rtems-tools: DWARF5 support in Elftoolchain

Covoar has been unable to run after upgrading to GCC 12 because by default it generates DWARF5 debug info. Covoar uses libdwarf from Elftoolchain to read the debug info, and libdwarf only supports DWARF versions 2 through 4. I submitted a ticket notifying Elftoolchain about this issue (https:

[PATCH v1] bsps/amd64: remove -Werror from ABI flags

From: Stephen Clark The ABI flags for the amd64 BSP contain the -Werror=return-type flag. There is no reason for this to be there so it has been removed. The same option has also been removed amd64.cfg file. --- bsps/x86_64/amd64/config/amd64.cfg | 1 - spec/build/bsps/x86_64/amd64/abi.yml |

Re: [PATCH] bsps/xilinx/versal: Remove IPL32 BSPs, add aiedge and change defaults

Looks ok. Push it. On Mon, Aug 22, 2022 at 1:54 AM wrote: > From: Chris Johns > > - Versal has IO mapped to the upper 64bit address space and > needs full 64bit addresses. > > - Add xilinx_versal_aiedge for custom hardware > > - Make the hardware settings the defaults and qemu as variants > >

[PATCH v1 0/2] Change to QEMU 5

Hi, For this patchset I changed the hash of the glib hash because it has changed and it doesn't work on Debian 11, Ubuntu 22.04, and CentOS 8 Stream. After fixing the glib hash, I continued to run into other errors. So I brought in the config files for QEMU 5 since all the previously mentioned OS

[PATCH v1 2/2] Add QEMU 5 configuration files

--- bare/config/devel/qemu-5.2.0-1.cfg | 42 +++ bare/config/devel/qemu-couverture.bset | 2 +- bare/config/devel/qemu.bset | 2 +- source-builder/config/qemu-5-1.cfg | 9 ++ source-builder/config/qemu-common-2.cfg | 145 5 files changed,

[PATCH v1 1/2] glib-2.48.2-1.cfg: Fix checksum of patch

--- bare/config/devel/glib-2.48.2-1.cfg | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/bare/config/devel/glib-2.48.2-1.cfg b/bare/config/devel/glib-2.48.2-1.cfg index 76927e9..397a2c9 100644 --- a/bare/config/devel/glib-2.48.2-1.cfg +++ b/bare/config/devel/glib-2.48.2-1.cfg @

[PATCH] Change DWARF version to 5

This is a termporary fix while we wait for it to be fixed up stream. Covoar will not run without this at the moment. We don't know if the error occurred with any other Elf tools. Updates #4707 --- rtemstoolkit/elftoolchain/libdwarf/libdwarf_info.c | 2 +- 1 file changed, 1 insertion(+), 1 deletio

Re: [PATCH] Change DWARF version to 5

On 23/8/2022 6:31 am, Ryan Long wrote: > This is a termporary fix while we wait for it to be fixed up stream. Covoar > will not run without this at the moment. We don't know if the error occurred > with any other Elf tools. > > Updates #4707 > --- > rtemstoolkit/elftoolchain/libdwarf/libdwarf_inf

Re: rtems-tools: DWARF5 support in Elftoolchain

On 23/8/2022 2:39 am, Ryan Long wrote: > Covoar has been unable to run after upgrading to GCC 12 because by default it > generates DWARF5 debug info. Covoar uses libdwarf from Elftoolchain to read > the > debug info, and libdwarf only supports DWARF versions 2 through 4. > > I submitted a ticket

Re: [PATCH v1 0/2] Change to QEMU 5

OK Thanks Chris On 23/8/2022 6:29 am, Ryan Long wrote: > Hi, > > For this patchset I changed the hash of the glib hash because it has changed > and it doesn't work on Debian 11, Ubuntu 22.04, and CentOS 8 Stream. > > After fixing the glib hash, I continued to run into other errors. So I brought