[PATCH] Trivial pthread testsuite fixes

Redirected to here from [EMAIL PROTECTED] since I now have posting privileges. Thanks Elfyn McBratney. 2003-03-26 Brian Ford <[EMAIL PROTECTED]> * winsup.api/pthread/condvar7.c (mythread): Cast pthread_mutex_unlock argument of pthread_cleanup_push to void *, preven

gethostid and GetDiskFreeSpaceEx on NT4

The attached patch fixes the Cygwin testsuite failure I mentioned here: http://www.cygwin.com/ml/cygwin-developers/2003-09/msg00019.html 2003-09-16 Brian Ford <[EMAIL PROTECTED]> * syscalls.cc (gethostid): GetDiskFreeSpaceEx call on NT4 requires lpFreeBytesAvailable ar

Re: gethostid and GetDiskFreeSpaceEx on NT4

4.00.1381 It should be NT 4 with all the latest windows update stuff patched. But, this is an SGI 320, and so subject to all kinds of possibly weird stuff. Sorry, I really don't mean to be difficult. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightS

fhandler_tty_slave::ioctl (FIONBIO) return status

I don't really understand the code here in depth, but could someone explain to me why the attached trivial patch would'nt fix a minor bug? Thanks. 2003-10-23 Brian Ford <[EMAIL PROTECTED]> * fhandler_tty.c (fhandler_tty_slave::ioctl): Assure correct return v

Re: fhandler_tty_slave::ioctl (FIONBIO) return status

t;if (retval < 0) > { >set_errno (-retval); > You might want to move this if statement up too, as an optimization. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International Phone: 314-551-8460 Fax: 314-551-8444

fhandler_base::fcntl (F_SETFL)

I'm working on my stupid pedantic patch tricks. Can you tell? :) 2003-10-23 Brian Ford <[EMAIL PROTECTED]> * fhandler.cc (fhandler_base::fcntl): Don't clobber O_APPEND when both O_NONBLOCK/O_NDELAY are set for F_SETFL. -- Brian Ford Senior Realtime Softwar

fhandler_base::ioctl (FIONBIO)

Any reason not to support this? It seams to me that this patch just parallels what is already in fhandler_base::fcntl (F_SETFL) for O_NONBLOCK. I was trying to fix this issue: http://www.cygwin.com/ml/cygwin/2003-10/msg01159.html 2003-10-23 Brian Ford <[EMAIL PROTEC

Re: fhandler_base::ioctl (FIONBIO)

On Fri, 24 Oct 2003, Corinna Vinschen wrote: > On Thu, Oct 23, 2003 at 06:06:09PM -0500, Brian Ford wrote: > > Any reason not to support this? It seams to me that this patch just > > parallels what is already in fhandler_base::fcntl (F_SETFL) for > > O_NONBLOCK. > >

Re: Add PAGE_SIZE, PAGE_SHIFT, PAGE_MASK to sys/param.h

define PAGEOFFSET (PAGESIZE - 1) #define PAGEMASK(~PAGEOFFSET) It might be good to have them just for compatability. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International Phone: 314-551-8460 Fax: 314-551-8444

dtable.cc typo

I don't know c++ much/at all, but this looks wrong to me. I don't understand how it even compiled before? Feel free to slap me in the face because you can switch on a struct in c++? :) 2003-11-11 Brian Ford <[EMAIL PROTECTED]> * dtable.cc (build_fh_pc): Fix typo

dtable.cc (build_fh_pc): serial port handling

Here is one I think I do understand. 2003-11-11 Brian Ford <[EMAIL PROTECTED]> * dtable.cc (build_fh_pc): Use DEV_SERIAL_MAJOR to catch all serial ports. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International Phon

Re: dtable.cc (build_fh_pc): serial port handling

Revised patch attached. I noticed that there was a redundant cygdrive case too, so I removed it. 2003-11-12 Brian Ford <[EMAIL PROTECTED]> * dtable.cc (build_fh_pc): Use DEV_SERIAL_MAJOR to catch all serial ports. Remove redundant FH_CYGDRIVE case since it is h

Re: dtable.cc (build_fh_pc): serial port handling

On Thu, 13 Nov 2003, Christopher Faylor wrote: > On Wed, Nov 12, 2003 at 01:24:21PM -0600, Brian Ford wrote: > >2003-11-12 Brian Ford <[EMAIL PROTECTED]> > > > > * dtable.cc (build_fh_pc): Use DEV_SERIAL_MAJOR to catch all > > serial ports. Remove redun

(fhandler_base::lseek): Include high order bits in return.

