git: vkernel - Remove unneeded -T flag, that was added in 7b21e5e4a2ba4592a.

2018-12-28 Thread Imre Vadasz
commit fab26d8f4e1b08a36d7666ce136cf26d197deadc Author: Imre Vadász Date: Fri Dec 28 15:50:31 2018 +0100 vkernel - Remove unneeded -T flag, that was added in 7b21e5e4a2ba4592a. * Instead the -e hw.tsc_cputimer_enable=0 flag can be used, to set the tunable that disables use

git: vkernel - Delete unused/ancient timer/rtc function declarations in clock.h.

2018-12-28 Thread Imre Vadasz
commit 15729594a6d6a7f467d968f996658679fb9ee13b Author: Imre Vadász Date: Fri Dec 28 11:50:57 2018 +0100 vkernel - Delete unused/ancient timer/rtc function declarations in clock.h. * Also get rid of tsc_is_broken flag, that is completely unused. Summary of changes: sys/platfor

git: net - Simplify net/altq TSC timer code, get rid of its TSC calibration code.

2018-12-27 Thread Imre Vadasz
commit adceedafb85f72d6c3d71e4eae3266b57c24025e Author: Imre Vadász Date: Thu Dec 27 22:41:25 2018 +0100 net - Simplify net/altq TSC timer code, get rid of its TSC calibration code. * There is no use for the custom TSC calibration code in the ALTQ code. If our bootup TSC c

git: kernel - Factor out TSC cputimer into common x86_64 code, use for vkernel.

2018-12-27 Thread Imre Vadasz
commit 7b21e5e4a2ba4592a7e27970929718a29bcbb614 Author: Imre Vadász Date: Tue Dec 25 16:02:38 2018 +0100 kernel - Factor out TSC cputimer into common x86_64 code, use for vkernel. * This adds a command line flag -T to the vkernel, to force disable use of the TSC cputimer.

git: syscons - Increase static buffer size for ttyv0.

2018-11-15 Thread Imre Vadasz
commit 6fd42cc50e47866c551d4da5a6ec73cc6e1ef5d7 Author: Imre Vadász Date: Thu Nov 15 13:53:02 2018 +0100 syscons - Increase static buffer size for ttyv0. * This allows the console on the EFI framebuffer to fill the whole screen on "odd" resolutions. This only takes up slig

git: kernel - Fix build when neither INET nor INET6 options are set.

2018-11-06 Thread Imre Vadasz
commit 0f367864cd2af5f00ad1ae8dbe4fb0b090294959 Author: Imre Vadász Date: Sat Nov 3 20:51:06 2018 +0100 kernel - Fix build when neither INET nor INET6 options are set. Summary of changes: sys/kern/vfs_subr.c | 2 ++ 1 file changed, 2 insertions(+) http://gitweb.dragonflybsd.org/dragon

git: crypto - Avoid including opensslconf.h when using openssl/md5.h in kernel.

2018-11-06 Thread Imre Vadasz
commit 580e6763435e7d71e8e969089faa2273cf43dbe9 Author: Imre Vadász Date: Mon Nov 5 23:14:40 2018 +0100 crypto - Avoid including opensslconf.h when using openssl/md5.h in kernel. Summary of changes: crypto/libressl/include/openssl/md5.h | 2 ++ 1 file changed, 2 insertions(+) http://g

git: kernel - Fix typo from b227f3f50d5dc0f when filling in kpmap header.

2018-11-06 Thread Imre Vadasz
commit 6b0da239f8a087706b7df3aacf2e39c48315852f Author: Imre Vadász Date: Tue Nov 6 20:52:36 2018 +0100 kernel - Fix typo from b227f3f50d5dc0f when filling in kpmap header. Summary of changes: sys/kern/init_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) http://gitweb

git: kernel - Make the wdog.h and gpio.h includes conditional in kern_shutdown.c.

2018-11-06 Thread Imre Vadasz
commit c9336c7e2b224b103ba4de9e3fbe59cb9559426a Author: Imre Vadász Date: Mon Nov 5 23:13:00 2018 +0100 kernel - Make the wdog.h and gpio.h includes conditional in kern_shutdown.c. Summary of changes: sys/kern/kern_shutdown.c | 4 1 file changed, 4 insertions(+) http://gitweb.dra

git: sound - Allow /dev/sndstat to be opened multiple times from one process.

2018-10-27 Thread Imre Vadasz
commit b809e1148793bbe33af2adb09af395668a8dae22 Author: Imre Vadász Date: Fri Oct 26 22:14:47 2018 +0200 sound - Allow /dev/sndstat to be opened multiple times from one process. * This still isn't great, but at least it's enough to make pulseaudio's oss module detect the a

git: sound - Fix mmap handling for /dev/dsp* devices.

2018-10-25 Thread Imre Vadasz
commit 827eaa6e4ed034bcb0c621588cb8fb393600a5f1 Author: Imre Vadász Date: Fri Oct 26 00:53:08 2018 +0200 sound - Fix mmap handling for /dev/dsp* devices. * The d_mmap function is expected to return a page index, not an address, the the atop() function for this as usual. S

git: ichwd - Fix typo in TCO2_CNT register #define.

2018-10-04 Thread Imre Vadasz
commit ba59fa7650ff2f210db5f38168ab9d7a1e925937 Author: Imre Vadász Date: Thu Oct 4 21:08:28 2018 +0200 ichwd - Fix typo in TCO2_CNT register #define. * The TCO2_CNT isn't actually used in the driver at the moment. Summary of changes: sys/dev/misc/ichwd/ichwd.h | 2 +- 1 file

git: ichsmb - Add additional PCI IDs to ichsmb.

