Using LLVM 11.0.0 Toolset
Installing and using LLVM 11.0.0 toolset
Eva-Lotte Gebhardt
egebhard@redhat.com
Zuzana Zoubkova
zzoubkov@redhat.com
Olga Tikhomirova
otikhomi@redhat.com
Peter Macko
Kevin Owen
Vladimir Slavik
devtools-docs@redhat.com
Abstract
Making open source more inclusive
Red Hat is committed to replacing problematic language in our code, documentation, and web properties. We are beginning with these four terms: master, slave, blacklist, and whitelist. Because of the enormity of this endeavor, these changes will be implemented gradually over several upcoming releases. For more details, see our CTO Chris Wright’s message.
Chapter 1. LLVM
1.1. About LLVM Toolset
LLVM Toolset is a Red Hat offering for developers on the Red Hat Enterprise Linux platform. It provides the LLVM compiler infrastructure framework, the Clang compiler for the C and C++ languages, the LLDB debugger, and related tools for code analysis.
LLVM Toolset is distributed as a part of Red Hat Developer Tools for Red Hat Enterprise Linux 7. LLVM Toolset is available as a module for Red Hat Enterprise Linux 8.
The following components are available as a part of LLVM Toolset:
Table 1.1. LLVM Components
Name | Version | Description |
---|---|---|
clang |
RHEL 7 — 11.0.1 | An LLVM compiler front end for C and C++. |
lldb |
RHEL 7 — 11.0.1 | A C and C++ debugger using portions of LLVM. |
compiler-rt |
RHEL 7 — 11.0.1 | Runtime libraries for LLVM. |
llvm |
RHEL 7 — 11.0.1 | A collection of modular and reusable compiler and toolchain technologies. |
libomp |
RHEL 7 — 11.0.1 | A library for utilization of Open MP API specification for parallel programming. |
lld |
RHEL 7 — 11.0.1 | An LLVM linker. |
python-lit |
RHEL 7 — 0.11.1 | A software testing tool for LLVM- and Clang-based test suites. |
LLVM Toolset for Red Hat Enterprise Linux 7 also provides CMake as a separate package. On Red Hat Enterprise Linux 8, CMake is available in the system repository. For more information on how to install CMake, see Section 1.4, “Installing LLVM Toolset”.
1.2. Compatibility
LLVM Toolset is available for Red Hat Enterprise Linux 7 and Red Hat Enterprise Linux 8 on the following architectures:
- AMD and Intel 64-bit architectures
- The 64-bit ARM architecture (Only RHEL 8)
- IBM Power Systems, Little Endian
- IBM Power Systems, Big Endian (Only RHEL 7)
- 64-bit IBM Z
1.3. Getting access to LLVM Toolset on Red Hat Enterprise Linux 7
This chapter lists the steps to perform before installing LLVM Toolset on a Red Hat Enterprise Linux 7 system. Complete the following steps to attach a subscription that provides access to the repository for Red Hat Developer Tools, and then enable the Red Hat Developer Tools and Red Hat Software Collections repositories.
Prerequisites
Verify that
wget
is installed on your system. The tool is available from the default Red Hat Enterprise Linux repositories. To install it, run the following command as root:# yum install wget
Procedure
Get the latest subscription data from the server:
# subscription-manager refresh
Use the following command to register the system:
# subscription-manager register
You can also register the system by following the appropriate steps in Registering and Unregistering a System in the Red Hat Subscription Management document.
Display a list of all subscriptions that are available for your system and identify the pool ID for the subscription:
# subscription-manager list --available
This command displays the subscription name, unique identifier, expiration date, and other details related to it. The pool ID is listed on a line beginning with
Pool ID
.Attach the subscription that provides access to the
Red Hat Developer Tools
repository. Use the pool ID you identified in the previous step.# subscription-manager attach --pool=<appropriate pool ID from the subscription>
Verify the list of subscriptions attached to your system:
# sudo subscription-manager list --consumed
Enable the
rhel-7-variant-devtools-rpms
repository:# subscription-manager repos --enable rhel-7-variant-devtools-rpms
Replace
variant
with the Red Hat Enterprise Linux system variant (server
orworkstation
).Consider using Red Hat Enterprise Linux Server to access the widest range of the development tools.
