact when updated, except for the occasional deprecation warnings and removal of legacy interfaces from public header files.
- Release engineering: A mass rebuild is strongly encouraged;
[https://pagure.io/releng/issue/XX #XX] ** Filed after approval.
- Policies and guidelines: N/A (not needed for this Change)
- Trademark approval: N/A (not needed for this Change)
- Alignment with Objectives: N/A
== Upgrade/compatibility impact == The compiler, the static linker and the the library are backwards compatible with the previous version of Fedora.
Some source changes may be required for the gcc 13 update. Please refer to the latest changes here: https://gcc.gnu.org/gcc-13/changes.html
Any source level changes required for glibc 2.37 will be noted here: https://sourceware.org/glibc/wiki/Release/2.37#Packaging_Changes
== How To Test == The GNU Compiler Collection has its own testsuite which is run during the package build and examined by the gcc developers before being uploaded.
The GNU C Library has its own testsuite which is run during the package build and examined by the glibc developers before being uploaded. This test suite has over 6200 tests that run to verify the correct operation of the library. In the future we may also run the microbenchmark to look for performance regressions.
The GNU Binutils has its own testsuite which is run during the package build and examined by binutils developers before being uploaded. The regression testsuite is run to verify the correct operation of the static linker and attendant utilities.
The GNU Debugger has its own testsuite which is run during the package build and examined by gdb developers before being uploaded. The regression testsuite is run to verify the correct operation of the debugger.
== User Experience == Fedora developers as well as developers using the distribution will be able to use and develop using the new features offered by the updated components. Developers will need to enable specific compiler features as required e.g. '-mcpu=neoverse-v2'.
== Dependencies == All packages do not need to be rebuilt due to backwards compatibility. However, it is advantageous if a mass rebuild is performed during the Fedora 38 cycle. The mass rebuild would ensure all packages can be built with the newer compiler and core runtime.
== Contingency Plan ==
- Contingency mechanism glibc: If glibc 2.37 proves too disruptive to
compiling the distribution we could revert to 2.36, but given that Rawhide has started tracking glibc 2.37, no show-stopper problems are expected. At this point we can still revert to upstream version 2.36 if insurmountable problems appear, but to do so may require a mass rebuild to remove new symbols from the ABI/API.
- Contingency mechanism binutils: If binutils 2.39 proves too
distruptive to assembling and linking the distribution we could revert to 2.38, but given that Rawhide is using 2.39, no show-stopper problems are expected. At this point we can still revert if insurmountable problems appear, but to do so may require a mass rebuild if the defects involve generated binaries.
- Contingency mechanism for gcc: If gcc 13 proves too disruptive to
compiling the distribution we could revert to gcc 12.
- Contingency mechanism for gdb: The gdb 12.1 release is already in
all the Fedora releases and it would not be reverted. If any gcc, glibc or binutils changes cause gdb to fail then that would need to be reviewed on a case-by-case basis.
- Contingency deadline: Fedora mass rebuild on 2023-01-18.
- Blocks release?
** Yes, upgrading to gcc 13.0 does block the release. ** Yes, upgrading to binutils 2.39 does block the release. ** Yes, upgrading to glibc 2.37 does block the release. ** No, upgrading to gdb 12.1 does block the release (already released).
== Documentation == The gcc manual contains the documentation for the release and doesn't need any more additional work.
The binutils manual contains the documentation for the release and doesn't need any more additional work.
The glibc manual contains the documentation for the release and doesn't need any more additional work.
The gdb manual contains the documentation for the release and doesn't need any more additional work.
== Release Notes ==
<!-- Use this text for GCC updates: -->
See https://gcc.gnu.org/gcc-13/changes.html for the GNU Compiler Collection version 13 release notes.
<!-- Use this text for GLIBC updates: -->
The GNU C Library version 2.37 will be released at the beginning of February 2023. The current NEWS notes can be seen here as they are added: https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS;hb=HEAD
The GNU Binary Utilities version 2.39 was released August 2022. The current release notes will be sent to the developer mailing list: https://sourceware.org/pipermail/binutils/2022-August/122246.html.