2018-10-03 Thread Imre Vadasz
commit f07e6439028a10ad336b9be636dcfeb6a43b76b3 Author: Imre Vadász Date: Wed Oct 3 22:56:19 2018 +0200 ichsmb - Add additional PCI IDs to ichsmb. * This moves from a long switch statement to a lookup table. Taken-From: FreeBSD Summary of changes: sys/bus/smbus/ichsmb

git: nmalloc - Try bigcache in posix_memalign() for big PAGE_SIZE aligned allocs.

2018-10-01 Thread Imre Vadasz
commit c8a21d032b25a10c79005c89f88a46985cad31de Author: Imre Vadász Date: Mon Oct 1 21:14:01 2018 +0200 nmalloc - Try bigcache in posix_memalign() for big PAGE_SIZE aligned allocs. * This improves performance (by reducing the amount of mmap() calls required and avoiding so

git: iwm - Remove unused REPLY_MAX

2018-09-22 Thread Imre Vadasz
commit 92a727c99d6ec5abf14bb6853e95e3a187a0cd4e Author: Imre Vadász Date: Fri Sep 21 15:58:54 2018 +0200 iwm - Remove unused REPLY_MAX Taken-From: Linux git e4eb275ac5cfe71686612d929a9829345b2a4ada Summary of changes: sys/dev/netif/iwm/if_iwmreg.h | 2 -- 1 file changed, 2 del

git: iwm - Remove unused TX_CMD_NEXT_FRAME_*

2018-09-22 Thread Imre Vadasz
commit b0c6116f364a121ab6b9d634ca1997d4167fa747 Author: Imre Vadász Date: Fri Sep 21 16:00:35 2018 +0200 iwm - Remove unused TX_CMD_NEXT_FRAME_* Taken-From: Linux git b1e06c65fb69c5e3fddcd91987561e225eaa9bfa Summary of changes: sys/dev/netif/iwm/if_iwmreg.h | 25 +-

git: iwm - Update alive response handling, add v4 and remove old versions.

2018-09-22 Thread Imre Vadasz
commit 3820e2bf3331ced3541d1811a38c5a5136dfab93 Author: Imre Vadász Date: Fri Sep 21 13:43:11 2018 +0200 iwm - Update alive response handling, add v4 and remove old versions. Summary of changes: sys/dev/netif/iwm/if_iwm.c| 101 -- sys/dev/net

git: if_iwm - Stop iwm_watchdog callout when idle.

2018-09-20 Thread Imre Vadasz
commit 6a8683b0e9d734f23bd9647e117da198c2b9a74e Author: Imre Vadász Date: Thu May 24 23:05:10 2018 +0200 if_iwm - Stop iwm_watchdog callout when idle. Summary of changes: sys/dev/netif/iwm/if_iwm.c | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) http://gitweb.dragonflyb

git: iwm - Fix race during detach, where a callout is left after driver is gone.

2018-09-20 Thread Imre Vadasz
commit ba3b4ff9a1fc04a349df05d6d3449f4d9b15c4be Author: Imre Vadász Date: Wed Jun 6 23:32:36 2018 +0200 iwm - Fix race during detach, where a callout is left after driver is gone. Summary of changes: sys/dev/netif/iwm/if_iwm.c | 7 +-- sys/dev/netif/iwm/if_iwm_led.c | 3 +++ 2

git: nfs - Avoid memory leakage when unloading the nfs module.

2018-09-20 Thread Imre Vadasz
commit f4aeef67f55881dd79267697a869634dcfbf3590 Author: Imre Vadász Date: Wed Sep 5 21:59:55 2018 +0200 nfs - Avoid memory leakage when unloading the nfs module. * This adds some missing objcache_destroy and hashdestroy calls in the nfs module unload codepath. Summary of

git: kernel - Avoid a race between dfly_schedulerclock and usched_dfly_cpu_init.

2018-09-05 Thread Imre Vadasz
commit 3979974961a6ce6281d5b1425e99b0945281f39c Author: Imre Vadász Date: Tue Sep 4 22:47:17 2018 +0200 kernel - Avoid a race between dfly_schedulerclock and usched_dfly_cpu_init. * This race was repeatedly occuring during startup, running with multiple virtual cpu cores i

git: ig4 - We actually need to set the Rx threshold register one smaller.

2018-08-18 Thread Imre Vadasz
commit 02f0bf2a2a28ad502f5568df747bcb10ed1315a9 Author: Imre Vadász Date: Sat Aug 18 18:41:47 2018 +0200 ig4 - We actually need to set the Rx threshold register one smaller. * Setting the IG4_REG_RX_TL register to 1 was actually generating an interrupt after 2 bytes were a

git: ig4 - Stop I2C controller after checking that it's kind of functional.

2018-08-18 Thread Imre Vadasz
commit 0b3eedb01d28925fed4a60d5732a8e566d65d93b Author: Imre Vadász Date: Sat Aug 18 18:42:55 2018 +0200 ig4 - Stop I2C controller after checking that it's kind of functional. Summary of changes: sys/bus/smbus/ichiic/ig4_iic.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-)

git: ig4 - Only enable interrupts when we want them. Otherwise keep mask at 0.

2018-08-18 Thread Imre Vadasz
commit d7c855546f690f05cd952ead7f4f09313a428240 Author: Imre Vadász Date: Sat Aug 18 12:58:40 2018 +0200 ig4 - Only enable interrupts when we want them. Otherwise keep mask at 0. * This avoids possible interrupt storms, depending on the state of the I2C controller before t

git: ig4 - Use serializer instead of lockmgr lock.

2018-08-18 Thread Imre Vadasz
commit ab1e5d7f1c9718cc7eb60cdc9299231cb9933a04 Author: Imre Vadász Date: Sat Aug 18 12:49:40 2018 +0200 ig4 - Use serializer instead of lockmgr lock. * This allows us to use lwkt_serialize_handler_disable when stopping the interrupt, which should be safer in case of detac

git: pc64 - If appropriate, determine TSC frequency via CPUID.

