This is next in my series of "fix the racey tests that fail on
greendragon" addressing the failure of TestConcurrentManyBreakpoints.py
where we set a breakpoint in a function that 100 threads execute, and we
check that we hit the breakpoint 100 times. But sometimes it is only hit
99 times, and the test fails.
When we hit a software breakpoint, the pc value for the thread is the
address of the breakpoint instruction - as if it had not been hit yet.
And because a user might ADD a breakpoint for the current pc from the
commandline, when we go to resume execution, any thread that is sitting
at a breakpoint site will be silently advanced past the breakpoint
instruction (disable bp, instruction step that thread, re-enable bp)
before resuming -- whether that thread has hit its breakpoint or not.
What this test is exposing is that there is another corner case, a
thread that is sitting at a breakpoint site but has not yet executed the
breakpoint instruction. The thread will have no stop reason, no mach
exception, so it will not be recorded as having hit the breakpoint
(because it hasn't yet). But when we resume execution, because it is
sitting at a breakpoint site, we advance past it and miss the breakpoint
hit.
In 2016 Abhishek Aggarwal handled a similar issue with a patch in
`ProcessGDBRemote::SetThreadStopInfo()`, adding a breakpoint StopInfo
for a thread sitting at a breakpoint site that has no stop reason.
debugserver's `jThreadsInfo` would not correctly execute Abhishek's code
though because it would respond with `"reason":"none"` for a thread with
no stop reason, and `SetThreadStopInfo()` expected an empty reason here.
The first part of my patch is to clear the `reason` if it is `"none"` so
we flow through the code correctly.
On Darwin, though, our stop reply packet (Txx...) includes the
`threads`, `thread-pcs`, and `jstopinfo` keys, which give us the tids
for all current threads, the pc values for those threads, and
`jstopinfo` has a JSON dictionary with the mach exceptions for all
threads that have a mach exception. In
`ProcessGDBRemote::CalculateThreadStopInfo()` we set the StopInfo for
each thread for a private stop and if we have `jstopinfo` it is the
source of all the StopInfos. I have to add the same logic here, to give
the thread a breakpoint StopInfo even though it hasn't executed the
breakpoint yet. In this case we are very early in thread construction
and I only have the information in the Txx stop reply packet -- tids,
pcs, and jstopinfo, so I can't use the normal general mechanisms of
going through the RegisterContext to get the pc, it's a bit different.
If I hack debugserver to not issue `jstopinfo`,
`CalculateThreadStopInfo` will fall back to sending `qThreadStopInfo`
for each thread and going through
`ProcessGDBRemote::SetThreadStopInfo()` to set the stop infos (and with
the `reason:none` fix, use Abhishek's code).
rdar://110549165
debugserver on arm64 devices can manage both Byte Address Select
watchpoints (1-8 bytes) and MASK watchpoints (8 bytes-2 gigabytes). This
adds a SupportedWatchpointTypes key to the QSupported response from
debugserver with a list of these, so lldb can take full advantage of
them when creating larger regions with a single hardware watchpoint.
Also add documentation for this, and two other lldb extensions, to the
lldb-gdb-remote.txt documentation.
Re-enable TestLargeWatchpoint.py on Darwin systems when testing with the
in-tree built debugserver. I can remove the "in-tree built debugserver"
in the future when this new key is handled by an Xcode debugserver.
This patch is the next piece of work in my Large Watchpoint proposal,
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
This patch breaks a user's watchpoint into one or more
WatchpointResources which reflect what the hardware registers can cover.
This means we can watch objects larger than 8 bytes, and we can watched
unaligned address ranges. On a typical 64-bit target with 4 watchpoint
registers you can watch 32 bytes of memory if the start address is
doubleword aligned.
Additionally, if the remote stub implements AArch64 MASK style
watchpoints (e.g. debugserver on Darwin), we can watch any power-of-2
size region of memory up to 2GB, aligned to that same size.
I updated the Watchpoint constructor and CommandObjectWatchpoint to
create a CompilerType of Array<UInt8> when the size of the watched
region is greater than pointer-size and we don't have a variable type to
use. For pointer-size and smaller, we can display the watched granule as
an integer value; for larger-than-pointer-size we will display as an
array of bytes.
I have `watchpoint list` now print the WatchpointResources used to
implement the watchpoint.
I added a WatchpointAlgorithm class which has a top-level static method
that takes an enum flag mask WatchpointHardwareFeature and a user
address and size, and returns a vector of WatchpointResources covering
the request. It does not take into account the number of watchpoint
registers the target has, or the number still available for use. Right
now there is only one algorithm, which monitors power-of-2 regions of
memory. For up to pointer-size, this is what Intel hardware supports.
AArch64 Byte Address Select watchpoints can watch any number of
contiguous bytes in a pointer-size memory granule, that is not currently
supported so if you ask to watch bytes 3-5, the algorithm will watch the
entire doubleword (8 bytes). The newly default "modify" style means we
will silently ignore modifications to bytes outside the watched range.
