Bug 10267 - Windows 8 printing via local printer drivers does not work
Summary: Windows 8 printing via local printer drivers does not work
Status: RESOLVED FIXED
Alias: None
Product: Samba 4.1 and newer
Classification: Unclassified
Component: Printing (show other bugs)
Version: 4.1.0
Hardware: All All
: P5 normal (vote)
Target Milestone: ---
Assignee: Karolin Seeger
QA Contact: Samba QA Contact
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-11-14 09:51 UTC by Arvid Requate
Modified: 2013-11-26 19:30 UTC (History)
0 users

See Also:

Attachments
windows8_printer_driver_v4_XPS_PASS.patch (587 bytes, patch)
2013-11-14 09:51 UTC, Arvid Requate 		no flags Details
windows8_printer_driver_v4_XPS_PASS.patch (2.81 KB, patch)
2013-11-18 17:15 UTC, Arvid Requate 		no flags Details
patch for 4.1, straight cherry-pick from master (1.46 KB, patch)
2013-11-21 14:26 UTC, David Disseldorp 		kseeger: review? (gd)
asn: review+
Details
patch for 4.0, straight cherry-pick from master (1.46 KB, patch)
2013-11-21 14:27 UTC, David Disseldorp 		kseeger: review? (gd)
asn: review+
Details
patch for 3.6, straight cherry-pick from master (1.46 KB, patch)
2013-11-21 14:27 UTC, David Disseldorp 		kseeger: review? (gd)
asn: review+
Details
Show Obsolete (2) View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Arvid Requate 2013-11-14 09:51:45 UTC 
Created attachment 9420 [details]
windows8_printer_driver_v4_XPS_PASS.patch

Sppolss silently drops print jobs sent from a Windows 8 machine when generated using a printer which was installed with Windows 8 native local printer drivers. The attached patch fixes this problem.

Analysis: The new v4 driver model used in Windows 8 declares print jobs intended to bypass the XPS processing layer by setting datatype to "XPS_PASS" instead of "RAW" ( see e.g. http://www.microsoft.com/whdc/device/print/XPSDrv_FilterPipe.mspx ).

The attached patch makes srv_spoolss_nt accept this datatype.



This is the log.smbd level 10 output of the issue without the patch:
=============================================================================
+[2013/11/07 07:38:48.432465,  1, pid=20933, effective(0, 5000), real(0, 0)] ../librpc/ndr/ndr.c:333(ndr_print_function_debug)
+       spoolss_StartDocPrinter: struct spoolss_StartDocPrinter
+          in: struct spoolss_StartDocPrinter
+              handle                   : *
+                  handle: struct policy_handle
+                      handle_type              : 0x00000000 (0)
+                      uuid                     : 000000ed-0000-0000-7b52-7735c5510000
+              info_ctr                 : *
+                  info_ctr: struct spoolss_DocumentInfoCtr
+                      level                    : 0x00000001 (1)
+                      info                     : union spoolss_DocumentInfo(case 1)
+                      info1                    : *
+                          info1: struct spoolss_DocumentInfo1
+                              document_name            : *
+                                  document_name            : 'Testseite'
+                              output_file              : NULL
+                              datatype                 : *
+                                  datatype                 : 'XPS_PASS'
+[2013/11/07 07:38:48.433186,  6, pid=20933, effective(0, 5000), real(0, 0), class=rpc_srv] ../source3/rpc_server/rpc_handles.c:337(find_policy_by_hnd_internal)
+  Found policy hnd[2] [0000] 00 00 00 00 ED 00 00 00   00 00 00 00 7B 52 77 35   ........ ....{Rw5
+  [0010] C5 51 00 00                                       .Q.. 
+[2013/11/07 07:38:48.433348,  1, pid=20933, effective(0, 5000), real(0, 0)] ../librpc/ndr/ndr.c:333(ndr_print_function_debug)
+       spoolss_StartDocPrinter: struct spoolss_StartDocPrinter
+          out: struct spoolss_StartDocPrinter
+              job_id                   : *
+                  job_id                   : 0x00000000 (0)
+              result                   : WERR_INVALID_DATATYPE
=============================================================================
Comment 1 David Disseldorp 2013-11-14 10:14:34 UTC 
Thanks for your analysis and patch Arvid!

A couple of minor comments on the patch:
- please add a code comment with your analysis
- split the check across multiple lines

e.g.
                /*
                 * The v4 driver model used in Windows 8 declares print jobs 
                 * intended to bypass the XPS processing layer by setting
                 * datatype to "XPS_PASS" instead of "RAW".
                 */ 
                if ((strcmp(info_1->datatype, "RAW") != 0)
                 && (strcmp(info_1->datatype, "XPS_PASS") != 0)) {

A corresponding smbtorture test would also be ideal.
Comment 2 David Disseldorp 2013-11-18 14:27:40 UTC 
(In reply to comment #1)
> Thanks for your analysis and patch Arvid!
> 
> A couple of minor comments on the patch:
> - please add a code comment with your analysis
> - split the check across multiple lines

Ping, were you able to address these points Arvid? I'd like to get your fix upstream as soon as possible.
Comment 3 Arvid Requate 2013-11-18 17:15:05 UTC 
Created attachment 9433 [details]
windows8_printer_driver_v4_XPS_PASS.patch

I'm not sure about the torture test, I tried to follow the wiki page but didn't have time to test this yet.
Comment 4 David Disseldorp 2013-11-21 14:26:23 UTC 
Created attachment 9458 [details]
patch for 4.1, straight cherry-pick from master
Comment 5 David Disseldorp 2013-11-21 14:27:20 UTC 
Created attachment 9459 [details]
patch for 4.0, straight cherry-pick from master
Comment 6 David Disseldorp 2013-11-21 14:27:44 UTC 
Created attachment 9460 [details]
patch for 3.6, straight cherry-pick from master
Comment 7 David Disseldorp 2013-11-21 14:30:23 UTC 
Arvid's fix is now in master as b2815b4c8c3e436a79fb7f07be285a417fd6e8cb.

@Karolin, please merge the attached patches to the corresponding maintenance branches.
Comment 8 Karolin Seeger 2013-11-22 10:32:00 UTC 
(In reply to comment #7)
> Arvid's fix is now in master as b2815b4c8c3e436a79fb7f07be285a417fd6e8cb.
> 
> @Karolin, please merge the attached patches to the corresponding maintenance
> branches.

Sorry, I need second review first...
Comment 9 Karolin Seeger 2013-11-22 10:52:44 UTC 
Pushed to autobuild-v4-1-test, autobuild-v4-0-test and v3-6-test.
Comment 10 Karolin Seeger 2013-11-26 19:30:28 UTC 
Pushed to v4-1-test and v4-0-test.
Closing out bug report.

Thanks!