2022年10月19日(水) 6:06 Thomas Munro :
>
> On Wed, Apr 6, 2022 at 10:40 AM Victor Spirin wrote:
> > Updated patch: we use the posix semantic features in Windows build 17763
> > and up.
> > We found an issue with this feature on Windows Server 2016 without
> > updates (Windows 1607 Build 14393)
>
> Hi
On Wed, Apr 6, 2022 at 10:40 AM Victor Spirin wrote:
> Updated patch: we use the posix semantic features in Windows build 17763
> and up.
> We found an issue with this feature on Windows Server 2016 without
> updates (Windows 1607 Build 14393)
Hi Victor,
I rebased and simplified this, and added
Hi,
On April 13, 2022 8:30:33 AM PDT, Robert Haas wrote:
>On Wed, Apr 13, 2022 at 11:03 AM Andres Freund wrote:
>> > Next decade's hot new processor design might do things
>> > differently enough that it matters that we use SpinLockInit()
>> > not memset-to-zero. This is not academic either,
On Wed, Apr 13, 2022 at 11:03 AM Andres Freund wrote:
> > Next decade's hot new processor design might do things
> > differently enough that it matters that we use SpinLockInit()
> > not memset-to-zero. This is not academic either, as we've had
> > exactly such bugs in the past.
>
> FWIW, I'l
Andres Freund writes:
> On 2022-04-13 10:19:44 -0400, Tom Lane wrote:
>> Next decade's hot new processor design might do things
>> differently enough that it matters that we use SpinLockInit()
>> not memset-to-zero. This is not academic either, as we've had
>> exactly such bugs in the past.
>
Hi,
On 2022-04-13 10:19:44 -0400, Tom Lane wrote:
> Meh. I agree that it seems unlikely that anyone will come out with a
> new processor design that lacks the ability to do spinlocks or atomics.
> It's substantially more likely though that someone would want those
> configure switches
On Wed, Apr 13, 2022 at 10:19 AM Tom Lane wrote:
> Meh. I agree that it seems unlikely that anyone will come out with a
> new processor design that lacks the ability to do spinlocks or atomics.
> It's substantially more likely though that someone would want those
> configure switches temporarily
Robert Haas writes:
> On Wed, Apr 13, 2022 at 3:04 AM Michael Paquier wrote:
>> Agreed. I think that things can be usually helpful. Now, I am not
>> really convinced that there is a strong need in running a VAX if you
>> are worrying about timing issues so this is a matter of balance. You
>>
On Wed, Apr 13, 2022 at 3:04 AM Michael Paquier wrote:
> Agreed. I think that things can be usually helpful. Now, I am not
> really convinced that there is a strong need in running a VAX if you
> are worrying about timing issues so this is a matter of balance. You
> could get down to the same
On Fri, Apr 08, 2022 at 11:44:31AM -0400, Greg Stark wrote:
> Generally I think supporting older systems that do funny things is
> helpful in avoiding problems that either 1) Can happen on newer
> systems but rarely 2) Can happen on other systems that people are
> using but we don't know about and
On Fri, Apr 8, 2022 at 11:45 AM Greg Stark wrote:
> But that's useful for some things and not for others. Like, it's
> useful to be sure we don't have odd dependencies on timing quirks of
> the specific machines that are currently common, or depend on gcc/llvm
> compiler behaviour that isn't
On Fri, 8 Apr 2022 at 11:30, Robert Haas wrote:
>
> On Fri, Apr 8, 2022 at 10:12 AM Greg Stark wrote:
> > On Thu, 9 Dec 2021 at 23:36, Tom Lane wrote:
> > > I'm not for dropping support for some platform just because it's old.
> >
> > I guess I'll have to spin up the Vax again :)
>
> This is a
On Fri, Apr 8, 2022 at 10:12 AM Greg Stark wrote:
> On Thu, 9 Dec 2021 at 23:36, Tom Lane wrote:
> > I'm not for dropping support for some platform just because it's old.
>
> I guess I'll have to spin up the Vax again :)
This is a pretty good summary of what's wrong with our current
deprecation
On Thu, 9 Dec 2021 at 23:36, Tom Lane wrote:
>
> I'm not for dropping support for some platform just because it's old.
I guess I'll have to spin up the Vax again :)
Hi
Updated patch: we use the posix semantic features in Windows build 17763
and up.
We found an issue with this feature on Windows Server 2016 without
updates (Windows 1607 Build 14393)
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian Postgres Company
05.07.2021
On Thu, Dec 09, 2021 at 11:33:17PM -0500, Tom Lane wrote:
> My general approach to platform compatibility is that when we
> break compatibility with old versions of something, we should do so
> because it will bring concrete benefits. If we can plausibly
> drop support for Windows versions that
Thomas Munro writes:
> Playing the devil's advocate here: why shouldn't we routinely drop
> support for anything that'll be EOL'd when a given PostgreSQL major
> release ships?
I don't like the word "routinely" here. Your next bit is a better
argument:
> Something EOL'd over a year ago that
On Fri, Dec 10, 2021 at 5:23 PM Thomas Munro wrote:
> Playing the devil's advocate here: why shouldn't we routinely drop
> support for anything that'll be EOL'd when a given PostgreSQL major
> release ships? The current policy seems somewhat extreme in the other
> direction: our target OS
On Tue, Jul 6, 2021 at 1:43 PM Michael Paquier wrote:
> This is a large bump for Studio >= 2015 I am afraid. That does not
> seem acceptable, as it means losing support for GetLocaleInfoEx()
> across older versions.
