Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
--- Davide Libenzi wrote: > Charles, would you mind trying the patch below > against -git13 on your > machine. I tested it with wine and firefox on a 32 > bit P4 with HT and it's > working fine. i applied the patch against git13. starcraft, pokerstars, and firefox under wine have not frozen. looks good. Never miss an email again! Yahoo! Toolbar alerts you the instant new Mail arrives. http://tools.search.yahoo.com/toolbar/features/mail/ - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Davide Libenzi wrote:
> On Thu, 10 May 2007, Charles Gagalac wrote:
>
> > i just installed the windows version of firefox
> > 2.0.0.3 under wine and firefox also froze after a few
> > minutes of use.
> >
> > also, wine version is wine-0.9.36.
>
> Thank you Charles. Would you mind giving the patch below a try?
Charles, would you mind trying the patch below against -git13 on your
machine. I tested it with wine and firefox on a 32 bit P4 with HT and it's
working fine.
Andrew, do not even look at it. Comments are totally broken, and for sure
does not apply over the latest tree after you pushed the other epoll bits.
I'll repost a polished version once it has been verfied.
- Davide
Index: linux-2.6.21/fs/eventpoll.c
===
--- linux-2.6.21.orig/fs/eventpoll.c2007-05-10 11:47:10.0 -0700
+++ linux-2.6.21/fs/eventpoll.c 2007-05-11 12:18:20.0 -0700
@@ -27,7 +27,6 @@
#include
#include
#include
-#include
#include
#include
#include
@@ -39,7 +38,6 @@
#include
#include
#include
-#include
/*
@@ -47,7 +45,7 @@
* There are three level of locking required by epoll :
*
* 1) epmutex (mutex)
- * 2) ep->sem (rw_semaphore)
+ * 2) ep->mtx (mutex)
* 3) ep->lock (rw_lock)
*
* The acquire order is the one listed above, from 1 to 3.
@@ -58,19 +56,19 @@
* a spinlock. During the event transfer loop (from kernel to
* user space) we could end up sleeping due a copy_to_user(), so
* we need a lock that will allow us to sleep. This lock is a
- * read-write semaphore (ep->sem). It is acquired on read during
- * the event transfer loop and in write during epoll_ctl(EPOLL_CTL_DEL)
- * and during eventpoll_release_file(). Then we also need a global
- * semaphore to serialize eventpoll_release_file() and ep_free().
- * This semaphore is acquired by ep_free() during the epoll file
+ * mutex (ep->mtx). It is acquired during the event transfer loop,
+ * during epoll_ctl(EPOLL_CTL_DEL) and during eventpoll_release_file().
+ * Then we also need a global mutex to serialize eventpoll_release_file()
+ * and ep_free().
+ * This mutex is acquired by ep_free() during the epoll file
* cleanup path and it is also acquired by eventpoll_release_file()
* if a file has been pushed inside an epoll set and it is then
- * close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
- * It is possible to drop the "ep->sem" and to use the global
- * semaphore "epmutex" (together with "ep->lock") to have it working,
- * but having "ep->sem" will make the interface more scalable.
+ * close()d without a previous call to epoll_ctl(EPOLL_CTL_DEL).
+ * It is possible to drop the "ep->ntx" and to use the global
+ * mutex "epmutex" (together with "ep->lock") to have it working,
+ * but having "ep->mtx" will make the interface more scalable.
* Events that require holding "epmutex" are very rare, while for
- * normal operations the epoll private "ep->sem" will guarantee
+ * normal operations the epoll private "ep->mtx" will guarantee
* a greater scalability.
*/
@@ -106,6 +104,7 @@
#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
+#define EP_UNACTIVE_PTR ((void *) -1L)
struct epoll_filefd {
struct file *file;
@@ -135,6 +134,44 @@
};
/*
+ * Each file descriptor added to the eventpoll interface will
+ * have an entry of this type linked to the "rbr" RB tree.
+ */
+struct epitem {
+ /* RB-Tree node used to link this structure to the eventpoll rb-tree */
+ struct rb_node rbn;
+
+ /* List header used to link this structure to the eventpoll ready list
*/
+ struct list_head rdllink;
+
+ /* The file descriptor information this item refers to */
+ struct epoll_filefd ffd;
+
+ /* Number of active wait queue attached to poll operations */
+ int nwait;
+
+ /* List containing poll wait queues */
+ struct list_head pwqlist;
+
+ /* The "container" of this item */
+ struct eventpoll *ep;
+
+ /* The structure that describe the interested events and the source fd
*/
+ struct epoll_event event;
+
+ /*
+* Used to keep track of the usage count of the structure. This avoids
+* that the structure will desappear from underneath our processing.