Enable the
rhel-variant-rhscl-7-rpms
repository:# subscription-manager repos --enable rhel-variant-rhscl-7-rpms
Replace
variant
with the Red Hat Enterprise Linux system variant (server
orworkstation
).Add the Red Hat Developer Tools GPG key to your system:
# cd /etc/pki/rpm-gpg # wget -O RPM-GPG-KEY-redhat-devel https://www.redhat.com/security/data/a5787476.txt # rpm --import RPM-GPG-KEY-redhat-devel
Once the subscription is attached to the system and the repositories are enabled, install LLVM Toolset as described in Section 1.4, “Installing LLVM Toolset”.
Additional Resources
- For more information on how to register your system using Red Hat Subscription Management and associate it with subscriptions, see the Red Hat Subscription Management collection of guides.
1.4. Installing LLVM Toolset
LLVM Toolset is distributed as a collection of RPM packages that can be installed, updated, uninstalled, and inspected by using the standard package management tools that are included in Red Hat Enterprise Linux.
Note that a valid subscription that provides access to the Red Hat Developer Tools content set is required in order to install LLVM Toolset on your Red Hat Enterprise Linux 7 system. For detailed instructions on how to associate your Red Hat Enterprise Linux 7 system with an appropriate subscription and get access to LLVM Toolset, see Section 1.3, “Getting access to LLVM Toolset on Red Hat Enterprise Linux 7”.
Before installing LLVM Toolset, install all available Red Hat Enterprise Linux updates.
Install all of the components included in LLVM Toolset for your operating system:
On Red Hat Enterprise Linux 7, install the llvm-toolset 11.0 collection:
# yum install llvm-toolset-11.0
On Red Hat Enterprise Linux 8, install the llvm-toolset module:
# yum module install llvm-toolset
This installs all development and debugging tools, and other dependent packages to the system.
1.4.1. Installing CMake on Red Hat Enterprise Linux
CMake is available as a separate package. To install CMake:
On Red Hat Enterprise Linux 7, install the llvm-toolset-11.0-cmake package:
# yum install llvm-toolset-11.0-cmake llvm-toolset-11.0-cmake-doc
On Red Hat Enterprise Linux 8, install the cmake package:
# yum install cmake cmake-doc
1.4.1.1. CMake installable documentation
The cmake package contains installed documentation. On Red Hat Enterprise Linux 7, find the documentation in opt/rh/llvm-toolset-11.0/root/usr/share/doc/llvm-toolset-11.0-cmake-3.6.2/html/index.html
. On Red Hat Enterprise Linux 8, find the documentation in /usr/share/doc/llvm/html/index.html
.
1.4.2. Installable documentation
The following section describes how to install the LLVM Toolset installable documentation.
- On Red Hat Enterprise Linux 7, install the llvm-doc-11.0 package:
# yum install llvm-toolset-11.0-llvm-doc
The documentation is available in /opt/rh/llvm-toolset-11.0/root/usr/share/doc/llvm-toolset-11.0-llvm-11.0/html/index.html
.
- On Red Hat Enterprise Linux 8, install the llvm-doc package:
# yum install llvm llvm-doc
The documentation is available in /usr/share/doc/llvm/html/index.html
.
The documenation for CMake is not included in the LLVM documentation package. To install the documentation for CMake, see Section 1.4.1.1, “CMake installable documentation”.
1.5. Additional Resources
A detailed description of LLVM Toolset and all its features is beyond the scope of this document. For more information, see the resources listed below.
Online documentation
- LLVM documentation overview — The official LLVM documentation.
