2022-09-06 19:14:07 -04:00
|
|
|
// Copyright 2014 The Crashpad Authors
|
2014-12-17 14:35:18 -08:00
|
|
|
//
|
|
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
// you may not use this file except in compliance with the License.
|
|
|
|
// You may obtain a copy of the License at
|
|
|
|
//
|
|
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
//
|
|
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
// See the License for the specific language governing permissions and
|
|
|
|
// limitations under the License.
|
|
|
|
|
|
|
|
#include "util/file/file_io.h"
|
|
|
|
|
2020-06-18 15:35:28 +02:00
|
|
|
#include "base/check_op.h"
|
2014-12-17 14:35:18 -08:00
|
|
|
#include "base/logging.h"
|
|
|
|
#include "base/numerics/safe_conversions.h"
|
|
|
|
|
|
|
|
namespace crashpad {
|
|
|
|
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
namespace {
|
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
class FileIOReadExactly final : public internal::ReadExactlyInternal {
|
|
|
|
public:
|
|
|
|
explicit FileIOReadExactly(FileHandle file)
|
|
|
|
: ReadExactlyInternal(), file_(file) {}
|
2021-09-20 12:55:12 -07:00
|
|
|
|
|
|
|
FileIOReadExactly(const FileIOReadExactly&) = delete;
|
|
|
|
FileIOReadExactly& operator=(const FileIOReadExactly&) = delete;
|
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
~FileIOReadExactly() {}
|
|
|
|
|
|
|
|
private:
|
|
|
|
// ReadExactlyInternal:
|
|
|
|
FileOperationResult Read(void* buffer, size_t size, bool can_log) override {
|
|
|
|
FileOperationResult rv = ReadFile(file_, buffer, size);
|
|
|
|
if (rv < 0) {
|
|
|
|
PLOG_IF(ERROR, can_log) << internal::kNativeReadFunctionName;
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return rv;
|
|
|
|
}
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
FileHandle file_;
|
|
|
|
};
|
|
|
|
|
|
|
|
class FileIOWriteAll final : public internal::WriteAllInternal {
|
|
|
|
public:
|
|
|
|
explicit FileIOWriteAll(FileHandle file) : WriteAllInternal(), file_(file) {}
|
2021-09-20 12:55:12 -07:00
|
|
|
|
|
|
|
FileIOWriteAll(const FileIOWriteAll&) = delete;
|
|
|
|
FileIOWriteAll& operator=(const FileIOWriteAll&) = delete;
|
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
~FileIOWriteAll() {}
|
|
|
|
|
|
|
|
private:
|
|
|
|
// WriteAllInternal:
|
|
|
|
FileOperationResult Write(const void* buffer, size_t size) override {
|
|
|
|
return internal::NativeWriteFile(file_, buffer, size);
|
|
|
|
}
|
|
|
|
|
|
|
|
FileHandle file_;
|
|
|
|
};
|
|
|
|
|
|
|
|
} // namespace
|
|
|
|
|
|
|
|
namespace internal {
|
|
|
|
|
|
|
|
bool ReadExactlyInternal::ReadExactly(void* buffer, size_t size, bool can_log) {
|
2020-12-09 22:21:38 +00:00
|
|
|
uintptr_t buffer_int = reinterpret_cast<uintptr_t>(buffer);
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
size_t total_bytes = 0;
|
|
|
|
size_t remaining = size;
|
|
|
|
while (remaining > 0) {
|
2020-12-09 22:21:38 +00:00
|
|
|
FileOperationResult bytes_read =
|
|
|
|
Read(reinterpret_cast<char*>(buffer_int), remaining, can_log);
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
if (bytes_read < 0) {
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
DCHECK_LE(static_cast<size_t>(bytes_read), remaining);
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
if (bytes_read == 0) {
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2020-12-09 22:21:38 +00:00
|
|
|
buffer_int += bytes_read;
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
remaining -= bytes_read;
|
|
|
|
total_bytes += bytes_read;
|
2014-12-17 14:35:18 -08:00
|
|
|
}
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
if (total_bytes != size) {
|
|
|
|
LOG_IF(ERROR, can_log) << "ReadExactly: expected " << size << ", observed "
|
|
|
|
<< total_bytes;
|
2014-12-17 14:35:18 -08:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
bool WriteAllInternal::WriteAll(const void* buffer, size_t size) {
|
2020-12-09 22:21:38 +00:00
|
|
|
uintptr_t buffer_int = reinterpret_cast<uintptr_t>(buffer);
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
|
|
|
|
while (size > 0) {
|
2020-12-09 22:21:38 +00:00
|
|
|
FileOperationResult bytes_written =
|
|
|
|
Write(reinterpret_cast<const char*>(buffer_int), size);
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
if (bytes_written < 0) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
DCHECK_NE(bytes_written, 0);
|
|
|
|
|
2020-12-09 22:21:38 +00:00
|
|
|
buffer_int += bytes_written;
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
size -= bytes_written;
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
} // namespace internal
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
|
|
|
|
bool ReadFileExactly(FileHandle file, void* buffer, size_t size) {
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
FileIOReadExactly read_exactly(file);
|
|
|
|