Can we please have gcc-rs also built (even though it's experimental)?
gcc-rs is not capable of even compiling hello world, so it would be rather pointless to package it. Even once gcc-rs is finished, I suspect rustc-codegen-gcc will be a better long-term solution for architectures LLVM does not support, as it reuses the existing rustc frontend.
On Fri, Dec 30, 2022 at 7:57 PM Demi Marie Obenour demiobenour@gmail.com wrote:
On 12/30/22 14:12, Neal Gompa wrote:
On Fri, Dec 30, 2022 at 2:02 PM Ben Cotton bcotton@redhat.com wrote:
https://fedoraproject.org/wiki/Changes/GNUToolchainF38
This document represents a proposed Change. As part of the Changes process, proposals are publicly announced in order to receive community feedback. This proposal will only be implemented if approved by the Fedora Engineering Steering Committee.
== Summary == Update the Fedora 38 GNU Toolchain to gcc 13.0, binutils 2.39, and glibc 2.37.
The existing gdb 12.1 will be used as-is.
The set of core GNU Toolchain packages for Fedora 38 are as follows:
- GNU C Compiler 13.0
** Associated runtimes for C++ (libstdc++), Go (gccgo), OpenMP (gomp), Fortran (gfortran), D (phobos), Objective C/C++.
- GNU Binary Utilities 2.39
- GNU C Library 2.37
- GNU Debugger 12.1 (immediately available in Fedora 37)
The gcc 13.0 change will be tracked in this top-level GNU Toolchain system-wide update.
The binutils 2.39 change will be tracked in this top-level GNU Toolchain system-wide update.
The glibc 2.37 change will be tracked in this top-level GNU Toolchain system-wide update.
== Owner ==
- Name: [[User:codonell|Carlos O'Donell]]
- Email: carlos@redhat.com
== Detailed Description == The GNU Compiler Collection, GNU Binary Utilities, GNU C Library, and the GNU Debugger make up the core part of the GNU Toolchain and it is useful for our users to transition these components as a complete implementation when making a new release of Fedora.
The GNU Compiler Collection is expected to release version 13.0, after the Fedora 38 release. It will contain many new features, documented here: https://gcc.gnu.org/gcc-13/changes.html. The latest release candidate for gcc 13 will be included in Fedora 38 and will be updated when released.
The GNU Binutils version 2.39 was released before Fedora 38; and we have already been using this version of binutils in Fedora Rawhide successfully to build the distribution for the last 4 months. Given the present schedule for Fedora 38 we will continue to use Binutils 2.39.
The GNU C Library version 2.37 is expected to be release before Fedora 38; we have started closely tracking the glibc 2.37 development code in Fedora Rawhide and are addressing any issues as they arise. Given the present schedule Fedora 38 will branch after the release of glibc 2.37. However, the mass rebuild schedule means Fedora 38 will mass rebuild (if required) before the final release of glibc 2.37, but after the ABI is frozen.
The GNU Debugger version 12.1 was released before Fedora 38; and we plan to continue to use this version of the debugger.
== Benefit to Fedora == Stays up to date with latest features, improvements, security and bug fixes from gcc, glibc, binutils, and gdb upstream.
The goal is to track and transition to the latest components of the GNU Toolchain.
== Scope ==
- Proposal owners: Fedora Toolchain Team (gcc, glibc, binutils, gdb,
...) developers need to ensure that gcc, glibc, binutils, and gdb in rawhide are stable and ready for the Fedora 38 branch.
- Other developers: Given that glibc is backwards compatible and we
have been testing the new glibc in rawhide it should make very little impact when updated, except for the occasional deprecation warnings and removal of legacy interfaces from public header files.