This bug fix got our app past its first problem with > 2 Gig files, but then it tripped over ftello. I'm still trying to figure that one out. It looks like it got a 32 bit sign extended value somewhere. Any help would be appreciated. Thanks. 2003-11-17 Brian Ford <[EMAI

Re: (fhandler_base::lseek): Include high order bits in return.

On Mon, 17 Nov 2003, Brian Ford wrote: > This bug fix got our app past its first problem with > 2 Gig files, but > then it tripped over ftello. I'm still trying to figure that one out. > > It looks like it got a 32 bit sign extended value somewhere. Any help would >

Re: (fhandler_base::lseek): Include high order bits in return.

On Mon, 17 Nov 2003, Corinna Vinschen wrote: > On Mon, Nov 17, 2003 at 03:40:46PM -0600, Brian Ford wrote: > > On Mon, 17 Nov 2003, Brian Ford wrote: > > > > > This bug fix got our app past its first problem with > 2 Gig files, but > > > then it tripped over

Re: (fhandler_base::lseek): Include high order bits in return.

On Mon, 17 Nov 2003, Brian Ford wrote: > On Mon, 17 Nov 2003, Corinna Vinschen wrote: > > > On Mon, Nov 17, 2003 at 03:40:46PM -0600, Brian Ford wrote: > > > On Mon, 17 Nov 2003, Brian Ford wrote: > > > > > > > This bug fix got our app past its first pr

Fix tcflush hang with streaming devices

it too if the reporter supplies more info. 2003-12-10 Brian Ford <[EMAIL PROTECTED]> * fhandler_serial.cc (fhandler_serial::tcflush): Simplify. Remove read polling loop to avoid a hang with streaming devices. -- Brian Ford Senior Realtime Software Engineer VITAL - V

termios.cc: Restore setting of EBADF appropriately throughout

Ford <[EMAIL PROTECTED]> * termios.cc: Restore setting of EBADF appropriately throughout. Lost in version 1.16. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International Phone: 314-551-8460 Fax: 314-551-8444Index: term

Re: termios.cc: Restore setting of EBADF appropriately throughout

On Fri, 12 Dec 2003, Christopher Faylor wrote: > On Fri, Dec 12, 2003 at 01:14:32PM -0600, Brian Ford wrote: > >I noticed this while digging through other serial port problems. It > >appears to have been lost in version 1.16, although there did not appear > >to be a reason

Re: lstat symbolic link size

member shall contain meaningful > information when used with the file type macros, and the st_size member > shall contain the length of the pathname contained in the symbolic link. > > So, this is a PTC situation. > Ok, here it is. 2004-01-07 Brian Ford <[EMAIL PROT

Re: lstat symbolic link size

On Wed, 7 Jan 2004, Christopher Faylor wrote: > On Wed, Jan 07, 2004 at 11:33:18AM -0600, Brian Ford wrote: > >Ok, here it is. > > > >2004-01-07 Brian Ford <[EMAIL PROTECTED]> > > > > * fhandler_disk_file.cc (fhandler_base::fstat_helper): Comply wit

Fix write deadlock with streaming serial devices

-22 Brian Ford <[EMAIL PROTECTED]> * fhandler_serial.cc (fhandler_serial::raw_write): Prevent a deadlock when the input buffer overflows. (fhandler_serial::raw_read): Correct to print the actual error and only call PurgeComm when necessary. -- Brian Ford