Chapter 2. Using clang
Clang is a LLVM compiler front end for C-based languages: C, C++, Objective C/C++, OpenCL, and Cuda.
LLVM Toolset is distributed with clang 11.0.1 on RHEL 7 and clang 11.0.0 on RHEL 8.
You can execute any command using the scl
utility on Red Hat Enterprise Linux 7, causing it to be run with the LLVM binaries available. To use LLVM Toolset on Red Hat Enterprise Linux 7 without a need to use scl enable
with every command, run a shell session with:
$ scl enable llvm-toolset-11.0 'bash'
2.1. Compiling a C source file to a binary file
To compile a C program to a binary file:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -o output_file source_file'
For Red Hat Enterprise Linux 8:
$ clang -o output_file source_file
This creates a binary file named output_file
in the current working directory. If the -o
option is omitted, the compiler creates a binary file named a.out
by default.
Example 2.1. Compiling a C Program with clang
Consider a source file named hello.c
with the following contents:
#include <stdio.h> int main(int argc, char *argv[]) { printf("Hello, World!\n"); return 0; }
Compile this source code on the command line by using the clang
compiler from LLVM Toolset:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -o hello hello.c'
For Red Hat Enterprise Linux 8:
$ clang -o hello hello.c
This creates a new binary file called hello
in the current working directory.
2.2. Compiling a C source file to an object file
When you are working on a project that consists of several source files, it is common to compile an object file for each of the source files first and then link these object files together. This way, when you change a single source file, you can recompile only this file without having to compile the entire project.
To compile a C source file to an object file:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -o object_file -c source_file'
For Red Hat Enterprise Linux 8:
$ clang -o object_file -c source_file
This creates an object file named object_file
. If the -o
option is omitted, the compiler creates a file named after the source file with the .o
file extension.
2.3. Linking C object files to a binary File
To link object files together and create a binary file:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -o output_file object_file ...'
For Red Hat Enterprise Linux 8:
$ clang -o output_file object_file ...
Certain more recent library features are statically linked into applications built with LLVM Toolset to support execution on multiple versions of Red Hat Enterprise Linux. This creates an additional minor security risk as standard Red Hat Enterprise Linux errata do not change this code. If the need arises for developers to rebuild their applications due to this risk, Red Hat will communicate this using a security erratum.
Because of this additional security risk, developers are strongly advised not to statically link their entire application for the same reasons.
2.4. Using the clang Integrated Assembler
To produce an object file from an assembly language program, run the clang
tool as follows:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -o object_file source_file'
For Red Hat Enterprise Linux 8:
$ clang -o object_file source_file
This creates an object file named object_file
in the current working directory.
2.5. Running a C program
When clang
compiles a program, it creates an executable binary file. To run this program on the command line, change to the directory with the executable file and run the program:
$ ./file_name
Example 2.2. Running a C program on the command line
Assuming that you have successfully compiled the hello
binary file as shown in Example 2.1, “Compiling a C Program with clang”, you can run it by typing the following command:
$ ./hello Hello, World!
2.6. Additional Resources
A detailed description of the clang compiler and its features is beyond the scope of this document. For more information, see the resources listed below.
Installed documentation
clang(1) — The manual page for the
clang
compiler provides detailed information on its usage; with few exceptions,clang++
accepts the same command line options asclang
. To display the manual page for the version included in LLVM Toolset:For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'man clang'
For Red Hat Enterprise Linux 8:
$ man clang
Online documentation
-
clang — The clang compiler documentation provides detailed information about use of
clang
.
See Also
- Chapter 1, LLVM — An overview of LLVM and more information on how to install it on your system.
Chapter 3. Using clang++
3.1. Compiling a C++ Source File to a Binary File
To compile a C++ program on the command line, run the clang++
compiler as follows:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang++ -o output_file source_file ...'
For Red Hat Enterprise Linux 8:
$ clang++ -o output_file source_file ...
