Re: simple-mtpfs panics
On 21/01/19(Mon) 21:18, Ed Ahlsen-Girard wrote: > On Mon, 21 Jan 2019 21:32:55 -0200 > Martin Pieuchot wrote: > > > On 21/01/19(Mon) 09:36, Ed Ahlsen-Girard wrote: > > > After this command sequence while a Galaxy S8 running > > > R16NW.G950USQU5CSA4 is plugged in: > > > > Does it also happen when you unplug your phone w/o using mtp-connect? > > No, the phone unplugs with no effect. In that case could you try disabling umodem(4) at boot? I'd like to know if this is due to ugen(4) being open by mtp-connect or/and if it has to do with having two drivers attached to a single USB device. So, boot with -c, disable umodem, quit. Then save the dmesg, your device should only attach at ugen(4). If that's not the case, no need to continue. Then try to repeat the panic and look for detach messages. Thanks, Martin
Re: simple-mtpfs panics
On 21/01/19(Mon) 09:36, Ed Ahlsen-Girard wrote: > After this command sequence while a Galaxy S8 running > R16NW.G950USQU5CSA4 is plugged in: Does it also happen when you unplug your phone w/o using mtp-connect? How does your phone appear in the dmesg? Do you see any 'detach' message before the panic? Could you include them in your report? > > doas mtp-connect > doas simple-mtpfs /mnt > > I eventually end up here: > > login: panic: free: size too small 16 <= 64 / 2 (0x80da7ac0) > type USB Stopped at db_enter+0x12: popq %r11 > TIDPIDUID PRFLAGS PFLAGS CPU COMMAND > 154398 12758 350x12 01 Xorg > *409957 13277 0 0x14000 0x2000K usbtask > db_enter() at db_enter+0x12 > panic() at panic+0x120 > free(cc5dea5c80df9709,80e2c000,80dfbe00) at free+0x3cf > usbd_free_device(0) at usb_free_device+0xf6 > usbd_detach(268907f288667fb8,80904500) at usbd_detach+0x81 > uhub_port_connect(b1a3e43ba98bd4a1,80904500,19,6) at > uhub_port_connect+0x66 uhub_explore(280398b005991987) at > uhub_explore+0x208 uhub_explore(280398b00585e61e) at uhub_explore+0x24e > usb_explore(89e789df119a3f2) at usb_explore+0x87 > usb_task_thread(0) at usb_task_thread+0x100 > end trace frame: 0x0, count: 5 > > Sometimes I have useful connectivity for a while, possibly > associated with the state of the snapshot being run. I did not have > that today. Eventually, though, it always ends with something > like this. The keyboard is unresponsive at this point, so I have run no > debugger commands. dmesg below signature. > > > My dmesg: > > > -- > > Edward Ahlsen-Girard > Ft Walton Beach, FL > > > OpenBSD 6.4-current (GENERIC.MP) #623: Fri Jan 18 22:57:22 MST 2019 > dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP > real mem = 4176125952 (3982MB) > avail mem = 4039950336 (3852MB) > mpath0 at root > scsibus0 at mpath0: 256 targets > mainbus0 at root > bios0 at mainbus0: SMBIOS rev. 2.8 @ 0xec530 (36 entries) > bios0: vendor AMI version "80.06" date 04/01/2015 > bios0: Hewlett-Packard 550-036 > acpi0 at bios0: rev 2 > acpi0: sleep states S0 S3 S4 S5 > acpi0: tables DSDT FACP APIC FPDT FIDT MSDM SSDT SSDT MCFG HPET SSDT > SSDT DBGP acpi0: wakeup devices PXSX(S4) RP01(S4) PXSX(S4) PXSX(S4) > PXSX(S4) RP04(S4) PXSX(S4) PXSX(S4) RP06(S4) PXSX(S4) RP07(S4) PXSX(S4) > GLAN(S4) EHC1(S3) EHC2(S3) XHC_(S3) [...] acpitimer0 at acpi0: 3579545 > Hz, 24 bits acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at > mainbus0: apid 0 (boot processor) cpu0: Intel(R) Core(TM) i3-4170 CPU @ > 3.70GHz, 3691.98 MHz, 06-3c-03 cpu0: > FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,SDBG,FMA3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,MOVBE,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,PAGE1GB,RDTSCP,LONG,LAHF,ABM,PERF,ITSC,FSGSBASE,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,IBRS,IBPB,STIBP,L1DF,SSBD,SENSOR,ARAT,XSAVEOPT,MELTDOWN > cpu0: 256KB 64b/line 8-way L2 cache cpu0: smt 0, core 0, package 0 > mtrr: Pentium Pro MTRR support, 10 var ranges, 88 fixed ranges > cpu0: apic clock running at 99MHz > cpu0: mwait min=64, max=64, C-substates=0.2.1.2.4, IBE > cpu1 at mainbus0: apid 2 (application processor) > cpu1: Intel(R) Core(TM) i3-4170 CPU @ 3.70GHz, 3691.46 MHz, 06-3c-03 > cpu1: > FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,SDBG,FMA3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,MOVBE,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,PAGE1GB,RDTSCP,LONG,LAHF,ABM,PERF,ITSC,FSGSBASE,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,IBRS,IBPB,STIBP,L1DF,SSBD,SENSOR,ARAT,XSAVEOPT,MELTDOWN > cpu1: 256KB 64b/line 8-way L2 cache cpu1: smt 0, core 1, package 0 > cpu2 at mainbus0: apid 1 (application processor) > cpu2: Intel(R) Core(TM) i3-4170 CPU @ 3.70GHz, 3691.45 MHz, 06-3c-03 > cpu2: > FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,SDBG,FMA3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,MOVBE,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,PAGE1GB,RDTSCP,LONG,LAHF,ABM,PERF,ITSC,FSGSBASE,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,IBRS,IBPB,STIBP,L1DF,SSBD,SENSOR,ARAT,XSAVEOPT,MELTDOWN > cpu2: 256KB 64b/line 8-way L2 cache cpu2: smt 1, core 0, package 0 > cpu3 at mainbus0: apid 3 (application processor) > cpu3: Intel(R) Core(TM) i3-4170 CPU @ 3.70GHz, 3691.45 MHz, 06-3c-03 > cpu3: > FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,SDBG,FMA3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,MOVBE,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,PAGE1GB,RDTSCP,LONG,LAHF,ABM,PERF,ITSC,FSGSBASE,BMI1,AVX2,SMEP,BMI2,ERMS,INVPCID,IBRS,IBPB,STIBP,L
Re: [PATCH] ntfs-3g needs OpenBSD specific groupmember() function
On 04/06/18(Mon) 14:26, Helg wrote:
> Hi Ports,
>
> I have an upcoming patch to FUSE that passes the current process tid,
> uid, gid and umask to the file system. This has highlighted a bug in the
> port where the groupmember() function in libntfs-3g/security.c assumes
> it's runing on Linux where thread information is available in /proc.
>
> This diff adds an OpenBSD specific implementation of this function.
We should refrain linking to libkvm. In this particular case you should
be able to call the KERN_PROC sysctl(2) directly.
> Index: Makefile
> ===
> RCS file: /cvs/ports/sysutils/ntfs-3g/Makefile,v
> retrieving revision 1.9
> diff -u -p -r1.9 Makefile
> --- Makefile 22 Jun 2016 09:51:33 - 1.9
> +++ Makefile 4 Jun 2018 06:20:59 -
> @@ -15,7 +15,7 @@ MAINTAINER =Paul Irofti # GPLv2
> PERMIT_PACKAGE_CDROM = Yes
>
> -WANTLIB += c uuid fuse
> +WANTLIB += c uuid fuse kvm
>
> MASTER_SITES = ${HOMEPAGE}/opensource/
> EXTRACT_SUFX = .tgz
> @@ -24,7 +24,7 @@ LIB_DEPENDS = sysutils/e2fsprogs
>
> CONFIGURE_STYLE =gnu
> CONFIGURE_ENV = CPPFLAGS="-I${LOCALBASE}/include" \
> - LDFLAGS="-L${LOCALBASE}/lib"
> + LDFLAGS="-L${LOCALBASE}/lib -lkvm"
> CONFIGURE_ARGS = --disable-ldconfig \
> --disable-mtab \
> --enable-extras \
> Index: patches/patch-libntfs-3g_security_c
> ===
> RCS file: patches/patch-libntfs-3g_security_c
> diff -N patches/patch-libntfs-3g_security_c
> --- /dev/null 1 Jan 1970 00:00:00 -
> +++ patches/patch-libntfs-3g_security_c 4 Jun 2018 06:20:59 -
> @@ -0,0 +1,78 @@
> +$OpenBSD$
> +
> +Index: libntfs-3g/security.c
> +--- libntfs-3g/security.c.orig
> libntfs-3g/security.c
> +@@ -47,6 +47,11 @@
> + #ifdef HAVE_SYS_STAT_H
> + #include
> + #endif
> ++#ifdef __OpenBSD__
> ++#include
> ++#include
> ++#include
> ++#endif
> +
> + #include
> + #include
> +@@ -1228,6 +1233,60 @@ static BOOL groupmember(struct SECURITY_CONTEXT *scx,
> + close(fd);
> + }
> + }
> ++return (ismember);
> ++}
> ++
> ++#elif defined(__OpenBSD__)
> ++
> ++static BOOL groupmember(struct SECURITY_CONTEXT *scx, uid_t uid, gid_t gid)
> ++{
> ++kvm_t *kd;
> ++int k;
> ++int pcnt;
> ++gid_t *p;
> ++BOOL ismember;
> ++pid_t tid;
> ++static char errbuf[_POSIX2_LINE_MAX];
> ++struct kinfo_proc *kp;
> ++
> ++if (scx->vol->secure_flags & (1 << SECURITY_STATICGRPS))
> ++ismember = staticgroupmember(scx, uid, gid);
> ++else {
> ++ismember = FALSE; /* default return */
> ++tid = scx->tid;
> ++kd = kvm_openfiles(NULL, NULL, NULL, KVM_NO_FILES, errbuf);
> ++if (kd == NULL)
> ++ntfs_log_error("%s\n", errbuf);
> ++else {
> ++kp = kvm_getprocs(kd, KERN_PROC_ALL |
> ++KERN_PROC_SHOW_THREADS, NULL, sizeof(*kp), &pcnt);
> ++if (kp == NULL)
> ++ntfs_log_error("%s\n", kvm_geterr(kd));
> ++else if (pcnt > 0) {
> ++k = 0;
> ++while ((kp->p_tid != tid) && (k < pcnt)) {
> ++k++;
> ++kp++;
> ++}
> ++if (k < pcnt) {
> ++if (kp->p_gid == gid)
> ++ismember = TRUE;
> ++p = kp->p_groups;
> ++k = 0;
> ++while (!ismember
> ++&& (k < kp->p_ngroups)
> ++&& (*p != gid)) {
> ++k++;
> ++p++;
> ++}
> ++if (k < kp->p_ngroups)
> ++ismember = TRUE;
> ++}
> ++}
> ++kvm_close(kd);
> ++}
> ++}
> ++
> + return (ismember);
> + }
> +
>
Re: glib2: gio kqueue fixes
On 13/01/18(Sat) 16:13, Martin Pieuchot wrote:
> Diff below is a redesign of the kqueue(2) backend for GIO/glib.
> It fixes BZ #739424 and a couple of more bugs and races & reduces
> the size of the backend by 1K+ lines.
>
> The backend is still not generating all the events it should, but
> that's a different problem. More details about my work can be found
> in the git commit below.
>
> I'd appreciate test reports before I push that upstream.
Anyone?
> Index: Makefile
> ===
> RCS file: /cvs/ports/devel/glib2/Makefile,v
> retrieving revision 1.268
> diff -u -p -r1.268 Makefile
> --- Makefile 9 Jan 2018 12:31:54 - 1.268
> +++ Makefile 13 Jan 2018 14:55:24 -
> @@ -9,6 +9,7 @@ COMMENT= general-purpose utility librar
> GNOME_PROJECT= glib
> GNOME_VERSION= 2.54.3
> PKGNAME= ${DISTNAME:S/glib/glib2/}
> +REVISION=0
>
> CATEGORIES= devel
>
> @@ -54,6 +55,12 @@ CONFIGURE_ARGS=--enable-static \
> --with-pcre=system \
> --disable-fam
>
> +AUTOCONF_VERSION=2.69
> +AUTOMAKE_VERSION=1.15
> +BUILD_DEPENDS += ${MODGNU_AUTOCONF_DEPENDS} \
> + ${MODGNU_AUTOMAKE_DEPENDS} \
> + devel/libtool
> +
> .include
> .if ${PROPERTIES:Mgcc3}
> PATCH_LIST= patch-* gcc3-*
> @@ -71,5 +78,9 @@ pre-configure:
> sed -i "s,/usr/share,${PREFIX}/share,g" \
> ${WRKSRC}/docs/reference/gio/html/glib-compile-schemas.html \
> ${WRKSRC}/docs/reference/gio/glib-compile-schemas.xml
> + cd ${WRKSRC} && env -i \
> + AUTOCONF_VERSION=${AUTOCONF_VERSION} \
> + AUTOMAKE_VERSION=${AUTOMAKE_VERSION} \
> + autoreconf -fi
>
> .include
> Index: patches/patch-gio_kqueue_gkqueuefilemonitor_c
> ===
> RCS file: patches/patch-gio_kqueue_gkqueuefilemonitor_c
> diff -N patches/patch-gio_kqueue_gkqueuefilemonitor_c
> --- patches/patch-gio_kqueue_gkqueuefilemonitor_c 8 May 2017 16:57:08
> - 1.2
> +++ /dev/null 1 Jan 1970 00:00:00 -
> @@ -1,50 +0,0 @@
> -$OpenBSD: patch-gio_kqueue_gkqueuefilemonitor_c,v 1.2 2017/05/08 16:57:08
> ajacoutot Exp $
> -
> -Prevent a use-after-free resulting in a crash of all apps depending
> -on glib/GIO mostly at "pkg_add -u" time.
> -https://bugzilla.gnome.org/show_bug.cgi?id=739424
> -
> -Index: gio/kqueue/gkqueuefilemonitor.c
> gio/kqueue/gkqueuefilemonitor.c.orig
> -+++ gio/kqueue/gkqueuefilemonitor.c
> -@@ -30,6 +30,16 @@
> - #include
> -
> -
> -+/*
> -+ * Because `kqueue_sub'' are not refcounted, we need
> -+ * ensure no other thread is getting a reference to
> -+ * the element we want to free.
> -+ *
> -+ * That's why _kh_cancel_sub() must be called with
> -+ * this lock held to prevent a race.
> -+ */
> -+G_LOCK_EXTERN (hash_lock);
> -+
> - struct _GKqueueFileMonitor
> - {
> - GLocalFileMonitor parent_instance;
> -@@ -80,9 +90,11 @@ g_kqueue_file_monitor_finalize (GObject *object)
> -
> - if (kqueue_monitor->sub)
> - {
> -+ G_LOCK (hash_lock);
> - _kh_cancel_sub (kqueue_monitor->sub);
> - _kh_sub_free (kqueue_monitor->sub);
> - kqueue_monitor->sub = NULL;
> -+ G_UNLOCK (hash_lock);
> - }
> -
> - if (kqueue_monitor->fallback)
> -@@ -181,9 +193,11 @@ g_kqueue_file_monitor_cancel (GFileMonitor *monitor)
> -
> - if (kqueue_monitor->sub)
> - {
> -+ G_LOCK (hash_lock);
> - _kh_cancel_sub (kqueue_monitor->sub);
> - _kh_sub_free (kqueue_monitor->sub);
> - kqueue_monitor->sub = NULL;
> -+ G_UNLOCK (hash_lock);
> - }
> - else if (kqueue_monitor->fallback)
> - {
> Index: patches/patch-gio_kqueue_kqueue-helper_c
> ===
> RCS file: patches/patch-gio_kqueue_kqueue-helper_c
> diff -N patches/patch-gio_kqueue_kqueue-helper_c
> --- patches/patch-gio_kqueue_kqueue-helper_c 9 Jan 2018 12:31:54 -
> 1.4
> +++ /dev/null 1 Jan 1970 00:00:00 -
> @@ -1,64 +0,0 @@
> -$OpenBSD: patch-gio_kqueue_kqueue-helper_c,v 1.4 2018/01/09 12:31:54
> ajacoutot Exp $
> -
> -Prevent a use-after-free resulting in a crash of all apps depending
> -on glib/GIO mostly at "pkg_add -u" time.
> -https://bugzilla.gnome.org/show_bug.cgi?id=739424
> -
> -Index: gio/kqueue/kqueue-helper.c
> gio/kqueue/kqueue
glib2: gio kqueue fixes
KH_W ("Got a notification for a deleted or non-existing subscription
%d",
- n.fd);
- return TRUE;
-@@ -338,6 +338,7 @@ process_kqueue_notifications (GIOChannel *gioc,
- g_file_monitor_source_handle_event (source, mask, NULL, NULL, NULL,
g_get_monotonic_time ());
- }
-
-+ G_UNLOCK (hash_lock);
- return TRUE;
- }
-
-@@ -453,13 +454,14 @@ _kh_start_watching (kqueue_sub *sub)
-
- G_LOCK (hash_lock);
- g_hash_table_insert (subs_hash_table, GINT_TO_POINTER (sub->fd), sub);
-- G_UNLOCK (hash_lock);
-
- _kqueue_thread_push_fd (sub->fd);
-
- /* Bump the kqueue thread. It will pick up a new sub entry to monitor */
- if (!_ku_write (kqueue_socket_pair[0], "A", 1))
- KH_W ("Failed to bump the kqueue thread (add fd, error %d)", errno);
-+ G_UNLOCK (hash_lock);
-+
- return TRUE;
- }
-
-@@ -506,9 +508,7 @@ _kh_cancel_sub (kqueue_sub *sub)
-
- _km_remove (sub);
-
-- G_LOCK (hash_lock);
- removed = g_hash_table_remove (subs_hash_table, GINT_TO_POINTER (sub->fd));
-- G_UNLOCK (hash_lock);
-
- if (removed)
- {
Index: patches/patch-gio_tests_file_c
===
RCS file: /cvs/ports/devel/glib2/patches/patch-gio_tests_file_c,v
retrieving revision 1.7
diff -u -p -r1.7 patch-gio_tests_file_c
--- patches/patch-gio_tests_file_c 24 Sep 2015 10:23:42 - 1.7
+++ patches/patch-gio_tests_file_c 13 Jan 2018 14:59:18 -
@@ -1,7 +1,6 @@
$OpenBSD: patch-gio_tests_file_c,v 1.7 2015/09/24 10:23:42 ajacoutot Exp $
-XXX why isn't this working with kqueue?
-push upstream when fixed
+The current kqueue backend does not monitor non-existing file correctly.
--- gio/tests/file.c.orig Wed Aug 19 05:35:30 2015
+++ gio/tests/file.c Wed Sep 23 09:25:35 2015
Index: patches/patch-kqueue_fix
===
RCS file: patches/patch-kqueue_fix
diff -N patches/patch-kqueue_fix
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-kqueue_fix13 Jan 2018 14:59:47 -
@@ -0,0 +1,2083 @@
+commit 59fcc304e9f29467e0cc68c4ec6d4b3c0d0fcd59
+Author: Martin Pieuchot
+Date: Sat Jan 13 15:25:30 2018 +0100
+
+kqueue: Multiple fixes and simplifications.
+
+- Stop using a custom thread for listening to kqueue(2) events. Instead
+call kevent(2) in non blocking mode in a monitor callback. Under the
+hood poll(2) is used to figure out if new events are available.
+
+- Do not use a socketpair with a custom protocol requiring 2 supplementary
+context switches per event to commicate between multiple threads. Calling
+kevent(2), in non blocking mode, to add/remove events is fine from any
+context.
+
+- Add kqueue(2) events without the EV_ONESHOT flag. This removes a race
+were some notifications were lost because events had to be re-added for
+every new notification.
+
+- Get rid of the global hash table and its associated lock and races. Use
+the 'cookie' argument of kevent(2) to pass the associated descriptor when
+registering an event.
+
+- Fix _kh_file_appeared_cb() by properly passing a monitor instead of a
+source to g_file_monitor_emit_event().
+
+- Porperly refcount sources.
+
+- Remove a lot of abstraction making it harder to fix the remaining issues.
+
+This diff does not improve the correctness of the generate events, a lot
+can still be done.
+
+Fix https://bugzilla.gnome.org/show_bug.cgi?id=739424
+
+diff --git gio/kqueue/Makefile.am gio/kqueue/Makefile.am
+index d5657d7e4..24e9724e5 100644
+--- gio/kqueue/Makefile.am
gio/kqueue/Makefile.am
+@@ -4,19 +4,9 @@ noinst_LTLIBRARIES += libkqueue.la
+
+ libkqueue_la_SOURCES = \
+gkqueuefilemonitor.c \
+- gkqueuefilemonitor.h \
+kqueue-helper.c \
+kqueue-helper.h \
+- kqueue-thread.c \
+- kqueue-thread.h \
+- kqueue-sub.c \
+- kqueue-sub.h \
+kqueue-missing.c \
+- kqueue-missing.h \
+- kqueue-utils.c \
+- kqueue-utils.h \
+- kqueue-exclusions.c \
+- kqueue-exclusions.h \
+dep-list.c \
+dep-list.h \
+$(NULL)
+diff --git gio/kqueue/gkqueuefilemonitor.c gio/kqueue/gkqueuefilemonitor.c
+index 78b749637..a9107b0a9 100644
+--- gio/kqueue/gkqueuefilemonitor.c
gio/kqueue/gkqueuefilemonitor.c
+@@ -22,33 +22,72 @@
+
+ #include "config.h"
+
+-#include "gkqueuefilemonitor.h"
+-#include "kqueue-helper.h"
+-#include "kqueue-exclusions.h"
++#include
++#include
++#include
++#include
++#include
++
++#include
++#include
++#include
++
++#include
++#include
++#include
++#include
+ #include
+ #include
+-#include
++#include
++#include "glib-private.h"
++
++#include "kqueue-helper.h"
++#include "dep-list.h"
++
++st
Fix audio/tagtool
This port is currently broken. At run time no button work and the app
spits a lot of the following warnings:
libglade-WARNING **: could not find signal handler 'id3_info'.
Adding "-Wl,--export-dynamic" to the linker fixes the problem and I can
finally tag my mp3.
ok?
Index: Makefile
===
RCS file: /cvs/ports/audio/tagtool/Makefile,v
retrieving revision 1.34
diff -u -p -r1.34 Makefile
--- Makefile21 Nov 2017 00:12:59 - 1.34
+++ Makefile8 Dec 2017 10:40:42 -
@@ -3,7 +3,7 @@
COMMENT= Ogg Vorbis and MP3 files tagger
DISTNAME= tagtool-0.12.3
-REVISION= 20
+REVISION= 21
CATEGORIES=audio
MAINTAINER=Vlad Glagolev
@@ -36,7 +36,7 @@ RUN_DEPENDS= devel/desktop-file-utils \
CONFIGURE_STYLE=gnu
CONFIGURE_ENV= CPPFLAGS="-I${LOCALBASE}/include" \
- LDFLAGS="-L${LOCALBASE}/lib -lm -logg"
+ LDFLAGS="-L${LOCALBASE}/lib -lm -logg -Wl,--export-dynamic"
USE_GMAKE= Yes
Fix dhcpcd(8) breakage
Conditionally issues ioctl(2)s the kernel no longer supports. There's no need to work around the kernel expiring prefixes and routers anymore on OpenBSD. This fix a breakage introduced by my recent removal of such ioctl(2), reported by aja@. ok? Index: patches/patch-if-bsd_c === RCS file: /cvs/ports/net/dhcpcd/patches/patch-if-bsd_c,v retrieving revision 1.2 diff -u -p -r1.2 patch-if-bsd_c --- patches/patch-if-bsd_c 10 Aug 2017 06:58:39 - 1.2 +++ patches/patch-if-bsd_c 27 Oct 2017 08:56:30 - @@ -5,6 +5,10 @@ Fixed with netinet6/in6.c r1.202. netinet6/nd6.h r1.72 removed ND6_IFF_PERFORMNUD +Also conditionnaly flush prefix and router lists, OpenBSD no longer +keep track of them in kernel. So there's no risk for dhcpcd(8) to +have its advertised routers and prefixes expired. + Index: if-bsd.c --- if-bsd.c.orig +++ if-bsd.c @@ -33,3 +37,20 @@ Index: if-bsd.c static int if_raflush(int s) +@@ -1720,9 +1722,14 @@ if_raflush(int s) + char dummy[IFNAMSIZ + 8]; + + strlcpy(dummy, "lo0", sizeof(dummy)); +- if (ioctl(s, SIOCSRTRFLUSH_IN6, (void *)&dummy) == -1 || +- ioctl(s, SIOCSPFXFLUSH_IN6, (void *)&dummy) == -1) ++#ifdef SIOCSRTRFLUSH_IN6 ++ if (ioctl(s, SIOCSRTRFLUSH_IN6, (void *)&dummy) == -1) + return -1; ++#endif ++#ifdef SIOCSPFXFLUSH_IN6 ++ if (ioctl(s, SIOCSPFXFLUSH_IN6, (void *)&dummy) == -1) ++ return -1; ++#endif + return 0; + } +
Re: Port bulk with
On 05/10/17(Thu) 16:02, Christian Weisgerber wrote: > Martin Pieuchot: > > > > On amd64, two ports failed to build: devel/libdwarf and devel/valgrind. > > > > Thanks, here's an updated diff that should fix those. > > Nope, those two still fail. Full logs attached. I added yesterday the missing defines to . While doing so I figured that devel/libelf will pick if it is present. So I'd like to bump it when I introduce this header. This will make it easier to deal with possible fallout. Full diff below, ok? Index: src/include/Makefile === RCS file: /cvs/src/include/Makefile,v retrieving revision 1.220 diff -u -p -r1.220 Makefile --- src/include/Makefile6 Oct 2017 19:58:37 - 1.220 +++ src/include/Makefile9 Oct 2017 07:54:16 - @@ -10,9 +10,10 @@ .include FILES= a.out.h ar.h asr.h assert.h bitstring.h blf.h bsd_auth.h \ - complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h \ - dlfcn.h elf_abi.h err.h errno.h fenv.h float.h fnmatch.h fstab.h fts.h \ - ftw.h getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ + complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h dlfcn.h \ + elf.h elf_abi.h err.h errno.h \ + fenv.h float.h fnmatch.h fstab.h fts.h ftw.h \ + getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ iso646.h kvm.h langinfo.h libgen.h limits.h link.h link_elf.h \ locale.h login_cap.h math.h md5.h memory.h ndbm.h netdb.h netgroup.h \ nlist.h nl_types.h paths.h poll.h pwd.h ranlib.h readpassphrase.h \ Index: src/include/elf.h === RCS file: src/include/elf.h diff -N src/include/elf.h --- /dev/null 1 Jan 1970 00:00:00 - +++ src/include/elf.h 9 Aug 2017 15:49:08 - @@ -0,0 +1,12 @@ +/* $OpenBSD$ */ + +/* + * Public domain. + */ + +#ifndef _ELF_H_ +#define _ELF_H_ + +#include + +#endif /* _ELF_H_ */ Index: ports/devel/libelf/Makefile === RCS file: /cvs/ports/devel/libelf/Makefile,v retrieving revision 1.12 diff -u -p -r1.12 Makefile --- ports/devel/libelf/Makefile 15 Aug 2015 08:03:20 - 1.12 +++ ports/devel/libelf/Makefile 11 Oct 2017 12:07:10 - @@ -6,7 +6,7 @@ PORTROACH_COMMENT=0.8.14 is actually 0.8 COMMENT= read, modify, create ELF files on any arch DISTNAME= libelf-0.8.13 -REVISION= 3 +REVISION= 4 MAINTAINER=Jasper Lievisse Adriaanse SHARED_LIBS= elf 1.1
Re: Port bulk with
On 12/09/17(Tue) 16:23, Christian Weisgerber wrote:
> Martin Pieuchot:
>
> > > So here's a first step, introducing /usr/include/elf.h. Could some of
> > > you run a bulk with it and report the possible breakages?
> >
> > Now that the offending function declaration has been remove from libc
> > libelf builds as before.
> >
> > Could you guys tell us if there's any other fallout from this diff?
>
> On amd64, two ports failed to build: devel/libdwarf and devel/valgrind.
Thanks, here's an updated diff that should fix those.
Index: sys/sys/exec_elf.h
===
RCS file: /cvs/src/sys/sys/exec_elf.h,v
retrieving revision 1.75
diff -u -p -r1.75 exec_elf.h
--- sys/sys/exec_elf.h 5 Sep 2017 06:35:19 - 1.75
+++ sys/sys/exec_elf.h 3 Oct 2017 13:54:57 -
@@ -187,12 +187,14 @@ typedef struct {
#define EM_PARISC 15 /* HPPA */
#define EM_SPARC32PLUS 18 /* Enhanced instruction set SPARC */
#define EM_PPC 20 /* PowerPC */
+#define EM_PPC64 21 /* PowerPC 64 */
#define EM_ARM 40 /* Advanced RISC Machines ARM */
#define EM_ALPHA 41 /* DEC ALPHA */
-#defineEM_SH 42 /* Hitachi/Renesas Super-H */
+#define EM_SH 42 /* Hitachi/Renesas Super-H */
#define EM_SPARCV9 43 /* SPARC version 9 */
#define EM_IA_64 50 /* Intel IA-64 Processor */
#define EM_AMD64 62 /* AMD64 architecture */
+#define EM_X86_64 EM_AMD64
#define EM_VAX 75 /* DEC VAX */
#define EM_AARCH64 183 /* ARM 64-bit architecture (AArch64) */
Index: include/Makefile
===
RCS file: /cvs/src/include/Makefile,v
retrieving revision 1.219
diff -u -p -r1.219 Makefile
--- include/Makefile17 Apr 2017 15:53:21 - 1.219
+++ include/Makefile9 Aug 2017 15:46:46 -
@@ -10,9 +10,10 @@
.include
FILES= a.out.h ar.h asr.h assert.h bitstring.h blf.h bsd_auth.h \
- complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h \
- dlfcn.h elf_abi.h err.h errno.h fenv.h float.h fnmatch.h fstab.h fts.h \
- ftw.h getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \
+ complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h dlfcn.h \
+ elf.h elf_abi.h err.h errno.h \
+ fenv.h float.h fnmatch.h fstab.h fts.h ftw.h \
+ getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \
iso646.h kvm.h langinfo.h libgen.h limits.h link.h link_elf.h \
locale.h login_cap.h math.h md5.h memory.h ndbm.h netdb.h netgroup.h \
nlist.h nl_types.h paths.h poll.h pwd.h ranlib.h readpassphrase.h \
Index: include/elf.h
===
RCS file: include/elf.h
diff -N include/elf.h
--- /dev/null 1 Jan 1970 00:00:00 -
+++ include/elf.h 9 Aug 2017 15:49:08 -
@@ -0,0 +1,12 @@
+/* $OpenBSD$ */
+
+/*
+ * Public domain.
+ */
+
+#ifndef _ELF_H_
+#define _ELF_H_
+
+#include
+
+#endif /* _ELF_H_ */
Re: mednafen 0.9.39.2 -> 0.9.46
On 20/09/17(Wed) 09:13, Anthony J. Bentley wrote:
> Martin Pieuchot writes:
> > Do you know if the games are multi-threaded? Could you run "top -H" and
> > "kdump -H"?
>
> top -H shows a single line for gambatte.
Thanks. Could you try the diff below? It includes some debug stuff and
a potential fix.
I'm guessing that the problem is that my last change made uhid_do_ioctl()
return an error for FIOASYNC. This makes the following syscall fall:
390:/* The poll blocks the event thread. */
391:fcntl(fd, F_SETFL, O_NONBLOCK)
As a result FNONBLOCK is not set on the 'struct file' and uhid_do_read()
block, or "freeze", when there's nothing to read instead of returning
EWOULDBLOCK.
Index: uhid.c
===
RCS file: /cvs/src/sys/dev/usb/uhid.c,v
retrieving revision 1.68
diff -u -p -r1.68 uhid.c
--- uhid.c 20 Jul 2017 16:54:45 - 1.68
+++ uhid.c 22 Sep 2017 12:31:56 -
@@ -59,10 +59,12 @@
#include
+#define UHID_DEBUG
+
#ifdef UHID_DEBUG
#define DPRINTF(x) do { if (uhiddebug) printf x; } while (0)
#define DPRINTFN(n,x) do { if (uhiddebug>(n)) printf x; } while (0)
-intuhiddebug = 0;
+intuhiddebug = 6;
#else
#define DPRINTF(x)
#define DPRINTFN(n,x)
@@ -179,7 +181,7 @@ uhid_intr(struct uhidev *addr, void *dat
struct uhid_softc *sc = (struct uhid_softc *)addr;
#ifdef UHID_DEBUG
- if (uhiddebug > 5) {
+ if (uhiddebug > 6) {
u_int32_t i;
DPRINTF(("uhid_intr: data ="));
@@ -359,6 +361,7 @@ uhid_do_ioctl(struct uhid_softc *sc, u_l
switch (cmd) {
case FIONBIO:
+ case FIOASYNC:
/* All handled in the upper FS layer. */
break;
Re: mednafen 0.9.39.2 -> 0.9.46
On 20/09/17(Wed) 01:02, Anthony J. Bentley wrote:
> Martin Pieuchot writes:
> > On 17/09/17(Sun) 17:05, Bryan Linton wrote:
> > > [ CCing all potentially involved parties, because I'd rather CC
> > > more people than necessary than leave out an interested party.
> > > Please ignore this mail if it's not relevant to you. ]
> > >
> > > [...]
> > >
> > > Reverting the above commit fixes the issue for me. Without the
> > > reversion, SDL games freeze at a black screen if a joypad is
> > > connected. Invoking "kill -9" or unplugging the joypad is the
> > > only way to unfreeze the program.
> >
> > When the program "freeze" what does top says? Where is it sleeping?
>
> PID USERNAME PRI NICE SIZE RES STATE WAIT TIMECPU COMMAND
> 30135 anthony00 2616K 5596K idle uhidrea 0:00 0.00%
> gambatte_sdl
>
> > Could you please ktrace the freezing programme when this happen? What
> > do you see?
>
> Once it's frozen ktrace seems to show no output. Here's the last few
> lines if I run ktrace from the start:
> [...]
> 90169 gambatte_sdl CALL read(5,0x18f94585a804,0x4)
So it seems that the sleeping thread isn't awaken. Bryan Linton
provided some additional information, he said that emulators/mednafen
still work for him.
Do you know if the games are multi-threaded? Could you run "top -H" and
"kdump -H"?
Could you also run a kernel compiled with the following diff and see if
something is printed in the dmesg when the program "freeze"?
diff --git sys/dev/usb/uhid.c sys/dev/usb/uhid.c
index 184f30879e9..97c265a7ec2 100644
--- sys/dev/usb/uhid.c
+++ sys/dev/usb/uhid.c
@@ -59,10 +59,12 @@
#include
+#define UHID_DEBUG
+
#ifdef UHID_DEBUG
#define DPRINTF(x) do { if (uhiddebug) printf x; } while (0)
#define DPRINTFN(n,x) do { if (uhiddebug>(n)) printf x; } while (0)
-intuhiddebug = 0;
+intuhiddebug = 5;
#else
#define DPRINTF(x)
#define DPRINTFN(n,x)
@@ -179,7 +181,7 @@ uhid_intr(struct uhidev *addr, void *data, u_int len)
struct uhid_softc *sc = (struct uhid_softc *)addr;
#ifdef UHID_DEBUG
- if (uhiddebug > 5) {
+ if (uhiddebug > 6) {
u_int32_t i;
DPRINTF(("uhid_intr: data ="));
Re: mednafen 0.9.39.2 -> 0.9.46
On 17/09/17(Sun) 17:05, Bryan Linton wrote: > [ CCing all potentially involved parties, because I'd rather CC > more people than necessary than leave out an interested party. > Please ignore this mail if it's not relevant to you. ] > > [...] > > Reverting the above commit fixes the issue for me. Without the > reversion, SDL games freeze at a black screen if a joypad is > connected. Invoking "kill -9" or unplugging the joypad is the > only way to unfreeze the program. When the program "freeze" what does top says? Where is it sleeping? Could you please ktrace the freezing programme when this happen? What do you see? By looking at SDL sources I cannot spot how this commit can affect it, so it'd help if you could provide me more info.
Re: Port bulk with
On 09/08/17(Wed) 11:58, Martin Pieuchot wrote: > As discussed recently on tech@ with Karel, some ports would benefit > from having a header. This header is the Solaris version of > our and has been adopted by FreeBSD. > > I'd like to migrate our base system from to . This > would make our base tools dealing with ELF more portable. > > So here's a first step, introducing /usr/include/elf.h. Could some of > you run a bulk with it and report the possible breakages? Now that the offending function declaration has been remove from libc libelf builds as before. Could you guys tell us if there's any other fallout from this diff? Index: Makefile === RCS file: /cvs/src/include/Makefile,v retrieving revision 1.219 diff -u -p -r1.219 Makefile --- Makefile17 Apr 2017 15:53:21 - 1.219 +++ Makefile9 Aug 2017 15:46:46 - @@ -10,9 +10,10 @@ .include FILES= a.out.h ar.h asr.h assert.h bitstring.h blf.h bsd_auth.h \ - complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h \ - dlfcn.h elf_abi.h err.h errno.h fenv.h float.h fnmatch.h fstab.h fts.h \ - ftw.h getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ + complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h dlfcn.h \ + elf.h elf_abi.h err.h errno.h \ + fenv.h float.h fnmatch.h fstab.h fts.h ftw.h \ + getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ iso646.h kvm.h langinfo.h libgen.h limits.h link.h link_elf.h \ locale.h login_cap.h math.h md5.h memory.h ndbm.h netdb.h netgroup.h \ nlist.h nl_types.h paths.h poll.h pwd.h ranlib.h readpassphrase.h \ Index: elf.h === RCS file: elf.h diff -N elf.h --- /dev/null 1 Jan 1970 00:00:00 - +++ elf.h 9 Aug 2017 15:49:08 - @@ -0,0 +1,12 @@ +/* $OpenBSD$ */ + +/* + * Public domain. + */ + +#ifndef _ELF_H_ +#define _ELF_H_ + +#include + +#endif /* _ELF_H_ */
Port bulk with
As discussed recently on tech@ with Karel, some ports would benefit from having a header. This header is the Solaris version of our and has been adopted by FreeBSD. I'd like to migrate our base system from to . This would make our base tools dealing with ELF more portable. So here's a first step, introducing /usr/include/elf.h. Could some of you run a bulk with it and report the possible breakages? Thanks, Martin Index: Makefile === RCS file: /cvs/src/include/Makefile,v retrieving revision 1.219 diff -u -p -r1.219 Makefile --- Makefile17 Apr 2017 15:53:21 - 1.219 +++ Makefile9 Aug 2017 15:46:46 - @@ -10,9 +10,10 @@ .include FILES= a.out.h ar.h asr.h assert.h bitstring.h blf.h bsd_auth.h \ - complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h \ - dlfcn.h elf_abi.h err.h errno.h fenv.h float.h fnmatch.h fstab.h fts.h \ - ftw.h getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ + complex.h cpio.h ctype.h curses.h db.h dirent.h disktab.h dlfcn.h \ + elf.h elf_abi.h err.h errno.h \ + fenv.h float.h fnmatch.h fstab.h fts.h ftw.h \ + getopt.h glob.h grp.h icdb.h ieeefp.h ifaddrs.h inttypes.h \ iso646.h kvm.h langinfo.h libgen.h limits.h link.h link_elf.h \ locale.h login_cap.h math.h md5.h memory.h ndbm.h netdb.h netgroup.h \ nlist.h nl_types.h paths.h poll.h pwd.h ranlib.h readpassphrase.h \ Index: elf.h === RCS file: elf.h diff -N elf.h --- /dev/null 1 Jan 1970 00:00:00 - +++ elf.h 9 Aug 2017 15:49:08 - @@ -0,0 +1,12 @@ +/* $OpenBSD$ */ + +/* + * Public domain. + */ + +#ifndef _ELF_H_ +#define _ELF_H_ + +#include + +#endif /* _ELF_H_ */
Re: git 2.13.0 broken on sparc64
On 26/06/17(Mon) 11:56, Landry Breuil wrote: > On Mon, Jun 26, 2017 at 11:51:15AM +0200, Martin Pieuchot wrote: > > No debug symbols, so I can't find/fix the problem. > > > > OpenBSD 6.1-current (GENERIC.MP) #18: Mon Jun 26 11:20:49 CEST 2017 > > m...@oliva.grenadille.net:/sys/arch/sparc64/compile/GENERIC.MP > > > > $ git --version > > git version 2.13.0 > > > > Maybe it's the sha1 thing that was broken in 2.13.0 and got fixed in > 2.13.0p0 and the sparc64 havent catched up yet ? Dunno, I updated my packages before reporting: # pkg_add -u git quirks-2.325 signed on 2017-06-18T10:35:43Z # pkg_info |grep git git-2.13.0p0GIT - Tree History Storage Tool
git 2.13.0 broken on sparc64
No debug symbols, so I can't find/fix the problem. OpenBSD 6.1-current (GENERIC.MP) #18: Mon Jun 26 11:20:49 CEST 2017 m...@oliva.grenadille.net:/sys/arch/sparc64/compile/GENERIC.MP $ git --version git version 2.13.0 $ git clone https://github.com/mpieuchot/ctfconvert ctfconv Cloning into 'ctfconv'... remote: Counting objects: 380, done. error: index-pack died of signal 10 fatal: index-pack failed $ gdb /usr/local/bin/git git.core (gdb) bt #0 0x003eb449bb84 in sha1_compression_states () from /usr/local/bin/git #1 0x003eb449dbc0 in sha1_process () from /usr/local/bin/git #2 0x003eb44a03ac in SHA1DCUpdate () from /usr/local/bin/git #3 0x003eb4341a2c in parse_pack_objects () from /usr/local/bin/git #4 0x003eb43437f4 in cmd_index_pack () from /usr/local/bin/git #5 0x003eb43011f8 in handle_builtin () from /usr/local/bin/git #6 0x003eb43017d0 in cmd_main () from /usr/local/bin/git #7 0x003eb43939d8 in main () from /usr/local/bin/git
Re: deadbeef/libsndiod NULL dereference
On 29/12/16(Thu) 16:55, Martin Pieuchot wrote:
> I am running GNOME3 and my music player is audio/deadbeef. Since the
> last package update very time my screen saver starts the music stop.
> Previously it was only the first time per session.
>
> Here's my last output of /var/log/messages. The screen saver started
> at 16:35, 16:42 and 16:49.
>
> Dec 29 16:30:04 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
> device: %s"
> Dec 29 16:35:51 oliva pulseaudio[70213]: [sndio] module-sndio.c: POLLHUP!
> Dec 29 16:35:51 oliva gnome-settings-daemon: vfprintf %s NULL in
> "gvc-mixer-output-set-property - card port name: %s "
> Dec 29 16:35:51 oliva gnome-shell: vfprintf %s NULL in
> "gvc-mixer-output-set-property - card port name: %s "
> Dec 29 16:42:30 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
> device: %s"
> Dec 29 16:49:24 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
> device: %s"
>
> The last time deadbeef crashed, here's the backtrace:
>
> #0 sio_write (hdl=0x0, buf=0x171416be2230, len=3528)
> at /usr/src/lib/libsndio/sio.c:330
> 330 if (hdl->eof) {
> (gdb) bt
> #0 sio_write (hdl=0x0, buf=0x171416be2230, len=3528)
> at /usr/src/lib/libsndio/sio.c:330
> #1 0x1714690edd3e in sndio_thread () from
> /usr/local/lib/deadbeef/sndio.so
> #2 0x17147941e7ee in _rthread_start (v=Variable "v" is not available.
> )
> at /usr/src/lib/librthread/rthread.c:114
> #3 0x1714184daf8b in __tfork_thread ()
> at /usr/src/lib/libc/arch/amd64/sys/tfork_thread.S:75
> #4 0x in ?? ()
>
> Anything else I can provide?
Another backtrace:
#0 0x12763f3f0d4a in thrkill () at :2
#1 0x12763f3c9169 in *_libc_abort ()
at /usr/src/lib/libc/stdlib/abort.c:52
#2 0x12763f3bf202 in wrterror (d=Variable "d" is not available.
)
at /usr/src/lib/libc/stdlib/malloc.c:296
#3 0x12763f3c0458 in ofree (argpool=0x12761cef5b00, p=0x127661c4b230)
at /usr/src/lib/libc/stdlib/malloc.c:1367
#4 0x12763f3c06bb in free (ptr=0x127661c4b230)
at /usr/src/lib/libc/stdlib/malloc.c:1400
#5 0x1275deea982d in sndio_free () from /usr/local/lib/deadbeef/sndio.so
#6 0x1275deea9d7d in sndio_thread () from /usr/local/lib/deadbeef/sndio.so
#7 0x1275f92a87ee in _rthread_start (v=Variable "v" is not available.
)
at /usr/src/lib/librthread/rthread.c:114
#8 0x12763f3a6f8b in __tfork_thread ()
at /usr/src/lib/libc/arch/amd64/sys/tfork_thread.S:75
deadbeef/libsndiod NULL dereference
I am running GNOME3 and my music player is audio/deadbeef. Since the
last package update very time my screen saver starts the music stop.
Previously it was only the first time per session.
Here's my last output of /var/log/messages. The screen saver started
at 16:35, 16:42 and 16:49.
Dec 29 16:30:04 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
device: %s"
Dec 29 16:35:51 oliva pulseaudio[70213]: [sndio] module-sndio.c: POLLHUP!
Dec 29 16:35:51 oliva gnome-settings-daemon: vfprintf %s NULL in
"gvc-mixer-output-set-property - card port name: %s "
Dec 29 16:35:51 oliva gnome-shell: vfprintf %s NULL in
"gvc-mixer-output-set-property - card port name: %s "
Dec 29 16:42:30 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
device: %s"
Dec 29 16:49:24 oliva gdm: vfprintf %s NULL in "GdmSession: Setting display
device: %s"
The last time deadbeef crashed, here's the backtrace:
#0 sio_write (hdl=0x0, buf=0x171416be2230, len=3528)
at /usr/src/lib/libsndio/sio.c:330
330 if (hdl->eof) {
(gdb) bt
#0 sio_write (hdl=0x0, buf=0x171416be2230, len=3528)
at /usr/src/lib/libsndio/sio.c:330
#1 0x1714690edd3e in sndio_thread () from /usr/local/lib/deadbeef/sndio.so
#2 0x17147941e7ee in _rthread_start (v=Variable "v" is not available.
)
at /usr/src/lib/librthread/rthread.c:114
#3 0x1714184daf8b in __tfork_thread ()
at /usr/src/lib/libc/arch/amd64/sys/tfork_thread.S:75
#4 0x in ?? ()
Anything else I can provide?
gnome-session-failed crash
At least once per day in GNOME3 I get a core dump in my $HOME. Here's the backtrace: (gdb) bt #0 0x15d7ab302e9d in _gtk_style_provider_private_get_settings () from /usr/local/lib/libgtk-3.so.2200.2 #1 0x15d7ab184ded in gtk_css_value_initial_compute () from /usr/local/lib/libgtk-3.so.2200.2 #2 0x15d7ab19a1cc in gtk_css_static_style_compute_value () from /usr/local/lib/libgtk-3.so.2200.2 #3 0x15d7ab1860ee in _gtk_css_lookup_resolve () from /usr/local/lib/libgtk-3.so.2200.2 #4 0x15d7ab19a49c in gtk_css_static_style_new_compute () from /usr/local/lib/libgtk-3.so.2200.2 #5 0x15d7ab19a514 in gtk_css_static_style_get_default () from /usr/local/lib/libgtk-3.so.2200.2 #6 0x15d7ab1888c6 in gtk_css_node_init () from /usr/local/lib/libgtk-3.so.2200.2 #7 0x15d7c3e2037b in g_type_create_instance () from /usr/local/lib/libgobject-2.0.so.4200.4 #8 0x15d7c3e0622d in g_object_new_internal () from /usr/local/lib/libgobject-2.0.so.4200.4 #9 0x15d7c3e07375 in g_object_newv () from /usr/local/lib/libgobject-2.0.so.4200.4 #10 0x15d7c3e0756c in g_object_new () from /usr/local/lib/libgobject-2.0.so.4200.4 #11 0x15d7ab1a2d02 in gtk_css_widget_node_new () ---Type to continue, or q to quit--- from /usr/local/lib/libgtk-3.so.2200.2 #12 0x15d7ab3a857d in gtk_widget_init () from /usr/local/lib/libgtk-3.so.2200.2 #13 0x15d7c3e2037b in g_type_create_instance () from /usr/local/lib/libgobject-2.0.so.4200.4 #14 0x15d7c3e0622d in g_object_new_internal () from /usr/local/lib/libgobject-2.0.so.4200.4 #15 0x15d7c3e07375 in g_object_newv () from /usr/local/lib/libgobject-2.0.so.4200.4 #16 0x15d7c3e0756c in g_object_new () from /usr/local/lib/libgobject-2.0.so.4200.4 #17 0x15d59b900b6c in __register_frame_info () from /usr/local/libexec/gnome-session-failed #18 0x15d59b900892 in ?? () from /usr/local/libexec/gnome-session-failed #19 0x in ?? ()
liferea/GDK crash
liferea crash every second time when I use it under GNOME. Most of the time when refreshing feeds, sometimes at startup. I would spend more time chasing the issue if I had debug symbols in glib/gdk. Here's the trace: #0 0x0ba9a561a393 in gdk_pixbuf_get_from_surface () from /usr/local/lib/libgdk-3.so.2200.2 (gdb) bt #0 0x0ba9a561a393 in gdk_pixbuf_get_from_surface () from /usr/local/lib/libgdk-3.so.2200.2 #1 0x0baa27ae5fec in gtk_status_icon_update_image () from /usr/local/lib/libgtk-3.so.2200.2 #2 0x0ba77233a529 in ui_tray_get_count () from /usr/local/bin/liferea #3 0x0ba77233a974 in ui_tray_update () from /usr/local/bin/liferea #4 0x0ba77230e937 in feedlist_update_new_item_count () from /usr/local/bin/liferea #5 0x0ba77230d2c6 in feed_get_node_type () from /usr/local/bin/liferea #6 0x0ba772319687 in subscription_update_favicon () from /usr/local/bin/liferea #7 0x0ba772319e8e in update_request_free () from /usr/local/bin/liferea #8 0x0ba9729e5532 in g_main_context_dispatch () from /usr/local/lib/libglib-2.0.so.4200.4 #9 0x0ba9729e76eb in g_main_context_iterate () from /usr/local/lib/libglib-2.0.so.4200.4 #10 0x0ba9729e77b7 in g_main_context_iteration () from /usr/local/lib/libglib-2.0.so.4200.4 #11 0x0baa58a3e14e in g_application_run () from /usr/local/lib/libgio-2.0.so.4200.4 #12 0x0ba77231b024 in main () from /usr/local/bin/liferea (gdb)
Re: blink1: openbsd backend / remove libusb dependency
On 06/12/16(Tue) 16:40, Patrick Wildt wrote: > [...] > blink1 uses libusb-compat to talk to a USB device that speaks hid. I > have come to realize that the usage of libusb by blink1 makes libusb > send 4 packets to each connected USB device when it tries to find which > USB devices are connected. That's a bug in libusb/kernel interface and should be fixed. libusb1 properly documents which functions MUST NOT generate traffics, and we do not respect that. > [...] I have USB sticks that abort operation as > soon as they receive a "get config descriptor" while doing block read/ > write operations. This means that EHCI transfers time out and might > even throw a panic, simply by calling blink1-tool. > > Instead of using and depending on libusb (especially the old "compat" > version) I propose to change blink1 to talk to USB hid devices only > through uhid(4). This has the benefits that we don't depend on libusb, > only talk to uhid(4) devices and not _all_ USB devices, don't stress the > USB bus, and simplify the implementation. Unfortunately this means we > implement another backend (that can probably be used my more BSDs). Basically what you say is "let's not fix the bugs in libusb and keep the old unused interface". I disagree with this approach. libusb *is* a standard so better fix and improve it rather that work around it. Improving libusb1 will result in better USB support for everybody.
Re: New port devel/libuv, a new dependency for CMake-3.7.0
On 26/11/16(Sat) 14:51, David Coppa wrote: > On Sat, Nov 26, 2016 at 2:47 PM, Stuart Henderson > wrote: > > On 2016/11/26 13:59, David Coppa wrote: > >> > >> Hi. > >> > >> Here's a new port for libuv-1.10.1, a new dependency for the upcoming > >> update to CMake-3.7.0. > >> > >> Feedback welcome! > > > >> WANTLIB += kvm pthread > > > > ugh kvm.. > > Yes, the openbsd backend uses kvm. Are you sure? $ grep kvm libuv/src/unix/openbsd.c #include It looks like only the netbsd and freebsd backends do kvm(3) functions.
gtk2 vs gtk3 HiDPI
Is there a reason to use gtk2 over gtk3 in a port? Diff below switch
audio/deadbeef to gtk3 which makes it useable on a HiDPI screen. ok?
Index: Makefile
===
RCS file: /cvs/ports/audio/deadbeef/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile18 Mar 2016 23:12:15 - 1.23
+++ Makefile21 Aug 2016 17:44:45 -
@@ -2,7 +2,7 @@
COMMENT = DeaDBeeF is an audio player
DISTNAME = deadbeef-0.6.2
-REVISION = 7
+REVISION = 8
CATEGORIES = audio
HOMEPAGE = http://deadbeef.sourceforge.net/
@@ -10,16 +10,16 @@ HOMEPAGE = http://deadbeef.sourceforge.
# GPLv2 & LGPLv2.1
PERMIT_PACKAGE_CDROM = Yes
-WANTLIB += FLAC Imlib2 X11 Xcomposite Xcursor Xdamage Xext Xfixes
-WANTLIB += Xi Xinerama Xrandr Xrender atk-1.0 avcodec avformat
-WANTLIB += avutil c cairo cddb cdio crypto curl dbus-1 expat faad
-WANTLIB += ffi fontconfig freetype gdk-x11-2.0 gdk_pixbuf-2.0 gio-2.0
-WANTLIB += glib-2.0 gmodule-2.0 gobject-2.0 gthread-2.0 gtk-x11-2.0
-WANTLIB += glib-2.0 gmodule-2.0 gobject-2.0 graphite2 gthread-2.0
-WANTLIB += gtk-x11-2.0 harfbuzz idn m mad mms mp4ff mpcdec nghttp2 ogg
-WANTLIB += pango-1.0 pangocairo-1.0 pangoft2-1.0 pcre pixman-1 png
-WANTLIB += pthread pthread-stubs sndfile sndio ssl vorbis vorbisenc
-WANTLIB += vorbisfile wavpack xcb xcb-render xcb-shm z zip
+WANTLIB += FLAC ICE Imlib2 SM X11 Xcomposite Xcursor Xdamage Xext
+WANTLIB += Xfixes Xi Xinerama Xrandr Xrender atk-1.0 atk-bridge-2.0
+WANTLIB += atspi avcodec avformat avutil c cairo cairo-gobject
+WANTLIB += cddb cdio crypto curl dbus-1 epoxy expat faad ffi fontconfig
+WANTLIB += freetype gdk-3 gdk_pixbuf-2.0 gio-2.0 glib-2.0 gmodule-2.0
+WANTLIB += gobject-2.0 graphite2 gthread-2.0 gtk-3 harfbuzz idn
+WANTLIB += m mad mms mp4ff mpcdec nghttp2 ogg pango-1.0 pangocairo-1.0
+WANTLIB += pangoft2-1.0 pcre pixman-1 png pthread pthread-stubs
+WANTLIB += sndfile sndio ssl vorbis vorbisenc vorbisfile wavpack
+WANTLIB += xcb xcb-render xcb-shm z zip
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=deadbeef/}
EXTRACT_SUFX = .tar.bz2
@@ -52,7 +52,7 @@ LIB_DEPENDS = archivers/libzip \
multimedia/libmms \
net/curl \
x11/dbus \
- x11/gtk+2
+ x11/gtk+3
USE_GMAKE =Yes
USE_LIBTOOL = gnu
@@ -70,7 +70,7 @@ CONFIGURE_ARGS += --disable-static \
--disable-alsa \
--disable-coreaudio \
--disable-gme \
- --disable-gtk3 \
+ --disable-gtk2 \
--disable-mono2stereo \
--disable-oss \
--disable-pulse \
Index: pkg/PLIST
===
RCS file: /cvs/ports/audio/deadbeef/pkg/PLIST,v
retrieving revision 1.7
diff -u -p -r1.7 PLIST
--- pkg/PLIST 23 Sep 2014 13:44:04 - 1.7
+++ pkg/PLIST 21 Aug 2016 17:02:23 -
@@ -9,7 +9,7 @@ lib/deadbeef/alac.so
lib/deadbeef/artwork.so
lib/deadbeef/cdda.so
lib/deadbeef/converter.so
-lib/deadbeef/converter_gtk2.so
+lib/deadbeef/converter_gtk3.so
lib/deadbeef/convpresets/
lib/deadbeef/convpresets/AAC_(Nero_FAAC).txt
lib/deadbeef/convpresets/ALAC.txt
@@ -26,7 +26,7 @@ lib/deadbeef/convpresets/WavPack.txt
lib/deadbeef/dca.so
lib/deadbeef/ddb_ao.so
lib/deadbeef/ddb_dumb.so
-lib/deadbeef/ddb_gui_GTK2.so
+lib/deadbeef/ddb_gui_GTK3.so
lib/deadbeef/ddb_shn.so
lib/deadbeef/ffap.so
lib/deadbeef/ffmpeg.so
@@ -39,9 +39,9 @@ lib/deadbeef/mpgmad.so
lib/deadbeef/musepack.so
lib/deadbeef/notify.so
lib/deadbeef/nullout.so
-lib/deadbeef/pltbrowser_gtk2.so
+lib/deadbeef/pltbrowser_gtk3.so
lib/deadbeef/shellexec.so
-lib/deadbeef/shellexecui_gtk2.so
+lib/deadbeef/shellexecui_gtk3.so
lib/deadbeef/sndfile.so
lib/deadbeef/sndio.so
lib/deadbeef/tta.so
@@ -104,8 +104,6 @@ share/locale/it/LC_MESSAGES/deadbeef.mo
share/locale/ja/LC_MESSAGES/deadbeef.mo
share/locale/kk/LC_MESSAGES/deadbeef.mo
share/locale/km/LC_MESSAGES/deadbeef.mo
-share/locale/lg/
-share/locale/lg/LC_MESSAGES/
share/locale/lg/LC_MESSAGES/deadbeef.mo
share/locale/lt/LC_MESSAGES/deadbeef.mo
share/locale/nl/LC_MESSAGES/deadbeef.mo
Re: sysutils/apcupsd with USB cable broken in latest snap
On 06/06/16(Mon) 19:01, Johan Huldtgren wrote: > hello, > > I have an APC Smart-UPS-2200 which is connected via USB, as long as > I've followed the instructions in pkg-readme about disabling uhidev > and upd it's always worked fine. When I updated to the latest snap > (June 4th) however I am now always seeing "COMMLOST" in the output of > 'apcaccess status' I tried booting the previous kernel (May 19th, > #1709), but that did not resolve it, it seems the breakage occurred > when apcupsd went from 3.14.13p1 to 3.14.13p2 which seems to have been > because the removal of USB_GET_STRING_DESC. I did try the newly > committed (but as yet unpackaged) 3.14.14 but no more luck there. > > I tried to build 3.14.13p1 to see if I can confirm what broke it, but > I've been unable to succeed. I backed out the USB_GET_STRING_DESC > changes and built a new kernel based on sources from around > 2016-05-20, I tried to do the same thing with the ports tree but > clearly I've missed something, because I never make it further than: > > make[5]: Entering directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/drivers/usb/bsd' > CXX src/drivers/usb/bsd/hidutils.c > if test ! -d .obj; then mkdir -p .obj; fi > c++ -x c++ -I/usr/local/include > -I/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/libusbhid > -I/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/include -O2 -pipe > -fno-exceptions -fno-rtti -Wall -pthread -c -o .obj/hidutils.o hidutils.c > hidutils.c: In function 'const char* hidu_get_string(int, int)': > hidutils.c:354: error: 'USB_GET_STRING_DESC' was not declared in this scope > ../../../../autoconf/targets.mak:141: recipe for target '.obj/hidutils.o' > failed > gmake[5]: *** [.obj/hidutils.o] Error 1 > gmake[5]: Leaving directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/drivers/usb/bsd' > ../../../../autoconf/targets.mak:55: recipe for target 'all' failed > gmake[4]: *** [all] Error 2 > gmake[4]: Leaving directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/drivers/usb/bsd' > ../../../autoconf/targets.mak:105: recipe for target 'bsd_DIR' failed > gmake[3]: *** [bsd_DIR] Error 2 > gmake[3]: Leaving directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/drivers/usb' > ../../autoconf/targets.mak:105: recipe for target 'usb_DIR' failed > gmake[2]: *** [usb_DIR] Error 2 > gmake[2]: Leaving directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src/drivers' > ../autoconf/targets.mak:105: recipe for target 'drivers_DIR' failed > gmake[1]: *** [drivers_DIR] Error 2 > gmake[1]: Leaving directory > '/usr/ports/pobj/apcupsd-3.14.13/apcupsd-3.14.13/src' > autoconf/targets.mak:105: recipe for target 'src_DIR' failed > gmake: *** [src_DIR] Error 2 > *** Error 2 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2675 > '/usr/ports/pobj/apcupsd-3.14.13/.build_done') > *** Error 1 in /usr/ports/sysutils/apcupsd > (/usr/ports/infrastructure/mk/bsd.port.mk:2397 'all') > > any clues? Does it work if you use the system libusbhid instead of the bundled one?
Re: Spending time for macppc - taken from [Re: lang/gcc on macppc]
On 04/02/16(Thu) 18:57, Christoph R. Murauer wrote: > > macppc is starting to be unusable as desktop because nobody > > is spending time to ensure the ports build and run correctly. > > > > Seriously asked, what is the reason, that noone spent time ? My opinion is that there are so many things that can be improved in OpenBSD that I now prefer to spend my time in something else. > If I am interested spending time, could I get some help (specially as > if I decided to learn coding) ? Well first thing to do is to try to use a macppc as desktop. Then figure out what's missing/not working then find the bugs and fix them. It's an insane amount of work since freesotfware application are no longer tested on bigendian and/or powerpc machines.
Re: lang/gcc on macppc
On 04/02/16(Thu) 00:29, Michael McConville wrote: > I'm getting the below error when building the GCC port on a Power Mac G5 > running the most recent snapshot. It fails reliably when I retry the > build, so it doesn't seem like an Act of God(TM) caused by a kernel or > hardware bug. > > The build is successful with Ada disabled. > > Is this a known issue? Let me know if a full build log would be useful. Are you using GENERIC.MP or GENERIC? Does it make a difference? Could it be related to a bug in our binutils? macppc is starting to be unusable as desktop because nobody is spending time to ensure the ports build and run correctly.
Re: Firefox performance regressions
On 06/01/16(Wed) 11:19, Landry Breuil wrote: > [...] > i've had multiple ppl coming to me privately about this - Yes, > performance with firefox has been steadily degrading in the past > releases, since around 39 or so - i'm aware of this, but nobody is > actually working on it - i won't/can't since this is way out my skills > and i have no time nor motivation for that. > > Debugging this requires profiling / running the browser within ktrace > and figuring out why apparently it does way much more syscalls than > before, which might be a clue, or not. Of course since you cant really > use traditional tools like gdb, your toolbox is empty. Or that could be > graphical stack regressions. Or so many other things. The builtin > profiler needs specific code to work on OpenBSD. > > Anybody is welcome to look into it - or 5.9 will ship with firefox 43 > as is. I started looking at this but didn't go far. It seems that the problem is related to/exposed by the use of pthread_mutex_lock(3) & friends. I tried to analyze ltrace(1) outputs, but I got lost in Firefox's sources. I really don't know where to look at. Here's what I wrote two months ago: On 18/11/15(Wed) 20:44 +0100, Martin Pieuchot wrote: > The actual firefox port (not the -esr one) might be exposing a bug in > librthread resulting in a storm of sched_yield(2) calls when multiples > threads are trying to access the same lock, see _spinlock(). > > I generated two ltrace(1) dumps for librthread hackers to have a look > at. They are both generate with a hacked version of ltrace(1) with: > > $ ltrace -p $pid -t cu -u libpthread ; sleep 2; ktrace -C > > After having started firefox with: > > $ LD_TRACE_PLT="" LD_TRACE_PLTSPEC="libpthread" DISPLAY=:0 firefox > > > The dumps are generate while firefox is sitting in its own custom home > page. > > > kump-esr.txt a dump of the firefox-esr which work "not so bad" > > kdump-nightly.txt a dump of firefox's trunk which expose the problem > > > $ grep sched_yield kdump-esr.txt |wc -l >4 > $ grep sched_yield kdump-nightly.txt |wc -l > 89418 > > > The sequence below is an extract of what you can find there, when you > have multiple threads fighting for the malloc lock... this goes on > forever and gets even worse when thread 1019109 returns from thrsleep... > > Apparently they are all waiting for thread 1010468 to release the malloc > lock or... is it bad doctor? > > > 13288/1032189 firefox-bin RET sched_yield 0 > 13288/1010095 firefox-bin USER .plt symbol: 7 bytes"__errno" > 13288/1032189 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1010095 firefox-bin USER .plt symbol: 19 bytes > "_thread_malloc_lock" > 13288/1032189 firefox-bin CALL sched_yield() > 13288/1027370 firefox-bin USER .plt symbol: 11 bytes"_spinunlock" > 13288/1010095 firefox-bin USER .plt symbol: 9 bytes"_spinlock" > 13288/1027370 firefox-bin USER .plt symbol: 7 bytes"__errno" > 13288/1010095 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1010095 firefox-bin CALL sched_yield() > 13288/1027370 firefox-bin USER .plt symbol: 19 bytes > "_thread_malloc_lock" > 13288/1010095 firefox-bin RET sched_yield 0 > 13288/1027370 firefox-bin USER .plt symbol: 9 bytes"_spinlock" > 13288/1010095 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1027370 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1010095 firefox-bin CALL sched_yield() > 13288/1027370 firefox-bin CALL sched_yield() > 13288/1032189 firefox-bin RET sched_yield 0 > 13288/1032189 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1032189 firefox-bin CALL sched_yield() > 13288/1027370 firefox-bin RET sched_yield 0 > 13288/1027370 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1027370 firefox-bin CALL sched_yield() > 13288/1032189 firefox-bin RET sched_yield 0 > 13288/1032189 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1032189 firefox-bin CALL sched_yield() > 13288/1027370 firefox-bin RET sched_yield 0 > 13288/1010095 firefox-bin RET sched_yield 0 > 13288/1027370 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1010095 firefox-bin USER .plt symbol: 12 bytes"_atomic_lock" > 13288/1027370 firefox-bin CALL sched_yield() > 13288/1010095 firefox-
Re: plan9port: use getifaddrs(3) instead of kvm(3)
On 07/12/15(Mon) 18:51, Gleydson Soares wrote:
> Hi,
>
> plan9/plan9port is broken after network stack changes in net/if_var.h
> here is a diff that convert plan9port to use getifaddrs(3) instead of kvm(3)
> in p9p auxstats.
>
> builds and runs fine(auxstats is incrementing and seems ok), but I would
> appreciate other eyes(review),
Diff reads fine, ok with me.
> Index: Makefile
> ===
> RCS file: /cvs/ports/plan9/plan9port/Makefile,v
> retrieving revision 1.10
> diff -u -p -u -p -r1.10 Makefile
> --- Makefile 4 Dec 2015 04:26:33 - 1.10
> +++ Makefile 7 Dec 2015 21:36:31 -
> @@ -6,7 +6,7 @@ BROKEN-powerpc = threading issues
> COMMENT =Plan 9 from user space
>
> DISTNAME = plan9port-2015
> -REVISION = 0
> +REVISION = 1
>
> GH_ACCOUNT = 9fans
> GH_PROJECT = plan9port
> Index: patches/patch-src_cmd_auxstats_OpenBSD_c
> ===
> RCS file: patches/patch-src_cmd_auxstats_OpenBSD_c
> diff -N patches/patch-src_cmd_auxstats_OpenBSD_c
> --- /dev/null 1 Jan 1970 00:00:00 -
> +++ patches/patch-src_cmd_auxstats_OpenBSD_c 7 Dec 2015 21:36:31 -
> @@ -0,0 +1,157 @@
> +$OpenBSD$
> +
> +replace kvm(3) with getifaddrs(3)
> +
> +--- src/cmd/auxstats/OpenBSD.c.orig Wed Nov 11 12:09:34 2015
> src/cmd/auxstats/OpenBSD.c Mon Dec 7 02:20:58 2015
> +@@ -1,14 +1,12 @@
> + #include
> +-#include
> +-#include
> + #include
> + #include
> + #include
> + #include
> ++#include
> + #include
> + #include
> + #include
> +-#include
> + #include
> + #include
> + #include
> +@@ -24,11 +22,9 @@ void xcpu(int);
> + void xswap(int);
> + void xsysctl(int);
> + void xnet(int);
> +-void xkvm(int);
> +
> + void (*statfn[])(int) =
> + {
> +-xkvm,
> + xapm,
> + xloadavg,
> + xcpu,
> +@@ -37,14 +33,6 @@ void (*statfn[])(int) =
> + 0
> + };
> +
> +-static kvm_t *kvm;
> +-
> +-static struct nlist nl[] = {
> +-{ "_ifnet" },
> +-{ "_cp_time" },
> +-{ "" }
> +-};
> +-
> + void
> + xloadavg(int first)
> + {
> +@@ -76,78 +64,37 @@ xapm(int first)
> + Bprint(&bout, "battery =%d 100\n", ai.battery_life);
> + }
> +
> +-
> + void
> +-kvminit(void)
> +-{
> +-char buf[_POSIX2_LINE_MAX];
> +-
> +-if(kvm)
> +-return;
> +-kvm = kvm_openfiles(nil, nil, nil, O_RDONLY, buf);
> +-if(kvm == nil) {
> +-fprint(2, "kvm open error\n%s", buf);
> +-return;
> +-}
> +-if(kvm_nlist(kvm, nl) < 0 || nl[0].n_type == 0){
> +-kvm = nil;
> +-return;
> +-}
> +-}
> +-
> +-void
> +-xkvm(int first)
> +-{
> +-if(first)
> +-kvminit();
> +-}
> +-
> +-int
> +-kread(ulong addr, char *buf, int size)
> +-{
> +-if(kvm_read(kvm, addr, buf, size) != size){
> +-memset(buf, 0, size);
> +-return -1;
> +-}
> +-return size;
> +-}
> +-
> +-void
> + xnet(int first)
> + {
> + ulong out, in, outb, inb, err;
> +-static ulong ifnetaddr;
> +-ulong addr;
> +-struct ifnet ifnet;
> +-struct ifnet_head ifnethead;
> +-char name[16];
> ++struct ifaddrs *ifa, *ifap;
> ++struct if_data *ifd = NULL;
> +
> +-if(first)
> ++if (first)
> + return;
> +
> +-if(ifnetaddr == 0){
> +-ifnetaddr = nl[0].n_value;
> +-if(ifnetaddr == 0)
> +-return;
> +-}
> ++out = in = outb = inb = err = 0;
> +
> +-if(kread(ifnetaddr, (char*)&ifnethead, sizeof ifnethead) < 0)
> ++if (getifaddrs(&ifap) == -1)
> + return;
> +
> +-out = in = outb = inb = err = 0;
> +-addr = (ulong)TAILQ_FIRST(&ifnethead);
> +-while(addr){
> +-if(kread(addr, (char*)&ifnet, sizeof ifnet) < 0
> +-|| kread((ulong)ifnet.if_xname, name, 16) < 0)
> +-return;
> +-name[15] = 0;
> +-addr = (ulong)TAILQ_NEXT(&ifnet, if_list);
> +-out += ifnet.if_opackets;
> +-in += ifnet.if_ipackets;
> +-outb += ifnet.if_obytes;
> +-inb += ifnet.if_ibytes;
> +-err += ifnet.if_oerrors+ifnet.if_ierrors;
> ++for (ifa = ifap; ifa != NULL; ifa = ifa->ifa_next) {
> ++if (ifa->ifa_addr == NULL ||
> ++ifa->ifa_addr->sa_family != AF_LINK)
> ++continue;
> ++
> ++ifd = ifa->ifa_data;
> ++
> ++if (ifd != NULL) {
> ++out += ifd->ifi_opackets;
> ++in += ifd->ifi_ipackets;
> ++outb += ifd->ifi_obytes;
> ++inb += ifd->ifi_ibytes;
> ++err += ifd->ifi_ierrors;
> ++}
> + }
> ++
> + Bprint(&bout, "etherin %lud 1000\n", in);
> + Bprint(&bout, "etherout %lud 1000\n", out);
> + Bprint(&bout, "etherinb %lud 100\n", i
Re: Firefox 41.0.2 with W^X
On 02/11/15(Mon) 13:21, David Coppa wrote: > On Fri, Oct 23, 2015 at 2:01 PM, Martin Pieuchot wrote: > > On 22/10/15(Thu) 21:40, Amit Kulkarni wrote: > >> On Thu, Oct 22, 2015 at 12:26 PM, David Coppa wrote: > >> > >> > On Thu, Oct 22, 2015 at 3:45 PM, Ed Ahlsen-Girard > >> > wrote: > >> > > I have noticed a performance hit since the switch was flipped. Firefox > >> > > stays at the top of top most of the time, and its CPU percentages have > >> > > spiked to 175% if multiple tabs were being opened. dmesg below the sig. > >> > > >> > Can you try if the attached patch is an improvement? > >> > > >> > >> Hi, > >> > >> This CPU spike is present with October 11 packages (Firefox 41.0.1) on > >> amd64, so it will be difficult to isolate the performance impact of the W > >> ^X vs the existing situation. > > > > FWIW I found that firefox is (ab)using pthread_mutex_trylock(3) a lot > > resulting in a storm of sched_yield(2) triggering a lot (dozen to > > hundreds of thousands) of IPIs on my x220. > > > > I tried to look at the source code but couldn't figure out where the > > call of pthread_mutex_trylock(3) are coming from. Firefox is just a > > monster. > > > > I'm sorry but I agree that if nobody is taking care of this regression > > it will be really hard to measure the impact of the W^X change. > > > > Martin, > > Has this problem manifested itself with firefox-41 or was it already > present with 40.x ? I don't remember, I can try to figure out but I'd be happy if somebody else would do it 8)
Re: Firefox 41.0.2 with W^X
On 22/10/15(Thu) 21:40, Amit Kulkarni wrote: > On Thu, Oct 22, 2015 at 12:26 PM, David Coppa wrote: > > > On Thu, Oct 22, 2015 at 3:45 PM, Ed Ahlsen-Girard > > wrote: > > > I have noticed a performance hit since the switch was flipped. Firefox > > > stays at the top of top most of the time, and its CPU percentages have > > > spiked to 175% if multiple tabs were being opened. dmesg below the sig. > > > > Can you try if the attached patch is an improvement? > > > > Hi, > > This CPU spike is present with October 11 packages (Firefox 41.0.1) on > amd64, so it will be difficult to isolate the performance impact of the W > ^X vs the existing situation. FWIW I found that firefox is (ab)using pthread_mutex_trylock(3) a lot resulting in a storm of sched_yield(2) triggering a lot (dozen to hundreds of thousands) of IPIs on my x220. I tried to look at the source code but couldn't figure out where the call of pthread_mutex_trylock(3) are coming from. Firefox is just a monster. I'm sorry but I agree that if nobody is taking care of this regression it will be really hard to measure the impact of the W^X change.
Re: [wip] Firefox 40.0
On 30/09/15(Wed) 15:08, patrick keshishian wrote: > On 9/30/15, Erling Westenvik wrote: > > On Tue, Aug 11, 2015 at 09:23:49AM +0200, Landry Breuil wrote: > >> On Thu, Jul 23, 2015 at 06:09:12AM +0200, Landry Breuil wrote: > >> > >> And fx 40.0 is almost there. Will be in cvs when ports unlock.. > > > > Hi, > > Since the move from 39.0 to 40.0, and lately 41.0, Firefox has become > > unbearable slow/unrepsonsive on my Thinkpad T500. See dmesg below. > > Just confirming a noticeable drop in responsiveness of FF 40.x. > Haven't tried 41 yet. I confirm this regression. All I have been able to observe right now with ltrace/ktrace is that a lot of librthread's _spinlock are called which results in an insanely high number of sched_yield(2) syscalls...
link_addr(3)
I just removed link_addr(3) from libc, we'd like to remove the kernel routing interface decoding interfaces names. Last time I asked there was one port using this function: net/dhcpcd. Apparently it does not use link_addr(3) anymore, if somebody finds another port using this function and need some help fixing it. Please let me know. Martin
libusb 1.0.20rc1
Here's a diff to test first release candidate of the next libusb
version.
I'd appreciate if people using libusb could test, especially if
your port was preventing an upgrade. I don't plan to commit an
upgrade to a RC but know would be a good time to find & report
potential issues.
Index: Makefile
===
RCS file: /cvs/ports/devel/libusb1/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile16 Mar 2015 18:07:43 - 1.23
+++ Makefile14 Aug 2015 11:57:51 -
@@ -2,15 +2,15 @@
COMMENT = library for USB device access from userspace
-VERSION = 1.0.9
+VERSION = 1.0.20-rc1
DISTNAME = libusb-${VERSION}
PKGNAME = libusb1-${VERSION}
-REVISION = 9
-SHARED_LIBS += usb-1.0 1.0 # 1.0
+
+SHARED_LIBS += usb-1.0 1.1 # 1.0
CATEGORIES = devel
-HOMEPAGE = http://www.libusb.org
+HOMEPAGE = http://www.libusb.info
MAINTAINER = Martin Pieuchot
@@ -19,8 +19,26 @@ PERMIT_PACKAGE_CDROM = Yes
WANTLIB = pthread
+.if ${VERSION:S/-rc//} == ${VERSION}
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=libusb/}
EXTRACT_SUFX = .tar.bz2
+.else
+GH_ACCOUNT = libusb
+GH_PROJECT = libusb
+GH_COMMIT =d73bc723aaeb4e043bb079198b37f5e0b0726168
+
+
+BUILD_DEPENDS= devel/libtool \
+ ${MODGNU_AUTOCONF_DEPENDS} \
+ ${MODGNU_AUTOMAKE_DEPENDS}
+
+AUTOCONF_VERSION= 2.69
+AUTOMAKE_VERSION= 1.11
+
+pre-configure:
+ cd ${WRKSRC} && env AUTOCONF_VERSION=${AUTOCONF_VERSION} \
+ AUTOMAKE_VERSION=${AUTOMAKE_VERSION} sh ./bootstrap.sh
+.endif
CONFIGURE_STYLE = gnu
CONFIGURE_ARGS += ${CONFIGURE_SHARED}
Index: distinfo
===
RCS file: /cvs/ports/devel/libusb1/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo18 Jan 2015 03:13:17 - 1.3
+++ distinfo14 Aug 2015 11:45:33 -
@@ -1,2 +1,2 @@
-SHA256 (libusb-1.0.9.tar.bz2) = 6SDu3C0GsJYGYRyZ7HMEQTxnhMum4zko54JD0yMZX5s=
-SIZE (libusb-1.0.9.tar.bz2) = 421971
+SHA256 (libusb-1.0.20-rc1.tar.gz) =
lmlBTmwtinvE5Rg584rKMhD9l3wJP6GQ7HnE5W4eiZ0=
+SIZE (libusb-1.0.20-rc1.tar.gz) = 361819
Index: patches/patch-libusb_core_c
===
RCS file: /cvs/ports/devel/libusb1/patches/patch-libusb_core_c,v
retrieving revision 1.3
diff -u -p -r1.3 patch-libusb_core_c
--- patches/patch-libusb_core_c 19 Jun 2012 21:10:59 - 1.3
+++ patches/patch-libusb_core_c 16 Jun 2015 11:45:17 -
@@ -4,9 +4,9 @@ On OpenBSD USB controllers are shown as
itinial limit too small. On a recent machine this value is almost
always exceeded, so bump it.
libusb/core.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/core.c Tue Jun 12 15:48:05 2012
-@@ -468,7 +468,7 @@ libusb_free_device_list(list, 1);
+--- libusb/core.c.orig Sun Jan 26 00:06:33 2014
libusb/core.c Mon Jan 27 10:08:37 2014
+@@ -448,7 +448,7 @@ libusb_free_device_list(list, 1);
* which grows when required. it can be freed once discovery has completed,
* eliminating the need for a list node in the libusb_device structure
* itself. */
Index: patches/patch-libusb_os_openbsd_usb_c
===
RCS file: /cvs/ports/devel/libusb1/patches/patch-libusb_os_openbsd_usb_c,v
retrieving revision 1.5
diff -u -p -r1.5 patch-libusb_os_openbsd_usb_c
--- patches/patch-libusb_os_openbsd_usb_c 22 Apr 2013 08:37:20 -
1.5
+++ patches/patch-libusb_os_openbsd_usb_c 16 Jun 2015 11:45:17 -
@@ -1,570 +0,0 @@
-$OpenBSD: patch-libusb_os_openbsd_usb_c,v 1.5 2013/04/22 08:37:20 mpi Exp $
-
-Add support for non ugen(4) attached devices through usb(4) buses.
-
libusb/os/openbsd_usb.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/os/openbsd_usb.cSun Apr 21 12:39:40 2013
-@@ -1,5 +1,5 @@
- /*
-- * Copyright (c) 2011 Martin Pieuchot
-+ * Copyright (c) 2011-2013 Martin Pieuchot
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
-@@ -32,8 +32,8 @@
- #include "libusbi.h"
-
- struct device_priv {
-- char devnode[16];
-- int fd;
-+ char *devname; /* name of the ugen(4) node */
-+ int fd; /* device file descriptor */
-
- unsigned char *cdesc; /* active config descriptor */
- usb_device_descriptor_t ddesc; /* usb device descriptor */
-@@ -82,11 +82,14 @@ static int obsd_clock_gettime(int, struct timespec *);
- * Private functions
- */
- static in
Re: Chrome 40+ FIDO U2F Security Keys
On 14/08/15(Fri) 12:22, Alexey Suslikov wrote: > [...] > I have a question about USB. Is there any sort of multipath in USB > standard, like in SCSI, so USB stack can see same device attached > to different controllers/buses? Is it possible? Not that I know.
Re: Chrome 40+ FIDO U2F Security Keys
On 14/08/15(Fri) 10:17, Stuart Henderson wrote: > On 2015/08/14 11:00, Martin Pieuchot wrote: > > On 13/08/15(Thu) 20:35, Alexey Suslikov wrote: > > > On Tue, Aug 11, 2015 at 11:20 PM, Brandon Mercer > > > wrote: > > > >> Another thing that bothers me. These keys are USB HIDs, right? Is it > > > >> safe > > > >> enough to let browser access USB bus (USB keyboard is HID and people > > > >> can type different things on it). > > > > What do you mean? You're already typing in your browser, right? AFAIK > > these devices act like standard keyboards. > > You're thinking of classic yubikey, which in normal use just emulates > a keyboard and just sends a sequence that looked like it was typed. > (Even for these, programming the device, and using it in other modes, > does require 2-way comms). > > The problem there is that the private key needs to be known by anyone > who allows your key to be used to authenticate. So it's great when you > want a centrally controlled location to gate authentication, but you > wouldn't want to hand this around to random websites. > > U2F is a different protocol, though still presenting as an HID. > It does challenge/response instead, so it needs 2-way comms in normal > use so the challenge can be sent to the key.. > > https://www.yubico.com/wp-content/uploads/2015/03/U2F.png Thanks for the pointer. It seems they have a very simple client-side library which is nothing else than a HID driver in userland with some trendy JSON parsing: https://github.com/Yubico/libu2f-host So if you write a kernel driver for the important bits you can completely abstract the fact that this is a USB device. But the problem becomes then how to interface it with your application(s)? I don't know how Chrome uses this, but if that's the reason for a browser to be linked to the libusb, I feel this is completely wrong. But since a lot of people write USB drivers in userland I understand that applications are following this road.
Re: Chrome 40+ FIDO U2F Security Keys
On 13/08/15(Thu) 20:35, Alexey Suslikov wrote: > On Tue, Aug 11, 2015 at 11:20 PM, Brandon Mercer > wrote: > >> Another thing that bothers me. These keys are USB HIDs, right? Is it safe > >> enough to let browser access USB bus (USB keyboard is HID and people > >> can type different things on it). What do you mean? You're already typing in your browser, right? AFAIK these devices act like standard keyboards. "Is is safe enough" depends on a lot of factors. And I don't know if nor why the browser needs to access your USB bus. > > Well, that part of it is a completely different animal. It's probably worth > > a separate discussion about how the protocol works. You are suggesting that > > this couldn't even be made to work in a secure fashion, and I'm not going to > > disagree with you. > > Not exactly what I suggested. > > My idea was a sort of USB proxy (emulator) to allow software like Chrome > to access USB bus in secure way (like if you can configure what do you want > to expose and what you don't). > > Smth like vscsi midlayer accessible from user-land which user in control of. > > Chrome uses only sort of vusb hub/bus with only permitted devices behind. > > Also useful for usb camera/sound access control. Why not put the same amount of effort in the existing userland interface of the USB stack? In the end what's complicated is the answer to "which user can to what". Honestly I doubt that another layer of abstraction will help, especially if you considering the sate of our USB stack. I don't know if you looked at Chrome's sources but it uses the libusb (don't ask me why) and that's already a "proxy" to use your words.
Re: [tobi...@tmux.org: mark net/rtorrent as broken on powerpc]
On 31/07/15(Fri) 15:31, Stuart Henderson wrote:
> On 2015/07/31 15:53, Tobias Ulmer wrote:
> > Is David around? Any objections should be voiced in the next few hours,
> > before I commit this.
>
> It's actually libtorrent not rtorrent that should be marked broken.
>
> Would it be worth building on i386 with compiler flags to fix that,
> at the expense of not running on crappy old machines that you probably
> wouldn't want to run this on anyway?
There's only one atomic operation on a 64bit value and it is just for
instrumentation. So by adding "--disable-instrumentation" to the
configure arguments libtorrent builds just fine on macppc. I did not
try i386 but it should be the same.
Index: Makefile
===
RCS file: /cvs/ports/net/libtorrent/Makefile,v
retrieving revision 1.41
diff -u -p -r1.41 Makefile
--- Makefile31 Jul 2015 16:08:44 - 1.41
+++ Makefile31 Jul 2015 16:45:13 -
@@ -11,7 +11,7 @@ NOT_FOR_ARCHS=${GCC3_ARCHS}
DISTNAME= libtorrent-0.13.4
EPOCH= 0
-REVISION= 0
+REVISION= 1
SHARED_LIBS += torrent 21.0# .18.0
CATEGORIES=net devel
@@ -32,14 +32,7 @@ CONFIGURE_ARGS= ${CONFIGURE_SHARED} \
--enable-static \
--with-kqueue \
--without-epoll \
+ --disable-instrumentation \
--disable-debug
-
-.include
-
-# for 64-bit atomic ops
-.if ${ARCH:Mi386}
-CFLAGS+= -march=i586 -fomit-frame-pointer
-CXXFLAGS+= -march=i586 -fomit-frame-pointer
-.endif
.include
Re: Update devel/libusb1 to 1.0.19
On 23/06/15(Tue) 14:04, Brandon Mercer wrote:
> On Tue, Jun 23, 2015 at 5:17 AM Martin Pieuchot wrote:
>
> > On 22/06/15(Mon) 23:56, Jérémie Courrèges-Anglas wrote:
> > > Martin Pieuchot writes:
> > >
> > > > Here's an update to the latest version. This diff is basically the
> > > > same as last year and hopefully the regression exposed in the 1.0.18
> > > > are now gone.
> > > >
> > > > Please test with your favorite ports and report back.
> > >
> > > FWIW, the update looks sane ports-wise.
> >
> > Thanks.
> >
> > > If you're still looking for test reports then running a test bulk
> > > (on exopi) in the meantime would be a good idea to avoid any fallout.
> >
> > I doubt this will cause any breakage. But more importantly dcoppa@ told
> > me in private that the regression introduced in 1.0.18 is still present.
> > However is has been fixed in master.
> >
> > Since the changes are not trivial to backport I prefer to wait for the
> > next release.
> >
>
> Is this a good candidate for tracking their github repo?
You can use the diff below if you want. But I'd prefer to stick to
stable releases. Thing is OpenBSD's backend is synchronous whereas
the libusb advertise synchronous transfer submission. This difference
does not really matter for the port ecosystem as most of the ports
use the synchronous libusb-compat wrapper. But this difference matters
as of the 1.0.18 release when libusbx started adding some locks that
trigger a deadlock with one application on OpenBSD.
Now we currently have a GSoC student working on improving our kernel
interface to support asynchronous transfers so I'm not in a rush of
upgrading the current port.
Index: Makefile
===
RCS file: /cvs/ports/devel/libusb1/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile16 Mar 2015 18:07:43 - 1.23
+++ Makefile18 Jun 2015 16:33:06 -
@@ -2,15 +2,19 @@
COMMENT = library for USB device access from userspace
-VERSION = 1.0.9
+VERSION = 1.0.20
DISTNAME = libusb-${VERSION}
PKGNAME = libusb1-${VERSION}
-REVISION = 9
-SHARED_LIBS += usb-1.0 1.0 # 1.0
+
+GH_ACCOUNT = libusb
+GH_PROJECT = libusb
+GH_COMMIT =c141457debff6156b83786eb69b46d873634e5bd
+
+SHARED_LIBS += usb-1.0 1.1 # 1.0
CATEGORIES = devel
-HOMEPAGE = http://www.libusb.org
+HOMEPAGE = http://www.libusb.info
MAINTAINER = Martin Pieuchot
@@ -19,8 +23,15 @@ PERMIT_PACKAGE_CDROM = Yes
WANTLIB = pthread
-MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=libusb/}
-EXTRACT_SUFX = .tar.bz2
+#MASTER_SITES =${MASTER_SITE_SOURCEFORGE:=libusb/}
+#EXTRACT_SUFX =.tar.bz2
+
+BUILD_DEPENDS= devel/libtool \
+ ${MODGNU_AUTOCONF_DEPENDS} \
+ ${MODGNU_AUTOMAKE_DEPENDS}
+
+AUTOCONF_VERSION= 2.69
+AUTOMAKE_VERSION= 1.11
CONFIGURE_STYLE = gnu
CONFIGURE_ARGS += ${CONFIGURE_SHARED}
@@ -29,5 +40,9 @@ CONFIGURE_ARGS += ${CONFIGURE_SHARED}
CONFIGURE_ARGS += --enable-debug-log \
--enable-examples-build
.endif
+
+pre-configure:
+ cd ${WRKSRC} && env AUTOCONF_VERSION=${AUTOCONF_VERSION} \
+ AUTOMAKE_VERSION=${AUTOMAKE_VERSION} sh ./bootstrap.sh
.include
Index: distinfo
===
RCS file: /cvs/ports/devel/libusb1/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo18 Jan 2015 03:13:17 - 1.3
+++ distinfo18 Jun 2015 16:26:48 -
@@ -1,2 +1,2 @@
-SHA256 (libusb-1.0.9.tar.bz2) = 6SDu3C0GsJYGYRyZ7HMEQTxnhMum4zko54JD0yMZX5s=
-SIZE (libusb-1.0.9.tar.bz2) = 421971
+SHA256 (libusb-1.0.20.tar.gz) = qs7ke/y3BBAm5JRbrOHGjFSm0bRrRD4JxlL1cJVDgxc=
+SIZE (libusb-1.0.20.tar.gz) = 361223
Index: patches/patch-libusb_core_c
===
RCS file: /cvs/ports/devel/libusb1/patches/patch-libusb_core_c,v
retrieving revision 1.3
diff -u -p -r1.3 patch-libusb_core_c
--- patches/patch-libusb_core_c 19 Jun 2012 21:10:59 - 1.3
+++ patches/patch-libusb_core_c 16 Jun 2015 11:45:17 -
@@ -4,9 +4,9 @@ On OpenBSD USB controllers are shown as
itinial limit too small. On a recent machine this value is almost
always exceeded, so bump it.
libusb/core.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/core.c Tue Jun 12 15:48:05 2012
-@@ -468,7 +468,7 @@ libusb_free_device_list(list, 1);
+--- libusb/core.c.orig Sun Jan 26 00:06:33
Re: Update devel/libusb1 to 1.0.19
On 22/06/15(Mon) 23:56, Jérémie Courrèges-Anglas wrote: > Martin Pieuchot writes: > > > Here's an update to the latest version. This diff is basically the > > same as last year and hopefully the regression exposed in the 1.0.18 > > are now gone. > > > > Please test with your favorite ports and report back. > > FWIW, the update looks sane ports-wise. Thanks. > If you're still looking for test reports then running a test bulk > (on exopi) in the meantime would be a good idea to avoid any fallout. I doubt this will cause any breakage. But more importantly dcoppa@ told me in private that the regression introduced in 1.0.18 is still present. However is has been fixed in master. Since the changes are not trivial to backport I prefer to wait for the next release. Thanks to all testers. Martin
Stop linking ports to libusbhid
Various ports in our tree support USB devices through the libusbhid. This library, only available on BSD systems, depends on the original model of 1 kernel driver per reportID. Nowadays it is common to plug mouses, keyboards or joysticks that show up as multiples uhid(4) devices. Because of that you might have to bump the number of /dev/uhid* nodes to be able to use your joystick with SDL. Last year with the import of upd(4) we started moving away from the model of 1 driver per reportID. To ease this transition it would be nice to reduce the number of ports depending on libusbhid and instead make them depend on devel/libusb1 or devel/libusb-compat. So if you maintain a port that currently depends on libusbhid, please try to use the libusb instead and let me know if something does not work. Thanks, Martin
Update devel/libusb1 to 1.0.19
Here's an update to the latest version. This diff is basically the
same as last year and hopefully the regression exposed in the 1.0.18
are now gone.
Please test with your favorite ports and report back.
Index: Makefile
===
RCS file: /cvs/ports/devel/libusb1/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile16 Mar 2015 18:07:43 - 1.23
+++ Makefile11 Jun 2015 13:03:28 -
@@ -2,15 +2,15 @@
COMMENT = library for USB device access from userspace
-VERSION = 1.0.9
+VERSION = 1.0.19
DISTNAME = libusb-${VERSION}
PKGNAME = libusb1-${VERSION}
-REVISION = 9
-SHARED_LIBS += usb-1.0 1.0 # 1.0
+
+SHARED_LIBS += usb-1.0 1.1 # 1.0
CATEGORIES = devel
-HOMEPAGE = http://www.libusb.org
+HOMEPAGE = http://www.libusb.info
MAINTAINER = Martin Pieuchot
Index: distinfo
===
RCS file: /cvs/ports/devel/libusb1/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo18 Jan 2015 03:13:17 - 1.3
+++ distinfo11 Jun 2015 13:03:51 -
@@ -1,2 +1,2 @@
-SHA256 (libusb-1.0.9.tar.bz2) = 6SDu3C0GsJYGYRyZ7HMEQTxnhMum4zko54JD0yMZX5s=
-SIZE (libusb-1.0.9.tar.bz2) = 421971
+SHA256 (libusb-1.0.19.tar.bz2) = bFAsgWAC+Q1PdgUKZCnDp+DYQgQiLL/y3Old13O6aEA=
+SIZE (libusb-1.0.19.tar.bz2) = 521496
Index: patches/patch-libusb_core_c
===
RCS file: /cvs/ports/devel/libusb1/patches/patch-libusb_core_c,v
retrieving revision 1.3
diff -u -p -r1.3 patch-libusb_core_c
--- patches/patch-libusb_core_c 19 Jun 2012 21:10:59 - 1.3
+++ patches/patch-libusb_core_c 11 Jun 2015 12:57:54 -
@@ -4,9 +4,9 @@ On OpenBSD USB controllers are shown as
itinial limit too small. On a recent machine this value is almost
always exceeded, so bump it.
libusb/core.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/core.c Tue Jun 12 15:48:05 2012
-@@ -468,7 +468,7 @@ libusb_free_device_list(list, 1);
+--- libusb/core.c.orig Sun Jan 26 00:06:33 2014
libusb/core.c Mon Jan 27 10:08:37 2014
+@@ -448,7 +448,7 @@ libusb_free_device_list(list, 1);
* which grows when required. it can be freed once discovery has completed,
* eliminating the need for a list node in the libusb_device structure
* itself. */
Index: patches/patch-libusb_os_openbsd_usb_c
===
RCS file: patches/patch-libusb_os_openbsd_usb_c
diff -N patches/patch-libusb_os_openbsd_usb_c
--- patches/patch-libusb_os_openbsd_usb_c 22 Apr 2013 08:37:20 -
1.5
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,570 +0,0 @@
-$OpenBSD: patch-libusb_os_openbsd_usb_c,v 1.5 2013/04/22 08:37:20 mpi Exp $
-
-Add support for non ugen(4) attached devices through usb(4) buses.