Re: munmap slowness; IsBadReadPtr considered harmful

On Wed, 25 Feb 2004, Corinna Vinschen wrote: > On Feb 24 17:23, Brian Ford wrote: > > I guess it all depends on your interpretation of the following lines from: > > > > http://www.opengroup.org/onlinepubs/007904975/functions/munmap.html > > > > ERRORS: > >

sigproc.cc (proc_subproc): make -j hang

new 0x63C So, looking there, I stumbled onto the following: 2004-03-08 Brian Ford <[EMAIL PROTECTED]> * sigproc.cc (proc_subproc): Only call sync_proc_subproc->release() once for exec(). I'm not sure this is a bug, and it doesn't appear to fix the make hang I

Re: sigproc.cc (proc_subproc): make -j hang

On Mon, 8 Mar 2004, Christopher Faylor wrote: > On Mon, Mar 08, 2004 at 03:05:12PM -0600, Brian Ford wrote: > >While trying to analyze my own strace example of a "make -j hang" ala this > >ugly thread: > > > >http://www.cygwin.com/ml/cygwin/2004-03/msg00376.h

Re: sigproc.cc (proc_subproc): make -j hang

On Mon, 8 Mar 2004, Brian Ford wrote: > BTW, did moving the ProtectHandle1 call at line 349 outside the lock cause > a race with the ForceCloseHandle1 at line 516 in proc_terminate? My > limited testing shows much greater stability when it is moved back in. > Just curious. > Sam

Re: [Patch]: path.cc

;re saying, or the following from the Cygwin User's Guide is too dificult to interpret correctly? http://cygwin.com/cygwin-ug-net/using-cygwinenv.html: (no)smbntsec - if set, use ntsec on remote drives as well (this is the default). -- Brian Ford Senior Realtime Software Engineer VITAL -

wingdi.h (ENUMLOGFONTEXDV[AW]): breaks Cygwin (fwd)

ption? Thanks. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International Phone: 314-551-8460 Fax: 314-551-8444 -- Forwarded message -- Date: Mon, 19 Apr 2004 17:43:48 -0500 To: mingw-patches=lists.sourceforge.net Cc: Filip N

Re: [Patch] Fix gethwnd race

s, but I think the attached patch addresses the root of your objection to the previous patch. 2004-05-13 Brian Ford <[EMAIL PROTECTED]> * window.cc (window_started): Make NO_COPY. (gethwnd): Fix initialization race. (window_init): New function to initialize window_s

Re: [Patch] Fix gethwnd race

n.com/ml/cygwin/2004-05/msg00232.html > I'd like to see if what I've done actually solves anything. Me too ;-). -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained pilot...

Re: [Patch] Fix gethwnd race

On Thu, 13 May 2004, Brian Ford wrote: > On Thu, 13 May 2004, Christopher Faylor wrote: > > I'd like to see if what I've done actually solves anything. > > Me too ;-). PS. In case it wasn't obvious, none of that was meant to offend. I'm genuinely curious to s

Re: [Patch] Fix gethwnd race

object/handle. Yes, I thought of that but it didn't seem as clean. Once again, I don't care, but I did have a good reason for choosing the method I presented. (At least, IMHO it was a good reason). -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSaf

yRe: [Patch] Fix gethwnd race

On Thu, 13 May 2004, Brian Ford wrote: > On Thu, 13 May 2004, Christopher Faylor wrote: > > If it does exist, just return it. No locking required. > > Ok, add: > > if (ourhwnd) > return ourhwnd; > > to the beginning of my patch if your worried about the interlock

Re: [Patch] Fix gethwnd race

this just be a fatal condition? That would sure make things simpler. I don't think any calling code is prepared to deal with that failure anyway. I'll cook up a muto based solution in the next few minutes now that I understand your preference. But, I'd like an opinion on the questio

Re: [Patch] Fix gethwnd race

On Fri, 14 May 2004, Christopher Faylor wrote: > On Fri, May 14, 2004 at 10:59:42AM -0500, Brian Ford wrote: > >I'll cook up a muto based solution in the next few minutes now > >that I understand your preference. But, I'd like an opinion on the > >question abov

Re: [Patch] Fix gethwnd race

st have trusted Sergey on this one. > > So, nevermind on this wonderful enhancement. Just in case it might still be "interesting", here is a revised no muto patch with the appropriate fatal error changes. I think it is pretty straight forward. If you still prefer the muto route, ple

Re: [Patch] Fix gethwnd race

Ugh! Ok, this falls under the "too much list noise" category, so I'll just shut up now. On Fri, 14 May 2004, Brian Ford wrote: + HANDLE ws; + + if (InterlockedDecrement (&window_waiters) == 0 + && (ws = (HANDLE) InterlockedExchange ((long *) &window_sta

Re: cygwin1.dll 1.5.9-1 and pthread_exit: UNSTABLE

e. We would prefer that you didn't tell us what "a good idea" is. Thanks. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained pilot...

