summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* [intel] Force RX polling on VMware emulated 82545emMichael Brown2015-04-212-1/+13
| | | | | | | | | | | | The emulated Intel 82545em in some versions of VMware (observed with ESXi v5.1) seems to sometimes fail to set the RXT0 bit in the interrupt cause register (ICR), causing iPXE to stop receiving packets. Work around this problem (for the 82545em only) by always polling the receive queue regardless of the state of the ICR. Reported-by: Slava Bendersky <volga629@networklab.ca> Tested-by: Slava Bendersky <volga629@networklab.ca> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [intel] Report any unexpected interrupt causesMichael Brown2015-04-212-0/+10
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [intel] Update PCI device IDs for Intel 82599 and X540 10G NICsEd Swierk2015-04-211-4/+9
| | | | | | | | | | | | Identifiers are based on defines in Linux ixgbe_type.h. Descriptive names are based on https://www-ssl.intel.com/content/www/us/en/ethernet-controllers/82599-10-gbe-controller-spec-update.html and https://www-ssl.intel.com/content/www/us/en/network-adapters/10-gigabit-network-adapters/ethernet-x540-spec-update.html Signed-off-by: Ed Swierk <eswierk@skyportsystems.com> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [build] Use a single call to parserom.pl to speed up buildingMichael Brown2015-04-181-21/+60
| | | | | | Inspired-by: Robin Smidsrød <robin@smidsrod.no> Tested-by: Robin Smidsrød <robin@smidsrod.no> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [build] Work around binutils quirk on OpenBSD 5.7Michael Brown2015-04-181-0/+3
| | | | | | | | | | | | | The assembler on OpenBSD 5.7 seems not to correctly handle the combinations of .struct and .previous used in unlzma.S, and ends up complaining about an "attempt to allocate data in absolute section". Work around this problem by explicitly resetting the section after the data structure definitions. Reported-by: Jiri B <jirib@devio.us> Tested-by: Jiri B <jirib@devio.us> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [build] Fix compiler warning on OpenBSD 5.7Michael Brown2015-04-181-1/+1
| | | | | | Reported-by: Jiri B <jirib@devio.us> Tested-by: Jiri B <jirib@devio.us> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [build] Allow building PCI ROMs with device ID listsMichael Brown2015-04-153-5/+38
| | | | | | | | | | | | | | PCI v3.0 supports a "device list" which allows the ROM to claim support for multiple PCI device IDs (but only a single vendor ID). Add support for building such ROMs by scanning the build target element list and incorporating any device IDs into the ROM's device list header. For example: make bin/8086153a--8086153b.mrom would build a ROM claiming support for both 8086:153a and 8086:153b. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [prism2] Remove duplicate PCI_ROM() linesMichael Brown2015-04-151-2/+0Star
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [eepro100] Remove duplicate PCI_ROM() lineMichael Brown2015-04-151-1/+0Star
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Use the EFI_RNG_PROTOCOL as an entropy source if availableMichael Brown2015-04-142-3/+234
| | | | | | | | | | | | | | | Entropy gathering via timer ticks is slow under UEFI (of the order of 20-30 seconds on some machines). Use the EFI_RNG_PROTOCOL if available, to speed up the process of entropy gathering. Note that some implementations (including EDK2) will fail if we request fewer than 32 random bytes at a time, and that the RNG protocol provides no guarantees about the amount of entropy provided by a call to GetRNG(). We take the (hopefully pessimistic) view that a 32-byte block returned by GetRNG() will contain at least the 1.3 bits of entropy claimed by min_entropy_per_sample(). Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Poll for TX completions only when there is an outstanding TX bufferMichael Brown2015-04-141-8/+4Star
| | | | | | | | | At least one NII implementation (in a Microsoft Surface tablet) seems to fail to report the absence (sic) of TX completions properly. Work around this by checking for TX completions only when we expect to see one. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Poll media status only if advertised as supportedMichael Brown2015-04-141-2/+14
| | | | | | | | Some NII implementations will fail the GET_STATUS operation if we request the media status. Fix by doing so only if GET_INIT_INFO reported that media status is supported. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Provide a dummy data block in nii_initialise()Michael Brown2015-04-141-1/+6
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Add EFI time sourceMichael Brown2015-04-145-1/+98
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Add EFI entropy sourceMichael Brown2015-04-145-2/+189
| | | | | Originally-implemented-by: Jarrod Johnson <jbjohnso@us.ibm.com> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [util] Add ability to dump PCI device ID listMichael Brown2015-04-132-0/+24