I've temporarily skipped TestLargeWatchpoint.py for all targets. It was
only run on Darwin when using the in-tree debugserver, which was a proxy
for "debugserver supports MASK watchpoints". I'll be adding the
aforementioned feature flag from the stub and enabling full mask
watchpoints when a debugserver with that feature is enabled, and
re-enable this test.
I added a new TestUnalignedLargeWatchpoint.py which only has one test
but it's a great one, watching a 22-byte range that is unaligned and
requires four 8-byte watchpoints to cover.
I also added a unit test, WatchpointAlgorithmsTests, which has a number
of simple tests against WatchpointAlgorithms::PowerOf2Watchpoints. I
think there's interesting possible different approaches to how we cover
these; I note in the unit test that a user requesting a watch on address
0x12e0 of 120 bytes will be covered by two watchpoints today, a
128-bytes at 0x1280 and at 0x1300. But it could be done with a 16-byte
watchpoint at 0x12e0 and a 128-byte at 0x1300, which would have fewer
false positives/private stops. As we try refining this one, it's helpful
to have a collection of tests to make sure things don't regress.
I tested this on arm64 macOS, (genuine) x86_64 macOS, and AArch64
Ubuntu. I have not modifed the Windows process plugins yet, I might try
that as a standalone patch, I'd be making the change blind, but the
necessary changes (see ProcessGDBRemote::EnableWatchpoint) are pretty
small so it might be obvious enough that I can change it and see what
the Windows CI thinks.
There isn't yet a packet (or a qSupported feature query) for the gdb
remote serial protocol stub to communicate its watchpoint capabilities
to lldb. I'll be doing that in a patch right after this is landed,
having debugserver advertise its capability of AArch64 MASK watchpoints,
and have ProcessGDBRemote add eWatchpointHardwareArmMASK to
WatchpointAlgorithms so we can watch larger than 32-byte requests on
Darwin.
I haven't yet tackled WatchpointResource *sharing* by multiple
Watchpoints. This is all part of the goal, especially when we may be
watching a larger memory range than the user requested, if they then add
another watchpoint next to their first request, it may be covered by the
same WatchpointResource (hardware watchpoint register). Also one "read"
watchpoint and one "write" watchpoint on the same memory granule need to
be handled, making the WatchpointResource cover all requests.
As WatchpointResources aren't shared among multiple Watchpoints yet,
there's no handling of running the conditions/commands/etc on multiple
Watchpoints when their shared WatchpointResource is hit. The goal beyond
"large watchpoint" is to unify (much more) the Watchpoint and Breakpoint
behavior and commands. I have a feeling I may be slowly chipping away at
this for a while.
Re-landing this patch after fixing two undefined behaviors in
WatchpointAlgorithms found by UBSan and by failures on different
CI bots.
rdar://108234227
This patch is the next piece of work in my Large Watchpoint proposal,
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
This patch breaks a user's watchpoint into one or more
WatchpointResources which reflect what the hardware registers can cover.
This means we can watch objects larger than 8 bytes, and we can watched
unaligned address ranges. On a typical 64-bit target with 4 watchpoint
registers you can watch 32 bytes of memory if the start address is
doubleword aligned.
Additionally, if the remote stub implements AArch64 MASK style
watchpoints (e.g. debugserver on Darwin), we can watch any power-of-2
size region of memory up to 2GB, aligned to that same size.
I updated the Watchpoint constructor and CommandObjectWatchpoint to
create a CompilerType of Array<UInt8> when the size of the watched
region is greater than pointer-size and we don't have a variable type to
use. For pointer-size and smaller, we can display the watched granule as
an integer value; for larger-than-pointer-size we will display as an
array of bytes.
I have `watchpoint list` now print the WatchpointResources used to
implement the watchpoint.
I added a WatchpointAlgorithm class which has a top-level static method
that takes an enum flag mask WatchpointHardwareFeature and a user
address and size, and returns a vector of WatchpointResources covering
the request. It does not take into account the number of watchpoint
registers the target has, or the number still available for use. Right
now there is only one algorithm, which monitors power-of-2 regions of
memory. For up to pointer-size, this is what Intel hardware supports.
AArch64 Byte Address Select watchpoints can watch any number of
contiguous bytes in a pointer-size memory granule, that is not currently
supported so if you ask to watch bytes 3-5, the algorithm will watch the
entire doubleword (8 bytes). The newly default "modify" style means we
will silently ignore modifications to bytes outside the watched range.
I've temporarily skipped TestLargeWatchpoint.py for all targets. It was
only run on Darwin when using the in-tree debugserver, which was a proxy
for "debugserver supports MASK watchpoints". I'll be adding the
aforementioned feature flag from the stub and enabling full mask
watchpoints when a debugserver with that feature is enabled, and
re-enable this test.
I added a new TestUnalignedLargeWatchpoint.py which only has one test
but it's a great one, watching a 22-byte range that is unaligned and
requires four 8-byte watchpoints to cover.