[Patch] Fix gethwnd race

only pursued it to the point of being sure it is unrelated. I'll look at it more later if time permits. 2004-05-06 Brian Ford <[EMAIL PROTECTED]> * window.cc (window_started): Change type to long and make NO_COPY. (Winmain): Use InterlockedExchange instead

Re: [Patch] Fix gethwnd race

Uh, scratch that. Original patch stands. With this change, close could still get called multiple times. Digs hole and hides... :-( On Fri, 14 May 2004, Brian Ford wrote: > Ugh! Ok, this falls under the "too much list noise" category, so I'll > just shut up now. > >

Re: [Patch] To handle Win32 pipe names

or a slightly modified version of is_unc_share? I'm confused here. Are you suggesting that UNC paths can't contain symlinks? I use UNC paths to access Samba shares containing Cygwin symlinks regularly. I hope that won't become unsupported. -- Brian Ford Senior Realtime Softwa

[UG Patch] kmem and check_case typo

atch to the users guide? (plus a typo fix) 2004-05-21 Brian Ford <[EMAIL PROTECTED]> * pathnames.sgml: Remove /dev/kmem from the supported POSIX device list. * cygwinenv.sgml: Fix typo in check_case description. -- Brian Ford Senior Realtime Software Engineer VITA

Re: [PATCH]: Still path.cc

t I didn't have time to make. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained pilot...

Re: [PATCH]: Still path.cc

On Wed, 22 Sep 2004, Christopher Faylor wrote: > On Wed, Sep 22, 2004 at 09:20:50AM -0500, Brian Ford wrote: > >On Tue, 21 Sep 2004, Pierre A. Humblet wrote: > >>Avoid infinite loop with names starting in double dots. > > > >This may not be appropriate for this

[OT] Re: [PATCH]: Still path.cc

On Wed, 22 Sep 2004, Christopher Faylor wrote: > On Wed, Sep 22, 2004 at 11:07:09AM -0500, Brian Ford wrote: > >You can't win here. If you're a developer like person, the answer to > >bug reports is usually "fix it yourslef if it's important to you". If &

Re: [Patch] cygcheck: warn about empty path-components

the path list. etc... I believe this is a valid construct and I have used it frequently. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained pilot...

Re: Need help with procps-010801-2 patch.

have a cygwin build environment set up and was wondering whether someone has > a patched procps-010801-2 executable tarball or if could point me to where I could > download one. > > Thanks very much in advance. > > Eli Lauris > -- Brian Ford Senior Realtime Software Engineer

Re: [Patch] Deimpersonate while accessing HKLM

libiberty/fibheap.c:285: error: for each function it appears in.) ../../cygwin/libiberty/fibheap.c: In function `fibheap_consolidate': ../../cygwin/libiberty/fibheap.c:395: warning: implicit declaration of function `memset' which I'm sure has nothing to do with this patch. I'll loo

Re: [Patch] Deimpersonate while accessing HKLM

On Thu, 28 Oct 2004, Brian Ford wrote: > This may very well be user error, but I thought it important to report. > After compiling current CVS I get a slew of: > > 53 [main] sh 3064 cygheap_user::reimpersonate: ImpersonateLoggedOnUser: > Win32 error 6 Yup, user error. After a cl

[PATCH] Re: IP_MULTICAST_IF et all / Winsock[2] value conflict

On Fri, 30 Sep 2005, Corinna Vinschen wrote: > On Sep 30 10:07, Brian Ford wrote: > > We can simply translate the current constant Winsock 1 values to Winsock 2 > > ones when necessary in cygwin_[set|get]sockopt. There are only 8 values > > that need changing, I think. >