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [romprefix] Allow autoboot device filter to be disabledMichael Brown2015-04-132-0/+3
| | | | | | | | | | | | | | | | | Our current behaviour when booting as a ROM is to autoboot only from devices which are attached via the PCI bus:dev.fn address passed to the ROM's initialisation vector. Add a build configuration option AUTOBOOT_ROM_FILTER (enabled by default) to control this behaviour. This allows for ROMs to be built which will attempt to boot from any detected device, even if not attached via the original PCI bus:dev.fn address. (This is particularly useful when building combined EHCI/xHCI ROMs for USB network boot, since the BIOS may request a boot via the EHCI controller but the xHCI driver will reroute the root hub ports to the xHCI controller.) Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [xhci] Always reset root hub portsMichael Brown2015-04-131-7/+5Star
| | | | | | | | In theory USB3 ports do not require a reset to enable the port. Experimentation shows that this is sometimes required, particularly when rerouting ports from EHCI to xHCI and switching speeds. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [build] Rewrite parserom.pl to support multiple source filesRobin Smidsrød2015-04-131-51/+244
| | | | | | | | | | | | | | | | | Running util/parserom.pl on all source files (637) one by one takes approximately 35 seconds because of the startup cost of each invocation. With the utility rewritten to support multiple source files it now takes approximately 1 second to scan all source files for ROM declarations. The --exclude-driver and --exclude-driver-class options have been added, making it possible to skip certain source files from being scanned at all. In addition --debug option has been added to more easily trace progress. Finally --help option was added to show usage information. Signed-off-by: Robin Smidsrød <robin@smidsrod.no> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [peerdist] Add support for decoding PeerDist Content InformationMichael Brown2015-04-135-0/+1757
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [efi] Ensure drivers are disconnected when ExitBootServices() is calledMichael Brown2015-04-131-0/+35
| | | | | | | | | | | | | | | | | | We hook the UEFI ExitBootServices() event and use it to trigger a call to shutdown_boot(). This does not automatically cause drivers to be disconnected from their devices, since device enumeration is now handled by the UEFI core rather than by iPXE. (Under the old and dubiously compatible device model, iPXE used to perform its own device enumeration and so the call to shutdown_boot() would indeed have caused drivers to be disconnected.) Fix by replicating parts of the dummy "EFI root device" from efiprefix.c to efidrvprefix.c, so that the call to shutdown_boot() will call efi_driver_disconnect_all(). Originally-fixed-by: Laszlo Ersek <lersek@redhat.com> Tested-by: Laszlo Ersek <lersek@redhat.com> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [crypto] Add SHA-512/224 algorithmMichael Brown2015-04-124-0/+122
| | | | | | | | | | SHA-512/224 is almost identical to SHA-512, with differing initial hash values and a truncated output length. This implementation has been verified using the NIST SHA-512/224 test vectors. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [crypto] Add SHA-512/256 algorithmMichael Brown2015-04-124-0/+122
| | | | | | | | | | SHA-512/256 is almost identical to SHA-512, with differing initial hash values and a truncated output length. This implementation has been verified using the NIST SHA-512/256 test vectors. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [crypto] Add SHA-384 algorithmMichael Brown2015-04-124-3/+129
| | | | | | | | | | SHA-384 is almost identical to SHA-512, with differing initial hash values and a truncated output length. This implementation has been verified using the NIST SHA-384 test vectors. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [crypto] Add SHA-512 algorithmMichael Brown2015-04-127-4/+500
| | | | | | | This implementation has been verified using the NIST SHA-512 test vectors. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [crypto] Add SHA-224 algorithmMichael Brown2015-04-125-16/+167
| | | | | | | | | | SHA-224 is almost identical to SHA-256, with differing initial hash values and a truncated output length. This implementation has been verified using the NIST SHA-224 test vectors. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [test] Simplify digest algorithm self-testsMichael Brown2015-04-125-200/+235
| | | | | | | | Update the digest self-tests to use okx(), and centralise concepts and data shared between tests for multiple algorithms to reduce duplicated code. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [virtio] Downgrade per-iobuf debug messages to DBGC2Laszlo Ersek2015-04-101-6/+6
