Re: [9fans] Mouse jumps in vmware
Maybe someone will be motivated to fill in the rest using the latest and greatest VMware API instead of one from seven and a half years ago. Russ, is this the right API? http://open-vm-tools.sourceforge.net/
Re: [9fans] Mouse jumps in vmware
On Thu, Dec 11, 2008 at 10:47 AM, Skip Tavakkolian 9...@9netics.com wrote: Maybe someone will be motivated to fill in the rest using the latest and greatest VMware API instead of one from seven and a half years ago. Russ, is this the right API? http://open-vm-tools.sourceforge.net/ Presumably that code uses the right API. Russ
Re: [9fans] Mouse jumps in vmware
Ummm, I am not sure if the problem is vmware-tools, probably not (and Plan 9 is not the problem too). I am using linux + plan9port + acme and I get the same problem ... I am confused... BSOD @ me Skip Tavakkolian escribió: Maybe someone will be motivated to fill in the rest using the latest and greatest VMware API instead of one from seven and a half years ago. Russ, is this the right API? http://open-vm-tools.sourceforge.net/
Re: [9fans] Mouse jumps in vmware
Wrong: http://open-vm-tools.sourceforge.net/
Now maybe somebody will release the code for the /bin/aux/vm* vmware
tools so we can fix them.
In order to get the information that I used to
write aux/vmware, I signed an NDA that prohibits
me from releasing the source code, regardless
of what VMware has done in the interim. At the
time, that was the only way to get the information.
That said, the mechanism for backdoor calls into
VMware has completely changed at least twice
since I wrote aux/vmware, which is why it no longer
works very well. It's no loss to start from scratch.
If I thought the program was still useful today, I might
try to get permission to release it. But it's just not
useful anymore in any real sense.
Also, the program is pretty small:
vx32% wc *.[chs] mkfile
84 2061683 all.h
27 49 289 asmbackdoor.s
34 83 539 backdoor.c
16 33 217 calls.c
6421303 10148 fs.c
176 3262224 inout.c
33 46 392 isvmware.c
10 13 98 mousepoll.c
224 5513677 msgchan.c
44 69 554 vmmousepoll.c
12 11 177 mkfile
13022690 19998 total
vx32%
I have attached fs.c, which is quite generic
and has no VMware-specific goo. Maybe
someone will be motivated to fill in the rest
using the latest and greatest VMware API
instead of one from seven and a half years ago.
Or maybe Uriel will just keep whining.
Russ
#include all.h
#include fcall.h
#include thread.h
#include 9p.h
char *mtpt = /mnt/vmware;
char *srvname;
uint time0;
enum
{
Qroot = 0,
Qmousepoint,
Qsnarf,
Qgui,
Qdev,
Qtime,
Qbintime,
Qmsg,
};
typedef struct Tab Tab;
struct Tab
{
int qid;
char *name;
uint perm;
uint vers;
void (*open)(Req*);
void (*read)(Req*);
void (*write)(Req*);
void (*close)(Fid*);
};
static void
mousepointread(Req *r)
{
char buf[32];
Point p;
p = getmousepoint();
snprint(buf, sizeof buf, %11d %11d , p.x, p.y);
readstr(r, buf);
respond(r, nil);
}
static void
mousepointwrite(Req *r)
{
char buf[32], *f[3];
int nf, n;
Point p;
n = r-ifcall.count;
if(n = sizeof buf){
respond(r, write too large);
return;
}
memmove(buf, r-ifcall.data, n);
buf[n] = '\0';
nf = tokenize(buf, f, nelem(f));
if(nf != 2){
respond(r, bad point format);
return;
}
p.x = atoi(f[0]);
p.y = atoi(f[1]);
setmousepoint(p);
respond(r, nil);
}
static void
timeread(Req *r)
{
char buf[32];
uint sec, microsec, lag;
gettime(sec, microsec, lag);
snprint(buf, sizeof buf, %11d , sec);
readstr(r, buf);
respond(r, nil);
}
static uvlong uvorder = 0x0001020304050607ULL;
static uchar*
vlong2le(uchar *t, vlong from)
{
uchar *f, *o;
int i;
f = (uchar*)from;
o = (uchar*)uvorder;
for(i = 0; i sizeof(vlong); i++)
t[i] = f[o[i]];
return t+sizeof(vlong);
}
static void
bintimeread(Req *r)
{
uchar *b;
int i, n;
uint sec, microsec, lag;
vlong nsec;
b = (uchar*)r-ofcall.data;
n = r-ifcall.count;
i = 0;
if(n = 8){
gettime(sec, microsec, lag);
nsec = sec*10LL+microsec*1000LL;
vlong2le(b, nsec);
i = 8;
}
if(n = 16){
vlong2le(b+8, nsec);
i = 16;
}
if(n = 24){
vlong2le(b+16, 10LL);
i = 24;
}
r-ofcall.count = i;
respond(r, nil);
}
char *snarf;
int nsnarf;
char *tsnarf;
int ntsnarf;
static void
snarfread(Req *r)
{
int i;
if(r-ifcall.offset == 0){
if(snarf)
free(snarf);
nsnarf = getsnarflength();
snarf = emalloc9p(nsnarf+4+1);
for(i=0; insnarf; i+=4)
*(uint*)(snarf+i) = getsnarfpiece();
snarf[nsnarf] = '\0';
nsnarf = strlen(snarf); /* there's extra crap because we have to transfer 4 bytes at a time */
}
readbuf(r, snarf, nsnarf);
respond(r, nil);
}
static void
snarfwrite(Req *r)
{
if(r-ifcall.offset == 0){
free(tsnarf);
tsnarf = nil;
ntsnarf = 0;
}
if(r-ifcall.offset 100*1024){
respond(r, snarf buffer too long);
return;
}
tsnarf = erealloc9p(tsnarf, ntsnarf+r-ifcall.count);
memmove(tsnarf+ntsnarf, r-ifcall.data, r-ifcall.count);
ntsnarf += r-ifcall.count;
r-ofcall.count = r-ifcall.count;
respond(r, nil);
}
static void
snarfclose(Fid *fid)
{
int i;