-
libusb/os/openbsd_usb.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/os/openbsd_usb.cSun Apr 21 12:39:40 2013
-@@ -1,5 +1,5 @@
- /*
-- * Copyright (c) 2011 Martin Pieuchot
-+ * Copyright (c) 2011-2013 Martin Pieuchot
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
-@@ -32,8 +32,8 @@
- #include "libusbi.h"
-
- struct device_priv {
-- char devnode[16];
-- int fd;
-+ char *devname; /* name of the ugen(4) node */
-+ int fd; /* device file descriptor */
-
- unsigned char *cdesc; /* active config descriptor */
- usb_device_descriptor_t ddesc; /* usb device descriptor */
-@@ -82,11 +82,14 @@ static int obsd_clock_gettime(int, struct timespec *);
- * Private functions
- */
- static int _errno_to_libusb(int);
--static int _cache_active_config_descriptor(struct libusb_device *, int);
-+static int _cache_active_config_descriptor(struct libusb_device *);
- static int _sync_control_transfer(struct usbi_transfer *);
- static int _sync_gen_transfer(struct usbi_transfer *);
- static int _access_endpoint(struct libusb_transfer *);
-
-+static int _bus_open(int);
-+
-+
- const struct usbi_os_backend openbsd_backend = {
- "Synchronous OpenBSD backend",
- NULL, /* init() */
-@@ -128,75 +131,105 @@ const struct usbi_os_backend openbsd_backend = {
- 0, /* add_iso_packet_size */
- };
-
-+#define DEVPATH "/dev/"
-+#define USBDEVDEVPATH "usb"
-+
- int
- obsd_get_device_list(struct libusb_context * ctx,
- struct discovered_devs **discdevs)
- {
-+ struct discovered_devs *ddd;
- struct libusb_device *dev;
- struct device_priv *dpriv;
- struct usb_devic
Re: vmmeter.h, VM_UVMEXP & breakage
On 10/04/14(Thu) 15:24, Martin Pieuchot wrote:
> Thanks to naddy, here's the list of ports that will break when
> will stop to include . I'd
> appreciate if the maintainers of the ports below could have a
> look and fix their ports.
>
> If the application is only using the VM_METER sysctl(3) to get
> memory statics, then it should be converted to the VM_UVMEXP
> one. See the attached diff for examples.
>
> If the application also uses the process stats from the "struct
> vmtotal", then you should also include . You can
> have a look at usr.sbin/vmstat/vmstat.c if you want an example.
>
> Attached are three diffs for the following ports, I'm looking
> for reviews/oks.
>
> databases/pg_top
> x11/xfce4/xfce4-taskmanager
These two have been committed.
> sysutils/libstatgrab
This one is below, ok?
> These are the remaining ports to fix:
>
> lang/erlang
> sysutils/monit
> sysutils/symon
> sysutils/toprump
> sysutils/wmmon
> sysutils/xstatbar
> telephony/asterisk
> www/chromium
> x11/kde/base3
> x11/kde4/workspace
I didn't see any activity for these, I sent this mail a month ago,
should I commit the header change?
Martin
Index: Makefile
===
RCS file: /home/ncvs/ports/sysutils/libstatgrab/Makefile,v
retrieving revision 1.9
diff -u -p -r1.9 Makefile
--- Makefile21 Mar 2013 08:48:54 - 1.9
+++ Makefile10 Apr 2014 12:44:14 -
@@ -4,6 +4,7 @@ COMMENT=system statistics gathering lib
DISTNAME= libstatgrab-0.17
CATEGORIES=sysutils
+REVISION= 0
HOMEPAGE= http://www.i-scream.org/libstatgrab/
Index: patches/patch-src_libstatgrab_memory_stats_c
===
RCS file:
/home/ncvs/ports/sysutils/libstatgrab/patches/patch-src_libstatgrab_memory_stats_c,v
retrieving revision 1.1
diff -u -p -r1.1 patch-src_libstatgrab_memory_stats_c
--- patches/patch-src_libstatgrab_memory_stats_c6 Jul 2011 12:45:04
- 1.1
+++ patches/patch-src_libstatgrab_memory_stats_c10 Apr 2014 12:44:14
-
@@ -1,15 +1,82 @@
$OpenBSD: patch-src_libstatgrab_memory_stats_c,v 1.1 2011/07/06 12:45:04
jasper Exp $
-Fix implicit declarations of sysconf() and bzero().
+Prefer the UVM way to the vmmeter one to get memory stats.
src/libstatgrab/memory_stats.c.origWed Jul 6 14:37:07 2011
-+++ src/libstatgrab/memory_stats.c Wed Jul 6 14:38:26 2011
-@@ -50,6 +50,8 @@
- #include
- #include
- #include
-+#include
-+#include
+--- src/libstatgrab/memory_stats.c.origSun Feb 21 11:04:26 2010
src/libstatgrab/memory_stats.c Thu Apr 10 14:43:24 2014
+@@ -91,15 +91,9 @@ sg_mem_stats *sg_get_mem_stats(){
+ u_long inactive_count;
+ int pagesize;
#endif
- #ifdef HPUX
- #include
+-#if defined(NETBSD)
++#if defined(NETBSD) || defined(OPENBSD)
+ struct uvmexp *uvm;
+ #endif
+-#if defined(OPENBSD)
+- int mib[2];
+- struct vmtotal vmtotal;
+- size_t size;
+- int pagesize, page_multiplier;
+-#endif
+ #ifdef WIN32
+ MEMORYSTATUSEX memstats;
+ #endif
+@@ -240,53 +234,15 @@ sg_mem_stats *sg_get_mem_stats(){
+ mem_stat.free = uvm->pagesize * (uvm->free + uvm->inactive);
+ mem_stat.used = mem_stat.total - mem_stat.free;
+ #endif
+-
+ #if defined(OPENBSD)
+- /* The code in this section is based on the code in the OpenBSD
+- * top utility, located at src/usr.bin/top/machine.c in the
+- * OpenBSD source tree.
+- *
+- * For fun, and like OpenBSD top, we will do the multiplication
+- * converting the memory stats in pages to bytes in base 2.
+- */
+-
+- /* All memory stats in OpenBSD are returned as the number of pages.
+- * To convert this into the number of bytes we need to know the
+- * page size on this system.
+- */
+- pagesize = sysconf(_SC_PAGESIZE);
+-
+- /* The pagesize gives us the base 10 multiplier, so we need to work
+- * out what the base 2 multiplier is. This means dividing
+- * pagesize by 2 until we reach unity, and counting the number of
+- * divisions required.
+- */
+- page_multiplier = 0;
+-
+- while (pagesize > 1) {
+- page_multiplier++;
+- pagesize >>= 1;
+- }
+-
+- /* We can now ret the the raw VM stats (in pages) using the
+- * sysctl interface.
+- */
+- mib[0] = CTL_VM;
+- mib[1] = VM_METER;
+- size = sizeof(vmtotal);
+-
+- if (sysctl(mib, 2, &vmtotal, &size, NULL, 0) < 0) {
+- bzero(&vmtotal, sizeof(vmtotal));
+- sg_set_error_with_errno(SG_ERROR_SYSCTL, "CTL_VM.VM_METER");
++ if ((uvm = sg_get_uvme
vmmeter.h, VM_UVMEXP & breakage
Thanks to naddy, here's the list of ports that will break when
will stop to include . I'd
appreciate if the maintainers of the ports below could have a
look and fix their ports.
If the application is only using the VM_METER sysctl(3) to get
memory statics, then it should be converted to the VM_UVMEXP
one. See the attached diff for examples.
If the application also uses the process stats from the "struct
vmtotal", then you should also include . You can
have a look at usr.sbin/vmstat/vmstat.c if you want an example.
Attached are three diffs for the following ports, I'm looking
for reviews/oks.
databases/pg_top
sysutils/libstatgrab
x11/xfce4/xfce4-taskmanager
These are the remaining ports to fix:
lang/erlang
sysutils/monit
sysutils/symon
sysutils/toprump
sysutils/wmmon
sysutils/xstatbar
telephony/asterisk
www/chromium
x11/kde/base3
x11/kde4/workspace
And here's the diff that I'd like to commit but breaks these ports:
Index: sys/arch/amd64/amd64/fpu.c
===
RCS file: /home/ncvs/src/sys/arch/amd64/amd64/fpu.c,v
retrieving revision 1.28
diff -u -p -r1.28 fpu.c
--- sys/arch/amd64/amd64/fpu.c 29 Mar 2014 18:09:28 - 1.28
+++ sys/arch/amd64/amd64/fpu.c 8 Apr 2014 14:06:24 -
@@ -40,7 +40,6 @@
#include
#include
#include
-#include
#include
#include
Index: sys/kern/kern_fork.c
===
RCS file: /home/ncvs/src/sys/kern/kern_fork.c,v
retrieving revision 1.162
diff -u -p -r1.162 kern_fork.c
--- sys/kern/kern_fork.c30 Mar 2014 21:54:48 - 1.162
+++ sys/kern/kern_fork.c8 Apr 2014 14:06:24 -
@@ -48,6 +48,7 @@
#include
#include
#include
+#include
#include
#include
#include
Index: sys/miscfs/procfs/procfs_vfsops.c
===
RCS file: /home/ncvs/src/sys/miscfs/procfs/procfs_vfsops.c,v
retrieving revision 1.29
diff -u -p -r1.29 procfs_vfsops.c
--- sys/miscfs/procfs/procfs_vfsops.c 15 Apr 2013 15:32:19 - 1.29
+++ sys/miscfs/procfs/procfs_vfsops.c 8 Apr 2014 14:06:24 -
@@ -169,13 +169,10 @@ procfs_start(struct mount *mp, int flags
int
procfs_statfs(struct mount *mp, struct statfs *sbp, struct proc *p)
{
- struct vmtotal vmtotals;
-
- uvm_total(&vmtotals);
sbp->f_bsize = PAGE_SIZE;
sbp->f_iosize = PAGE_SIZE;
- sbp->f_blocks = vmtotals.t_vm;
- sbp->f_bfree = vmtotals.t_vm - vmtotals.t_avm;
+ sbp->f_blocks = uvmexp.npages - uvmexp.free + uvmexp.swpginuse;
+ sbp->f_bfree = uvmexp.npages - uvmexp.free - uvmexp.active;
sbp->f_bavail = 0;
sbp->f_files = maxprocess; /* approx */
sbp->f_ffree = maxprocess - nprocesses; /* approx */
Index: sys/uvm/uvm_extern.h
===
RCS file: /home/ncvs/src/sys/uvm/uvm_extern.h,v
retrieving revision 1.112
diff -u -p -r1.112 uvm_extern.h
--- sys/uvm/uvm_extern.h3 Apr 2014 21:40:10 - 1.112
+++ sys/uvm/uvm_extern.h8 Apr 2014 14:06:24 -
@@ -394,7 +394,6 @@ extern struct uvmexp uvmexp;
/*
* Finally, bring in standard UVM headers.
*/
-#include
#include
#include
#include
@@ -484,7 +483,6 @@ struct loadavg;
struct proc;
struct pmap;
struct vmspace;
-struct vmtotal;
struct mount;
struct vnode;
struct core;
@@ -677,7 +675,6 @@ voiduvmspace_share(struct proc *,
str
void uvm_meter(void);
intuvm_sysctl(int *, u_int, void *, size_t *,
void *, size_t, struct proc *);
-void uvm_total(struct vmtotal *);
/* uvm_mmap.c */
intuvm_mmap(vm_map_t, vaddr_t *, vsize_t,
Index: sys/uvm/uvm_meter.c
===
RCS file: /home/ncvs/src/sys/uvm/uvm_meter.c,v
retrieving revision 1.30
diff -u -p -r1.30 uvm_meter.c
--- sys/uvm/uvm_meter.c 23 Mar 2013 16:12:31 - 1.30
+++ sys/uvm/uvm_meter.c 8 Apr 2014 14:06:24 -
@@ -45,7 +45,7 @@
#include
#include
#include
-#include
+#include
#include
#include
@@ -79,11 +79,8 @@ static fixpt_t cexp[3] = {
0.9944598480048967 * FSCALE,/* exp(-1/180) */
};
-/*
- * prototypes
- */
-
static void uvm_loadav(struct loadavg *);
+void uvm_total(struct vmtotal *);
/*
* uvm_meter: calculate load average and wake up the swapper (if needed)
Index: Makefile
===
RCS file: /home/ncvs/ports/sysutils/libstatgrab/Makefile,v
retrieving revision 1.9
diff -u -p -r1.9 Makefile
--- Makefile21 Mar 2013 08:48:54 - 1.9
+++ Makefile10 Apr 2014 12:44:14 -
@@ -4,6 +4,7 @@ COMMENT=system statistics gathering lib
DISTNAME= li
Re: Update: devel/libusb1
David, On 13/03/14(Thu) 17:04, David Coppa wrote: > On Thu, Mar 13, 2014 at 4:45 PM, Stuart Henderson wrote: > > On 2014/03/13 09:20, David Coppa wrote: > >> And, please, tell me how to further debug this issue... > > > > See libusb1's Makefile; build libusb1 with DEBUG defined and > > you get some additional logging. > > Here are the logs with DEBUG enabled. There's nothing obvious in the log and no error from the libusb point of view. I tried to look at the pcsc-lite source and spend some time without understanding... of course, the USB driver are in the ccid ports. All I can see is that the last completed transfers have a different length with the new release (10 instead of 21 and 13), what does that mean? Maybe an error, I don't know. M.
Update: sysutils/usbutils
Here's an update to the last usbutils package, ok?
Index: Makefile
===
RCS file: /home/ncvs/ports/sysutils/usbutils/Makefile,v
retrieving revision 1.6
diff -u -p -r1.6 Makefile
--- Makefile21 Mar 2013 08:48:54 - 1.6
+++ Makefile11 Mar 2014 15:43:13 -
@@ -2,12 +2,10 @@
COMMENT = utility used to display information about USB devices
-VERSION = 005
+VERSION = 007
DISTNAME = usbutils_${VERSION}.orig
PKGNAME = usbutils-${VERSION}
-REVISION = 1
-
CATEGORIES = sysutils
HOMEPAGE = http://www.linux-usb.org
@@ -17,12 +15,13 @@ MAINTAINER = Martin Pieuchot http://ftp.de.debian.org/debian/pool/main/u/usbutils/
+EXTRACT_SUFX = .tar.xz
WANTLIB = c usb-1.0 z
LIB_DEPENDS = devel/libusb1
AUTOMAKE_VERSION = 1.11
-AUTOCONF_VERSION = 2.68
+AUTOCONF_VERSION = 2.69
BUILD_DEPENDS += ${MODGNU_AUTOCONF_DEPENDS} \
${MODGNU_AUTOMAKE_DEPENDS}
Index: distinfo
===
RCS file: /home/ncvs/ports/sysutils/usbutils/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo15 May 2012 13:44:52 - 1.1.1.1
+++ distinfo11 Mar 2014 15:43:13 -
@@ -1,5 +1,2 @@
-MD5 (usbutils_005.orig.tar.gz) = LpkCZdRy4vbwZiNW1lRoOw==
-RMD160 (usbutils_005.orig.tar.gz) = UP7nRFbOxj1u4l5pQxATWm5k1nQ=
-SHA1 (usbutils_005.orig.tar.gz) = 5nri5uCxT7d4srV2yPswK/pD0K8=
-SHA256 (usbutils_005.orig.tar.gz) =
KlM+WNWaojWGEOQPgHQowxiYSEWGT5QnukjSs06dXRc=
-SIZE (usbutils_005.orig.tar.gz) = 234751
+SHA256 (usbutils_007.orig.tar.xz) =
dZOgFyS7wP2f5I5ivHIc62HHZlTx17Ixs8Zfbfu676Q=
+SIZE (usbutils_007.orig.tar.xz) = 418264
Index: patches/patch-Makefile_am
===
RCS file: /home/ncvs/ports/sysutils/usbutils/patches/patch-Makefile_am,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-Makefile_am
--- patches/patch-Makefile_am 15 May 2012 13:44:52 - 1.1.1.1
+++ patches/patch-Makefile_am 11 Mar 2014 15:43:13 -
@@ -2,8 +2,8 @@ $OpenBSD: patch-Makefile_am,v 1.1.1.1 20
No usbhid-dump(1) nor 'lsusb -t' option under OpenBSD
Makefile.am.orig Sun Oct 30 20:35:50 2011
-+++ Makefile.amTue May 15 14:28:34 2012
+--- Makefile.am.orig Wed Dec 12 00:01:00 2012
Makefile.amTue Mar 11 16:28:10 2014
@@ -1,6 +1,3 @@
-SUBDIRS = \
- usbhid-dump
@@ -11,11 +11,11 @@ No usbhid-dump(1) nor 'lsusb -t' option
AM_LDFLAGS = \
-Wl,--as-needed
-@@ -19,7 +16,6 @@ bin_SCRIPTS = \
+@@ -17,7 +14,6 @@ bin_SCRIPTS = \
lsusb_SOURCES = \
lsusb.c lsusb.h \
- lsusb-t.c \
list.h \
names.c names.h \
- devtree.c devtree.h \
+ usbmisc.c usbmisc.h
Index: patches/patch-lsusb_c
===
RCS file: /home/ncvs/ports/sysutils/usbutils/patches/patch-lsusb_c,v
retrieving revision 1.2
diff -u -p -r1.2 patch-lsusb_c
--- patches/patch-lsusb_c 16 Jul 2012 20:17:03 - 1.2
+++ patches/patch-lsusb_c 11 Mar 2014 15:43:13 -
@@ -2,48 +2,46 @@ $OpenBSD: patch-lsusb_c,v 1.2 2012/07/16
Disable unsupported options.
lsusb.c.orig Sun Oct 30 20:35:50 2011
-+++ lsusb.cMon Jul 16 18:05:14 2012
-@@ -3902,6 +3902,7 @@ void devtree_devdisconnect(struct usbdevnode *dev)
- dev = dev; /* reduce compiler warnings */
- }
+--- lsusb.c.orig Fri Jun 7 01:13:43 2013
lsusb.cTue Mar 11 16:42:07 2014
+@@ -3856,6 +3856,7 @@ static void dumpdev(libusb_device *dev)
+
+ /* -- */
+#ifndef __OpenBSD__
- static int treedump(void)
+ static int dump_one_device(libusb_context *ctx, const char *path)
{
- int fd;
-@@ -3919,6 +3920,7 @@ static int treedump(void)
- devtree_dump(verblevel);
+ libusb_device *dev;
+@@ -3877,6 +3878,7 @@ static int dump_one_device(libusb_context *ctx, const
+ dumpdev(dev);
return 0;
}
+#endif
- /* -- */
+ static int list_devices(libusb_context *ctx, int busnum, int devnum, int
vendorid, int productid)
+ {
+@@ -3947,7 +3949,7 @@ int main(int argc, char *argv[])
-@@ -3937,7 +3939,7 @@ int main(int argc, char *argv[])
- char *cp;
- int status;
+ setlocale(LC_CTYPE, "");
-- while ((c = getopt_long(argc, argv, "D:vxtP:p:s:d:V",
-+ while ((c = getopt_long(argc, argv, "D:vxP:p:s:d:V",
+- while ((c = getopt_long(argc, argv, "D:vtP:p:s:d:Vh",
++ while ((c = getopt_long(argc, argv, "vP:p:s:d:Vh",
long_options, NULL)) != EOF) {
Update: devel/libusb1
Diff below is a major update to the last libusb (previously libusbx)
release. Our backend has been updated upstream and this update
includes various fixes. But since *a lot* of development happened
since our previous packaged version, I'd like to hear from you, USB
users, to know if this new version introduces any regression.
Comments and oks are also welcome :)
Martin
Index: Makefile
===
RCS file: /home/ncvs/ports/devel/libusb1/Makefile,v
retrieving revision 1.22
diff -u -p -r1.22 Makefile
--- Makefile7 Aug 2013 15:37:09 - 1.22
+++ Makefile11 Mar 2014 15:22:10 -
@@ -2,15 +2,14 @@
COMMENT = library for USB device access from userspace
-VERSION = 1.0.9
+VERSION = 1.0.18
DISTNAME = libusb-${VERSION}
PKGNAME = libusb1-${VERSION}
-REVISION = 8
-SHARED_LIBS += usb-1.0 1.0 # 1.0
+SHARED_LIBS += usb-1.0 1.1 # 1.0
CATEGORIES = devel
-HOMEPAGE = http://www.libusb.org
+HOMEPAGE = http://www.libusb.info
MAINTAINER = Martin Pieuchot
Index: distinfo
===
RCS file: /home/ncvs/ports/devel/libusb1/distinfo,v
retrieving revision 1.2
diff -u -p -r1.2 distinfo
--- distinfo22 Apr 2012 09:06:37 - 1.2
+++ distinfo11 Mar 2014 15:22:10 -
@@ -1,5 +1,2 @@
-MD5 (libusb-1.0.9.tar.bz2) = f1oCN1rZYNTjOm2ufWPPyw==
-RMD160 (libusb-1.0.9.tar.bz2) = VBBXD4zjF4METrBK66rnreOyGmE=
-SHA1 (libusb-1.0.9.tar.bz2) = AlWC/y9iFuLbwmEK4WsuBz4bM0Y=
-SHA256 (libusb-1.0.9.tar.bz2) = 6SDu3C0GsJYGYRyZ7HMEQTxnhMum4zko54JD0yMZX5s=
-SIZE (libusb-1.0.9.tar.bz2) = 421971
+SHA256 (libusb-1.0.18.tar.bz2) = xz9c7EWl3pRBjaThUbcjKVhXGSaYSs+5vOArlCToNyA=
+SIZE (libusb-1.0.18.tar.bz2) = 502160
Index: patches/patch-libusb_core_c
===
RCS file: /home/ncvs/ports/devel/libusb1/patches/patch-libusb_core_c,v
retrieving revision 1.3
diff -u -p -r1.3 patch-libusb_core_c
--- patches/patch-libusb_core_c 19 Jun 2012 21:10:59 - 1.3
+++ patches/patch-libusb_core_c 11 Mar 2014 15:22:10 -
@@ -4,9 +4,9 @@ On OpenBSD USB controllers are shown as
itinial limit too small. On a recent machine this value is almost
always exceeded, so bump it.
libusb/core.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/core.c Tue Jun 12 15:48:05 2012
-@@ -468,7 +468,7 @@ libusb_free_device_list(list, 1);
+--- libusb/core.c.orig Sun Jan 26 00:06:33 2014
libusb/core.c Mon Jan 27 10:08:37 2014
+@@ -448,7 +448,7 @@ libusb_free_device_list(list, 1);
* which grows when required. it can be freed once discovery has completed,
* eliminating the need for a list node in the libusb_device structure
* itself. */
Index: patches/patch-libusb_os_openbsd_usb_c
===
RCS file: patches/patch-libusb_os_openbsd_usb_c
diff -N patches/patch-libusb_os_openbsd_usb_c
--- patches/patch-libusb_os_openbsd_usb_c 22 Apr 2013 08:37:20 -
1.5
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,570 +0,0 @@
-$OpenBSD: patch-libusb_os_openbsd_usb_c,v 1.5 2013/04/22 08:37:20 mpi Exp $
-
-Add support for non ugen(4) attached devices through usb(4) buses.
-
libusb/os/openbsd_usb.c.orig Fri Apr 20 08:44:27 2012
-+++ libusb/os/openbsd_usb.cSun Apr 21 12:39:40 2013
-@@ -1,5 +1,5 @@
- /*
-- * Copyright (c) 2011 Martin Pieuchot
-+ * Copyright (c) 2011-2013 Martin Pieuchot
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
-@@ -32,8 +32,8 @@
- #include "libusbi.h"
-
- struct device_priv {
-- char devnode[16];
-- int fd;
-+ char *devname; /* name of the ugen(4) node */
-+ int fd; /* device file descriptor */
-
- unsigned char *cdesc; /* active config descriptor */
- usb_device_descriptor_t ddesc; /* usb device descriptor */
-@@ -82,11 +82,14 @@ static int obsd_clock_gettime(int, struct timespec *);
- * Private functions
- */
- static int _errno_to_libusb(int);
--static int _cache_active_config_descriptor(struct libusb_device *, int);
-+static int _cache_active_config_descriptor(struct libusb_device *);
- static int _sync_control_transfer(struct usbi_transfer *);
- static int _sync_gen_transfer(struct usbi_transfer *);
- static int _access_endpoint(struct libusb_transfer *);
-
-+static int _bus_open(int);
-+
-+
- const struct usbi_os_backend openbsd_backend = {
- "Synchronous OpenBSD backend",
- NULL, /* init() */
-@@ -128,75 +131,105 @@ const struct usbi_os_backend openbsd_backend = {
- 0,
Re: Epson ESC/P-R driver
Hello Maurice, On 29/01/14(Wed) 18:55, Maurice McCarthy wrote: > On Wed, Jan 29, 2014 at 06:31:55PM +0100 or thereabouts, Antoine Jacoutot > wrote: > [...] > > Thank you Antoine, I've never actually tried to configure cups through ulpt > after reading your pkg-readme. But after seeing this is dmesg: > > $ dmesg | grep ugen > ugen0 at uhub0 port 4 configuration 1 "EPSON USB2.0 MFP(Hi-Speed)" rev > 2.00/1.00 addr 3 > $ dmesg | grep ulpt > ulpt0 at uhub0 port 4 configuration 1 interface 1 "EPSON USB2.0 > MFP(Hi-Speed)" rev 2.00/1.00 addr 3 > ulpt0: using bi-directional mode > > ... it made me wonder whether ulpt is enabled in the default 5.4 kernel or > not. The reason behind enabling ulpt(4) in the default kernel is that you might be able to use your printer without 3rd party programs. ugen(4) is here to provide an interface to use userland drivers, and that's what CUPS is all about, it bundles a lot of drivers for printers. With some coding efforts, it should be possible to use both at the same time, but somebody has to do it. > $ sudo usbdevs -vd > Controller /dev/usb0: > addr 1: high speed, self powered, config 1, EHCI root hub(0x), > ATI(0x1002), rev 1.00 > uhub0 > port 1 addr 2: high speed, self powered, config 1, USB2.0 Hub(0x0605), > Genesys Logic(0x05e3), rev 6.0b >uhub10 > port 1 powered > port 2 powered > port 3 powered > port 4 powered > port 2 powered > port 3 powered > port 4 addr 3: high speed, self powered, config 1, USB2.0 > MFP(Hi-Speed)(0x0849), EPSON(0x04b8), rev 1.00, iSerialNumber > 4B4C4E4B3032373907 >ulpt0 >umass0 >ugen0 Here comes all the fun :) You still have ulpt0 attached to your device, and that is certainly why you get an USB write error: > [...] > D [29/Jan/2014:18:40:39 +] [Job 2] libusb write operation returned > ff9d. Could you try disabling ulpt(4) in your kernel like it is said in cups' README? Then make sure only ugen(4) and umass(4) are attached to your device after reboot, check your /dev/ugen* permissions and try again. M.
Re: xkb-qwerty-fr
On 01/10/13(Tue) 12:03, Stuart Henderson wrote:
> On 2013/10/01 12:20, Jérémie Courrèges-Anglas wrote:
> >
> > Here's an old mail...
> >
> > Tristan Le Guern writes:
> >
> > > Hi,
> >
> > Hi Tristan,
> >
> > > This is an updated version of my previous submission of xbk-qwerty-fr.
> > > The keymap file is now installed under
> > > "/usr/local/share/X11/xkb/symbols/us_qwerty-fr" and then @sample'd to
> > > "/usr/X11R6/share/X11/xkb/symbols/us_qwerty-fr".
> > >
> > > Ok?
> >
> > Hmm, almost. I wouldn't use ${PREFIX}/share/doc/${PKGNAME}, but
> > ${PREFIX}/share/doc/xkb-qwerty-fr/ in do-install. This way the path
> > doesn't change with updates, and the make update-plist won't try to use
> > ${FULLPKGNAME} (which is wrong). But I can fix this before importing.
> >
> > I'd like to hear at least another porter about the /usr/X11R6 handling.
> > Does @sample make sense? That's the only way we found for setxkbmap to
> > work.
>
> I'm not totally keen on a port installing files into /usr/X11R6, does
> a symlink work? If so, we could ask people to add that themselves (a quick
> 1-line MESSAGE would probably be enough for this).
IMHO the best solution would be to make it part of xkb, this way it
will be directly available in xenocara.
Did you or the original author tried to submit it upstream? I guess it
should go in xkeyboard-config.
Re: pstree(1) dumps core on macppc/current
On 18/08/13(Sun) 19:40, Jan Stary wrote: > On Aug 18 19:38:29, h...@stare.cz wrote: > > After I upgraded a MacMini to current/macppc, > > this is what happens when I try to run pstree: > > > > $ pstree > > Bad system call (core dumped) > > > > I made a ktrace of it (attached). > > It looks like the ps(1) child is exiting prematurely. > > Could this be related to the recent ABI change? Yes. > > > > This installation of pstree was done from the port; > > the prebuilt package (July 20) is too old for current. > > Maybe I should say explicitly that I removed and reinstalled all packages, > as per the current.html note about the recent ABI change. powerpc snapshots still predate the ABI change, just wait for a new one ;) Martin
Re: devel/libdnet: fix for upcoming SIOCSIF{NET,DST}MASK change
On 10/05/13(Fri) 12:59, Giovanni Bechis wrote: > On 05/10/13 12:49, Stuart Henderson wrote: > > I doubt this is actually used in nmap, but it makes sense to > > keep it in-sync. OK. > > > actually it is never called. Event if it's not called, it think that it makes sense to commit it as long as nmap use its on copy of libdnet. It's some kind of documentation and to make sure it wont break if it starts to be used. Obviously ok mpi@. M.
devel/libdnet: fix for upcoming SIOCSIF{NET,DST}MASK change
The diff below prevents libdnet to issue an unneeded SIOCSIFNETMASK
ioctl(2) prior to SIOCSIFADDR.
I'd like to commit this change because I'm about to change the behavior
of the SIOCSIFNETMASK and SIOCSIFDSTMASK ioctl(2)s in such way that they
will fail if the interface has no previous configured IPv4 address.
This port is the only one that I've seen that makes an incorrect use of
this ioctl(2)s, if you are aware of another, please let me know.
Ok?
Index: Makefile
===
RCS file: /home/ncvs/ports/net/libdnet/Makefile,v
retrieving revision 1.36
diff -u -p -r1.36 Makefile
--- Makefile21 Mar 2013 08:46:34 - 1.36
+++ Makefile8 May 2013 14:38:10 -
@@ -7,7 +7,7 @@ MODPY_EGG_VERSION= 1.12
DISTNAME= libdnet-${MODPY_EGG_VERSION}
PKGNAME-main= ${DISTNAME}
PKGNAME-python= py-${DISTNAME}
-REVISION= 6
+REVISION= 7
SHARED_LIBS= dnet1.0
Index: patches/patch-src_intf_c
===
RCS file: /home/ncvs/ports/net/libdnet/patches/patch-src_intf_c,v
retrieving revision 1.1
diff -u -p -r1.1 patch-src_intf_c
--- patches/patch-src_intf_c16 Nov 2012 17:05:13 - 1.1
+++ patches/patch-src_intf_c8 May 2013 14:38:10 -
@@ -1,6 +1,10 @@
$OpenBSD: patch-src_intf_c,v 1.1 2012/11/16 17:05:13 sthen Exp $
src/intf.c.origWed Nov 14 04:15:30 2012
-+++ src/intf.c Wed Nov 14 04:16:33 2012
+
+Do not issue an SIOCSIFNETMASK before SIOCSIFADDR it is not
+needed even if the XXX says otherwise.