2018-08-04 Thread Imre Vadasz
commit 8b257adbda32b59bb93cb76a5daa122a9252f6c0 Author: Imre Vadász Date: Sat Aug 4 17:09:26 2018 +0200 pc64 - If appropriate, determine TSC frequency via CPUID. On Intel systems with Skylake and newer CPUs, and on modern Atom CPUs this avoids the ca. 200ms TSC calibration p

git: corepower - Add support for Platform Energy Counter.

2018-08-04 Thread Imre Vadasz
commit 214ee9b53298179f8906c527caa25e5a212ec8c6 Author: Imre Vadász Date: Sat Aug 4 10:58:17 2018 +0200 corepower - Add support for Platform Energy Counter. Summary of changes: sys/cpu/x86_64/include/specialreg.h| 1 + sys/dev/powermng/corepower/corepower.c | 21 ++

git: kbdmux - Simplify taskqueue handling, and get rid of the 1HZ callout.

2018-04-12 Thread Imre Vadasz
commit 51b03d73b4910d235bb78db099ab7603138c526e Author: Imre Vadász Date: Tue Feb 27 20:50:53 2018 +0100 kbdmux - Simplify taskqueue handling, and get rid of the 1HZ callout. Summary of changes: sys/dev/misc/kbdmux/kbdmux.c | 62 +++- 1 file chang

git: if_vtnet - Add polling support.

2018-03-30 Thread Imre Vadasz
commit 34c7d7ddfd3916373730d51801c1e2358bbcde07 Author: Imre Vadász Date: Thu Mar 29 23:34:20 2018 +0200 if_vtnet - Add polling support. * We need to tear down the interrupts, to reliably stop them from firing, when enabling polling. So when disabling polling, we need to ho

git: if_vtnet - Factor out the virtio_bind_intr() calls into a separate function.

2018-03-30 Thread Imre Vadasz
commit 4bf21f5c5175bb629fdf4c2997e182b8c3f1eb33 Author: Imre Vadász Date: Thu Mar 29 21:23:10 2018 +0200 if_vtnet - Factor out the virtio_bind_intr() calls into a separate function. * And most importantly, store the bindings in the vtnet_softc struct, so we can restore the

git: DragonFly_RELEASE_5_2 virtio_pci - Fix vtpci_reinit MSI-X vector reinitialization.

2018-03-29 Thread Imre Vadasz
commit beac0aaed7266839e181c0707761973b955c54ab Author: Imre Vadász Date: Thu Mar 29 22:10:07 2018 +0200 virtio_pci - Fix vtpci_reinit MSI-X vector reinitialization. Summary of changes: sys/dev/virtual/virtio/pci/virtio_pci.c | 13 - 1 file changed, 8 insertions(+), 5 deleti

git: virtio_pci - Fix vtpci_reinit MSI-X vector reinitialization.

2018-03-29 Thread Imre Vadasz
commit 92d9e66d550e56fad63067121682475b11638f09 Author: Imre Vadász Date: Thu Mar 29 22:10:07 2018 +0200 virtio_pci - Fix vtpci_reinit MSI-X vector reinitialization. Summary of changes: sys/dev/virtual/virtio/pci/virtio_pci.c | 13 - 1 file changed, 8 insertions(+), 5 deleti

git: if_vtnet - Fixup serialized assertion in vtnet_is_link_up().

2018-03-29 Thread Imre Vadasz
commit 5f772200d64c6bef92b3a6fdc8872706b907c402 Author: Imre Vadász Date: Fri Mar 30 00:33:03 2018 +0200 if_vtnet - Fixup serialized assertion in vtnet_is_link_up(). * We check vtnet_is_link_up() from the config interrupt, which runs with just the sc->vtnet_slz serializer.

git: DragonFly_RELEASE_5_2 iwm - Always clear watchdog timer, when bringing down firmware state.

2018-03-29 Thread Imre Vadasz
commit e942fa89d8ff92ac991c8b1bf03530b5f0752034 Author: Imre Vadász Date: Wed Mar 28 22:46:14 2018 +0200 iwm - Always clear watchdog timer, when bringing down firmware state. Summary of changes: sys/dev/netif/iwm/if_iwm.c | 7 +++ 1 file changed, 3 insertions(+), 4 deletions(-) htt

git: DragonFly_RELEASE_5_2 iwm - Clear Time Event active state, when receiving End Notification.

2018-03-29 Thread Imre Vadasz
commit 16e2947a0571007985f41fbaee0566b8b4c3a372 Author: Imre Vadász Date: Wed Mar 28 22:26:18 2018 +0200 iwm - Clear Time Event active state, when receiving End Notification. * This hopefully avoids some firmware panics, I was occasionally seeing, when iwm disconnects upon

git: iwm - Clear Time Event active state, when receiving End Notification.

2018-03-28 Thread Imre Vadasz
commit e8cb71584a6a72232c13151d60e57f7f229220eb Author: Imre Vadász Date: Wed Mar 28 22:26:18 2018 +0200 iwm - Clear Time Event active state, when receiving End Notification. * This hopefully avoids some firmware panics, I was occasionally seeing, when iwm disconnects upon

git: iwm - Always clear watchdog timer, when bringing down firmware state.

2018-03-28 Thread Imre Vadasz
commit 8abdc2b36a45c4e9c95fc8263ca532ea26633dcb Author: Imre Vadász Date: Wed Mar 28 22:46:14 2018 +0200 iwm - Always clear watchdog timer, when bringing down firmware state. Summary of changes: sys/dev/netif/iwm/if_iwm.c | 7 +++ 1 file changed, 3 insertions(+), 4 deletions(-) htt

git: vtnet - Split up serializers, now has separate rx and tx serializers.

