Michael137 wrote:
> Why do we need to touch source files to add a source file attribution from
> debug info to a declaration? We don't require the actual presence of source
> files in order to make source file attributions anywhere else in the debug
> info handling. Jim
A `clang::SourceLocati
https://github.com/SuibianP updated
https://github.com/llvm/llvm-project/pull/121269
>From 9f89e9c59d444057abe517e3c912f6af00c69a2b Mon Sep 17 00:00:00 2001
From: Jialun Hu
Date: Mon, 24 Feb 2025 22:10:17 +0800
Subject: [PATCH] [lldb-dap] Implement runInTerminal for Windows
Currently, the name
https://github.com/labath approved this pull request.
This looks really nice. I also think the Input/OutputStreams could go away but
that can be a separate patch.
*However*, this part
> properly detect shutdown requests when the Socket is closed
really scares me. I must have missed this dur
Author: Pavel Labath
Date: 2025-02-26T11:22:11+01:00
New Revision: 13245cea11050f875891389ce36115c78aaedd4a
URL:
https://github.com/llvm/llvm-project/commit/13245cea11050f875891389ce36115c78aaedd4a
DIFF:
https://github.com/llvm/llvm-project/commit/13245cea11050f875891389ce36115c78aaedd4a.diff
https://github.com/labath closed
https://github.com/llvm/llvm-project/pull/128505
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
https://github.com/labath updated
https://github.com/llvm/llvm-project/pull/128719
>From 9116f3110146965d18eab7e50465855954c87f26 Mon Sep 17 00:00:00 2001
From: Pavel Labath
Date: Tue, 25 Feb 2025 13:30:12 +0100
Subject: [PATCH 1/2] [lldb] Assorted improvements to the Pipe class
The main motiv
@@ -93,15 +93,13 @@ Status SharedSocket::CompleteSending(lldb::pid_t child_pid)
{
"WSADuplicateSocket() failed, error: %d", last_error);
}
- size_t num_bytes;
- Status error =
- m_socket_pipe.WriteWithTimeout(&protocol_info, sizeof(protocol_info),
-
@@ -55,8 +55,6 @@ TEST_F(PipeTest, OpenAsReader) {
}
#endif
-// This test is flaky on Windows on Arm.
labath wrote:
Yeah it was 2fae58e9c7becc376454005da69acb3fa993350e. I'm pretty sure this will
fix it, as it's the exact same problem that was fixed for the
@@ -1154,17 +1156,25 @@ Status GDBRemoteCommunication::StartDebugserverProcess(
if (socket_pipe.CanWrite())
socket_pipe.CloseWriteFileDescriptor();
if (socket_pipe.CanRead()) {
-// The port number may be up to "65535\0".
-char port_cstr[6] =
@@ -55,8 +55,6 @@ TEST_F(PipeTest, OpenAsReader) {
}
#endif
-// This test is flaky on Windows on Arm.
-#ifndef _WIN32
TEST_F(PipeTest, WriteWithTimeout) {
labath wrote:
Good idea. I should have done that originally.
https://github.com/llvm/llvm-project/pull
@@ -268,29 +267,24 @@ PipeWindows::GetReadNativeHandle() { return m_read; }
HANDLE
PipeWindows::GetWriteNativeHandle() { return m_write; }
-Status PipeWindows::ReadWithTimeout(void *buf, size_t size,
-const std::chrono::microseconds &duratio
@@ -1154,17 +1156,25 @@ Status GDBRemoteCommunication::StartDebugserverProcess(
if (socket_pipe.CanWrite())
socket_pipe.CloseWriteFileDescriptor();
if (socket_pipe.CanRead()) {
-// The port number may be up to "65535\0".
-char port_cstr[6] =
@@ -300,70 +304,51 @@ void PipePosix::CloseWriteFileDescriptorUnlocked() {
}
}
-Status PipePosix::ReadWithTimeout(void *buf, size_t size,
- const std::chrono::microseconds &timeout,
- size_t &bytes_read) {
+ll
@@ -24,41 +30,95 @@ using namespace llvm;
namespace lldb_dap {
-FifoFile::FifoFile(StringRef path) : m_path(path) {}
+std::error_code EC;
+FifoFile::FifoFile(StringRef path)
+: m_path(path), m_file(fopen(path.data(), "r+")) {
+ if (m_file == nullptr) {
+EC = std::e
@@ -24,41 +30,95 @@ using namespace llvm;
namespace lldb_dap {
-FifoFile::FifoFile(StringRef path) : m_path(path) {}
+std::error_code EC;
+FifoFile::FifoFile(StringRef path)
+: m_path(path), m_file(fopen(path.data(), "r+")) {
+ if (m_file == nullptr) {
+EC = std::e
@@ -24,41 +30,95 @@ using namespace llvm;
namespace lldb_dap {
-FifoFile::FifoFile(StringRef path) : m_path(path) {}
-
+FifoFile::FifoFile(StringRef path)
+: m_path(path), m_file(fopen(path.data(), "r+")) {
+ std::error_code EC;
+ if (m_file == nullptr) {
+EC = std:
@@ -24,41 +30,95 @@ using namespace llvm;
namespace lldb_dap {
-FifoFile::FifoFile(StringRef path) : m_path(path) {}
-
+FifoFile::FifoFile(StringRef path)
+: m_path(path), m_file(fopen(path.data(), "r+")) {
+ std::error_code EC;
+ if (m_file == nullptr) {
+EC = std:
https://github.com/SuibianP edited
https://github.com/llvm/llvm-project/pull/121269
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 60cc3af0d93ecb8bfc9d6bebc6cbc395df3bb4b6
f62993f2aa1ee48685c775529a98ce69aedef2c1 --e
@@ -1154,17 +1156,25 @@ Status GDBRemoteCommunication::StartDebugserverProcess(
if (socket_pipe.CanWrite())
socket_pipe.CloseWriteFileDescriptor();
if (socket_pipe.CanRead()) {
-// The port number may be up to "65535\0".
-char port_cstr[6] =
DavidSpickett wrote:
All my questions answered, I leave it to @ashgti as the relative expert to be
the approver.
https://github.com/llvm/llvm-project/pull/128719
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/
https://github.com/labath updated
https://github.com/llvm/llvm-project/pull/128719
>From 9116f3110146965d18eab7e50465855954c87f26 Mon Sep 17 00:00:00 2001
From: Pavel Labath
Date: Tue, 25 Feb 2025 13:30:12 +0100
Subject: [PATCH 1/3] [lldb] Assorted improvements to the Pipe class
The main motiv
101 - 122 of 122 matches
Mail list logo