+
+--- src/intf.c.origMon Jan 9 08:09:49 2006
src/intf.c Thu Apr 11 16:16:02 2013
@@ -66,7 +66,13 @@
/* XXX - superset of ifreq, for portable SIOC{A,D}IFADDR */
struct dnet_ifaliasreq {
@@ -16,3 +20,12 @@ $OpenBSD: patch-src_intf_c,v 1.1 2012/11
struct sockaddr ifra_brdaddr;
struct sockaddr ifra_mask;
int ifra_cookie;/* XXX - IRIX!@#$ */
+@@ -281,7 +287,7 @@ intf_set(intf_t *intf, const struct intf_entry *entry)
+ }
+ /* Set interface address. */
+ if (entry->intf_addr.addr_type == ADDR_TYPE_IP) {
+-#ifdef BSD
++#if defined(BSD) && !defined(__OpenBSD__)
+ /* XXX - why must this happen before SIOCSIFADDR? */
+ if (addr_btos(entry->intf_addr.addr_bits,
+ &ifr.ifr_addr) == 0) {
Re: tests invited: rtl-sdr
On 20/04/13(Sat) 13:08, Stuart Henderson wrote: > On 2013/04/20 11:24, Stuart Henderson wrote: > > ok so I have some hardware now. bulk transfers with asynchronous > > events don't work via libusb yet (mglocker has been looking at this) but > > when used in synchronous mode this is doing something which looks kind-of > > correct. > > > > I need to make some comparisons with non-OpenBSD machines to > > see if it it's really working though; I'm in a bit of a radio black > > hole and the crappy stock aerial I'm currently using don't make it > > easy to tell what's working and what isn't ;) > > > > attached tar.gz (to unpack in /usr/ports/mystuff/comms) has these 3 ports: > > > > py-rtlsdr - python library making librtlsdr functions available from python. > > includes a demo "waterfall" program (which requires newer py-matplotlib > > from openbsd-wip), dependency on matplotlib not enforced for this as it's > > mostly intended as a library. > > > > rtlsdr-scanner: scanner program making use of above library - > > this one does work with -current py-matplotlib. > > > > updated rtl-sdr port: for now, remove programs which can *only* use > > async mode. > > > so... given the limitations, this definitely works: > > http://junkpile.org/rtlscanner.png > > any OKs to import these? > > http://junkpile.org/rtl-sdr-ports.tgz I've seen by looking at the code that there's a "-S" flag to use libusb in a synchronous fashion. Does it work better when you use this flag? If so, do you think it would make any sense to make it the default on OpenBSD and add a comment about our current status of asynchronous transfer submission? M.
Re: UPDATE: pyflakes
On 12/03/13(Tue) 06:16, Benoit Lecocq wrote: > On 03/12/13 02:13, frantisek holop wrote: > >hmm, on Sat, Mar 09, 2013 at 12:13:16PM +0100, Martin Pieuchot said that > >>On 08/03/13(Fri) 23:03, frantisek holop wrote: > >>>hmm, on Fri, Mar 08, 2013 at 11:01:23PM +0100, frantisek holop said that > >>>>update to latest. > >>>>please review and commit. > >>> > >>>please note the regress depends on the new port > >>>i submitted, unittest2 > >> > >>Looks to me that you forgot the actual diff (: > >> > >>By the way, if you want to take the maintainership of this port feel > >>free. > > > >ok and ok :] > > > >actual patch included. > > > >-f > > > > The new test run fine. > Need to import the new port before committing your update. Go ahead, ok mpi@ M.
Re: UPDATE: pyflakes
On 08/03/13(Fri) 23:03, frantisek holop wrote: > hmm, on Fri, Mar 08, 2013 at 11:01:23PM +0100, frantisek holop said that > > update to latest. > > please review and commit. > > please note the regress depends on the new port > i submitted, unittest2 Looks to me that you forgot the actual diff (: By the way, if you want to take the maintainership of this port feel free. M.
Re: One more thought on the pthread patch
On 07/03/13(Thu) 11:11, Jim Razmus wrote: > Hello Martin, > > I noticed the same missing wantlib with py-yaml. So I'm questioning if > my system is screwing something up. Or if indeed, we need to add > pthread to both ports wantlib. > > Your thoughts? It looks to me that since libpython.* started depending on libpthread (I've seen the sync from r1.93 of lang/python/Makefile.inc) all the python packages depending on this lib should now also have their WANTLIB updated. Anybody have a better understanding of the situation and/or bumping session? M.
Re: massive E17 update to 0.17.0 alpha
On 09/11/12(Fri) 20:27, David Coppa wrote: > On Fri, Nov 9, 2012 at 7:29 PM, Landry Breuil wrote: > > > On ppc/ati, only composite is auto-enabled, and e produces garbled > > colors, the bg/transparent stuff is bright blue, see > > http://rhaalovely.net/~landry/shared/e17-openbsd-ppc-shot-2012-11-09_18-49-45.jpg > > Probably a littleendian issue. > > Probably: I have the same with openbox on sparc64. What graphic card are you using? It looks like the same issue I ran into with my radeon r200 and clutter (gnome-shell) on macppc and i386. M.
New: sysutils/usbutils
I'd like to import this port, it provides the lsusb(1) utility that has no real interest by itself compared to usbdevs(1), but it is really useful for testing the libusb and/or getting strings of some USB devices. Description: Utility used to display information about USB devices. Any objection? Ok? Martin usbutils.tgz Description: application/tar-gz
Update: libusb-compat to 0.1.4
Update to 0.1.4 and take maintainership to keep it in sync with devel/libusb1. This is a really small bugfix and documentation release. Ok? Martin Index: Makefile === RCS file: /cvs/ports/devel/libusb-compat/Makefile,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 Makefile --- Makefile13 Nov 2011 13:49:50 - 1.1.1.1 +++ Makefile9 May 2012 09:20:22 - @@ -2,7 +2,7 @@ COMMENT= libusb-0.1 compatibility layer for libusb1 -DISTNAME= libusb-compat-0.1.3 +DISTNAME= libusb-compat-0.1.4 EXTRACT_SUFX= .tar.bz2 SHARED_LIBS += usb 10.0 # 8.4 @@ -12,6 +12,8 @@ MODGNU_SHARED_LIBS= usb '-export-dynamic CATEGORIES=devel HOMEPAGE= http://libusb.sourceforge.net/ + +MAINTAINER= Martin Pieuchot # LGPL PERMIT_PACKAGE_CDROM= Yes Index: distinfo === RCS file: /cvs/ports/devel/libusb-compat/distinfo,v retrieving revision 1.1.1.1 diff -u -p -r1.1.1.1 distinfo --- distinfo13 Nov 2011 13:49:50 - 1.1.1.1 +++ distinfo24 Apr 2012 07:49:14 - @@ -1,5 +1,5 @@ -MD5 (libusb-compat-0.1.3.tar.bz2) = VwrC6ghbgNH3Tdx8apPA6w== -RMD160 (libusb-compat-0.1.3.tar.bz2) = 0A4AuflzPvxD605EuEKYVA6JZpE= -SHA1 (libusb-compat-0.1.3.tar.bz2) = 1XENW8S2fFNE53lHW3YWjHzMXmk= -SHA256 (libusb-compat-0.1.3.tar.bz2) = pZCgO2GIAw7hyhoK9VaF/N4AXKgHuWOXD4Ob53YDHZQ= -SIZE (libusb-compat-0.1.3.tar.bz2) = 251218 +MD5 (libusb-compat-0.1.4.tar.bz2) = LKUh//rdDCj98XTm7HOGWw== +RMD160 (libusb-compat-0.1.4.tar.bz2) = hKeqOilKYGWDj8tGSTWSa2WeT+4= +SHA1 (libusb-compat-0.1.4.tar.bz2) = /cHfb1z3tx3np0KSruoaoqOVUq4= +SHA256 (libusb-compat-0.1.4.tar.bz2) = 7VvdFgx7Ae92f7kxqBtFT0YibR4s9YUCztdY0+Wp/cQ= +SIZE (libusb-compat-0.1.4.tar.bz2) = 242379
Update: devel/libusb1 1.0.9
The diff below updates devel/libusb1 to the recent 1.0.9 release.
All OpenBSD specific code has been commited upsteam and I don't expect
any regression but if you are using any application relying on the
libusb, please test this update.
Comments and oks welcome.
Martin
Index: Makefile
===
RCS file: /cvs/ports/devel/libusb1/Makefile,v
retrieving revision 1.7
diff -u -p -r1.7 Makefile
--- Makefile8 Apr 2012 13:14:54 - 1.7
+++ Makefile20 Apr 2012 12:44:05 -
@@ -2,11 +2,10 @@
COMMENT = library for USB device access from userspace
-VERSION = 1.0.8
+VERSION = 1.0.9
DISTNAME = libusb-${VERSION}
PKGNAME = libusb1-${VERSION}
-REVISION = 4
-SHARED_LIBS += usb-1.0 0.0 # 0.0
+SHARED_LIBS += usb-1.0 1.0 # 1.0
CATEGORIES = devel
@@ -27,29 +26,7 @@ EXTRACT_SUFX = .tar.bz2
BUILD_DEPENDS =devel/libtool
USE_LIBTOOL = Yes
-AUTOCONF_VERSION = 2.65
-AUTOMAKE_VERSION = 1.11
-CONFIGURE_STYLE = automake autoconf
-CONFIGURE_ARGS += ${CONFIGURE_SHARED}
-
-.ifdef DEBUG
-CONFIGURE_ARGS += --enable-log \
- --enable-debug-log \
- --enable-examples-build
-.endif
-
-AUTO_ENV= AUTOCONF_VERSION=${AUTOCONF_VERSION} \
- AUTOMAKE_VERSION=${AUTOMAKE_VERSION}
-
-post-extract:
- cp ${FILESDIR}/openbsd_ugen.c ${WRKSRC}/libusb/os/
-
-post-patch:
- @echo "Running aclocal-${AUTOMAKE_VERSION} in ${WRKSRC}"
- @cd ${WRKSRC}; ${AUTO_ENV} aclocal
-
-pre-configure:
- @echo "Running automake-${AUTOMAKE_VERSION} in ${WRKSRC}"
- @cd ${WRKSRC}; ${AUTO_ENV} automake --foreign
+CONFIGURE_STYLE = gnu
+USE_GMAKE =Yes
.include
Index: distinfo
===
RCS file: /cvs/ports/devel/libusb1/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo2 Nov 2011 14:31:56 - 1.1.1.1
+++ distinfo20 Apr 2012 12:37:44 -
@@ -1,5 +1,5 @@
-MD5 (libusb-1.0.8.tar.bz2) = N9NObqpppLZFoZ/0ymPO7w==
-RMD160 (libusb-1.0.8.tar.bz2) = WkE6d+JAoVSA7xQ31fU5O6266YU=
-SHA1 (libusb-1.0.8.tar.bz2) = VIQ5eGD3Ccm1FhHSJIGfjtWZQGM=
-SHA256 (libusb-1.0.8.tar.bz2) = IdDTpXEPf0IRxZUQLGuezLQkNaF6T1vSw/QWarG626k=
-SIZE (libusb-1.0.8.tar.bz2) = 331820
+MD5 (libusb-1.0.9.tar.bz2) = f1oCN1rZYNTjOm2ufWPPyw==
+RMD160 (libusb-1.0.9.tar.bz2) = VBBXD4zjF4METrBK66rnreOyGmE=
+SHA1 (libusb-1.0.9.tar.bz2) = AlWC/y9iFuLbwmEK4WsuBz4bM0Y=
+SHA256 (libusb-1.0.9.tar.bz2) = 6SDu3C0GsJYGYRyZ7HMEQTxnhMum4zko54JD0yMZX5s=
+SIZE (libusb-1.0.9.tar.bz2) = 421971
Index: files/openbsd_ugen.c
===
RCS file: files/openbsd_ugen.c
diff -N files/openbsd_ugen.c
--- files/openbsd_ugen.c8 Apr 2012 13:14:54 - 1.6
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,715 +0,0 @@
-/* $OpenBSD: openbsd_ugen.c,v 1.6 2012/04/08 13:14:54 mpi Exp $*/
-/*
- * Copyright (c) 2011 Martin Pieuchot
- *
- * Permission to use, copy, modify, and distribute this software for any
- * purpose with or without fee is hereby granted, provided that the above
- * copyright notice and this permission notice appear in all copies.
- *
- * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
- * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
- * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
- * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
- * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
- * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
- * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
- */
-
-#include
-#include
-
-#include
-#include
-#include
-#include
-#include
-#include
-
-#include
-
-#include "libusb.h"
-#include "libusbi.h"
-
-struct device_priv {
- char devnode[16];
- int fd;
-
- unsigned char *cdesc; /* active config descriptor */
- usb_device_descriptor_t ddesc; /* usb device descriptor */
-};
-
-struct handle_priv {
- int pipe[2];/* for event notification */
- int endpoints[USB_MAX_ENDPOINTS];
-};
-
-/*
- * Backend functions
- */
-int ugen_get_device_list(struct libusb_context *, struct discovered_devs **);
-int ugen_open(struct libusb_device_handle *);
-void ugen_close(struct libusb_device_handle *);
-
-int ugen_get_device_descriptor(struct libusb_device *, unsigned char *, int *);
-int ugen_get_active_config_descriptor(struct libusb_device *, unsigned char *,
-size_t, int *);
-int ugen_get_config_descriptor(struct libusb_device *,
Update: py-setproctitle
Update py-setproctitle to 1.1.5 and add a python3 flavor.
Ok?
Martin
Index: Makefile
===
RCS file: /cvs/ports/devel/py-setproctitle/Makefile,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 Makefile
--- Makefile19 Dec 2011 10:37:41 - 1.1.1.1
+++ Makefile8 Apr 2012 14:00:54 -
@@ -2,7 +2,7 @@
COMMENT = library for getting/setting process names
-MODPY_EGG_VERSION= 1.1.3
+MODPY_EGG_VERSION= 1.1.5
DISTNAME = setproctitle-${MODPY_EGG_VERSION}
PKGNAME = py-${DISTNAME}
@@ -22,6 +22,9 @@ MASTER_SITES= ${MASTER_SITE_PYPI:=s/set
MODULES = lang/python
+FLAVORS = python3
+FLAVOR ?=
+
WANTLIB += ${MODPY_WANTLIB}
BUILD_DIR =build/lib.${MODPY_VERSION}
@@ -29,10 +32,10 @@ MODPY_DISTUTILS_BUILDARGS = --build-lib
REGRESS_DEPENDS = devel/gmake
-MAKE_ENV +=PYTHON="${MODPY_BIN}" PYCONFIG="${MODPY_BIN}-config"
+MAKE_ENV +=PYTHON="${MODPY_BIN}" PYCONFIG="${MODPY_BIN}-config" \
+ PY2TO3="2to3-${MODPY_VERSION}"
do-regress:
- cd ${WRKSRC} && gmake ${MAKE_ENV} tests/pyrun && \
- PYTHONPATH=${BUILD_DIR} ${MODPY_BIN} tests/setproctitle_test.py
+ cd ${WRKSRC} && gmake ${MAKE_ENV} check
.include
Index: distinfo
===
RCS file: /cvs/ports/devel/py-setproctitle/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo19 Dec 2011 10:37:41 - 1.1.1.1
+++ distinfo8 Apr 2012 13:20:29 -
@@ -1,5 +1,5 @@
-MD5 (setproctitle-1.1.3.tar.gz) = ZAKFIYUNxoxbkNRTd94/XA==
-RMD160 (setproctitle-1.1.3.tar.gz) = ZGFBQZmOS1MVB4d4p1b2JK4g5CY=
-SHA1 (setproctitle-1.1.3.tar.gz) = efUzcaNwfXMojxcDjeepfuAzuFY=
-SHA256 (setproctitle-1.1.3.tar.gz) =
Svl+sBov67gQoItG9/U7Rgwe2mvroHiVwT6+dW9CIwY=
-SIZE (setproctitle-1.1.3.tar.gz) = 19286
+MD5 (setproctitle-1.1.5.tar.gz) = jcI2wjcnpKygMfzSm3+Kcw==
+RMD160 (setproctitle-1.1.5.tar.gz) = ycs2DpTxwBqma4SnvfhR4kpBjbI=
+SHA1 (setproctitle-1.1.5.tar.gz) = AH4b9Av5URvaLS+J44KX/pY0HvQ=
+SHA256 (setproctitle-1.1.5.tar.gz) =
xDRkVBKlVvckyP/lVaLptYfNESdCamHJh5SvI0eD9Fg=
+SIZE (setproctitle-1.1.5.tar.gz) = 22382
Index: patches/patch-Makefile
===
RCS file: patches/patch-Makefile
diff -N patches/patch-Makefile
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile 8 Apr 2012 13:59:40 -
@@ -0,0 +1,27 @@
+$OpenBSD$
+
+Copy missing file for python3 flavor and remove non supported cp option.
+
+--- Makefile.orig Mon Feb 6 13:49:36 2012
Makefile Sun Apr 8 15:58:50 2012
+@@ -29,7 +29,7 @@
+ check: build tests/pyrun2
+ PYTHONPATH=`pwd`/$(BUILD_DIR):$$PYTHONPATH \
+ ROOT_PATH=$(ROOT_PATH) \
+- $(PYTHON) `which nosetests` -v -s -w tests
++ $(PYTHON) tests/setproctitle_test.py -v
+
+ tests/pyrun2: tests/pyrun.c
+ $(CC) $(PYINC) -o $@ $< $(PYLIB)
+@@ -48,9 +48,10 @@
+ $(MKDIR) py3
+ $(MKDIR) py3/src
+ $(MKDIR) py3/tests
+- for f in `grep -v "#" MANIFEST`; do cp -v $$f py3/$$f; done
++ for f in `grep -v "#" MANIFEST`; do cp $$f py3/$$f; done
+ # setup.py should be executable with python3 as distribute
+ # currenlty doesn't seem to try to convert it
++ cp tests/setproctitle_test.py py3/tests
+ $(PY2TO3) -w --no-diffs py3/tests
+
+ tests/pyrun3: tests/pyrun.c
Index: patches/patch-tests_pyrun_c
===
RCS file: /cvs/ports/devel/py-setproctitle/patches/patch-tests_pyrun_c,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-tests_pyrun_c
--- patches/patch-tests_pyrun_c 19 Dec 2011 10:37:41 - 1.1.1.1
+++ patches/patch-tests_pyrun_c 8 Apr 2012 13:57:12 -
@@ -1,6 +1,6 @@
-$OpenBSD: patch-tests_pyrun_c,v 1.1.1.1 2011/12/19 10:37:41 mpi Exp $
tests/pyrun.c Sun Dec 18 15:07:17 2011
-+++ tests/pyrun.c Sun Dec 18 15:07:28 2011
+$OpenBSD$
+--- tests/pyrun.c.orig Mon Feb 6 13:49:36 2012
tests/pyrun.c Sun Apr 8 15:56:57 2012
@@ -10,6 +10,7 @@
*-
*/
Index: patches/patch-tests_setproctitle_test_py
===
RCS file:
/cvs/ports/devel/py-setproctitle/patches/patch-tests_setproctitle_test_py,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-tests_setproctitle_test_py
--- patches/patch-tests_setproctitle_test_py19 Dec 2011 10:37:41 -
1.1.1.1
+++ patches/patch-tests_setproctitle_test_py8 Apr 2012 13:56:58 -
@@ -1,7 +1,7 @@
-$OpenBSD: patch-tests_setproctitle_test_py,v 1.1.1.1 2011/12/19 10:37:41 mpi
Exp $
tests/setproctitle_test.py Wed Nov 9 02:54:18 2011
-+++ tests/setproctitle_test.py Sun Dec 18 15:47:52 2011
-@@ -431
Re: devel/vte* without gnome-pty-helper
On 23/02/12(Thu) 20:22, Landry Breuil wrote:
> On Thu, Feb 23, 2012 at 06:05:59PM +0100, Martin Pieuchot wrote:
> > Diff below adds a new option for the libvte to rely on openpty(3) to
> > allocate a pseudo-tty. With it, we no longer need the gnome-pty-helper.
> >
> > If you use a terminal relying on the libvte, please test this diff and
> > report any breakage and/or improvement. Oks are also welcome ;)
>
> Works after some fiddling with xfce4-terminal using devel/vte, i386 &
> ppc.
> Your diff lacks patch-configure_in in devel/vte/patches, and it doesn't
> apply (?) in vte3/Makefile.
Here's a new diff that should fix that. Thanks for spotting it.
Martin
Index: vte/Makefile
===
RCS file: /cvs/ports/devel/vte/Makefile,v
retrieving revision 1.95
diff -u -p -r1.95 Makefile
--- vte/Makefile6 Feb 2012 22:07:09 - 1.95
+++ vte/Makefile23 Feb 2012 19:38:17 -
@@ -14,7 +14,7 @@ SHARED_LIBS+= vte 14.1# 2609.0
FULLPKGNAME-main= vte-${GNOME_VERSION}
FULLPKGNAME-python=py-${DISTNAME}
-REVISION-main =5
+REVISION-main =6
REVISION-python = 3
CATEGORIES=devel
@@ -39,7 +39,7 @@ WANTLIB += Xi Xinerama Xrandr Xrender Xx
WANTLIB += expat ffi fontconfig freetype gdk_pixbuf-2.0 gio-2.0
WANTLIB += glib-2.0 gmodule-2.0 gobject-2.0 gthread-2.0 m ncurses
WANTLIB += pango-1.0 pangocairo-1.0 pangoft2-1.0 pcre pixman-1
-WANTLIB += png pthread-stubs xcb xcb-render xcb-shm z
+WANTLIB += png pthread-stubs stdc++ util xcb xcb-render xcb-shm z
MODPY_RUNDEP= No
@@ -63,8 +63,13 @@ CONFIGURE_ARGS+= --enable-python
MODGNOME_TOOLS=goi
-CONFIGURE_STYLE= gnu
+AUTOCONF_VERSION= 2.68
+AUTOMAKE_VERSION= 1.11
+
+CONFIGURE_STYLE= autoconf
CONFIGURE_ARGS+= --disable-Bsymbolic \
+ --disable-gnome-pty-helper \
+ --enable-bsd-pty \
--with-gtk=2.0 \
--program-suffix=2
Index: vte/patches/patch-configure_in
===
RCS file: vte/patches/patch-configure_in
diff -N vte/patches/patch-configure_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ vte/patches/patch-configure_in 23 Feb 2012 19:33:56 -
@@ -0,0 +1,51 @@
+$OpenBSD$
+
+Support for the openpty(3) backend.
+
+--- configure.in
configure.in
+@@ -313,7 +313,7 @@ AC_SUBST(VTE_DEFAULT_EMULATION)
+ AM_CONDITIONAL(VTE_DEFAULT_EMULATION, [test "$emulation" != xterm])
+
+ # Check for headers.
+-AC_CHECK_HEADERS(sys/select.h sys/syslimits.h sys/termios.h sys/un.h
sys/wait.h stropts.h termios.h wchar.h)
++AC_CHECK_HEADERS(sys/select.h sys/syslimits.h sys/termios.h sys/un.h
sys/wait.h stropts.h termios.h util.h wchar.h)
+ AC_HEADER_TIOCGWINSZ
+
+ # Check for PTY handling functions.
+@@ -324,6 +324,7 @@ AC_CHECK_FUNCS([cfmakeraw fork setsid setpgid getpgid
getpt grantpt unlockpt pts
+
AC_CHECK_FUNC(socket,[have_socket=1],AC_CHECK_LIB(socket,socket,[have_socket=1;
LIBS="$LIBS -lsocket"]))
+
AC_CHECK_FUNC(socketpair,[have_socketpair=1],AC_CHECK_LIB(socket,socketpair,[have_socketpair=1;
LIBS="$LIBS -lsocket"]))
+
AC_CHECK_FUNC(recvmsg,[have_recvmsg=1],AC_CHECK_LIB(socket,recvmsg,[have_recvmsg=1;
LIBS="$LIBS -lsocket -lnsl"]))
++AC_CHECK_FUNC(openpty,[have_openpty=1],AC_CHECK_LIB(util,openpty,[have_openpty=1;
LIBS="$LIBS -lutil"]))
+ if test x$have_socket = x1 ; then
+ AC_DEFINE(HAVE_SOCKET,1,[Define if you have the socket function.])
+ fi
+@@ -333,6 +334,9 @@ fi
+ if test x$have_recvmsg = x1 ; then
+ AC_DEFINE(HAVE_RECVMSG,1,[Define if you have the recvmsg function.])
+ fi
++if test x$have_openpty = x1 ; then
++ AC_DEFINE(HAVE_OPENPTY,1,[Define if you have the openpty function.])
++fi
+ AC_CHECK_FUNC(floor,,AC_CHECK_LIB(m,floor,LIBS=["$LIBS -lm"]))
+ AC_CHECK_FUNCS([ceil floor])
+
+@@ -423,6 +427,17 @@ fi
+ AM_CONDITIONAL(BUILD_GNOME_PTY_HELPER,[test "$enable_gnome_pty_helper" != no])
+
+
++# BSD PTY
++
++
++AC_ARG_ENABLE(bsd-pty, [AS_HELP_STRING(--enable-bsd-pty,Support bsd rather
than unix98 ptys [default=no])], enable_bsd_pty="$enableval",
enable_bsd_pty=yes)
++if test "$enable_bsd_pty" != no; then
++ AC_DEFINE(VTE_USE_BSD_PTY,1,[Define if you intend to use bsd-pty.])
++ AC_CONFIG_SUBDIRS(bsd-pty)
++fi
++AM_CONDITIONAL(BUILD_BSD_PTY,[test "$enable_bsd_pty" != no])
++
++
+ # Glade catalogue
+
devel/vte* without gnome-pty-helper
Diff below adds a new option for the libvte to rely on openpty(3) to
allocate a pseudo-tty. With it, we no longer need the gnome-pty-helper.
If you use a terminal relying on the libvte, please test this diff and
report any breakage and/or improvement. Oks are also welcome ;)
Martin
Index: vte/Makefile
===
RCS file: /cvs/ports/devel/vte/Makefile,v
retrieving revision 1.95
diff -u -p -r1.95 Makefile
--- vte/Makefile6 Feb 2012 22:07:09 - 1.95
+++ vte/Makefile23 Feb 2012 16:08:42 -
@@ -14,7 +14,7 @@ SHARED_LIBS+= vte 14.1# 2609.0
FULLPKGNAME-main= vte-${GNOME_VERSION}
FULLPKGNAME-python=py-${DISTNAME}
-REVISION-main =5
+REVISION-main =6
REVISION-python = 3
CATEGORIES=devel
@@ -39,7 +39,7 @@ WANTLIB += Xi Xinerama Xrandr Xrender Xx
WANTLIB += expat ffi fontconfig freetype gdk_pixbuf-2.0 gio-2.0
WANTLIB += glib-2.0 gmodule-2.0 gobject-2.0 gthread-2.0 m ncurses
WANTLIB += pango-1.0 pangocairo-1.0 pangoft2-1.0 pcre pixman-1
-WANTLIB += png pthread-stubs xcb xcb-render xcb-shm z
+WANTLIB += png pthread-stubs stdc++ util xcb xcb-render xcb-shm z
MODPY_RUNDEP= No
@@ -63,8 +63,13 @@ CONFIGURE_ARGS+= --enable-python
MODGNOME_TOOLS=goi
-CONFIGURE_STYLE= gnu
+AUTOCONF_VERSION= 2.68
+AUTOMAKE_VERSION= 1.11
+
+CONFIGURE_STYLE= autoconf
CONFIGURE_ARGS+= --disable-Bsymbolic \
+ --disable-gnome-pty-helper \
+ --enable-bsd-pty \
--with-gtk=2.0 \
--program-suffix=2
Index: vte/patches/patch-src_pty_c
===
RCS file: /cvs/ports/devel/vte/patches/patch-src_pty_c,v
retrieving revision 1.9
diff -u -p -r1.9 patch-src_pty_c
--- vte/patches/patch-src_pty_c 26 Sep 2011 14:25:09 - 1.9
+++ vte/patches/patch-src_pty_c 23 Feb 2012 16:07:28 -
@@ -1,94 +1,90 @@
-$OpenBSD: patch-src_pty_c,v 1.9 2011/09/26 14:25:09 jasper Exp $
+$OpenBSD$
-- Party revert to previous behaviour of returning -1, instead of failing
- to compile.
+Use openpty(3) backend instead of gnome-pty-helper
-- Adjust gnome-pty-helper2 binary name.
-
src/pty.c.orig Tue Aug 16 23:52:48 2011
-+++ src/pty.c Mon Sep 26 15:44:22 2011
-@@ -795,7 +795,8 @@ _vte_pty_ptsname(int master,
- "%s failed: %s", "ioctl(TIOCGPTN)", g_strerror(errno));
- return NULL;
- #else
--#error no ptsname implementation for this platform
-+#warning no ptsname implementation for this platform
-+ return NULL;
+--- src/pty.c
src/pty.c
+@@ -54,6 +54,9 @@
+ #include
#endif
+ #include
++#ifdef HAVE_UTIL_H
++#include
++#endif
+ #ifdef HAVE_STROPTS_H
+ #include
+ #endif
+@@ -734,6 +737,7 @@ vte_pty_get_size(VtePty *pty,
+ }
}
-@@ -903,7 +904,8 @@ _vte_pty_unlockpt(int fd,
- }
++#ifndef VTE_USE_BSD_PTY
+ /*
+ * _vte_pty_ptsname:
+ * @master: file descriptor to the PTY master
+@@ -951,6 +955,44 @@ _vte_pty_open_unix98(VtePty *pty,
return TRUE;
- #else
--#error no unlockpt implementation for this platform
-+#warning no unlockpt implementation for this platform
-+ return -1;
- #endif
}
-@@ -1181,8 +1183,8 @@ _vte_pty_start_helper(GError **error)
- close(tunnel);
- close(_vte_pty_helper_tunnel);
- /* Exec our helper. */
-- execl(LIBEXECDIR "/gnome-pty-helper",
--"gnome-pty-helper", NULL);
-+ execl(LIBEXECDIR "/gnome-pty-helper2",
-+"gnome-pty-helper2", NULL);
- /* Bail. */
- _exit(1);
- }
-@@ -1197,7 +1199,7 @@ failure:
-
- g_set_error(error, VTE_PTY_ERROR,
- VTE_PTY_ERROR_PTY_HELPER_FAILED,
--"Failed to start gnome-pty-helper: %s",
-+"Failed to start gnome-pty-helper2: %s",
- g_strerror (errsv));
++#else
++/*
++ * _vte_pty_open_bsd:
++ * @pty: a #VtePty
++ * @error: a location to store a #GError, or %NULL
++ *
++ * Opens new file descriptors to a new PTY master and slave.
++ *
++ * Returns: %TRUE on success, %FALSE on failure with @error filled in
++ */
++static gboolean
++_vte_pty_open_bsd(VtePty *pty,
++ GError **error)
++{
++#ifdef HAVE_OPENPTY
++VtePtyPrivate *priv = pty->priv;
++ int parentfd, childfd;
++
++ if (openpty(&parentfd, &childfd, NULL, NULL, NULL) != 0) {
++int errsv = errno;
++g_set_error(error, VTE_PTY_ERROR, VTE_PTY_ERROR_PTY98_FAILED,
++"%s failed: %s", "openpty", g_strerror(errsv));
++errno = errsv;
++ return FALSE;
++ }
++
++priv->pty_fd = parentfd;
++priv->child_setup_data.mode = TTY_OPEN_BY_FD;
++priv->child_setup_data.tt
Re: UPDATE: py-crypto 2.4.1
On 26/12/11(Mon) 10:18, Martin Pieuchot wrote:
> Update py-crypto to 2.4.1 and install doc under py-XXX for upcoming
> python 2 and 3 compatibility.
Better update, honores CFLAGS and picks gmp for compiling the
fastmath module, pointed out by rpointel@
Ok?