2018-03-28 Thread Imre Vadasz
commit 5bd18522b0e3a6420b7cbaddad305b5a038a2315 Author: Imre Vadász Date: Sun Sep 10 22:27:50 2017 +0200 vtnet - Split up serializers, now has separate rx and tx serializers. * This uses the MULTI SERIALIZERS MODE in the DragonFly net framework. * This allows the rx and

git: if_vtnet - Cleanup ifalt_subque handling a bit.

2018-03-27 Thread Imre Vadasz
commit 1ede37801a5378f10ce9610b5e8ee69aa0d28610 Author: Imre Vadász Date: Tue Mar 27 20:40:14 2018 +0200 if_vtnet - Cleanup ifalt_subque handling a bit. Summary of changes: sys/dev/virtual/virtio/net/if_vtnet.c | 16 +--- 1 file changed, 9 insertions(+), 7 deletions(-) http

git: if_vtnet - Get rid of separate vtnet_{rx, tx}_intr_task functions.

2018-03-26 Thread Imre Vadasz
commit 85c6f9e57caa9d6b3de3b5be7c863828d6e3b600 Author: Imre Vadász Date: Mon Mar 26 23:44:30 2018 +0200 if_vtnet - Get rid of separate vtnet_{rx,tx}_intr_task functions. * Just inline them into vtnet_rx_vq_intr and vtnet_tx_vq_intr respectively. Summary of changes: sys/dev/vir

git: if_vtnet - Fixup previous commit.

2018-03-26 Thread Imre Vadasz
commit bce0b58796f92bdcdb1c3e98cd44a00c9ce36e3a Author: Imre Vadász Date: Mon Mar 26 22:56:29 2018 +0200 if_vtnet - Fixup previous commit. Summary of changes: sys/dev/virtual/virtio/net/if_vtnet.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) http://gitweb.dragonflybsd.or

git: if_vtnet - Use ifsq_watchdog_* functions as the watchdog.

2018-03-26 Thread Imre Vadasz
commit f7e8a8e7217c502bdfc9b765bdfc3e21f1f042f0 Author: Imre Vadász Date: Mon Mar 26 01:17:40 2018 +0200 if_vtnet - Use ifsq_watchdog_* functions as the watchdog. * The existing watchdog code was commented out. Summary of changes: sys/dev/virtual/virtio/net/if_vtnet.c| 37 +

git: if_vtnet - Don't drop vtnet_slz in some of the virtqueue_notify() calls.

2018-03-25 Thread Imre Vadasz
commit 36ad0f356d159f9a99f74f2fd4559443ef9af7f9 Author: Imre Vadász Date: Sun Mar 25 21:59:43 2018 +0200 if_vtnet - Don't drop vtnet_slz in some of the virtqueue_notify() calls. * There are several cases where the vtnet_slz serializer is dropped around virtqueue_notify() ca

git: if_vtnet - Remove vtnet_tick_ch and vtnet_cfgchg_task from softc struct.

2018-03-25 Thread Imre Vadasz
commit af62becb69fced07907e16f75095de297a34eaba Author: Imre Vadász Date: Sun Mar 25 21:05:04 2018 +0200 if_vtnet - Remove vtnet_tick_ch and vtnet_cfgchg_task from softc struct. * Forgot to remove the softc variables in the changes c5f8e1538bcc13f7a39b805bcc5b774780 and 603

git: if_vtnet - Pass the vtnet serializer in ether_ifattach().

2018-03-25 Thread Imre Vadasz
commit 9528162f66f9f659e14f42b3f1bb70798da1755d Author: Imre Vadász Date: Sun Mar 25 20:55:23 2018 +0200 if_vtnet - Pass the vtnet serializer in ether_ifattach(). * This is likely just an intermediate step, the plan is to eventually use the MULTI SERIALIZER MODE. Summary o

git: if_vtnet - Remove unused vtnet_tick() code.

2018-03-25 Thread Imre Vadasz
commit 603be9646446bdb4292b6cd27544788c2c00c988 Author: Imre Vadász Date: Sun Mar 25 20:54:41 2018 +0200 if_vtnet - Remove unused vtnet_tick() code. Summary of changes: sys/dev/virtual/virtio/net/if_vtnet.c | 29 ++--- 1 file changed, 2 insertions(+), 27 deletion

git: if_vtnet - No need to use the taskqueue to run vtnet_update_link_status().

2018-03-25 Thread Imre Vadasz
commit c5f8e1538bcc13f7a39b805bcc5b77478066e3e5 Author: Imre Vadász Date: Sun Mar 25 16:35:18 2018 +0200 if_vtnet - No need to use the taskqueue to run vtnet_update_link_status(). Summary of changes: sys/dev/virtual/virtio/net/if_vtnet.c | 17 + 1 file changed, 1 inserti

git: virtio - Refactor virtio bus API, to untangle virtuques from IRQ allocation.

2018-03-24 Thread Imre Vadasz
commit 2f2405bb182023ff0ff0fb5a4d9fc82a5231807c Author: Imre Vadász Date: Fri Sep 8 00:22:49 2017 +0200 virtio - Refactor virtio bus API, to untangle virtuques from IRQ allocation. * These API changes allow each virtio device driver to explicitly control the assignment of v

git: virtio - Move virtqueue_pending() checks into the child driver.

2018-03-24 Thread Imre Vadasz
commit f852cf8223a804cdf37fb5cc685a71a14e5b2388 Author: Imre Vadász Date: Wed Sep 13 21:59:25 2017 +0200 virtio - Move virtqueue_pending() checks into the child driver. * This makes it easier for now, to deal with virtqueue locking, when multiple virtqueues are mapped to a

git: virtio - Get rid of config_change devmethod, treat like vq interrupts.

