Fix ptsname() for big-endian architectures#20
Merged
estesp merged 2 commits intocontainerd:masterfrom Feb 20, 2018
pmorjan:fix-ptsname2
Merged
Fix ptsname() for big-endian architectures#20estesp merged 2 commits intocontainerd:masterfrom pmorjan:fix-ptsname2
estesp merged 2 commits intocontainerd:masterfrom
pmorjan:fix-ptsname2
Conversation
On big-endian architectures unix.IoctlGetInt() leads to a wrong result because a 32 bit value is stored into a 64 bit buffer. When dereferencing the result is left shifted by 32. Without this patch ptsname() returns a wrong path from the second pty onwards. To protect syscalls against re-arranging by the GC the conversion from unsafe.Pointer to uintptr must occur in the Syscall expression itself. See the documentation of the unsafe package for details. Signed-off-by: Peter Morjan <peter.morjan@de.ibm.com>
Member
|
The Solaris build break with Go 1.9.4 is a known issue** (fixed in 1.10 which is why travis "tip" succeeds). Should we: a) fix Golang 1.9.x to 1.9.3 in travis, b) move travis to 1.10, or c) remove Solaris from travis unless we have known use? |
Member
|
Maybe pin to 1.9.3 so we don't lose solaris builds. I know we have some solaris users of this package so it would be nice to keep the build. |
Member
|
By pin, i mean 1.9.3 and 1.10/tip |
Contributor
Author
|
You can allow Travis to fail for a particular Go version: |
Member
|
Works for me. Maybe allow failures only for 1.9.4? |
Contributor
Author
|
but then you can't use the 1.9.x glob any more |
Member
|
Ok, that works for me. Thanks for debugging |
Solaris builds fail when using Go 1.9.4. Signed-off-by: Peter Morjan <peter.morjan@de.ibm.com>
Member
|
LGTM |
Closed
3 tasks
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 14, 2018
This PR fixes the issue kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166. The console package has been add in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 14, 2018
This PR fixes the issue kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166. The console package has been add in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 14, 2018
This PR fixes the issue kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 15, 2018
The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 15, 2018
Fixes: kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 15, 2018
Fixes: kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
alicefr
pushed a commit
to alicefr/agent
that referenced
this pull request
Nov 15, 2018
Fixes: kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
jshachm
pushed a commit
to jshachm/agent
that referenced
this pull request
Nov 22, 2018
Fixes: kata-containers#410 The console package in vendor contains a big endian bug. The issue has already been solved in runc. See PRs: - opencontainers/runc#1727 - containerd/console#20 The console version has been updated from commit 84eeaae905 to commit 2748ece166 (commit that fixed the big-endian issue) The console package has been added in the Gopkg.toml because the console package is not directly used in the code and the dep command doesn't automatically update the console package. Signed-off-by: Alice Frosi <afrosi@de.ibm.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
On big-endian architectures unix.IoctlGetInt() leads to a wrong result
because a 32 bit value is stored into a 64 bit buffer. When dereferencing
the result is left shifted by 32. Without this patch ptsname() returns
a wrong path from the second pty onwards.
To protect syscalls against re-arranging by the GC the conversion from
unsafe.Pointer to uintptr must occur in the Syscall expression itself.
See the documentation of the unsafe package for details.
Signed-off-by: Peter Morjan peter.morjan@de.ibm.com