Index: Makefile
===
RCS file: /cvs/ports/security/py-crypto/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile4 Nov 2011 13:36:16 - 1.23
+++ Makefile26 Dec 2011 13:38:37 -
@@ -4,13 +4,11 @@ SHARED_ONLY = Yes
COMMENT = cryptographic tools for Python
-MODPY_EGG_VERSION =2.3
+MODPY_EGG_VERSION =2.4.1
DISTNAME = pycrypto-${MODPY_EGG_VERSION}
PKGNAME = py-crypto-${MODPY_EGG_VERSION}
CATEGORIES = security devel
-REVISION = 0
-
HOMEPAGE = https://www.dlitz.net/software/pycrypto/
MAINTAINER = Martin Pieuchot
@@ -24,10 +22,15 @@ PERMIT_DISTFILES_FTP = Yes
MASTER_SITES = http://ftp.dlitz.net/pub/dlitz/crypto/pycrypto/
MODULES = lang/python
-LIB_DEPENDS = ${MODPY_LIB_DEPENDS}
-WANTLIB = ${MODPY_WANTLIB}
-
-BUILD_DEPENDS =devel/gmp
+LIB_DEPENDS = ${MODPY_LIB_DEPENDS} \
+ devel/gmp
+WANTLIB = ${MODPY_WANTLIB} gmp
+
+# XXX Don't set CONFIGURE_STYLE=simple otherwise python.port.mk will
+# not define do-build and do-install
+CONFIGURE_ARGS = CFLAGS="-I${LOCALBASE}/include" \
+ LDFLAGS="-L${LOCALBASE}/lib"
+CFLAGS += -I${LOCALBASE}/include
DOCSDIR = ${PREFIX}/share/doc/py-crypto
@@ -35,6 +38,9 @@ pre-configure:
@${SUBST_CMD} ${WRKSRC}/setup.py
rm ${WRKSRC}/lib/Crypto/SelfTest/Cipher/test_ARC2.py
rm ${WRKSRC}/lib/Crypto/SelfTest/Hash/test_MD2.py
+
+do-configure:
+ ${MODSIMPLE_configure}
post-install:
${INSTALL_DATA_DIR} ${DOCSDIR}
Index: distinfo
===
RCS file: /cvs/ports/security/py-crypto/distinfo,v
retrieving revision 1.6
diff -u -p -r1.6 distinfo
--- distinfo8 Oct 2011 14:56:34 - 1.6
+++ distinfo23 Dec 2011 18:28:51 -
@@ -1,5 +1,5 @@
-MD5 (pycrypto-2.3.tar.gz) = K4Ec+/w0LYPuYUCX7/uBAQ==
-RMD160 (pycrypto-2.3.tar.gz) = W3sVqrP14zETuhl6Lt1inHJZU3Q=
-SHA1 (pycrypto-2.3.tar.gz) = lrM7d+cpiTvQg3oJpqp6ErUHADc=
-SHA256 (pycrypto-2.3.tar.gz) = TxHoX7zxOWA3NlD8La6PCI+bAB8H+20++y/LUzSYcYI=
-SIZE (pycrypto-2.3.tar.gz) = 331476
+MD5 (pycrypto-2.4.1.tar.gz) = wqFASoSHl/sIBvPhHCnvFQ==
+RMD160 (pycrypto-2.4.1.tar.gz) = fb87DKEOj3KHcvCygmZ+S6lGYcU=
+SHA1 (pycrypto-2.4.1.tar.gz) = 2gKDWZhskSW4gLW2K3yLoGtzMoI=
+SHA256 (pycrypto-2.4.1.tar.gz) = T/X3AIhpyFCdFqMfKFufY0DEqf9wPIUpiDKihpIQsVQ=
+SIZE (pycrypto-2.4.1.tar.gz) = 383032
Index: patches/patch-lib_Crypto_SelfTest_Cipher___init___py
===
RCS file:
/cvs/ports/security/py-crypto/patches/patch-lib_Crypto_SelfTest_Cipher___init___py,v
retrieving revision 1.2
diff -u -p -r1.2 patch-lib_Crypto_SelfTest_Cipher___init___py
--- patches/patch-lib_Crypto_SelfTest_Cipher___init___py9 Oct 2011
15:09:55 - 1.2
+++ patches/patch-lib_Crypto_SelfTest_Cipher___init___py26 Dec 2011
08:51:36 -
@@ -1,10 +1,11 @@
lib/Crypto/SelfTest/Cipher/__init__.py.origFri Aug 27 06:41:55 2010
-+++ lib/Crypto/SelfTest/Cipher/__init__.py Sun Oct 9 17:06:47 2011
-@@ -29,7 +29,6 @@ __revision__ = "$Id:
patch-lib_Crypto_SelfTest_Cipher___init___py,v 1.2 2011/10/09 15:09:55
ajacoutot Exp $"
+$OpenBSD$
+--- lib/Crypto/SelfTest/Cipher/__init__.py.origFri Nov 4 20:17:40 2011
lib/Crypto/SelfTest/Cipher/__init__.py Mon Dec 26 09:48:45 2011
+@@ -29,7 +29,6 @@
def get_tests(config={}):
tests = []
- import test_AES; tests += test_AES.get_tests(config=config)
--import test_ARC2; tests += test_ARC2.get_tests(config=config)
- import test_ARC4; tests += test_ARC4.get_tests(config=config)
- import test_Blowfish; tests += test_Blowfish.get_tests(config=config)
- import test_CAST; tests += test_CAST.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_AES; tests +=
test_AES.get_tests(config=config)
+-from Crypto.SelfTest.Cipher import test_ARC2; tests +=
test_ARC2.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_ARC4; tests +=
test_ARC4.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_Blowfish; tests +=
test_Blowfish.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_CAST; tests +=
test_CAST.get_tests(config=config)
Index: patches/patch-lib_Crypto_SelfTest_Hash___init__py
===
RCS file:
/cvs/ports/security/py-crypto/patches/patch-lib_Crypto_SelfTest_Hash___i
UPDATE: py-crypto 2.4.1
Update py-crypto to 2.4.1 and install doc under py-XXX for upcoming
python 2 and 3 compatibility.
Ok?
Martin
Index: Makefile
===
RCS file: /cvs/ports/security/py-crypto/Makefile,v
retrieving revision 1.23
diff -u -p -r1.23 Makefile
--- Makefile4 Nov 2011 13:36:16 - 1.23
+++ Makefile26 Dec 2011 09:10:51 -
@@ -4,13 +4,11 @@ SHARED_ONLY = Yes
COMMENT = cryptographic tools for Python
-MODPY_EGG_VERSION =2.3
+MODPY_EGG_VERSION =2.4.1
DISTNAME = pycrypto-${MODPY_EGG_VERSION}
PKGNAME = py-crypto-${MODPY_EGG_VERSION}
CATEGORIES = security devel
-REVISION = 0
-
HOMEPAGE = https://www.dlitz.net/software/pycrypto/
MAINTAINER = Martin Pieuchot
@@ -38,7 +36,7 @@ pre-configure:
post-install:
${INSTALL_DATA_DIR} ${DOCSDIR}
- ${INSTALL_DATA} ${WRKSRC}/README ${DOCSDIR}
+ ${INSTALL_DATA} ${WRKSRC}/README ${PREFIX}/share/doc/py-crypto
do-regress: fake
cd ${WRKSRC} && ${MAKE_ENV} \
Index: distinfo
===
RCS file: /cvs/ports/security/py-crypto/distinfo,v
retrieving revision 1.6
diff -u -p -r1.6 distinfo
--- distinfo8 Oct 2011 14:56:34 - 1.6
+++ distinfo23 Dec 2011 18:28:51 -
@@ -1,5 +1,5 @@
-MD5 (pycrypto-2.3.tar.gz) = K4Ec+/w0LYPuYUCX7/uBAQ==
-RMD160 (pycrypto-2.3.tar.gz) = W3sVqrP14zETuhl6Lt1inHJZU3Q=
-SHA1 (pycrypto-2.3.tar.gz) = lrM7d+cpiTvQg3oJpqp6ErUHADc=
-SHA256 (pycrypto-2.3.tar.gz) = TxHoX7zxOWA3NlD8La6PCI+bAB8H+20++y/LUzSYcYI=
-SIZE (pycrypto-2.3.tar.gz) = 331476
+MD5 (pycrypto-2.4.1.tar.gz) = wqFASoSHl/sIBvPhHCnvFQ==
+RMD160 (pycrypto-2.4.1.tar.gz) = fb87DKEOj3KHcvCygmZ+S6lGYcU=
+SHA1 (pycrypto-2.4.1.tar.gz) = 2gKDWZhskSW4gLW2K3yLoGtzMoI=
+SHA256 (pycrypto-2.4.1.tar.gz) = T/X3AIhpyFCdFqMfKFufY0DEqf9wPIUpiDKihpIQsVQ=
+SIZE (pycrypto-2.4.1.tar.gz) = 383032
Index: patches/patch-lib_Crypto_SelfTest_Cipher___init___py
===
RCS file:
/cvs/ports/security/py-crypto/patches/patch-lib_Crypto_SelfTest_Cipher___init___py,v
retrieving revision 1.2
diff -u -p -r1.2 patch-lib_Crypto_SelfTest_Cipher___init___py
--- patches/patch-lib_Crypto_SelfTest_Cipher___init___py9 Oct 2011
15:09:55 - 1.2
+++ patches/patch-lib_Crypto_SelfTest_Cipher___init___py26 Dec 2011
08:51:36 -
@@ -1,10 +1,11 @@
lib/Crypto/SelfTest/Cipher/__init__.py.origFri Aug 27 06:41:55 2010
-+++ lib/Crypto/SelfTest/Cipher/__init__.py Sun Oct 9 17:06:47 2011
-@@ -29,7 +29,6 @@ __revision__ = "$Id:
patch-lib_Crypto_SelfTest_Cipher___init___py,v 1.2 2011/10/09 15:09:55
ajacoutot Exp $"
+$OpenBSD$
+--- lib/Crypto/SelfTest/Cipher/__init__.py.origFri Nov 4 20:17:40 2011
lib/Crypto/SelfTest/Cipher/__init__.py Mon Dec 26 09:48:45 2011
+@@ -29,7 +29,6 @@
def get_tests(config={}):
tests = []
- import test_AES; tests += test_AES.get_tests(config=config)
--import test_ARC2; tests += test_ARC2.get_tests(config=config)
- import test_ARC4; tests += test_ARC4.get_tests(config=config)
- import test_Blowfish; tests += test_Blowfish.get_tests(config=config)
- import test_CAST; tests += test_CAST.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_AES; tests +=
test_AES.get_tests(config=config)
+-from Crypto.SelfTest.Cipher import test_ARC2; tests +=
test_ARC2.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_ARC4; tests +=
test_ARC4.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_Blowfish; tests +=
test_Blowfish.get_tests(config=config)
+ from Crypto.SelfTest.Cipher import test_CAST; tests +=
test_CAST.get_tests(config=config)
Index: patches/patch-lib_Crypto_SelfTest_Hash___init__py
===
RCS file:
/cvs/ports/security/py-crypto/patches/patch-lib_Crypto_SelfTest_Hash___init__py,v
retrieving revision 1.2
diff -u -p -r1.2 patch-lib_Crypto_SelfTest_Hash___init__py
--- patches/patch-lib_Crypto_SelfTest_Hash___init__py 9 Oct 2011 15:09:55
- 1.2
+++ patches/patch-lib_Crypto_SelfTest_Hash___init__py 26 Dec 2011 08:51:39
-
@@ -1,10 +1,11 @@
lib/Crypto/SelfTest/Hash/__init__.py.orig Fri Aug 27 06:41:55 2010
-+++ lib/Crypto/SelfTest/Hash/__init__.py Sun Oct 9 17:06:47 2011
-@@ -29,7 +29,6 @@ __revision__ = "$Id:
patch-lib_Crypto_SelfTest_Hash___init__py,v 1.2 2011/10/09 15:09:55 ajacoutot
Exp $"
+$OpenBSD$
+--- lib/Crypto/SelfTest/Hash/__init__.py.orig Fri Nov 4 20:17:40 2011
lib/Crypto/SelfTest/Hash/__init__.py Mon Dec 26 09:50:24 2011
+@@ -29,7 +29,6 @@
def get_tests(config={}):
tests = []
- import test_HMAC; tests += test_HMAC.get_tests(config=config)
--
Re: [NEW] devel/py-setproctitle
On 13/12/11(Tue) 11:47, Martin Pieuchot wrote: > Description: > setproctitle allows a process to change its title (as displayed by > system tools such as ps and top). Updated version that fixes the regress tests, pointed ou by benoit@. Ok? Martin py-setproctitle.tgz Description: application/tar-gz
[NEW] devel/py-xmlrunner
Description: PyUnit-based test runner with JUnit like XML reporting. Comments, Ok? Martin py-xmlrunner.tgz Description: application/tar-gz
[NEW] devel/py-country
Description: pycountry provides ISO country, subdivision, language, currency and script definitions and their translations, taken from the iso-codes package. Comments, Ok? Martin py-country.tgz Description: application/tar-gz
[NEW] devel/py-setproctitle
Description: setproctitle allows a process to change its title (as displayed by system tools such as ps and top). Comments? Ok? py-setproctitle.tgz Description: application/tar-gz
Re: [UPDATE] libgpod 0.8.0
On 14/11/11(Mon) 16:43, Martin Pieuchot wrote:
> Slowly updating gtkpod's family... Here's libgpod with a new
> dependency devel/libplist 1.4:
New diff with a correct library bump, pointed out by sthen@
Ok?
Index: Makefile
===
RCS file: /cvs/ports/audio/libgpod/Makefile,v
retrieving revision 1.28
diff -u -p -r1.28 Makefile
--- Makefile12 Oct 2011 12:54:30 - 1.28
+++ Makefile15 Nov 2011 11:09:06 -
@@ -3,13 +3,11 @@
COMMENT-main= library to access the contents of an iPod
COMMENT-python= python bindings for libgpod
-VERSION= 0.7.2
+VERSION= 0.8.0
DISTNAME= libgpod-${VERSION}
PKGNAME-main= ${DISTNAME}
-REVISION-main =7
PKGNAME-python= py-gpod-${VERSION}
-REVISION-python = 7
-SHARED_LIBS= gpod720.0
+SHARED_LIBS= gpod721.0 # 7.1
CATEGORIES=audio
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=gtkpod/}
@@ -38,21 +36,26 @@ RUN_DEPENDS-python= audio/py-mutagen
LIB_DEPENDS-python=${LIB_DEPENDS} \
devel/libffi \
${BUILD_PKGPATH},-main
-LIB_DEPENDS= devel/glib2 \
+LIB_DEPENDS= databases/sqlite3 \
+ devel/glib2 \
+ devel/libplist \
+ devel/py-gobject \
graphics/gdk-pixbuf2 \
textproc/libxml
WANTLIB += gio-2.0 glib-2.0 gmodule-2.0 gobject-2.0 gthread-2.0 m z pcre
-WANTLIB += gdk_pixbuf-2.0 xml2 png ffi
+WANTLIB += gdk_pixbuf-2.0 xml2 png ffi sqlite3 plist
+WANTLIB-main += ${WANTLIB} c pthread
WANTLIB-python += ${WANTLIB} gpod
CONFIGURE_STYLE=gnu
-CONFIGURE_ARGS+=--disable-hal \
- --with-python="${MODPY_BIN}"
-CONFIGURE_ENV+= CPPFLAGS="-I${LOCALBASE}/include" \
- LDFLAGS="-L${LOCALBASE}/lib" \
- CXXFLAGS="${CXXFLAGS} -ftrampolines" \
- CFLAGS="${CFLAGS} -ftrampolines"
+CONFIGURE_ARGS+=--enable-gdk-pixbuf \
+ --enable-pygobject \
+ --without-hal \
+ --without-libimobiledevice \
+ --without-mono \
+ --with-python="${MODPY_BIN}" \
+ --localedir=${TRUEPREFIX}/share/locale
pre-configure:
@find ${WRKDIST} -name \*.py | \
Index: distinfo
===
RCS file: /cvs/ports/audio/libgpod/distinfo,v
retrieving revision 1.8
diff -u -p -r1.8 distinfo
--- distinfo4 Sep 2009 01:04:54 - 1.8
+++ distinfo15 Nov 2011 08:46:01 -
@@ -1,5 +1,5 @@
-MD5 (libgpod-0.7.2.tar.gz) = HsacOhn7BxsWOc3K9oRjwQ==
-RMD160 (libgpod-0.7.2.tar.gz) = IxtO/Um4s25bZE6ASizQlYsdP78=
-SHA1 (libgpod-0.7.2.tar.gz) = Zslc7J+KupLzRy33Qp6HPTDQvXk=
-SHA256 (libgpod-0.7.2.tar.gz) = Lo8x/DftDJyJAECX/bVo6JcWJ+eu37V/sik5nA9JB3c=
-SIZE (libgpod-0.7.2.tar.gz) = 959142
+MD5 (libgpod-0.8.0.tar.gz) = ZmD3TMUyk9zIR0B6pfZyzg==
+RMD160 (libgpod-0.8.0.tar.gz) = RG1QRfbJHlXolMXVNwZYr8ir6O8=
+SHA1 (libgpod-0.8.0.tar.gz) = 3e9/NYNTUkK0koswDriqa8mg5tw=
+SHA256 (libgpod-0.8.0.tar.gz) = G4uu+OWnGPahHqrqVhm5BhmHZtmZ6NVDCPUUtSJXFgI=
+SIZE (libgpod-0.8.0.tar.gz) = 997674
Index: patches/patch-po_Makefile_in_in
===
RCS file: patches/patch-po_Makefile_in_in
diff -N patches/patch-po_Makefile_in_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-po_Makefile_in_in 15 Nov 2011 08:46:01 -
@@ -0,0 +1,13 @@
+$OpenBSD$
+--- po/Makefile.in.in Sat Nov 12 18:35:15 2011
po/Makefile.in.in Sat Nov 12 18:36:14 2011
+@@ -33,8 +33,7 @@
+ datadir = @datadir@
+ datarootdir = @datarootdir@
+ libdir = @libdir@
+-DATADIRNAME = @DATADIRNAME@
+-itlocaledir = $(prefix)/$(DATADIRNAME)/locale
++itlocaledir = @localedir@
+ subdir = po
+ install_sh = @install_sh@
+ # Automake >= 1.8 provides @mkdir_p@.
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/audio/libgpod/pkg/PLIST-main,v
retrieving revision 1.1
diff -u -p -r1.1 PLIST-main
--- pkg/PLIST-main 24 Apr 2009 09:45:48 - 1.1
+++ pkg/PLIST-main 15 Nov 2011 08:46:01 -
@@ -1,5 +1,6 @@
@comment $OpenBSD: PLIST-main,v 1.1 2009/04/24 09:45:48 jasper Exp $
@pkgpath audio/libgpod
+@bin bin/ipod-read-sysinfo-extended
%%SHARED%%
include/gpod-1.0/
include/gpod-1.0/gpod/
@@ -7,6 +8,7 @@ include/gpod-1.0/gpod/itdb.h
lib/libgpod.a
lib/libgpod.la
lib/pkgconfig/libgpod-1.0.pc
+@comment lib/pkgconfig/libgpod-sharp.pc
share/gtk-doc/html/libgpod/
share/gtk-doc/html/libgpod/ch01.html
share/gtk-doc/html/libgpod/home.png
[UPDATE] gtkpod 1.0.0
Update gtkpod to 1.0.0 (latest headache proof version). This update also
enable flac and ogg conversion, tested on amd64.
Comments, Ok?
Martin
Index: Makefile
===
RCS file: /cvs/ports/audio/gtkpod/Makefile,v
retrieving revision 1.41
diff -u -p -r1.41 Makefile
--- Makefile18 Oct 2011 18:57:15 - 1.41
+++ Makefile14 Nov 2011 15:31:06 -
@@ -1,8 +1,7 @@
# $OpenBSD: Makefile,v 1.41 2011/10/18 18:57:15 sthen Exp $
COMMENT= GTK+2 program to synchronize an Apple iPod
-DISTNAME= gtkpod-0.99.14
-REVISION= 7
+DISTNAME= gtkpod-1.0.0
CATEGORIES=audio x11
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=gtkpod/}
@@ -21,14 +20,15 @@ WANTLIB += Xcomposite Xdamage c cairo ex
WANTLIB += pthread-stubs xcb z xml2 png glib-2.0 gmodule-2.0 gobject-2.0
WANTLIB += gthread-2.0 pango-1.0 pangoft2-1.0 pangocairo-1.0 atk-1.0 pixman-1
WANTLIB += gio-2.0 id3tag gpod gdk-x11-2.0 gdk_pixbuf-2.0 gtk-x11-2.0 curl
-WANTLIB += glade-2.0 mp4v2 xcb-render GL xcb-shm
+WANTLIB += glade-2.0 mp4v2 xcb-render GL xcb-shm FLAC ogg vorbis vorbisfile
USE_GMAKE= Yes
-USE_GROFF =Yes
RUN_DEPENDS= devel/desktop-file-utils
LIB_DEPENDS= audio/libid3tag \
audio/libgpod \
+ audio/flac \
+ audio/libvorbis \
devel/libglade2 \
net/curl \
multimedia/libmp4v2
@@ -36,11 +36,9 @@ LIB_DEPENDS= audio/libid3tag \
MODULES= devel/gettext textproc/intltool
CONFIGURE_STYLE=gnu
-CONFIGURE_ARGS+=--without-hal \
- --without-gnome-vfs
-# trampolines needed for display_playlists.c
-CONFIGURE_ENV= LDFLAGS="-L${LOCALBASE}/lib" \
- CFLAGS="${CFLAGS} -ftrampolines" \
- CPPFLAGS="-I${LOCALBASE}/include"
+CONFIGURE_ARGS+= --with-curl \
+ --with-ogg \
+ --with-flac
+CONFIGURE_ENV+=LDFLAGS="-L${LOCALBASE}/lib -logg -lvorbis -lm"
.include
Index: distinfo
===
RCS file: /cvs/ports/audio/gtkpod/distinfo,v
retrieving revision 1.12
diff -u -p -r1.12 distinfo
--- distinfo6 Mar 2009 20:44:24 - 1.12
+++ distinfo11 Nov 2011 18:23:13 -
@@ -1,5 +1,5 @@
-MD5 (gtkpod-0.99.14.tar.gz) = 95SOzrlVswL0xH2kwOHsEg==
-RMD160 (gtkpod-0.99.14.tar.gz) = 7EgGkYMQswccM9Rhrzo1S1gmqEw=
-SHA1 (gtkpod-0.99.14.tar.gz) = kpd0IMC1ToXdyA++LpnNSl5Ogbk=
-SHA256 (gtkpod-0.99.14.tar.gz) = c+EoNIJnmuDMd0qkoHrslJFZgoizaYZQXsPRxqB37D0=
-SIZE (gtkpod-0.99.14.tar.gz) = 1856919
+MD5 (gtkpod-1.0.0.tar.gz) = yt1ALc0c++3aA1e/JJZafA==
+RMD160 (gtkpod-1.0.0.tar.gz) = Jx6a6yVJO4mTPyA+vOTABM4NpXM=
+SHA1 (gtkpod-1.0.0.tar.gz) = d7doZjRYBppBTFm3qtKFAgyToDU=
+SHA256 (gtkpod-1.0.0.tar.gz) = 2AiUVdCJuFOQ2u/FIrfWBAPM8kWFpXyun0ewUvTyXxI=
+SIZE (gtkpod-1.0.0.tar.gz) = 1899869
Index: patches/patch-scripts_gtkpod-convert-common_sh
===
RCS file: patches/patch-scripts_gtkpod-convert-common_sh
diff -N patches/patch-scripts_gtkpod-convert-common_sh
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-scripts_gtkpod-convert-common_sh 13 Nov 2011 18:06:36
-
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- scripts/gtkpod-convert-common.sh Sun Nov 13 19:05:33 2011
scripts/gtkpod-convert-common.sh Sun Nov 13 19:05:45 2011
+@@ -90,7 +90,7 @@
+ case "$filetype" in
+ flac) decoder="flac" ; options="-d -c --" ;;
+ oga|ogg|ogv|ogx)
+- decoder="oggdec" ; options="--output - --" ;;
++ decoder="oggdec" ; options="-Q --output - --" ;;
+ m4a)decoder="faad" ; options="-o -" ;;
+ wav)decoder="" ;;
+ *) exit 4 ;;
Index: patches/patch-src_Makefile_in
===
RCS file: patches/patch-src_Makefile_in
diff -N patches/patch-src_Makefile_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_Makefile_in 13 Nov 2011 18:07:04 -
@@ -0,0 +1,12 @@
+$OpenBSD$
+--- src/Makefile.inSun Nov 13 17:32:12 2011
src/Makefile.inSun Nov 13 17:33:08 2011
+@@ -282,7 +282,7 @@
+ rb_rating_helper.c rb_rating_helper.h \
+ rb_cell_renderer_rating.c rb_cell_renderer_rating.h
+
+-gtkpod_LDADD = $(LIBS) -lm $(INTLLIBS) -ldl @LIBOBJS@
++gtkpod_LDADD = $(LIBS) -lm $(INTLLIBS) @LIBOBJS@
+ EXTRA_DIST = getopt.h
+ CLEANFILES = date_parser.c date_parser2.c
+ all: all-am
Index: patches/patch-src_mp4file_c
===
RCS file: patches/patch-src_mp4file_c
diff -N patches/patch-src_mp4file_c
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-src_mp4file_c 13 Nov 2011 18:07:16 -
@@ -0,0 +1,28 @@
+$OpenBSD$
+--- src/mp4file.c Sun Nov 13 17:27:50 2011
src/mp4file.c Sun Nov 13 17:47:00 2011
+@@ -35,6 +35,10 @@
+ # include /* for be32toh () */
+ #endif
+
++#
[UPDATE] libgpod 0.8.0
Slowly updating gtkpod's family... Here's libgpod with a new
dependency devel/libplist 1.4:
Comment:
library to handle Apple Property List format
Required by:
libgpod-0.8.0
Ok?
Index: Makefile
===
RCS file: /cvs/ports/audio/libgpod/Makefile,v
retrieving revision 1.28
diff -u -p -r1.28 Makefile
--- Makefile12 Oct 2011 12:54:30 - 1.28
+++ Makefile14 Nov 2011 09:06:40 -
@@ -3,13 +3,11 @@
COMMENT-main= library to access the contents of an iPod
COMMENT-python= python bindings for libgpod
-VERSION= 0.7.2
+VERSION= 0.8.0
DISTNAME= libgpod-${VERSION}
PKGNAME-main= ${DISTNAME}
-REVISION-main =7
PKGNAME-python= py-gpod-${VERSION}
-REVISION-python = 7
-SHARED_LIBS= gpod720.0
+SHARED_LIBS= gpod800.0 # 7.1
CATEGORIES=audio
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE:=gtkpod/}
@@ -38,21 +36,26 @@ RUN_DEPENDS-python= audio/py-mutagen
LIB_DEPENDS-python=${LIB_DEPENDS} \
devel/libffi \
${BUILD_PKGPATH},-main
-LIB_DEPENDS= devel/glib2 \
+LIB_DEPENDS= databases/sqlite3 \
+ devel/glib2 \
+ devel/libplist \
+ devel/py-gobject \
graphics/gdk-pixbuf2 \
textproc/libxml
WANTLIB += gio-2.0 glib-2.0 gmodule-2.0 gobject-2.0 gthread-2.0 m z pcre
-WANTLIB += gdk_pixbuf-2.0 xml2 png ffi
+WANTLIB += gdk_pixbuf-2.0 xml2 png ffi sqlite3 plist
+WANTLIB-main += ${WANTLIB} c pthread
WANTLIB-python += ${WANTLIB} gpod
CONFIGURE_STYLE=gnu
-CONFIGURE_ARGS+=--disable-hal \
- --with-python="${MODPY_BIN}"
-CONFIGURE_ENV+= CPPFLAGS="-I${LOCALBASE}/include" \
- LDFLAGS="-L${LOCALBASE}/lib" \
- CXXFLAGS="${CXXFLAGS} -ftrampolines" \
- CFLAGS="${CFLAGS} -ftrampolines"
+CONFIGURE_ARGS+=--enable-gdk-pixbuf \
+ --enable-pygobject \
+ --without-hal \
+ --without-libimobiledevice \
+ --without-mono \
+ --with-python="${MODPY_BIN}" \
+ --localedir=${TRUEPREFIX}/share/locale
pre-configure:
@find ${WRKDIST} -name \*.py | \
Index: distinfo
===
RCS file: /cvs/ports/audio/libgpod/distinfo,v
retrieving revision 1.8
diff -u -p -r1.8 distinfo
--- distinfo4 Sep 2009 01:04:54 - 1.8
+++ distinfo12 Nov 2011 17:08:28 -
@@ -1,5 +1,5 @@
-MD5 (libgpod-0.7.2.tar.gz) = HsacOhn7BxsWOc3K9oRjwQ==
-RMD160 (libgpod-0.7.2.tar.gz) = IxtO/Um4s25bZE6ASizQlYsdP78=
-SHA1 (libgpod-0.7.2.tar.gz) = Zslc7J+KupLzRy33Qp6HPTDQvXk=
-SHA256 (libgpod-0.7.2.tar.gz) = Lo8x/DftDJyJAECX/bVo6JcWJ+eu37V/sik5nA9JB3c=
-SIZE (libgpod-0.7.2.tar.gz) = 959142
+MD5 (libgpod-0.8.0.tar.gz) = ZmD3TMUyk9zIR0B6pfZyzg==
+RMD160 (libgpod-0.8.0.tar.gz) = RG1QRfbJHlXolMXVNwZYr8ir6O8=
+SHA1 (libgpod-0.8.0.tar.gz) = 3e9/NYNTUkK0koswDriqa8mg5tw=
+SHA256 (libgpod-0.8.0.tar.gz) = G4uu+OWnGPahHqrqVhm5BhmHZtmZ6NVDCPUUtSJXFgI=
+SIZE (libgpod-0.8.0.tar.gz) = 997674
Index: patches/patch-po_Makefile_in_in
===
RCS file: patches/patch-po_Makefile_in_in
diff -N patches/patch-po_Makefile_in_in
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-po_Makefile_in_in 14 Nov 2011 15:22:54 -
@@ -0,0 +1,13 @@
+$OpenBSD$
+--- po/Makefile.in.in Sat Nov 12 18:35:15 2011
po/Makefile.in.in Sat Nov 12 18:36:14 2011
+@@ -33,8 +33,7 @@
+ datadir = @datadir@
+ datarootdir = @datarootdir@
+ libdir = @libdir@
+-DATADIRNAME = @DATADIRNAME@
+-itlocaledir = $(prefix)/$(DATADIRNAME)/locale
++itlocaledir = @localedir@
+ subdir = po
+ install_sh = @install_sh@
+ # Automake >= 1.8 provides @mkdir_p@.
Index: pkg/PLIST-main
===
RCS file: /cvs/ports/audio/libgpod/pkg/PLIST-main,v
retrieving revision 1.1
diff -u -p -r1.1 PLIST-main
--- pkg/PLIST-main 24 Apr 2009 09:45:48 - 1.1
+++ pkg/PLIST-main 14 Nov 2011 09:05:44 -
@@ -1,5 +1,6 @@
@comment $OpenBSD: PLIST-main,v 1.1 2009/04/24 09:45:48 jasper Exp $
@pkgpath audio/libgpod
+@bin bin/ipod-read-sysinfo-extended
%%SHARED%%
include/gpod-1.0/
include/gpod-1.0/gpod/
@@ -7,6 +8,7 @@ include/gpod-1.0/gpod/itdb.h
lib/libgpod.a
lib/libgpod.la
lib/pkgconfig/libgpod-1.0.pc
+@comment lib/pkgconfig/libgpod-sharp.pc
share/gtk-doc/html/libgpod/
share/gtk-doc/html/libgpod/ch01.html
share/gtk-doc/html/libgpod/home.png
libplist.tgz
Description: application/tar-gz
Re: [UPDATE] offlineimap 6.3.4
On 09/10/11(Sun) 14:50, Martin Pieuchot wrote:
> Simple update... Ok?
I missed a patch in the previous diff...