| | | | | Signed-off-by: Laszlo Ersek <lersek@redhat.com> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [test] Add setjmp()/longjmp() self-testsMichael Brown2015-04-072-0/+172
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [libc] Add x86_64 versions of setjmp() and longjmp()Michael Brown2015-04-073-0/+100
| | | | | | | | None of the x86_64 builds currently have any way of invoking these functions. They are included only to avoid introducing unnecessary architecture-specific dependencies into the self-test suite. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [libc] Fix typo in longjmp()Michael Brown2015-04-071-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 8ab4b00 ("[libc] Rewrite setjmp() and longjmp()") introduced a regression in which the saved values of %ebx, %esi, and %edi were all accidentally restored into %esp. The result is that the second and subsequent returns from setjmp() would effectively corrupt %ebx, %esi, %edi, and the stack pointer %esp. Use of setjmp() and longjmp() is generally discouraged: our only use occurs as part of the implementation of PXENV_RESTART_TFTP, since the PXE API effectively mandates its use here. The call to setjmp() occurs at the start of pxe_start_nbp(), where there are almost certainly no values held in %ebx, %esi, or %edi. The corruption of these registers therefore had no visible effect on program execution. The corruption of %esp would have been visible on return from pxe_start_nbp(), but there are no known PXE NBPs which first call PXENV_RESTART_TFTP and subsequently attempt to return to the PXE base code. The effect on program execution was therefore similar to that of moving the stack to a pseudo-random location in the 32-bit address space; this will often allow execution to complete successfully since there is a high chance that the pseudo-random location will be unused. The regression therefore went undetected for around one month. Fix by restoring the correct registers from the saved jmp_buf structure. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [xhci] Support USB1 devices attached via transaction translatorsMichael Brown2015-03-235-33/+164
| | | | | | | | | | | | | | | | | | | | xHCI provides a somewhat convoluted mechanism for specifying details of a transaction translator. Hubs must be marked as such in the device slot context. The only opportunity to do so is as part of a Configure Endpoint command, which can be executed only when opening the hub's interrupt endpoint. We add a mechanism for host controllers to intercept the opening of hub devices, providing xHCI with an opportunity to update the internal device slot structure for the corresponding USB device to indicate that the device is a hub. We then include the hub-specific details in the input context whenever any Configure Endpoint command is issued. When a device is opened, we record the device slot and port for its transaction translator (if any), and supply these as part of the Address Device command. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Improve debug messages for failed control transactionsMichael Brown2015-03-231-6/+5Star
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [ehci] Support USB1 devices attached via transaction translatorsMichael Brown2015-03-232-11/+53
| | | | | | | | | | | | | | | Support low-speed and full-speed devices attached to a USB2 hub. Such devices use a transaction translator (TT) within the USB2 hub, which asynchronously initiates transactions on the lower-speed bus and returns the result via a split completion on the high-speed bus. We make the simplifying assumption that there will never be more than sixteen active interrupt endpoints behind a single transaction translator; this assumption allows us to schedule all periodic start splits in microframe 0 and all periodic split completions in microframes 2 and 3. (We do not handle isochronous endpoints.) Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Clear transaction translator buffers when applicableMichael Brown2015-03-232-4/+72
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Add clear_tt() hub method to clear transaction translator bufferMichael Brown2015-03-235-2/+114
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Reset endpoints without waiting for a new transfer to be enqueuedMichael Brown2015-03-232-31/+70
| | | | | | | | | | | | | | | The current endpoint reset logic defers the reset until the caller attempts to enqueue a new transfer to that endpoint. This is insufficient when dealing with endpoints behind a transaction translator, since the transaction translator is a resource shared between multiple endpoints. We cannot reset the endpoint as part of the completion handling, since that would introduce recursive calls to usb_poll(). Instead, we add the endpoint to a list of halted endpoints, and perform the reset on the next call to usb_step(). Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [xhci] Ring doorbell as part of endpoint resetMichael Brown2015-03-231-0/+3
