[Bug 1001959] New: ARM call-graph profiling

[Bug 1001959] New: ARM call-graph profiling

[bugzilla-daemon]

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001959

            Bug ID: 1001959
           Summary: ARM call-graph profiling
           Product: eCos
           Version: CVS
            Target: Other (please specify)
  Architecture/Host ARM
                OS:
            Status: NEW
          Severity: enhancement
          Priority: low
         Component: Patches and contributions
          Assignee: unassigned@bugs.ecos.sourceware.org
          Reporter: john@dallaway.org.uk
                CC: ecos-patches@ecos.sourceware.org

Created attachment 2484
  --> http://bugs.ecos.sourceware.org/attachment.cgi?id=2484&action=edit
arm-mcount-140314.patch

ARM mcount functions to support call-graph profiling.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug 1001959] ARM call-graph profiling

[bugzilla-daemon]

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001959

--- Comment #1 from Ilija Kocho [Илија Кочо] <ilijak@siva.com.mk> ---
Comment on attachment 2484
  --> http://bugs.ecos.sourceware.org/attachment.cgi?id=2484
arm-mcount-140314.patch

John,

Just for my education (should have asked in bug 1001954 a week ago): When using
hardware FP, do we need to also put FPU registers?

Ilija

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug 1001959] ARM call-graph profiling

[bugzilla-daemon]

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001959

--- Comment #2 from John Dallaway <john@dallaway.org.uk> ---
(In reply to comment #1)

> Just for my education (should have asked in bug 1001954 a week ago): When
> using hardware FP, do we need to also put FPU registers?

Ilija, by inspection, we know that the mcount functions do not use any FP
registers. These functions call __profile_mcount() which is self-contained and
does not perform any floating point calculations. So there is no possibility
that function arguments passed in FP registers will be modified within the call
to mcount() or __gnu_mcount_nc() and therefore no need for the mcount functions
to save/restore FP registers.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug 1001959] ARM call-graph profiling

[bugzilla-daemon]

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001959

Ilija Kocho [Илија Кочо] <ilijak@siva.com.mk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ilijak@siva.com.mk

--- Comment #3 from Ilija Kocho [Илија Кочо] <ilijak@siva.com.mk> ---
(In reply to comment #2)
> (In reply to comment #1)
> 
> > Just for my education (should have asked in bug 1001954 a week ago): When
> > using hardware FP, do we need to also put FPU registers?
> 
> Ilija, by inspection, we know that the mcount functions do not use any FP
> registers. These functions call __profile_mcount() which is self-contained
> and does not perform any floating point calculations. So there is no
> possibility that function arguments passed in FP registers will be modified
> within the call to mcount() or __gnu_mcount_nc() and therefore no need for
> the mcount functions to save/restore FP registers.

Many thanks John.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

[Bug 1001959] ARM call-graph profiling

[bugzilla-daemon]

Please do not reply to this email, use the link below.

http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001959

John Dallaway <john@dallaway.org.uk> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
                 CC|                            |ecos-bugs@ecos.sourceware.o
                   |                            |rg
          Component|Patches and contributions   |HAL
         Resolution|---                         |CURRENTRELEASE
           Assignee|unassigned@bugs.ecos.source |john@dallaway.org.uk
                   |ware.org                    |

--- Comment #4 from John Dallaway <john@dallaway.org.uk> ---
Patch checked-in.

-- 
You are receiving this mail because:
You are on the CC list for the bug.