aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorhpa <hpa>1999-09-27 02:09:55 +0000
committerhpa <hpa>1999-09-27 02:09:55 +0000
commit7e44c9e8c7ee1a10b88ea2015bc9e043794eaa61 (patch)
tree110c9ababa37684bd3c671184c4b81eeac78a44c
parent2e9a9df72034e367d2bd6e4a71ab4d93fcc66b2a (diff)
downloadsyslinux.git-7e44c9e8c7ee1a10b88ea2015bc9e043794eaa61.tar.gz
syslinux.git-7e44c9e8c7ee1a10b88ea2015bc9e043794eaa61.tar.xz
syslinux.git-7e44c9e8c7ee1a10b88ea2015bc9e043794eaa61.zip
Update docs.
-rw-r--r--NEWS5
-rw-r--r--pxelinux.doc25
2 files changed, 12 insertions, 18 deletions
diff --git a/NEWS b/NEWS
index 9db096e9..a1869f90 100644
--- a/NEWS
+++ b/NEWS
@@ -1,9 +1,12 @@
Changes in 1.48:
* PXELINUX: Workaround for PXE ROMs based on the Intel PXE PDK
- 2.0 build 071 and earlier: missing !PXE structure pointer.
+ 3.0 build 071 and earlier: missing !PXE structure pointer.
* PXELINUX: Handle larger BOOTP/DHCP packages.
* PXELINUX: The command line passing was broken; fix.
* PXELINUX: Make COMBOOT images work.
+ * PXELINUX: Documentation on how to make booting work using
+ the PDK 3.0-derived clients, which aren't so generous as to
+ allow booting with only "PXEClient" specified.
Changes in 1.47:
* PXELINUX: RFC 1123 states that a TFTP implementation MUST
diff --git a/pxelinux.doc b/pxelinux.doc
index 43515399..eef86110 100644
--- a/pxelinux.doc
+++ b/pxelinux.doc
@@ -112,15 +112,16 @@ that the filename parameter should not include the /tftpboot prefix.
I have successfully booted an Intel N440BX ("Nightshade") motherboard
with onboard Ethernet and "Intel LANDesk(R) Service Agent II version
-0.99c" (derived from the Intel PXE PDK 1.0) and the Intel PRO/100+
+0.99c" (derived from the Intel PXE PDK V2.0) and the Intel PRO/100+
Management Adapter "Boot agent 2.6 (build 071)" (derived from the
Intel PXE PDK V3.0 bld 071) using the configuration above and ISC
dhcpd 2.0.
Unfortunately, the Intel LANDesk Service Agent II prior to version
-0.99h seems to have a rather serious bug : it requests the TFTP
-"blksize" option, but will be mortally confused if this option is
-actually accepted! There are two possible workarounds for this bug:
+0.99h (PXE PDK V2.4) seems to have a rather serious bug : it requests
+the TFTP "blksize" option, but will be mortally confused if this
+option is actually accepted by the server! There are two possible
+workarounds for this bug:
1. Use a TFTP server with doesn't support "blksize".
@@ -135,18 +136,8 @@ actually accepted! There are two possible workarounds for this bug:
... and on any kernel.org mirror (see http://www.kernel.org/mirrors/).
-2. Add the following DHCP option:
-
- THIS IS AN ILLEGAL OPTION - FIX THIS SECTION
- "option vendor-encapsulated-options 08:00:00:01:7f:00:00:01 ;"
-
- That line is not supposed to be necessary, and in fact makes
- booting significantly slower, since it will attempt (and fail) an
- MTFTP transfer before falling back to TFTP. For some reason, the
- boot PROM doesn't request the "blksize" option if it has already
- failed an MTFTP transfer, and thus the bug is avoided.
-
-It appears this bug was fixed in version 0.99h.
+2. Use MTFTP for the initial bootstrap. You need an MTFTP server to
+ do this.
++++ SOME NOTES ++++
@@ -169,7 +160,7 @@ Currently known problems:
+ Requires a TFTP server which supports the "tsize" option.
+ The error recovery routine doesn't work quite right. For right now,
- just do a hard reset.
+ it just does a hard reset - seems good enough.
+ There may be funnies with memory management. The PXE spec has no
decent way of telling it to free up all memory and unchain any
interrupts; it allows the base stack to be unloaded, but not the