I also added a unit test, WatchpointAlgorithmsTests, which has a number
of simple tests against WatchpointAlgorithms::PowerOf2Watchpoints. I
think there's interesting possible different approaches to how we cover
these; I note in the unit test that a user requesting a watch on address
0x12e0 of 120 bytes will be covered by two watchpoints today, a
128-bytes at 0x1280 and at 0x1300. But it could be done with a 16-byte
watchpoint at 0x12e0 and a 128-byte at 0x1300, which would have fewer
false positives/private stops. As we try refining this one, it's helpful
to have a collection of tests to make sure things don't regress.
I tested this on arm64 macOS, (genuine) x86_64 macOS, and AArch64
Ubuntu. I have not modifed the Windows process plugins yet, I might try
that as a standalone patch, I'd be making the change blind, but the
necessary changes (see ProcessGDBRemote::EnableWatchpoint) are pretty
small so it might be obvious enough that I can change it and see what
the Windows CI thinks.
There isn't yet a packet (or a qSupported feature query) for the gdb
remote serial protocol stub to communicate its watchpoint capabilities
to lldb. I'll be doing that in a patch right after this is landed,
having debugserver advertise its capability of AArch64 MASK watchpoints,
and have ProcessGDBRemote add eWatchpointHardwareArmMASK to
WatchpointAlgorithms so we can watch larger than 32-byte requests on
Darwin.
I haven't yet tackled WatchpointResource *sharing* by multiple
Watchpoints. This is all part of the goal, especially when we may be
watching a larger memory range than the user requested, if they then add
another watchpoint next to their first request, it may be covered by the
same WatchpointResource (hardware watchpoint register). Also one "read"
watchpoint and one "write" watchpoint on the same memory granule need to
be handled, making the WatchpointResource cover all requests.
As WatchpointResources aren't shared among multiple Watchpoints yet,
there's no handling of running the conditions/commands/etc on multiple
Watchpoints when their shared WatchpointResource is hit. The goal beyond
"large watchpoint" is to unify (much more) the Watchpoint and Breakpoint
behavior and commands. I have a feeling I may be slowly chipping away at
this for a while.
rdar://108234227
There are 3 ways to create an EventDataBytes object: (const char *),
(llvm::StringRef), and (const void *, size_t len). All of these cases
can be handled under `llvm::StringRef`. Additionally, this allows us to
remove the otherwise unused `SetBytes`, `SwapBytes`, and
`SetBytesFromCString` methods.
BroadcastEvent currently takes its EventData* param and shoves it into
an Event object, which takes ownership of the pointer and places it into
a shared_ptr to manage the lifetime.
Instead of relying on `new` and passing raw pointers around, I think it
would make more sense to create the shared_ptr up front.
This patch replaces uses of StringRef::{starts,ends}with with
StringRef::{starts,ends}_with for consistency with
std::{string,string_view}::{starts,ends}_with in C++20.
I'm planning to deprecate and eventually remove
StringRef::{starts,ends}with.
We need to generate events when finalizing, or we won't know that we
succeeded in stopping the process to detach/kill. Instead, we stall and
then after our 20 interrupt timeout, we kill the process (even if we
were supposed to detach) and exit.
OTOH, we have to not generate events when the Process is being
destructed because shared_from_this has already been torn down, and
using it will cause crashes.
This patch is rearranging code a bit to add WatchpointResources to
Process. A WatchpointResource is meant to represent a hardware
watchpoint register in the inferior process. It has an address, a size,
a type, and a list of Watchpoints that are using this
WatchpointResource.
This current patch doesn't add any of the features of
WatchpointResources that make them interesting -- a user asking to watch
a 24 byte object could watch this with three 8 byte WatchpointResources.
Or a Watchpoint on 1 byte at 0x1002 and a second watchpoint on 1 byte at
0x1003, these must both be served by a single WatchpointResource on that
doubleword at 0x1000 on a 64-bit target, if two hardware watchpoint
registers were used to track these separately, one of them may not be
hit. Or if you have one Watchpoint on a variable with a condition set,
and another Watchpoint on that same variable with a command defined or
different condition, or ignorecount, both of those Watchpoints need to
evaluate their criteria/commands when their WatchpointResource has been
hit.
There's a bit of code movement to rearrange things in the direction I'll
need for implementing this feature, so I want to start with reviewing &
landing this mostly NFC patch and we can focus on the algorithmic
choices about how WatchpointResources are shared and handled as they're
triggeed, separately.
This patch also stops printing "Watchpoint <n> hit: old value: <x>, new
vlaue: <y>" for Read watchpoints. I could make an argument for print
"Watchpoint <n> hit: current value <x>" but the current output doesn't
make any sense, and the user can print the value if they are
particularly interested. Read watchpoints are used primarily to
understand what code is reading a variable.
This patch adds more fallbacks for how to print the objects being
watched if we have types, instead of assuming they are all integral
values, so a struct will print its elements. As large watchpoints are
added, we'll be doing a lot more of those.
To track the WatchpointSP in the WatchpointResources, I changed the
internal API which took a WatchpointSP and devolved it to a Watchpoint*,
which meant touching several different Process files. I removed the
watchpoint code in ProcessKDP which only reported that watchpoints
aren't supported, the base class does that already.