+*/
+ atomic_t usecnt;
+
+ /* List header used to link this item to the "struct file" items list */
+ struct list_head fllink;
+
+ struct epitem *next;
+};
+
+/*
* This structure is stored inside the "private_data" member of the file
* structure and rapresent the main data sructure for the eventpoll
* interface.
@@ -144,12 +181,10 @@
rwlock_t lock;
/*
-* This semaphore is used to ensure that files are not removed
-* while epoll is using them. This is read-held during the event
-* collection loop and it is write-held during the file cleanup
-* path, the epoll file exit
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Davide Libenzi wrote:
On Thu, 10 May 2007, Charles Gagalac wrote:
i just installed the windows version of firefox
2.0.0.3 under wine and firefox also froze after a few
minutes of use.
also, wine version is wine-0.9.36.
Thank you Charles. Would you mind giving the patch below a try?
Charles, would you mind trying the patch below against -git13 on your
machine. I tested it with wine and firefox on a 32 bit P4 with HT and it's
working fine.
Andrew, do not even look at it. Comments are totally broken, and for sure
does not apply over the latest tree after you pushed the other epoll bits.
I'll repost a polished version once it has been verfied.
- Davide
Index: linux-2.6.21/fs/eventpoll.c
===
--- linux-2.6.21.orig/fs/eventpoll.c2007-05-10 11:47:10.0 -0700
+++ linux-2.6.21/fs/eventpoll.c 2007-05-11 12:18:20.0 -0700
@@ -27,7 +27,6 @@
#include linux/hash.h
#include linux/spinlock.h
#include linux/syscalls.h
-#include linux/rwsem.h
#include linux/rbtree.h
#include linux/wait.h
#include linux/eventpoll.h
@@ -39,7 +38,6 @@
#include asm/io.h
#include asm/mman.h
#include asm/atomic.h
-#include asm/semaphore.h
/*
@@ -47,7 +45,7 @@
* There are three level of locking required by epoll :
*
* 1) epmutex (mutex)
- * 2) ep-sem (rw_semaphore)
+ * 2) ep-mtx (mutex)
* 3) ep-lock (rw_lock)
*
* The acquire order is the one listed above, from 1 to 3.
@@ -58,19 +56,19 @@
* a spinlock. During the event transfer loop (from kernel to
* user space) we could end up sleeping due a copy_to_user(), so
* we need a lock that will allow us to sleep. This lock is a
- * read-write semaphore (ep-sem). It is acquired on read during
- * the event transfer loop and in write during epoll_ctl(EPOLL_CTL_DEL)
- * and during eventpoll_release_file(). Then we also need a global
- * semaphore to serialize eventpoll_release_file() and ep_free().
- * This semaphore is acquired by ep_free() during the epoll file
+ * mutex (ep-mtx). It is acquired during the event transfer loop,
+ * during epoll_ctl(EPOLL_CTL_DEL) and during eventpoll_release_file().
+ * Then we also need a global mutex to serialize eventpoll_release_file()
+ * and ep_free().
+ * This mutex is acquired by ep_free() during the epoll file
* cleanup path and it is also acquired by eventpoll_release_file()
* if a file has been pushed inside an epoll set and it is then
- * close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
- * It is possible to drop the ep-sem and to use the global
- * semaphore epmutex (together with ep-lock) to have it working,
- * but having ep-sem will make the interface more scalable.
+ * close()d without a previous call to epoll_ctl(EPOLL_CTL_DEL).
+ * It is possible to drop the ep-ntx and to use the global
+ * mutex epmutex (together with ep-lock) to have it working,
+ * but having ep-mtx will make the interface more scalable.
* Events that require holding epmutex are very rare, while for
- * normal operations the epoll private ep-sem will guarantee
+ * normal operations the epoll private ep-mtx will guarantee
* a greater scalability.
*/
@@ -106,6 +104,7 @@
#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
+#define EP_UNACTIVE_PTR ((void *) -1L)
struct epoll_filefd {
struct file *file;
@@ -135,6 +134,44 @@
};
/*
+ * Each file descriptor added to the eventpoll interface will
+ * have an entry of this type linked to the rbr RB tree.
+ */
+struct epitem {
+ /* RB-Tree node used to link this structure to the eventpoll rb-tree */
+ struct rb_node rbn;
+
+ /* List header used to link this structure to the eventpoll ready list
*/
+ struct list_head rdllink;
+
+ /* The file descriptor information this item refers to */
+ struct epoll_filefd ffd;
+
+ /* Number of active wait queue attached to poll operations */
+ int nwait;
+
+ /* List containing poll wait queues */
+ struct list_head pwqlist;
+
+ /* The container of this item */
+ struct eventpoll *ep;
+
+ /* The structure that describe the interested events and the source fd
*/
+ struct epoll_event event;
+
+ /*
+* Used to keep track of the usage count of the structure. This avoids
+* that the structure will desappear from underneath our processing.
