DOSBox-X branch

Reply 860 of 2412, by ykhwong

Posted on 2015-01-27, 07:06

ykhwong Offline

Rank Oldbie

Rank: Oldbie
Posts: 627
Joined: 2004-03-17, 15:35

SedrynTyros, could you remove my email address, please?
Malicious spam bots circulating on the web can collect the address, making me hard to keep junk mail out of my inbox.
Few years ago, I replaced the plain text with an image on my website to prevent the situation.

I do not have my own source control repository but have been maintaining patches that can be directly applied to the source from the official svn branch. They still need to be cleaned up for a better management.
I am considering to upload a source archive to my website with previous tarballs kept every time a new build is released.
This allows to generate all the changes between the two points.

Reply 861 of 2412, by Dominus

Posted on 2015-01-27, 08:05

Dominus Offline

Rank DOSBox Moderator

Rank: DOSBox Moderator
Posts: 9165
Joined: 2002-10-03, 09:54
Location: Ludwigsburg

I edited Sedryn's post.

A quick diff between your releases would probably help as well. I still need to find out why an app bundle of your code works, while it crashes with dosbox-x code...

Windows 3.1x guide for DOSBox
60 seconds guide to DOSBox
DOSBox SVN snapshot for macOS (10.4-11.x ppc/intel 32/64bit) notarized for gatekeeper

Reply 862 of 2412, by ykhwong

Posted on 2015-01-27, 09:15

ykhwong Offline

Rank Oldbie

Rank: Oldbie
Posts: 627
Joined: 2004-03-17, 15:35

Dominus wrote:
I edited Sedryn's post.

Thank you.

Reply 863 of 2412, by ykhwong

Posted on 2015-01-27, 09:15

ykhwong Offline

Rank Oldbie

Rank: Oldbie
Posts: 627
Joined: 2004-03-17, 15:35

A regression regarding a recent dosbox-x change is found: case sensitivity issue in the environment variables set.
Lower-case variables should overwrite the upper-case variables.

For example, typing "set path=c:\" makes the SET command show both PATH and path variables.
The following patch should fix the problem.

1--- ./src/misc/programs.cpp	2015-01-24 14:29:27.871925500 +0900
2+++ ./src/misc/programs.cpp	2015-01-24 15:30:45.392255100 +0900
3@@ -296,13 +297,16 @@
4 		return false;
5 	}
6 
7+	std::string bigentry(entry);
8+	for (std::string::iterator it = bigentry.begin(); it != bigentry.end(); ++it) *it = toupper(*it);
9+
10 	env_scan = env_base;
11 	while (env_scan < env_fence) {
12 		/* "NAME" + "=" + "VALUE" + "\0" */
13 		/* end of the block is a NULL string meaning a \0 follows the last string's \0 */
14 		if (mem_readb(env_scan) == 0) break; /* normal end of block */
15 
16-		if (EnvPhys_StrCmp(env_scan,env_fence,entry) == 0) {
17+		if (EnvPhys_StrCmp(env_scan,env_fence,bigentry.c_str()) == 0) {
18 			EnvPhys_StrCpyToCPPString(result,env_scan,env_fence);
19 			return true;
20 		}
21@@ -387,7 +391,10 @@
22 		return false;
23 	}
24 
25-	el = strlen(entry);
26+	std::string bigentry(entry);
27+	for (std::string::iterator it = bigentry.begin(); it != bigentry.end(); ++it) *it = toupper(*it);
28+
29+	el = strlen(bigentry.c_str());
30 	if (*new_string != 0) nsl = strlen(new_string);
31 	needs = nsl+1+el+1+1; /* entry + '=' + new_string + '\0' + '\0' */
32 
33@@ -396,7 +403,7 @@
34 	while (env_scan < env_fence) {
35 		if (mem_readb(env_scan) == 0) break;
36 
37-		if (EnvPhys_StrCmp(env_scan,env_fence,entry) == 0) {
38+		if (EnvPhys_StrCmp(env_scan,env_fence,bigentry.c_str()) == 0) {
39 			/* found it. remove by shifting the rest of the environment block over */
40 			int zeroes=0;
41 			PhysPt s,d;
42@@ -404,7 +411,7 @@
43 			/* before we remove it: is there room for the new value? */
44 			if (nsl != 0) {
45 				if ((env_scan+needs) > env_fence) {
46-					LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",entry,new_string);
47+					LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",bigentry.c_str(),new_string);
48 					return false;
49 				}
50 			}
51@@ -433,12 +440,12 @@
52 	/* add the string to the end of the block */
53 	if (*new_string != 0) {
54 		if ((env_scan+needs) > env_fence) {
55-			LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",entry,new_string);
56+			LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",bigentry.c_str(),new_string);
57 			return false;
58 		}
59 
60 		assert(env_scan < env_fence);