This creates a binary file named output_file
in the current working directory. If the -o
option is omitted, the clang++
compiler creates a file named a.out
by default.
3.2. Compiling a C++ source file to an object file
When you are working on a project that consists of several source files, it is common to compile an object file for each of the source files first and then link these object files together. This way, when you change a single source file, you can recompile only this file without having to compile the entire project.
To compile an object file on the command line:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang++ -o object_file -c source_file'
For Red Hat Enterprise Linux 8:
$ clang++ -o object_file -c source_file
This creates an object file named object_file
. If the -o
option is omitted, the clang++
compiler creates a file named after the source file with the .o
file extension.
3.3. Linking C++ object files to a binary file
To link object files together and create a binary file:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang++ -o output_file object_file ...'
For Red Hat Enterprise Linux 8:
$ clang++ -o output_file object_file ...
Certain more recent library features are statically linked into applications built with LLVM Toolset to support execution on multiple versions of Red Hat Enterprise Linux. This creates an additional minor security risk as standard Red Hat Enterprise Linux errata do not change this code. If the need arises for developers to rebuild their applications due to this risk, Red Hat will communicate this using a security erratum.
Because of this accitional security risk, developers are strongly advised not to statically link their entire application for the same reasons.
Example 3.1. Compiling a C++ Program on the Command Line
Consider a source file named hello.cpp
with the following contents:
#include <iostream> using namespace std; int main(int argc, char *argv[]) { cout << "Hello, World!" << endl; return 0; }
Compile this source code on the command line by using the clang++
compiler from LLVM:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang++ -o hello hello.cpp'
For Red Hat Enterprise Linux 8:
$ clang++ -o hello hello.cpp
This creates a new binary file called hello
in the current working directory.
3.4. Running a C++ program
When clang++
compiles a program, it creates an executable binary file. Change to the directory with the executable file and run this program:
./file_name
Example 3.2. Running a C++ program on the command line
Assuming that you have successfully compiled the hello
binary file as shown in Example 3.1, “Compiling a C++ Program on the Command Line”, you can run it by typing the following at a shell prompt:
$ ./hello Hello, World!
3.5. Additional Resources
A detailed description of the clang compiler and its features is beyond the scope of this document. For more information, see the resources listed below.
Installed documentation
clang(1) — The manual page for the
clang
compiler provides detailed information on its usage; with few exceptions,clang++
accepts the same command line options asclang
. To display the manual page for the version included in LLVM Toolset:For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'man clang'
For Red Hat Enterprise Linux 8:
$ man clang
Online documentation
-
clang — The clang compiler documentation provides detailed information about use of
clang
.
See Also
- Chapter 1, LLVM — An overview of LLVM and more information on how to install it on your system.
Chapter 4. lldb
lldb is a command line tool you can use to debug programs written in C and C++. It allows you to inspect memory within the code being debugged, control the execution state of the code, detect the execution of particular sections of code, and much more.
LLVM Toolset is distributed with lldb 11.0.1 on RHEL 7 lldb 11.0.0 on RHEL 8.
You can execute any command using the scl
utility on Red Hat Enterprise Linux 7, causing it to be run with the LLVM binaries available. To use LLVM Toolset on Red Hat Enterprise Linux 7 without a need to use scl enable
with every command, run a shell session with:
$ scl enable llvm-toolset-11.0 'bash'
4.1. Preparing a program for debugging
To compile a C or C++ program with debugging information that lldb can read, make sure the compiler you use is instructed to create debug information.
- For instructions on suitably configuring clang, see the section Controlling Debug Information in Clang Compiler User’s Manual.
- For instructions on suitably configuring GCC, see Red Hat Developer Toolset User Guide, Section 8.2. Preparing a Program for Debugging.
4.2. Running lldb
To run lldb on a program you want to debug:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'lldb program_file_name'
For Red Hat Enterprise Linux 8:
$ lldb program_file_name
This command starts lldb
in an interactive mode and displays the default prompt, (lldb)
.
