Messages in this thread | | | Date | Tue, 18 Jan 2000 22:57:33 +0100 | From | Luca Montecchiani <> | Subject | Re: [2.3.3x] ALI M15x3 chipset support (EXPERIMENTAL) |
| |
Tom> On a related issue: I find kernel 2.3.37 which I am running at the moment Tom> does not report the "ide: Assuming xxMHz system bus speed..." at all. Tom> Previously I was running 2.3.24 which did report 40MHz assumed speed. I Tom> did a quick grep of the two sources and the routines for this in Tom> drivers/block/{ide.c,ali15x3.c} are the same. Any idea whats going on?
Alim15x3 driver prior to kernel 2.3.28 was unable to properly detect our chipset rev ;) Anyway you can read your PCI bus speed from /proc/ide/ali
Tom> I hope to try this once I have backed-up my root HD. I'm puzzled by the Tom> argument to idebus. Where do you get the 30 from? I cannot find any
Reading the source, idebus is used only if you are PIOing wich is not your case and that's why you don't see "..assuming 40Mhz" message anymore.
Tom> Does anybody know the nature of the 'instability' ?
I've hack the driver to _force_ udma33 with my revision 0x20, please don't do that !! This completely _fuck_ my hdc WDC AC24300L, and I need to repartitioning but with my new IBM-DJNA-371350 I've got _very_ good result and right now I didn't fall back to dma2 :|
I've moved several gbytes of data, compiled to many kernels and run several benchmarks without any problems:
Machine ------- This is an AMD AMD-K6 3D 450mhz with 64M fastpage ram
00:00.0 Host bridge: Acer Laboratories Inc. [ALi] M1541 (rev 04) 00:01.0 PCI bridge: Acer Laboratories Inc. [ALi] M5243 (rev 04) 00:07.0 ISA bridge: Acer Laboratories Inc. [ALi] M1533 PCI to ISA Bridge [Aladdin IV] (rev b4) 00:0f.0 IDE interface: Acer Laboratories Inc. [ALi] M5229 IDE (rev 20)
hda: IBM-DJNA-371350, 12949MB w/1966kB Cache, CHS=1650/255/63, (U)DMA multcount = 0 (off) I/O support = 1 (32-bit) unmaskirq = 1 (on) using_dma = 1 (on) keepsettings = 0 (off) nowerr = 0 (off) readonly = 0 (off) readahead = 8 (on) geometry = 1650/255/63, sectors = 26520480, start = 0
+-----------------+-----------------------+------------------------+ |hdparm -tT | buffer-cache (MB/sec) | buffered disk (MB/sec) | +-----------------+-----------------------+------------------------+ |2.2.14 DMA2 | 36.16 | 7.88 | |2.2.14 UDMA33 | 36.6 | 14.35 | |2.3.40 DMA2 | 32.24 | 6.99 | |2.3.40 UDMA33 | 37.65 | 13.62 | +-----------------+-----------------------+------------------------+
I don't know why I'm so slow in buffer-cache, maybe the fastpage ram ?
+-----------------+-----------Sequential Output------------Sequential Input-----Random--+ |bonnie -s 200 | -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---| +-----------------+ MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU| |2.2.14 DMA2 |200 4119 98.2 11835 26.5 4869 25.0 4091 95.0 10531 15.1 126.7 1.8| |2.2.14 UDMA33 |200 4128 98.6 15013 36.7 6238 36.2 4133 96.0 15570 29.5 120.5 1.8| |2.3.40 DMA2 |100 3494 96.9 11692 32.8 4373 12.0 2895 75.7 8636 15.5 102.1 2.6| |2.3.40 UDMA33 |200 4164 99.3 15032 37.1 6605 17.2 3884 90.2 15866 24.6 118.7 1.8| +-----------------+---------------------------------------------------------------------+
+-----------------+------+-------+-------+----+ |time make bzImage|user |system |elapsed|%CPU| +-----------------+------+-------+-------+----+ |2.2.14 DMA2 |309.96| 33.22|5:47.96| 98| |2.2.14 UDMA33 |304.79| 31.94|5:40.85| 98| |2.3.40 DMA2 | | | | | |2.3.40 UDMA33 |299.88| 33.64|5:48.18| 95| +-----------------+------+-------+-------+----+
ciao, luca ------------------------------------------------------------------ E-mail......: Luca Montecchiani <m.luca@iname.com> W.W.W.......: http://i.am/m.luca - http://luca.myip.org Speakfreely.: sflwl -hlwl.fourmilab.ch luca@ I.C.Q.......: 17655604 -----------------------=(Linux since 1995)=-----------------------
Non esiste vento favorevole per il marinaio che non sa dove andare Seneca
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |