Added dynamic TargetDescription
#3780
Conversation
theIDinside
force-pushed
the
dynamic-target-descriptions
branch
from
a571187 to
c276f0e
Compare
|
@rocallahan || @khuey |
|
Prior to this patch, I have no idea how RR informed GDB on |
src/GdbServerConnection.cc
Outdated
| @@ -111,7 +112,8 @@ unique_ptr<GdbServerConnection> GdbServerConnection::await_connection( | |||
| Task* t, ScopedFd& listen_fd, const GdbServerConnection::Features& features) { | |||
| auto dbg = unique_ptr<GdbServerConnection>( | |||
| new GdbServerConnection(t->thread_group()->tguid(), features)); | |||
| dbg->set_cpu_features(get_cpu_features(t->arch())); | |||
| const auto arch = t->arch(); | |||
| dbg->set_cpu_features(arch, get_cpu_features(arch)); | |||
There was a problem hiding this comment.
You can just drop this hunk.
src/TargetDescription.cc
Outdated
| } | ||
| } | ||
|
|
||
| static constexpr auto Header = R"(<?xml version="1.0"?> |
There was a problem hiding this comment.
Don't need to make this constexpr or auto. Also, name should be header or possibly HEADER. E.g. use
static const char header[] = ...
There was a problem hiding this comment.
... or else just write it inline, that's fine
There was a problem hiding this comment.
Changed it to header, but refrained from inlining it directly as raw string literals make code look weird, though that's a personal taste though, and I'll gladly change if it you like.
|
Also, it looks like tests are failing in CI on x86-64? |
Where can I see these tests? It says "All checks have passed" here. |
| #include "TargetDescription.h" | ||
| #include "GdbServerConnection.h" | ||
| #include "kernel_abi.h" | ||
| #include <sstream> |
There was a problem hiding this comment.
Order #includes like our other files: first TargetDescription.h, then the C++ library headers, then the rr headers.
Yep, that looks good! |
|
@theIDinside as far as I understood the review that's nearly finished and only needs minor tweaks. It would be nice if this could be finished as it is specified as a dependency of #3776 which (after rebase including these changes) is a real nice addition. |
TargetDescription dynamically generates the "master" target description contents (target.xml), so that we only have to manage the individual concrete CPU features (like foo-sse.xml, for instance). TargetDescription will generate the xml contents that includes the used features. Removed combinatorial target.xml files These are now handled by `TargetDescription` instead. This unblocks the rr-debugger#3776 PR (AVX512 support) from getting pulled in, though that PR will need to take this PR into account.
I don't even know how we have sent this before or why this hasn't been a bug prior to this.
theIDinside
force-pushed
the
dynamic-target-descriptions
branch
from
b194918 to
1155d34
Compare
I'm on it |
|
I fixed the final nits, squshed and merged: f7f4c29 Thanks!!! |
|
currently this is in a separate branch only... the referenced #3776 needs to be rebased on this (and likely still target the master) |
|
Oops I pushed to the wrong branch. Fixed. |
TargetDescription dynamically generates the "master" target description contents (target.xml), so that we only have to manage the individual concrete CPU features (like foo-sse.xml, for instance). TargetDescription will generate the xml contents that includes the used features.
Removed combinatorial target.xml files
These are now handled by
TargetDescriptioninstead.This unblocks the #3776 PR (AVX512 support) from getting pulled in, though that PR will need to take this PR into account.
Note:
The
GdbServerRegistershould probably be rewritten/removed and then rely on aTargetDescriptionthat provides a "static" array ofGdbServerValuesince once we know the target description, it won't change during a replay, so dynamically allocating like we do now (indispatch_regs_request) is not really necessary. But these are just loose thoughts I have, I haven't really verified that this would be a better approach.