return read_exactly.ReadExactly(buffer, size, false);
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
}
|
|
|
|
|
|
|
|
bool LoggingReadFileExactly(FileHandle file, void* buffer, size_t size) {
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
FileIOReadExactly read_exactly(file);
|
|
|
|
return read_exactly.ReadExactly(buffer, size, true);
|
|
|
|
}
|
|
|
|
|
|
|
|
bool WriteFile(FileHandle file, const void* buffer, size_t size) {
|
|
|
|
FileIOWriteAll write_all(file);
|
|
|
|
return write_all.WriteAll(buffer, size);
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
}
|
|
|
|
|
2014-12-17 14:35:18 -08:00
|
|
|
bool LoggingWriteFile(FileHandle file, const void* buffer, size_t size) {
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
if (!WriteFile(file, buffer, size)) {
|
|
|
|
PLOG(ERROR) << internal::kNativeWriteFunctionName;
|
2014-12-17 14:35:18 -08:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
Make file_io reads more rational and predictable
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
2017-03-16 13:36:38 -04:00
|
|
|
void CheckedReadFileExactly(FileHandle file, void* buffer, size_t size) {
|
|
|
|
CHECK(LoggingReadFileExactly(file, buffer, size));
|
2014-12-17 14:35:18 -08:00
|
|
|
}
|
|
|
|
|
|
|
|
void CheckedWriteFile(FileHandle file, const void* buffer, size_t size) {
|
|
|
|
CHECK(LoggingWriteFile(file, buffer, size));
|
|
|
|
}
|
|
|
|
|
|
|
|
void CheckedReadFileAtEOF(FileHandle file) {
|
|
|
|
char c;
|
2015-10-22 16:14:18 -07:00
|
|
|
FileOperationResult rv = ReadFile(file, &c, 1);
|
2014-12-17 14:35:18 -08:00
|
|
|
if (rv < 0) {
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
PCHECK(rv == 0) << internal::kNativeReadFunctionName;
|
2014-12-17 14:35:18 -08:00
|
|
|
} else {
|
util/file: Handle oversized reads and writes gracefully
file_io and the FileReader family had a few loose ends regarding big
reads and writes. It’s not likely that we’ve experienced these
conditions yet, but they’d be likely to appear in a potential future
involving full memory dumps. This specifies the behavior with large
reads and writes, consolidates some logic, and improves some interfaces.
ReadFile() should always return without retrying after a short read, and
in fact does return after short reads since 00b64427523b. It is
straightforward to limit the maximum read size based on a parameter
limitation of the underlying operation, or a limitation of the type used
for FileOperationResult.
In contrast, WriteFile() should always retry after a short write,
including a write shortened because of a parameter limitation of the
underlying operation, or a limitation of the type used for
FileOperationResult. This allows its return value to be simplified to a
“bool”.
The platform-specific WriteFile() code has been moved to
internal::NativeWriteFile(), and the platform-independent loop that
retries following a short write has been refactored into
internal::WriteAllInternal so that it can be used by a new test.
The platform-agnostic ReadFileExactlyInternal() implementation has been
refactored into internal::ReadExactlyInternal so that it can be used by
a new test and by FileReaderInterface::ReadExactly(), which had a nearly
identical implementation.
Test: crashpad_util_test FileIO.ReadExactly_*:FileIO.WriteAll_*:FileReader.ReadExactly_*
Change-Id: I487450322ab049c6f2acd4061ea814037cc9a864
Reviewed-on: https://chromium-review.googlesource.com/456824
Reviewed-by: Scott Graham <scottmg@chromium.org>
2017-03-21 15:08:05 -04:00
|
|
|
CHECK_EQ(rv, 0) << internal::kNativeReadFunctionName;
|
2014-12-17 14:35:18 -08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-04-27 17:00:02 -07:00
|
|
|
bool LoggingReadToEOF(FileHandle file, std::string* contents) {
|
2017-05-01 20:40:01 -07:00
|
|
|
char buffer[4096];
|
|
|
|
FileOperationResult rv;
|
|
|
|
std::string local_contents;
|
2018-04-27 17:00:02 -07:00
|
|
|
while ((rv = ReadFile(file, buffer, sizeof(buffer))) > 0) {
|
2017-05-02 13:52:55 -04:00
|
|
|
DCHECK_LE(static_cast<size_t>(rv), sizeof(buffer));
|
2017-05-01 20:40:01 -07:00
|
|
|
local_contents.append(buffer, rv);
|
|
|
|
}
|
|
|
|
if (rv < 0) {
|
|
|
|
PLOG(ERROR) << internal::kNativeReadFunctionName;
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
contents->swap(local_contents);
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2018-04-27 17:00:02 -07:00
|
|
|
bool LoggingReadEntireFile(const base::FilePath& path, std::string* contents) {
|
|
|
|
ScopedFileHandle handle(LoggingOpenFileForRead(path));
|
|
|
|
if (!handle.is_valid()) {
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
return LoggingReadToEOF(handle.get(), contents);
|
|
|
|
}
|
|
|
|
|
2014-12-17 14:35:18 -08:00
|
|
|
void CheckedCloseFile(FileHandle file) {
|
|
|
|
CHECK(LoggingCloseFile(file));
|
|
|
|
}
|
|
|
|
|
|
|
|
} // namespace crashpad
|