I haven't yet changed how we receive a watchpoint to identify the
WatchpointResource responsible for the trigger, and identify all
Watchpoints that are using this Resource to evaluate their conditions
etc. This is the same work that a BreakpointSite needs to do when it has
been tiggered, where multiple Breakpoints may be at the same address.
There is not yet any printing of the Resources that a Watchpoint is
implemented in terms of ("watchpoint list", or
SBWatchpoint::GetDescription).
"watchpoint set var" and "watchpoint set expression" take a size
argument which was previously 1, 2, 4, or 8 (an enum). I've changed this
to an unsigned int. Most hardware implementations can only watch 1, 2,
4, 8 byte ranges, but with Resources we'll allow a user to ask for
different sized watchpoints and set them in hardware-expressble terms
soon.
I've annotated areas where I know there is work still needed with
LWP_TODO that I'll be working on once this is landed.
I've tested this on aarch64 macOS, aarch64 Linux, and Intel macOS.
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
(cherry picked from commit fc6b72523f3d73b921690a713e97a433c96066c6)
Currently when you interrupt a:
(lldb) process attach -w -n some_process
lldb just closes the connection to the stub and kills the
lldb_private::Process it made for the attach. The stub at the other end
notices the connection go down and exits because of that. But when
communication to a device is handled through some kind of proxy server
which isn't as well behaved as one would wish, that signal might not be
reliable, causing debugserver to persist on the machine, waiting to
steal the next instance of that process.
We can work around those failures by sending an explicit interrupt
before closing down the connection. The stub will also have to be
waiting for the interrupt for this to make any difference. I changed
debugserver to do that.
I didn't make the equivalent change in lldb-server. So long as you
aren't faced with a flakey connection, this should not be necessary.
...and follow ups.
As it has caused test failures on Linux Arm and AArch64:
https://lab.llvm.org/buildbot/#/builders/96/builds/49126https://lab.llvm.org/buildbot/#/builders/17/builds/45824
```
lldb-shell :: Subprocess/clone-follow-child-wp.test
lldb-shell :: Subprocess/fork-follow-child-wp.test
lldb-shell :: Subprocess/vfork-follow-child-wp.test
```
This reverts commit a6c62bf1a4717accc852463b664cd1012237d334,
commit a0a1ff3ab40e347589b4e27d8fd350c600526735 and commit
fc6b72523f3d73b921690a713e97a433c96066c6.
This patch is rearranging code a bit to add WatchpointResources to
Process. A WatchpointResource is meant to represent a hardware
watchpoint register in the inferior process. It has an address, a size,
a type, and a list of Watchpoints that are using this
WatchpointResource.
This current patch doesn't add any of the features of
WatchpointResources that make them interesting -- a user asking to watch
a 24 byte object could watch this with three 8 byte WatchpointResources.
Or a Watchpoint on 1 byte at 0x1002 and a second watchpoint on 1 byte at
0x1003, these must both be served by a single WatchpointResource on that
doubleword at 0x1000 on a 64-bit target, if two hardware watchpoint
registers were used to track these separately, one of them may not be
hit. Or if you have one Watchpoint on a variable with a condition set,
and another Watchpoint on that same variable with a command defined or
different condition, or ignorecount, both of those Watchpoints need to
evaluate their criteria/commands when their WatchpointResource has been
hit.
There's a bit of code movement to rearrange things in the direction I'll
need for implementing this feature, so I want to start with reviewing &
landing this mostly NFC patch and we can focus on the algorithmic
choices about how WatchpointResources are shared and handled as they're
triggeed, separately.
This patch also stops printing "Watchpoint <n> hit: old value: <x>, new
vlaue: <y>" for Read watchpoints. I could make an argument for print
"Watchpoint <n> hit: current value <x>" but the current output doesn't
make any sense, and the user can print the value if they are
particularly interested. Read watchpoints are used primarily to
understand what code is reading a variable.
This patch adds more fallbacks for how to print the objects being
watched if we have types, instead of assuming they are all integral
values, so a struct will print its elements. As large watchpoints are
added, we'll be doing a lot more of those.
To track the WatchpointSP in the WatchpointResources, I changed the
internal API which took a WatchpointSP and devolved it to a Watchpoint*,
which meant touching several different Process files. I removed the
watchpoint code in ProcessKDP which only reported that watchpoints
aren't supported, the base class does that already.
I haven't yet changed how we receive a watchpoint to identify the
WatchpointResource responsible for the trigger, and identify all
Watchpoints that are using this Resource to evaluate their conditions
etc. This is the same work that a BreakpointSite needs to do when it has
been tiggered, where multiple Breakpoints may be at the same address.
There is not yet any printing of the Resources that a Watchpoint is
implemented in terms of ("watchpoint list", or
SBWatchpoint::GetDescription).
"watchpoint set var" and "watchpoint set expression" take a size
argument which was previously 1, 2, 4, or 8 (an enum). I've changed this
to an unsigned int. Most hardware implementations can only watch 1, 2,
4, 8 byte ranges, but with Resources we'll allow a user to ask for
different sized watchpoints and set them in hardware-expressble terms
soon.