if((fid-omode3) == OREAD)
return;
setsnarflength(ntsnarf);
for(i=0; intsnarf; i+=4)
setsnarfpiece(*(uint*)(tsnarf+i));
free(tsnarf);
tsnarf = nil;
ntsnarf = 0;
}
typedef struct Bit Bit;
struct Bit
{
char *name;
uint bit;
};
Bit guibit[] = {
autograb, 1,
autorelease, 2,
autoscroll, 4,
autoraise, 8,
copypaste, 0x10,
hidecursor, 0x20,
fullscreen, 0x40,
tofullscreen, 0x80,
towindow, 0x100,
autoraise-disabled, 0x200,
synctime, 0x400,
};
static void
guiread(Req *r)
{
int i;
char *s;
Fmt fmt;
uint val;
val = getguistate();
fmtstrinit(fmt);
for(i=0; inelem(guibit); i++)
fmtprint(fmt, %s %s\n, guibit[i].name, (val guibit[i].bit) ? on : off);
s = fmtstrflush(fmt);
readstr(r, s);
free(s);
respond(r, nil);
}
static void
guiwrite(Req *r)
{
int i, on;
uint v;
Cmdbuf *cb;
cb = parsecmd(r-ifcall.data, r-ifcall.count);
Re: [9fans] Mouse jumps in vmware
Hi Lucio, I am happy with vmware. Is very good environment for my plan9, because my wifi card is not supported, ... the problem is vmware is to avoid comments about it. [EMAIL PROTECTED] escribió: Yes, I know, the problem is vmware :-) No, the problem is _not_ vmware. At least to the extent that vmware is just another hardware model. The only really unfortunate thing about vmware is that they are holding a few cards close to their chest with a view to making a bit of a profit. It may not be to one's greatest advantage, but it seems legitimate to me and above the of criticism levelled in the above. ++L PS: I'm a contented small time user of vmware's ESX Server 3i. All the products from vmware (not many, I must confess) I have had occasion to use and in keeping with impressions gained while monitoring dicussions on the Xen developers' mailing list, convince me that vmware strive for professionalism, understandably within the limits set by the marketplace. It's my personal opinion that none of the equivalent tools available today are in the same league, so I can't resist defending what I consider to be the market leader; feel free to correct me (privately, preferably).
Re: [9fans] Mouse jumps in vmware
I am happy with vmware. Is very good environment for my plan9, because my wifi card is not supported, ... No offence intended or taken, I just thought that vmware generally get an undeservedly bad rap on this list :-) ++L
Re: [9fans] Mouse jumps in vmware
I tryed xen, qemu and vmware ... and IMO vmware is better for my plan9 virtual machines. In other thread (probably long) ... we need think about to support any emulated software. We are not many people and write drivers is a hard work. I am not sure what is the rate of 9fans using real/emulated hardware. [EMAIL PROTECTED] escribió: I am happy with vmware. Is very good environment for my plan9, because my wifi card is not supported, ... No offence intended or taken, I just thought that vmware generally get an undeservedly bad rap on this list :-) ++L
Re: [9fans] Mouse jumps in vmware
On Sun, Dec 7, 2008 at 9:44 AM, [EMAIL PROTECTED] wrote: Yes, I know, the problem is vmware :-) No, the problem is _not_ vmware. At least to the extent that vmware is just another hardware model. The only really unfortunate thing about vmware is that they are holding a few cards close to their chest with a view to making a bit of a profit. It may not be to one's greatest advantage, but it seems legitimate to me and above the of criticism levelled in the above. Wrong: http://open-vm-tools.sourceforge.net/ Now maybe somebody will release the code for the /bin/aux/vm* vmware tools so we can fix them. Peace uriel - Not holding his breath. ++L PS: I'm a contented small time user of vmware's ESX Server 3i. All the products from vmware (not many, I must confess) I have had occasion to use and in keeping with impressions gained while monitoring dicussions on the Xen developers' mailing list, convince me that vmware strive for professionalism, understandably within the limits set by the marketplace. It's my personal opinion that none of the equivalent tools available today are in the same league, so I can't resist defending what I consider to be the market leader; feel free to correct me (privately, preferably).
Re: [9fans] Mouse jumps in vmware
We are not many people and write drivers is a hard work. why not give it a shot? writing drivers is the most straightforward and rewarding part of my job. - erik
Re: [9fans] Mouse jumps in vmware
writing drivers is the most straightforward Have you found some devices with fully documented programming interfaces, which actually behave as the documentation describes? Lucky man. and rewarding part of my job. There I agree with you. But I like solving puzzles.
Re: [9fans] Mouse jumps in vmware
writing drivers is the most straightforward Have you found some devices with fully documented programming interfaces, which actually behave as the documentation describes? Lucky man. sure. intel writes good documentation these days. - erik