- Release engineering: A mass rebuild is strongly encouraged;
[https://pagure.io/releng/issue/XX #XX] ** Filed after approval.
- Policies and guidelines: N/A (not needed for this Change)
- Trademark approval: N/A (not needed for this Change)
- Alignment with Objectives: N/A
== Upgrade/compatibility impact == The compiler, the static linker and the the library are backwards compatible with the previous version of Fedora.
Some source changes may be required for the gcc 13 update. Please refer to the latest changes here: https://gcc.gnu.org/gcc-13/changes.html
Any source level changes required for glibc 2.37 will be noted here: https://sourceware.org/glibc/wiki/Release/2.37#Packaging_Changes
== How To Test == The GNU Compiler Collection has its own testsuite which is run during the package build and examined by the gcc developers before being uploaded.
The GNU C Library has its own testsuite which is run during the package build and examined by the glibc developers before being uploaded. This test suite has over 6200 tests that run to verify the correct operation of the library. In the future we may also run the microbenchmark to look for performance regressions.
The GNU Binutils has its own testsuite which is run during the package build and examined by binutils developers before being uploaded. The regression testsuite is run to verify the correct operation of the static linker and attendant utilities.
The GNU Debugger has its own testsuite which is run during the package build and examined by gdb developers before being uploaded. The regression testsuite is run to verify the correct operation of the debugger.
== User Experience == Fedora developers as well as developers using the distribution will be able to use and develop using the new features offered by the updated components. Developers will need to enable specific compiler features as required e.g. '-mcpu=neoverse-v2'.
== Dependencies == All packages do not need to be rebuilt due to backwards compatibility. However, it is advantageous if a mass rebuild is performed during the Fedora 38 cycle. The mass rebuild would ensure all packages can be built with the newer compiler and core runtime.
== Contingency Plan ==
- Contingency mechanism glibc: If glibc 2.37 proves too disruptive to
compiling the distribution we could revert to 2.36, but given that Rawhide has started tracking glibc 2.37, no show-stopper problems are expected. At this point we can still revert to upstream version 2.36 if insurmountable problems appear, but to do so may require a mass rebuild to remove new symbols from the ABI/API.
- Contingency mechanism binutils: If binutils 2.39 proves too
distruptive to assembling and linking the distribution we could revert to 2.38, but given that Rawhide is using 2.39, no show-stopper problems are expected. At this point we can still revert if insurmountable problems appear, but to do so may require a mass rebuild if the defects involve generated binaries.
- Contingency mechanism for gcc: If gcc 13 proves too disruptive to
compiling the distribution we could revert to gcc 12.
- Contingency mechanism for gdb: The gdb 12.1 release is already in
all the Fedora releases and it would not be reverted. If any gcc, glibc or binutils changes cause gdb to fail then that would need to be reviewed on a case-by-case basis.
- Contingency deadline: Fedora mass rebuild on 2023-01-18.
- Blocks release?
** Yes, upgrading to gcc 13.0 does block the release. ** Yes, upgrading to binutils 2.39 does block the release. ** Yes, upgrading to glibc 2.37 does block the release. ** No, upgrading to gdb 12.1 does block the release (already released).
== Documentation == The gcc manual contains the documentation for the release and doesn't need any more additional work.
The binutils manual contains the documentation for the release and doesn't need any more additional work.
The glibc manual contains the documentation for the release and doesn't need any more additional work.
The gdb manual contains the documentation for the release and doesn't need any more additional work.
== Release Notes ==
<!-- Use this text for GCC updates: -->
See https://gcc.gnu.org/gcc-13/changes.html for the GNU Compiler Collection version 13 release notes.
<!-- Use this text for GLIBC updates: -->
The GNU C Library version 2.37 will be released at the beginning of February 2023. The current NEWS notes can be seen here as they are added: https://sourceware.org/git/?p=glibc.git;a=blob;f=NEWS;hb=HEAD
The GNU Binary Utilities version 2.39 was released August 2022. The current release notes will be sent to the developer mailing list: https://sourceware.org/pipermail/binutils/2022-August/122246.html.
