plan9front.git - my patches for 9front; forked from http://git.9front.org/plan9front/plan9front/HEAD/info.html

Age	Commit message (Collapse)	Author
2013-05-25	graphics(2): remove non existing char *mousedir argument from geninitdraw ↵	cinap_lenrek
	(from sources / geninitdraw-no-mousedir patch)
2013-05-25	libdraw: fix agefont crash when display == nil (from sources / ↵	cinap_lenrek
	agefont-nodisplay patch) If you run this: {for (i in `{seq 1 70000}) echo '½•α·'} \| mc \| wc mc will eventually suicide. The error comes from /sys/src/libdraw/font.c:/^agefont which assumes it has a display. Normally, this is set up by geninitdraw (usually via initdraw), but mc.c avoids that to avoid the screen redraws (flicker, at a minimum, but worse in long-running cases). graphics(2) does not prohibit this use, so I made the least intrusive change to make it work.
2013-05-25	kernel: fix floating point exceptions (was broken by sse kernel changes)	cinap_lenrek
	the fpenv() instruction stores in x87 format, using mathstate() would interpret fpsave as FPssestate in case it was enabled! instead, pass the status word and fppc explicitely to mathnote() in matherror(). get rid of m->fpsavealign buffer, as we can just use FPssesave struct which has enougth padding so rounding up base pointer will not overflow.
2013-05-21	kernel: cleanup /sys/src/9/pc/mkfile PORT= and OBJ= lists	cinap_lenrek

2013-05-21	kernel: prevent alarm(2) from returning instantly when the tick timer wraps. ↵	cinap_lenrek
	(from erik quanstros alarmwrap patch)
2013-05-21	wifi: properly size internal wifi input processing queue	cinap_lenrek
	we only used wifi->iq to process management frames only, now its also used to forward encrypted data frames, so size it accordingly.
2013-05-21	merge	cinap_lenrek

2013-05-22	merge	jpathy

2013-05-22	sse kernel support (sources)	jpathy

2013-05-21	add 6(a\|l) sse support to 8(a\|l)	jpathy

2013-05-21	kernel: make allocb() wait instead of panic() when possible	cinap_lenrek
	as long as our process doesnt hold any locks or ilocks, we can try to wait for the memory to become available instead of panicing.
2013-05-20	uarti8250: print on malloc failure	cinap_lenrek

2013-05-20	merge	cinap_lenrek

2013-05-20	pcuart: malloc error handling, cleanup	cinap_lenrek

2013-05-20	kbmap/ua: fix uppercase ь	ftrvxmtrx

2013-05-20	devether: handle malloc error in etherprobe()	cinap_lenrek

2013-05-20	wifi: handle malloc errors in wifiattach()	cinap_lenrek

2013-05-20	use resrcwait() when waiting for memory to become available	cinap_lenrek
	use resrcwait() when waiting for memory to become available. randomize the sleep time and properly restore old process status in case tsleep() gets interrupted.
2013-05-19	devsd: don't raise Enomem error if sdmalloc() fails, instead wait for the ↵	cinap_lenrek
	memory to become available filesystems do not handle i/o errors well (cwfs will abandon the blocks), and temporary exhaustion of kernel memory (because of too many i/o's in parallel) causes read and write on the partition to fail. i think it is better to wait for the memory to become available in this case. the single allocation is at max SDmaxio bytes, which makes it likely to become available. if we havnt even enought fo that, then rebooting the machine would be the best option. (aux/reboot)
2013-05-19	kbmap/ru: fix uppercase ь	ftrvxmtrx

2013-05-18	nusb/kb: simplify repeat logic as scancode now contains the info about esc1 ↵	cinap_lenrek
	extension
2013-05-18	nusb/kb: oops, fixup	ftrvxmtrx

2013-05-18	nusb/kb: use a flag to indicate extended code; fix the most unix-friendly ↵	ftrvxmtrx
	key on se/fi kbd layout
2013-05-15	faces: just say no to dejavu	stanley lieber

2013-05-14	netaudit: check if ipgw= is an ip address	cinap_lenrek

2013-05-14	make clear that the "ipgw" attribute has to be an ip address	cinap_lenrek

2013-05-14	fortunes: Subject: [9fans] [gsoc] Dart9P	stanley lieber

2013-05-13	rio: place text at beginning of line before cursor and host point	cinap_lenrek

2013-05-13	rio: fix completion and readback and make it possible to complete history	cinap_lenrek

2013-05-12	toppm: no need to initialize err	ppatience0

2013-05-12	toppm: checking for err is useless; it was initialized to nil earlier	ppatience0

2013-05-12	toppm: multichan converts to RGB24, not RGBV	ppatience0

2013-05-12	writeppm: replace WriteGIF by WritePPM	ppatience0

2013-05-13	writeppm: simplify	cinap_lenrek

2013-05-12	writeppm: for bitmaps, 1 is black, 0 is white	ppatience0

2013-05-12	toppm: fix style	ppatience0

2013-05-12	toppm: add -r flag for raw ppm	ppatience0

2013-05-12	writeppm: fix tab/spaces	cinap_lenrek

2013-05-12	merge	cinap_lenrek

2013-05-12	devaoe: avoid panic on malloc() error by using smalloc() instead.	cinap_lenrek

2013-05-12	file: recognize bitmap and graymap ppm images	ppatience0
	writeppm: do not print extra space after each pixel for graymap formats to save space
2013-05-12	ether8169: do phywakeup magic only for specific mac versions (from openbsd)	cinap_lenrek

2013-05-12	ether8169: and phy wakeup and reset phy before enabling auto negotiation	cinap_lenrek
	this brings up the 8198B on BurnZeZ's ASRock Z68 Professional Gen3 board.
2013-05-11	devmnt: fix mount device leak and allocation error handling in mntversion()	cinap_lenrek
	the fist problem is that qopen() might return nil and that kstrdup() will sleep, so we should try to avoid holding the mntalloc lock. so we move the kstrdup() and qopen() calls before the Mnt allocation, and properly recover the memory if we fail later. the second problem was that we error(Eshort) after we already created the Mnt when returnlen < sizeof(f.version). this check has to happen before we even attempt to allocate the Mnt structures. note that we only copy the version string once everything is in the clear, so the semantics of the user buffer not being modified in case of error is not changed. a little cleanup in muxclose(), getting rid of mntptfree()...
2013-05-11	sdide: add Intel SERIES 6 SATA controller pci ids	cinap_lenrek

2013-05-11	acme: apply nemos acmediskread patch (from sources)	cinap_lenrek
	pread does not guarantee that it would read all the data asked for. But acme usage of disk assumes that. This issues as many reads as needed to make acme work when read returns less data than it wanted.
2013-05-11	ape: change types from unsigned long to void* for rendezvous() and ↵	cinap_lenrek
	segattach()/segbrk() (erik), use uniqueue rendezvous tags for _buf and listen
2013-05-09	awk: proctab.c is always regenerated from maketab, so it is unnecessary to ↵	ppatience0
	keep it
2013-05-10	srv: be nice, close the previously duped pipe filedescriptor	cinap_lenrek

2013-05-08	fortunes: all the world's a social vax cloud	khm