| | | | | | | | | The endpoint may already have enqueued TRBs at the time that xhci_endpoint_reset() is called. Ring the doorbell to resume processing these TRBs immediately, rather than waiting until the next call to xhci_endpoint_message() or xhci_endpoint_stream(). Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [libprefix] Fix building on 64-bit FreeBSD 8.4Michael Brown2015-03-231-1/+1
| | | | | Reported-by: Pavel Antonov <holly@istu.edu> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Fix USB timeouts to match specificationMichael Brown2015-03-183-7/+37
| | | | | | | | | | | | | | | Several of the USB timeouts were chosen on the principle of "pick an arbitrary but ridiculously large value, just to be safe". It turns out that some of the timeouts permitted by the USB specification are even larger: for example, control transactions are allowed to take up to five seconds to complete. Fix up these USB timeout values to match those found in the USB2 specification. Debugged-by: Robin Smidsrød <robin@smidsrod.no> Tested-by: Robin Smidsrød <robin@smidsrod.no> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [netdevice] Add missing bus types to netdev_fetch_bustype()Michael Brown2015-03-181-0/+4
| | | | | Reported-by: Robin Smidsrød <robin@smidsrod.no> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [ehci] Add support for EHCI host controllersMichael Brown2015-03-187-0/+2370
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [xhci] Do not release ownership back to BIOS when booting an OSMichael Brown2015-03-181-0/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | xHCI (and EHCI) nominally provide a mechanism for releasing ownership of the host controller back to the BIOS, which can then potentially restore legacy USB keyboard functionality. This is a rarely used code path, since most operating systems claim ownership and never attempt to later return to the BIOS. On some systems (observed with a Lenovo X1 Carbon), this code path leads to obscure and interesting bugs: if the xHCI and EHCI controllers are both claimed and later released back to the BIOS, then a subsequent call to INT 16,0305 to set the keyboard repeat rate to a non-default value will lock the system. Obscure though this sequence of operations may sound, it is exactly what happens when using iPXE to boot a Linux kernel via a USB network card. There is old and probably unwanted code in Linux's arch/x86/boot/main.c which sets the keyboard repeat rate (with the accompanying comment "Set keyboard repeat rate (why?)"). When booting Linux via a USB network card on a Lenovo X1 Carbon, the system therefore locks up immediately after jumping to the kernel's entry point. Work around this problem by preventing the release of ownership back to the BIOS if it is known that we are shutting down to boot an OS. This should allow legacy USB keyboard functionality to be restored if the user chooses to exit iPXE, while avoiding the rarely used code paths (and corresponding BIOS bugs) if the user chooses instead to boot an OS. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Add config/usb.h for USB configuration optionsMichael Brown2015-03-183-1/+26
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [autoboot] Match against parent devices when matching by bus type and locationMichael Brown2015-03-171-2/+7
| | | | | | | | | | | | | | | | | When using iPXE as an option ROM for a PCI USB controller (e.g. via qemu's "-device nec-usb-xhci,romfile=..." syntax), the ROM prefix will set the PCI bus:dev.fn address of the USB controller as the PCI autoboot device. This will cause iPXE to fail to boot from any detected USB network devices, since they will not match the autoboot bus type (or location). Fix by allowing the autoboot bus type and location to match against the network device or any of its parent devices. This allows the match to succeed for USB network devices attached to the selected PCI USB controller. Reported-by: Dan Ellis <Dan.Ellis@displaylink.com> Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [xhci] Forcibly disable SMIs if BIOS fails to release ownershipMichael Brown2015-03-161-10/+10
| | | | | | | | If the BIOS fails to gracefully release ownership of the xHCI controller, we can forcibly claim it by disabling all SMIs via the USB legacy support control/status register. Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Add functions for manual device address assignmentMichael Brown2015-03-162-0/+70
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [ncm] Respect maximum transfer size of the busMichael Brown2015-03-161-0/+3
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [usb] Add the concept of a USB bus maximum transfer sizeMichael Brown2015-03-164-3/+12
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
* [libc] Add ffs(), ffsl(), and ffsll()Michael Brown2015-03-164-6/+233
| | | | Signed-off-by: Michael Brown <mcb30@ipxe.org>
generated by cgit v1.2.3-55-g7522 (git 2.39.0) at 2026-02-27 12:30:55 +0100