Re: [PATCH] Re: IP_MULTICAST_IF et all / Winsock[2] value conflict

On Mon, 3 Oct 2005, Corinna Vinschen wrote: > On Sep 30 18:39, Brian Ford wrote: > > On Fri, 30 Sep 2005, Corinna Vinschen wrote: > > > Actually we have two states, applications built before we changed the > > > header file and applications built after we changed the

Re: [PATCH] Re: IP_MULTICAST_IF et all / Winsock[2] value conflict

On Mon, 17 Oct 2005, Corinna Vinschen wrote: > On Oct 3 12:25, Brian Ford wrote: > > On Mon, 3 Oct 2005, Corinna Vinschen wrote: > > > No, let's change it. Winsock2 is the way to go. Winsock1 is just old > > > stuff. Since Cygwin is using Winsock2 when runni

Re: [PATCH] Re: IP_MULTICAST_IF et all / Winsock[2] value conflict

at Winsock1 has been dropped. I hope so too. I'll test as soon as it is possible. Thanks again. -- Brian Ford Senior Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained pilot...

[PATCH] Translate INSUFFICIENT_RESOURCES errno

Trivial patch to do subject. This would have made the error in this thread more consistent: http://cygwin.com/ml/cygwin/2005-11/msg00110.html 2005-11-05 Brian Ford <[EMAIL PROTECTED]> * errno.cc (errmap): Handle INSUFFICIENT_RESOURCES. -- Brian Ford Senior Realtime So

Re: [PATCH] Translate INSUFFICIENT_RESOURCES errno

On Sat, 5 Nov 2005, Christopher Faylor wrote: > On Sat, Nov 05, 2005 at 10:41:35AM -0600, Brian Ford wrote: > > * errno.cc (errmap): Handle INSUFFICIENT_RESOURCES. > > I don't see an ERROR_INSUFFICIENT_RESOURCES Um..., nevermind. Mea culpa. Sorry to waste your time.

Re: [Patch] regtool: Add load/unload commands and --binary option

On Wed, 25 Jan 2006, Igor Peshansky wrote: > I'm not aware of any program that does the reverse (hex dump->binary), > but writing a perl script for that is trivial. xxd -r ;-). -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety Internat

clock_[get|set]time timespec conversions

It's late and I haven't had time to test this, but I thought it deserved a look. There appears to be some confusion (at least in my head) about the units of gtod.resolution() and minperiod. 2006-02-10 Brian Ford <[EMAIL PROTECTED]> * times.cc (clock_gettime): Pro

Re: clock_[get|set]res timespec conversions

I said it was late. Oh, well. Make that clock_[get|set]RES... On Fri, 10 Feb 2006, Brian Ford wrote: > It's late and I haven't had time to test this, but I thought it deserved a > look. There appears to be some confusion (at least in my head) about the > units of gt

raw email addresses and ChangeLog entries?

out the above dichotomy. Is it ok to obfuscate ChangeLog email addresses? Otherwise, what's the point of PCYMTNQREAIYR requests in cygwin-patches? Sorry for being slightly OT. -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the bes

Re: raw email addresses and ChangeLog entries?

On Fri, 17 Feb 2006, Corinna Vinschen wrote: > On Feb 17 09:52, Brian Ford wrote: > > On Fri, 17 Feb 2006, Corinna Vinschen wrote: > > > On Feb 16 10:46, Jerry D. Hedden wrote: > > > > > > Would you please don't copy raw email addresses in your

allow read into untouched noreserve mappings

, I'll consider doing something similar for recvmsg and recvfrom. That should cover the majority of cases, I believe. 2006-07-17 Brian Ford <[EMAIL PROTECTED]> * winsup.h (mmap_commit_noreserve_pages): New prototype. * mmap.cc (fhandler_base::raw_read): New function.