+*/
+ atomic_t usecnt;
+
+ /* List header used to link this item to the struct file items list */
+ struct list_head fllink;
+
+ struct epitem *next;
+};
+
+/*
* This structure is stored inside the private_data member of the file
* structure and rapresent the main data sructure for the eventpoll
* interface.
@@ -144,12 +181,10 @@
rwlock_t lock;
/*
-* This semaphore is used to ensure that files are not removed
-* while epoll is using them. This is read-held during the event
-* collection
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
--- Davide Libenzi wrote: Charles, would you mind trying the patch below against -git13 on your machine. I tested it with wine and firefox on a 32 bit P4 with HT and it's working fine. i applied the patch against git13. starcraft, pokerstars, and firefox under wine have not frozen. looks good. Never miss an email again! Yahoo! Toolbar alerts you the instant new Mail arrives. http://tools.search.yahoo.com/toolbar/features/mail/ - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Charles Gagalac wrote: > > --- Davide Libenzi <[EMAIL PROTECTED]> wrote: > > > Thank you Charles. Would you mind giving the patch > > below a try? > > > > i applied the patch and i'm still encountering > freezing problems with starcraft, pokerstars, and > firefox under wine. Ok, thanks for testing. I've installing wine and firefox on a 32 bit box and will test by myself ... - Davide - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
--- Davide Libenzi <[EMAIL PROTECTED]> wrote: > Thank you Charles. Would you mind giving the patch > below a try? > i applied the patch and i'm still encountering freezing problems with starcraft, pokerstars, and firefox under wine. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Charles Gagalac wrote: > i just installed the windows version of firefox > 2.0.0.3 under wine and firefox also froze after a few > minutes of use. > > also, wine version is wine-0.9.36. Thank you Charles. Would you mind giving the patch below a try? - Davide Index: linux-2.6.21/fs/eventpoll.c === --- linux-2.6.21.orig/fs/eventpoll.c2007-05-10 11:47:10.0 -0700 +++ linux-2.6.21/fs/eventpoll.c 2007-05-10 11:49:47.0 -0700 @@ -1410,13 +1410,14 @@ * the ep_poll_callback() can requeue the item again, * but we don't care since we are already past it. */ + epi->rdllink.prev = >rdllink; smp_mb(); - INIT_LIST_HEAD(>rdllink); + epi->rdllink.next = >rdllink; } } error = 0; - errxit: +errxit: /* * If the re-injection list or the txlist are not empty, re-splice - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
> Is it possible to have more information? Like CPU, processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 3 model name : Intel(R) Pentium(R) 4 CPU 3.00GHz stepping: 4 cpu MHz : 3000.000 cache size : 1024 KB physical id : 0 siblings: 2 core id : 0 cpu cores : 1 fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 3 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe constant_tsc pni monitor ds_cpl cid xtpr bogomips: 5989.96 clflush size: 64 > kernel 32/64, userspace > 32/64 ...? kernel is 32 userspace is 32 > Does the problem happens with some other publicly > available software that > runs under wine, so that I can test it? > i just installed the windows version of firefox 2.0.0.3 under wine and firefox also froze after a few minutes of use. also, wine version is wine-0.9.36. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Wed, 9 May 2007, Andrew Morton wrote: > On Wed, 9 May 2007 19:37:05 -0700 [EMAIL PROTECTED] wrote: > > > http://bugzilla.kernel.org/show_bug.cgi?id=8462 Is it possible to have more information? Like CPU, kernel 32/64, userspace 32/64 ...? I've spotted something here: + smp_mb(); + INIT_LIST_HEAD(>rdllink); The epoll callback uses list_empty() as test if to queue the item as ready, and list_empty() checks ->next. INIT_LIST_HEAD() sets ->next before, then ->prev. Whoops :) This should be: epi->rdllink.prev = >rdllink; smp_mb(); epi->rdllink.next = >rdllink; Does the problem happens with some other publicly available software that runs under wine, so that I can test it? - Davide - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Wed, 9 May 2007, Andrew Morton wrote: On Wed, 9 May 2007 19:37:05 -0700 [EMAIL PROTECTED] wrote: http://bugzilla.kernel.org/show_bug.cgi?id=8462 Is it possible to have more information? Like CPU, kernel 32/64, userspace 32/64 ...? I've spotted something here: + smp_mb(); + INIT_LIST_HEAD(epi-rdllink); The epoll callback uses list_empty() as test if to queue the item as ready, and list_empty() checks -next. INIT_LIST_HEAD() sets -next before, then -prev. Whoops :) This should be: epi-rdllink.prev = epi-rdllink; smp_mb(); epi-rdllink.next = epi-rdllink; Does the problem happens with some other publicly available software that runs under wine, so that I can test it? - Davide - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