Can we please have gcc-rs also built (even though it's experimental)?
gcc-rs is not capable of even compiling hello world, so it would be rather pointless to package it. Even once gcc-rs is finished, I suspect rustc-codegen-gcc will be a better long-term solution for architectures LLVM does not support, as it reuses the existing rustc frontend.
I am perfectly aware of the state of gcc-rs. I've been following it quite closely for years. I explicitly asked for it *despite* its current state. It already doesn't let you use it unless you pass a really long flag that states you know that this is experimental and shouldn't be relied on yet.
I want gcc-rs available so that people get exposed to a Rust compiler that deliberately *does not* use the rustc interface.
-- 真実はいつも一つ!/ Always, there's only one truth!
Neal Gompa wrote:
Can we please have gcc-rs also built (even though it's experimental)?
Will gcc-rs be able to generate usable shared libraries for Rust crates?
Kevin Kofler
On Fri, Dec 30, 2022 at 9:37 PM Kevin Kofler via devel devel@lists.fedoraproject.org wrote:
Neal Gompa wrote:
Can we please have gcc-rs also built (even though it's experimental)?
Will gcc-rs be able to generate usable shared libraries for Rust crates?
If someone were to spend the time to build the functionality into its code generator, sure. I don't think that's high on anyone's list right now, though.
On Sat, Dec 31, 2022 at 3:48 AM Neal Gompa ngompa13@gmail.com wrote:
On Fri, Dec 30, 2022 at 9:37 PM Kevin Kofler via devel devel@lists.fedoraproject.org wrote:
Neal Gompa wrote:
Can we please have gcc-rs also built (even though it's experimental)?
Will gcc-rs be able to generate usable shared libraries for Rust crates?
If someone were to spend the time to build the functionality into its code generator, sure. I don't think that's high on anyone's list right now, though.
rustc can already produce shared libraries - they're just pretty useless due to two factors:
- lack of stable ABI: for every compiler and dependency update, you'd need to recompile everything. And unless work on a stable ABI progresses in upstream Rust, I doubt that gcc-rs can do anything about this. - incompatible compile-time options (i.e. resulting in conditional compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
Fabio
Fabio Valentini wrote:
- incompatible compile-time options (i.e. resulting in conditional
compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
The application can pick the options with which each library is compiled? What a stupid idea! Now I understand why the language is called "Rust".
Kevin Kofler
On Mon, Jan 02, 2023 at 01:59:46AM +0100, Kevin Kofler via devel wrote:
The application can pick the options with which each library is compiled? What a stupid idea! Now I understand why the language is called "Rust".
Okay. no. This is not how we do things here.
Matthew Miller wrote:
Okay. no. This is not how we do things here.
Apologies for my snide remark that visibly came out rude, sorry.
Kevin Kofler
On Mon, Jan 02, 2023 at 01:06:52PM +0100, Kevin Kofler via devel wrote:
Matthew Miller wrote:
Okay. no. This is not how we do things here.
Apologies for my snide remark that visibly came out rude, sorry.
Thank you, Kevin.
Fabio Valentini wrote:
- incompatible compile-time options (i.e. resulting in conditional
compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
Let me try formulating my criticism more constructively (since my previous reply failed both at being polite and at getting my point through, sorry again for that):
I am really surprised to read above that Rust apparently allows applications to pick the flag with which the libraries they depend on are compiled. I really have to wonder why anyone would think that allowing that would be a good idea, but then again I guess I know the answer: Whoever added this feature was so set in a mindset where everything is compiled on demand and statically linked that they figured: why not?
And if you are in that mindset, that actually sounds like a reasonable call to make. Source-based software distributions do have the advantage of offering this kind of flexibility on demand, see also the USE flags in Gentoo. Those are in fact one of the main reasons some people decide to compile an entire GNU/Linux distribution from source (and hence pick a distribution such as Gentoo) to begin with. Likewise, the Rust way of compiling dependencies on demand allows applications to make this kind of settings for them.