…Show last 5 lines

61-		for (const char *s=entry;*s != 0;) mem_writeb(env_scan++,*s++);
62+		for (const char *s=bigentry.c_str();*s != 0;) mem_writeb(env_scan++,*s++);
63 		mem_writeb(env_scan++,'=');
64 
65 		assert(env_scan < env_fence);

Reply 864 of 2412, by Alegend45

Posted on 2015-01-27, 16:13

Alegend45 Offline

Rank Newbie

Rank: Newbie
Posts: 77
Joined: 2012-06-23, 18:18

Hey, would you be open to GitHub pull requests?

Reply 865 of 2412, by Stiletto

Posted on 2015-01-27, 16:32

Stiletto Offline

Rank l33t++

Rank: l33t++
Posts: 5133
Joined: 2002-07-01, 21:57

Sidenote: the number of times ykhwong's name has been misspelled on VOGONS by posters is appalling, I'm going to see if I can fix some of them 😉

"I see a little silhouette-o of a man, Scaramouche, Scaramouche, will you
do the Fandango!" - Queen

Stiletto

Reply 866 of 2412, by SedrynTyros

Posted on 2015-01-27, 16:58

SedrynTyros Offline

Rank Member

Rank: Member
Posts: 119
Joined: 2012-07-24, 21:44

ykhwong wrote:
SedrynTyros, could you remove my email address, please? Malicious spam bots circulating on the web can collect the address, maki […]
Show full quote
SedrynTyros, could you remove my email address, please?
Malicious spam bots circulating on the web can collect the address, making me hard to keep junk mail out of my inbox.
Few years ago, I replaced the plain text with an image on my website to prevent the situation.

I do not have my own source control repository but have been maintaining patches that can be directly applied to the source from the official svn branch. They still need to be cleaned up for a better management.
I am considering to upload a source archive to my website with previous tarballs kept every time a new build is released.
This allows to generate all the changes between the two points.

Hey, sorry about that. Obviously, I didn't consider that. I won't do that again.

Reply 867 of 2412, by TheGreatCodeholio

Posted on 2015-01-27, 16:59

TheGreatCodeholio Offline

Rank Oldbie

Rank: Oldbie
Posts: 817
Joined: 2011-08-18, 20:15
Location: Seattle, WA

ykhwong wrote:

A regression regarding a recent dosbox-x change is found: case sensitivity issue in the environment variables set. Lower-case va […]
Show full quote

A regression regarding a recent dosbox-x change is found: case sensitivity issue in the environment variables set.
Lower-case variables should overwrite the upper-case variables.

For example, typing "set path=c:\" makes the SET command show both PATH and path variables.
The following patch should fix the problem.

1--- ./src/misc/programs.cpp	2015-01-24 14:29:27.871925500 +0900
2+++ ./src/misc/programs.cpp	2015-01-24 15:30:45.392255100 +0900
3@@ -296,13 +297,16 @@
4 		return false;
5 	}
6 
7+	std::string bigentry(entry);
8+	for (std::string::iterator it = bigentry.begin(); it != bigentry.end(); ++it) *it = toupper(*it);
9+
10 	env_scan = env_base;
11 	while (env_scan < env_fence) {
12 		/* "NAME" + "=" + "VALUE" + "\0" */
13 		/* end of the block is a NULL string meaning a \0 follows the last string's \0 */
14 		if (mem_readb(env_scan) == 0) break; /* normal end of block */
15 
16-		if (EnvPhys_StrCmp(env_scan,env_fence,entry) == 0) {
17+		if (EnvPhys_StrCmp(env_scan,env_fence,bigentry.c_str()) == 0) {
18 			EnvPhys_StrCpyToCPPString(result,env_scan,env_fence);
19 			return true;
20 		}
21@@ -387,7 +391,10 @@
22 		return false;
23 	}
24 
25-	el = strlen(entry);
26+	std::string bigentry(entry);
27+	for (std::string::iterator it = bigentry.begin(); it != bigentry.end(); ++it) *it = toupper(*it);
28+
29+	el = strlen(bigentry.c_str());
30 	if (*new_string != 0) nsl = strlen(new_string);
31 	needs = nsl+1+el+1+1; /* entry + '=' + new_string + '\0' + '\0' */
32 
33@@ -396,7 +403,7 @@
34 	while (env_scan < env_fence) {
35 		if (mem_readb(env_scan) == 0) break;
36 
37-		if (EnvPhys_StrCmp(env_scan,env_fence,entry) == 0) {
38+		if (EnvPhys_StrCmp(env_scan,env_fence,bigentry.c_str()) == 0) {
39 			/* found it. remove by shifting the rest of the environment block over */
40 			int zeroes=0;
41 			PhysPt s,d;
42@@ -404,7 +411,7 @@
43 			/* before we remove it: is there room for the new value? */
44 			if (nsl != 0) {
45 				if ((env_scan+needs) > env_fence) {
46-					LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",entry,new_string);
47+					LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",bigentry.c_str(),new_string);
48 					return false;
49 				}
50 			}
51@@ -433,12 +440,12 @@
52 	/* add the string to the end of the block */
53 	if (*new_string != 0) {
54 		if ((env_scan+needs) > env_fence) {
55-			LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",entry,new_string);
56+			LOG_MSG("Program::SetEnv() error, insufficient room for environment variable %s=%s\n",bigentry.c_str(),new_string);
57 			return false;
58 		}
59 
60 		assert(env_scan < env_fence);