Index: Makefile
===
RCS file: /cvs/ports/mail/offlineimap/Makefile,v
retrieving revision 1.22
diff -u -p -r1.22 Makefile
--- Makefile16 Sep 2011 10:31:21 - 1.22
+++ Makefile10 Oct 2011 08:25:19 -
@@ -2,11 +2,10 @@
COMMENT= powerful IMAP/Maildir synchronization and reader support
-MODPY_EGG_VERSION= 6.3.2
+MODPY_EGG_VERSION= 6.3.4
DISTNAME= offlineimap_${MODPY_EGG_VERSION}.orig
CATEGORIES=mail
PKGNAME= offlineimap-${MODPY_EGG_VERSION}
-REVISION= 1
HOMEPAGE= http://offlineimap.org/
@@ -26,7 +25,7 @@ MODULES= lang/python
BUILD_DEPENDS= textproc/py-docutils
-WRKDIST= ${WRKDIR}/nicolas33-offlineimap-b49802f
+WRKDIST= ${WRKDIR}/nicolas33-offlineimap-94450e9
EXAMPLESDIR= ${PREFIX}/share/examples/offlineimap
Index: distinfo
===
RCS file: /cvs/ports/mail/offlineimap/distinfo,v
retrieving revision 1.13
diff -u -p -r1.13 distinfo
--- distinfo27 Apr 2011 12:59:27 - 1.13
+++ distinfo10 Oct 2011 08:25:28 -
@@ -1,5 +1,5 @@
-MD5 (offlineimap_6.3.2.orig.tar.gz) = c+9rRoXXZRJh+VkZSfaBvg==
-RMD160 (offlineimap_6.3.2.orig.tar.gz) = fKHo6nWpKlaxXzVLwk/wBZ3fw/0=
-SHA1 (offlineimap_6.3.2.orig.tar.gz) = fQxfibCX8U9il1HWuklPQ33QTcQ=
-SHA256 (offlineimap_6.3.2.orig.tar.gz) =
JAgqcOx6bS4mybA4UpW0iPGZNgf5lBAWD0kFpvmfF0M=
-SIZE (offlineimap_6.3.2.orig.tar.gz) = 92784
+MD5 (offlineimap_6.3.4.orig.tar.gz) = hMQgiP/McJJHP+924XwIKw==
+RMD160 (offlineimap_6.3.4.orig.tar.gz) = HlmyU19qYGvP+CtlPMjpF40fZss=
+SHA1 (offlineimap_6.3.4.orig.tar.gz) = K6PRFFo+NtDoG6DupzvUxbSGMeY=
+SHA256 (offlineimap_6.3.4.orig.tar.gz) =
oqxppYYFM46XFljpMOCV5MJhE4ODepN4uIwHSc03QzM=
+SIZE (offlineimap_6.3.4.orig.tar.gz) = 137313
Index: patches/patch-docs_MANUAL_rst
===
RCS file: patches/patch-docs_MANUAL_rst
diff -N patches/patch-docs_MANUAL_rst
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-docs_MANUAL_rst 10 Oct 2011 08:29:27 -
@@ -0,0 +1,21 @@
+$OpenBSD$
+--- docs/MANUAL.rst.orig Mon Oct 10 10:25:54 2011
docs/MANUAL.rstMon Oct 10 10:26:53 2011
+@@ -380,7 +380,7 @@
+ default.
+
+ Certificate checking
+-
++
+
+ Unfortunately, by default we will not verify the certificate of an IMAP
+ TLS/SSL server we connect to, so connecting by SSL is no guarantee
+@@ -397,7 +397,7 @@
+ certificate and CA certificate.
+
+ StartTLS
+-
++
+
+ If you have not configured your account to connect via SSL anyway,
+ OfflineImap will still attempt to set up an SSL connection via the
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/offlineimap/pkg/PLIST,v
retrieving revision 1.9
diff -u -p -r1.9 PLIST
--- pkg/PLIST 27 Apr 2011 12:59:27 - 1.9
+++ pkg/PLIST 10 Oct 2011 08:25:19 -
@@ -8,6 +8,8 @@ lib/python${MODPY_VERSION}/site-packages
lib/python${MODPY_VERSION}/site-packages/offlineimap/__init__.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/accounts.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/accounts.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/error.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/error.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Base.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Base.pyc
@@ -17,12 +19,16 @@ lib/python${MODPY_VERSION}/site-packages
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/IMAP.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatus.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatus.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatusSQLite.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatusSQLite.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Maildir.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Maildir.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/UIDMaps.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/UIDMaps.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/__init__.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/__init__.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplib2.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplib2.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplibutil.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplibutil.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/imapserver.py
[UPDATE] offlineimap 6.3.4
Simple update... Ok?
Index: Makefile
===
RCS file: /cvs/ports/mail/offlineimap/Makefile,v
retrieving revision 1.22
diff -u -p -r1.22 Makefile
--- Makefile16 Sep 2011 10:31:21 - 1.22
+++ Makefile8 Oct 2011 16:18:19 -
@@ -2,11 +2,10 @@
COMMENT= powerful IMAP/Maildir synchronization and reader support
-MODPY_EGG_VERSION= 6.3.2
+MODPY_EGG_VERSION= 6.3.4
DISTNAME= offlineimap_${MODPY_EGG_VERSION}.orig
CATEGORIES=mail
PKGNAME= offlineimap-${MODPY_EGG_VERSION}
-REVISION= 1
HOMEPAGE= http://offlineimap.org/
@@ -26,7 +25,7 @@ MODULES= lang/python
BUILD_DEPENDS= textproc/py-docutils
-WRKDIST= ${WRKDIR}/nicolas33-offlineimap-b49802f
+WRKDIST= ${WRKDIR}/nicolas33-offlineimap-94450e9
EXAMPLESDIR= ${PREFIX}/share/examples/offlineimap
Index: distinfo
===
RCS file: /cvs/ports/mail/offlineimap/distinfo,v
retrieving revision 1.13
diff -u -p -r1.13 distinfo
--- distinfo27 Apr 2011 12:59:27 - 1.13
+++ distinfo8 Oct 2011 16:17:28 -
@@ -1,5 +1,5 @@
-MD5 (offlineimap_6.3.2.orig.tar.gz) = c+9rRoXXZRJh+VkZSfaBvg==
-RMD160 (offlineimap_6.3.2.orig.tar.gz) = fKHo6nWpKlaxXzVLwk/wBZ3fw/0=
-SHA1 (offlineimap_6.3.2.orig.tar.gz) = fQxfibCX8U9il1HWuklPQ33QTcQ=
-SHA256 (offlineimap_6.3.2.orig.tar.gz) =
JAgqcOx6bS4mybA4UpW0iPGZNgf5lBAWD0kFpvmfF0M=
-SIZE (offlineimap_6.3.2.orig.tar.gz) = 92784
+MD5 (offlineimap_6.3.4.orig.tar.gz) = hMQgiP/McJJHP+924XwIKw==
+RMD160 (offlineimap_6.3.4.orig.tar.gz) = HlmyU19qYGvP+CtlPMjpF40fZss=
+SHA1 (offlineimap_6.3.4.orig.tar.gz) = K6PRFFo+NtDoG6DupzvUxbSGMeY=
+SHA256 (offlineimap_6.3.4.orig.tar.gz) =
oqxppYYFM46XFljpMOCV5MJhE4ODepN4uIwHSc03QzM=
+SIZE (offlineimap_6.3.4.orig.tar.gz) = 137313
Index: pkg/PLIST
===
RCS file: /cvs/ports/mail/offlineimap/pkg/PLIST,v
retrieving revision 1.9
diff -u -p -r1.9 PLIST
--- pkg/PLIST 27 Apr 2011 12:59:27 - 1.9
+++ pkg/PLIST 8 Oct 2011 16:24:41 -
@@ -8,6 +8,8 @@ lib/python${MODPY_VERSION}/site-packages
lib/python${MODPY_VERSION}/site-packages/offlineimap/__init__.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/accounts.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/accounts.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/error.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/error.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Base.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Base.pyc
@@ -17,12 +19,16 @@ lib/python${MODPY_VERSION}/site-packages
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/IMAP.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatus.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatus.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatusSQLite.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/LocalStatusSQLite.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Maildir.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/Maildir.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/UIDMaps.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/UIDMaps.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/__init__.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/folder/__init__.pyc
+lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplib2.py
+lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplib2.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplibutil.py
lib/python${MODPY_VERSION}/site-packages/offlineimap/imaplibutil.pyc
lib/python${MODPY_VERSION}/site-packages/offlineimap/imapserver.py
[UPDATE] py-crypto-2.3
Diff below update py-crypto to the lastest version, all tests pass on
amd64.
Comments? Ok?
Martin
Index: Makefile
===
RCS file: /cvs/ports/security/py-crypto/Makefile,v
retrieving revision 1.19
diff -u -p -r1.19 Makefile
--- Makefile16 Sep 2011 11:41:39 - 1.19
+++ Makefile30 Sep 2011 10:35:12 -
@@ -4,13 +4,12 @@ SHARED_ONLY = Yes
COMMENT = cryptographic tools for Python
-MODPY_EGG_VERSION =2.0.1
+MODPY_EGG_VERSION =2.3
DISTNAME = pycrypto-${MODPY_EGG_VERSION}
PKGNAME = py-crypto-${MODPY_EGG_VERSION}
-REVISION= 9
CATEGORIES = security devel
-HOMEPAGE = http://www.amk.ca/python/code/crypto.html
+HOMEPAGE = https://www.dlitz.net/software/pycrypto/
MAINTAINER = Will Maier
@@ -20,7 +19,7 @@ PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM = Yes
PERMIT_DISTFILES_FTP = Yes
-MASTER_SITES = http://www.amk.ca/files/python/crypto/
+MASTER_SITES = http://ftp.dlitz.net/pub/dlitz/crypto/pycrypto/
MODULES = lang/python
LIB_DEPENDS = devel/gmp
@@ -37,6 +36,6 @@ post-install:
do-regress: fake
cd ${WRKSRC} && ${MAKE_ENV} \
- PYTHONPATH=${WRKINST}${MODPY_SITEPKG} ${MODPY_BIN} test.py
+ PYTHONPATH=${WRKINST}${MODPY_SITEPKG} ${MODPY_BIN} setup.py test
.include
Index: distinfo
===
RCS file: /cvs/ports/security/py-crypto/distinfo,v
retrieving revision 1.5
diff -u -p -r1.5 distinfo
--- distinfo5 Apr 2007 17:26:09 - 1.5
+++ distinfo30 Sep 2011 10:01:01 -
@@ -1,5 +1,5 @@
-MD5 (pycrypto-2.0.1.tar.gz) = TVZ084mKVzaR/7M16NdJzQ==
-RMD160 (pycrypto-2.0.1.tar.gz) = XOk4ok939BTkJoDBfvm23I3pSi4=
-SHA1 (pycrypto-2.0.1.tar.gz) = x3ze/fsG5HSWkAE6mp4WAKsU4m8=
-SHA256 (pycrypto-2.0.1.tar.gz) = sI1O1UyUA8d3eKOAPlOk8z81m0LZT28+FKuxv0lB5uo=
-SIZE (pycrypto-2.0.1.tar.gz) = 154292
+MD5 (pycrypto-2.3.tar.gz) = K4Ec+/w0LYPuYUCX7/uBAQ==
+RMD160 (pycrypto-2.3.tar.gz) = W3sVqrP14zETuhl6Lt1inHJZU3Q=
+SHA1 (pycrypto-2.3.tar.gz) = lrM7d+cpiTvQg3oJpqp6ErUHADc=
+SHA256 (pycrypto-2.3.tar.gz) = TxHoX7zxOWA3NlD8La6PCI+bAB8H+20++y/LUzSYcYI=
+SIZE (pycrypto-2.3.tar.gz) = 331476
Index: patches/patch-Cipher___init___py
===
RCS file: patches/patch-Cipher___init___py
diff -N patches/patch-Cipher___init___py
--- patches/patch-Cipher___init___py11 Jun 2008 02:59:32 - 1.2
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,26 +0,0 @@
-$OpenBSD: patch-Cipher___init___py,v 1.2 2008/06/11 02:59:32 wcmaier Exp $
Cipher/__init__.py.origFri Feb 28 09:28:35 2003
-+++ Cipher/__init__.py Wed May 28 20:55:20 2008
-@@ -11,7 +11,6 @@ If you don't know which algorithm to choose, use AES b
- standard and has undergone a fair bit of examination.
-
- Crypto.Cipher.AES Advanced Encryption Standard
--Crypto.Cipher.ARC2Alleged RC2
- Crypto.Cipher.ARC4Alleged RC4
- Crypto.Cipher.Blowfish
- Crypto.Cipher.CAST
-@@ -19,12 +18,11 @@ Crypto.Cipher.DES The Data Encryption Standard
- in the past, but today its 56-bit keys are too
small.
- Crypto.Cipher.DES3Triple DES.
- Crypto.Cipher.IDEA
--Crypto.Cipher.RC5
- Crypto.Cipher.XOR The simple XOR cipher.
- """
-
--__all__ = ['AES', 'ARC2', 'ARC4',
-- 'Blowfish', 'CAST', 'DES', 'DES3', 'IDEA', 'RC5',
-+__all__ = ['AES', 'ARC4',
-+ 'Blowfish', 'CAST', 'DES', 'DES3',
-'XOR'
-]
-
Index: patches/patch-Hash_MD5_py
===
RCS file: patches/patch-Hash_MD5_py
diff -N patches/patch-Hash_MD5_py
--- patches/patch-Hash_MD5_py 29 Aug 2009 15:17:48 - 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,29 +0,0 @@
-$OpenBSD: patch-Hash_MD5_py,v 1.1 2009/08/29 15:17:48 fgsch Exp $
Hash/MD5.py.orig Sat Aug 29 14:48:13 2009
-+++ Hash/MD5.pySat Aug 29 14:48:34 2009
-@@ -3,11 +3,19 @@
-
- __revision__ = "$Id: patch-Hash_MD5_py,v 1.1 2009/08/29 15:17:48 fgsch Exp $"
-
--from md5 import *
-+try:
-+# The md5 module is deprecated in Python 2.6, so use hashlib when
possible.
-+import hashlib
-+def new(data=""):
-+return hashlib.md5(data)
-+digest_size = new().digest_size
-
--import md5
--if hasattr(md5, 'digestsize'):
--digest_size = digestsize
--del digestsize
--del md5
-+except ImportError:
-+from md5 import *
-+
-+import md5
-+if hasattr(md5, 'digestsize'):
-+digest_size = digestsize
-+del digestsize
-+del md5
-
Index: patches/patch-Hash_SHA_py
===
RCS file: patches/patch-Hash_SHA_py
diff -N patches/patch-Hash_SHA_py
--- patches/patch-Hash_SHA_py 29 Aug 2009 15:17:48 - 1.1
+++ /dev/null 1 Jan 1970 00:00
[NEW] net/py-ipy 0.75
Description: IPy is a Python module for handling IPv4 and IPv6 Addresses and Networks in a fashion similar to perl's Net::IP and friends. The IP class allows a comfortable parsing and handling for most notations in use for IPv4 and IPv6 Addresses and Networks. It can detect about a dozen different ways of expressing IP addresses and networks, parse them and distinguish between IPv4 and IPv6 addresses. Ok? py-ipy.tgz Description: application/tar-gz
[NEW] py-monthdelta 0.9.1
Simple python port... Description: Pythonic date computations with months Ok? py-monthdelta.tgz Description: application/tar-gz
[Update] pyflakes 0.5.0
Ok?
Index: Makefile
===
RCS file: /cvs/ports/devel/pyflakes/Makefile,v
retrieving revision 1.2
diff -u -p -r1.2 Makefile
--- Makefile16 Sep 2011 09:24:56 - 1.2
+++ Makefile26 Sep 2011 09:11:12 -
@@ -2,7 +2,7 @@
COMMENT = passive checker of Python programs
-MODPY_EGG_VERSION =0.4.0
+MODPY_EGG_VERSION =0.5.0
DISTNAME = pyflakes-${MODPY_EGG_VERSION}
REVISION= 0
Index: distinfo
===
RCS file: /cvs/ports/devel/pyflakes/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo18 Aug 2011 12:22:31 - 1.1.1.1
+++ distinfo9 Sep 2011 14:21:48 -
@@ -1,5 +1,5 @@
-MD5 (pyflakes-0.4.0.tar.gz) = YwpyUQquh1j0jPYOT6Fxdg==
-RMD160 (pyflakes-0.4.0.tar.gz) = spYmZEFMQRKLFl0ttZd7/I7Ld3c=
-SHA1 (pyflakes-0.4.0.tar.gz) = Q5tUe+O+Qx3pDr6ha3/So/Yg4/k=
-SHA256 (pyflakes-0.4.0.tar.gz) = 8S2Y66uuDo2j4GnZxwYh0VT2iL0G4HZoGzWZ7Xn7JsQ=
-SIZE (pyflakes-0.4.0.tar.gz) = 14794
+MD5 (pyflakes-0.5.0.tar.gz) = Vo2rJ8QuWCJ4eqimA4mGcg==
+RMD160 (pyflakes-0.5.0.tar.gz) = 9kq7/Bn8Xc166U9FdMRsMR2U3PY=
+SHA1 (pyflakes-0.5.0.tar.gz) = zjaI4mKX5RjIuSIgnDvs4z1RI1I=
+SHA256 (pyflakes-0.5.0.tar.gz) = U/l5SqE7mcygQlm0of5L0gfcOy192FunFYFP3q4u+M0=
+SIZE (pyflakes-0.5.0.tar.gz) = 17635
Re: [NEW] devel/py-pep8
On 17/08/11(Wed) 10:41, Martin Pieuchot wrote: > Another simple python port. > > Comment: > python style guide checker > > Description: > pep8 is a tool to check your Python code against some of the style > conventions in PEP 8. Resubmit as "pep8" instead of "py-pep8" because as rpointel@ pointed out, the py- prefix is needed for python libraries. Ok? pep8.tgz Description: application/tar-gz
[NEW] devel/py-pep8
Another simple python port. Comment: python style guide checker Description: pep8 is a tool to check your Python code against some of the style conventions in PEP 8. Comments? Ok? Martin py-pep8.tgz Description: application/tar-gz
[NEW] devel/pyflakes
Simple python port: Comment: passive checker of Python programs Description: Pyflakes is a program to analyze Python programs and detect various errors. It works by parsing the source file, not importing it, so it is safe to use on modules with side effects. Comments? Ok? pyflakes.tgz Description: application/tar-gz
Re: NEW: gnome-colors-icon-theme
On 05/05/11(Thu) 15:47, David Coppa wrote: > On Thu, May 5, 2011 at 3:16 PM, Martin Pieuchot > wrote: > > > Updated archive, no dependence and make use of pax -rw. > > Ok for me. I like this icon theme a lot. More comments or this can go in?
Resubmit: NEW: graphics/pngnq
On 14/01/11(Fri) 14:46, Martin Pieuchot wrote: > Hello, > > pngnq is a tool for quantizing PNG images in RGBA format, it's a > useful tool for saving some Kb when using png images on your website. > Description: Pngnq is a tool for quantizing PNG images in RGBA format. Pngnq is an adaptation by Stuart Coyle of Greg Roelf's pnqquant using Anthony Dekker's neuquant algorithm. The neuquant algorithm uses a neural network to optimise the color map selection. This is fast and quite accurate, giving good results on many types of images. Comments? Ok? Regards, Martin pngnq.tgz Description: application/tar-gz
Re: UPDATE: tig
On 12/04/11(Tue) 08:37, Landry Breuil wrote: > On Tue, Apr 12, 2011 at 12:47:03AM +0200, Tobias Ulmer wrote: > > Survived cursory testing on amd64, diff looks good. Ok by me. > > Fails to build on sparc64 with a weird error : > > ===> Building for tig-0.17 > cc -O2 -pipe -Wall -I/usr/local/include -DHAVE_CONFIG_H > '-DTIG_VERSION="0.17"' '-DSYSCONFDIR="/etc"' -c -o tig.o tig.c > tig.c: In function 'toggle_option': > tig.c:1916: error: subscripted value is neither array nor pointer Have you updated your base system (at least the includes) after miod's commits changing NULL from 0L to void * ? I think the problem comes from a weird use of nitems(NULL). Let me know, Martin
UPDATE: tig
Hello Tobias,
I know you are without internet but here is an update for tig. It works
well here on amd64. I've add the tigmanual.7 because when I look for
some documentation I always start with manuals. And there is no need for
the libncursw detection hack.
Note that you need a recent -current in order to build it (NULL should be
defined as a void *.
tig-0.17
Improvements:
Start rewrite of the revision graph renderer. Three modes are
supported UTF-8, ncurses line graphics, and ASCII. Also, enable
revision graph rendering by default.
Bug fixes:
Fix ncurses 2.8 issue by passing non-zero column size to
newwin(1).
Fix opening of diffs when browsing branches.
Regards,
Martin
Index: Makefile
===
RCS file: /cvs/ports/devel/tig/Makefile,v
retrieving revision 1.5
diff -u -p -r1.5 Makefile
--- Makefile15 Jan 2011 21:01:20 - 1.5
+++ Makefile8 Apr 2011 16:19:49 -
@@ -1,7 +1,7 @@
# $OpenBSD: Makefile,v 1.5 2011/01/15 21:01:20 jasper Exp $
COMMENT = ncurses GIT repository browser
-DISTNAME = tig-0.16.2
+DISTNAME = tig-0.17
CATEGORIES = devel
HOMEPAGE = http://jonas.nitro.dk/tig/
MAINTAINER = Tobias Ulmer
@@ -16,7 +16,7 @@ WANTLIB = c ncursesw
MODULES = converters/libiconv
RUN_DEPENDS = devel/git
-MASTER_SITES = ${HOMEPAGE}/releases/
+MASTER_SITES = ${HOMEPAGE}releases/
USE_GMAKE =Yes
CONFIGURE_STYLE = gnu
@@ -29,5 +29,6 @@ post-install:
${INSTALL_DATA} ${WRKBUILD}/manual.* ${PREFIX}/share/doc/tig
${INSTALL_MAN} ${WRKBUILD}/tig.1 ${PREFIX}/man/man1
${INSTALL_MAN} ${WRKBUILD}/tigrc.5 ${PREFIX}/man/man5
+ ${INSTALL_MAN} ${WRKBUILD}/tigmanual.7 ${PREFIX}/man/man7
.include
Index: distinfo
===
RCS file: /cvs/ports/devel/tig/distinfo,v
retrieving revision 1.3
diff -u -p -r1.3 distinfo
--- distinfo15 Jan 2011 21:01:20 - 1.3
+++ distinfo8 Apr 2011 16:12:04 -
@@ -1,5 +1,5 @@
-MD5 (tig-0.16.2.tar.gz) = 1ytdNDfbxTjqb2bHSYjXXg==
-RMD160 (tig-0.16.2.tar.gz) = LHmHXVyGZAJbtWp7y0DjQ2SsVKQ=
-SHA1 (tig-0.16.2.tar.gz) = d366EaT1On3Ms2Qf7aRNbv+2oC4=
-SHA256 (tig-0.16.2.tar.gz) = fTKQZB3LFXlJGgybN5pbaJcHaqqEWCb7NwJRDwdAbqI=
-SIZE (tig-0.16.2.tar.gz) = 162809
+MD5 (tig-0.17.tar.gz) = 83M0MZlCLFlRh3bbRI3sDg==
+RMD160 (tig-0.17.tar.gz) = MDwJIOfyuLi7dNZc8Ls+kUUB7NQ=
+SHA1 (tig-0.17.tar.gz) = pI6dCNlvi0m2XPhmAhEsfI4aSAQ=
+SHA256 (tig-0.17.tar.gz) = RNnJ1N7GiRHg9AJ1IYIGA5+EO8Oa93cYd35wCOlolPs=
+SIZE (tig-0.17.tar.gz) = 168918
Index: patches/patch-configure
===
RCS file: patches/patch-configure
diff -N patches/patch-configure
--- patches/patch-configure 15 Jan 2011 21:01:20 - 1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,19 +0,0 @@
-$OpenBSD: patch-configure,v 1.1 2011/01/15 21:01:20 jasper Exp $
-
-Correctly detect the wchar ncurses library.
-
configure.orig Tue Sep 21 14:44:10 2010
-+++ configure Sat Jan 15 19:58:35 2011
-@@ -3203,10 +3203,10 @@ done
-
- for ac_header in ncursesw/ncurses.h
- do :
-- ac_fn_c_check_header_mongrel "$LINENO" "ncursesw/ncurses.h"
"ac_cv_header_ncursesw_ncurses_h" "$ac_includes_default"
-+ ac_fn_c_check_header_mongrel "$LINENO" "ncurses.h"
"ac_cv_header_ncursesw_ncurses_h" "$ac_includes_default"
- if test "x$ac_cv_header_ncursesw_ncurses_h" = x""yes; then :
- cat >>confdefs.h <<_ACEOF
--#define HAVE_NCURSESW_NCURSES_H 1
-+/* #define HAVE_NCURSESW_NCURSES_H 1 */
- _ACEOF
- { $as_echo "$as_me:${as_lineno-$LINENO}: checking for library containing
initscr" >&5
- $as_echo_n "checking for library containing initscr... " >&6; }
Index: pkg/PLIST
===
RCS file: /cvs/ports/devel/tig/pkg/PLIST,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 PLIST
--- pkg/PLIST 24 Oct 2008 22:49:21 - 1.1.1.1
+++ pkg/PLIST 8 Apr 2011 17:05:19 -
@@ -1,7 +1,9 @@
@comment $OpenBSD: PLIST,v 1.1.1.1 2008/10/24 22:49:21 bernd Exp $
+@bin bin/test-graph
@bin bin/tig
@man man/man1/tig.1
@man man/man5/tigrc.5
+@man man/man7/tigmanual.7
share/doc/tig/
share/doc/tig/manual.html
share/doc/tig/manual.txt
NEW: keybinder & Hotot
Hi,
Want to try hotot [0]? Here is a port with its dependence {py-,}keybinder.
There is no release for hotot atm so I've mirrored the last hg checkout.
If you plan to commit it, can you mirror the archive? I'm not sure I
want to maintain this port.
I've a question though. Is it needed to add a "gtk-update-icon-cache"
when a port put an image under "share/pixmaps/" ?
Description (keybinder):
Keybinder is a library for registering global keyboard shortcuts.
It works with GTK-based applications using the X Window System.
This library originates in Tomboy and has been widely reused without
having a separate release. This package has taken the python bindings
from the Deskbar Applet project, and broken it out to a standalone
project. The library was subsequently rewritten in major parts.
Description (hotot):
Lightweight Microblogging Client based on Python, Gtk2 and Webkit.
Comments?
Regards,
Martin
[0] http://hotot.org
keybinder.tgz
Description: application/tar-gz
hotot.tgz
Description: application/tar-gz
Re: NEW: audio/soundconverter
On 05/04/11(Tue) 22:02, Landry Breuil wrote: > On Tue, Apr 05, 2011 at 04:37:41PM +0530, Martin Pieuchot wrote: > > [...] > > Comments? Ok? > > - lacks gtk-update-icon-cache and update-desktop-database > @exec/@unexec-delete Good to know. Is there any documentation explaining when I should add those? > [...] > - added dependency on gstreamer-lame, as it complains upon startup : > 'lame' gstreamer element not found, disabling MP3. > Not a strong need though.. It's Ok for me, I don't really care, I don't use mp3. By the way, what is the politic about adding dependencies on something that isn't distribuable on the CDs, is there any? > So here's the fixed version. Ok for me, thanks. Martin
NEW: audio/soundconverter
Hi, Description: SoundConverter is a simple GTK+ audio file converter for the GNOME environment. It reads anything GStreamer can read and writes to Ogg Vorbis, FLAC, AAC, WAV or MP3 formats. Comments? Ok? Martin soundconverter.tgz Description: application/tar-gz
Re: UPDATE: audio/py-mpd
On 16/01/11(Sun) 19:38, Martin Pieuchot wrote:
> Hi,
>
> Update for py-mpd to the latest version 0.3.0 & update the homepage.
>
> Tested here on amd64 with sonata.
>
Resubmit, works well here with sonata & mpd-add-similar (and with
mpd-0.16.2)
Comments? OK?
Martin
Index: Makefile
===
RCS file: /cvs/ports/audio/py-mpd/Makefile,v
retrieving revision 1.4
diff -u -p -r1.4 Makefile
--- Makefile30 Aug 2010 21:36:14 - 1.4
+++ Makefile5 Apr 2011 10:45:14 -
@@ -1,30 +1,30 @@
# $OpenBSD: Makefile,v 1.4 2010/08/30 21:36:14 fgsch Exp $
-COMMENT= Python MPD client library
+COMMENT = Python MPD client library
-MODPY_EGG_VERSION=0.2.0
-DISTNAME= python-mpd-${MODPY_EGG_VERSION}
-PKGNAME= py-mpd-${MODPY_EGG_VERSION}
-REVISION = 1
-CATEGORIES=audio
+MODPY_EGG_VERSION =0.3.0
+DISTNAME = python-mpd-${MODPY_EGG_VERSION}
+PKGNAME = py-mpd-${MODPY_EGG_VERSION}
+CATEGORIES = audio
-HOMEPAGE= http://www.musicpd.org/~jat/python-mpd/
+HOMEPAGE = http://jatreuman.indefero.net/p/python-mpd/
+
+MAINTAINER = Martin Pieuchot
# GPLv3
-PERMIT_PACKAGE_CDROM= Yes
-PERMIT_PACKAGE_FTP=Yes
-PERMIT_DISTFILES_CDROM= Yes
-PERMIT_DISTFILES_FTP= Yes
+PERMIT_PACKAGE_CDROM = Yes
+PERMIT_PACKAGE_FTP = Yes
+PERMIT_DISTFILES_CDROM =Yes
+PERMIT_DISTFILES_FTP = Yes
-MASTER_SITES= ${MASTER_SITE_PYPI:=p/python-mpd/}
+MASTER_SITES = ${MASTER_SITE_PYPI:=p/python-mpd/}
-MODULES= lang/python
+MODULES = lang/python
-NO_REGRESS=Yes
+NO_REGRESS = Yes
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/py-mpd
- ${INSTALL_DATA} ${WRKSRC}/doc/commands.txt \
- ${PREFIX}/share/doc/py-mpd
+ ${INSTALL_DATA} ${WRKSRC}/doc/commands.txt ${PREFIX}/share/doc/py-mpd
.include
Index: distinfo
===
RCS file: /cvs/ports/audio/py-mpd/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo19 Apr 2008 15:08:30 - 1.1.1.1
+++ distinfo5 Apr 2011 02:47:29 -
@@ -1,5 +1,5 @@
-MD5 (python-mpd-0.2.0.tar.gz) = mDUH7jCLWJVggVbaEMPPyw==
-RMD160 (python-mpd-0.2.0.tar.gz) = e7S+mHEaF8p1AQOuSIJPaEEksfA=
-SHA1 (python-mpd-0.2.0.tar.gz) = d69Y1N+pbBZlyw33PLJk4c4YLXQ=
-SHA256 (python-mpd-0.2.0.tar.gz) = tmF0k0JZMDtbIzThinoDTG2GRwdzGShvkTFR/h46DMc=
-SIZE (python-mpd-0.2.0.tar.gz) = 17383
+MD5 (python-mpd-0.3.0.tar.gz) = WzhJsTHi+xLyUUNFl9ZWNQ==
+RMD160 (python-mpd-0.3.0.tar.gz) = LUVVOQygXVDbPC6wfhkkFGoP4Jo=
+SHA1 (python-mpd-0.3.0.tar.gz) = v1wh4i04BdDh34Zd/eUDTQHVNCA=
+SHA256 (python-mpd-0.3.0.tar.gz) = AoEuuh0uD0bjdFf1pvojuiA2IuS8qwoZsmXmawjNIbQ=
+SIZE (python-mpd-0.3.0.tar.gz) = 20507
Index: patches/patch-mpd_py
===
RCS file: patches/patch-mpd_py
diff -N patches/patch-mpd_py
--- patches/patch-mpd_py19 Apr 2008 15:08:31 - 1.1.1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,12 +0,0 @@
-$OpenBSD: patch-mpd_py,v 1.1.1.1 2008/04/19 15:08:31 ajacoutot Exp $
mpd.py.origSat Apr 19 12:10:21 2008
-+++ mpd.py Sat Apr 19 12:10:43 2008
-@@ -293,7 +293,7 @@ class MPDClient(object):
- msg = "getaddrinfo returns an empty list"
- for res in socket.getaddrinfo(host, port, socket.AF_UNSPEC,
- socket.SOCK_STREAM, socket.IPPROTO_TCP,
-- socket.AI_ADDRCONFIG):
-+ socket.AF_INET):
- af, socktype, proto, canonname, sa = res
- try:
- self._sock = socket.socket(af, socktype, proto)
UPDATE: audio/mpd
Hi Tobias,
Here is an updated version of your diff for mpd-0.16.2 . Previously discussed
issues (http://marc.info/?l=openbsd-ports&m=130114470610474&w=2) seem to have
been fixed.