2018-03-24 Thread Imre Vadasz
commit b817dce2e30a257b86b253a698e63a418758d76d Author: Imre Vadász Date: Wed Sep 13 00:32:22 2017 +0200 virtio - Get rid of config_change devmethod, treat like vq interrupts. Summary of changes: sys/conf/files| 1 - sys/dev/virtual/virtio/block/Makefile

git: virtio - Specify handler and argument in bind_intr, instead of vq alloc.

2018-03-24 Thread Imre Vadasz
commit 9d96478c5a4cae6157aeb8f1ebf8845d910fde14 Author: Imre Vadász Date: Mon Sep 11 22:59:58 2017 +0200 virtio - Specify handler and argument in bind_intr, instead of vq alloc. Summary of changes: sys/dev/virtual/virtio/block/virtio_blk.c | 14 ++-- sys/dev/virtual/virtio/net/if_vt

git: virtio - Get rid of unused flags option in virtio_alloc_virtqueues().

2018-03-24 Thread Imre Vadasz
commit 099c4d8e901ac568bc7b1b6cc7e646b7d498e22b Author: Imre Vadász Date: Fri Mar 23 00:08:43 2018 +0100 virtio - Get rid of unused flags option in virtio_alloc_virtqueues(). Summary of changes: sys/dev/virtual/virtio/block/virtio_blk.c | 2 +- sys/dev/virtual/virtio/net/if_vtnet.c

git: virtio - Explicitly teardown interrupts in driver detach methods.

2018-03-24 Thread Imre Vadasz
commit 6ea187fdf56cbf76e1aae26f2ebd723669e6a033 Author: Imre Vadász Date: Sat Mar 24 14:44:19 2018 +0100 virtio - Explicitly teardown interrupts in driver detach methods. * This seems to be an easy way to avoid potential races during driver detachements. Summary of changes

git: virtio_blk - Inline vtblk_complete into vtblk_vq_intr. Dedup sglist_reset.

2018-03-22 Thread Imre Vadasz
commit 2cc8f29fc3365da0cab26773a48f179a100807b0 Author: Imre Vadász Date: Sat Mar 11 21:13:19 2017 +0100 virtio_blk - Inline vtblk_complete into vtblk_vq_intr. Dedup sglist_reset. Summary of changes: sys/dev/virtual/virtio/block/virtio_blk.c | 30 +- 1 file c

git: virtio_pci - Remove int return value from interrupt handler functions.

2018-03-22 Thread Imre Vadasz
commit df2df2fc398f99295f0c44100f8c54bc00bae3f1 Author: Imre Vadász Date: Mon Mar 6 20:33:56 2017 +0100 virtio_pci - Remove int return value from interrupt handler functions. Summary of changes: sys/dev/virtual/virtio/pci/virtio_pci.c | 43 --- sys/dev/virt

git: atkbdc - Add additional ACPI PNP IDs specified by Microsoft for Keyboards.

2018-03-19 Thread Imre Vadasz
commit b276d6fe06e6cb191f75486f1608f0dbc5822485 Author: Imre Vadász Date: Mon Mar 19 23:39:17 2018 +0100 atkbdc - Add additional ACPI PNP IDs specified by Microsoft for Keyboards. These IDs are from Microsofts device IDs list at http://download.microsoft.com/download/1/6/1/1

git: pc64 - Allow for initializing other cputimers than i8254 in early boot.

2018-03-18 Thread Imre Vadasz
commit 1a3a6ceef07ae114bff26280fd583cee2103b1d6 Author: Imre Vadász Date: Sun Feb 18 14:27:46 2018 +0100 pc64 - Allow for initializing other cputimers than i8254 in early boot. Summary of changes: sys/platform/pc64/apic/lapic.c| 14 -- sys/platform/pc64/include/clock.h | 14

git: pc64 - Improve TSC and LAPIC timer calibration code.

2018-03-18 Thread Imre Vadasz
commit 4098a6e56c77c5ea1fb3cd4cb9afe1bbb3739ff3 Author: Imre Vadász Date: Sun Feb 25 22:51:26 2018 +0100 pc64 - Improve TSC and LAPIC timer calibration code. * The hw.tsc_calibrate_test=1 and hw.lapic_calibrate_test=1 tunables can be specified to test results of the calibra

git: acpi_hpet - Implement early HPET initialization for TSC calibration.

2018-03-18 Thread Imre Vadasz
commit c4984bac94803f7c1962edfe7f66263e51c8e371 Author: Imre Vadász Date: Mon Feb 5 22:55:23 2018 +0100 acpi_hpet - Implement early HPET initialization for TSC calibration. Summary of changes: sys/dev/acpica/acpi_hpet.c | 200 + 1 file changed

git: iwm - Improve firmware Time Event handling.

2018-03-18 Thread Imre Vadasz
commit 706a3044afd27c3fecfdf57bec1695310e53e228 Author: Imre Vadász Date: Sun Mar 18 15:24:22 2018 +0100 iwm - Improve firmware Time Event handling. * This is a mix of the OpenBSD Git 7fd9664469d1b717a307eebd74aeececbd3c41cc change, and syncing with the Linux iwlwifi code.

git: iwm - Avoid Tx watchdog timeout, when dropping a connection.

2018-03-18 Thread Imre Vadasz
commit 3e12596fb5c55351517cdd741d72979388a8c75c Author: Imre Vadász Date: Sun Mar 18 11:04:49 2018 +0100 iwm - Avoid Tx watchdog timeout, when dropping a connection. Summary of changes: sys/dev/netif/iwm/if_iwm.c | 4 1 file changed, 4 insertions(+) http://gitweb.dragonflybsd.org/

git: iwm - Track firmware state better, and improve handling in iwm_newstate().