I've annotated areas where I know there is work still needed with
LWP_TODO that I'll be working on once this is landed.
I've tested this on aarch64 macOS, aarch64 Linux, and Intel macOS.
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
The Watchpoint and Breakpoint objects try to track the hardware index
that was used for them, if they are hardware wp/bp's. The majority of
our debugging goes over the gdb remote serial protocol, and when we set
the watchpoint/breakpoint, there is no (standard) way for the remote
stub to communicate to lldb which hardware index was used. We have an
lldb-extension packet to query the total number of watchpoint registers.
When a watchpoint is hit, there is an lldb extension to the stop reply
packet (documented in lldb-gdb-remote.txt) to describe the watchpoint
including its actual hardware index,
<addr within wp range> <wp hw index> <actual accessed address>
(the third field is specifically needed for MIPS). At this point, if the
stub reported these three fields (the stub is only required to provide
the first), we can know the actual hardware index for this watchpoint.
Breakpoints are worse; there's never any way for us to be notified about
which hardware index was used. Breakpoints got this as a side effect of
inherting from StoppointSite with Watchpoints.
We expose the watchpoint hardware index through "watchpoint list -v" and
through SBWatchpoint::GetHardwareIndex.
With my large watchpoint support, there is no *single* hardware index
that may be used for a watchpoint, it may need multiple resources. Also
I don't see what a user is supposed to do with this information, or an
IDE. Knowing the total number of watchpoint registers on the target, and
knowing how many Watchpoint Resources are currently in use, is helpful.
Knowing how many Watchpoint Resources
a single user-specified watchpoint needed to be implemented is useful.
But knowing which registers were used is an implementation detail and
not available until we hit the watchpoint when using gdb remote serial
protocol.
So given all that, I'm removing watchpoint hardware index numbers. I'm
changing the SB API to always return -1.
This removes AArch64 specific code from the GDB* classes.
To do this I've added 2 new methods to Architecture:
* RegisterWriteCausesReconfigure to check if what you are about to do
will trash the register info.
* ReconfigureRegisterInfo to do the reconfiguring. This tells you if
anything changed so that we only invalidate registers when needed.
So that ProcessGDBRemote can call ReconfigureRegisterInfo in
SetThreadStopInfo,
I've added forwarding calls to GDBRemoteRegisterContext and the base
class
RegisterContext.
(which removes a slightly sketchy static cast as well)
RegisterContext defaults to doing nothing for both the methods
so anything other than GDBRemoteRegisterContext will do nothing.
This completes the conversion of LocateSymbolFile into a SymbolLocator
plugin. The only remaining function is DownloadSymbolFileAsync which
doesn't really fit into the plugin model, and therefore moves into the
SymbolLocator class, while still relying on the plugins to do the
underlying work.
This removes explicit invalidation of vg and svg that was done in
`GDBRemoteRegisterContext::AArch64Reconfigure`. This was in fact
covering up a bug elsehwere.
Register information says that a write to vg also invalidates svg (it
does not unless you are in streaming mode, but we decided to keep it
simple and say it always does).
This invalidation was not being applied until *after* AArch64Reconfigure
was called. This meant that without those manual invalidates this
happened:
* vg is written
* svg is not invalidated
* Reconfigure uses the written vg value
* Reconfigure uses the *old* svg value
I have moved the AArch64Reconfigure call to after we've processed the
invalidations caused by the register write, so we no longer need the
manual invalidates in AArch64Reconfigure.
In addition I have changed the order in which expedited registers as
parsed. These registers come with a stop notification and include,
amongst others, vg and svg.
So now we:
* Parse them and update register values (including vg and svg)
* AArch64Reconfigure, which uses those values, and invalidates every
register, because offsets may have changed.
* Parse the expedited registers again, knowing that none of the values
will have changed due to the scaling.
This means we use the expedited registers during the reconfigure, but
the invalidate does not mean we throw all of them away.
The cost is we parse them twice client side, but this is cheap compared
to a network packet, and is limited to AArch64 targets only.
On a system with SVE and SME, these are the packets sent for a step:
```
(lldb) b-remote.async> < 803> read packet:
$T05thread:p1f80.1f80;name:main.o;threads:1f80;thread-pcs:000000000040056c<...>a1:0800000000000000;d9:0400000000000000;reason:trace;#fc
intern-state < 21> send packet: $xfffffffff200,200#5e
intern-state < 516> read packet:
$e4f2ffffffff000000<...>#71
intern-state < 15> send packet: $Z0,400568,4#4d
intern-state < 6> read packet: $OK#9a
dbg.evt-handler < 16> send packet: $jThreadsInfo#c1
dbg.evt-handler < 224> read packet:
$[{"name":"main.o","reason":"trace","registers":{"161":"0800000000000000",<...>}],"signal":5,"tid":8064}]]#73
```
You can see there are no extra register reads which means we're using
the expedited registers.