Is it possible to have more information? Like CPU, processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 3 model name : Intel(R) Pentium(R) 4 CPU 3.00GHz stepping: 4 cpu MHz : 3000.000 cache size : 1024 KB physical id : 0 siblings: 2 core id : 0 cpu cores : 1 fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 3 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe constant_tsc pni monitor ds_cpl cid xtpr bogomips: 5989.96 clflush size: 64 kernel 32/64, userspace 32/64 ...? kernel is 32 userspace is 32 Does the problem happens with some other publicly available software that runs under wine, so that I can test it? i just installed the windows version of firefox 2.0.0.3 under wine and firefox also froze after a few minutes of use. also, wine version is wine-0.9.36. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Charles Gagalac wrote: i just installed the windows version of firefox 2.0.0.3 under wine and firefox also froze after a few minutes of use. also, wine version is wine-0.9.36. Thank you Charles. Would you mind giving the patch below a try? - Davide Index: linux-2.6.21/fs/eventpoll.c === --- linux-2.6.21.orig/fs/eventpoll.c2007-05-10 11:47:10.0 -0700 +++ linux-2.6.21/fs/eventpoll.c 2007-05-10 11:49:47.0 -0700 @@ -1410,13 +1410,14 @@ * the ep_poll_callback() can requeue the item again, * but we don't care since we are already past it. */ + epi-rdllink.prev = epi-rdllink; smp_mb(); - INIT_LIST_HEAD(epi-rdllink); + epi-rdllink.next = epi-rdllink; } } error = 0; - errxit: +errxit: /* * If the re-injection list or the txlist are not empty, re-splice - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
--- Davide Libenzi [EMAIL PROTECTED] wrote: Thank you Charles. Would you mind giving the patch below a try? i applied the patch and i'm still encountering freezing problems with starcraft, pokerstars, and firefox under wine. __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Thu, 10 May 2007, Charles Gagalac wrote: --- Davide Libenzi [EMAIL PROTECTED] wrote: Thank you Charles. Would you mind giving the patch below a try? i applied the patch and i'm still encountering freezing problems with starcraft, pokerstars, and firefox under wine. Ok, thanks for testing. I've installing wine and firefox on a 32 bit box and will test by myself ... - Davide - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Wed, 9 May 2007 19:37:05 -0700 [EMAIL PROTECTED] wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=8462 Let's do this via emailed reply-to-all, please. bugzilla is more appropriate for tracking of longer-term bugs. >Summary: applications under wine freezes > Kernel Version: 2.6.21-git > Status: NEW > Severity: normal > Owner: [EMAIL PROTECTED] > Submitter: [EMAIL PROTECTED] > > > Most recent commit where this bug did *NOT* occur: > commit a989705c4cf6e6c1a339c95f9daf658b4ba88ca8 > Merge: 2d56d3c... d291825... > Author: Linus Torvalds <[EMAIL PROTECTED]> > Date: Mon May 7 12:34:57 2007 -0700 > > Distribution: slackware 11.0 > > Problem Description: > applications, namely starcraft and pokerstars, running under wine freezes > shortly after starting the app. > > bisect produced the following commit: > commit 6192bd536f96c6a0d969081bc71ae24f9319bfdc > Author: Davide Libenzi <[EMAIL PROTECTED]> > Date: Tue May 8 00:25:41 2007 -0700 Thanks for doing the bisection. That's "epoll: optimizations and cleanups". - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [Bugme-new] [Bug 8462] New: applications under wine freezes
On Wed, 9 May 2007 19:37:05 -0700 [EMAIL PROTECTED] wrote: http://bugzilla.kernel.org/show_bug.cgi?id=8462 Let's do this via emailed reply-to-all, please. bugzilla is more appropriate for tracking of longer-term bugs. Summary: applications under wine freezes Kernel Version: 2.6.21-git Status: NEW Severity: normal Owner: [EMAIL PROTECTED] Submitter: [EMAIL PROTECTED] Most recent commit where this bug did *NOT* occur: commit a989705c4cf6e6c1a339c95f9daf658b4ba88ca8 Merge: 2d56d3c... d291825... Author: Linus Torvalds [EMAIL PROTECTED] Date: Mon May 7 12:34:57 2007 -0700 Distribution: slackware 11.0 Problem Description: applications, namely starcraft and pokerstars, running under wine freezes shortly after starting the app. bisect produced the following commit: commit 6192bd536f96c6a0d969081bc71ae24f9319bfdc Author: Davide Libenzi [EMAIL PROTECTED] Date: Tue May 8 00:25:41 2007 -0700 Thanks for doing the bisection. That's epoll: optimizations and cleanups. - To unsubscribe from this list: send the line unsubscribe linux-kernel in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
