The commit a22a0ad7c4f0 was not entirely correct. Even with the patch,
some hangs still occur. This patch overrides the previous commit along
with the patch that makes cygwait() reentrant, to fix these hangs.
Addresses: https://cygwin.com/pipermail/cygwin/2024-December/256954.html
Addresses: https
To allow cygwait() to be called in the signal handler, a locally
created timer is used instead of _cygtls::locals.cw_timer if it is
in use.
Co-Authored-By: Corinna Vinschen
Signed-off-by: Takashi Yano
---
winsup/cygwin/cygtls.cc | 2 ++
winsup/cygwin/cygwait.cc | 22
This reverts commit a22a0ad7c4f0 to apply a new patch for the same
purpose.
Signed-off-by: Takashi Yano
---
winsup/cygwin/release/3.5.6 | 3 ---
winsup/cygwin/sigproc.cc| 20 +---
2 files changed, 5 insertions(+), 18 deletions(-)
diff --git a/winsup/cygwin/release/3.5.6 b/w
v5 -> v6:
Revise cygwait()/select() a bit
Takashi Yano (3):
Revert "Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is
sent"
Cygwin: cygwait: Make cygwait() reentrant
Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is sent
winsup/cygwin/cygtls.cc | 2
The commit a22a0ad7c4f0 was not exactly the correct thing. Even with
the patch, some hangs still happen. This patch overrides the previous
commit togerther with the patch making cygwait() reentrant to fix these
hangs.
Addresses: https://cygwin.com/pipermail/cygwin/2024-December/256954.html
Address
This reverts commit a22a0ad7c4f0 to apply a new patch for the same
purpose.
Signed-off-by: Takashi Yano
---
winsup/cygwin/release/3.5.6 | 3 ---
winsup/cygwin/sigproc.cc| 20 +---
2 files changed, 5 insertions(+), 18 deletions(-)
diff --git a/winsup/cygwin/release/3.5.6 b/w
To allow cygwait() to be called in the signal handler, a locally
created timer is used instead of _cygtls::locals.cw_timer if it is
in use.
Co-Authored-By: Corinna Vinschen
Signed-off-by: Takashi Yano
---
winsup/cygwin/cygtls.cc | 2 ++
winsup/cygwin/cygwait.cc | 23
Takashi Yano (3):
Revert "Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is
sent"
Cygwin: cygwait: Make cygwait() reentrant
Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is sent
winsup/cygwin/cygtls.cc | 2 ++
winsup/cygwin/cygwait.cc | 2
On Mon, 20 Jan 2025 09:03:01 -0800 (PST)
Jeremy Drake wrote:
> On Tue, 21 Jan 2025, Takashi Yano wrote:
>
> > diff --git a/winsup/cygwin/cygwait.cc b/winsup/cygwin/cygwait.cc
> > index 80c0e971c..8613638f6 100644
> > --- a/winsup/cygwin/cygwait.cc
> > +++ b/winsup/cygwin/cygwait.cc
> > @@ -58,16 +
On Tue, 21 Jan 2025, Takashi Yano wrote:
> diff --git a/winsup/cygwin/cygwait.cc b/winsup/cygwin/cygwait.cc
> index 80c0e971c..8613638f6 100644
> --- a/winsup/cygwin/cygwait.cc
> +++ b/winsup/cygwin/cygwait.cc
> @@ -58,16 +58,22 @@ cygwait (HANDLE object, PLARGE_INTEGER timeout, unsigned
> mask)
On Mon, 20 Jan 2025 16:53:34 +0100
Corinna Vinschen wrote:
> Hi Takashi,
>
> do you want to keep it this way, or do you rather want to change
> cygwait to use a local timer?
I prefer changing cygwait() to use a local timer because:
1) It makes the __SIGFLUSHFAST patch much simpler.
2) I wonder si
This reverts commit a22a0ad7c4f0 to apply a new patch for the same
purpose.
Signed-off-by: Takashi Yano
---
winsup/cygwin/release/3.5.6 | 3 ---
winsup/cygwin/sigproc.cc| 20 +---
2 files changed, 5 insertions(+), 18 deletions(-)
diff --git a/winsup/cygwin/release/3.5.6 b/w
Hi Takashi,
do you want to keep it this way, or do you rather want to change
cygwait to use a local timer?
Thanks,
Corinna
On Jan 20 17:52, Takashi Yano wrote:
> The commit a22a0ad7c4f0 was not exactly the correct thing. Even with
> the patch, some hangs still happen. This patch overrides the p
To allow cygwait() to be called in the signal handler, a locally
created timer is used instead of _cygtls::locals.cw_timer if it is
in use.
Co-Authored-By: Corinna Vinschen
Signed-off-by: Takashi Yano
---
winsup/cygwin/cygtls.cc | 2 ++
winsup/cygwin/cygwait.cc | 22
The commit a22a0ad7c4f0 was not exactly the correct thing. Even with
the patch, some hangs still happen. This patch overrides the previous
commit togerther with the patch making cygwait() reentrant to fix these
hangs.