For a write to vg:
```
(lldb) register write vg 4
lldb < 37> send packet:
$Pa1=0400000000000000;thread:1f80;#4a
lldb < 6> read packet: $OK#9a
lldb < 20> send packet: $pa1;thread:1f80;#29
lldb < 20> read packet: $0400000000000000#04
lldb < 20> send packet: $pd9;thread:1f80;#34
lldb < 20> read packet: $0400000000000000#04
```
There is the initial P write, and lldb correctly assumes that SVG is
invalidated by this also so we read back the new vg and svg values
afterwards.
On an SVE/SME the register context configuration may change after the
inferior process has executed. This was handled via
https://reviews.llvm.org/D159504 but it is reconfiguring and clearing
the register context after we've parsed any expedited reigster values
from the stop reply packet. That results in lldb having to read each
register value one at a time while at that stop location, which will be
a performance problem on non-local debug setups.
The configuration & clearing needs to happen first. Also, update the
names of the local variables for a little clarity.
[lldb] Part 2 of 2 - Refactor `CommandObject::DoExecute(...)` to return
`void` instead of ~~`bool`~~
Justifications:
- The code doesn't ultimately apply the `true`/`false` return values.
- The methods already pass around a `CommandReturnObject`, typically
with a `result` parameter.
- Each command return object already contains:
- A more precise status
- The error code(s) that apply to that status
Part 1 refactors the `CommandObject::Execute(...)` method.
- See
[https://github.com/llvm/llvm-project/pull/69989](https://github.com/llvm/llvm-project/pull/69989)
rdar://117378957
This reverts commit a7b78cac9a77e3ef6bbbd8ab1a559891dc693401.
With updates to the tests.
TestWatchTaggedAddress.py: Updated the expected watchpoint types,
though I'm not sure there should be a differnt default for the two
ways of setting them, that needs to be confirmed.
TestStepOverWatchpoint.py: Skipped this everywhere because I think
what used to happen is you couldn't put 2 watchpoints on the same
address (after alignment). I guess that this is now allowed because
modify watchpoints aren't accounted for, but likely should be.
Needs investigating.
This reverts commit 933ad5c897ee366759a54869b35b2d7285a92137.
This caused 1 test failure and an unexpected pass on AArch64 Linux:
https://lab.llvm.org/buildbot/#/builders/96/builds/45765
Wasn't reported because the bot was already red at the time.
Watchpoints in lldb can be either 'read', 'write', or 'read/write'. This
is exposing the actual behavior of hardware watchpoints. gdb has a
different behavior: a "write" type watchpoint only stops when the
watched memory region *changes*.
A user is using a watchpoint for one of three reasons:
1. Want to find what is changing/corrupting this memory.
2. Want to find what is writing to this memory.
3. Want to find what is reading from this memory.
I believe (1) is the most common use case for watchpoints, and it
currently can't be done in lldb -- the user needs to continue every time
the same value is written to the watched-memory manually. I think gdb's
behavior is the correct one. There are some use cases where a developer
wants to find every function that writes/reads to/from a memory region,
regardless of value, I want to still allow that functionality.
This is also a bit of groundwork for my large watchpoint support
proposal
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
where I will be adding support for AArch64 MASK watchpoints which watch
power-of-2 memory regions. A user might ask to watch 24 bytes, and a
MASK watchpoint stub can do this with a 32-byte MASK watchpoint if it is
properly aligned. And we need to ignore writes to the final 8 bytes of
that watched region, and not show those hits to the user.
This patch adds a new 'modify' watchpoint type and it is the default.
Re-landing this patch after addressing testsuite failures found in CI on
Linux, Intel machines, and windows.
rdar://108234227
The size of ZA depends on the streaming vector length regardless
of the active mode. So in addition to vg (which reports the active
mode) we must send the client svg.
Otherwise the mechanics are the same as for non-streaming SVE.
Use the svg value to update the defined size of ZA, accounting
for the fact that ZA is not a single vector but a suqare matrix.
So if svg is 8, a single streaming vector would be 8*8 = 64 bytes.
ZA is that squared, so 64*64 = 4096 bytes.
Testing is included in a later patch.
Reviewed By: omjavaid
Differential Revision: https://reviews.llvm.org/D159504
Watchpoints in lldb can be either 'read', 'write', or 'read/write'. This
is exposing the actual behavior of hardware watchpoints. gdb has a
different behavior: a "write" type watchpoint only stops when the
watched memory region *changes*.
A user is using a watchpoint for one of three reasons:
1. Want to find what is changing/corrupting this memory.
2. Want to find what is writing to this memory.
3. Want to find what is reading from this memory.
I believe (1) is the most common use case for watchpoints, and it
currently can't be done in lldb -- the user needs to continue every time
the same value is written to the watched-memory manually. I think gdb's
behavior is the correct one. There are some use cases where a developer
wants to find every function that writes/reads to/from a memory region,
regardless of value, I want to still allow that functionality.