…Show last 5 lines

61-		for (const char *s=entry;*s != 0;) mem_writeb(env_scan++,*s++);
62+		for (const char *s=bigentry.c_str();*s != 0;) mem_writeb(env_scan++,*s++);
63 		mem_writeb(env_scan++,'=');
64 
65 		assert(env_scan < env_fence);

Got it. Done.

I recommend getting some form of source control so that you can track revisions in your code as you apply patches. If you use git, it also makes it easier for other people to incorporate your changes by pulling your branch and then using git's merge functions, and you can git pull from other's branches as well. There's a reason the Linux kernel is developed in that fashion. It's far easier than making diffs between two tarballs and trying to figure out how to merge them together without screwing up the code.

DOSBox-X project: more emulation better accuracy.
DOSLIB and DOSLIB2: Learn how to tinker and hack hardware and software from DOS.

Reply 868 of 2412, by TheGreatCodeholio

Posted on 2015-01-27, 17:00

TheGreatCodeholio Offline

Rank Oldbie

Rank: Oldbie
Posts: 817
Joined: 2011-08-18, 20:15
Location: Seattle, WA

Alegend45 wrote:
Hey, would you be open to GitHub pull requests?

Sure! Just tell me where to pull/merge from and I'll do it!

DOSBox-X project: more emulation better accuracy.
DOSLIB and DOSLIB2: Learn how to tinker and hack hardware and software from DOS.

Reply 869 of 2412, by Alegend45

Posted on 2015-01-27, 17:09

Alegend45 Offline

Rank Newbie

Rank: Newbie
Posts: 77
Joined: 2012-06-23, 18:18

TheGreatCodeholio wrote:
Alegend45 wrote:
Hey, would you be open to GitHub pull requests?

Sure! Just tell me where to pull/merge from and I'll do it!

Alright, thanks.

Reply 870 of 2412, by Firtasik

Posted on 2015-01-27, 22:15

Firtasik Offline

Rank Oldbie

Rank: Oldbie
Posts: 537
Joined: 2013-07-21, 19:07

Maybe a wrong thread, but DOSBox SVN Daum 20150125 gives me less fps in Blood. 😕

11 1 111 11 1 1 1 1 1 11 1 1 111 1 111 1 1 1 1 111

Reply 871 of 2412, by Alegend45

Posted on 2015-01-28, 20:28

Alegend45 Offline

Rank Newbie

Rank: Newbie
Posts: 77
Joined: 2012-06-23, 18:18

Firtasik wrote:
Maybe a wrong thread, but DOSBox SVN Daum 20150125 gives me less fps in Blood. 😕

This is the wrong thread. It doesn't have Daum in the title.

Reply 872 of 2412, by ykhwong

Posted on 2015-01-28, 22:49

ykhwong Offline

Rank Oldbie

Rank: Oldbie
Posts: 627
Joined: 2004-03-17, 15:35

TheGreatCodeholio wrote:
I recommend getting some form of source control so that you can track revisions in your code as you apply patches. If you use git, it also makes it easier for other people to incorporate your changes by pulling your branch and then using git's merge functions, and you can git pull from other's branches as well. There's a reason the Linux kernel is developed in that fashion. It's far easier than making diffs between two tarballs and trying to figure out how to merge them together without screwing up the code.