Re: allow read into untouched noreserve mappings

alled. I was trying to be less intrusive and that looked messy. Oh, well... Untested this time because I have to run to an appointment. 2006-07-17 Brian Ford <[EMAIL PROTECTED]> * winsup.h (mmap_region_status): New enum. (mmap_is_attached_or_noreserve_page): Adjust

Re: allow read into untouched noreserve mappings

On Mon, 17 Jul 2006, Brian Ford wrote: > Untested this time because I have to run to an appointment. Now tested and working fine with no changes. I guess I can infer from cgf's mail to cygwin-developers that this will not make it into 1.5.21 :-(. -- Brian Ford Lead Realtime Software

Re: allow read into untouched noreserve mappings

after the check for an empty mmap region list. > Thanks for the patch. It's available for further digestion and patches > in the cv-branch. I guess it'll be a while then before this hits a release then :-(. Thanks for applying anyway. -- Brian Ford Lead Realtime Software Engineer V

Re: allow read into untouched noreserve mappings

On Tue, 18 Jul 2006, Christopher Faylor wrote: > On Tue, Jul 18, 2006 at 08:22:38AM -0500, Brian Ford wrote: > > >I guess I can infer from cgf's mail to cygwin-developers that this will > >not make it into 1.5.21 :-(. > > I guess the fact that Corinna mentioned

Re: allow read into untouched noreserve mappings

On Tue, 18 Jul 2006, Corinna Vinschen wrote: > On Jul 18 11:59, Brian Ford wrote: > > I confess to not understanding the purpose and inner workings of either > > search_record implementation. I first tried the two parameter version, > > but then realized it was searching

check_iovec cleanup

I think this removes some redundancy and a questionable check while being more straight forward and covering more possible fault cases. Please let me know if I missed a reason it had to be that way. 2006-07-26 Brian Ford <[EMAIL PROTECTED]> * miscfuncs.cc (dummytest):

Re: check_iovec cleanup

cases), and a minuscule performance/elegance improvement to the general case justified the change. Maybe not. -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained crew...

[send|recv]msg tidy

Ok, I'll try this one. 2006-07-26 Brian Ford <[EMAIL PROTECTED]> * fhandler_socket.cc (fhandler_socket::recvmsg): Remove unused tot argument. All callers changed. (fhandler_socket::sendmsg): Likewise. * net.cc (cygwin_recvmsg): Likewise. (cyg

get_readahead; remove duplicate

One more tiny one. 2006-07-26 Brian ford <[EMAIL PROTECTED]> * fhandler.cc (fhandler_base::read): Call get_readahead_into_buffer instead of duplicating it. -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International th

Prevent closing a NULL pinfo handle

fo_spawn():125, Win32 error 6 2006-08-17 Brian Ford <[EMAIL PROTECTED]> * child_info.h (~child_info_spawn): Prevent closing a NULL handle. Although, I suspect if the correct thing to do were that simple, it would already have been noticed and fixed? And yes, I know that functionall

Increase st_blksize to 64k

As suggested here: http://cygwin.com/ml/cygwin/2006-12/msg00911.html 2007-01-02 Brian Ford <[EMAIL PROTECTED]> * fhandler.cc (fhandler_base::fstat): Use system page size (64k) as the st_blksize prefered I/O size for improved performance. * fhandler_disk_f

Re: Increase st_blksize to 64k

block units reported in the stat structure. S_BLKSIZE has historically been 512 bytes. These are actually two different things. -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained crew...

Re: Increase st_blksize to 64k

blcm, overriding the above adjustments. Watch out for overflow. */ buf_size += blcm - 1; buf_size -= buf_size % blcm; if (buf_size == 0 || blcm_max < buf_size) buf_size = blcm; } -- Brian Ford Lead Realtime Software Engineer VITAL -

Re: Increase st_blksize to 64k

difies that 8KB of data. This means that for each buffered random write includes one or more 64KB reads. The right side of Figure 11 shows this 100% IO penalty. -- Brian Ford Lead Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in an

Re: [PATCHes] Misc aliasing fixes for building DLL with gcc-4.5.0

ne who tries to figure out what this means years down the road. Without the other context of this thread, I would not have any idea what software that PR applied to. -- Brian Ford Staff Realtime Software Engineer VITAL - Visual Simulation Systems FlightSafety International the best safety device in any aircraft is a well-trained crew...