http://etherboot.org/wiki/ 2026-05-04T17:49:41-07:00 http://etherboot.org/wiki/ http://etherboot.org/wiki/lib/tpl/monobook/images/favicon.ico text/html 2010-08-07T04:15:21-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/start?rev=1281179721&do=diff Guo-Fu Tseng: Implement and Port Ethernet Drivers for gPXE Journal * Week -4: [ Apr 18 - May 24 ] Debug hardware problem * Week -3: [ Apr 25 - May 2 ] Trace gPXE code base * Week -2: [ May 3 - May 9 ] Implement JMicron driver * Week -1: [ May 10 - May 16 ] Modify JMicron driver, RX offloading * Week 0: [ May 17 - May 23 ] TCP performance test and tunning * Week 1: [ May 24 - May 30 ] Update wiki, TCP tunning * Week 2: [ May 31 - Jun 6 ] Update jme driver, Trace memory rela… text/html 2010-05-26T10:10:10-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week0?rev=1274893810&do=diff Week 0: TCP performance test and tunning May 13th to May 23th * Implement TCP receive queue * Implement TCP SACK [RFC 2018] * Implement TCP Window Scale [RFC 1323] The test results can be found at these notes: * TCP out-of-order receive dstat test logs * Benchmarks for downloading image from boot.kernel.org Found TCP FIN ACK number issue text/html 2010-05-28T06:15:53-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week1?rev=1275052553&do=diff Week 1: Update wiki, TCP tunning May 27th. * Update wiki for week -4 to week 0 progress. May 28th. Weekly meeting * Post JMicron driver to gpxe-devel as a patch. * Start a discussion on gpxe-devel regarding to TCP and memory changes. * Update plan to reflect my current work. text/html 2010-06-04T05:47:04-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week2?rev=1275655624&do=diff Week 2 [ 31 May - 6 Jun 2010 ]: Discuss TCP and memory changes, update jme driver jme driver * From Joshua Oreman * jme_check_link(struct net_device *netdev, int testonly) * In this function, I believe the code and string data used in constructing linkmsg will make it into even non-debug versions of the executable. If that's true it would be better if you could perform all the manipulations in the final DBG() line, using the ?: operator and such. It's possible the compiler is clever … text/html 2010-06-07T15:36:15-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week3?rev=1275950175&do=diff Week 3 [ Jun 7 - Jun 13 ] Trace memory related codes, misc things * Take some time to trace more gPXE/iPXE memory/bootstrap related codes. * Tried to setup DOS/Windows XP/Linux AoE sanboot. * Implemented copy-on-write support for vblade-20. Git tree text/html 2010-07-19T19:11:24-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week8?rev=1279591884&do=diff Several TCP fixes Separating the xfer interface closing routines from tcp_close The original Receive flow of TCP stack pass the data to upper layer while the TCP state transition and response still in progress. But when upper layer received some specific data, it might send data through the current TCP connection, or close the current TCP connection. This issue makes the TCP state hard to maintain. text/html 2010-07-29T07:08:37-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week9?rev=1280412517&do=diff TCP cleanup Cleanup TCP close actions * Reduce duplicate code by separating tcp_terminate() from tcp_close(). * Use tcp_terminate to immediately free all tcp resources. * Use tcp_close() to bing TCP connection to closing state. * Make tcp_close() function simpler, not doing different job in different state. text/html 2010-07-30T04:14:59-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week10?rev=1280488499&do=diff Create a branch to merge latest iPXE TCP related commits based on gPXE master Shortlog of iPXE merge commits Git-branch * Guo-Fu Tseng (2): * [tcp] Randomize TCP bind port * [hci] Continue processing while prompting for shell banner * Michael Brown (8): text/html 2010-08-07T04:03:02-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week11?rev=1281178982&do=diff Benchmark plan * Performance test sites * <http://netboot.me/> * <http://boot.fedoraproject.org/index> * <http://boot.kernel.org/> * Test env: * KVM with different kind of emulated NIC. * Direct link. * User mode network stack. * With TUN/TAP * Bridge/NAT with TUN/TAP. * Native gPXE. * Trace the issue that gPXE download speed slow on KVM. text/html 2010-08-13T01:31:24-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/week12?rev=1281688284&do=diff Port wscale branch against current gPXE master Recent TCP updates is a lot different from previous base of wscale branch. Hit a GCC optimization trap For the following code #include <stdio.h> struct foo { char huge_gap[0x4005d0]; int v; }; void store_and_print(int *val, struct foo *f) { *val = f->v; if ( !f ) { printf("Value is wrong.\n"); } else { printf("Value is correct.\n"); } } int main() { int u… text/html 2010-05-26T09:57:46-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/weekb1?rev=1274893066&do=diff Week -1: Modify JMicron driver, RX offloading May 2nd to May 12th Chanloading with undionly image Stefanha pointed out that I should try if JMicron driver can be chanloaded like: gPXE(jme driver)->gPXE(UNDI driver). Due to the UNDI interface would trigger .irq() function of net_device_operations. text/html 2010-05-26T09:36:05-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/weekb2?rev=1274891765&do=diff Week -2: Implement JMicron driver Apr. 28th to May 1st Day 1 Start to write JMicron Driver. * Helper functions * jme_reload_eeprom(): Which cause the MAC processor to place MAC Address to register. * j{read,write}32(): Register reading/writing wrapper. text/html 2010-05-26T09:18:38-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/weekb3?rev=1274890718&do=diff Week -3: Trace gPXE code base Apr. 22th to Apr. 28th * Trace gPXE driver code: rtl8139.c * Trace gPXE build system. * Review JMicron Ethernet controller spec. text/html 2010-05-26T09:18:00-07:00 http://etherboot.org/wiki/soc/2010/cooldavid/journal/weekb4?rev=1274890680&do=diff Week -4: Debug hardware problem Before Apr. 22th Debug boot crash problem I had problem for boot gPXE with my testing computer. It keeps reboot(crash) right after gPXE relocats itself to high memory location. I've traced partial codes of src/arch/i386/prefix/*prefix.S