Thanks for the tip, TheGreatCodeholio.
I will take that into consideration.

Reply 873 of 2412, by ykhwong

Posted on 2015-01-28, 22:57

ykhwong Offline

Rank Oldbie

Rank: Oldbie
Posts: 627
Joined: 2004-03-17, 15:35

Firtasik wrote:
Maybe a wrong thread, but DOSBox SVN Daum 20150125 gives me less fps in Blood. 😕

Not exactly sure why that happens.
However, I assume that that is caused by some vga_draw updates from DOSBox-X which are incorporated into my build.
AFAIK, the linewise (drawing line by line) is forced and the option was completely removed, possibly leading to the performance degradation.

Reply 874 of 2412, by Firtasik

Posted on 2015-02-02, 16:19

Firtasik Offline

Rank Oldbie

Rank: Oldbie
Posts: 537
Joined: 2013-07-21, 19:07

I've found that launching NOLFBLIM makes that fps drop. Anyway, it looks like Blood doesn't flicker anymore, so no NOLFBLIM needed. 😀

11 1 111 11 1 1 1 1 1 11 1 1 111 1 111 1 1 1 1 111

Reply 875 of 2412, by TheGreatCodeholio

Posted on 2015-02-08, 17:57

TheGreatCodeholio Offline

Rank Oldbie

Rank: Oldbie
Posts: 817
Joined: 2011-08-18, 20:15
Location: Seattle, WA

I've set up a web site and wiki for DOSBox-X

Web page: http://dosbox-x.software/
Wiki: https://github.com/joncampbell123/dosbox-x/wiki
Issues tracker: https://github.com/joncampbell123/dosbox-x/issues

DOSBox-X project: more emulation better accuracy.
DOSLIB and DOSLIB2: Learn how to tinker and hack hardware and software from DOS.

Reply 876 of 2412, by Jorpho

Posted on 2015-02-09, 14:49

Jorpho Offline

Rank l33t++

Rank: l33t++
Posts: 8168
Joined: 2003-02-14, 19:50
Location: Canada

Your continued efforts in this endeavor are much appreciated.

Reply 877 of 2412, by Synoptic

Posted on 2015-02-09, 21:10

Synoptic Offline

Rank Member

Rank: Member
Posts: 320
Joined: 2010-01-24, 15:56
Location: Canada

TheGreatCodeholio wrote:
I've set up a web site and wiki for DOSBox-X […]
Show full quote
I've set up a web site and wiki for DOSBox-X

Web page: http://dosbox-x.software/
Wiki: https://github.com/joncampbell123/dosbox-x/wiki
Issues tracker: https://github.com/joncampbell123/dosbox-x/issues

Trying to build today's release : #include "afxres.h" not found in wines.rc

the file is indeed absent from the branch.

Reply 878 of 2412, by leileilol

Posted on 2015-02-12, 17:49

leileilol Offline

Rank l33t++

Rank: l33t++
Posts: 12147
Joined: 2006-12-16, 18:03

That's a standard header and MinGW or MSVC should come with it.

long live PCem

Reply 879 of 2412, by emendelson

Posted on 2015-02-12, 23:03

emendelson Offline

Rank Oldbie

Rank: Oldbie
Posts: 843
Joined: 2010-02-14, 02:00

Synoptic wrote:
Trying to build today's release : #include "afxres.h" not found in wines.rc

the file is indeed absent from the branch.

Replace "afxres.h" with "windows.h" and try again.

Last edited by emendelson on 2015-02-15, 00:11. Edited 1 time in total.

Main menu

Topic actions

Reply 860 of 2412, by ykhwong

Reply 861 of 2412, by Dominus

Reply 862 of 2412, by ykhwong

Reply 863 of 2412, by ykhwong

Reply 864 of 2412, by Alegend45

Reply 865 of 2412, by Stiletto

Reply 866 of 2412, by SedrynTyros

Reply 867 of 2412, by TheGreatCodeholio

Reply 868 of 2412, by TheGreatCodeholio

Reply 869 of 2412, by Alegend45

Reply 870 of 2412, by Firtasik

Reply 871 of 2412, by Alegend45

Reply 872 of 2412, by ykhwong

Reply 873 of 2412, by ykhwong

Reply 874 of 2412, by Firtasik

Reply 875 of 2412, by TheGreatCodeholio

Reply 876 of 2412, by Jorpho

Reply 877 of 2412, by Synoptic

Reply 878 of 2412, by leileilol

Reply 879 of 2412, by emendelson