This is also a bit of groundwork for my large watchpoint support
proposal
https://discourse.llvm.org/t/rfc-large-watchpoint-support-in-lldb/72116
where I will be adding support for AArch64 MASK watchpoints which watch
power-of-2 memory regions. A user might ask to watch 24 bytes, and a
MASK watchpoint stub can do this with a 32-byte MASK watchpoint if it is
properly aligned. And we need to ignore writes to the final 8 bytes of
that watched region, and not show those hits to the user.
This patch adds a new 'modify' watchpoint type and it is the default.
rdar://108234227
The remaining use of ConstString in StructuredData is the Dictionary
class. Internally it's backed by a `std::map<ConstString, ObjectSP>`.
I propose that we replace it with a `llvm::StringMap<ObjectSP>`.
Many StructuredData::Dictionary objects are ephemeral and only exist for
a short amount of time. Many of these Dictionaries are only produced
once and are never used again. That leaves us with a lot of string data
in the ConstString StringPool that is sitting there never to be used
again. Even if the same string is used many times for keys of different
Dictionary objects, that is something we can measure and adjust for
instead of assuming that every key may be reused at some point in the
future.
Quick comparisons of key data is likely not a concern with Dictionary,
but the use of `llvm::StringMap` means that lookups should be fast with
its hashing strategy.
Switching to a llvm::StringMap meant that the iteration order may be
different. To account for this when serializing/dumping the dictionary,
I added some code to sort the output by key before emitting anything.
Differential Revision: https://reviews.llvm.org/D159313
ConstString can be implicitly converted into a llvm::StringRef. This is
very useful in many places, but it also hides places where we are
creating a ConstString only to use it as a StringRef for the entire
lifespan of the ConstString object.
I locally removed the implicit conversion and found some of the places we
were doing this.
Differential Revision: https://reviews.llvm.org/D159237
I'm rewriting this for 2 reasons:
1) Although a 1024 char buffer is probably enough space, the error
string may grow beyond that and we could lose some information. Using
StringStream (as we do in the rest of ProcessGDBRemote) seems like a
sensible solution.
2) I am planning on changing `UnixSignals::GetSignalAsCString`,
rewriting things with `llvm::formatv` will make that process easier.
Differential Revision: https://reviews.llvm.org/D158017
Add support for the `low_mem_addressing_bits` and
`high_mem_addressing_bits` keys in the stop reply packet,
in addition to the existing `addressing_bits`. Same
behavior as in the qHostInfo packet.
Clean up AddressableBits so we don't need to check if
any values have been set in the object before using it
to potentially update the Process address masks.
Differential Revision: https://reviews.llvm.org/D158041
On AArch64 systems, we may have different page table setups for
low memory and high memory, and therefore a different number of
bits used for addressing depending on which half of memory the
address is in.
This patch extends the qHostInfo and LC_NOTE "addrable bits" so
that it can specify the number of addressing bits in high memory
and in low memory separately. It builds on the patch I added in
https://reviews.llvm.org/D151292 where Process tracks the separate
address masks, and there is a user setting to set them manually.
Differential Revision: https://reviews.llvm.org/D157667
rdar://113225907
StreamFile subclasses Stream (from lldbUtility) and is backed by a File
(from lldbHost). It does not depend on anything from lldbCore or any of its
sibling libraries, so I think it makes sense for this to live in
lldbHost instead.
Differential Revision: https://reviews.llvm.org/D157460
DynamicLoader::LoadBinaryWithUUIDAndAddress can create a Module based
on the binary image in memory, which in some cases contains symbol
names and can be genuinely useful. If we don't have a filename, it
creates a name in the form `memory-image-0x...` with the header address.
In practice, this is most useful with Darwin userland corefiles
where the binary was stored in the corefile in whole, and we can't
find a binary with the matching UUID. Using the binary out of
the corefile memory in this case works well.
But in other cases, akin to firmware debugging, we merely end up
with an oddly named binary image and no symbols.
Add a flag to control whether we will create these memory images
and add them to the Target or not; only set it to true when working
with a userland Mach-O image with the "all image infos" LC_NOTE for
a userland corefile.
Differential Revision: https://reviews.llvm.org/D157167
I need to call this to figure out why the assert in
StopInfoMachException::CreateStopReasonWithMachException is triggering, but
it isn't appropriate to directly access the GDBRemoteCommunication there. And
dumping whatever history the process plugin has collected during the run isn't
gdb-remote specific...
Differential Revision: https://reviews.llvm.org/D154992
Fix incorrect uses of LLDB_LOG_ERROR. The macro doesn't automatically
inject the error in the log message: it merely passes the error as the
first argument to formatv and therefore must be referenced with {0}.
Thanks to Nicholas Allegra for collecting a list of places where the
macro was misused.
rdar://111581655
Differential revision: https://reviews.llvm.org/D154530
In ProcessMachCore::LoadBinariesViaMetadata(), if we did
load some binaries via metadata in the core file, don't
then search for a userland dyld in the corefile / kernel
and throw away that binary list. Also fix a little bug
with correctly recognizing corefiles using a `main bin spec`
LC_NOTE that explicitly declare that this is a userland
corefile.