Still, I can see several issues with that approach, e.g., what if an application depends on two libraries A and B that both depend on library C, but with conflicting flags? But the main issue is that, as you point out, it makes binary distribution of shared libraries highly impractical. That is why I think this was a short-sighted design decision.
But we will have to work around this one way or another, because I doubt anyone will be willing to remove that questionable feature now that developers have come to rely on it. (And no, I do not think the current Fedora approach of packaging crates in source form only is the optimal approach, for reasons I have already pointed out in other threads on this list.)
I hope that the above now brings my point across constructively.
Kevin Kofler
On 1/2/23 22:30, Kevin Kofler via devel wrote:
Fabio Valentini wrote:
- incompatible compile-time options (i.e. resulting in conditional
compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
Let me try formulating my criticism more constructively (since my previous reply failed both at being polite and at getting my point through, sorry again for that):
I am really surprised to read above that Rust apparently allows applications to pick the flag with which the libraries they depend on are compiled. I really have to wonder why anyone would think that allowing that would be a good idea, but then again I guess I know the answer: Whoever added this feature was so set in a mindset where everything is compiled on demand and statically linked that they figured: why not?
One of the major uses is to allow code that requires a particular dependency to be disabled when that dependency is not available. In particular, Rust targets platforms (such as OS kernels and embedded systems) where the standard library is not available. This would be extremely difficult without Cargo features.
And if you are in that mindset, that actually sounds like a reasonable call to make. Source-based software distributions do have the advantage of offering this kind of flexibility on demand, see also the USE flags in Gentoo. Those are in fact one of the main reasons some people decide to compile an entire GNU/Linux distribution from source (and hence pick a distribution such as Gentoo) to begin with. Likewise, the Rust way of compiling dependencies on demand allows applications to make this kind of settings for them.
Still, I can see several issues with that approach, e.g., what if an application depends on two libraries A and B that both depend on library C, but with conflicting flags?
Last I checked, Cargo features are additive, so the answer is that C will be compiled with the union of all flags used by A and B.
But the main issue is that, as you point out, it makes binary distribution of shared libraries highly impractical. That is why I think this was a short-sighted design decision.
Cargo features are supposed to be additive, so one can sometimes ship a single package with the union of all features used by its reverse dependencies. This must be handled on a case-by-case basis, though.
On Mon, Jan 2, 2023 at 10:52 PM Demi Marie Obenour demiobenour@gmail.com wrote:
On 1/2/23 22:30, Kevin Kofler via devel wrote:
Fabio Valentini wrote:
- incompatible compile-time options (i.e. resulting in conditional
compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
Let me try formulating my criticism more constructively (since my previous reply failed both at being polite and at getting my point through, sorry again for that):
I am really surprised to read above that Rust apparently allows applications to pick the flag with which the libraries they depend on are compiled. I really have to wonder why anyone would think that allowing that would be a good idea, but then again I guess I know the answer: Whoever added this feature was so set in a mindset where everything is compiled on demand and statically linked that they figured: why not?
One of the major uses is to allow code that requires a particular dependency to be disabled when that dependency is not available. In particular, Rust targets platforms (such as OS kernels and embedded systems) where the standard library is not available. This would be extremely difficult without Cargo features.
platforms != features. That said, some crates do support "nostd" as a feature flag, others don't. It depends.
And if you are in that mindset, that actually sounds like a reasonable call to make. Source-based software distributions do have the advantage of offering this kind of flexibility on demand, see also the USE flags in Gentoo. Those are in fact one of the main reasons some people decide to compile an entire GNU/Linux distribution from source (and hence pick a distribution such as Gentoo) to begin with. Likewise, the Rust way of compiling dependencies on demand allows applications to make this kind of settings for them.