To quit the debugging session and return to the shell prompt, run the following command at any time:
(lldb) quit
Example 4.1. Running the lldb Utility on the fibonacci Binary File
Consider a C source file named fibonacci.c
with the following content:
#include <stdio.h> #include <limits.h> int main (int argc, char *argv[]) { unsigned long int a = 0; unsigned long int b = 1; unsigned long int sum; while (b < LONG_MAX) { printf("%ld ", b); sum = a + b; a = b; b = sum; } return 0; }
Enable the debug information and compile the fibonacci.c
source file with the following command:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'clang -g -o fibonacci fibonacci.c'
For Red Hat Enterprise Linux 8:
$ clang -g -o fibonacci fibonacci.c
Refer to Section 4.1, “Preparing a program for debugging” for information about controlling debug information using GCC or clang.
Start debugging the program with lldb
:
For Red Hat Enterprise Linux 7:
$ scl enable llvm-toolset-11.0 'lldb fibonacci' (lldb) target create "fibonacci" Current executable set to 'fibonacci' (x86_64). (lldb)
For Red Hat Enterprise Linux 8:
$ lldb fibonacci (lldb) target create "fibonacci" Current executable set to 'fibonacci' (x86_64). (lldb)
The output indicates that the program fibonacci
is ready for debugging.
4.3. Listing the source code
To view the source code of the program you are debugging:
(lldb) list
As a result, the first ten lines of the source code are displayed.
To display the code from a particular line:
(lldb) list source_file_name:line_number
Additionally, lldb displays source code listing automatically in the following situations:
-
Before you start the execution of the program you are debugging,
lldb
displays the first ten lines of the source code. -
Each time the execution of the program is stopped,
lldb
displays the lines that surround the line on which the execution stops.
4.4. Using breakpoints
Setting a New Breakpoint
To set a new breakpoint at a certain line:
(lldb) breakpoint source_file_name:line_number
To set a breakpoint on a certain function:
(lldb) breakpoint source_file_name:function_name
Example 4.2. Setting a new breakpoint
This example assumes that you have successfully compiled the fibonacci.c
file as shown in Example 4.1, “Running the lldb Utility on the fibonacci Binary File”.
Set two breakpoints at line 10 by running the following commands:
(lldb) b 10 Breakpoint 1: where = fibonacci`main + 33 at fibonacci.c:10, address = 0x000000000040054e
(lldb) breakpoint set -f fibonacci.c --line 10 Breakpoint 2: where = fibonacci`main + 33 at fibonacci.c:10, address = 0x000000000040054e
In lldb, the command b
is not an alias to breakpoint
. You can use both commands to set breakpoints, but b
uses a subset of the syntax supported by the gdb break
command, and breakpoint
uses lldb syntax for setting breakpoints.
Listing Breakpoints
To display a list of currently set breakpoints:
(lldb) breakpoint list
Example 4.3. Listing Breakpoints
This example assumes that you have successfully followed the instructions in Example 4.2, “Setting a new breakpoint”.
Display the list of currently set breakpoints:
(lldb) breakpoint list Current breakpoints: 1: file = 'fibonacci.c', line = 10, exact_match = 0, locations = 1 1.1: where = fibonacci`main + 33 at fibonacci.c:10, address = fibonacci[0x000000000040054e], unresolved, hit count = 0 2: file = 'fibonacci.c', line = 10, exact_match = 0, locations = 1 2.1: where = fibonacci`main + 33 at fibonacci.c:10, address = fibonacci[0x000000000040054e], unresolved, hit count = 0
Deleting Existing Breakpoints
To delete a breakpoint that is set at a certain line:
(lldb) breakpoint clear -f source_file_name -l line_number
Example 4.4. Deleting an Existing Breakpoint
This example assumes that you have successfully compiled the fibonacci.c
file.