Playing the devil's advocate here: why shouldn't we routinely drop
support for
Hi
The flags for calling the CreateFile function have been changed.
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian Postgres Company
05.07.2021 16:53, Victor Spirin пишет:
Hi
I used the SetFileInformationByHandle function with the
FILE_RENAME_FLAG_POSIX_SEMANTICS
Hi
Added the pgunlink_windows_posix_semantics function and modified the
pgunlink function
I used FILE_DISPOSITION_POSIX_SEMANTICS flag for unlink files on Windows
10 (1607) and above.
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian Postgres Company
05.07.2021
Hi
Added a small fix for calling the GetFileAttributes function
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian Postgres Company
05.07.2021 16:53, Victor Spirin пишет:
Hi
I used the SetFileInformationByHandle function with the
FILE_RENAME_FLAG_POSIX_SEMANTICS flag
Thank you
Thank you
In this version of patch:
1. Made function isWindows1607OrGreater() without manifest
2. To open a directory using CreateFile, have to specify the
FILE_FLAG_BACKUP_SEMANTICS flag as part of dwFlagsAndAttributes. Checks
that file is a directory by the GetFileAttributes
On Thu, Sep 30, 2021 at 11:00 PM Victor Spirin
wrote:
>
> IsWindowsVersionOrGreater(10,0,1607) always returns false
>
> Only IsWindowsVersionOrGreater(10,0,0) is a valid call. (There are no
> service packs in Windows 10.)
>
> I haven't found a way to determine the Windows 10 release ID.
> The
Thanks.
IsWindowsVersionOrGreater(10,0,1607) always returns false
Only IsWindowsVersionOrGreater(10,0,0) is a valid call. (There are no
service packs in Windows 10.)
I haven't found a way to determine the Windows 10 release ID.
The RtlGetVersion function returns dwBuildNumber = 19042 on my
Thank you,
Fixed FILE_RENAME_INFO structure
I prepared 2 versions of the patch:
1) with manifest and IsWindows10OrGreater() function
2) without manifest and RtlGetVersion function from ntdll.dll
What's better?
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian
On Wed, Sep 8, 2021 at 10:13 PM Juan José Santamaría Flecha
wrote:
> On Thu, Jul 8, 2021 at 12:32 AM Victor Spirin wrote:
>> > #if defined(_MSC_VER) && _MSC_VER >= 1900
>> > -#define MIN_WINNT 0x0600
>> > +#define MIN_WINNT 0x0A00
>> > #else
>> > #define MIN_WINNT 0x0501
>> > #endif
>> >
On Wed, Sep 8, 2021 at 9:40 AM Victor Spirin wrote:
> Is this code better? Maybe there is another correct method?
Hmm, if we want to use the system header's struct definition, add some
space for a path at the end, and avoid heap allocation, perhaps we
could do something like:
struct {
On Tue, Sep 7, 2021 at 11:40 PM Victor Spirin
wrote:
>
> I checked the pgrename_windows_posix_semantics() function on Windows 7.
> It returns error 87: Parameter is incorrect. Hence, it is necessary to
> check the Windows version and call the old pgrename function for old
> Windows.
>
> The
On Thu, Jul 8, 2021 at 12:32 AM Victor Spirin
wrote:
>
> > #if defined(_MSC_VER) && _MSC_VER >= 1900
> > -#define MIN_WINNT 0x0600
> > +#define MIN_WINNT 0x0A00
> > #else
> > #define MIN_WINNT 0x0501
> > #endif
> > This is a large bump for Studio >= 2015 I am afraid. That does not
> >
Thank you,
In this variant:
1) renamed file win10.manifest to windows.manifest
2) renamed function pgrename_win10 to pgrename_windows_posix_semantics
3) Function pgrename returns result of pgrename_windows_posix_semantics
function and not contiue run old version of function.
4) Added call
On Tue, Sep 7, 2021 at 5:44 AM Victor Spirin wrote:
> I have changed the way I add the manifest to projects. I used the
> AdditionalManifestFiles option for a VS project.
Hi Victor,
Thanks for working on this!
I wonder if POSIX-style rename is used automatically on recent
Windows, based on the
Hello.
I have changed the way I add the manifest to projects. I used the
AdditionalManifestFiles option for a VS project.
Victor Spirin
Postgres Professional:http://www.postgrespro.com
The Russian Postgres Company
08.07.2021 1:32, Victor Spirin пишет:
Thanks!
In this version of the patch,
Thanks!
In this version of the patch, calls to malloc have been removed.
Hopefully MAX_PATH is long enough for filenames.
How does that cope with durable_rename_excl() where rename() is used
on Windows? The problems that 909b449 has somewhat "fixed" were
annoying for the users as it
On Mon, Jul 05, 2021 at 04:53:06PM +0300, Victor Spirin wrote:
> This patch related to this post:
> https://www.postgresql.org/message-id/CAEepm%3D0FV-k%2B%3Dd9z08cW%3DZXoR1%3Dkw9wdpkP6WAuOrKJdz-8ujg%40mail.gmail.com
How does that cope with durable_rename_excl() where rename() is used
on Windows?
Hi
I used the SetFileInformationByHandle function with the
FILE_RENAME_FLAG_POSIX_SEMANTICS flag for the file rename function..
1) The _WIN32_WINNT variable needs to be increased to 0x0A00 (Windows
10). Fixed conflict with #undef CHECKSUM_TYPE_NONE
2) The SetFileInformationByHandle
36 matches
Mail list logo
