Closed Bug 146722 Opened 22 years ago Closed 21 years ago

Enable HP IPF (ia64) XPTcall code

Categories

(Core :: XPConnect, defect)

HP
HP-UX
defect
Not set
normal

Tracking

()

VERIFIED FIXED

People

(Reporter: jdunn, Assigned: jdunn)

References

Details

(Keywords: fixedOEM)

Attachments

(2 files, 1 obsolete file)

We received code from HP for getting the xptcall code working
on HP-UXB.11.20 IPF (ia64 machine) and part of the code was
modifying mozilla/xpcom/reflect/xptcall/public/genstubs.pl

I don't fully understand it, but if we are going to get mozilla
working on IPF (Itanium Processor Family) then we are going to 
either have to use it or come up with something else.

The code is (NOTE: this is not the final patch, I realize that
I need to ifdef for HP & ia64, but I wanted to get feedback)

Index: xpcom/reflect/xptcall/public/genstubs.pl
===================================================================
RCS file: /cvsroot/mozilla/xpcom/reflect/xptcall/public/genstubs.pl,v
retrieving revision 1.5
diff -u -r1.5 genstubs.pl
--- xpcom/reflect/xptcall/public/genstubs.pl    19 May 2001 00:03:00 -0000      1.5
+++ xpcom/reflect/xptcall/public/genstubs.pl    23 May 2002 11:58:51 -0000
@@ -38,9 +38,17 @@
 print OUTFILE "*  1 is AddRef\n";
 print OUTFILE "*  2 is Release\n";
 print OUTFILE "*/\n";
+print OUTFILE "#ifndef __ia64\n";
 for($i = 0; $i < $entry_count; $i++) {
     print OUTFILE "XPTC_EXPORT NS_IMETHOD Stub",$i+3,"();\n";
 }
+print OUTFILE "#else\n";
+for($i = 0; $i < $entry_count; $i++) {
+    print OUTFILE "XPTC_EXPORT NS_IMETHOD Stub",$i+3,"(PRUint64,\n";
+    print OUTFILE " PRUint64,PRUint64,PRUint64,PRUint64,PRUint64,PRUint64);\n";
+
+}
+print OUTFILE "#endif\n";
 
 print OUTFILE "\n/* declarations of sentinel stubs */\n";
 
----------
Apparently this matches up with xpcom/xptcall/src/md/unix/xptcstubs_ipf32.cpp
because at the bottom, the definition of STUBS_ENTRY has a comment:

/* Variable a0-a7 were put there so we can have access to 
 *  the 8 input registers on Stubxyz entry
 */
updating...
Assignee: dbradley → jdunn
Status: NEW → ASSIGNED
Blocks: 136163
note: for the trunk we need to work on this...
Whiteboard: branchOEM+
Changes checked into OEM branch.
Whiteboard: branchOEM+ → branchOEM+, fixedOEM
Whiteboard: branchOEM+, fixedOEM → fixedOEM
Keywords: fixedOEM
Whiteboard: fixedOEM
Attached patch trunk patch, better #ifdefs (obsolete) — Splinter Review
Patch updated for trunk check in.   Added an extra #ifdef to more specifically
target only HPUX ia64 machines.
Extraneous lines removed
Attachment #102976 - Attachment is obsolete: true
Comment on attachment 102978 [details] [diff] [review]
Trunk patch revised

r=dbradley

Can't vouch for the details, but looks ok, and will only affect the ia64 HP
systems.
Attachment #102978 - Flags: review+
Attachment #102978 - Flags: superreview?(shaver)
Comment on attachment 102978 [details] [diff] [review]
Trunk patch revised

This is fine, yeah.  Drivers: let's take this for 1.4 -- simple, safe port fix.
Attachment #102978 - Flags: superreview?(shaver)
Attachment #102978 - Flags: superreview+
Attachment #102978 - Flags: approval1.4?
Comment on attachment 102978 [details] [diff] [review]
Trunk patch revised

since this only affects HPUX, and the HPUX folks want it, and they want to ship
1.4, a=sspitzer
Attachment #102978 - Flags: approval1.4? → approval1.4+
fix checked in
Status: ASSIGNED → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
Checkin verified -
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: