This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
GDB port for RISC-V
- From: Palmer Dabbelt <palmer at dabbelt dot com>
- To: Andrew Waterman <andrew at sifive dot com>
- To: gdb-patches at sourceware dot org
- To: amodra at gmail dot com
- Date: Sat, 22 Oct 2016 16:39:50 -0700
- Subject: GDB port for RISC-V
- Authentication-results: sourceware.org; auth=none
- References: <1477179037-32066-1-git-send-email-palmer@dabbelt.com>
This message is "In-Reply-To" a larger patch set that contains a whole binutils
port for RISC-V. During our original patch submission it was suggested we send
the patches relevant to GDB to this mailing list, which is why you might not
have seen the v1 patches.
While there are a few things left to do on our end, I'd like to submit a v2 of
the patch set (the first submission to the GDB mailing list) just so we're all
on the same page. The exact commit I'm sending is availiable here:
https://github.com/riscv/riscv-binutils-gdb/commit/a949ca81ca7eca890e27d85d729203253d7862dd
while the whole binutils port, including both the patches submitted to the
binutils mailing list and the autogenerated files necessary for me to build the
RISC-V port, is availiable here:
https://github.com/riscv/riscv-binutils-gdb/tree/riscv-for-upstream-v2
Aside from review on this mailing list, there are a few problems pointed out by
our first submission that haven't been addresses yet:
* riscv-opc.h is autogenerated from a Python script. We will either clean up
this script enough to be suitable for submission, or will write some CPP
that produces a similar result.
* I forgot to fix copyright dates, but I've already made it this far into the
email... The plan is to only extend copyrights, and mark every committed
file as changed in 2016 if it's not already.
Additionally, there are a few issues on our github that are relevant, but I'm
not sure if they block upstreaming:
* https://github.com/riscv/riscv-gnu-toolchain/issues/124 : "sizeof(long
double)" is 64-bit on RISC-V, which precludes the use of the Q extension for
128-bit floating point. This may cause an ABI-breakage.
* https://github.com/riscv/riscv-gnu-toolchain/pull/136 : The path for ld.so
might change. This would definately be an ABI change.
Thanks to everyone who has provided feedback from the other list, hopefully
this one will be as painless :).
[PATCH 1/2] RISC-V GDB Port
[PATCH 2/2] RISC-V GDB Simulator Port