2018-03-18 Thread Imre Vadasz
commit 8a41b10ac639d0609878696808387a6799d39b57 Author: Imre Vadász Date: Sat Mar 17 15:48:02 2018 +0100 iwm - Track firmware state better, and improve handling in iwm_newstate(). * This avoids firmware resets in all the cases in iwm_newstate(). Instead iwm_bring_down_firmw

git: syscons - Fix NULL pointer access in 0d7c8a4d1cafae68239

2018-03-04 Thread Imre Vadasz
commit e3657467019c4c01c48e39f5eedb819571d07159 Author: Imre Vadász Date: Sun Mar 4 22:49:59 2018 +0100 syscons - Fix NULL pointer access in 0d7c8a4d1cafae68239 Summary of changes: sys/dev/misc/syscons/syscons.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) http://gitweb.

git: pc64 - Unmask some AMD Family 15h and 16h CPUs for TSC mpsync test.

2018-03-04 Thread Imre Vadasz
commit 8a93c79fd605ae9811f0e11aeaf695e5fef0d49a Author: Imre Vadász Date: Sun Mar 4 13:03:57 2018 +0100 pc64 - Unmask some AMD Family 15h and 16h CPUs for TSC mpsync test. * The problematic Erratum 778 "Processor Core Time Stamp Counters May Experience Drift" is only

git: syscons - Wait for drm modesetting at appropriate places.

2018-03-04 Thread Imre Vadasz
commit 0d7c8a4d1cafae68239fdb96d2e0432a9e98a551 Author: Imre Vadász Date: Sun Mar 4 14:51:34 2018 +0100 syscons - Wait for drm modesetting at appropriate places. * This avoids races between userland using DRM, and the drm modesetting callbacks called by syscons. *

git: sysmouse - Fix sigio handling after c2962d9da8200947ee394b190eea39d437fb1af7

2018-03-03 Thread Imre Vadasz
commit 2a070c438663edc8253ede1162ec24613a804dc2 Author: Imre Vadász Date: Wed Feb 28 21:47:08 2018 +0100 sysmouse - Fix sigio handling after c2962d9da8200947ee394b190eea39d437fb1af7 Summary of changes: sys/dev/misc/syscons/sysmouse.c | 29 + 1 file changed, 2

git: syscons - No special minor device number is needed anymore for sysmouse.

2018-02-22 Thread Imre Vadasz
commit 095f0dab393e9460c772f589858779f7bf10c3ac Author: Imre Vadász Date: Thu Feb 22 22:54:10 2018 +0100 syscons - No special minor device number is needed anymore for sysmouse. Summary of changes: sys/dev/misc/syscons/sysmouse.c | 10 +- 1 file changed, 1 insertion(+), 9 deleti

git: syscons - Detangle sysmouse from tty handling, implement a custom FIFO.

2018-02-22 Thread Imre Vadasz
commit c2962d9da8200947ee394b190eea39d437fb1af7 Author: Imre Vadász Date: Thu Feb 22 22:54:33 2018 +0100 syscons - Detangle sysmouse from tty handling, implement a custom FIFO. * This avoids unnecessary IPIs and other overhead from the tty handling. * This also makes rea

git: virtio - Remove explicit load of virtio.ko. Remove device virtio_pci line.

2018-02-18 Thread Imre Vadasz
commit b37a48c947a68883cbd9c74b4863efa6c4dc6862 Author: Imre Vadász Date: Sun Mar 12 13:54:48 2017 +0100 virtio - Remove explicit load of virtio.ko. Remove device virtio_pci line. * virtio.ko never needs to be loaded explicitly, since virtio_pci.ko depends on it. *

git: stage created virtio - Remove explicit load of virtio.ko. Remove device virtio_pci line.

2018-02-18 Thread Imre Vadasz
at 05ad4752a1e5675638241c40375746b6a09be1d3 (commit) -- DragonFly BSD source repository

git: stage virtio - Remove explicit load of virtio.ko. Remove device virtio_pci line.

2018-02-18 Thread Imre Vadasz
commit 05ad4752a1e5675638241c40375746b6a09be1d3 Author: Imre Vadász Date: Sun Mar 12 13:54:48 2017 +0100 virtio - Remove explicit load of virtio.ko. Remove device virtio_pci line. * virtio.ko never needs to be loaded explicitly, since virtio_pci.ko depends on it. *

git: devinfo - Print info, whether a device is flagged as disabled.

2018-02-18 Thread Imre Vadasz
commit 9f05f13f0fc9e04c6ab7af6a94af3643b5e740fe Author: Imre Vadász Date: Mon Feb 5 18:19:52 2018 +0100 devinfo - Print info, whether a device is flagged as disabled. * Also make -v print only enabled devices, specifying -v twice will also print disabled devices. This shoul

git: libdevinfo - Fix enum devinfo_state.

2018-02-18 Thread Imre Vadasz
commit 652e7e9f433a1caebbea9b477b183a4ec052d622 Author: Imre Vadász Date: Mon Feb 5 18:19:33 2018 +0100 libdevinfo - Fix enum devinfo_state. Summary of changes: lib/libdevinfo/devinfo.h | 1 + usr.sbin/devinfo/devinfo.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) http://

git: kern - Add D_MEM flag to /dev/mem and /dev/kmem character devices.

2018-02-18 Thread Imre Vadasz
commit 7e81bb374d263400b16daa9a11d684287db36e78 Author: Imre Vadász Date: Mon Feb 5 21:24:11 2018 +0100 kern - Add D_MEM flag to /dev/mem and /dev/kmem character devices. * This makes seeking actually work on these character devices, which fixes performance of various tools

git: atkbdc - Do not attach PS2 controller via legacy ISA bus, if FADT says so.