LocateSymbolFileMacOSX.cpp's Symbols::DownloadObjectAndSymbolFile
clarify the comments on how the force_lookup and how the
dbgshell_command local both have the same effect.
In PlatformDarwinKernel::LoadPlatformBinaryAndSetup, don't
log a message unless we actually found a kernel fileset.
Reorganize ObjectFileMachO::LoadCoreFileImages so that it delegates
binary searching to DynamicLoader::LoadBinaryWithUUIDAndAddress and
doesn't duplicate those searches. For searches that fail, we would
perform them multiple times in both methods. When we have the
mach-o segment vmaddrs for a binary, don't let LoadBinaryWithUUIDAndAddress
load the binary first at its mach-o header address in the Target;
we'll load the segments at the correct addresses individually later
in this method.
DynamicLoaderDarwin::ImageInfo::PutToLog fix a LLDB_LOG logging
formatter.
In DynamicLoader::LoadBinaryWithUUIDAndAddress, instead of using
Target::GetOrCreateModule as a way to find a binary already registered
in lldb's global module cache (and implicitly add it to the Target
image list), use ModuleList::GetSharedModule() which only searches
the global module cache, don't add it to the Target. We may not
want to add an unstripped binary to the Target.
Add a call to Symbols::DownloadObjectAndSymbolFile() even if
"force_symbol_search" isn't set -- this will turn into a
DebugSymbols call / Spotlight search on a macOS system, which
we want.
Only set the Module's LoadAddress if the caller asked us to do that.
Differential Revision: https://reviews.llvm.org/D150928
rdar://109186357
This patch refactors the `StructuredData::Integer` class to make it
templated, makes it private and adds 2 public specialization for both
`int64_t` & `uint64_t` with a public type aliases, respectively
`SignedInteger` & `UnsignedInteger`.
It adds new getter for signed and unsigned interger values to the
`StructuredData::Object` base class and changes the implementation of
`StructuredData::Array::GetItemAtIndexAsInteger` and
`StructuredData::Dictionary::GetValueForKeyAsInteger` to support signed
and unsigned integers.
This patch also adds 2 new `Get{Signed,Unsigned}IntegerValue` to the
`SBStructuredData` class and marks `GetIntegerValue` as deprecated.
Finally, this patch audits all the caller of `StructuredData::Integer`
or `StructuredData::GetIntegerValue` to use the proper type as well the
various tests that uses `SBStructuredData.GetIntegerValue`.
rdar://105575764
Differential Revision: https://reviews.llvm.org/D150485
Signed-off-by: Med Ismail Bennani <ismail@bennani.ma>
Use templates to simplify {Get,Set}PropertyAtIndex. It has always
bothered me how cumbersome those calls are when adding new properties.
After this patch, SetPropertyAtIndex infers the type from its arguments
and GetPropertyAtIndex required a single template argument for the
return value. As an added benefit, this enables us to remove a bunch of
wrappers from UserSettingsController and OptionValueProperties.
Differential revision: https://reviews.llvm.org/D149774
If a remote stub provides the addressing_bits kv pair in
the stop reply packet, update the Process address masks with
that value as it possibly changes during the process runtime.
This is an unusual situation, most likely a JTAG remote stub
and some very early startup code that is setting up the page
tables. Nearly all debug sessions will have a single address
mask that cannot change during the lifetime of a Process.
Differential Revision: https://reviews.llvm.org/D149803
rdar://61900565
The majority of call sites are nullptr as the execution context.
Refactor OptionValueProperties to make the argument optional and
simplify all the callers.
Similar to fdbe7c7faa54, refactor OptionValueProperties to return a
std::optional instead of taking a fail value. This allows the caller to
handle situations where there's no value, instead of being unable to
distinguish between the absence of a value and the value happening the
match the fail value. When a fail value is required,
std::optional::value_or() provides the same functionality.
In the particular case I was looking at I autogenerated a 128 bit set
of flags that is only 64 bit. This doesn't crash lldb but it was certainly
not expected.
I suspect that we would have crashed if the top 64 bits weren't
marked as unused (or at least invoked some very undefined behaviour).
When this happens, log the details and ignore the flags. Like this:
```
Size of register flags TTBR0_EL1_flags (16 bytes) for register TTBR0_EL1 does not match the register size (8 bytes). Ignoring this set of flags.
```
Turns out a few of the tests relied on this bug so I have updated
them and added a specific test for this case.
Reviewed By: jasonmolenda
Differential Revision: https://reviews.llvm.org/D148715
These probably do not need to be in the ConstString StringPool as they
don't really need any of the advantages that ConstStrings offer.
Lifetime for these things is always static and we never need to perform
comparisons for setting descriptions.
Differential Revision: https://reviews.llvm.org/D148679
Instead of taking a `const std::string &` we can take an
`llvm::StringRef`. The motivation for this change is that many of the
callers of `ParseJSON` end up creating a temporary `std::string` from an existing
`StringRef` or `const char *` in order to satisfy the API. There's no
reason we need to do this.
Differential Revision: https://reviews.llvm.org/D148579