Addresses: https://cygwin.com/pipermail/cygwin/2024-December/256954.html
Address
Takashi Yano (3):
Revert "Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is
sent"
Cygwin: cygwait: Make cygwait() reentrant
Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is sent
winsup/cygwin/cygtls.cc | 2 ++
winsup/cygwin/cygwait.cc | 2
On Mon, 20 Jan 2025 12:38:24 +0100
Corinna Vinschen wrote:
> On Jan 20 18:08, Takashi Yano wrote:
> > On Mon, 20 Jan 2025 00:33:26 +0900
> > Takashi Yano wrote:
> > > On Sun, 19 Jan 2025 09:40:14 +0900
> > > Takashi Yano wrote:
> > > > However, I wonder if cw_timer is re-set by NtSetTimer() in the
On 1/20/2025 6:49 AM, Corinna Vinschen wrote:
Nice idea, but this may not do what is expected if the mapping is an
anonymous mapping, leaving the protection or mapping of trailing pages
in a wrong state, isn't it?
Can we easily make sure the type of mapping (file vs anon) is known
at the time of
On Mon, 20 Jan 2025 12:43:07 +0100
Corinna Vinschen wrote:
> On Jan 19 19:42, Takashi Yano wrote:
> > Hi Corinna,
> >
> > On Sun, 19 Jan 2025 11:49:58 +0900
> > Takashi Yano wrote:
> > > On Sat, 18 Jan 2025 17:06:50 -0800 (PST)
> > > Jeremy Drake wrote:
> > > > On Sat, 18 Jan 2025, Takashi Yano wr
It seems that current _cygtls::handle_SIGCONT() code sometimes falls
into a deadlock due to frequent TLS lock/unlock operation in the
yield() loop. With this patch, the yield() in the wait loop is placed
outside the TLS lock to avoid frequent TLS lock/unlock.
Fixes: 9ae51bcc51a7 ("Cygwin: signal:
On Jan 17 18:22, Ken Brown wrote:
> On 1/15/2025 12:31 PM, Corinna Vinschen wrote:
> > > Ouch. It looks like we can't go to 64K bookkeeping. Windows files are
> > > not length-aligned to 64K allocation granularity, but to 4K pagesize.
> > > Thus, if we align the length to 64K in mprotect or
> > >
On Jan 19 19:42, Takashi Yano wrote:
> Hi Corinna,
>
> On Sun, 19 Jan 2025 11:49:58 +0900
> Takashi Yano wrote:
> > On Sat, 18 Jan 2025 17:06:50 -0800 (PST)
> > Jeremy Drake wrote:
> > > On Sat, 18 Jan 2025, Takashi Yano wrote:
> > >
> > > > While debugging this problem, I encountered another han
On Jan 20 18:08, Takashi Yano wrote:
> On Mon, 20 Jan 2025 00:33:26 +0900
> Takashi Yano wrote:
> > On Sun, 19 Jan 2025 09:40:14 +0900
> > Takashi Yano wrote:
> > > However, I wonder if cw_timer is re-set by NtSetTimer() in the
> > > cygwait(), it will be set to WSSC (60 sec) (or 10msec) in the
> >
On Mon, 20 Jan 2025 00:33:26 +0900
Takashi Yano wrote:
> On Sun, 19 Jan 2025 09:40:14 +0900
> Takashi Yano wrote:
> > On Sat, 18 Jan 2025 20:41:37 +0900
> > Takashi Yano wrote:
> > > On Fri, 17 Jan 2025 18:52:41 +0900
> > > Takashi Yano wrote:
> > > > On Wed, 8 Jan 2025 18:05:53 -0800 (PST)
> > > >
This reverts commit a22a0ad7c4f0 to apply a new patch for the same
purpose.
Signed-off-by: Takashi Yano
---
winsup/cygwin/release/3.5.6 | 3 ---
winsup/cygwin/sigproc.cc| 20 +---
2 files changed, 5 insertions(+), 18 deletions(-)
diff --git a/winsup/cygwin/release/3.5.6 b/w
The commit a22a0ad7c4f0 was not exactly the correct thing. Even with
the patch, some hangs still happen. This patch overrides the previous
commit to fix these hangs.
Addresses: https://cygwin.com/pipermail/cygwin/2024-December/256954.html
Addresses: https://cygwin.com/pipermail/cygwin/2024-Decembe
Takashi Yano (2):
Revert "Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is
sent"
Cygwin: signal: Do not handle signal when __SIGFLUSHFAST is sent
winsup/cygwin/release/3.5.6 | 3 ---
winsup/cygwin/sigproc.cc| 43 +++--
2 files changed, 27 in
Commentary wording now refers to tasks (i.e., threads) rather than
processes. This makes it somewhat easier to justify adding two kinds of
counters together. After researching what "load average" has meant over
time, we have what seems like a reasonable implementation, modulo
Windows differences
28 matches
Mail list logo