Set a new breakpoint at line 7:
(lldb) b 7 Breakpoint 3: where = fibonacci`main + 31 at fibonacci.c:9, address = 0x000000000040054c
Remove this breakpoint:
(lldb) breakpoint clear -l 7 -f fibonacci.c 1 breakpoints cleared: 3: file = 'fibonacci.c', line = 7, exact_match = 0, locations = 1
4.5. Starting Execution
To start an execution of the program you are debugging:
(lldb) run
If the program accepts command-line arguments, you can provide them as arguments to the run
command:
(lldb) run argument …
The execution stops when the first breakpoint is reached, when an error occurs, or when the program terminates.
Example 4.5. Executing the fibonacci Binary File in lldb
This example assumes that you have successfully followed the instructions in Example 4.2, “Setting a new breakpoint”.
Execute the fibonacci
binary file in lldb:
(lldb) run Process 21054 launched: 'fibonacci' (x86_64) Process 21054 stopped * thread #1, name = 'fibonacci', stop reason = breakpoint 1.1 frame #0: fibonacci`main(argc=1, argv=0x00007fffffffdeb8) at fibonacci.c:10 7 unsigned long int sum; 8 9 while (b < LONG_MAX) { -> 10 printf("%ld ", b); 11 sum = a + b; 12 a = b; 13 b = sum;
Execution of the program stops at the breakpoint set in Example 4.2, “Setting a new breakpoint”.
4.6. Displaying Current Program Data
The lldb
tool enables you to display data relevant to the program state, including:
- Variables of any complexity
- Any valid expressions
- Function call return values
The common usage is to display the value of a variable. To display the current value of a certain variable:
(lldb) print variable_name
Example 4.6. Displaying the current values of variables
This example assumes that you have successfully followed the instructions in Example 4.5, “Executing the fibonacci Binary File in lldb”. Execution of the fibonacci
binary stopped after reaching the breakpoint at line 10.
Display the current values of variables a
and b
:
(lldb) print a $0 = 0 (lldb) print b $1 = 1
4.7. Continuing execution after a breakpoint
To resume the execution of the program you are debugging after it reached a breakpoint:
(lldb) continue
The execution stops again when it reaches another breakpoint.
To skip a certain number of breakpoints, typically when you are debugging a loop, run the continue
command in the following form:
(lldb) continue -i number_of_breakpoints_to_skip
If the breakpoint is set on a loop, in order to skip the whole loop, you will have to set the number_of_breakpoints_to_skip to match the loop iteration count.
The lldb
tool enables you to execute a single line of code from the current line pointer with step
:
(lldb) step
To execute a certain number of lines:
(lldb) step -c number
Example 4.7. Continuing the execution of the fibonacci binary file after a breakpoint
This example assumes that you have successfully followed the instructions in Example 4.5, “Executing the fibonacci Binary File in lldb”. The execution of the fibonacci
binary stopped after reaching the breakpoint at line 10.
Resume the execution:
(lldb) continue Process 21580 resuming Process 21580 stopped * thread #1, name = 'fibonacci', stop reason = breakpoint 1.1 frame #0: fibonacci`main(argc=1, argv=0x00007fffffffdeb8) at fibonacci.c:10 7 unsigned long int sum; 8 9 while (b < LONG_MAX) { -> 10 printf("%ld ", b); 11 sum = a + b; 12 a = b; 13 b = sum;
The execution stops the next time it reaches a breakpoint. In this case, it is the same breakpoint. Execute the next three lines of code:
(lldb) step -c 3 Process 21580 stopped * thread #1, name = 'fibonacci', stop reason = step in frame #0: fibonacci`main(argc=1, argv=0x00007fffffffdeb8) at fibonacci.c:11 8 9 while (b < LONG_MAX) { 10 printf("%ld ", b); -> 11 sum = a + b; 12 a = b; 13 b = sum; 14 }
Verify the current value of the sum
variable:
(lldb) print sum $2 = 2
4.8. Additional Resources
A detailed description of the lldb debugger and all its features is beyond the scope of this document. For more information, see the resources listed below.