2018-02-18 Thread Imre Vadasz
commit 131acb0380e93664f106117ebf8ba322a3d1b7c4 Author: Imre Vadász Date: Thu Feb 1 13:16:55 2018 +0100 atkbdc - Do not attach PS2 controller via legacy ISA bus, if FADT says so. * If the FADT BootFlags value tells us that we shouldn't expect the PS2 controller to be at the

git: acpi - Implement a hack to recheck whether devices are present a few times.

2018-01-30 Thread Imre Vadasz
commit ebb7869e549bcacf2537fb61e21749e3f4588a45 Author: Imre Vadász Date: Sun Jan 28 22:06:29 2018 +0100 acpi - Implement a hack to recheck whether devices are present a few times. * This allows various ACPI devices to attach, which depend on another device. Often the _STA

git: acpi - Use KOBJ_GPRI_ACPI driver priorities for ig4 and gpio_intel.

2018-01-30 Thread Imre Vadasz
commit df21e16d16b51cf8e1828be094b4866e5d0a1bc8 Author: Imre Vadász Date: Sun Jan 28 22:04:55 2018 +0100 acpi - Use KOBJ_GPRI_ACPI driver priorities for ig4 and gpio_intel. * Since we currently don't handle the ACPI _DEP methods appropriately, use the "global priority hack"

git: led(4): Turn this into a module. Only used by acpi_thinkpad and acpi_asus.

2018-01-28 Thread Imre Vadasz
commit f2c400e8d295d33c95df1bc50e36c0585233955e Author: Imre Vadász Date: Sun Jan 21 17:48:53 2018 +0100 led(4): Turn this into a module. Only used by acpi_thinkpad and acpi_asus. Summary of changes: sys/conf/files | 2 +- sys/config/LINT64

git: syscons - Untangle device attachement from isa?, attach to nexus? instead.

2018-01-28 Thread Imre Vadasz
commit 20e2cd35cbe6ee16ff636fba5635b7c5f96b7744 Author: Imre Vadász Date: Sun Jan 28 16:11:10 2018 +0100 syscons - Untangle device attachement from isa?, attach to nexus? instead. * The syscons device was still attaching via the isa? bus for mostly historical reasons. Inste

git: syscons - Use COL and ROW #define constants.

2018-01-13 Thread Imre Vadasz
commit 1ba5dedead07cbd24fd7b016c3ef65e88b1c55f4 Author: Imre Vadász Date: Sat Jan 13 21:11:40 2018 +0100 syscons - Use COL and ROW #define constants. Summary of changes: sys/dev/misc/syscons/syscons.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) http://gitweb.dragonflybs

git: syscons - Skip drawing of the border, when switching to a graphics vty.

2018-01-13 Thread Imre Vadasz
commit 8a5e307d170a0da955a03766f8d402188d545efd Author: Imre Vadász Date: Sat Jan 13 21:37:40 2018 +0100 syscons - Skip drawing of the border, when switching to a graphics vty. * When the target vty in a vty switch is in graphics mode, and we have a dumb framebuffer (sc->fb

git: syscons - Partly implement FBIO_BLANK ioctl, for display powersaving.

2017-12-01 Thread Imre Vadasz
commit 1743eeb476593a2463740afc618c5272b68275aa Author: Imre Vadász Date: Sat Nov 25 18:25:54 2017 +0100 syscons - Partly implement FBIO_BLANK ioctl, for display powersaving. * Only implement the parts relevant for display powersaving when a KMS graphics driver is loaded. A

git: kernel - Get rid of old KDSETRAD keyboard ioctl, and some __i386__ kbd code.

2017-11-24 Thread Imre Vadasz
commit 27332c1b34b2ee966854708d0ed957eeb6d61baa Author: Imre Vadász Date: Fri Nov 24 20:19:50 2017 +0100 kernel - Get rid of old KDSETRAD keyboard ioctl, and some __i386__ kbd code. * The KDSETREPEAT ioctl was implemented as a replacement more than 18 years ago in 1999 (Fre

git: syscons - Add minimal fbio support for "scfb" xorg driver with sc->fbi fb.

2017-11-03 Thread Imre Vadasz
commit bc7273cfc7e394a5944a4f30683254daadd8b6b7 Author: Imre Vadász Date: Wed Oct 18 23:40:12 2017 +0200 syscons - Add minimal fbio support for "scfb" xorg driver with sc->fbi fb. * At the moment there is no support for safely handling the case where userspace has mapped th

git: syscons - Add 24bit pixel format support for EFI framebuffer.

2017-10-26 Thread Imre Vadasz
commit a6dad826d931cdc52e6e2dfde1facf9b69eb3ca4 Author: Imre Vadász Date: Sun Oct 22 22:47:24 2017 +0200 syscons - Add 24bit pixel format support for EFI framebuffer. Summary of changes: sys/dev/misc/syscons/sckmsrndr.c | 237 ++--- sys/platform/pc64/x8

git: boot - Abort boot if EFI-framebuffer format is unsupported.

2017-10-26 Thread Imre Vadasz
commit ce0b01dd911aa9d20918c80828569ce2e882aa96 Author: Imre Vadász Date: Wed Oct 25 21:11:08 2017 +0200 boot - Abort boot if EFI-framebuffer format is unsupported. * At the moment we support 24-bit and 32-bit pixel formats, make sure we notify the user and abort booting, w

git: DragonFly_RELEASE_4_8 if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload.

2017-10-22 Thread Imre Vadasz
commit bbb6850667ab7baddb5db2876fd095aa4ca42d5d Author: Imre Vadász Date: Tue Oct 17 22:11:08 2017 +0200 if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload. * Ignoring the checksum offloading in the receive path of the driver isn't sufficient, since

git: DragonFly_RELEASE_4_8 if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff.