Still, I can see several issues with that approach, e.g., what if an application depends on two libraries A and B that both depend on library C, but with conflicting flags?
Last I checked, Cargo features are additive, so the answer is that C will be compiled with the union of all flags used by A and B.
But the main issue is that, as you point out, it makes binary distribution of shared libraries highly impractical. That is why I think this was a short-sighted design decision.
Cargo features are supposed to be additive, so one can sometimes ship a single package with the union of all features used by its reverse dependencies. This must be handled on a case-by-case basis, though.
Even if it wasn't, building a library in various feature modes would be possible. Cargo would just need to know which binary object to select to link with, which I imagine it would learn how to do if someone cared about shared libraries in Rust upstream.
Dne 03. 01. 23 v 4:30 Kevin Kofler via devel napsal(a):
Fabio Valentini wrote:
- incompatible compile-time options (i.e. resulting in conditional
compilation): different packages depend on crates with different sets of features enabled, sometimes with conflicting options. Even with a stable ABI, you'd need to build crates for all necessary combinations of configurations, and that matrix quickly explodes (i.e. usually exponentially - 2^n builds for for n independent flags). This is a deal-breaker for shared libraries in most cases, and also can't be solved by using a different compiler. (Unless you want to figure out *which* combinations to build, and *only* build these.)
I wish this was specific to Rust. We more or less deal with this issue in every ecosystem. We just sometimes choose to ignore the whole parts of the matrix, e.g. statically/dynamically linked libraries. But also support for some extensions, such as e.g. language bindings and what not. Or maybe platform support.
Let me try formulating my criticism more constructively (since my previous reply failed both at being polite and at getting my point through, sorry again for that):
I am really surprised to read above that Rust apparently allows applications to pick the flag with which the libraries they depend on are compiled. I really have to wonder why anyone would think that allowing that would be a good idea, but then again I guess I know the answer: Whoever added this feature was so set in a mindset where everything is compiled on demand and statically linked that they figured: why not?
And if you are in that mindset, that actually sounds like a reasonable call to make. Source-based software distributions do have the advantage of offering this kind of flexibility on demand, see also the USE flags in Gentoo. Those are in fact one of the main reasons some people decide to compile an entire GNU/Linux distribution from source (and hence pick a distribution such as Gentoo) to begin with. Likewise, the Rust way of compiling dependencies on demand allows applications to make this kind of settings for them.
Still, I can see several issues with that approach, e.g., what if an application depends on two libraries A and B that both depend on library C, but with conflicting flags? But the main issue is that, as you point out, it makes binary distribution of shared libraries highly impractical. That is why I think this was a short-sighted design decision.
But we will have to work around this one way or another, because I doubt anyone will be willing to remove that questionable feature now that developers have come to rely on it. (And no, I do not think the current Fedora approach of packaging crates in source form only is the optimal approach, for reasons I have already pointed out in other threads on this list.)
We don't need to remove this feature, just limit the scope into acceptable size.
Vít
I hope that the above now brings my point across constructively.
Kevin Kofler
devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-leave@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
I wonder if we should drop the cloog package from Fedora. It was separated from gcc so that both gcc and cross-gcc could use it. However neither of them now do.
Also, the isl package was split out from gcc so that cross-gcc could use it also, but gcc now seems to be carrying its own isl package of a different version (0.12.2 rather than 0.16.1). Do we still build with isl support or could this also be dropped?
David
On Tue, Jan 10, 2023 at 8:14 AM David Howells dhowells@redhat.com wrote:
Also, the isl package was split out from gcc so that cross-gcc could use it also, but gcc now seems to be carrying its own isl package of a different version (0.12.2 rather than 0.16.1). Do we still build with isl support or could this also be dropped?
The sagemath package would like to build with isl support, but needs version 0.12.0 or later. There is a "BuildRequires: pkgconfig(isl)" in sagemath.spec in case the system version is ever updated, but right now the build system rejects it as being too old.