Online documentation
- lldb Tutorial — The official lldb tutorial.
- gdb to lldb command map — A list of GDB commands and their lldb equivalents.
See also
- Chapter 1, LLVM — An overview of LLVM and more information on how to install it.
Chapter 5. Container images with LLVM Toolset
You can build your own LLVM Toolset container images on top of Red Hat Universal Base Images (UBI) containers. For more information on using UBIs as the foundation for container images of LLVM, see Section 5.1, “Using UBI repositories”.
- For more information on Red Hat UBI images, see Building, Running, and Managing Containers - Chapter 3, Working with Container Images.
5.1. Using UBI repositories
LLVM Toolset packages are part of the Red Hat Universal Base Images (UBIs) repositories. You need a set up UBI to access UBI repositories and build a container image of LLVM.
5.2. Example: Building a container image of LLVM on RHEL 8 using a Dockerfile
Prerequisites
- A set up UBI.
To set up a UBI, complete the following steps:
- Visit the Red Hat Container Catalog and search for UBI 8.
- Select a UBI.
- Click Get this image and follow the instructions.
For more information on using UBI images, see Building, Running, and Managing Containers - Chapter 3, Working with Container Images.
Procedure
To build a container of LLVM Toolset on Red Hat Enterprise Linux 8, your Dockerfile should contain the following lines:
FROM registry.access.redhat.com/ubi8/ubi:latest
RUN yum install -y llvm-toolset
To keep the container size as small as possible, install individual packages. For example, to install only the Clang compiler in a container image, use:
RUN yum install clang
5.3. Example: Building a container image of LLVM on RHEL 7 using a Dockerfile
Prerequisites
- A set up UBI.
To set up a UBI, complete the following steps:
- Visit the Red Hat Container Catalog and search for UBI 7.
- Select a UBI.
- Click Get this image and follow the instructions.
For more information on using UBI images, see Building, Running, and Managing Containers - Chapter 3, Working with Container Images.
Procedure
To build a container containing LLVM Toolset on Red Hat Enterprise Linux 7, your Dockerfile should contain the following lines:
FROM registry.access.redhat.com/ubi7/ubi:ilatest RUN yum install -y llvm-toolset-${version}
To keep the container size as small as possible, install individual packages. For example, to install only the Clang compiler in a container image, use:
RUN yum install llvm-toolset-${_version_}-clang
5.4. Additional resources
- For more information on Red Hat UBI images, see Building, Running, and Managing Containers - Chapter 3, Working with Container Images.
- For more information on Red Hat UBI repositories, see Universal Base Images (UBI): Images, repositories, packages, and source code.
Chapter 6. Changes in LLVM 11.0.0 Toolset
LLVM Toolset has been updated from version 10.0.1 to 11.0.0 on Red Hat Enterprise Linux 8 and to 11.0.1 on Red Hat Enterprise Linux 7. Notable changes include:
-
Support for the
-fstack-clash-protection
command-line option has been added to the AMD and Intel 64-bit architectures, IBM Power Systems, Little Endian, and IBM Z. This new compiler flag protects from stack-clash attacks by automatically checking each stack page. -
The new compiler flag
ffp-exception-behavior={ignore,maytrap,strict}
enables the specification of floating-point exception behavior. The default setting isignore
. -
The new compiler flag
ffp-model={precise,strict,fast}
allows the simplification of single purpose floating-point options. The default setting isprecise
. -
The new compiler flag
-fno-common
is now enabled by default. With this enhancement, code written in C using tentative variable definitions in multiple translation units now triggers multiple-definition linker errors. To disable this setting, use the-fcommon
flag. - Container images for LLVM Toolset have been deprecated and LLVM Toolset has been added to the Universal Base Images (UBI) repositories.
For more information, see the upstream LLVM 11.0.0 Release Notes.