2017-10-22 Thread Imre Vadasz
commit e3d658a46a0da6a8a2d290b9a191143976b2f039 Author: Imre Vadász Date: Tue Oct 17 22:06:23 2017 +0200 if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff. * The if_vtnet driver used to define the IFCAP_LRO, IFCAP_VLAN_HWFILTER and IFCAP_VLAN_HWTSO

git: DragonFly_RELEASE_5_0 if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff.

2017-10-22 Thread Imre Vadasz
commit d2565295f6d8d86a3b9e236a3a650a2b77ef0c15 Author: Imre Vadász Date: Tue Oct 17 22:06:23 2017 +0200 if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff. * The if_vtnet driver used to define the IFCAP_LRO, IFCAP_VLAN_HWFILTER and IFCAP_VLAN_HWTSO

git: DragonFly_RELEASE_5_0 if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload.

2017-10-22 Thread Imre Vadasz
commit 849f77c77acc4d5041fa638b364cb6a296a8051b Author: Imre Vadász Date: Tue Oct 17 22:11:08 2017 +0200 if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload. * Ignoring the checksum offloading in the receive path of the driver isn't sufficient, since

git: if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff.

2017-10-17 Thread Imre Vadasz
commit d07f5241eb2c4ceea221840fae247bb2e2a47f5e Author: Imre Vadász Date: Tue Oct 17 22:06:23 2017 +0200 if_vtnet - Handle missing IFCAP_VLAN_* flags nicer. Comment IFCAP_LOR stuff. * The if_vtnet driver used to define the IFCAP_LRO, IFCAP_VLAN_HWFILTER and IFCAP_VLAN_HWTSO

git: if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload.

2017-10-17 Thread Imre Vadasz
commit dddbf4cd54036c88ab90de54668f542d70b771b0 Author: Imre Vadász Date: Tue Oct 17 22:11:08 2017 +0200 if_vtnet - Disable rx csum offload due to unsupported ipv6 rx csum offload. * Ignoring the checksum offloading in the receive path of the driver isn't sufficient, since

git: virtio_blk - Fix capacity calculation, when host sets large disk block size.

2017-10-16 Thread Imre Vadasz
commit 4846c10471333bbb8549928a19e1928788509360 Author: Imre Vadász Date: Tue Oct 17 00:00:32 2017 +0200 virtio_blk - Fix capacity calculation, when host sets large disk block size. * The disk capacity in the virtio configuration space is always specified in 512 byte sector

git: Revert "Update drm/radeon to Linux 4.7.10 as much as possible..."

2017-09-30 Thread Imre Vadasz
commit ee479021092456b32f018d8b9dc5d3e1f4fb9363 Author: Imre Vadász Date: Sat Sep 30 21:08:51 2017 +0200 Revert "Update drm/radeon to Linux 4.7.10 as much as possible..." This caused severe issues on some modern radeon models (e.g. Kaveri APUs). Not reverting the 62dc643ef61b

git: sound - Don't need to use swi_taskq to call KNOTE from the interrupt.

2017-09-28 Thread Imre Vadasz
commit 52b1a712425f78997df963f067f86f7e251fd912 Author: Imre Vadász Date: Fri Sep 29 00:15:08 2017 +0200 sound - Don't need to use swi_taskq to call KNOTE from the interrupt. * Also the dsp_read_filterops and dsp_write_filterops kqueue filters can be flagged FILTEROP_MPSAFE

git: microuptime.9 microtime.9: Fix documentation of the get* function versions.

2017-09-23 Thread Imre Vadasz
commit 2f3ba5dc82b34ec1f5295ea0d3e49d254c66792d Author: Imre Vadász Date: Sat Sep 23 17:04:38 2017 +0200 microuptime.9 microtime.9: Fix documentation of the get* function versions. The kern.timecounter sysctl tree doesn't exist nowadays, the getmicrotime(), getnanotime(), get

git: psm: Drop bpsm%d device files. Instead support non-blocking reads on psm%d.

2017-09-23 Thread Imre Vadasz
commit baa896597c4cebf6e2a3bafa7a3040b6fc6d3eec Author: Imre Vadász Date: Sat Sep 23 13:12:34 2017 +0200 psm: Drop bpsm%d device files. Instead support non-blocking reads on psm%d. The /dev/psm%d vs. /dev/bpsm%d separation doesn't serve any clear purpose nowadays. Userland ca

git: psm: Get rid of PSM_LEVEL_NATIVE, and the psmwrite method used with that.

2017-09-22 Thread Imre Vadasz
commit 5b246b1705a0eaf9593195f71335082547c8edc1 Author: Imre Vadász Date: Fri Sep 22 17:46:45 2017 +0200 psm: Get rid of PSM_LEVEL_NATIVE, and the psmwrite method used with that. * Nothing in userspace ever uses this feature. This apparently was intended to allow implementi

git: psm: Remove dead unused code: psmpoll(), enable_lordless(), is_a_mouse().

2017-09-22 Thread Imre Vadasz
commit 617f6bd6553bb7523bb0198e7aaa5d9c36fb6e05 Author: Imre Vadász Date: Fri Sep 22 14:21:52 2017 +0200 psm: Remove dead unused code: psmpoll(), enable_lordless(), is_a_mouse(). * The is_a_mouse() check method was already disabled in the original FreeBSD commit, which adde

git: if_vtnet - Fix potential vtnet_set_hwaddr call before virtqueues are set up.

2017-09-10 Thread Imre Vadasz
commit 216dcef0df73be65b8df15124fe87c388e390440 Author: Imre Vadász Date: Mon Sep 11 00:21:24 2017 +0200 if_vtnet - Fix potential vtnet_set_hwaddr call before virtqueues are set up. * When VIRTIO_NET_F_MAC isn't negotiated (i.e. the host doesn't give us a macaddress), we ge

  1   2   3   4   5   6   >