Bug 959013 - x86_64 vDSO has grown 4KB -> 8KB
Summary: x86_64 vDSO has grown 4KB -> 8KB
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 18
Hardware: x86_64
OS: Linux
low
low
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On: 312011
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-05-03 01:30 UTC by Jan Kratochvil
Modified: 2013-07-24 20:22 UTC (History)
7 users (show)

Fixed In Version:
Clone Of: 312011
Environment:
Last Closed: 2013-07-24 20:22:27 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Jan Kratochvil 2013-05-03 01:30:08 UTC
+++ This bug was initially created as a clone of Bug #312011 +++

Description of problem:
on recent x86_64 kernels the kernel VDSO has grown from 4KB to 8KB.
This affects reading debug info of vDSO in various GDB variants.

Version-Release number of selected component (if applicable):
kernel-3.8.8-202.fc18.x86_64

How reproducible:
Always.

Steps to Reproduce:
cat /proc/self/maps | grep vdso

Actual results:
7fff9db67000-7fff9db69000 r-xp 00000000 00:00 0 [vdso]

Expected results:
7fff9db67000-7fff9db68000 r-xp 00000000 00:00 0 [vdso]

Comment 2 Josh Boyer 2013-07-02 16:23:59 UTC
As far as I can tell, the text section itself of vdso.so is almost 4K.  When combined with the rest of the sections, the overall size of the vdso is 5560 bytes (on a 3.9.6-301 f19 kernel).  That is then page aligned, so we get 8K total.

I might be missing something, but unlike the original bug I don't see the vDSO being fully contained within a single page?  So it seems comment #3 from that bug might still be applicable.

If I've missed something, please let me know.

Comment 3 Jan Kratochvil 2013-07-24 20:22:27 UTC
At least kernel-3.10.0-2.el7.x86_64 has:

Program Headers:
  Type           Offset             VirtAddr           PhysAddr
                 FileSiz            MemSiz              Flags  Align
  LOAD           0x0000000000000000 0xffffffffff700000 0xffffffffff700000
                 0x000000000000102d 0x000000000000102d  R E    1000

so it should not be a problem.

It was not a case for kernel-3.9.9-201.fc18.x86_64 but let's forget about F-18.

Problem was when the runtime (segment) informatio was 4KB but some sections (debug) information were 8KB, they were cut then.

Confirming that it seems vDSO really contains so much code nowadays.


Note You need to log in before you can comment on or make changes to this bug.