I've tested the libao and the httpd output without any problem on amd64.
The manpage is also well displayed with mandoc.
Regards,
Martin
Index: Makefile
===
RCS file: /cvs/ports/audio/mpd/Makefile,v
retrieving revision 1.37
diff -u -p -r1.37 Makefile
--- Makefile14 Mar 2011 09:17:26 - 1.37
+++ Makefile4 Apr 2011 17:44:05 -
@@ -1,8 +1,7 @@
# $OpenBSD: Makefile,v 1.37 2011/03/14 09:17:26 sthen Exp $
COMMENT = Music Player Daemon
-DISTNAME = mpd-0.15.12
-REVISION = 6
+DISTNAME = mpd-0.16.2
CATEGORIES = audio
HOMEPAGE = http://www.musicpd.org/
MAINTAINER = Tobias Ulmer
@@ -13,8 +12,8 @@ PERMIT_PACKAGE_FTP = Yes
PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
-WANTLIB = c m pthread ogg z stdc++ sndio curl glib-2.0 gthread-2.0
-WANTLIB += audiofile ao samplerate FLAC id3tag mad mikmod
+WANTLIB += c m pthread ogg z stdc++ sndio curl glib-2.0 gthread-2.0
+WANTLIB += audiofile ao samplerate sndfile FLAC id3tag mad mikmod
MASTER_SITES = ${MASTER_SITE_SOURCEFORGE:=musicpd/}
@@ -51,11 +50,19 @@ CONFIGURE_ARGS += --enable-ao \
--disable-modplug \
--disable-sidplay \
--disable-ffmpeg \
- --enable-mpc
+ --enable-mpc \
+ --disable-libwrap \
+ --enable-lsr \
+ --enable-audiofile \
+ --disable-gme \
+ --enable-mad \
+ --disable-mpg123 \
+ --enable-sndfile \
+ --disable-twolame-encoder
+
# OggFLAC and tremor tests, initgroups(3)
CONFIGURE_ENV =LDFLAGS="-L${LOCALBASE}/lib" \
CPPFLAGS="-I${LOCALBASE}/include -D_BSD_SOURCE"
-USE_GROFF =Yes
.if ${FLAVOR:L:Mtremor}
CONFIGURE_ARGS += --with-tremor \
@@ -76,7 +83,9 @@ LIB_DEPENDS +=audio/libvorbis \
net/libshout \
audio/faad>=2.6.1p0 \
audio/libmpcdec
-CONFIGURE_ARGS += --enable-aac
+CONFIGURE_ARGS += --enable-aac \
+ --enable-shout \
+ --enable-vorbis-encoder
.endif
USE_LIBTOOL = Yes
Index: distinfo
===
RCS file: /cvs/ports/audio/mpd/distinfo,v
retrieving revision 1.10
diff -u -p -r1.10 distinfo
--- distinfo23 Aug 2010 14:46:55 - 1.10
+++ distinfo4 Apr 2011 16:59:28 -
@@ -1,5 +1,5 @@
-MD5 (mpd-0.15.12.tar.gz) = W9YcISTUOIsTmLMvLHFDig==
-RMD160 (mpd-0.15.12.tar.gz) = QUgnNCADCYPdKuKiN6qiOc9kO9M=
-SHA1 (mpd-0.15.12.tar.gz) = fZObgk2YmjG+NmA/jVA+N3RJxj8=
-SHA256 (mpd-0.15.12.tar.gz) = p2oa1OAxLL2EwJ9S0/eLpcEvLrmRRSSJsEsXlkrftBc=
-SIZE (mpd-0.15.12.tar.gz) = 483928
+MD5 (mpd-0.16.2.tar.gz) = p1rXFcw4HcXv7/bb8xOpVA==
+RMD160 (mpd-0.16.2.tar.gz) = ma4X2KyatcMiNwiLrjp0EZ4ksd8=
+SHA1 (mpd-0.16.2.tar.gz) = kL9F9FZ+XW+C0nn3jaGIFLD0OsQ=
+SHA256 (mpd-0.16.2.tar.gz) = RWec8xqZ3Fuv0lMcrVrK6QTFy0xsuaLiUyfglLjotkE=
+SIZE (mpd-0.16.2.tar.gz) = 619586
Index: patches/patch-Makefile_in
===
RCS file: /cvs/ports/audio/mpd/patches/patch-Makefile_in,v
retrieving revision 1.2
diff -u -p -r1.2 patch-Makefile_in
--- patches/patch-Makefile_in 23 Aug 2010 14:46:55 - 1.2
+++ patches/patch-Makefile_in 4 Apr 2011 17:07:20 -
@@ -1,12 +1,12 @@
$OpenBSD: patch-Makefile_in,v 1.2 2010/08/23 14:46:55 jasper Exp $
Makefile.in.orig Wed Jul 21 02:42:27 2010
-+++ Makefile.inSun Aug 22 18:45:03 2010
-@@ -1447,7 +1447,7 @@ SPARSE_CPPFLAGS = $(DEFAULT_INCLUDES) -I$(shell $(CC)
+--- Makefile.inMon Apr 4 22:31:44 2011
Makefile.inMon Apr 4 22:31:30 2011
+@@ -2254,7 +2254,7 @@
# Documentation
#
man_MANS = doc/mpd.1 doc/mpd.conf.5
-doc_DATA = AUTHORS COPYING NEWS README UPGRADING doc/mpdconf.example
+doc_DATA = AUTHORS COPYING NEWS README UPGRADING
- DOCBOOK_FILES = doc/protocol.xml doc/user.xml doc/developer.xml
doc/sticker.xml
+ DOCBOOK_FILES = doc/protocol.xml doc/user.xml doc/developer.xml
@ENABLE_DOCUMENTATION_TRUE@protocoldir = $(docdir)/protocol
@ENABLE_DOCUMENTATION_TRUE@protocol_DATA = $(wildcard doc/protocol/*.html)
Index: patches/patch-configure
===
RCS file: /cvs/ports/audio/mpd/patches/patch-configure,v
retrieving revision 1.12
diff -u -p -r1.12 patch-configure
--- patches/patch-configure 23 Aug 2010 14:46:55 -
UPDATE: audio/py-mpd
Hi,
Update for py-mpd to the latest version 0.3.0 & update the homepage.
Tested here on amd64 with sonata.
Comments? OK?
Martin
Index: Makefile
===
RCS file: /home/aer/cvs/ports/audio/py-mpd/Makefile,v
retrieving revision 1.4
diff -u -p -r1.4 Makefile
--- Makefile30 Aug 2010 21:36:14 - 1.4
+++ Makefile16 Jan 2011 13:56:49 -
@@ -2,13 +2,12 @@
COMMENT= Python MPD client library
-MODPY_EGG_VERSION=0.2.0
+MODPY_EGG_VERSION=0.3.0
DISTNAME= python-mpd-${MODPY_EGG_VERSION}
PKGNAME= py-mpd-${MODPY_EGG_VERSION}
-REVISION = 1
CATEGORIES=audio
-HOMEPAGE= http://www.musicpd.org/~jat/python-mpd/
+HOMEPAGE= http://jatreuman.indefero.net/p/python-mpd/
# GPLv3
PERMIT_PACKAGE_CDROM= Yes
@@ -24,7 +23,6 @@ NO_REGRESS= Yes
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/py-mpd
- ${INSTALL_DATA} ${WRKSRC}/doc/commands.txt \
- ${PREFIX}/share/doc/py-mpd
+ ${INSTALL_DATA} ${WRKSRC}/doc/commands.txt ${PREFIX}/share/doc/py-mpd
.include
Index: distinfo
===
RCS file: /home/aer/cvs/ports/audio/py-mpd/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo19 Apr 2008 15:08:30 - 1.1.1.1
+++ distinfo14 Jan 2011 14:50:35 -
@@ -1,5 +1,5 @@
-MD5 (python-mpd-0.2.0.tar.gz) = mDUH7jCLWJVggVbaEMPPyw==
-RMD160 (python-mpd-0.2.0.tar.gz) = e7S+mHEaF8p1AQOuSIJPaEEksfA=
-SHA1 (python-mpd-0.2.0.tar.gz) = d69Y1N+pbBZlyw33PLJk4c4YLXQ=
-SHA256 (python-mpd-0.2.0.tar.gz) = tmF0k0JZMDtbIzThinoDTG2GRwdzGShvkTFR/h46DMc=
-SIZE (python-mpd-0.2.0.tar.gz) = 17383
+MD5 (python-mpd-0.3.0.tar.gz) = WzhJsTHi+xLyUUNFl9ZWNQ==
+RMD160 (python-mpd-0.3.0.tar.gz) = LUVVOQygXVDbPC6wfhkkFGoP4Jo=
+SHA1 (python-mpd-0.3.0.tar.gz) = v1wh4i04BdDh34Zd/eUDTQHVNCA=
+SHA256 (python-mpd-0.3.0.tar.gz) = AoEuuh0uD0bjdFf1pvojuiA2IuS8qwoZsmXmawjNIbQ=
+SIZE (python-mpd-0.3.0.tar.gz) = 20507
Index: patches/patch-mpd_py
===
RCS file: patches/patch-mpd_py
diff -N patches/patch-mpd_py
--- patches/patch-mpd_py19 Apr 2008 15:08:31 - 1.1.1.1
+++ /dev/null 1 Jan 1970 00:00:00 -
@@ -1,12 +0,0 @@
-$OpenBSD: patch-mpd_py,v 1.1.1.1 2008/04/19 15:08:31 ajacoutot Exp $
mpd.py.origSat Apr 19 12:10:21 2008
-+++ mpd.py Sat Apr 19 12:10:43 2008
-@@ -293,7 +293,7 @@ class MPDClient(object):
- msg = "getaddrinfo returns an empty list"
- for res in socket.getaddrinfo(host, port, socket.AF_UNSPEC,
- socket.SOCK_STREAM, socket.IPPROTO_TCP,
-- socket.AI_ADDRCONFIG):
-+ socket.AF_INET):
- af, socktype, proto, canonname, sa = res
- try:
- self._sock = socket.socket(af, socktype, proto)
Re: update: audio/mpd
On 16/01/11(Sun) 05:31, Tobias Ulmer wrote: > Update mpd to version 0.16.1 > > This thing has about 17 billion features, so please test. Works well here on amd64 even http streaming. Martin
Re: NEW: x11/dmz-cursor-themes
On 31/12/10(Fri) 18:50, Martin Pieuchot wrote:
> Hi,
>
> Two very popular cursor themes:
>
> Cursor themes derived from the Industrial theme developed for the Ximian
> GNOME desktop. Black and White flavors.
>
> I borrowed the explanation of "how to use" the new cursors from the
> xcursor-themes port.
>
> Comments? OK?
Any interest in committing this [0] simple port?
Regards,
Martin
[0] http://marc.info/?l=openbsd-ports&m=12938019977&q=p3
# $OpenBSD$
COMMENT = style neutral scalable cursor themes
VERSION = 0.4
DISTNAME = vanilla-dmz-${VERSION}
PKGNAME = dmz-cursor-themes-${VERSION}
REVISION =
CATEGORIES =x11
HOMEPAGE = http://jimmac.musichall.cz/themes.php?skin=7
MAINTAINER =Martin Pieuchot
# CC by-sa 3.0
PERMIT_PACKAGE_CDROM = Yes
PERMIT_PACKAGE_FTP =Yes
PERMIT_DISTFILES_CDROM =Yes
PERMIT_DISTFILES_FTP = Yes
MASTER_SITES= http://jimmac.musichall.cz/zip/
DISTFILES = ${DISTNAME}.tar.bz2 \
vanilla-dmz-aa-${VERSION}.tar.bz2
PKG_ARCH= *
NO_BUILD= Yes
NO_REGRESS= Yes
WRKDIST=${WRKDIR}
DMZDIR= ${PREFIX}/lib/X11/icons/dmz/cursors
DMZAADIR= ${PREFIX}/lib/X11/icons/dmz-aa/cursors
do-install:
${INSTALL_DATA_DIR} ${DMZDIR}
${INSTALL_DATA} ${WRKSRC}/Vanilla-DMZ/cursors/* ${DMZDIR}
${INSTALL_DATA_DIR} ${DMZAADIR}
${INSTALL_DATA} ${WRKSRC}/Vanilla-DMZ-AA/cursors/* ${DMZAADIR}
.include
NEW: graphics/pngnq
Hello,
pngnq is a tool for quantizing PNG images in RGBA format, it's a
useful tool for saving some Kb when using png images on your website.
I have one question about setting some flags:
CONFIGURE_ENV+= LDFLAGS="`pkg-config --libs libpng`" \
CPPFLAGS="-I${LOCALBASE}/include/libpng"
Is there a better way to set the LDFLAGS? Because idiom used in the
original Makefile (${shell libpng-config}) is not understood.
Comments? Ok?
Martin
pngnq.tgz
Description: application/tar-gz
NEW: x11/dmz-cursor-themes
Hi, Two very popular cursor themes: Cursor themes derived from the Industrial theme developed for the Ximian GNOME desktop. Black and White flavors. I borrowed the explanation of "how to use" the new cursors from the xcursor-themes port. Comments? OK? Martin dmz-cursor-themes.tgz Description: application/tar-gz
NEW: audio/tagpy
Hello ports@, I made a port of tagpy which is a python wrapper for taglib. It's used by sonata to edit tags. It works well here on i386. Any comment will be appreciated as it's my first attempt to make a port. Martin tagpy.tgz Description: application/compressed-tar
No more spell check enchant/gtkspell?
Hello ports, Yesterday I upgraded my ports, I am runing a -current i386 box, and after that spell checking stop to work in the following applications: - claws-mail (enchant) - pidgin-gtkspell - gedit (gtkspell plugin) Aspell and Enchant work well on the command line. Claws-mail give me a curious error: Couldn't initialize None dictionary: (null) Couldn't initialize None speller. I also tried sylpheed and the problem was the same: no dictionary available. Here is the output of my "aspell dicts" [1] And my pkg_info [2] Does someone have a similar issue or know how to solve this problem? Thanks, Martin [1] en en-variant_0 en-variant_1 en-variant_2 en-w_accents en-wo_accents en_CA en_CA-w_accents en_CA-wo_accents en_GB en_GB-ise en_GB-ise-w_accents en_GB-ise-wo_accents en_GB-ize en_GB-ize-w_accents en_GB-ize-wo_accents en_GB-w_accents en_GB-wo_accents en_US en_US-w_accents en_US-wo_accents fr-40 fr fr-60 fr-80 fr-lrg fr-med fr-sml fr_CH-40 fr_CH fr_CH-60 fr_CH-80 fr_CH-lrg fr_CH-med fr_CH-sml fr_FR-40 fr_FR fr_FR-60 fr_FR-80 fr_FR-lrg fr_FR-med fr_FR-sml pt pt_BR pt_PT [2] ORBit2-2.14.16 aalib-1.4 abiword-2.4.6p5 aspell-0.60.6p3 aspell-fr-0.50_3p1 aspell-pt-0.50_2p1 atk-1.24.0 autoconf-2.13p1 autoconf-2.52p3 autoconf-2.59p3 autoconf-2.61p3 automake-1.4.6 automake-1.8.5p1 automake-1.9.6p2 babl-0.0.22 bash-3.2.48 bzip2-1.0.5 cairo-1.8.4 cdparanoia-3.a9.8p0 claws-mail-3.6.1p0-ldap colorls-4.2 cups-1.2.7p14 curl-7.19.2 cyrus-sasl-2.1.22p4 dbus-1.2.6p0 dbus-glib-0.78 dbus-python-0.83.0p0 desktop-file-utils-0.15 detex-2.8 djvulibre-3.5.21 docbook-4.4 docbook-dsssl-1.72 dvi2tty-5.3.1 enchant-1.4.2p0 esound-0.2.41v0 evince-2.24.2p2 exo-0.3.4p5 faac-1.26 faad-2.6.1 fam-2.7.0p6 feh-1.3.4 ffmpeg-20080620p5 flac-1.2.1p0 fribidi-0.10.4p0 gconf2-2.24.0 gedit-2.24.2p0 gegl-0.0.22 gettext-0.17p0 ghostscript-8.63p2-gtk ghostscript-fonts-8.11p0 giblib-1.2.4p3 gimp-2.6.3 glib-1.2.10p3 glib2-2.18.3p0 glitz-0.5.6p0 gmake-3.81p0 gnome-doc-utils-0.10.3p7 gnome-icon-theme-2.24.0 gnome-keyring-2.24.1 gnome-mime-data-2.18.0p0 gnome-user-docs-2.24.1 gnome-vfs2-2.24.0 gnugetopt-1.1.4p0 gnupg-1.4.9 gnutls-2.6.2 gperf-3.0.1 gpgme-1.1.5 gtk+-1.2.10p7 gtk+2-2.14.5p1 gtk2-clearlooks-engine-0.6.2p6 gtksourceview-2.4.1 gtkspell-2.0.15 help2man-1.29p0 hicolor-icon-theme-0.10p4 id3lib-3.8.3p2 ijs-0.35 imlib2-1.4.0p0 intltool-0.40.5 iodbc-3.52.4p2 iso-codes-3.3p0 iso8879-1986 jasper-1.900.1p0 jpeg-6bp3 lame-3.98.2 lcms-1.17p0 libIDL-0.8.12 liba52-0.7.4p2 libao-0.8.8p2 libart-2.3.20p0 libaudiofile-0.2.6p1 libbonobo-2.24.0p0 libbonoboui-2.24.0 libcroco-0.6.1p0 libdvbpsi-0.1.4p1 libdvd-0.3p2 libdvdnav-20051102p2 libdvdread-0.9.7p1 libebml-0.7.7 liberation-fonts-1.04 libetpan-0.57 libexif-0.6.16p0 libgadu-1.8.1v0 libgcrypt-1.4.3p0 libglade2-2.6.2p2 libgnome-2.24.1 libgnomecanvas-2.20.1.1p4 libgnomeprint-2.18.5 libgnomeprintui-2.18.3 libgnomeui-2.24.0 libgpg-error-1.5 libgsf-1.14.11 libiconv-1.12 libid3tag-0.15.1bp0 libidn-1.11 libltdl-1.5.26 libmad-0.15.1bp1 libmatroska-0.8.1 libmikmod-3.1.10p6 libmng-1.0.10 libmp4v2-1.5.0.1 libmpcdec-1.2.6 libmpeg2-0.5.1 libmspack-20060920a libnotify-0.4.5 libogg-1.1.3 librsvg-2.22.3 libsamplerate-0.1.4 libsexy-0.1.11p1 libshout-2.2.2p0 libsndfile-1.0.17 libspectre-0.2.2 libtasn1-1.5 libtheora-1.0beta3 libtool-1.5.26p0 libungif-4.1.4p1 libvorbis-1.2.0p0 libwmf-0.2.8.3p4 libxfce4mcs-4.4.3 libxfce4util-4.4.3 libxfcegui4-4.4.3 libxml-2.6.32p2 libxslt-1.1.24p0 lzma-4.32.7 lzo-1.08p1 meanwhile-1.0.2p1 metaauto-0.9 mozilla-firefox-3.0.5p0 mpd-0.13.2p2 mplayer-1.0rc2p16-esd mysql-client-5.0.75 nasm-2.05.01 ncmpc-20070917p1 nspr-4.7.3 nss-3.12 openbox-3.4.7.2p0 openldap-client-2.3.43 p5-XML-Parser-2.36p1 pango-1.22.3 pcre-7.8 pidgin-2.5.3-gtkspell png-1.2.33 poppler-0.10.0p0 popt-1.7p0 postgresql-client-8.3.5 postgresql-server-8.3.5 ps2eps-1.64 psutils-0.17p0-a4 py-Numeric-23.1p2 py-cairo-1.4.12p0 py-gnome-2.22.3p0 py-gobject-2.16.0 py-gtk2-2.12.1p2 py-gtksourceview-2.4.0 py-libxml-2.6.32p0 py-mpd-0.2.0 py-orbit-2.14.3p2 py-setuptools-0.6c9 py-xml-0.8.4p6 py-zsi-2.0 python-2.5.4 qt3-mt-3.8p2 rarian-0.8.1p2 screen-4.0.3p1 sdl-1.2.13p10 sdl-image-1.2.4p5 shared-mime-info-0.51p0 silc-toolkit-1.1.8 sonata-1.5.3 speex-1.2rc1 sqlite-2.8.17p1 sqlite3-3.6.4p0 startup-notification-0.9 t1lib-5.1.0p1 t1utils-1.32 tango-icon-theme-0.8.1p4 tango-icon-theme-extras-0.1.0p2 tcl-8.5.6 texlive_base-2008p8 texlive_texmf-full-2008p0 texlive_texmf-minimal-2008p0 thunar-0.9.3 tiff-3.8.2p2 transmission-1.42 unzip-5.52p0 uuid-1.6.0 vim-7.2.25-no_x11 vlc-0.8.6ip2 wpi-firmware-3.0 wxWidgets-gtk2-2.8.7p1 x264-20080713p2 xchat-2.8.6 xdg-utils-1.0.2p2 xfce-mcs-manager-4.4.3 xfce4-icon-theme-4.4.3p0 xfce4-panel-4.4.3 xmms-1.2.11p4 xulrunner-1.8.1.19p1 xvidcore-1.1.3 yasm-0.7.2 yelp-2.18.1p8 zip-3.0
Re: [systrace: deny] gstreamer-plugins-good
Hi,
On Sat, 2008-11-15 at 23:11 +, Jacob Meuser wrote:
> On Sat, Nov 15, 2008 at 05:39:54PM -0200, Martin Pieuchot wrote:
> > Hello ports,
> >
> > I tried to install gstreamer-plugins-good from port today and I
> > encountered a problem during "make configure", see the log below.
> >
> > The problem is that the autogen.sh script calls autopoint, that
> > try to initiate a cvs checkout forbidden by the systrace policy. I
> > wonder if this issue is related to my system, see my mk.conf below, or
> > if anyone had already encountered this problem?
> >
> > I fixed it by commenting the autopoint call. Is it really necessary for
> > the package creation? Here is the little patch...
> >
> > Thanks for your explanations,
>
> the patch is needed, autopoint maybe not. I've never seen this
> happen and I always use systrace. are you building ports as
> root?
No I am not, I am using a wheel/operator/wsrc user... Can you reproduce
that problem? Any clue?
>
> >
> > Martin
> >
> >
> > Index: patches/patch-autogen_sh
> > ===
> > RCS file: patches/patch-autogen_sh
> > diff -N patches/patch-autogen_sh
> > --- /dev/null 1 Jan 1970 00:00:00 -
> > +++ patches/patch-autogen_sh15 Nov 2008 20:19:44 -
> > @@ -0,0 +1,24 @@
> > +--- autogen.sh.origSat Nov 15 17:12:33 2008
> > autogen.sh Sat Nov 15 17:12:51 2008
> > +@@ -63,14 +63,14 @@
> > +
> > + # autopoint
> > + #older autopoint (< 0.12) has a tendency to complain about
> > mkinstalldirs
> > +-if test -x mkinstalldirs; then rm mkinstalldirs; fi
> > ++#if test -x mkinstalldirs; then rm mkinstalldirs; fi
> > + #first remove patch if necessary, then run autopoint, then reapply
> > +-if test -f po/Makefile.in.in;
> > +-then
> > +- patch -p0 -R < common/gettext.patch
> > +-fi
> > +-tool_run "$autopoint --force"
> > +-patch -p0 < common/gettext.patch
> > ++#if test -f po/Makefile.in.in;
> > ++#then
> > ++# patch -p0 -R < common/gettext.patch
> > ++#fi
> > ++#tool_run "$autopoint --force"
> > ++#patch -p0 < common/gettext.patch
> > +
> > + tool_run "$aclocal" "-I m4 -I common/m4 $ACLOCAL_FLAGS"
> > + tool_run "$libtoolize" "--copy --force"
> >
> >
> >
> > #
> > # /etc/mk.conf - Aer
> > #
> >
> > PIPE= -pipe
> > SUDO= /usr/bin/sudo
> > CLEANDEPENDS= No
> > PLIST_DB= ${PORTSDIR}/plist
> > WRKOBJDIR= /usr/obj/ports
> > USE_SYSTRACE= Yes
> > #INSTALL_STRIP= -s
> >
> > XENOCARA_BUILD_DRI=Yes
> >
> > KERBEROS5= Yes
> > #DEBUG=Yes
> > #DEBUGLIBS=Yes
> > #WARNINGS=Yes
> >
> >
> >
> > ===> Patching for gst-plugins-good-0.10.8
> > ===> Configuring for gst-plugins-good-0.10.8
> > cd /usr/obj/ports/gst-plugins-good-0.10.8/gst-plugins-good-0.10.8 &&
> > AUTOCONF_VERSION=2.61 AUTOMAKE_VERSION=1.9 NOCONFIGURE="true" ./autogen.sh
> > + check for build tools
> > checking for autoconf >= 2.52 ... found 2.61, ok.
> > checking for automake >= 1.7 ... found 1.9.6, ok.
> > checking for autopoint >= 0.11.5 ... found 0.17, ok.
> > checking for libtoolize >= 1.5.0 ... found 1.5.26, ok.
> > checking for pkg-config >= 0.8.0 ... found 0.21, ok.
> > + checking for autogen.sh options
> > This autogen script will automatically run ./configure as:
> > ./configure --enable-maintainer-mode --enable-gtk-doc --enable-plugin-docs
> > To pass any additional options, please specify them on the ./autogen.sh
> > command line.
> > Hmm... Looks like a unified diff to me...
> > The text leading up to this was:
> > --
> > |--- po/Makefile.in.in.orig 2006-01-07 12:03:45.0 +0100
> > |+++ po/Makefile.in.in 2006-01-07 12:04:23.0 +0100
> > --
> > Patching file po/Makefile.in.in using Plan A...
> > Hunk #1 succeeded at 11.
> > Hunk #2 succeeded at 305.
> > done
> > + running autopoint --force ...
> > systrace: deny user: aer, prog: /usr/bin/ssh, pid: 586(0)[1815], policy:
> > /usr/bin/make, filters: 194, syscall: native-seteuid(183), uid: 1000,
> > uname: aer
> > PRIV_END: seteuid: Operation not permitted
> > cvs [checkout aborted]: end of file from server (consult above messages if
> > any)
> > find: archive: No such file or directory
> > find: archive: No such file or directory
> > autopoint: *** infrastructure files for version 0.11.5 not found; this is
> > autopoint from GNU gettext-tools 0.17
> > autopoint: *** Stop.
> >
> > autopoint --force failed
> > *** Error code 1
> >
> > Stop in /usr/ports/multimedia/gstreamer-0.10/plugins-good (line 173 of
> > Makefile).
> > *** Error code 1
> >
> > Stop in /usr/ports/multimedia/gstreamer-0.10/plugins-good (line 2144 of
> > /usr/ports/infrastructure/mk/bsd.port.mk).
> >
>
[systrace: deny] gstreamer-plugins-good
Hello ports,
I tried to install gstreamer-plugins-good from port today and I
encountered a problem during "make configure", see the log below.
The problem is that the autogen.sh script calls autopoint, that
try to initiate a cvs checkout forbidden by the systrace policy. I
wonder if this issue is related to my system, see my mk.conf below, or
if anyone had already encountered this problem?
I fixed it by commenting the autopoint call. Is it really necessary for
the package creation? Here is the little patch...
Thanks for your explanations,
Martin
Index: patches/patch-autogen_sh
===
RCS file: patches/patch-autogen_sh
diff -N patches/patch-autogen_sh
--- /dev/null 1 Jan 1970 00:00:00 -
+++ patches/patch-autogen_sh15 Nov 2008 20:19:44 -
@@ -0,0 +1,24 @@
+--- autogen.sh.origSat Nov 15 17:12:33 2008
autogen.sh Sat Nov 15 17:12:51 2008
+@@ -63,14 +63,14 @@
+
+ # autopoint
+ #older autopoint (< 0.12) has a tendency to complain about mkinstalldirs
+-if test -x mkinstalldirs; then rm mkinstalldirs; fi
++#if test -x mkinstalldirs; then rm mkinstalldirs; fi
+ #first remove patch if necessary, then run autopoint, then reapply
+-if test -f po/Makefile.in.in;
+-then
+- patch -p0 -R < common/gettext.patch
+-fi
+-tool_run "$autopoint --force"
+-patch -p0 < common/gettext.patch
++#if test -f po/Makefile.in.in;
++#then
++# patch -p0 -R < common/gettext.patch
++#fi
++#tool_run "$autopoint --force"
++#patch -p0 < common/gettext.patch
+
+ tool_run "$aclocal" "-I m4 -I common/m4 $ACLOCAL_FLAGS"
+ tool_run "$libtoolize" "--copy --force"
#
# /etc/mk.conf - Aer
#
PIPE= -pipe
SUDO= /usr/bin/sudo
CLEANDEPENDS= No
PLIST_DB= ${PORTSDIR}/plist
WRKOBJDIR= /usr/obj/ports
USE_SYSTRACE= Yes
#INSTALL_STRIP= -s
XENOCARA_BUILD_DRI=Yes
KERBEROS5= Yes
#DEBUG=Yes
#DEBUGLIBS=Yes
#WARNINGS=Yes
===> Patching for gst-plugins-good-0.10.8
===> Configuring for gst-plugins-good-0.10.8
cd /usr/obj/ports/gst-plugins-good-0.10.8/gst-plugins-good-0.10.8 &&
AUTOCONF_VERSION=2.61 AUTOMAKE_VERSION=1.9 NOCONFIGURE="true" ./autogen.sh
+ check for build tools
checking for autoconf >= 2.52 ... found 2.61, ok.
checking for automake >= 1.7 ... found 1.9.6, ok.
checking for autopoint >= 0.11.5 ... found 0.17, ok.
checking for libtoolize >= 1.5.0 ... found 1.5.26, ok.
checking for pkg-config >= 0.8.0 ... found 0.21, ok.
+ checking for autogen.sh options
This autogen script will automatically run ./configure as:
./configure --enable-maintainer-mode --enable-gtk-doc --enable-plugin-docs
To pass any additional options, please specify them on the ./autogen.sh
command line.
Hmm... Looks like a unified diff to me...
The text leading up to this was:
--
|--- po/Makefile.in.in.orig 2006-01-07 12:03:45.0 +0100
|+++ po/Makefile.in.in 2006-01-07 12:04:23.0 +0100
--
Patching file po/Makefile.in.in using Plan A...
Hunk #1 succeeded at 11.
Hunk #2 succeeded at 305.
done
+ running autopoint --force ...
systrace: deny user: aer, prog: /usr/bin/ssh, pid: 586(0)[1815], policy:
/usr/bin/make, filters: 194, syscall: native-seteuid(183), uid: 1000, uname: aer
PRIV_END: seteuid: Operation not permitted
cvs [checkout aborted]: end of file from server (consult above messages if any)
find: archive: No such file or directory
find: archive: No such file or directory
autopoint: *** infrastructure files for version 0.11.5 not found; this is
autopoint from GNU gettext-tools 0.17
autopoint: *** Stop.
autopoint --force failed
*** Error code 1
Stop in /usr/ports/multimedia/gstreamer-0.10/plugins-good (line 173 of
Makefile).
*** Error code 1
Stop in /usr/ports/multimedia/gstreamer-0.10/plugins-good (line 2144 of
/usr/ports/infrastructure/mk/bsd.port.mk).
Unable to compile abiword 2.4.6p4
Hi, I am running an i386 4.3-current box. After the last libiconv update[0] I was unable to compile abiword-2.4.6p4. The problem came from the patch introduced during the libiconv update[1]. Removing the patch solved the compilation problem. Best Regards, Martin [0]http://marc.info/?l=openbsd-ports&m=121183259820163&w=2 [1]http://www.openbsd.org/cgi-bin/cvsweb/ports/editors/abiword/patches/patch-abi_src_af_util_xp_ut_iconv_cpp
